Introduction

Foot-and-mouth disease (FMD) is a terrible viral infectious disease, attacks cloven-hoofed animals, including domestic livestock such as cattle, pigs, and sheep. FMD is enzootic in Africa, Asia, and South America (Perry and Rich 2007). The disease is highly contagious and outbreaks affect the economy through the loss of production, restriction of the trade, and tourism in affected regions and present a constant threat for FMD-free countries (Clavijo et al. 2004).

Foot-and-mouth disease virus (FMDV), the etiologic agent of FMD, is a non-enveloped, positive-single-stranded RNA virus, belonging to the genus of Aphthovirus. The full length of FMDV RNA is about 8.4 kb. It has seven significant serotypes that are not immunologically related (Jamal and Belsham 2013). FMDV fragment has a four-part structural protein and an eight-part non-structural protein (Du et al. 2007). VP1 as a structural protein is the major immunogenic protein of FMDV as it contains both T and B cell epitopes which trigger the immune system to induce a cellular immune response as well as sufficient levels of neutralizing antibodies for protection against infection (Wong et al. 2000; Peralta et al. 2013).

The traditional inactivated FMD vaccine currently is used in most countries to protect and eradicate the disease. Although it plays a great role in providing immune protection, it has several drawbacks: problems in manifestation procedures such as the improper viral inactivation (Doel 2003). Also, the vaccine antigen needs more purification process to prevent undesirable allergic reaction and allow the differentiation between vaccinated and infected animals throughout control management; in addition, it could not minimize the shedding of FMDV since it takes a long time to the outset of immunity (Cao et al. 2017). For all of this, researchers are trying to develop alternative vaccines to overcome these problems.

In previous reports, peptide vaccines based on immunogenic epitopes were developed to overcome the shortcomings of conventional FMD vaccines. However, peptide vaccines could not elicit adequate immune responses alone; it must be used with some specific adjuvants (Li et al. 2014). Live vector vaccines are the most attractive type among these modern vaccines since it combines the components of a subunit or DNA vaccine with the immunogenic characteristics of a living replicating organism to be more like an adjuvant. A potent attenuated bacterial strain as a vaccine vector has been established for its advantages over conventional vaccines or subunit vaccines. It has the ability to deliver heterologous antigens through its coding gene. The concept beyond this approach is using the immunological features of the vector to trigger an immune response against both its own and inserted heterologous antigen genes. Typically, this vaccine type has the ability to stimulate the immune system, as it happened in natural infections, to stimulate CD4+ and CD8+ T cell subsets. The disadvantage of this approach is the intensity of the immune response is directly related to the pathogenicity and ability of the vector to be replicated and it could be overcome via attenuation of the used vector (Nascimento and Leite 2012).

Listeria monocytogenes (LM), an intracellular facultative microorganism, possesses a wide range of mammalian hosts (Saklani-Jusforgues et al. 2003). The capability of LM to invade cytosol of the host cell after phagocytosis and deliver the heterologous antigen to the cytoplasm makes it a potent vaccine vector to conduct a specific immune response (Bierne et al. 2018; Dussurget et al. 2004)

In this study, we developed a live attenuated LM strain by knocking out the virulent genes of actA and plcB, and constructed a recombinant strain based on such attenuated LM containing FMDV vp1 gene and used mice as a model to assess both humoral and cellular immune responses that may offer the protection against FMDV infection as a novel vaccine (Fig. 1).

Fig. 1
figure 1

Schematic diagram of the current study. Several procedures to generate LM△actAplcB-vp1 recombinant strain as a novel vaccine against FMDV and evaluation of its immunological role

Full size image

Materials and methods

Bacterial strains and plasmids

Listeria monocytogenes 10403S is kindly offered by Dr. Hao Shen (Department of Microbiology, Perelman School of Medicine, University of Pennsylvania). Plasmids utilized in this paper are presented in Table 1.

Table 1 The plasmids used in this work
Full size table

Construction of attenuated mutant LM△actAplcB-lacZ

The integrative pCW182 plasmid carrying the lacZ gene (Accession No. J01636.1) was firstly constructed. The deletion mutant was achieved by chromosomal integration of pCW182 vector with LM wild-type strain. To construct pCW182, plasmid pCW108 was digested by SpeI and NotI restriction enzymes; then, LM-orfBAldh gene was ligated with T4 DNA ligase enzyme (TaKaRa Biotechnology, China) to construct pCW108-LMorfBAldh. With the same context, the plasmid pCW108-LM-orfBAldh was digested with XbaI and NotI restriction enzymes; then, LM-mpl gene was ligated to perform pCW108-LM-orfBAldh-mpl. Finally, the plasmid was digested with NotI restriction enzyme to ligate lacZ gene to generate pCW182 (Fig. 2).

Fig. 2
figure 2

Construction of pCW182 plasmid. The plasmid was constructed to be integrated with LM strain to generate LM△actAplcB-lacZ vector strain. (LM-mpl), a homologous fragment for homologous recombination translates to metalloprotease, has a role in vacuole escape. (LM-orfBldh), one of the characterized open reading frames of bacterial genome coding L-lactate dehydrogenase, is also a homologous fragment for homologous recombination

Full size image

According to the described protocol (Wang et al. 2014), the integrative plasmid pCW182 was introduced into LM competent cell by electroporation and grown in brain/heart infusion (BHI) selection media containing erythromycin. Presumptive positive blue colonies were grown on BHI plates with erythromycin at 42 °C for 48 h. Putative integrates were subsequently grown in BHI broth at 30 °C for six consecutive passages. Erythromycin-sensitive and white derivatives were identified and screened by genome PCR and gene sequencing.

Construction of antigen gene cassettes

The linearized pCW203 plasmid was generated with HindIII and XhoI restriction enzymes (TaKaRa Biotechnology, China). To generate pCW203-vp1, distinctively designed primers of about 44 bp were generated which included complementary sequences of both terminal ends of linearized pCW203 plasmid and vp1 gene (forward primer: 5′-GAGAAGTGAAAGACCACAAGCTTTGACAACAAGTACTGGTGAAT-3′ and reverse primer: 5′-CGATGCGGCCGCTCGAGTTGTTTTACTGGAGCAACAATTT-3′). PCR product of amplified vp1 gene from pUC57-vp1 utilizing these designed primes revealed to a new fragment of vp1 including the complementary sequence of both terminal ends of the linearized pCW203 plasmid. After purification of this new fragment of vp1, it was mixed with the linearized pCW203 plasmid to be ligated in an overlapping manner utilizing Sosoo cloning kits (Tsingke Company, China) at 50 °C for 15 min (Fig. 3).

Fig. 3
figure 3

Construction of pCW203-vp1 plasmid. (a) Amplification of FMDV-vp1fragment with the specially designed primers. (b) PCR fragment of vp1 includes both complementary sequences of pCW203 plasmid terminal ends. (c, d) Ligation of the new fragment of vp1with the linearized pCW203 plasmid at restriction site between HindIII and XhoI. (e) pCW203-vp1 plasmid encompasses (HA) human influenza hemagglutinin epitope (YPYDVPDYA) is a western blot antigen marker. (GST) is glutathione S-transferase for purified, visualized of the target protein. (AmpR) is ampicillin resistance gene. (lacZ a gene) encodes for the α fragment in the N terminal of bêtegalactosidase

Full size image

pCW154 and pCW182 were digested with Not I restriction enzyme; then, expression cassette from pCW154 was ligated in pCW182 NotI site with T4 DNA ligase enzyme (TaKaRa Biotechnology, China) to construct pCW702.

In order to construct the integrative plasmid pCW702-vp1 that contain antigen gene cassette, pCW702 was digested with XhoI and BamHI restriction enzymes to perform a linearized pCW702 plasmid and the vp1 fragment fused with HA gene was amplified from pCW203-vp1 with specially designed primers were generated (forward primer: 5′-TATAATTTTGCTACTATGAAGGATCCATACCATATGATGTTCCAGATTAG-3′, and reverse primer: 5′-CTGGTCCATTTAATCCCTCGAGTTGTTTTACTGGAGCAACAATTT-3′). Then, vp1 gene was ligated with pCW702 plasmid in an overlapping manner utilizing Sosoo cloning kits at 50 °C for 15 min (Fig. 4).

Fig. 4
figure 4

Generation of pCW702-vp1 that contain the antigen cassette. To build a target antigen cassette, vp1 must be ligated to pCW702 plasmid at XhoI and BamHI restriction sites. (a) Amplification of vp1-HA fragment with the designed primers. (b) The generated new fragment included the complementary sequence of linearized pCW702 at XhoI and BamHI restriction sites. (c, d) Overlap ligation of the vp1 fragment with the pCW702 plasmid to generate pCW702-vp1. (e) The antigen cassette to be integrated into LM genome contained (Phly) is the hly promoter of LM. (ss) is a secretory signal from the LM hly gene. (GP33) is the epitopes (KAVYNFATM) originated from LCMV for stimulation of CD8+ T cells. (HA) is the human influenza hemagglutinin epitope (YPYDVPDYA) as a western blot antigen marker. (vp1) is the main target gene of FMDV. (GP61) is the epitopes (GLKGPDIYKGVYQFKSVEFD) originated from LCMV for stimulation of CD4+ T cells. And (VSV-G) is a short peptide sequence (YTDIEMNRLGK) from vesicular stomatitis virus as a western blot antigen marker. Upstream of the antigen cassette is a gene (LM-mpl), a homologous fragment for homologous recombination translates to metalloprotease, has a role in vacuole escape. Downstream of the antigen cassette is a gene (LM-orfBldh), one of characterized open reading frames of bacterial genome coding L-lactate dehydrogenase, is also a homologous fragment for homologous recombination

Full size image

Integration of gene cassettes into LM genome

The target plasmid pCW702-vp1 was introduced to the LM△actAplcB-lacZ competent cell by gene pulser (electroporation) method and cultivated in BHI agar plates as mentioned before (Fig. 5, Fig. S1).

Fig. 5
figure 5

Construction of the recombinant LMΔactAplcB-vp1strain. For chromosomal integration, antigen cassette containing the target gene vp1 between two NotI restriction sites in pCW702-vp1 was inserted in LM△actAplcB-lacZ by electroporation to generate homologous recombination to construct stable LMΔactAplcB-vp1

Full size image

Western blot analysis

After cultivation of LM△actAplcB-vp1 recombinant strain, both intracellular and extracellular proteins were collected by TCA and sonication. Gel electrophoresis was used for separation of protein, then loaded in SDS-PAGE. In order to transfer protein from the gel, we used PVDF membrane in the buffer. The skimmed milk and Tween-20 were added for membrane blocking. Anti-HA tag mouse monoclonal antibody (Thermo Fisher, USA) was used as primary antibodies and horseradish peroxidase conjugate secondary antibodies (Sigma company, USA) was used. The accurate bands were seen after adding chemiluminescent substrate.

Characterization of bacterial growth in vivo

In order to determine the bacterial growth curve of LM△actAplcB-vp1 recombinant strain in vivo, groups of C57BL/6J mice (30 mice per group) were injected intravenously with 0.1 × LD50 (106 CFU) of LM△actAplcB-vp1 in a volume of 100 μl of normal saline; LD50 was calculated by improved Spearman-Karber method (Ramakrishnan 2016). Six mice from each group were sacrificed at 1, 2, 3, 5, and 7 days after injection. After homogenization of internal organs (liver and spleen) in PBS containing 0.1% Triton-X100, serial dilutions of homogenate were cultivated on BHI agar plates at 37 °C for 24–48 h. Counting of colonies was done after the incubation period.

Immunization of mice

Based on LD50 calculation, 0.1 ml of normal saline containing 106 CFU LM△actAplcB-vp1 (0.1 × LD50) was injected I.V. in 6–8 weeks old C57BL/6J mice groups (9 mice/group), while control groups (9 mice/group) were injected intravenously with 0.1 ml normal saline or 0.1 ml of normal saline containing 106 CFU LM△actAplcB-lacZ. Mice groups were immunized for a booster dose after 15 days from the first injection. Mice were kept under restricted hygienic conditions of Animal house at the School of Public Health, Sichuan University. Blood samples were collected at 14 days post the primary immunization and at 14 and 28 days post the booster dose. For intracellular cytokine staining, groups of mice were sacrificed at 9 days post the primary dose and another group at 9 days post the secondary does.

Intracellular cytokine staining

Splenocytes were harvested from both immunized and control mice groups after scarification. The cells were stimulated with 0.1 μg/ml GP-33 (KAVYNFATM) and 0.1 μg/ml GP 61(GLKGPDIYKGVYQFKSVEFD) or without peptides (10% FBS-1640) at 37 °C and 5% CO2 for 4 h. Cells were collected and stained with PerCP-anti-mouse CD4 and APC-CyTM7-anti-mouse CD8 antibodies (BD PharMingen, China) at 4 °C for 30 min. Afterward, The cells were fixed by the Cytofix/Cytoperm Kit (BD PharMingen, China). Intracellular staining was performed with PE-anti-mouse IFN-γ, APC-anti-mouse IL-2, and PE-CyTM7-anti-mouse TNF-α antibody (Biolegend, USA) at 4 °C for 45 min. The stained cells have been tested with BD FACSverse flow cytometer with Flowjo software.

Detection of specific antibody

The specific antibodies against FMDV-vp1 were captured by indirect ELISA. ELISA plates of 96 wells were coated with 100 μl of FMDV-VP1 peptide type “O1” diluted in NaHCO3 buffer (10 mmol/l, pH 9.6) at a final concentration (2 μg/ml), then incubated at 4 °C overnight. Following washing the plates with washing buffer, 200 μl of blocking buffer was appended and incubated at 37 °C for 1 h. Diluted sera (1:100) were added and incubated at 37 °C for 1 h. Following washing step, 100 μl/well of labeled goat anti-mouse IgG-horseradish peroxidase (1:3000) (Sigma company, USA) were added and incubated at 37 °C/30 min. After adding of 50 μl substrate buffer for 20 min, 50 μl of 1.2 mol/l H2SO4 was added to stop the reaction. The absorbance rate was detected by ELISA reader. The antibody levels were expressed of the optical density values (OD490) for the mice sample.

Statistical analysis and graph design

The obtained data were analyzed using the Holm Sidak’s multiple-comparison t test. And GraphPad prism v 7.0 (P < 0.05) was considered statistically significant. SnapGene software was used for plasmid design.

Gene availability

FMDV-vp1 gene serotype (O1) isolate of HKN/19/2010 (Accession No. JQ070305.1.). Codon optimization of FMDV-vp1 gene serotype (O1) (Accession No. MK050946). Listeria monocytogenes 10403S as the parent strain is publicly available at BEI Resources. Catalog No. NR-13223. Taxonomy ID: 393133 (NCBI:txid393133). The lacZ gene (Accession No. J01636.1).

Results

Construction affirmation of live vector platform; attenuated mutant LM△actAplcB-lacZ

The most virulent genes of LM actA and plcB were knocked out by chromosomal integration of pCW182 plasmid and replaced with lacZ gene. After introducing the integrative plasmid pCW182 into LM wild strain by gene pulser, cultivation was done to achieve the attenuated LM△actAplcB-lacZ vector platform. PCR and gene sequencing results confirmed the construction of LM△actAplcB-lacZ (Fig. S2, Fig. S3).

Confirmation of LM recombinant strain construction

After several procedures of gene digestion and ligation, plasmid pCW702-vp1 include the expression gene cassette was successfully constructed. pCW702-vp1 was introduced to LM△actAplcB-lacZ by electroporation, following several generations of cultivation to accomplish the homologous recombination of attenuated LM△actAplcB-lacZ with the heterologous antigen gene cassette to establish LM△actAplcB-vp1 recombinant strains Fig.4. Afterward, the cultivated bacterial colonies become white color and erythromycin sensitive as a successful construction. PCR results confirmed the successful integration of the heterologous gene into LM△actAplcB-lacZ genome as vp1 gene was amplified from the bacterial genome and erythromycin resistance gene was undetectable as in (Fig. S4) Moreover, the confirmation was achieved by gene sequencing and the required result was achieved (data not shown).

Protein expression analysis

The expression of LM△actAplcB-vp1 in vitro was so faint and weak, so it could not be detected by western blot (data not shown).

The load of LM△actAplcB-vp1 recombinant strain in organs post-immunization

In order to determine the growth and multiplication of LM△actAplcB-vp1 in the internal organs, particularly liver and spleen, C57BL/6 mice groups were inoculated I.V. with 0.1 × LD50 of LM△actAplcB-vp1. Data revealed that the post-infection load of LM△actAplcB-vp1 in liver reached its peak of more than 105 CFU at the first day post-injection and kept almost for 3 days. However, at the fifth day, post-injection was notably decreased to average about 10 times lower than that in the first 3 days and lowered to the undetectable level at the seventh day post-injection. The growth rate of LM△actAplcB-vp1 in the spleen was not the same as in the liver. The peak of the growth (105 CFU) was in the first day post-injection. A moderate decrease in load amount was observed from the second day and reached to undetectable in the fifth day post-injection (Fig. 6) (Table S1).

Fig. 6
figure 6

The growth curve of LMΔactAplcB-vp1 in mice liver and spleen. To detect LMΔactAplcB-vp1 load in liver and spleen of mice, groups of C57BL/6J mice were injected with 0.1 × LD50 (106 CFU) of LM△actAplcB-vp1 and were sacrificed at 1, 2, 3, 5, and 7 days after injection. a Bacterial load in the liver was at the peak at the first day post-injection and gradually decreased at the third day to be undetectable at the seventh day post-injection. b Spleen bacterial load showed the maximum level at the first day post-injection and slightly decrease on the third day and could not be detected on the fifth day post-injection. Results are expressed as means ± SEM per group. The interrupted line refers to the undetectable level of bacteria

Full size image

Neutralizing antibodies against FMDV-VP1

To evaluate the immunogenicity of LM△actAplcB-vp1, groups of C57BL/6J mice were immunized I.V. with 106 CFU LM△actAplcB-vp1 and 106 of LM△actAplcB-lacZ or saline as a control group. Sera were collected at 14dp first immunization and at 14 and 28dp booster dose and were analyzed for detection of specific IgG antibodies against the VP1 protein. The titers of antibody reached a detectable level in the vaccinated groups (0.675 ± 0.07734) while the control groups were (0.103 ± 0.01762) after the first immunization. Moreover, augmentation in antibody level was observed at the 14-day post the second dose (0.9625 ± 0.02631) while the control one was (0.107 ± 0.01412) and reached the peak at 28 days following the booster dose (1.467 ± 0.1646) with the highest level of 2.43 while the control groups were (0.129 ± 0.01252) (Fig. 7).

Fig. 7
figure 7

Detection of antibodies after immunizations with LMΔactAplcB-vp1. Specific antibody level against FMDV-vp1 was detected in sera of mice by indirect ELISA. a, b, c Representative comparative between the levels of antibody in immunized mice at different days and the control groups. d Antibody level in all immunized mice with LMΔactAplcB-vp1 after and before the booster dose. Data were expressed as means ± SEM per group and P values were determined by Holm Sidak’s multiple-comparison t test. ****P < 0.0001 and ***P < 0.001

Full size image

Cellular immunity against the expressed target gene

The assessment of cell-mediated immune response against LM△actAplcB-vp1 recombinant strain was performed by detection the levels of CD4+ and CD8+ T cell-specific secreted cytokines (IFN-γ, TNF-α, and IL-2) in vaccinated and control mice. The levels of CD4+ secreted IFN-γ, TNF-α, and IL-2 after the 1st dose was 0.259%, 0.246%, and 0.036%, respectively. The levels of CD8+ secreted IFN-γ, TNF-α, and IL-2 after the 1st dose were 0.402%, 0.371%, and 0.073%, respectively. After 9 days of the booster dose, the level of CD4+ secreted IFN-γ, TNF-α, and IL-2 became 0.501%, 0.283%, and 0.230%, respectively. The levels of CD8+ secreted IFN-γ, TNF-α, and IL-2 were 0.881%, 0.899%, and 0.213%, respectively. The current date revealed that the level of CD4+ and CD8+ secreted cytokines in vaccinated mice were significantly higher than control and naïve groups (p < 0.005). The level of CD8+ secreted cytokines was significantly higher than CD4+ secreted cytokines in vaccinated mice, while the level of all secreted cytokines T cells was raised after the booster dose in comparison with the level after the primary dose (Fig. 8).

Fig. 8
figure 8

Ag-specific CD4+ and CD8+ T cell cytokines. Specific secreted CD4+ and CD8+ T cell cytokine levels were estimated in both immunized and control mice groups at 9 days post the primer a and booster dose. a1, a2 The percentage of secreted CD4+ and CD8+ IFN-γ was revealed. b1, b2 The percentage of secreted CD4+ and CD8+ IL-2 was showed. c1, c2 The percentage of secreted CD4+ and CD8+ TNF-α was indicated. There was a significant increase in the percentage level of CD4+ and CD8+ secreted cytokines in immunized mice than control groups and a significant increase was observed in the level of CD4+ and CD8+ secreted cytokines after the second immunization. ****P < 0.0001

Full size image

Discussion

FMD is a contagious disease and causes loss of production, abortions, perinatal mortalities, and impediment of the global trade of livestock with a severe economic impact (James and Rushton 2002). The aim of this study is to provide a novel FMD vaccine to overcome the disadvantages of conventional one such as improper inactivation and the problem of differentiation between infected and vaccinated animals. Besides, the recombinant vaccine can be used in FMD-free countries which adopt killing policy of infected animals instead of vaccination with inactivated vaccine. The live attenuated LM strain in this experiment was constructed via deletion the virulent genes of LM actA and plcB. actA gene encodes actA protein, responsible for actin-based mobility and the spread of bacteria from cell to cell. While plcB gene encodes phospholipase C, it has an important role in cell invasion (Gründling et al. 2003). This gene mutation prevents the pathogenicity of LM and keeps its immunological characteristics.

LM△actAplcB-vp1 candidate vaccine was generated via homologous recombination, in which the gene expression cassette of integrative plasmid pCW702-vp1 was chromosomally integrated with genetically defined mutant LM (LM△actAplcB-lacZ) chromosome. Chromosomal integration technique has several advantages, including single copy number of the target gene is required to be integrated (Lauer et al. 2002); it provides stability and a specific directive site in a bacterial chromosome for inserted gene without the need of antibiotic marker (Gu et al. 2015).

Western blot analysis results indicated the LM△actAplcB-vp1 strain was unable to express the VP1 proteins in vitro. It attributed to LM-prfA gene, which is the key factor that organizes the transcription of other chromosomal genes (Scortti et al. 2007). Past findings indicated that prfA gene is scarce to be expressed in vitro but strongly expressed in vivo since the cultivation condition of LM in vitro is insufficient to produce detectable levels of prfA protein (Shetron-Rama et al. 2002; Toledo-Arana et al. 2009). Therefore, it was not surprising that the LM△actAplcB-vp1 could express the integrated proteins and induce anti-FMDV-VP1-specific immune response in vivo although it could not be expressed in vitro.

Safety of the attenuated bacterial strain carrying heterologous genes is of a great importance in efficacy evaluation of live vectors. The obtained data revealed that mutant LM△actAplcB-vp1 with deletion of the most virulent genes (actA and plcB) generate safe recombinant strain since the load of LM△actAplcB-vp1 in the liver and spleen of immunized mice was cleared and eliminated within 3–5 days. After intravenous injection of LM△actAplcB-vp1 near to 80% of the inoculum were cleared in the liver (Conlan 1999). Near to 80% of I/V injection of a sub-lethal dose of LM△actAplcB-vp1 were trapped by the Kupffer cells of the liver which conducted inactivation for most of ingested bacteria within the first few hours of infection. The growth rate of listeria in the liver kept its peak for 3–4 days, then decline upon the onset of specific immunity. Many cells contributed in the elimination of LM△actAplcB-vp1 including Kupffer cells, natural killer (NK) cells, immigrating macrophages, and neutrophils (Cousens and Wing 2000; Conlan 1999). While in the spleen where less than 20% of I/V inoculum trapped by splenic macrophages, Listeria-specific CD4+ and CD8+ T cells were triggered early during the second day, meaning that LM△actAplcB-vp1 were rapidly processed and presented with MHC class I and class II (Conlan 1996). This can explain the difference in the growth rate in the liver and spleen.

The humoral immune response against FMDV is important either in infected or vaccinated animals since it provides the desirable protection which persists for a long period. In the current study, high antibody level against FMDV-VP1 was detected in immunized mice with LM△actAplcB-vp1. The average level of antibodies after the prime immunization was 0.67 and reached 1.46 at 14 days followed by the booster dose, while the highest level was 2.43 at 28 days after the second immunization.

VP1 is a considerable immunogenic protein of FMDV which contains the most antigenic site making it able to induce sufficient neutralizing antibodies to provide strong protection against FMD infection. Previous reports indicated that LM has a minor effect on enhancing the humoral immunity since it is an intracellular bacterium and most of the researchers focused only on its role in cellular immunity. Our data revealed that, although we could not detect the expression of LM△actAplcB-vp1 in vitro, the existence of secretory signal (ss) in antigen cassette and prfA gene in bacterial genome lead to the full expression of the inserted cassette in vivo and production of the VP1 target protein which was neutralized by specific IgG antibodies. It was suggested that during bacteremia, the immune system induce specific antibodies against released extracellular antigens of LM△actAplcB-vp1 following the rapture of phagosome (Bhunia 1997; Edelson et al. 1999) which can explain the detection of neutralizing antibodies against the VP1 protein. These results support our new approach of FMD vaccine using mutant LM as a delivery platform.

Though the protection role of cellular immunity against FMDV has some argument, many studies indicated that specific T cell antiviral responses against FMDV, including CD4+ and CD8+ T cells, have been observed in vaccinated and infected animals (Bautista et al. 2003; Gerner et al. 2009). Moreover, it has been suggested cell-mediated immunity is implicated in the elimination of FMDV from persistently infected animals (Guzman et al. 2008; Juleff et al. 2009); these results suggested T lymphocytes play a great role in protection against FMDV. The current data showed that the immunized mice with LM△actAplcB-vp1 induced a detectable level of CD4+ and CD8+ secreted cytokines at 9 days post the first immunization and another elevation at 9 days following the booster dose. Such results could be explained in the context of “cross-presentation” in which antigen presenting cells (APCs) capture LM△actAplcB-vp1 by direct phagocytosis is important for stimulating T lymphocytes (Goldfine and Shen 2007; Heath and Carbone 2001). Also, MHC class I and class II have a great role in presenting the LM△actAplcB-vp1 to CD8 and CD4 cells through internal and external pathways, respectively, that led to more stimulation for a specific T cell immune response (Szalay et al. 1994). In comparison with previous studies to develop a new subunit vaccine against FMDV utilizing bacterial or viral live vectors, some live vectors induce cellular immunity only and the other induce both types of immunity but the level of specific antibodies was lower than ours (Li et al. 2007; Sanz-Parra et al. 1999) which make LM a promising bacterial vector platform.

In conclusion, LMΔactAplcB-vp1 candidate vaccine was successfully constructed to induce a specific immune response against FMDV utilizing attenuated LM mutant strain as a live vector. The immunized mice with LMΔactAplcB-vp1 induced both humoral and cellular immunity against the target protein (VP1). In addition, LMΔactAplcB-vp1 is well tolerant in organs and eliminated within 5 days. All these features make LMΔactAplcB-vp1 a promising vaccine against FMDV infection.