A vaccine candidate for post-weaning diarrhea in swine constructed with a live attenuated Salmonella delivering Escherichia coli K88ab, K88ac, FedA, and FedF fimbrial antigens and its immune responses in a murine model

Abstract

In order to construct a novel vaccine candidate for preventing post-weaning diarrhea in swine, the individual genes for Escherichia coli K88ab, K88ac, FedA, and FedF fimbriae were inserted into a secretion plasmid pBP244 containing asd, lepB, secA, and secB. These were transformed into Salmonella Typhimurium Δlon ΔcpxR Δasd. Secretion of the individual recombinant fimbrial antigens was confirmed by immunoblot analysis. Groups 1 and 2 mice received a single oral dose of the vaccine mixture and S. Typhimurium carrying pBP244 only as a control, respectively. In groups 3 and 4, mice were primed and boosted with the vaccine mixture and S. Typhimurium carrying pBP244 only as a control, respectively. In general, all immunized mice had significantly increased serum immunoglobulin (Ig)G (P < 0.05) and intestinal secretory IgA against the individual fimbrial antigens compared with those mice in the control group. In the IgG2a and IgG1 titer assay, only IgG2a titer was increased in group 1, while both IgG2a and IgG1 titers were increased in group 3. Furthermore, the vaccine strains were not detected in the excreted feces of any immunized mice. Thus, the vaccine candidate can be highly immunogenic and be safe to the environment.

Introduction

Enterotoxigenic Escherichia coli (ETEC) strains cause neonatal diarrhea and weaning and post-weaning diarrhea (PWD) in pigs, which can retard growth and even cause mortality (1–4). To cause the disease, ETEC must colonize the mucosal surface of the intestine using surface proteins known as fimbriae and produce heat-stable (STa, STb), heat-labile (LT) enterotoxins, or both (1,3,4). The known porcine ETEC fimbriae are F4 (K88), F5 (K99), F6 (987P), F18, and F41 (1,4). Among these fimbriae, the main ETEC adhesive factors associated with PWD are F4 (previously called K88) and F18 fimbriae (1,3,4).

Morphologically these fimbriae are straight, bent, or kinky proteinaceous appendages originating from the outer membrane of the bacterial cells. The K88 fimbriae are 2.1 nm in diameter and the molecular weight is 27.6 kDa. Three serological variants of K88 fimbriae are known: K88ab, K88ac, and K88ad (5). The most common variant isolated from diarrheic piglets is K88ac (6,7).

The F18 fimbrial E. coli strains adhere to the microvilli of small intestine epithelial cells in piglets and are associated with porcine PWD and pig edema disease, occurring after weaning or transfer to fattening premises. The genetic organization of the fed gene cluster, which is involved in the biosynthesis of F18 fimbriae, has been characterized (8,9). The FedA is the major protein and FedF is associated with adhesion of F18 fimbriae to epithelial cells (9,10).

Live vaccine vehicles offer a powerful approach for inducing protective immunity, such as mucosal and systemic immune responses, against pathogenic microorganisms (11). In particular, attenuated Salmonella strains have been modified to express a wide range of antigens from bacterial, parasitic, and viral sources (11–13). The present study aimed to construct a live vaccine candidate expressing and delivering recombinant E. coli K88ab, K88ac, FedA, and FedF fimbrial antigens via an attenuated S. Typhimurium system, and to investigate its immunegenecity. Individual genes for ETEC K88ab, K88ac, FedA, and FedF fimbrial antigens were inserted into the plasmid vector pBP244. These plasmids were transformed separately into the attenuated Salmonella for expression and delivery of the fimbrial antigens. Induction of immune responses by this vaccine construction was examined using a mouse model.

Materials and methods

Bacterial strains, plasmids, and growth conditions

Bacterial strains and plasmids used in this study are listed in Table I. Escherichia coli χ7213 was used as the host strain for construction of individual fimbrial antigen-Asd⁺ plasmids (14,15). The wild-type E. coli strains JOL416, JOL417, and JOL500 were used to amplify the genes for K88ab, K88ac, FedA, and FedF fimbriae. Escherichia coli JM109 and E. coli BL21(DE3)pLysS (hereafter referred to as E. coli BL21) were used as host cells for over-expression of recombinant fimbrial antigens. The attenuated S. Typhimurium strain, JOL912, was constructed by deletion of the lon, cpxR and asd genes in wild-type S. Typhimurium JOL401, as previously described (14,15). The strain JOL912 was used for delivery and expression of K88ab, K88ac, FedA, and FedF fimbrial proteins. The pQE31 and pET28a plasmids were used to overexpress fimbrial antigens. The pBP244 Asd⁺ plasmid was derived from pYA3493 Asd⁺ plasmid by incorporating the lepB, secA, and secB genes to improve secretion of expressed heterologous antigens in Salmonella (16). All strains were grown in Luria-Bertani media (LB; Becton, Dickinson and Company, Sparks, Maryland, USA) at 37°C. Diaminopimelic acid (DAP; Sigma-Aldrich, St. Louis, Missouri, USA) was added (50 μg/mL) to induce the growth of Asdnegative bacteria, such as E. coli χ7213 and JOL912.

Table I.

Bacterial strains and plasmids used in this study

Strain or plasmid	Description	Source or reference
Strain
E. coli
JOL416	Wild type F4 (K88ab)+ ETEC isolate from pig	This study
JOL417	Wild type F4 (K88ac)+ ETEC isolate from pig	This study
JOL500	Wild type F18+, LT+, STa+, STb+, stx2+, stx2e+ ETEC isolate from pig	This study
JM109	endA1, recA1, gyrA96, thi, hsdR17 (rk−, mk+), relA1, supE44, Δ(lac-proAB), [F′ traD36, proAB, laqIqZΔM15]	Promega
BL21(DE3)pLysS	F−, ompT, hsdSB (rB−, mB−), dcm, gal, λ(DE3), pLysS, Cmr	Promega
χ7213	λpir thi-1 thr-1 leuB6 supE44 towA21 lacY1 recA RP4-2 Tc::Mu ΔasdA4	16
JOL470	JM109 with pQE-K88ab	This study
JOL471	JM109 with pQE-K88ac	This study
JOL943	BL21 with pET-FedA	This study
JOL944	BL21 with pET-FedF	This study
JOL938	χ7213 with pBP-K88ab	This study
JOL939	χ7213 with pBP-K88ac	This study
JOL942	χ7213 with pBP-fedA	This study
JOL923	χ7213 with pBP-fedF	This study
JOL908	χ7213 with pBP244	This study
S. Typhimurium
JOL401	Wild type isolate from chicken	This study
JOL912	Δlon ΔcpxR Δasd, a derivative of JOL401	14
JOL940	JOL912 containing pBP-K88ab	This study
JOL941	JOL912 containing pBP-K88ac	This study
JOL934	JOL912 containing pBP-fedA	This study
JOL935	JOL912 containing pBP-fedF	This study
JOL932	JOL912 containing pBP244	This study
Plasmid
pQE31	IPTG-inducible expression vector; Amr	QIAGEN
pET28a	IPTG-inducible expression vector; Kmr	Novagen
pQE-K88ab	pQE31 derivative containing K88ab fimbrial gene	This study
pQE-K88ac	pQE31 derivative containing K88ac fimbrial gene	This study
pET-fedA	pET28a derivative containing fedA gene	This study
pET-fedF	pET28a derivative containing fedF gene	This study
pYA3493	Asd+, pBR322ori, β-lactamase signal sequence-based periplasmic secretion plasmid	16
pBP244	pYA3493 derivative containing lepB, secA and secB genes	16
pBP-K88ab	pBP244 derivative containing K88ab fimbrial gene	This study
pBP-K88ac	pBP244 derivative containing K88ac fimbrial gene	This study
pBP-fedA	pBP244 derivative containing fedA gene	This study
pBP-fedF	pBP244 derivative containing fedF gene	This study

Purification of individual recombinant fimbrial antigens

The genes for individual fimbrial antigens were amplified by polymerase chain reaction (PCR) using the fimbrial-specific primer sets (Table II). The K88ab and K88ac fimbriae-specific PCR products were digested with BamHI and HindIII, and then inserted into pQE31. These plasmids were used to transform E. coli JM109 and were designated as JOL470 and JOL471, respectively. The FedA and FedF fimbriae-specific PCR products were digested with EcoRI and HindIII and then inserted into pET28a. These plasmids were transformed into E. coli BL21 and were designated as JOL943 and JOL944, respectively. Recombinant K88ab, K88ac, FedA, and FedF proteins were prepared from JOL470, JOL471, JOL943, and JOL944, respectively. The proteins were purified by using an affinity purification process with Ni-nitrilotriacetic acid-agarose support (Peptron, Daejeon, South Korea) and stored at −70°C until required.

Table II.

Polymerase chain reaction primers and their product sizes

Gene	Primer	Sequence	Size (bp)	Accession number	Reference
faeF	K88ab-F	CCGCGGATCCGGCACATGCCTGGATGACT	814	V00292.1	This study
	K88ab-R	CCGCAAGCTTCCAGCAACTTTAGTAATAA
faeG	K88ac-F	CCGCGGATCCGGCACATGCCTGGATGACT	819	AJ616256.1	This study
	K88ac-R	CCGCAAGCTTAATTGGCAGCTCATCACG
fedA	FedA-F	CCGCGAATTCCAGCAAGGGGATGTTAAAT	455	M61713	This study
	FedA-R	CCGCAAGCTTGATGATTACTTGTAAGTA
fedF	FedF-F	CCGCGAATTCGCGTCTACTCTACAAGTA	846	DQ995282	This study
	FedF-R	CCGCAAGCTTTTACTGTATCTCGAAAACAA
cpxR	cpxR-F	CAGCGCCAGCGTCAACCAGAAGAT	304	AE006468	This study
	cpxR-R	GAGGCCATAACAGCAGCGGTAACT
lon	lon-F	ATTTTATCTCCCCTTTCGTTTTTC	244	AE006468	This study
	lon-R	CTGCCAGCCCTGTTTTTATTAGC

^aUnderlines indicate the sites of restriction enzymes, such as BamHI, EcoRI, and HindIII.

Construction of vaccine strains

To construct recombinant S. Typhimurium vector-mediated vaccine candidates delivering the recombinant K88ab, K88ac, FedA, and FedF fimbrial antigens, individual genes for the fimbrial antigens were inserted into pBP244, as previously described (14,16). These plasmids were transformed into E. coli χ7213 and colonies containing these plasmids were selected in the absence of DAP. Subsequently, the plasmids from the colonies were recovered and electroporated into JOL912. The different cell lines created by this process as the vaccine strains were designated as JOL940 for K88ab, JOL941 for K88ac, JOL934 for FedA, and JOL935 for FedF. The strain JOL932 was constructed by inserting only pBP244 into JOL912 and was used as a control.

Preparation of fimbrial antigens-specific antisera

To prepare specific antibodies against the individual fimbrial antigens, an emulsion containing approximately 250 μg of each purified recombinant fimbrial antigen in 0.5 mL of sterile phosphate buffered (saline) solution (PBS) and 0.5 mL of complete Freund adjuvant (Sigma-Aldrich, St. Louis, Missouri, USA) were subcutaneously injected into New Zealand white rabbits. Two boosters with the same antigen quantity in incomplete Freund adjuvant were given on days 14 and 28 post-prime immunization (PPI). Blood was collected to prepare antisera on day 14 after the last injection.

Immunoblot analysis

The individual recombinant K88ab, K88ac, FedA, and FedF fimbrial antigens secreted from JOL940, JOL941, JOL934, and JOL935, respectively, were identified by immunoblot analysis. In addition, JOL932 was used as a control for immunoblot analysis. The immunoblot analysis was carried out using a previously described method with slight modification (16). Briefly, each strain was cultured in LB broth at 37°C for preparation of secreted proteins. The culture was harvested at an optical density (OD) of 600 nm (OD₆₀₀) of 1.0. The bacterial culture was centrifuged at 3400 × g for 30 min in order to separate the cells from the supernatant. The collected supernatants were passed through a 0.22 μm pore-size filter and precipitated overnight at 4°C with 20% trichloroacetic acid (TCA). The precipitates were washed with acetone and re-suspended in cold PBS. These were boiled for 5 min and then separated by 15% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE). The resolved proteins were electrophoretically transferred to polyvinylidene fluoride (PVDF) membranes (Millipore; Billerica, Massachusetts, USA). The membranes were blocked for overnight at 4°C with 5% skim milk in PBS containing 0.1% Tween-20 and were sequentially incubated with each fimbrial antigen-specific polyclonal antibody and horse-radish peroxidase-conjugated goat anti-rabbit IgG (SouthernBiotech, Birmingham, Alabama, USA). Immunoreactive bands were developed with the addition of chemiluminescence dye using a Western Blot detection system (West-one; iNtRON, Seongnam, South Korea), and were observed using the multi-wavelength illumination system (KODAK Image Station 4000MM; Kodak, New Haven, Connecticut, USA).

Preparation of vaccine inoculum

The bacterial vaccine candidates and JOL932 (as control) were grown individually in LB broth overnight at 37°C. The living bacteria were diluted 1:20 in fresh LB broth, and were grown at 37°C to an OD₆₀₀ of 0.8. Cells were harvested by centrifugation at 3400 × g for 20 min. The pellets were resuspended in sterile PBS containing 20% sucrose (PBS-sucrose) to 1 × 10¹¹ colony-forming units (CFU)/mL. Mice were orally immunized with the live preparations on the same day, as previously described (15).

Immunization and sample collection

A total of 40 5-week-old female BALB/c mice were divided into 4 groups of 10. All groups of mice were primed at 6 wk of age, and mice in groups 3 and 4 were boosted at 9 wk of age. Group 1 mice received a single oral dose of a 20 μL mixture containing all the vaccine candidates. In group 2, a control group, mice received an oral dose of only 20 μL JOL932. In group 3, mice were orally primed and boosted with 20 μL of all bacterial vaccine candidates. In group 4, control group, mice were orally primed and boosted with 20 μL JOL932. Food and water were withdrawn 4 h prior to immunization and re-supplied 30 min after immunization. To evaluate serum immunoglobulin (Ig)G titers, blood samples were collected, and for mucosal secretory immunoglobulin (sIg)A titers, fecal samples were collected at 0, 2, 5, 7, 9, and 13 wk PPI. Sera were obtained from whole blood by centrifugation at 3400 × g for 5 min. The collected fecal samples were weighed and suspended at 100 mg/mL in PBS containing 0.1% sodium azide. All samples were stored at −70°C until use. The animal experiments described in this study were conducted with approval from the Chonbuk National University Animal Ethics Committee in accordance with the guidelines of the Korean Council on Animal Care.

Enzyme-linked immunosorbent assay (ELISA) for evaluation of immune responses

A standard ELISA was used to determine individual protein-specific IgG, IgG1, and IgG2a antibody titers in serum, and sIgA titers in fecal samples as previously described with slight modification (15). Briefly, ELISA plates (Greiner Bio-One, Frickenhausen, Germany) were coated with 0.5 μg of each antigen per well and incubated at 4°C for overnight. Sera were diluted 1:400 for examination of IgG, IgG1, and IgG2a titers. Fecal samples were diluted 1:3 for examination of sIgA titers. The plates were treated with horse-radish peroxidase-conjugated goat anti-mouse antibodies (Southern Biotechnology Associates). Enzymatic reactions were developed with o-phenylenediamine (Sigma-Aldrich) and measured with an automated ELISA spectrophotometer (TECAN, Salzburg, Austria) at 492 nm. A standard curve was generated describing the relationship between the concentration of the standards and their absorbance, and the concentration of antibodies for each sample was expressed as nanogram per milliliter (ng/mL).

Clinical signs and fecal shedding of vaccine strains after oral immunization

All mice orally immunized with the mixture of all vaccine strains were monitored daily for diarrhea, confusion, and abnormal behavior until day 21 post-oral immunization. In addition, to investigate shedding of the vaccine candidates in feces, 5 mice from groups 1 and 2 were tested daily from 3 d PPI until 3 wk PPI, while 5 mice from groups 3 and 4 were tested daily from 3 d post-booster immunization (PBI) until 3 wk PBI. The collected fecal samples were weighed and suspended as 100 mg/mL in buffered peptone water and were initially pre-enriched for 18 h at 37°C. Pre-enriched sample (100 μL) was transferred to 10 mL of Rappaport-Vassiliadis R10 broth (Becton, Dickinson and Company) and incubated under aerobic conditions for 24 h at 42°C. The enrichment medium (100 μL) was then streaked on brilliant green agar (BGA; Becton, Dickinson and Company) and presumptive Salmonella colonies were analyzed biochemically using the identification test panel (API 20E, bioMérieux, Marcy, I′Étoile, France). Identities of the vaccine candidates were confirmed by PCR using the following specific primer sets: OMPC (OMPCF: 5′-ATCGCTGACTTATGCAATCG-3′; OMPCR: 5′-CGGGTTGCGTTATAGGTCTG-3′) for Salmonella spp., TYPH (TYPHF: 5′-TTGTTCACTTTTTACCCCTGAA-3′; TYPHR: 5′-CCCTGACAGCCGTTAGATATT-3′) for S. Typhimurium, and cpxR and lon primer sets (Table II) for vaccine construction. Individual fimbrial specific primer sets were used to identify the individual vaccine strains (Table II).

Statistical analysis

Results are expressed as mean ± standard deviation (SD). The Mann–Whitney U-test using computer software (SPSS, version 16.0; SPSS, Chicago, Illinois, USA) was used to determine the differences in serum IgG and fecal sIgA titers between the vaccinated and vector control groups. Results were considered statistically significant if P < 0.05.

Results

Confirmation of vaccine strains

The 814-, 819-, 455-, and 846-bp DNA fragments for K88ab, K88ac, FedA, and FedF fimbriae, respectively, were amplified using fimbrial-specfic primers and cloned into pBP244 separately. Subsequently, these individual plasmids were transformed into Δlon ΔcpxR Δasd S. Typhimurium. Immunoblotting of TCA-precipitated culture supernatants was done to examine the expression and secretion of K88ab, K88ac, FedA, and FedF fimbrial antigens from the transformed cells. JOL932, a S. Typhimurium harboring only pPB244, was used as a control. The expected sizes (29 kDa for K88ab, 27 kDa for K88ac, 16 kDa for FedA, and 30 kDa for FedF) were observed in precipitated culture supernatants from the indivual transformants (Figure 1). In contrast, no band was observed from the control supernatant.

Figure 1.

Identification of secreted recombinant K88ab, K88ac, FedA, and FedF fimbrial antigens by immunoblot analysis. Recombinant K88ab, K88ac, FedA, and FedF fimbrial antigens expressed and secreted by JOL940, 941, 934, and 935, respectively, were detected by immunoblotting with the appropriate fimbrial-specific antibodies. Strain JOL932 carrying only pBP244 was used as a control. Lanes C, control; K88ab, recombinant K88ab fimbrial antigens secreted by JOL940; K88ac, K88ac fimbrial antigen secreted by JOL941; FedA, recombinant FedA fimbrial antigen secreted by JOL934; FedF, recombinant FedF fimbrial antigen secreted by JOL935.

Systemic and mucosal immune responses induced by inoculation of vaccine strains in mice

Antibody responses to each fimbrial antigen in the sera and fecal samples of mice immunized with the mixture of all vaccine strains and the control are presented in Figures 2 and 3. In groups 1 (single administration with the vaccine mixture) and 3 (prime-booster with the vaccine mixture), serum IgG titers against the individual fimbrial antigens were significantly increased compared with those of control groups 2 and 4, respectively, from week 5 PPI until the end of the study (P < 0.05) (Figure 2). In addition, in group 3 mice, the fecal sIgA titers against the individual fimbrial antigens were significantly higher than those of the control group 4 from week 2 PPI until the end of this study (P < 0.05), while the titers of group 1 mice were higher only at weeks 2 and 5 PPI (Figure 3).

Figure 2.

Titers of serum IgG against K88ab, K88ac, FedA, and FedF fimbrial antigens in mice immunized with the mixture of vaccine constructions. (A) K88ab-specific IgG; (B) K88ac-specific IgG; (C) FedA-specific IgG; and (D) FedF-specific IgG. Group 1 mice were given a single dose of the vaccine mixture. Group 2 mice were given a single dose of JOL932. Group 3 mice were primed and boosted with the vaccine mixture. Group 4 mice were primed and boosted with JOL932.

^a Significant differences between the antibody titers of groups immunized with the vaccine constructions and those of the control (P < 0.05).

Figure 3.

Titers of fecal sIgA against K88ab, K88ac, FedA, and FedF fimabrial antigens in mice immunized with the vaccine constructions. (A) K88ab-specific sIgA; (B) K88ac-specific sIgA; (C) FedA-specific sIgA; and (D) FedF-specific sIgA. Group 1 mice were given a single dose of the vaccine mixture. Group 2 mice were given a single dose of JOL932. Group 3 mice were primed and boosted with the vaccine mixture. Group 4 mice were primed and boosted with JOL932.

^a Significant differences between the antibody titers of groups immunized with the vaccine constructions and those of the control (P < 0.05).

ImmunoglobulinG isotype analysis

The nature of the immune response to the K88ab-, K88ac-, FedA-, and FedF-specific antigens was further examined by measuring the levels of IgG isotype subclasses IgG2a and IgG1. In group 1, the IgG2a titers to the individual fimbrial antigens were much higher than those of IgG1. On the other hand, in group 3, although the IgG2a titers prevailed, the titers of IgG1 to the antigens were also increased (Figure 4).

Figure 4.

Serum K88ab-, K88ac-, FedA-, and FedF-specific IgG1/IgG2a ratio. Group 1 mice were given a single dose of the vaccine mixture. Group 3 mice were primed and boosted with the vaccine mixture.

Clinical signs and shedding of the vaccine strains after oral immunization

Side effects, such as diarrhea, confusion, and abnormal behavior were not observed in any of the immunized group mice from day 1 post oral immunization until day 21. Shedding of the vaccine strains in feces was investigated from 3 d until 3 wk following oral inoculation. Five mice from each group were tested daily. The vaccine strains were not detected in the fecal samples from any of the immunized mice during the sampling period following inoculation.

Discussion

The key virulence factors of ETEC include bacterial adhesins or colonization factor antigens (17). Therefore, for protection against ETEC-induced diarrhea, specific antibodies, such as sIgA, that inhibit bacterial adherence to receptors on intestinal cells or that neutralize enterotoxins are important (3). In the absence of antibodies against adhesins, such as fimbriae, enterotoxin-neutralizing antibodies may be less effective due to the release of toxins directly on the surface of host cell membranes by adhering bacteria (3,18). Consequently, much effort has been directed at designing vaccines to stimulate mucosal immunity (19,20). Compartmentalization of the immune system blunts the effectiveness of parenteral immunization routes for inducing mucosal responses (21,22). Stimulating a mucosal response requires delivering the vaccine antigen to a mucosal inductive site such as Peyer’s patches in the gut via the oral route (23). However, oral mucosal vaccine approaches must preserve the integrity of the antigen during transit through the gut and deliver the antigen in such a way that enhances immunogenicity (23). Live and attenuated vaccines are capable of inducing effective immune responses (19). In particular, Salmonella has the advantage of having a tropism for the mucosal inductive tissues of Peyer’s patches, allowing in-situ antigen expression while enhancing immunogenicity. Salmonella Typhimurium is a promising tool for the generation of live recombinant vaccines (24–28). Attenuated Salmonella has been used to express over 50 different foreign antigens of various origins (24–28).

The proteins expressed on surface of bacteria and the antigens secreted from bacteria are highly immunogenic and/or more promptly interact with antigen-presenting cells (29). One previous study reported that the recombinant K99 fimbrial antigens expressed on the membrane surface of attenuated S. Typhimurium were delivered and elevated immune responses (30). However, the antigens expressed and embedded on membrane surface of the bacteria are less effective as vaccine candidates than the secreted antigens to host environments from the live Salmonella-based systems (29,31). In a previous study, attenuated S. Typhimurium Δlon ΔcpxR was found to induce protective immune responses and effectively provided protection against Salmonella infections (14,15). In this study, for delivery of recombinant K88ab, K88ac, FedA, or FedF fimbrial antigen, a S. Typhimurium Δlon ΔcpxR Δasd strain was constructed by deleting the 3 genes from wild-type S. Typhimurium and used as a delivery host. In addition, the vector plasmid pBP244, a secretion vector containing a β-lactamase signal sequence and lepB, secA, and secB to improve secretion of the expressed fimbrial antigens from the host (16), was used. The plasmid pBP244 also contains pBR ori; plasmids containing pBR ori have been stably maintained for over 50 generations in a S. Typhimurium strain grown in the presence of DAP (29). In general, the secretion of secretory proteins is accomplished using a type II Sec-dependent secretion system, or via the general secretion pathway (GSP), which is exploited by many Gram-negative bacteria (16). The secretion system described herein consisted of 2 distinct steps, as mentioned in a previous study (16). The initial step involved translocation to the periplasmic space by the Sec-dependent system, and the other involved secretion into the extracellular matrix using β-lactamase-dependent system. The immnuloblot assays that analyzed the individual vaccine strains in culture supernatants demonstrated robust production and secretion of the individual fimbrial antigens from the Salmonella strain. Therefore, we concluded that the pBP244 plasmid harboring the genes for K88ab, K88ac, FedA, or FedF was stably maintained and the expressed K88ab, K88ac, FedA, and FedF fimbrial antigens were efficiently secreted.

Along with the systemic immune response, mucosal antibody induction post-immunization is highly desirable for protection against ETEC infection (2,17). Our results showed that serum levels of IgG against K88ab, K88ac, FedA, and FedF in all groups of mice immunized with the mixture of vaccine candidates were significantly increased compared with those of the control mice (P < 0.05). In addition, fecal mucosal sIgA levels against all fimbrial antigens in the immunized mice were also increased. These data suggest that the fimbrial antigens secreted by the recombinant S. Typhimurium are immunogenic and effectively induced systemic and mucosal immune responses in mice. Our results showed that serum IgG titers against the individual fimbrial antigens in group 3 (prime-booster with the candidate) mice were higher than those of group 1 (single inoculation). Especially, mucosal sIgA titers against all the fimbrial antigens in group 3 mice were significantly higher than those of control group mice (P < 0.05). These data suggest that systemic and mucosal immune responses to the individual fimbrial antigens can be highly induced by booster with the candidate. This immune potency could efficiently inhibit the initial stage of ETEC attachment to the animal intestines and could thus result in effective protection against pathogenic E. coli infections.

ImmunoglobulinG subclass distribution is influenced by several factors including the cytokine environment during presenting antigens, type of antigen presenting cell and co-stimulatory interactions, as well as the nature and dose of antigen (23,32). Th1-helper T-cells direct cell-mediated immunity and promote class switching to IgG2a, while Th2-helper T-cells provide potent help for humoral immunity and promote class switching to IgG1 (29,32,33). Th1-type dominant immune responses are frequently observed after immunization with Salmonella (29,34,35). Our results showed that titers of IgG2a to the individual fimbrial antigens prevailed in groups 1 and 3 mice, suggesting that the vaccine strains may induce cellular immune responses. However, titers of IgG1 in group 3 mice were also highly elevated. These results suggest that booster administration of the candidate may effectively enhance both cellular and humoral immune responses.

To examine whether the vaccine constructs were excreted via the feces, we isolated the strains daily from fecal samples of immunized mice for 3 wk post-oral inoculation. No strains were identified in fecal samples from any mice, suggesting that the strain is excreted marginally, if at all, into the environment. Furthermore, adverse reactions including diarrhea and abnormal behavior were not observed in any of the immunized groups during the experimental period. This suggests that the constructs are not only efficient for inducing immune responses, but are also safe for the hosts and environment.