BIOTECHNOLOGICAL PRODUCTS AND PROCESS ENGINEERING

Lactobacillus acidophilus as a live vehicle for oralimmunization against chicken anemia virus

Hassan Moeini & Raha Abdul Rahim &

Abdul Rahman Omar & Norazizah Shafee &

Khatijah Yusoff

Received: 5 October 2010 /Revised: 29 November 2010 /Accepted: 30 November 2010 /Published online: 23 December 2010# Springer-Verlag 2010

Abstract The AcmA binding domains of Lactococcuslactis were used to display the VP1 protein of chickenanemia virus (CAV) on Lactobacillus acidophilus. One andtwo repeats of the cell wall binding domain of acmA genewere amplified from L. lactis MG1363 genome and theninserted into co-expression vector, pBudCE4.1. The VP1gene of CAV was then fused to the acmA sequences and theVP2 gene was cloned into the second MCS of the samevector before transformation into Escherichia coli. Theexpressed recombinant proteins were purified using a His-tag affinity column and mixed with a culture of L.acidophilus. Whole cell ELISA and immunofluorescenceassay showed the binding of the recombinant VP1 proteinon the surface of the bacterial cells. The lactobacilli cellscarrying the CAV VP1 protein were used to immunizespecific pathogen-free chickens through the oral route. Amoderate level of neutralizing antibody to CAV wasdetected in the serum of the immunized chickens. A VP1-specific proliferative response was observed in splenocytesof the chickens after oral immunization. The vaccinatedgroups also showed increased levels of Th1 cytokines

interleukin (IL)-2, IL-12, and IFN-γ. These observationssuggest that L. acidophilus can be used in the delivery ofvaccines to chickens.

Keywords Chicken anemia virus . Oral vaccine . AcmAanchor protein . Surface display . Lactobacillus acidophilus

Introduction

The oral administration of vaccines has some advantagesover other routes. Oral immunization is convenient,relatively inexpensive, easy boosting, and can be carriedout on a large scale. Furthermore, oral immunization hasthe capacity to induce immunity in the mucosal surface,where many viral and bacterial pathogens enter the body.Two main live bacterial delivery systems, either usinggenetically attenuated strains of pathogenic bacteria such asSalmonella sp. (Dertzbaugh 1998) and Mycobacterium(Stover et al. 1993) or using commensal or food-gradebacteria such as lactic acid bacteria (LAB), especiallylactococci and lactobacilli (Zegers et al. 1999; Maggi et al.2000; Lee et al. 2001; Ribeiro et al. 2002; Scheppler et al.2002; Dieye et al. 2003) have been developed for oralimmunization purposes. As live delivery vehicles, LABsare safer alternatives to live attenuated pathogens due totheir GRAS status and long history of use in the productionof fermented food.

The number of studies on oral immunization in chickensis very limited. One example is a report by Dieye et al.(2003) in which Lactococcus lactis was used as livemucosal vaccine against chicken infectious bursal diseasevirus (IBDV). They reported the development of alactococcal system for expressing and targeting the IBDVantigenic proteins, VP2 and VP3 to the cytoplasm, the cell

H. Moeini :N. Shafee :K. Yusoff (*)Department of Microbiology,Faculty of Biotechnology and Biomolecular Sciences,Universiti Putra Malysia (UPM),43400 Serdang, Selangor, Malaysiae-mail: kyusoff@biotech.upm.edu.my

R. A. RahimDepartment of Molecular Biology,Faculty of Biotechnology and Biomolecular Sciences,Serdang, Selangor, Malaysia

A. R. Omar :K. YusoffInstitute of Bioscience, Universiti Putra Malaysia,43400 Serdang, Selangor, Malaysia

Appl Microbiol Biotechnol (2011) 90:77–88DOI 10.1007/s00253-010-3050-0

wall, or the extracellular compartment of L. lactis.However, no immune response was detected to the IBDVVP2 and VP3 proteins in the immunized chickens.

In the present study, the development of a novel oralvaccine against chicken anemia virus (CAV) in chickenusing CAV VP1-carrying lactobacilli as live deliveryvehicle has been reported. The binding domain of AcmAanchor motif of L. lactis was used to display the VP1protein of CAV on the surface of the lactobacilli cells.AcmA, N-acetylmuramidase (Fig. 1a), is the major autolysinof L. lactis cell wall, consisting of an N-terminal active sitedomain and a C-terminal peptidoglycan-binding domaincontains three repeats of 44 amino acids which are involvedin cell wall binding (Buist et al. 1995). Only one of theserepeats is sufficient for cell wall binding (Buist 1997; Rahaet al. 2005). The AcmA binding domain of L. lactis has beenshown to be able to bind to the cell wall of a range of Gram-positive bacteria, including Lactobacillus, Clostridium,Listeria, and Bacillus, when added from the outside (Buist1997; Raha et al. 2005; Okano et al. 2008). Raha et al.(2005) reported that the VP1 protein of enterovirus 71 linkedto one repeat of AcmA binding domain purified from a

recombinant Escherichia coli can be successfully bound onthe surface of L. lactis, Lactobacillus plantarum, andBacillus sphericus by mixing it with the live cells. This E.coli expression-based system was chosen to be used toproduce and display the antigen-anchor fusion proteins onthe cell surface of L. acidophilus.

Material and methods

Microorganisms and vectors

L. lactis MG1363 was used for the amplification of theacmA gene. Lactobacillus acidophilus (ATCC 53672) wasused as carrier for viral proteins. E. coli BL21 (DE3) wasthe strain used as the expression host cell. ThepETDuetTM-1 (Novagen, Germany) vector was used forthe construction of the recombinant plasmids. ThepETDuetTM-1 contains two multiple cloning sites (MCS),each of which is preceded by a T7 promoter/lac operatorand a ribosome binding site. The Malaysian CAV isolateSMSC-1 (GenBank accession no. AF285882) and recom-

Fig. 1 a Schematic illustration of the AcmA protein. The AcmAprotein consists of two domains, an N-terminal catalytic domainincluding signal peptide (S.P.) and an active site (A.S.), and a C-terminal anchoring domain. The anchoring domain contains threerepeats (LysM1-3) that are highly homologous. b Schematic illustra-tion of the fusion proteins encoding by pETVP1/acmA1 and pETVP1/acmA2 plasmids. c Map of the recombinant plasmids pETVP1/

acmA1-VP2 and pETVP1/acmA2-VP2. The CAV VP1 gene wasinserted into EcoRI and PstI sites of pETacmA1 and pETacmA2plasmids at upstream of acmA gene to construct pETVP1/acmA1 andpETVP1/acmA2, respectively. The CAV VP2 gene was then clonedinto NdeI and XhoI of the MCS2 of pETVP1/acmA1 and pETVP1/acmA2 plasmids to construct pETVP1/acmA1-VP2 and pETVP1/acmA2-VP2

78 Appl Microbiol Biotechnol (2011) 90:77–88

binant plasmid pCR(VP1-VP2) containing VP1 and VP2genes of CAV isolate SMSC-1 (Chowdhury et al. 2003)was obtained from Prof. Abdul Rahman Omar, Faculty ofVeterinary Medicine, Universiti Putra Malaysia.

Construction of plasmids pETacmA1 and pETacmA2

One (acmA1) and two repeats (acmA2) of the cell wallbinding domain of acmA gene were amplified with theacmA-specific primers listed in Table 1, using the genomicDNA of L. lactis MG1363 (accession no. LLU17696) as atemplate. The amplified acmA1 and acmA2 fragments wereinserted into the PstI and HindIII sites of the MCS1 ofpETDuetTM-1 to construct pETacmA1 and pETacmA2,respectively. The recombinant plasmids were transformedinto E. coli BL21 (DE3) by the chemical transformationmethod (Sambrook and Russell 2001). The correct orienta-tion of the inserts was analyzed by restriction enzymeanalysis and polymerase chain reaction (PCR). The sequenceof the genes was confirmed by double-stranded sequencing(Medigene, Malaysia).

Cloning of the VP1 and VP2 genes of CAV

The CAV genes, VP1 and VP2, were amplified from therecombinant pCR(VP1-VP2) cloning vector containingVP1 and VP2 genes of CAV isolate SMSC-1 (accessionno. AF285882) using the specific primers listed in Table 1.The VP1 gene was inserted into the EcoRI and PstI sites ofthe recombinant plasmids, pETacmA1 and pETacmA2 inframe with the acmA gene to constructed pETVP1/acmA1and pETVP1/acmA2 encoding the fusion proteins VP1/AcmA1 and VP1/AcmA2 (Fig. 1b). To construct pETVP1/acmA1-VP2 and pETVP1/acmA2-VP2, the VP2 gene wascloned into the NdeI and XhoI sites of the MSC2 ofpETVP1/acmA1 and pETVP1/acmA2, respectively. Aftertransformation into E. coli and verification of the recombi-nant plasmids by restriction enzyme analysis and PCR, thesequence of the inserted genes was confirmed by double-stranded sequencing.

Protein expression in E. coli BL21 (DE3)

The protein expression of the genes was evaluated in therecombinant E. coli BL21 (DE3) containing the DNAplasmid constructs. Briefly, the cells were grown in LBbroth supplemented with 50 μg/ml ampicillin until theculture reached OD600 0.5–0.7. After induction by IPTG ata final concentration of 1 mM, the cells were harvested bycentrifugation at 2,000×g for 10 min. The cell pellet wasresuspended in 2× sample buffer (1.25 ml Tris–HCl,pH 6.8, 20% glycerol, 4% sodium dodecyl sulfate (SDS),0.02% (w/v) bromophenol blue, 10% 2-mercaptoethanol)prior to boiling at 95°C for 5 min. Insoluble materials wereremoved by centrifugation at 10,000×g for 5 min, and thesupernatant containing the proteins was used for SDS-polyacrylamide gel electrophoresis (PAGE) analysis. TheSDS-PAGE method was carried out according to theLaemmli discontinuous SDS system (Laemmli 1970) using12% polyacrylamide gel. For Western blot analysis, theseparated proteins on SDS-PAGE were transferred ontonitrocellulose membrane by semi-dry transfer cell (BioRad,USA). After blocking with 1:10 dilution of skim milk(KPL, USA) in TBS, the membrane was incubated withanti-His monoclonal antibody or chicken anti-CAV (isolateSMSC-1) serum as primary antibody for the detection ofthe fusion proteins, AcmA1/VP1 and AcmA2/VP1, ormonoclonal mouse S-tag antibody for the detection ofCAV VP2. The blot was then washed and soaked witheither secondary antibody conjugated to alkaline phospha-tase for 1 h at room temperature (RT) with gentle agitation.The membrane was further washed and the protein bandswere visualized by using developing a solution containingnitro blue tetrazolium, bromochloroindolyl phosphate inalkaline phosphatase buffer.

Purification of the fusion proteins and bindingto L. acidophilus

After synchronous expression of the VP1/AcmA1 and VP1/AcmA2 proteins with the CAV VP2, the fusion proteins

Table 1 Primers for PCR amplification of the genes

Gene Accession number Sequence (5′ to 3′) Expected product size (base pair)

CAV VP1 AF285882 Forward CTTAGGGAATTCGATGGCAAGACGAGCTCGC 1,335Reverse GATAGCCTGCAGCCAGTACATGGTGCTGTTGG

CAV VP2 AF285882 Forward CGCTAACATATGCACGGGAACGGCGGACAA 648Reverse TCATGGCTCGAGCACTATACGTACCGGGGC

acmA1 LLU17696 Forward CTTAGGCTGCAGGACGGAGCTTCTTCAGCT 249Reverse GATAGCAAGCTTTTAACCTGAATTTGTAGAAGAAG

acmA2 LLU17696 Forward CTTAGGCTGCAGGGTGGCTCGACAACCACAA 444Reverse GATAGCAAGCTTCGAAGGATTTGAAGCAGCA

Appl Microbiol Biotechnol (2011) 90:77–88 79

were purified by using the Ni-NTA purification system(Invitrogen, USA). After induction, the E. coli cells wereharvested; washed with pre-chilled dH2O; and then sus-pended in lysis buffer containing 50 mM NaH2PO4/Na2HPO4, 0.1 M NaCl, pH 7.6, 1 mM protease inhibitorphenylmethylsulfonyl fluoride and 0.2 mg/ml lysozyme,followed by 1 h of incubation at 4°C with gentleagitation. The cells were then sonicated using LabsonicU sonicator (B. Braun, Germany) at a power output of40 W for 10 cycles of 30-s bursts (10×, 30 s), withinterval cooling on ice. The crude lysate was centri-fuged at 20,000×g at 4°C for 30 min and the clearsupernatant containing soluble proteins was applied ontoa Ni2+-affinity column (Invitrogen, USA) to purify theproteins. The concentration of the purified proteins wasevaluated based on the Bradford method.

The cells in 5 ml of overnight culture of L. acidophilus(ATCC 53672) in MRS broth supplemented with 0.5%glucose (GMRS) (OD600 of 0.5–0.7) were harvested at2,000×g for 5 min; resuspended in 600 μl of fresh GMRSbroth; and then mixed with 200 μl of the purified proteins(5 mg/ml) followed by 2 h of incubation at 30°C. The cellswere precipitated at 2,000×g for 10 min and then washedwith 1X phosphate-buffered saline (PBS) three times. Thebinding was analyzed by ELISA and immunofluorescencemicroscopy.

To determine the binding stability of the fusion proteins,the cells formerly incubated with the fusion proteins wereharvested, and the cell pellets were resuspended in 1.2 mlof 1× PBS. The cell suspensions were stored at 4°C for5 days and after each 24-h interval, 200 μl of the sampleswere tested for the presence of the fusion proteins on thecell wall surface by ELISA.

ELISA and immunofluorescence microscopy

After binding procedure, cells were fixed with 4% (w/v)paraformaldehyde for 20 min at RT. The fixed cells werewashed with 1x PBS (×3); and then incubated with theblocking solution [3% (w/v) bovine serum albumin (BSA)in 1x PBS] for 30 min at RT followed by further washingwith 1x PBS (×3). The cells were then incubated with500 μl of primary chicken anti-CAV serum at a ratio of1:200 μl in 1% BSA (w/v) for 60 min at RT. The cells werewashed with 1x PBS (×3) and then incubated with 500 μlof horseradish peroxidase conjugated anti-chicken antibodydiluted at 1:500 in 1% BSA for further 60 min. Followingthree times washing, the cells were resuspended in 200 μlof 1x PBS. Finally, 10 μl of the suspension and 50 μl ofsubstrate (BM Blue, Roche) was mixed in the wells of theELISA plate and then incubated at RT for 20 min. Thereaction was stopped by adding 50 μl of stop solution (1 MH2SO4) and the absorbance was measured at 490 nm.

Lactobacilli cells incubated with BSA or 1x PBS were usedas control.

For immunofluorescence microscopy, the cells mixedwith the fusion proteins were placed on chamber slidesprecoated with poly-L-lysine and air-dried before incuba-tion with 4% (w/v) paraformaldehyde for 20 min at RT. Thefixed cells were washed (×3) with 1x PBS; and thenincubated with 3% (w/v) BSA in 1x PBS for 30 min at RTto block non-specific binding. The cells were thenincubated with chicken anti-CAV serum diluted at a ratioof 1:200 in 1% BSA, followed by incubation at RT for60 min. After washing three times with 1x PBS, the cellswere incubated with FITC anti-chicken IgY diluted at 1:200in 1% BSA at RT for 1 h. Following washing with 1x PBS(×3), the slide was air-dried and then analyzed by aninverted phase contrast microscope with fluorescence light.

Oral immunization of SPF chickens

The 3-week-old specific pathogen-free (SPF) chickens weredivided into four groups (n=10). Groups 1 and 2 wereorally fed with 109 cells of L. acidophilus carrying VP1/AcmA1 or VP1/AcmA2 protein, respectively. As control,chickens in groups 3 and 4 were fed with 109 cells ofnormal L. acidophilus not incubated with the fusionproteins or 1 ml of 1x PBS (pH 7.4), respectively. Thefeeding strategy was obtained (with some modification)from the report by Dieye et al. (2003) in which L. lactis wasused as live vehicle for mucosal delivery of infectiousbursal disease virus antigens in chickens. At the age of 21,the chickens were fed for five consecutive days. At the ageof 35 days, chickens again were fed for another fiveconsecutive days. Blood samples were taken from thechickens when they were 21, 35, and 46 days old. Thecollected sera were stored at −20°C for further analysis. Allprocedures were conducted with the protocols approved bythe Animal Care and Use Committee of the Faculty ofVeterinary Medicine, Universiti Putra Malaysia.

Antibody response after oral immunization

Serum antibody titer against CAV was determined using theIDEXX ELISA kit (IDEXX, Portland, ME, USA) at days 21(before the first feeding), 35 (2 weeks after the first feeding)and 46 (1 week after the second feeding). ELISA wasperformed according to the procedures recommended bythe company. Antibody titers higher than 1,000 wereconsidered positive.

Virus neutralization test

To determine the neutralization antibody titers in theimmunized chickens, a virus neutralization (VN) test was

80 Appl Microbiol Biotechnol (2011) 90:77–88

carried out in triplets in 96-well microplates by themeasurement of cell proliferation and viability using thecell proliferation reagent WST-1 (Roche, Germany) accord-ing to the method described by Lehtoranta et al. (2009)with some modifications. Anti-CAV serum was used aspositive control. Briefly, twofold serial dilutions of heat-inactivated (56°C for 30 min) sera were prepared in RPMImedium to achieve a dilution range from 1:4 to 1:2,048.The diluted sera (50 μl) were transferred into flat bottom96-well microplates and then mixed with an equal volumeof 105 TCID50 of CAV SMSC-1 strain. After 60 min ofincubation at 39°C, 200 μl MSB-1 (Marek’s disease spleenfrom chickens infected with the BC-1 strain of MDV) cellswere added to eachwell at a concentration of 1.5×105 cells permilliliter followed by further incubation at 39°C and 5% CO2

for 5 days. The RPMI medium with virus (non-neutralizedvirus) and or with positive anti-CAV (SMSC-1) serum wasserved as negative and positive controls, respectively. At5 days of incubation, 10 μl of WST-1 reagent (Roche,Germany) was added to each well, and the plates were furtherincubated at 37°C for 4 h in a humidified atmosphere to letthe color reaction develop. Subsequently, the plates werecentrifuged at 1,000×g for 10 min to remove the cells. Thesupernatants were transferred into new microplates and theabsorbance of the samples against a background control asblank was measured at 450 nm wavelength. The meanabsorbance value from the triplicates was measured for eachserum dilution and the controls. Neutralizing activity wascalculated as described by Lehtoranta et al. (2009).

Splenocyte proliferation assay in the immunized chickens

At day 10 post-immunization, the chickens were sacrificed,their spleens were harvested, and splenocytes were pre-pared in PBS–EDTA solution (1x PBS, 2 mg/ml EDTA)supplemented with 2% penicillin/streptomycin. Red bloodcells were removed using red blood cell lysis solution(0.84% NH4CL, 0.1% NaHCO3, and 1.8 ml of 5% EDTA).Splenocyte suspension was prepared in DMEM mediumsupplemented with 10% fetal bovine serum and 1%penicillin/streptomycin, and viable cells were counted bytrypan blue exclusion on a hemocytometer. The cells(0.1 ml at 2×104 cells/well) were placed into each well of96-well plates and then incubated in triplicate with CAVVP1 antigen (3 μg), mitogen phytohemagglutinin (PHA1 mg/ml, Sigma), or medium alone as negative control for3 days at 37°C in an atmosphere of 5% CO2. Theproliferation of splenocytes was evaluated by the BrdUcell proliferation assay kit (Exalpha Biologicals, USA).Data was reported as stimulation indices (SI), which wasthe mean of experimental wells/mean of antigen free wells(negative control). SI greater than two was consideredconsistent with an immunized response.

Th1 cytokines assay

The serum levels of Th1 cytokines interleukin (IL)-2, IL-12, and interferon (IFN)-γ were evaluated by commerciallyavailable chicken ELISA kits (Cusabio Biotech, USA). Theassay was carried out in 96-wells microplates according tothe procedures recommended by the manufacturer.

Statistical analysis

The data was analyzed by t test and statistical significancewas set at P<0.05. The results were expressed as means±standard error of mean.

Results

Construction of the recombinant plasmids

Two recombinant plasmids, namely pETVP1/acmA1-VP2and pETVP1/acmA2-VP2 were constructed. Figure 1cshows the map of the constructs. After transformation intoE. coli, the constructs were verified by restriction enzymeanalysis and PCR (results not shown). The sequence of thegenes was confirmed by sequencing (data not shown).

Expression studies

From the crude protein extracts of both pETVP1/acmA1-VP2 and pETVP1/acmA2-VP2-transformed E. coli, theexpression of VP2 gene was detected by Western blottingusing monoclonal antibody to S-tag polypeptide. The VP2protein was visualized as a single protein band with amolecular weight of about 27 kDa (Fig. 2a, b).

Western blot analysis also confirmed the expression ofthe fusion proteins, VP1/AcmA1 (∼62 kDa) and VP1/AcmA2 (∼69 kDa) using anti-His monoclonal antibody asprimary antibody (Fig. 3a, b, respectively). To explorewhether the VP1 protein linked to the AcmA proteins iscapable to recognize anti-CAV antibodies, when expressedsynchronously with VP2, the expression of the fusionproteins were studied using the anti-CAV serum. As shownin Fig. 3c, d, anti-CAV antibodies recognized the VP1protein linked to both AcmA1 and AcmA2 proteins whichwas co-expressed with the VP2 protein.

Binding of the purified fusion proteins on the cell wallsurface of L. acidophilus

VP1/AcmA1 and VP1/AcmA2 were purified on Ni2+

affinity columns and mixed with a culture of L. acidophilus(ATCC 53672) as described in “Material and methods.”Whole cell ELISA and immunofluorescence microscopy

Appl Microbiol Biotechnol (2011) 90:77–88 81

verified the binding of the fusion proteins on the cell wallsurface.

After immunofluorescence staining, lactobacilli cellsincubated with VP1/AcmA1 or VP1/AcmA2, bothexhibited bright fluorescence on the cell surface (Fig. 4a, b),indicating the presence of the fusion proteins on the cell wallsurface. The control bacterial cells showed no fluorescence(Fig. 4c).

According to the ELISA results summarized in Fig. 5,the cells incubated with VP1/AcmA1 or VP1/AcmA2, bothshowed high absorption values when compared with thecontrols. However, the absorbance value was significantly

(P<0.05) higher in the cells incubated with VP1/AcmA2compared to those from the cells incubated with VP1/AcmA1.

Stability of the fusion proteins on the surface of L.acidophilus

The binding stability of the fusion proteins on the surface ofL. acidophilus cells was tested for the duration of 5 days.As shown in Fig. 6, the presence of the fusion proteins onthe cell wall surface was detected even after a period of5 days. No significant differences were observed in theabsorbance values in this duration.

Fig. 3 Western blot analysis ofthe fusion proteins, VP1/AcmA1 (∼62 kDa) and VP1/AcmA2 (∼69 kDa) using a, banti-His monoclonal antibodyand c, d anti-CAV serum. a, bM, protein ladder, 10–200 kDa(Fermentas, Canada); lanes 1, aVP1/AcmA1 or b VP1/AcmA2;lane 3, negative control. c, d M,protein ladder; lanes 1 and 2, cVP1/AcmA1 or d VP1/AcmA2;lane 3, negative control

Fig. 2 Expression analysis ofthe VP2 protein. Expression ofthe VP2 protein in the E.coli-VP1/acmA1-VP2 (a)and -VP1/acmA2-VP2 (b) cellswas tested by Western blottingusing anti-S-tag monoclonalantibody. M, protein ladder,10–200 kDa (Fermentas,Canada); lanes 1 and 2, VP2protein; lane 3, negative control

82 Appl Microbiol Biotechnol (2011) 90:77–88

Antibody response to oral immunization

In order to study whether a humoral immune response hasbeen induced by oral immunization, serum antibody titeragainst CAV was tested using the IDEXX ELISA kit at

days 21 (before the first feeding), 35 (2 weeks after the firstfeeding), and 46 (1 week after the second feeding). Theantibody titers are summarized in Table 2. Although VP1/AcmA1- and VP1/AcmA2-vaccinated groups had nopositive antibody titers to CAV at day 14 after the firstfeeding, both groups showed positive antibody responsewith titer 1,439±62 and 1,580±77, respectively 7 days afterthe boosted feeding. The difference in antibody titerbetween VP1/AcmA1 and VP1/AcmA2 groups was not

Fig. 6 Stability analysis of the binding of the fusion proteins on thecell surface of L. acidophilus. The cells were incubated with the fusionprotein and then binding analysis was carried out by ELISA for theduration of 5 days. 0, initial value; 1–5, the days of assay

Fig. 5 Whole cell ELISA analysis. L. acidophilus cells incubatedwith the fusion proteins, VP1/AcmA1 or VP1/AcmA2 were subjectedto ELISA using chicken anti-CAV serum as primary and HRPconjugated anti-chicken IgY as secondary antibody. Controls-1 and -2 were the lactobacilli cells incubated with 3% BSA or 1X PBS,respectively

Fig. 4 Immunofluorescence micrographs of the binding of the fusionproteins to L. acidophilus. a, b L. acidophilus cells carrying VP1/AcmA1 and VP1/AcmA2, respectively. c Control. The cells wereobserved under ×100 objective. The arrows indicated the bacterialcells

Appl Microbiol Biotechnol (2011) 90:77–88 83

statistically significant. No positive antibody response wasdetected in the control groups.

Virus neutralizing antibody titers

The neutralization activity of the serum antibodies againstCAV infection in MSB-1 cells was studied in theimmunized groups. According to the results summarizedin Table 2, the tested serum samples from both VP1/AcmA1- and VP1/AcmA2-fed groups showed positiveneutralization activities against CAV with the titer of1:128 and 1:128–1:256, respectively after the boostedadministration. No neutralization activity was detected inthe controls fed by the normal cells or PBS.

Serum level of Th1 cytokines in the immunized chickens

The serum level of IL-2, IL-12, and IFN-γ were analyzedbefore and 21 days post-immunization (7 days after theboost) by ELISA. The VP1/AcmA1- and VP1/AcmA2-immunized groups, both induced considerable levels of allthree Th1 cytokines IL-2, IL-12, and IFN-γ compared tothose from the controls (Fig. 7). No significant differenceswere observed between the VP1/AcmA1 and VP1/AcmA2groups in the serum levels of the cytokines.

Antigen-stimulated proliferation response of splenocytes

Splenocyte proliferation assay was used to monitor T cellresponses after oral immunization. At day 10 post-immunization, the spleen of the chickens was removedand the splenocytes were prepared for the assay. The cellswere stimulated with the purified VP1 protein as antigenicprotein from CAV. PHA was used as positive control forstimulation. As shown in Fig. 8, a higher level of VP1induction proliferative response (p<0.05) was observed inthe VP1/AcmA1- and VP1/AcmA2-fed groups with the

mean SI of 4.97 and 5.61 respectively, when compared tothe control groups administrated with PBS (mean SI, 1.63)or the normal bacterial cells not carrying the fusion proteins(mean SI, 1.42). As expected, no significant difference wasobserved at SI between the groups when PHA was used asstimulator. These results showed the VP1-stimulated pro-liferative response of splenocytes in the chickens immu-nized with L. acidophilus cells carrying VP1/AcmA1 orVP1/AcmA2.

Discussion

In the present study, L. acidophilus (ATCC 53672) wasused as live delivery vehicle for oral immunization againstCAV because lactobacilli have the capacity to attach andcolonize at the certain regions of the intestine that may playan important role in stimulation of both non-specific andspecific immune responses against the antigens (Blomberget al. 1993; Vaughan et al. 1999).

Although lactobacilli seem potentially better candidatesfor oral immunization purposes, the number of positiveresults reported with Lactobacillus strains is lower incomparison to those from the lactococci (Scheppler et al.2002; Seegers 2002). This may be due to lower levels ofantigen expression in the lactobacilli. In addition, cell wall-anchored antigens have been shown to be more immuno-genic compared to the cytoplasmic or secreted antigens(Wells et al. 1993; Reveneau et al. 2002). For these reasons,the E. coli expression-based system adapted from Raha etal. (2005) was chosen to be used to display high levels ofthe antigen-anchor fusion proteins on the cell wall surfaceof L. acidophilus. In this system, target proteins can beproduced as fusion proteins with an anchor motif in E. coli.These antigen-anchor motif fusions are then able to bindnon-covalently onto the cell wall surface of a wide range ofGram-positive bacteria by simply mixing with the live cells.

Table 2 ELISA antibody titer and virus neutralizing titer after oral immunization

Vaccinated groups Serum antibody assays

Before immunization At day 14 after the firstimmunization

At day 7 after the second immunization

Ab titera

(mean±SD)ELISA result VN titer Ab titer

(mean±SD)ELISA result Ab titer

(mean±SD)ELISA result VN titer

VP1/AcmA2 <600 – – <1,000 – 1,580±77 + 1:128–1:252

VP1/AcmA1 <600 – – <1,000 – 1,439±62 + 1:128

Normal cells <600 – – <600 – <600 – –

PBS <600 – – <600 – <600 – –

a Antibody titers higher than 1,000 were considered positive

84 Appl Microbiol Biotechnol (2011) 90:77–88

The result is a live delivery vaccine that mimics pathogenswith the antigenic proteins on its surface.

The cell wall binding domains of the lactococcal AcmAhas been shown to bind to the cell wall of a range of Gram-positive bacteria, including Lactobacillus, Clostridium,Listeria, and Bacillus, when mixed to live cells (Buist1997; Raha et al. 2005; Steen et al. 2005). Therefore, in thepresent study, the cell binding domains of AcmA from L.lactis MG1363 were used as anchoring proteins to displaythe VP1 antigenic protein of CAVon the cell wall surface ofL. acidophilus. Briefly, one or two repeats of the cell wallbinding domain of AcmA were amplified and inserted intothe pETDuetTM-1 to construct pETacmA1 and pETacmA2.The VP1 gene was then fused to the acmA genes toconstruct pETVP1/acmA1 and pETVP1/acmA2 encodingthe fusion proteins, VP1/AcmA1 and VP1/AcmA2.According to the previous studies (Koch et al. 1995;Noteborn et al. 1998; Lacorte et al. 2007), synchronousexpression of the VP1 and VP2 proteins is required for theinduction of neutralizing antibodies against CAV. There-fore, the VP2 gene of CAV was cloned into the second

Fig. 8 Splenocyte proliferation response after oral immunization.Splenocytes from the immunized chickens were prepared andsubjected to a proliferation assay using the BrdU cell proliferationassay kit. Chickens administrated with PBS or normal bacterial cellsnot carrying the fusion proteins were used as negative controls.Mitogen phytohemagglutinin was used as a positive control for cellproliferation. Data was reported as SI. SI greater than two wasconsidered positive

Fig. 7 Serum level of IL-2, IL-12, and IFN-γ in the vaccinatedgroups. The serum levels of Th1 cytokines, IL-2 (a) IL-12 (b), andIFN-γ (c) were determined by ELISA. Chickens fed with PBS ornormal cells not carrying the fusion proteins were used as controls.The results showed an increased level of all three cytokines in theVP1/AcmA1 and VP1/AcmA2-fed chickens compared with thecontrols

R

Appl Microbiol Biotechnol (2011) 90:77–88 85

MCS of the recombinant plasmids, pETVP1/acmA1 andpETVP1/acmA2 to construct pETVP1/acmA1-VP2 andpETVP1/acmA2-VP2, respectively.

SDS-PAGE and Western blot analysis showed thesuccessful expression of the binding domains of acmAgene, VP2 and the fusion proteins, VP1/AcmA1 and VP1/AcmA2 in E. coli. Immunoblotting also showed theinteraction of anti-CAV polyclonal antibodies with theVP1 protein in both fusion proteins indicating the properstructural folding of the VP1 protein to induce immunesystem against CAV.

L. acidophilus (ATCC 53672) isolated from chickenintestine was chosen to be used as live delivery vehicle fororal immunization in SPF chickens because mucosalcolonization of lactobacilli has been proven to be hostspecific (Tannock et al. 1982; Lin and Savage 1984).

The purified VP1/AcmA1 and VP1/AcmA2 were mixedwith a culture of L. acidophilus followed by a binding studyby using immunofluorescence staining and whole cellELISA. Immunofluorescence staining using chicken anti-CAV serum revealed the presence of the VP1 protein on thecell surface of both bacterial groups, L. acidophilus cellsincubated with VP1/AcmA1 or VP1/AcmA2. Consistentwith the immunofluorescence staining results, whole cellELISA using anti-CAV serum verified the binding of thefusion proteins on the cell wall surface of L. acidophilus.However, the ELISA absorbance value was significantly(P<0.05) higher in the VP1/AcmA2 group compared to thosefrom the VP1/AcmA1 group suggesting that the bindingcapacity can be correlated with the number of the repeatdomains. This view is supported by the findings of Steen et al.(2005) and Okano et al. (2008) that with an increase in thenumber of the binding domains of AcmA, the bindingcapacity of the fusion proteins on the cell surface increases.

A stability study was carried out for a period of 5 daysand the result showed that the fusion proteins can bemaintained on the surface of the cells in this duration. Itwas in agreement with the previous study by Raha et al.(2005) that the binding of the fusion proteins containing thecell wall binding domains of AcmA on the surface ofGram-positive bacteria such as Lactococcus, Lactobacillus,and Bacillus are stable for the duration of several days,when added from the outside.

The lactobacilli cells carrying the VP1 protein of CAVwere used to immunize 3-week-old SPF chickens throughoral route. At the age of 21 days, chickens were fed for fiveconsecutive days and boosted with the same feedingprotocol 2 weeks later. Serum antibody titer to CAV wasanalyzed by ELISA at days 0, 14, and 21 following the firstadministration. According to the ELISA results, both VP1/AcmA1- and VP1/AcmA2-vaccinated groups showed posi-tive antibody titer to CAV without a significant difference atday 7 following the boosted administration.

It has been indicated that neutralizing antibodies canprovide complete protection to CAV infection (Yuasa etal. 1980; Rosenberger and Cloud 1998; Saif 2003; Schat2009). To determine the VN activity of the antibodiesproduced after oral immunization, the collected sera wereanalyzed for neutralization antibody titer against CAVinfection in MSB-1 cells. The serum samples from theVP1/AcmA2-fed chickens showed positive neutralizationtiters ranging from 1:125 to 1:256, while the VP1/AcmA1-fed group had a VN titer of 1:125. Accordingto Otaki et al. (1992), young chicks derived fromimmune hens against CAV with VN antibody titers aslow as 1:40 can survive from the virus challenge.However, Malo and Weingarten (1995) indicated thatneutralization antibody titers of at least 1:256 arenecessary to prevent virus shedding in the feces andvertical transmission (Malo and Weingarten 1995).Because of the development of age resistance to CAVduring the first week of life and becoming completewithin 3 weeks or even earlier (Yuasa et al. 1980;Rosenberger and Cloud 1989; Saif 2003), the protectionability of the neutralizing antibodies produced after oralimmunization was not possible to be tested. On the otherhand, chickens positive for anti-CAV maternal antibodieshave been shown to be protected from the disease (Yuasaet al. 1980; Vielitz and Landgraf 1988; Saif 2003).Therefore, further study is recommended in the vaccina-tion of breeder producing day old chicks with maternalantibodies followed by a virus challenge study.

Cell-mediated immune response was also detectedafter oral immunization. Positive VP1-stimulated prolif-erative response was observed in the splenocytes of thechickens immunized with the lactobacilli cells carryingVP1/AcmA1 or VP1/AcmA2 indicating the activation ofantigen-specific immune responses in the immunizedchickens. Splenocyte proliferation reaction has beenshown that is directly proportional to the cell-mediatedimmune response (Han et al. 2000; Kim et al. 2004;Chin’ombe et al. 2009; Feng et al. 2009). Furthermore,administration of the VP1/AcmA1- or VP1/AcmA2-carrying lactobacilli cells were able to induce increasesin the serum levels of Th1 cytokines IL-2, IL-12, andIFN-γ in the immunized chickens after 7 days post-immunization indicating that the cells carrying the VP1protein of CAV perhaps triggered Th1 immune responseswhich are known to be involved in cellular-mediatedimmunity (Goldsby et al. 2003).

In conclusion, consistent with the previous studies, theresults of the present study showed that the bindingdomains of AcmA protein are capable of binding to thecell wall surface of lactobacilli when added from theoutside. Furthermore, our data showed that the L.acidophilus cells carrying the VP1 protein of CAV

86 Appl Microbiol Biotechnol (2011) 90:77–88

induced immune response against CAV suggesting the useof lactobacilli as live delivery vehicle for oral immuniza-tion in chicken.

Acknowledgments We would like to thank Ms. Razieh Monjezi forher assistance in the oral immunization. Hassan Moeini is sponsoredunder the Graduate Research Fellowship, Universiti Putra Malaysia.

References

Blomberg L, Henriksson A, Conway P (1993) Inhibition of adhesionof Escherichia coli K88 to piglet ileal mucus by Lactobacillusspp. Appl Environ Microbiol 59(1):34–39

Buist G (1997) AcmA of Lactococcus lactis, a cell-binding majorautolysin. University of Groningen, Netherlands

Buist G, Kok J, Leenhouts K, DabrowskaM, Venema G, Haandrikman A(1995) Molecular cloning and nucleotide sequence of the geneencoding the major peptidoglycan hydrolase of Lactococcus lactis,a muramidase needed for cell separation. J Bacteriol 177(6):1554–1563

Chin’ombe N, Bourn W, Williamson A, Shephard E (2009) Oralvaccination with a recombinant Salmonella vaccine vectorprovokes systemic HIV-1 subtype C Gag-specific CD4+ Th1and Th2 cell immune responses in mice. Virol J 6(87):1–9

Chowdhury S, Omar A, Aini I, Hair-Bejo M, Jamaluddin A, Md-ZainB, Kono Y (2003) Pathogenicity, sequence and phylogeneticanalysis of Malaysian chicken anaemia virus obtained after lowand high passages in MSB-1 cells. Arch Virol 148(12):2437–2448

Dertzbaugh M (1998) Genetically engineered vaccines: an overview.Plasmid 39(2):100–113

Dieye Y, Hoekman A, Clier F, Juillard V, Boot H, Piard J (2003)Ability of Lactococcus lactis to export viral capsid antigens: acrucial step for development of live vaccines. Appl EnvironMicrobiol 69(12):7281–7288

Feng C, Li Q, Zhang X, Dong K, Hu B, Guo X (2009) Immunestrategies using single-component LipL32 and multi-componentrecombinant LipL32-41-OmpL1 vaccines against leptospira.Braz J Med Biol Res 42(9):796–803

Goldsby RA, Kindt TJ, Osborne BA, Kuby J (2003) Cytokines. In:Freeman WH (ed) Immunology, 5th edn. New York, NY

Han R, Cladel N, Reed C, Peng X, Budgeon L, Pickel M, ChristensenN (2000) DNA vaccination prevents and/or delays carcinomadevelopment of papillomavirus-induced skin papillomas onrabbits. J Virol 74(20):9712

Kim S, Sung H, Han J, Jackwood D, Kwon H (2004) Protectionagainst very virulent infectious bursal disease virus in chickensimmunized with DNA vaccines. Vet Microbiol 101(1):39–51

Koch G, van Roozelaar D, Verschueren C, van der Eb A, Noteborn M(1995) Immunogenic and protective properties of chickenanaemia virus proteins expressed by baculovirus. Vaccine 13(8):763–770

Lacorte C, Lohuis H, Goldbach R, Prins M (2007) Assessing theexpression of chicken anemia virus proteins in plants. Virus Res129(2):80–86

Laemmli U (1970) Cleavage of structural proteins during the assemblyof the head of bacteriophage T4. Nature 227(5259):680–685

Lee M, Roussel Y, Wilks M, Tabaqchali S (2001) Expression ofHelicobacter pylori urease subunit B gene in Lactococcus lactisMG1363 and its use as a vaccine delivery system against H.pylori infection in mice. Vaccine 19(28–29):3927–3935

Lehtoranta L, Villberg A, Santanen R, Ziegler T (2009) A novel,colorimetric neutralization assay for measuring antibodies toinfluenza viruses. J Virol Meth 159(2):271–276

Lin J, Savage D (1984) Host specificity of the colonization of murinegastric epithelium by lactobacilli. FEMS Microbiol Lett 24(1):67–71

Maggi T, Oggioni M, Medaglini D, Bianchi Bandinelli M, SoldateschiD, Wiesmüller K, Muller C, Valensin P, Pozzi G (2000)Expression of measles virus antigens in Streptococcus gordonii.New Microbiol 23(2):119–128

Malo A, Weingarten M (1995) Determination of the minimumprotective neutralizing antibody titer to CAV in adult chickens.Intervet VSD Newsletter 11:1–5

Noteborn M, Verschueren C, Koch G, Van der Eb A (1998)Simultaneous expression of recombinant baculovirus-encodedchicken anaemia virus (CAV) proteins VP1 and VP2 is requiredfor formation of the CAV-specific neutralizing epitope. J GenVirol 79(12):3073

Okano K, Zhang Q, Kimura S, Narita J, Tanaka T, Fukuda H, Kondo A(2008) System using tandem repeats of the cA peptidoglycan-binding domain from Lactococcus lactis for display of both N- andC-terminal fusions on cell surfaces of Lactic Acid Bacteria. ApplEnviron Microbiol 74(4):1117–1123

Otaki Y, Saito K, Tajima M, Momura Y (1992) Persistence ofmaternal antibody to chicken anemia agent and its effect on thesusceptibility of young chicken. Avian Pathol 21(1):147–151

Raha A, Varma N, Yusoff K, Ross E, Foo H (2005) Cell surfacedisplay system for Lactococcus lactis: a novel development fororal vaccine. Appl Microbiol Biotechnol 68(1):75–81

Reveneau N, Geoffroy M, Locht C, Chagnaud P, Mercenier A (2002)Comparison of the immune responses induced by local immuni-zations with recombinant Lactobacillus plantarum producingtetanus toxin fragment C in different cellular locations. Vaccine20(13–14):1769–1777

Ribeiro L, Azevedo V, Loir Y, Oliveira S, Dieye Y, Piard J, Gruss A,Langella P (2002) Production and targeting of the Brucellaabortus antigen L7/L12 in Lactococcus lactis: a first step towardsfood-grade live vaccines against brucellosis. Appl EnvironMicrobiol 68(2):910–916

Rosenberger J, Cloud S (1989) The isolation and characterization ofchicken anemia agent (CAA) from broilers in the United States.Avian Dis 33(4):707–713

Rosenberger J, Cloud S (1998) Chicken anemia virus. Poult Sci 77(8):1190

Saif YM (2003) Diseases of poultry, 11th edn. Iowa State UniversityPress, Ames, IA USA

Sambrook J, Russell D (2001) Molecular cloning: a laboratorymanual. CSHL press, New York

Schat K (2009) Chicken anemia virus. Curr Top Microbiol Immunol331:151–183

Scheppler L, Vogel M, Zuercher A, Zuercher M, Germond J, MiescherS, Stadler B (2002) Recombinant Lactobacillus johnsonii as amucosal vaccine delivery vehicle. Vaccine 20(23–24):2913–2920

Seegers J (2002) Lactobacilli as live vaccine delivery vectors: progressand prospects. Trends Biotechnol 20(12):508–515

Steen A, Buist G, Horsburgh G, Venema G, Kuipers O, Foste S, Kok J(2005) AcmA of Lactococcus lactis is an N-acetylglucosaminidasewith an optimal number of LysM domains for proper functioning.FEBS J 272(11):2854–2868

Stover C, Bansal G, Hanson M, Burlein J, Palaszynski S, Young J,Koenig S, Young D, Sadziene A, Barbour A (1993) Protectiveimmunity elicited by recombinant bacilli Calmette-Guerin (BCG)expressing outer surface protein A (OspA) lipoprotein: acandidate Lyme disease vaccine. J Exp Med 178(1):197–209

Tannock G, Szylit O, Duval Y, Raibaud P (1982) Colonisation oftissue surfaces in the gastrointestinal tract of gnotobiotic

Appl Microbiol Biotechnol (2011) 90:77–88 87

animals by Lactobacillus strains. Can J Microbiol 28(10):1196–1198

Vaughan E, Molletb B, deVos W (1999) Functionality of probioticsand intestinal lactobacilli: light in the intestinal tract tunnel. CurrOpin Biotechnol 10(5):505–510

Vielitz E, Landgraf H (1988) Anaemia-dermatitis of broilers: fieldobservations on its occurrence, transmission and prevention.Avian Pathol 17(1):113–120

Wells J, Wilson P, Norton P, Gasson M, Le Page R (1993)Lactococcus lactis: high-level expression of tetanus toxin

fragment C and protection against lethal challenge. Mol Micro-biol 8(6):1155–1162

Yuasa N, Noguchi T, Furuta K, Yoshida I (1980) Maternal antibodyand its effect on the susceptibility of chicks to chicken anemiaagent. Avian Dis 24(1):197–201

Zegers N, Kluter E, van der Stap H, van Dura E, van Dalen P, ShawM, Baillie L (1999) Expression of the protective antigen ofBacillus anthracis by Lactobacillus casei: towards the develop-ment of an oral vaccine against anthrax. J Appl Microbiol 87(2):309–314

88 Appl Microbiol Biotechnol (2011) 90:77–88