jou rn al hom ep age: www.elsev ier .com/ locat e/vacc ine
eneration of a safety enhanced Salmonella Gallinarum ghost usingntibiotic resistance free plasmid and its potential as an effectivenactivated vaccine candidate against fowl typhoid
hetan V. Jawale, Atul A. Chaudhari, John Hwa Lee ∗
ollege of Veterinary Medicine, Chonbuk National University, Jeonju 561-756, Republic of Korea
r t i c l e i n f o
rticle history:eceived 21 September 2013eceived in revised form0 December 2013ccepted 18 December 2013vailable online 7 January 2014
A safety enhanced Salmonella Gallinarum (SG) ghost was constructed using an antibiotic resistance genefree plasmid and evaluated its potential as fowl typhoid (FT) vaccine candidate. The antibiotic resistancefree pYA3342 plasmid possesses aspartate semialdehyde dehydrogenase gene which is complimentaryto the deletion of the chromosomal asd gene in the bacterial host. This plasmid was incorporated witha ghost cassette containing the bacteriophage PhiX174 lysis gene E, designated as pJHL101. The plasmidpJHL101 was transformed into a two virulence genes-deleted SG. The SG ghosts with tunnel formation andloss of cytoplasmic contents were observed by scanning electron microscopy and transmission electronmicroscopy. The cell viability of the culture solution was decreased to 0% at 24 h after the induction ofgene E expression by an increase in temperature from 37 ◦C to 42 ◦C. The safety and protective efficacyof the SG ghost vaccine was further examined in chickens which were divided into three groups: groupA (non-immunized control), group B (orally immunized), and group C (intramuscularly immunized).The birds were immunized at 7 d of age. No clinical symptoms associated with FT such as anorexia,depression and greenish diarrhea were observed in the immunized chickens. Upon challenge with avirulent SG strain at 3 week post-immunization, the chickens immunized with the SG ghost via variousroutes were efficiently protected, as shown by significantly lower mortality and post-mortem lesions
in comparison with control group. In addition, all the immunized chickens showed significantly higherantibody responses accompanied by a potent antigen-specific lymphocyte proliferative response alongwith significantly increased numbers of CD4+ and CD8+ T lymphocytes. Overall, our results provide apromising approach of generating SG ghosts using the antibiotic resistance free plasmid in order toprepare a non-living bacterial vaccine candidate which could be environmentally safe yet efficient to prevent FT in chickens.
. Introduction
Fowl typhoid (FT), a septicemic disease of poultry, causes acuteortality and induces severe inflammation of internal organs such
s liver and spleen, caecum and yolk sac which results in significantconomic losses to the poultry industry worldwide. FT is causedy facultative intracellular Gram-negative bacterium, Salmonellanterica serovar Gallinarum (SG), which also rarely causes illness
n humans [1–3].
Prevention of FT by means of vaccination has been demon-trated earlier [4–8]. However, the commercially available live
vaccine strain has several limitations such as its only limited usefor layer breeds older than 6 weeks, low body weight in vacci-nated poultry [4], and residual virulence that can cause hepatitisand splenic lesions in chicks [9]. The live Salmonella vaccine sharesthe disadvantages of the safety concerns for the animals and causeenvironmental contamination through fecal shedding [10–12]. Theuse of inactivated or killed vaccines could be an alternative forthese live vaccines. However, conventionally produced formalinor heat inactivated vaccine formulations can alter the physio-chemical/structural properties of the antigens thereby negativelyaffecting the development of protective immunity [13,14].
Bacterial ghost (BG) technology has been a new and progres-sive approach to construct the safe and immunogenic inactivatedvaccines against wide variety of infectious diseases [14–18]. BGs
are formed by protein E mediated lysis followed by expulsion ofcytoplasmic contents [15]. The most sensitive and fragile bacterialsurface structures are highly preserved following the ghost forma-tion and thus capable of inducing cellular and humoral immune
esponses [13,19–23]. In the recent few years we have shown thatse of bacterial ghosts is an effective measure to prevent salmonel-
osis in chickens [17,18].Despite its significance in disease prevention, BG technology still
emains debatable as most of the plasmids used for the develop-ent of ghost vaccines include an antibiotic resistance gene used
or the selection and maintenance in the host cell [14,17,18,24].he continued use of such plasmids is undesirable in the vaccineroduction due to the risk of spreading antibiotic resistance traitso environmental microbes [25].
In our previous study, we generated the SG using the geneticepression/expression system containing repressor cI857 and lysisene E; however the lysis plasmid contained the ampicillin resis-ance gene as a selectable marker and virulent SG wild-type strainas used as host bacterium [17]. Although we could observe the
omplete lysis in the ghost preparation of the previously publishedeport [17], but we cannot rule out the possibility of fewer cellopulations which may remain partially lysed or live or may leado the formation of the E-lysis resistant mutant. Although there ishe least possibility, but in the event that a ghost strain contain-ng an antibiotic-resistant gene bearing plasmid is not efficientlyysed and still used as a vaccine candidate, then the chicken mayecome infected with virulent SG and the external environmentight be at higher risk due to a possible fecal shedding of such
ntibiotic-resistant gene bearing strains virulent through animalody. Such strains could be a cause to generate new antibioticesistant pathogenic strains due to acquisition of antibiotic resis-ant plasmids through horizontal transfer. Although the possibilityf presence of few live SG wild-type cells in ghost preparationeading infection in immunized birds and shedding antibiotic resis-ance gene carrying strain is less likely to happen, this fact cannote overlooked. The intact plasmid DNA bearing antibiotic resis-ance genes are reported even in the completely lysed bacterialhost preparations [24,26]. In order to minimize or avoid theheses potential hazards, the non-antibiotic marker-based Asd+
aspartate-semialdehyde dehydrogenase) plasmid can be utilizedor carrying E-lysis system and the live attenuated non-pathogenicsd gene deleted SG strain can be used as host bacterium. Theelection of the Asd+ ghost plasmid containing SG ghost strainan be based on the balanced lethal host-vector system whereinhe plasmid contains asd gene that complements a host bacterialhromosomal asd gene deletion [27,28].
The reports are described in the literature where the efforts wereirected for minimization or the removal of genomic DNA contentnd antibiotic resistance gene containing plasmid DNA from thehost preparation by utilizing the host bacterial strain transformedith dual vector carrying staphylococcal nuclease A (SNA) gene and
ysis gene E [26,29], or two plasmids carrying lysis E lysis and SNAene separately [23,30]. Although this approach was successful foremoval of DNA content of ghost preparation, but still it needs theddition of antibiotics and chemical inducers in the culture [23,30].
The present study was carried out to develop a safety enhancedG ghost vaccine, in which an antibiotic resistance free plasmid con-aining the phiX174 lysis gene E was constructed and introducednto a two virulence genes-deleted SG mutant. The successful pro-uction of the SG ghost and its immune-potential was evaluated.
. Materials and methods
.1. Construction of SG ghost
Plasmid pHCE GAPDH ghost 37SDM carrying the parenthost cassette was constructed as previously described [31]. PCRmplification of the ghost cassette was performed using pHCEAPDH ghost 37SDM as a template and the primers ghost-F-XbaI
2 (2014) 1093– 1099
(5′-TCTAGAGACCAGAACACCTTGCCGATC-3′) and ghost-R-XbaI (5′-TCTAGAACATTACATCACTCCTTCCG-3′). The amplified DNA seg-ment was cloned onto the pGEM-T easy vector (Promega, Madison,WI, USA) and was designated pJHL99. To construct the antibioticgene free plasmid containing the ghost cassette, pYA3342 plasmidcontaining asd gene was used as asd-based balanced lethal hostvector system [32]. The plasmid pJHL99 was cut by the restrictionendonuclease XbaI, and a 1.2 kb DNA-fragment of the ghost cas-sette was isolated. The ghost cassette was comprised of the PhiX174lysis gene E and the lambda PR37-cI857 regulatory system. Then theantibiotic resistance gene free backbone plasmid pYA3342 was lin-earized by XbaI digestion, and the DNA fragment containing theghost cassette was inserted. The constructed antibiotic resistancegene free ghost plasmid was designated as pJHL101. For the con-struction of the safety enhanced SG ghost vaccine, the lon, cpxR andasdA16 deleted SG mutant strain JOL967 [28] was used. Transfor-mation of the plasmid pJHL101 into JOL967 cells was performed byelectroporation and transformants were selected on LB agar platesin deficiency of Diaminopimelic acid (DAP) (Sigma, St Louis, MO,USA). The resultant SG ghost strain was denoted as JOL1291. Thevector control strain was constructed by transforming JOL967 withbackbone plasmid pYA3342 (without lysis gene E).
2.2. Lysis pattern and characterization of SG ghost strain
A single colony of JOL1291 harboring pJHL101 was inoculatedinto 40 mL of LB broth. When the culture reached an optical den-sity (OD600) of 0.5–0.6, 20 mL of the culture was shifted from 37 ◦Cto 42 ◦C temperature to induce gene E-mediated lysis. To comparethe viable cell count of the culture at 37 ◦C and 42 ◦C, the remaining20 mL culture was further grown at 37 ◦C. The lysis was monitoredby performing the viable cell counts as colony forming units (CFU)at different time points (0 h, 3 h, 6 h, 18 h and 24 h). The vector con-trol strain was also grown in the same manner for comparison ofviable cell count at 37 ◦C and 42 ◦C. The morphological features ofthe SG ghost were characterized by scanning electron microscopy(SEM) and transmission electron microscopy (TEM) as previouslydescribed with minor modifications [17].
2.3. Immunization and observation of general condition inchickens after vaccination
The animal experiment described in this study was conductedwith approval (CBU 2011-0017) from the Chonbuk National Univer-sity Animal Ethics Committee in accordance with the guidelines ofthe Korean Council on Animal Care. One-day old female Brown Nickchickens were divided into three groups (n = 15 per group), and pro-vided with water and antibiotic-free food ad libitum. Chickens wereimmunized with the SG ghosts at 7-d of age. The dose and route ofthe vaccination were decided based on the data from our previousstudy [17]. In the control group A, chickens were orally inoculatedwith PBS. In group B, the birds were orally inoculated with the SGghosts at a concentration of 1 × 1010 cells/0.1 mL/chicken. GroupC birds were injected intramuscularly with 0.1 mL of the SG ghostsuspension containing 1 × 108 cells/chicken.
2.4. Collection of plasma and intestinal wash samples forantibody response assessment
For weekly determination of plasma IgG and intestinal secretory
IgA (sIgA), five birds per each group were used. The plasma sam-ples were obtained by centrifugation of the peripheral blood. Theintestinal wash samples were collected as described earlier [33].The samples were stored at −20 ◦C until use.
cine 32 (2014) 1093– 1099 1095
2
eqaMaE[
2
f1ab54si
2
icaB3ccdfl
2
wopcwas
2
uusdab
3
3
df[p
Table 1Comparative analysis of viable cell count of a SG ghost strain.
Incubation (h)a No. of CFU at incubation temperatureb
JOL1291c Vector controld
37 ◦C 42 ◦C 37 ◦C 42 ◦C
0 1.6 × 109 1.8 × 109 2.3 × 109 2.6 × 109
3 1.8 × 109 4.0 × 108 3 × 109 3.2 × 109
6 1.8 × 109 2.8 × 108 2.9 × 109 3 × 109
18 1.5 × 109 0 1.1 × 109 9.5 × 108
24 1.2 × 109 0 9.8 × 108 7.9 × 108
a The cultures were collected at different time intervals such as 0, 3, 6, 18 and 24 hafter the induction of lysis.
b The cultures were incubated at 37 ◦C and 42 ◦C, and the viable cell count was
C.V. Jawale et al. / Vac
.5. Enzyme-linked immunosorbant assay (ELISA)
An indirect-ELISA against the outer membrane protein (OMP)xtracted from the JOL394 SG wild type strain was performed touantify the plasma IgG and intestinal sIgA concentrations using
chicken IgG and IgA ELISA quantitation kit (Bethyl Laboratories,ontgomery, TX, USA). The plasma samples were diluted as 1:250
nd intestinal wash samples were diluted as 1:100. The indirectLISA was carried out using the procedure as described previously34].
The sonicated bacterial cell protein (sbcp) suspension extractedrom JOL394 was used as antigen for stimulation [34]. Briefly,
× 105 viable peripheral blood mononuclear cells (PBMCs) wasdded to complete RPMI-1640 medium. The suspension was incu-ated in triplicate in the wells of 96-well tissue culture plates with0 �L of medium alone or medium containing 4 �g/mL of (sbcp) at0 ◦C in a humidified 5% CO2 atmosphere for 72 h. Proliferation oftimulated lymphocytes was measured and the mean stimulationndex (SI) was determined as previously described [34].
.7. Analysis of CD4+ and CD8+ T cells by flow cytometry
Total 1 × 106 viable peripheral blood mononuclear cells werencubated with 0.1 mL of appropriately diluted fluorescein isothio-yanate (FITC)-labeled anti-CD3, biotin (BIOT)-labeled anti-CD4nd phycoerythrin (PE)-labeled anti-CD8a (all SouthernBiotech,irmingham, USA) monoclonal antibodies in the dark at 4 ◦C for0 min. For secondary staining of biotin (BIOT)-labeled anti-CD4 theells were incubated with 0.1 mL of appropriately diluted allophy-ocyanin (APC)-labeled streptavidin monoclonal antibody in theark at 4 ◦C for 30 min. All the cell samples were analyzed with aow cytometer (BD Biosciences, NJ, USA).
.8. Protection efficacy against virulent SG challenge
All the 15 birds present in each group were challenged. Each birdas inoculated with 100 �L of a suspension containing 1 × 106 CFU
f a wild-type Salmonella Gallinarum JOL394 strain on day 21ost-immunization. Mortality was observed daily for 14 d after thehallenge. After 14th day post challenge, the surviving birds thenere euthanized and examined for macroscopic lesions on liver
nd spleen. Lesion scores were determined and recorded using aystem similar to one previously described [17,34].
.9. Statistical analysis
All data are expressed as the mean ± standard deviation (S.D.)nless otherwise specified. A non-parametric chi-square test wassed to analyze significant differences in mortality and gross lesioncores. An independent sample t-test was used to analyze statisticalifferences in immune responses between the immunized groupsnd unimmunized control group. Differences were considered toe statistically significant when the p-values were ≤0.05 or 0.01.
. Results
.1. Construction of the ghost cassette and pJHL101 plasmid
The ghost cassette was isolated from pJHL99 by means of
igestion of the vector with XbaI enzyme. The isolated DNAragment containing the ghost cassette was cloned into pYA334232], thereby resulting in the antibiotic resistance gene-free lysislasmid pJHL101. The expression of gene E of the ghost cassette
measured as colony forming unit (CFU)/mL.c JOL1291 strain containing the pJHL101.d Vector control strain constructed by transforming JOL967 with backbone vector
pYA3342 (without lysis gene E).
used in the present study was under the control of rightwardʎPR promoter from bacteriophage lambda [31]. A mutation intothe operator 2 (OR2) of the PR promoter resulted into a modifiedexpression system, which stably repressed gene E expression attemperatures of up to 37 ◦C, but still allowed the induction of celllysis at a temperature 42 ◦C.
3.2. Lysis of SG bacterial ghost
The antibiotic gene free lysis plasmid pJHL101 was used to trans-form into a lon, cpxR and asd deleted mutant SG, JOL967 [28]. TheJOL1291 was inoculated in the LB broth and the culture was grownat 37 ◦C till it achieved the mid-logarithmic growth phase. SG ghostformation was carried out by shifting the mid-logarithmic phasegrown culture from 37 ◦C to 42 ◦C to activate gene E-mediated lysis.The viable number of cells reduced at 6 h after lysis induction andthe cell viability was decreased to 0% at 18 h after lysis induction(Table 1). The SG ghost culture that is incubated at 37 ◦C showed thepresence of viable cell count, which indicates the absence of geneE-mediated lysis activity at 37 ◦C (Table 1). The viable cell count invector control strain was not much affected by shift in the incu-bation temperature from 37 ◦C to 42 ◦C. So, data suggests that thereduction in the viable cell count of JOL1291 at 42 ◦C is due to thelytic activity of the protein E.
The scanning electron microscopic images showed the lysis tun-nel formation either in the pole region or central division regionof the SG ghost (indicated by arrows in Fig. 1A). In transmissionelectron micrographs, the SG ghost was clearly distinguished fromthe wild-type cell by their lower electron density and collapsedor round cell envelopes due to the loss of cytoplasmic material(Fig. 1C).
3.3. General conditions in chickens after immunization
Post-immunization the birds were observed for the vaccineassociated mortality and clinical symptoms in order to evaluatethe safety as a vaccine. There was no mortality in any of the immu-nized groups. All birds were apparently healthy and did not showany symptoms of anorexia, diarrhea, or depression for 3 week afterinoculation with the ghost strain. The vaccination status of the birdsdid not affect their body weight gain (Data not shown).
3.4. Antibody response analysis
The systemic IgG and mucosal sIgA antibodies to the spe-
cific antigen extracted from the wild-type SG were evaluated byindirect ELISA at each week post-immunization. On the second wpi,the immunized birds in group B (orally immunized) and group C(intramuscularly immunized) demonstrated significant elevation
ig. 1. Evaluation of Salmonella Gallinarum ghosts (JOL1291) and Salmonella Gallinrrow shows the trans-membrane lysis tunnel. (B) Naive Salmonella Gallinarum eaive Salmonella Gallinarum examined by TEM.
3.5 fold increase) in plasma IgG levels than the group A controlirds (Fig. 2A). The significantly increased plasma IgG levels in the
mmunized groups were observed on the third wpi (Fig. 2A). ThegG titers of group B and C birds were significantly increased at sec-nd and third wpi in comparison with those at first wpi, howeverhe IgG titers of group A chickens did not showed any significanteviation from first to third wpi (Fig. 2A). Secretory sIgA antibod-
es in the intestinal lavage were also significantly increased in the
rally immunized group B compared to the other groups (Fig. 2B).here was no significant increase in intestinal sIgA levels in group
(Fig. 2B). During the subsequent weeks after immunization, the
ig. 2. The plasma IgG and intestinal sIgA levels in chickens against the outerembrane protein (OMP). (A) Heparinized blood (1 mL) was collected from both
accinated and unvaccinated groups (n = 5). Plasma IgG levels against specific anti-ens were determined with the chicken IgG quantitative ELISA. Antibody levelsere expressed as mean ± standard deviation (S.D.) values for each week post-
mmunization. (B) The sIgA levels in intestinal wash fluids against OMP antigen wereuantified using the chicken IgA quantitative ELISA. Antibody levels were expresseds mean ± S.D. Statistical significance was defined at p values ≤0.05 or 0.01. * p < 0.05,
* p < 0.01 vs. unvaccinated control. group A, non immunized control; group B, orallymmunized; group C, intramuscularly immunized.
(JOL967) by SEM (A and B) and TEM (C and D). (A) Salmonella Gallinarum ghosts.ed by SEM. (C) Loss of cytoplasmic material of Salmonella Gallinarum ghosts. (D)
data shows that the sIgA titers of immunized group B chickens weresignificantly elevated from second to third wpi (Fig. 2B).
3.5. Lymphocyte proliferation assay
On the third wpi, all the immunized birds showed significantlyelevated lypmhocyte proliferative responses compared to the con-
trol group A (Fig. 3). Stimulation indices for the immunized groupsB and C were 4.3 and 5.69, respectively.
Fig. 3. The lymphocyte stimulation responses determined at 3-week-postimmunization against the sbcp Antigen: The stimulation index of lymphocytesample from the chickens was determined by the peripheral lymphocyte prolifera-tion assay. Statistical significance was defined at p values ≤0.05 or 0.01. * p < 0.05 vs.unvaccinated control. group A, non immunized control; group B, orally immunized;group C, intramuscularly immunized.
Fig. 4. Analysis of CD3+ CD4+ and CD8+ T-lymphocytes populations. (A) Flow cytometry scatter dot plots for CD3+, CD4+, CD8+ T-cell populations. The plots represent events foro cordinr chicka up. * pi
3c
glic(
ne representative chicken from each group. The gating and the quadrants are set acepresent CD3+ CD4+ T-lymphocytes population in immunized and non immunizednd non immunized chickens. Values are shown as mean ± S.D. of 5 chicken per grommunized; group C, intramuscularly immunized.
.6. Measurement of CD4+ and CD8+ T cells in immunizedhickens
Analysis of the T cell populations in the respective immunizedroups showed significantly increased number of CD4+ and CD8+
ymphocytes in all the immunized groups compared to the nonmmunized group A (Fig. 4). The immunized group had statisti-ally (p < 0.05) higher numbers of CD3+ CD4+, and CD3+ CD8+ T cellsFig. 4B and C).
g to the standard procedures of the BD-Biosciences flow cytometer. (B) Bar-graphsens. (C) Bar-graphs represent CD3+ CD8+ T-lymphocytes population in immunized
≤ 0.05 vs. unvaccinated control. Group A, non-immunized control; group B, orally
3.7. Protection against virulent challenge in chickens
The chickens in the immunized groups B and C were significantlyprotected against FT compared to the non-immunized group Abirds (Table 2). Group A displayed 60% mortality (Table 2) whereas
the groups B and C exhibited significantly lower mortality. Theprotection efficacy depended on the immunization routes as themortality rates for groups B (orally immunized), and C (immu-nized intramuscularly) were 26.6% and 13.3%, respectively. After
1098 C.V. Jawale et al. / Vaccine 3
Table 2Mortality and gross lesion in the chickens after challenge with wild type SG.
a Immunization was performed at day 7 of age with the SG ghost strain and thegroups were designated as group A (non vaccinated), group B (orally immunized),group C (immunized intramuscularly). Fifteen birds were allocated per group.
b Challenge was performed with a wild type Salmonella Gallinarum (JOl 394) strainusing 1 × 106 CFU after 21 days post immunization.
c Gross lesion was observed at day 14 post challenge.d Number of dead birds following challengee Group lesion score (mean ± SEM). All values were considered to be significant if
p
1otl
4
eavurpTdeRarcbss(etcrtfgtowptntckmmcdb
≤ 0.05 or 0.01.* p < 0.05,
** p < 0.01 vs. non vaccinated group A.
4th day post challenge, the control group A demonstrated severergan lesion scores as 1.86 and 1.80 for liver and spleen, respec-ively (Table 2). In contrast, the groups B and D had significantlyower lesion scores than those of the control.
. Discussion
For effective vaccination against FT, vaccine formulations shouldither contain live attenuated Salmonella vaccine strains or inactiv-ted organisms. However the major constraint with using liveaccines against FT is the safety as they are associated with resid-al virulence [12]. On the other hand, inactivated vaccines haveeduced the immunogenicity due to the traditional inactivationrocedures and thus are unable to offer effective immunity [35].he present study constructed a novel inactivated SG vaccine can-idate using a genetic engineering technique designed towardnhanced safety with high immune induction in animal hosts.etaining the virulence is a matter of concern with the attenu-ted bacterial vaccine strains. Disruption of the lon gene damageeplication of the Salmonella bacteria in the host cell whereaspxR gene deletion increases the adhesion and invasion of theacterial strain. Thus, the deletion of these two genes inducesatisfactory attenuation of SG and thus could be used to con-truct a safe vaccine formulation [28,36]. Although, bacterial ghostsBGs) have been shown to induce protective immunity in sev-ral animal models, most of the BG’s contain the plasmids withhe antibiotic resistance gene [14,17,18,24]. The use of such vac-ines is undesirable due to the risk of spreading the antibioticesistance genes among the environmental micro-organisms. Inhe present study, we successfully constructed an antibiotic generee plasmid containing the ghost cassette for the production of SGhost. In general, the ghost strains incorporated with the conven-ional ghost cassette are able to survive at 28 ◦C [37,38]. We andthers have successfully shown that the Salmonella ghost strainshen grown at 37 ◦C and then exposed to lysis induced by tem-erature upshift are immunogenic and could effectively inducehe immune responses in animal host possibly due to mainte-ance of their natural antigenic structures [14,17,18]. Althoughhe delayed onset of the lysis is observed in the present hostell than those reported in the literature, but the E-mediated cellilling activity is the same, indicated by the presence of trans-embrane lysis tunnels and absence (Fig. 1A) of the cytoplasmic
aterial (Fig. 1C) and the gradual decrease in the viable cell
ount over the time course (Table 1), and no viable cells wereetected after the completion of lysis hours. The weakened lysisehavior might have arisen from either reduced susceptibility
2 (2014) 1093– 1099
of the S. Gallinarum cell wall to the non-enzymatic E protein orreduced power of the lambda �pR promoter in currently used hostcell of S. Gallinarum.
The chickens immunized with the presently constructed SGghosts via various routes were efficiently protected againstexperimentally-induced FT. The least mortality and significantlylower gross lesion scores in all the immunized groups comparedto the non-immunized control was strongly suggestive of this pro-tective effect of the SG ghost in chickens (Table 2). In addition, ourprevious studies have shown that a live SG mutant effectively pro-tected the chickens against experimental FT infection [28,34]. Thepresent study SG (�lon, �cpxR and �asd) ghost vaccine candidatecontaining antibiotic resistance gene-free plasmid when inoculatedto chicken at day 7 of age showed approximately 80% survival ratewhich may suggests that the immunization with inactivated SGghost vaccine could be as efficient as its live counterparts. How-ever, the comparative evaluation of the presently constructed SGghost vaccine with its live counterparts is required for further theknowledge on this subject.
As reported in the literature, the total DNA content present in theghost preparation was inactivated by the expressing staphylococcalnuclease A (SNA) gene during the lysis process [26,29]. The nucleaseactivity of the SNA gene product is responsible for cleaning up resid-ual DNA in BGs and can lead to complete inactivation of the cultureas it degrades the host DNA into fragments no longer than 100 basepairs [30]. Addition of the alkylating agent �-propiolactone (BPL)after harvesting is also effective in fully inactivating all viable cellseither in combination with or as an alternative to SNA [39]. How-ever, in the development of veterinary vaccines, not only safetyenhancement but also reducing production cost is also importantaspect. Therefore, the use of expensive chemicals in the produc-tion of vaccines is not recommendable. Here, our results may alsosuggest that this approach of using antibiotic-gene free plasmid forconstruction of ghost plasmid in an attenuated SG strain is reli-able approach of the safety enhanced SG ghost production whichcompletely inhibits the possibility of the horizontal transfer of theantibiotic resistance gene from vaccine preparation to the otherresident bacteria.
BGs contain all the relevant antigenic determinants of their livecounterpart which are required for dendritic cells and macrophagesactivation in immunized animals and thus are capable of stimu-lating humoral and cell-mediated immune responses [14,24]. Ourdata indicated that all the immunized groups showed significantlyhigher systemic IgG responses whereas mucosal sIgA levels weresignificantly elevated in group B (Fig. 2A and B). The stronger sIgAresponse in the orally immunized group is likely due to the factthat the oral route of immunization induces significantly highermucosal secretory IgA levels compared to the other routes [40].This data is in agreement with our previous results where similarpattern of the sIgA response was observed in the chickens immu-nized via parenteral and oral route [17] and indicate that use of aghost strain constructed using an antibiotic gene free plasmid in anattenuated SG strain is efficient in terms of inducing humoral andmucosal antibody response.
The high survival rates in chickens after a virulent SG infec-tion have been correlated with the measurements of cell-mediatedimmune responses indicating the important role of these immuneresponses in protecting against Salmonella infections [17,34,41].The results of the present study demonstrated a potent lymphocyteproliferation response and dynamic changes in CD4+ and CD8+ T-lymphocyte populations in the immunized chickens compared tothe control. CD4+ T helper cells play an important role in establish-
ing and maximizing the capabilities of the immune system via theproduction of Th1 cytokines and increased antibody secretion [42].CD8+ cytotoxic T cells play a central role in immune protection inintracellular bacterial parasites such as Salmonella [43–45]. Overall,
cine 3
tr
iaFdcanc
A
o2
i
R
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
C.V. Jawale et al. / Vac
his data may indicate the development of cell mediated immuneesponses after vaccination.
In conclusion, the results of the present study underline themportance of generation of safety enhanced SG ghosts usingntibiotic gene free plasmid for developing a safe and effectiveT vaccine for chickens. The newly constructed SG ghost vaccineescribed presently is as efficient as previously reported ghost vac-ine [17] and could be added to the list of FT vaccine candidates,lthough further experiments, including optimization of the immu-ization strategy, clinical observation for more detail on safetyoncerns and its effects on egg production, are required.
cknowledgments
This work was supported by the National Research Foundationf Korea (NRF) grant funded by the Korea government (MISP) (No.013R1A4A1069486).
Conflict of interest statement: The authors declare no conflict ofnterest.
eferences
[1] Pomeroy BS, Nagaraja KV. Fowl typhoid. In: Calnek BW, Barnes HJ, Beard CW,Reid WM, Yoder Jr HW, editors. Diseases of poultry. Ames, IA: Iowa State Uni-versity Press; 1991. p. 87–99.
[2] Barrow PA, Wallis TS. Vaccination of food animals. In: In Wray C, Wray A, edit-ors. Salmonella in domestic animals. Wallingford, UK: CABI Publishing; 2000.p. 323–39.
[3] Shivaprasad HL. Fowl typhoid and pullorum disease. Rev Sci Tech Off Int Epiz2000;19(2):405–24.
[4] Bouzoubaa K, Nagaraja KV, Kabbaj FZ, Newman JA, Pomeroy BS. Feasibility ofusing proteins from Salmonella gallinarum vs. 9R live vaccine for the preventionof fowl typhoid in chickens. Avian Dis 1989;33:385–91.
[5] Rosu V, Chadfield MS, Santona A, Christensen JP, Thomsen LE, Rubino S, OlsenJE. Effects of crp deletion in Salmonella enterica serotype Gallinarum. Acta VetScand 2007;49:14.
[6] Shah DH, Shringi S, Desai AR, Heo EJ, Park JH, Chae JS. Effect of metC mutation onSalmonella Gallinarum virulence and invasiveness in 1-day-old White Leghornchickens. Vet Microbiol 2007;119:352–7.
[7] Smith HW. The use of live vaccines in experimental Salmonella gallinaruminfection in chickens with observation on their interference effect. J Hyg1956;54:419–32.
[8] Zhang-Barber L, Turner AK, Dougan G, Barrow PA. Protection of chickens againstexperimental fowl typhoid using a nuoG mutant of Salmonella serotype Galli-narum. Vaccine 1998;16:899–903.
[9] Silva EN, Snoeyenbos GH, Weinack OM, Smyser CF. Studies on the useof 9R strain of Salmonella gallinarum as a vaccine in chickens. Avian Dis1981;25:38–52.
10] Barrow PA, Huggins MB, Lovell MA. Host specificity of Salmonella infection inchickens and mice is expressed in vivo primarily at the level of the reticuloen-dothelial system. Infect Immun 1994;62:4602–10.
11] Jeong JH, Song M, Park SI, Cho KO, Rhee JH, Choy HE. Salmonella enterica serovarGallinarum requires ppGpp for internalization and survival in animal cells. JBacteriol 2008;190:6340–50.
12] Kwon HJ, Cho SH. Pathogenicity of SG 9R, a rough vaccine strain against fowltyphoid. Vaccine 2011;29:1311–8.
13] Jalava K, Hensel A, Szostak M, Resch S, Lubitz W. Bacterial ghosts as vaccinecandidates for veterinary applications. J Controlled Release 2002;85:17–25.
14] Peng W, Si W, Yin L, Liu H, Yu S, Liu S, Wang C, Chang Y, Zhang Z, Hu S, Du Y.Salmonella Enteritidis ghost vaccine induces effective protection against lethalchallenge in specific-pathogen-free chicks. Immunobiology 2011;216:558–65.
15] Szostak MP, Hensel A, Eko FO, Klein R, Auer T, Mader H, Haslberger A, BunkaS, Wanner G, Lubitz W. Bacterial ghosts: nonliving candidate vaccines. J Bio-technol 1996;44:61–170.
16] Lubitz W, Witte A, Eko FO, Kamal M, Jechlinger W, Brand E, Marchart J,Haidinger W, Huter V, Felnerova D, Stralis-Alves N, Lechleitner S, Melzer H,Szostak MP, Resch S, Mader H, Kuen B, Mayr B, Mayrhofer P, GeretschlägerR, Haslberger A, Hensel A. Extended recombinant bacterial ghost system. JBiotechnol 1999;73:261–73.
17] Chaudhari AA, Jawale CV, Kim SW, Lee JH. Construction of a Salmonella Galli-narum ghost as a novel inactivated vaccine candidate and its protective efficacyagainst fowl typhoid in chickens. Vet Res 2012;23(43(1)):44.
generated using a modified cI857/� PR/gene E expression system. Infect Immun2012;80:1502–9.
19] Witte A, Wanner G, Bläsi U, Halfmann G, Szostak M, Lubitz W. Endogenoustransmembrane tunnel formation mediated by phiX174 lysis protein E. J Bac-teriol 1990;172:4109–14.
20] Witte A, Wanner G, Sulzner M, Lubitz W. Dynamics of PhiX174 protein E-mediated lysis of E. coli. Arch Microbiol 1992;157:381–8.
21] Eko FO, Witte A, Huter V, Kuen B, Fürst-Ladani S, Haslberger A, Katinger A,Hensel A, Szostak MP, Resch S, Mader H, Raza P, Brand E, Marchart J, Jech-linger W, Haidinger W, Lubitz W. New strategies for combination vaccinesbased on the extended recombinant bacterial ghost system. Vaccine 1999;17:1643–9.
22] Eko FO, Mayr UB, Attridge SR, Lubitz W. Characterization and immunogenic-ity of Vibrio cholerae ghosts expressing toxin-coregulated pili. J Biotechnol2000;83:115–23.
23] Mayr UB, Haller C, Haidinger W, Atrasheuskaya A, Bukin E, Lubitz W, IgnatyevG. Bacterial ghosts as an oral vaccine: a single dose of Escherichia coliO157:H7 bacterial ghosts protects mice against lethal challenge. Infect Immun2005;73:4810–7.
24] Wang X, Lu C. Mice orally vaccinated with Edwardsiella tarda ghosts are signif-icantly protected against infection. Vaccine 2009;27:1571–8.
25] Ramos I, Alonso A, Peris A, Marcen JM, Abengozar MA, Alcolea PJ, Castillo JA,Larraga V. Antibiotic resistance free plasmid DNA expressing LACK proteinleads towards a protective Th1 response against Leishmania infantum infection.Vaccine 2009;12:6695–703.
26] Kwon SR, Kang YJ, Lee DJ, Lee EH, Nam YK, Kim SK, Kim KH. Generation of Vibrioanguillarum ghost by coexpression of PhiX 174 lysis E gene and staphylococcalnuclease A gene. Mol Biotechnol 2009;42:154–9.
27] Nakayama K, Kelly SM, Curtiss RIII. Construction of an Asd+ expression-cloningvector: stable maintenance and high level expression of cloned genes in aSalmonella vaccine strain. Nat Biotechnol 1988;6:693–7.
28] Chaudhari AA, Kim SW, Matsuda K, Lee JH. Safety evaluation and immunogenic-ity of arabinose-based conditional lethal Salmonella Gallinarum mutant unableto survive ex vivo as a vaccine candidate for protection against fowl typhoid.Avian Dis 2011;55:165–71.
29] Lee DJ, Kwon SR, Zenke K, Lee EH, Nam YK, Kim SK, Kim KH. Generation of safetyenhanced Edwardsiella tarda host vaccine. Dis Aquat Organ 2008;81:249–54.
30] Haidinger W, Mayr UB, Szostak MP, Resch S, Lubitz W. Escherichia coli ghostproduction by expression of lysis gene E and staphylococcal nuclease. ApplEnviron Microbiol 2003;69:6106–13.
31] Ra CH, Kim YJ, Park SJ, Jeong CW, Nam YK, Kim KH, Kim SK. Evaluation of optimalculture conditions for recombinant ghost bacteria vaccine production with theantigen of Streptococcus iniae GAPDH. J Microbiol Biotechnol 2009;19:982–6.
32] Kang HY, Srinivasan J, Curtiss 3rd R. Immune responses to recombinant pneu-mococcal PspA antigen delivered by live attenuated Salmonella enterica SerovarTyphimurium vaccine. Infect Immun 2002;70:1739–49.
33] Porter Jr RE, Holt PS. Use of a pilocarpine-based lavage procedure to study secre-tory immunoglobulin concentration in the alimentary tract of White Leghornchickens. Avian Dis 1992;36:529–36.
34] Matsuda K, Chaudhari AA, Kim SW, Lee KM, Lee JH. Physiology, pathogenicityand immunogenicity of lon and/or cpxR deleted mutants of Salmonella Galli-narum as vaccine candidates for fowl typhoid. Vet Res 2010;41:59.
35] Desin TS, Köster W, Potter AA. Salmonella vaccines in poultry: past, present andfuture. Expert Rev Vaccines 2013;12(1):87–96.
36] Matsuda K, Chaudhari AA, Lee JH. Evaluation of safety and protection efficacyon cpxR and lon deleted mutant of Salmonella Gallinarum as a live vaccinecandidate for fowl typhoid. Vaccine 2011;29:68–74.
37] Jechlinger W, Haller C, Resch S, Hofmann A, Szostak MP, Lubitz W. Comparativeimmunogenicity of the hepatitis B virus core 149 antigen displayed on the innerand outer membrane of bacterial ghosts. Vaccine 2005;23:3609–17.
38] Tu FP, Chu WH, Zhuang XY, Lu CP. Effect of oral immunization with Aeromonashydrophila ghosts on protection against experimental fish infection. Lett ApplMicrobiol 2010;50:13–7.
39] Perrin P, Morgeaux S. Inactivation of DNA by betapropiolactone. Biologicals1995;23:207–11.
40] Pasetti MF, Simon JK, Sztein MB, Levine MM. Immunology of gut mucosal vac-cines. Immunol Rev 2011;239:125–48.
41] Park SI, Jeong JH, Choy HE, Rhee JH, Na HS, Lee TH, Her M, Cho KO, HongY. Immune response induced by ppGpp-defective Salmonella enterica serovarGallinarum in chickens. J Microbiol 2010;48:674–81.
43] Flynn JL, Goldstein MM, Triebold KJ, Koller B, Bloom BR. Major histocom-patibility complex class I-restricted T cells are required for resistance to
