Evaluation of YadC protein delivered by live attenuated Salmonella as a vaccine against plague

Abstract

Yersinia pestis YadB and YadC are two new outer membrane proteins related with its pathogenicity. Here, codon optimized yadC, yadC810 (aa 32-551) or yadBC antigen genes delivered by live attenuated Salmonella strains are evaluated in mice for induction of protective immune responses against Y. pestis CO92 through subcutaneous (s.c.) or intranasal (i.n.) challenge. Our findings indicate that mice immunized with Salmonella synthesizing YadC, YadC810 or YadBC develop significant serum IgG responses to purified recombinant YadC protein. For s.c. challenge (~230 LD₅₀ of Y. pestis CO92), mice immunized with Salmonella synthesizing YadC or YadC810 are afforded 50% protection but no protection by immunization with the Salmonella strain synthesizing YadBC. None of these antigens provided protection against i.n. challenge (~31 LD₅₀ of Y. pestis CO92). In addition, s.c. immunization with purified YadC810 protein emulsified with Alum adjuvant does not elicit a protective response against Y. pestis administered by either challenge route.

1. Introduction

Yersinia pestis is the causative agent of both bubonic and pneumonic plague, primarily infects rats and wild rodents, and is generally transmitted to humans via the bite of an infected flea (Perry & Fetherston, 1997). Plague has resulted in over 200 million deaths over the course of human history (Wkly Epidemiol Rec, (2003). Due to aerosol transmission and the high infectivity, Y. pestis is generally considered to be a great potential biological weapon (Inglesby, et al., 2000, Riedel, 2005). Although the number of confirmed plague cases that occur worldwide has stabilized over the last 50 years to approximately 2000 cases annually, it remains a serious public health threat in some regions of the world and outbreaks still occur (Inglesby, et al., 2000, Riedel, 2005). Furthermore, the recent emergence of multiple antibiotic resistant strains poses potential therapeutic and prophylactic problems (Calhoun & Kwon, 2006). Although several active vaccine candidates are being developed (Leary, et al., 1995, Russell, et al., 1995, Williamson, et al., 1995, Anderson, et al., 1996, Heath, et al., 1998, Titball & Williamson, 2001, Garmory, et al., 2004, Wang, et al., 2004), there is currently no licensed plague vaccine available in the United States (Hart, et al., 2012). Those circumstances result in continuous research on Y. pestis and emphasize the need for development of effective vaccines for plague prevention. Live attenuated Salmonella strains were first developed as vaccines to prevent disease caused by Salmonella infections in humans and animals (Germanier & Fuer, 1975, Stocker, et al., 1983, Hassan & Curtiss, 1990). Subsequently, genetically modified attenuated Salmonella strains were constructed for delivery of heterologous antigens (Formal, et al., 1981, Clements, et al., 1986, Nakayama, et al., 1988). Orally administered Salmonella vaccines offer a variety of advantages over traditional vaccines, including stimulation of both systemic and mucosal responses, which are important for protection against pathogens such as plague, needle-free delivery and a relatively low cost of production (Curtiss, 2002). Two of the Y. pestis protective antigens F1 and V have been extensively tested in live attenuated Salmonella vaccine vectors. F1 antigen, the capsule-like F1 antigen, is unique to Y. pestis as an important protective antigen and appears to give this pathogen a high capability to resist phagocytosis (Du, et al., 2002). LcrV, a multifunctional virulence protein as a key antigen, is required by Y. pestis to resist phagocytosis (Burrows, 1956). Further, researches show that LcrV plays a role involving in translocation of Yops into host cells through the Ysc type III injection system (Cornelis, et al., 1998, Cornelis, 2002). Mice orally immunized with two doses of Salmonella Typhimurium SL3261 synthesizing F1 antigen on the bacterial surface were protected completely against 10⁷ LD₅₀ of virulent Y. pestis GB strain given by the subcutaneous route of challenge, whereas mice immunized with the recombinant Salmonella synthesizing F1 antigen intracellularly were only partially protected (33% survival) against 10⁵ LD₅₀ of Y. pestis (Titball, et al., 1997). Mice intravenously immunized with two dose of S. Typhimurium SL3261 synthesizing F1-V fusion protein were provided 85% protection against 7.4x10⁴ LD₅₀ of Y. pestis GB strain by the subcutaneous challenge (Leary, et al., 1997). Garmory, et al showed that mice orally immunized with five doses of S. Typhimurium SL3261 synthesizing LcrV antigen produces the anti-LcrV specific serum antibody responses but afforded 30% protection against subcutaneous challenge with 97 LD₅₀ of Y. pestis GB strain (Garmory, et al., 2003). Pascual's group demonstrated that mice immunized with S. Typhimurium H683 (Δasd ΔaroA) synthesizing F1 and LcrV antigens simultaneously were afforded similar protective efficacy against ~1000 LD₅₀ Y. pestis Madagascar 105 strain by the subcutaneous (>83% survival) or intranasal (>87.5% survival) challenge (Yang, et al., 2007). But F1-negative Y. pestis strains isolated from a number of different host species and from a human infections (Winter, et al., 1960, 1992) can still cause lethal infections in rats, mice and Africa green monkeys (Williams & Cavanaugh, 1983, Williams & Cavanaugh, 1984, Davis, et al., 1996). Furthermore, recent study indicated that the Δ(caf1) Y. pestis strain which is lack of F1 antigen synthesis was fully virulent in mice but broke through immune responses generated with live attenuated Y. pestis strains or F1 subunit vaccines (Quenee, et al., 2008). Thus F1 antigen may not be indispensable for vaccine combination. Researches also suggested that variations of LcrV in amino acid sequence still retain its function in pathogenicity of Yersinia (Anisimov, et al., 2009, Sun & Curtiss, 2012). Additionally, Branger et al showed that immunization of live attenuated Salmonella strain synthesizing engineered LcrV with five amino acid replacements (LcrV5214) did not provide any protections against bubonic challenge with 450 or 5630 CFU of Y. pestis CO92 in BALB/c mice (Branger, et al., 2010). This result raised a concern that variation of LcrV might affect its protective immunogenicity. Therefore, using only LcrV and/or F1 antigens for presentation by Salmonella might not be sufficient to combat weaponized or naturally occurring Y. pestis, which causes us to evaluate additional potential protective antigens.

In Y. pestis, the pesticin and yersiniabactin receptor (Psn) is the outer membrane receptor for yersiniabactin, a siderophore that is essential for in vivo growth and responsive to iron availability (Fetherston, et al., 1995). The Δpsn Y. pestis strain is avirulent when administered by the subcutaneous route (Bearden, et al., 1997, Fetherston, et al., 2010) and are highly attenuated when administered intranasally (Fetherston, et al., 2010). PsaA in Y. pestis, also called pH 6 antigen which forms a fibrillar structure on the Y. pestis cell surface (Ben-Efraim, et al., 1961). Mice immunized with 40 μg of PsaA along with an aluminum adjuvant exhibited a strong humoral immune response and a significant protection (70%) against a pneumonic plague intranasal infection with the Y. pestis KIM5 (Pgm⁻) strain (Galvan, et al., 2010). The lack of PsaA synthesis in the KIM5 strain reduces its virulence and increases its LD₅₀ at least 100 fold in retroorbitally injected mice (Lindler, et al., 1990). In addition to LcrV, our group has used attenuated S. Typhimurium strains χ8501 and/or χ9558 to vector above Y. pestis antigens, we demonstrated that Salmonella delivering LcrV provided complete protection and Psn elicited significant protective immunity (75% survival) against subcutaneous challenge with ~130 LD₅₀ of Y. pestis CO92, but partial protection against intranasal challenge ~40 LD₅₀ of Y. pestis CO92 (Branger, et al., 2010). PsaA was highly immunogenic, eliciting strong serum IgG and mucosal IgA antibodies. However, immunized mice were not protected from subcutaneous challenge and only partially protected from intranasal challenge without statistical significance (Torres-Escobar, et al., 2010). In all, Salmonella vectors delivering Y. pestis antigens produced protective immune response against plague, but their protective efficacies were not fully up to expectations for human use. Thus, we will improve vaccines based on live attenuated Salmonella from two aspects: 1) delivering multiple antigens (three or more) of Y. pestis simultaneously may provide the theoretical advantage of priming protective immunity against plague compared with single antigen (F1 or LcrV) or double antigens (F1+LcrV); 2) Increasing the safety of live attenuated Salmonella vaccines and retaining their immunogenicity.

Searching for new antigens for vaccines is a continuous endeavor in vaccinology. Straley's group found that Y. pestis YadB and YadC, two new members of the Oca (oligomeric coiled-coil adhesins) family of proteins (Roggenkamp, et al., 2003, Ackermann, et al., 2008), have the ability to form trimers and correlate with invasion of Y. pestis into epithelioid cells (Forman, et al., 2008). Loss of yadBC caused a modest loss of invasiveness for epithelioid cells and a subtle decrease in virulence for bubonic plague but not for pneumonic plague in mice (Forman, et al., 2008, 2013). But immunization with the GST-YadC_137-409 protein, which fused YadC aa 137 to 409 to C terminal of glutathione S-transferase (GST), provided partial protection against F1⁻ Y. pestis challenge in mice and was found to stimulate mixed Th1/Th2 responses (Murphy, et al., 2007). These evidences suggested that YadC might be a promising antigen candidate for vaccines. In this study, we evaluated the protective efficacy of different forms of yadC expressed in a live attenuated Salmonella strain χ11475, a newly improved vaccine delivery vector based on χ9558 (Bollen, et al., 2008, Xin, et al., 2008, Li, et al., 2009, Gunn, et al., 2010), which will enrich our antigen combination.

2. Materials and methods

2.1 Bacterial strains and growth media

Y. pestis strain CO92 was routinely grown in heart infusion broth (HIB) (Difco, Detroit, MI) at 28°C and used for challenge studies. All the strains and plasmids used in this study are listed in Table 1. Escherichia coli strains TOP10 and χ6212 and Salmonella Typhimurium strains were cultured in LB broth at 37°C and supplemented with 50 μg/ml of Diaminopimelic acid (DAP), 0.05% arabinose or 0.1% mannose when necessary for bacterial growth as described in our previous work (Branger, et al., 2010). S. Typhimurium vaccine strain χ11475 used here was derived from χ9558, in which the Δ(gmd-fcl)-26 and ΔagfBAC811 deletions in χ9558 were replaced by the Δ(wza-wcaM)-8 and Δ(agfC-agfG)-999 deletions, respectively. The first replacement blocks colanic acid production and enhances heterologous protein production (Curtiss, et al., 2011, Wang, et al., 2013) and the second eliminates both operons specifying synthesis of Agf fimbriae.

Table 1.

Strains and plasmids used in this study

Strain or Plasmid	Genotype or relevant characteristics	Derivation or source
Strains
E. coli
TOP10	F– mcrA Δ(mrr-hsdRMS-mcrBC) Φ80lacZΔM15 ΔlacX74 recA1 araD139 Δ(ara-leu) 7697 galU galKrpsL (StrR) endA1 nupG	Invitrogen
χ6212	F– X– ϕ80 Δ(lacZYA-argF) endA1 recA1 hsdR17 deoR thi-1 glnV44 gyrA96 relA1 ΔasdA4	(Kang, et al., 2002)
Salmonella
χ11475	Δpmi-2426 Δ(wza-wcaM)-8 ΔPfur81::TT araC PBAD fur ΔPcrp527::TT araC PBAD crp ΔasdA27::TT araC PBAD c2 ΔaraE25 ΔaraBAD23 ΔrelA198::araC PBAD lacI TT ΔsopB1925 Δ(agfC-agfG)-999	χ9558
χ9761	Δ(galE-uvrB)-1005 ΔmsbB48 ΔfliC2426 ΔpefA1225 ΔfimA2119 ΔfimH1019 ΔfljB217 ΔagfBAC811	Lab collection
χ9558	Δpmi-2426 Δ(gmd-fcl)-26 ΔPfur81::TT araC PBAD fur ΔPcrp527::TT araC PBAD crp ΔasdA27::TT araC PBAD c2 ΔaraE25 ΔaraBAD23 ΔrelA198::araC PBAD lacI TT ΔsopB1925 ΔagfBAC811	(Li, et al., 2009, Gunn, et al., 2010)
Y. pestis
CO92 Plasmids
pYA5136	Codon optimized yadB gene cloned into low copy plasmid pUC57	Genescript
pYA5137	Codon optimized yadC gene cloned into low copy plasmid pUC57	Genescript
pYA3342	Asd+; pBR ori	(Kang, et al., 2002)
pYA3620	Asd+; β-lactamase signal sequence-based periplasmic secretion	(Branger, et al., 2007)
pYA5047	pYA3342 yadC (codon optimized)	This study
pYA5048	pYA3620 yadC810 (codon optimized truncated YadC, aa 32-551)	This study
pYA5050	pBAD-HisB-yadC810	This study
pYA5067	pYA3342 yadBC (codon optimized)	This study
E6	Plasmid for protein expression	GenScript
E3	Codon optimized truncated yadC (encode aa 33-622) was subcloned into vector E6 with an N-terminal 6X His tag	GenScript

2.2 Construction of recombinant plasmids encoding yadC, yadC810 and yadBC genes

In order to improve heterologous protein expression by Salmonella, the codon usage of the foreign gene usually is optimized for Salmonella translation preferences (Andrew, 2000). The sequence of the yadB and yadC genes of Y. pestis CO92 (YPO1387 and YPO1388) were codon optimized for maximal expression in Salmonella. (Figure 1 and 2). Codon optimized yadB and yadC genes were synthesized by Genscript (Piscataway, NJ, USA) and cloned into the pUC57 plasmid to form plasmids pYA5136 and pYA5137, respectively. In plasmid pYA5136, the NcoI and NsiI sites were added to the N-terminal and C-terminal ends of the yadB gene fragment, respectively. In plasmid pYA5137, the NcoI and HindIII sites were added to the N-terminal and C-terminal ends of the yadC gene fragment, respectively. The yadC gene fragments were isolated by digesting plasmid pYA5137 using NcoI and HindIII and cloned into plasmid pYA3342 to construct plasmid pYA5047 (Fig. 3A). The yadB gene fragment was isolated by digesting plasmid pYA5136 using NcoI and NsiI and cloned into the same sites of pYA5047 to construct plasmid pYA5067 containing the yadBC operon (Fig. 3B). The E. coli strain χ6212 is a host strain for Asd⁺ plasmid construction. The yadC810 sequence (the truncated yadC gene, encoding a truncated YadC protein between aa 32-551) was amplified with plasmid pYA5047 as template using YadC-primer1: CGGGAATTCAATACGAATATTAATGGCTCCA (EcoRI) and YadC-primer2: CGGCTGCAGGTTTAATTGTTTCTGCTTTTGA (PstI), and cloned into the frame between the signal sequence (Bla_SS) and carboxy-terminal region (Bla_CT) of the TEM-1 β-lactamase at the EcoRI and PstI sites of plasmid pYA3620 to form plasmid pYA5048 (Fig. 3C), encoding a bla SS-yadC810-bla CT fusion. In order to purify YadC protein for immunization, the yadC810 fused to a C-terminal 6×His was amplified from pYA5047 using YadC-primer 3: CGGCCATGGGAAATACGAATATTAATGGCTCCA (NcoI) and YadC-primer 4: CGGAAGCTTTCAGTGATGATGATGATGATGGTTTAATTGTTTCTGCTTTTGATTAATA TC (HindIII) and cloned into the NcoI and HindIII sites of plasmid pBAD-HisB to form pYA5050. The E. coli TOP10 strain is a host strain for pYA5050 construction.

Fig. 1.

Alignment of wild-type yadB with codon optimized yadB.

Fig. 2.

Alignment of wild-type yadC with codon optimized yadC.

Fig. 3.

Construction of plasmids pYA5047, pYA5067 and pYA5048 and synthesis of YadC or YadC810 in live attenuated Salmonella vaccine strains. (A). Asd⁺ vector pYA5047 expressing yadC (encoding full-length YadC aa 1-662) and western blot showing YadC presence in different fractions from χ11475 harboring pYA5047. M, molecular-mass markers are labeled. (B) Asd⁺ secretion vector pYA5067 expressing yadBC (encoding full-length YadBC) and western blot showing YadC synthesis in whole cell lysate, supernatant and outer membrane fraction of χ11475 harboring pYA5067. (C). Asd⁺ vector pYA5048 expressing yadC810 (encoding truncated YadC aa 32-551) and western blot showing YadC810 synthesis in whole cell lysate, cytoplasm, periplasm and supernatant fraction of χ11475 harboring pYA5048.

The DNA sequences of all DNA inserts were confirmed by DNA sequence analysis. The plasmids pYA3620, pYA5047, pYA5048 and pYA5067 were electroporated into S. Typhimurium strain χ11475. Plasmid stability tests in strain χ11475 were performed as previously described (Kang, et al., 2002).

2.3 Protein expression and purification

E. coli TOP10 carrying pYA5050 (yadC810-6xHis) was grown overnight at 37°C in LB broth supplemented with 100 μg/ml ampicillin. The procedures for protein expression and purification were described in our previous study (Sun & Curtiss, 2012). Briefly, bacteria were grown at 37°C in a 2 L flask (agitation at 200 rpm) to an OD600 of 0.9 and then induced to production of the YadC810-6xHis protein for 3 hours through adding 0.1% arabinose. Bacteria were harvested by centrifugation at 6,000 × g for 10 min, resuspended in 50 ml of 50 mM sodium phosphate buffer (pH 8.0) containing 300 mM NaCl, and broken using ultra-sonication on ice. The extract was centrifuged at 12,000 × g for 20 min, and the soluble fraction was applied to a nickel nitrilotriacetic acid (Ni-NTA) column (Sigma). Purified YadC810-6xHis protein isolated from the nickel column was used for immunization. The C-terminal fused protein YadC810-6xHis tag protein was purified and used for immunization. In addition, we purchased the recombinant YadC protein (from aa 33-622) with an N-terminal 6xHis tag (designated as YadC-T) from GenScript for ELISA assays. The codon optimized truncated yadC (encode aa 33-622) was subcloned into vector E6 with an N-terminal 6xHis tag by GenScript to form plasmid E3. The YadC-T protein was obtained from inclusion bodies and one-step purification by Ni-NTA column.

2.4 SDS-PAGE and western blot analysis

S. Typhimurium strain χ11475 harboring pYA5047 (yadC), pYA5048 (yadC810), pYA5067 (yadBC) or empty plasmids were cultured in LB medium containing 0.05% arabinose and 0.1% mannose at 37°C. When the bacteria reached to an optical density at 600 nm (OD₆₀₀) of 0.8 (about 5 × 10⁸ CFU/ml), 1 mM IPTG was added into the cultures to induce heterologous protein synthesis. Protein samples from bacterial pellets and culture supernatants were separated and analyzed by SDS-polyacrylamide gel electrophoresis with transfer to nitrocellulose membranes as previously described (Branger, et al., 2010). The membranes were blocked with 3% skim milk in PBS and incubated with rabbit polyclonal anti-YadC antibodies for 1 h at room temperature. Then, the membranes were washed with 1X PBS/0.1% Tween-20 three times. Then, horseradish peroxidase-conjugated goat anti-rabbit immunoglobulin G (IgG) (Sigma) was added in 1X PBS/3% skim milk. Immunoreactive bands were detected by using a chemiluminescent detection system (ECL; Pierce, Rockford, IL). The reaction was captured by exposure to X-ray film for 1 minute.

2.5 Salmonella subcellular fractionation

Periplasmic fractions were prepared by a modification of the lysozyme-osmotic shock method (Dvorak, et al., 1970, Witholt, et al., 1976) as previously described (Kang, et al., 2002). S. Typhimurium strains χ11475 harboring pYA5047 (yadC), pYA5048 (yadC810), pYA5067 (yadBC) or empty plasmids were grown in LB broth to an OD₆₀₀ of 0.6 and centrifuged. The supernatant was separated, precipitated with 10% TCA and then resuspended in 0.1M Tris buffer for analysis of secreted proteins. The samples of whole cell lysates, periplasmic, cytoplasmic, and supernatant fractions with equal protein concentrations were separated by SDS-PAGE for western blot analysis. Salmonella outer membrane proteins (SOMPs) were prepared from χ9761 cells grown in LB broth without galactose for enzyme-linked immunosorbent assay (ELISA) assay as described previously (Kang, et al., 2002). The use of SOMPs obtained from χ9761 precludes LPS O-antigen contamination. Briefly, bacterial pellets from a 30 ml culture were suspended in 3 ml of 20 mM Tris-HCl (pH 8.6) containing 1% Sarkosyl and incubated for 30 min on ice. The outer membrane fraction was obtained as a pellet after centrifugation at 132,000 × g at 4°C for 1 h. The pellet was resuspended in 3 ml of 20 mM Tris-HCl buffer (pH 8.6).

2.6 Immunization of mice

All animal experiments were approved by the Arizona State University animal care and use committee. Female BALB/c mice, 7 weeks of age, were purchased from Charles River Laboratories (Wilmington, MA). Immunization procedures followed the previous description (Branger, et al., 2010). Briefly, mice were deprived of food and water for 4 h prior to immunization and re-supplied 30 min after oral immunization. S. Typhimurium vaccine strains were grown in LB broth with appropriate supplements at 37°C to an OD₆₀₀ of 0.9 and concentrated to a final concentration of 5 × 10¹⁰ CFU/ml in buffered saline with 0.01% gelatin (BSG). Groups of mice were orally immunized with 20 μl of RASV suspensions on days 0 and 10. Blood samples were collected on days 0, 14 and 28. In addition, one group of mice received two subcutaneous injections with 20 μg of YadC810-6xHis emulsified with Alum adjuvant (Thermo Scientific, West Palm Beach, FL, USA) in 100μl at two-week intervals.

2.7 Enzyme-linked immunosorbent assay (ELISA)

Nunc Immunoplate Maxisorb F96 plates (Nalge Nunc. Rochester, NY, USA) were coated with purified recombinant YadC (100 ng/well), or SOMPs (100 ng/well) and incubated overnight at 4°C. The procedures for measuring antibody titer were followed as our previous report (Branger, et al., 2010). Briefly, the plates were washed three times with PBS containing 0.1% Tween 20 (PBST), blocked with 1 × SEA BLOCK Blocking buffer (Pierce, Rockford, IL, USA) for 1 h at 37°C and then washed three times with PBST. The sera from mice were diluted in PBS and added to the plates for overnight incubation at 4°C. The plates were washed three times with PBST, followed by the addition of 100 μl of goat anti-mouse IgG(H+L), IgG1, IgG2a or IgA biotin conjugated antibodies (Southern Biotech) (1:10,000) and incubated 1 h at 37°C. The plates were washed again and streptavidin conjugated with alkaline phosphatase (Southern Biotech) (1:3000) was added. The plates were incubated for 1 h at 37°C, washed five times. The pnitrophenyl phosphate chromogenic substrate (Sigma) was added. After 30 min, the reaction was stopped by the addition of 3M NaOH. The optical density at 405 nm was measured using the automated ELISA plate reader (EL311SX, Biotech, Winooski, VT).

2.8 Challenge with Y. pestis

Challenge was performed 4 weeks after the second immunization (day 38). For subcutaneous challenge, each animal received a dose of approximately ~10³ CFU of Y. pestis CO92 freshly grown at 28°C in heart infusion broth (HIB) containing 0.2 % xylose. For intranasal challenge, mice were anesthetized with a cocktail of ketamine/xylazine. Then, each mouse received a dose of ~10³ CFU of Y. pestis CO92 freshly grown at 37°C in HIB containing 0.2% xylose and 2mM CaCl₂ administered intranasally. Mice were observed daily, and mortality was recorded for 15 days after the challenge. The LD₅₀s of virulent Y. pestis CO92 are around 10 CFU and 100 CFU by s.c. and i.n. administration, respectively.

2.9 Statistical analysis

Antibody responses were compared using two-way ANOVA. The log-rank (Mantel-Cox) test was used for analysis of the survival curves. Data are expressed as means ± SD. P-value <0.05 was considered significant.

3. Results

3.1 Synthesis of YadC and YadC810 in χ11475

YadC (the predicted mass, 65.3 kDa) is an outer membrane protein in Y. pestis (Forman, et al., 2008). We evaluated outer membrane fractions of χ11475(pYA5047) (yadC) and χ11475(pYA5067) (yadBC) to determine whether YadC was also able to insert into the Salmonella outer membrane. Outer membrane fractions from both RASVs contained protein that reacted with the anti-YadC antibody (Fig. 3A and B), indicating that the Y. pestis secretion and membrane insertion signals in the wild-type YadC can function in Salmonella. Strains carrying the empty vector plasmids did not produce any protein that reacted with anti-YadC antibody at a size around 56-72 kDa (Fig. 3A and B). But in whole cell lysates and outer membrane fraction, YadC expression level in χ11475(pYA5047) (yadC) seemed to be higher than that of χ11475(pYA5067) (yadBC). In addition, YadC was detected in the supernatant fraction of both χ11475(pYA5047) (yadC) and χ11475(pYA5067) (yadBC) (Fig. 3A and B). It is possible that over-expression of yadC in χ11475(pYA5047) and χ11475(pYA5067) may cause damage of bacterial membrane to release of YadC into supernatants. In addition, our results did not show the band of a trimer of YadC (186 kDa) like Forman et al observed. (Forman, et al., 2008).

Previous studies in our group have shown that the PspA of S. pneumoniae fused to a β-lactamase signal sequence delivered by an attenuated Salmonella strain can be transferred to the periplasmic space and subsequently released to the outside medium. Immunization of the Salmonella vaccine strain synthesizing β-lactamase signal peptide fused with PspA elicited higher PspA-specific immune responses than expression of the protein without a signal sequence and finally afforded improving protection for mice against virulent pneumococcal challenge (Kang, et al., 2002, Kang & Curtiss, 2003). So we fused YadC to β-lactamase signal peptides delivered by Salmonella to enhance its immunogenicity against plague. Protein structure analysis indicated that the YadC protein had two transmembrane regions in its N- (1-31aa) and C-terminal ends (552-622aa). Forman et al suggested that YadC is new member of the Oca family of adhesin, but its main role is unclear up to now (Forman, et al., 2008). The outer membrane part of YadC might have main immunogenicity. Thus, we removed N- (1-31aa) and C- (552-622aa) terminal transmembrane regions from yadC to form the yadC810 and cloned yadC810 into pYA3620 to construct pYA5048. Plasmid pYA5048 encodes fusions of yadC810 to the DNA sequence encoding amino-terminal and carboxy-terminal secretion signals from β-lactamase that direct the protein products to the periplasm and supernatant (Koshland & Botstein, 1980, Summers & Knowles, 1989). The predicted mass of YadC810 (32-551 aa) is 54.4 kDa. The supernatant and periplasm fractions from χ11475(pYA5048) contained more proteins that reacted with the anti-YadC antibody than that of cytoplasm fraction (Fig. 3C), indicating that the β-lactamase signal sequence could direct antigen secretion as expected. Supernatant fractions from strains carrying the empty vector plasmid pYA3620 did not react with the anti-YadC antibody at a size around 54.4 kDa (Fig. 3C).

3.2 Anti-YadC and anti-Salmonella OMPs responses in immunized mice

The IgG responses in immunized mice against the recombinant truncated YadC (33-662aa) protein purchased from Genscript were determined by ELISA (Fig. 4A). Comparing with control strain χ11475(pYA3620), mice subcutaneously immunized with YadC810 protein emulsified with Alum adjuvant or orally immunized with strains expressing either yadC, yadC810 or yadBC developed significant antibody titers against YadC-T (truncated YadC (33-662aa)) compared with mice immunized BSG at weeks 2 and 4. The YadC810 protein primed highest IgG titers (**, p < 0.01). The χ11475(pYA5047) (yadC) and χ11475(pYA5048) (yadC810) elicited higher levels of anti-YadC titers at 4 weeks than χ11475(pYA5067) (yadBC) did (*, p < 0.05). The antibody responses to Salmonella OMPs were also measured. Mice orally immunized with χ11475(pYA3620) (empty vector), χ11475(pYA5047) (yadC), χ11475(pYA5048) (yadC810) and χ11475(pYA5067) (yadBC) increasingly developed higher levels of anti-OMPs titers than mice immunized with protein YadC810 (**, p < 0.01) at weeks 2 and 4 (Fig. 4B).

Fig. 4.

Antibody responses to recombinant truncated YadC (33-622 aa) with an N-terminal 6xHis tag (YadC-T) and SOMPs determined by ELISA. The data represent IgG antibody levels induced in groups of 12 mice orally immunized with BSG, χ11475(pYA3620), χ11475(pYA5047), χ11475(pYA5048), χ11475(pYA5067) or YadC810 (32-551 aa) at the indicated weeks after immunization. (A) Total IgG responses to YadC-T; compared to χ11475(pYA3620)-immunized group. (B) Total IgG responses to SOMPs; compared to BSG (buffered saline with 0.01% gelatin) and YadC810-immunized groups. (C) Vaginal-wash IgA responses to recombinant YadC-T. compared to YadC810 and χ11475(pYA3620)-immunized groups. **, P < 0.01. The sera from 12 mice were individually analyzed and the experiments were performed twice with consistent results.

IgA antibody titers in vaginal washes to recombinant YadC-T were significantly higher in χ11475(pYA5047) (yadC), χ11475(pYA5048) (yadC810) or χ11475(pYA5067) (yadBC) -immunized mice than those in χ11475(pYA3620), BSG or YadC810 protein-immunized mice at 2 and 4 weeks (**, P<0.01) (Fig. 4C). Results also indicated that immunization with injectable YadC810 does not induce sIgA whereas delivery by orally administered RASVs does.

The levels of IgG isotype subclasses IgG1 and IgG2a were further measured. The Th1 cells direct cell mediated immunity and promote IgG class switching to IgG2a, and Th2 cells provide potent help for B-cell antibody production and promote IgG class switching to IgG1 (Gor, et al., 2003). We measured IgG1 and IgG2a against YadC in sera at 2 and 4 weeks post-immunization. All the titers of IgG1 and IgG2a to YadC were increased at 2 and 4 weeks post-immunization by YadC810 protein, χ11475(pYA5047) (yadC), χ11475(pYA5048) (yadC810) or χ11475(pYA5067) (yadBC) (Fig. 5). The levels of anti-YadC-T IgG1 isotype antibodies elicited by recombinant YadC810 protein immunization were significantly higher than the levels of IgG2a (**, P < 0.01) (Fig. 5A). The levels of IgG1 and IgG2a induced by immunization with χ11475(pYA5047), χ11475(pYA5048) or χ11475(pYA5067) were not significantly different at 2 weeks post-immunization (Fig. 5 B, C and D), while the levels of anti-YadC-T IgG2a isotype antibodies elicited by χ11475(pYA5047) were significantly higher than the levels of IgG1 at 4 weeks post-immunization (**, P < 0.01) (Fig. 5B). Immunization with χ11475(pYA5048) and χ11475(pYA5067) produced similar levels of IgG1 and IgG2a at 4 weeks (Fig. 5 C and D). These data indicate that the RASV synthesizing YadC induced a balanced or slightly Th1-biased response compared to the YadC810 protein vaccine, which induced significantly Th2-biased immune response.

Fig. 5.

Serum IgG1 and IgG2a responses to recombinant YadC-T. (A) IgG1 and IgG2a antibody levels to recombinant YadC-T in sera of mice subcutaneously immunized with YadC810; (B), (C) and (D) IgG1 and IgG2a antibody levels to recombinant YadC-T in sera of mice orally immunized with χ11475(pYA5047), χ11475(pYA5048) and χ11475(pYA5067), respectively. The data represent IgG antibody levels induced in groups of 12 mice orally immunized with YadC810, χ11475(pYA5047), χ11475(pYA5048) or χ11475(pYA5067) at the indicated weeks after immunization. **, P < 0.01. The sera from 12 mice were individually analyzed and the experiments were performed twice with consistent results.

3.3 Protection against Y. pestis challenge in mice immunized with χ11475 derivatives

To test our constructs delivered by strain χ11475, we immunized 3 groups of 20 mice per group with either χ11475(pYA5047) (yadC), χ11475(pYA5048) (yadC810) or χ11475(pYA5067) (yadBC) and 2 groups of 10 mice per group with χ11475(pYA3620) (control) or recombinant YadC810 protein emulsified with Alum adjuvant using the same immunization schedule at day 0 with a boost at day 10. In challenge experiments, we used 2300 CFU for subcutaneous challenge and 3100 CFU for intranasal challenge. Mortality was recorded for 15 days after challenge.

The challenge results are shown in Figure 6. Mice orally immunized with RASV strains χ11475(pYA5047) (yadC) or χ11475(pYA5048) (yadC810) were significantly protected against subcutaneous challenge with ~230 LD₅₀ of Y. pestis CO92 (P<0.01) (Fig. 6A), but such vaccination failed to protect against intranasal challenge with ~31 LD₅₀ of Y. pestis CO92 (Fig. 6B). All mice immunized with the YadC810 protein or orally immunized with the control strain χ9558(pYA3620) succumbed to both s.c. and i.n. challenge with the CO92 virulent strain. Unexpectedly, none of the mice orally immunized with χ11475(pYA5067) (yadBC) survived subcutaneous challenge.

Fig. 6.

Protective efficacy of oral immunization of BALB/c mice with Salmonella YadC synthesizing vaccines against Y. pestis challenge. Groups of 20 BALB/c mice were orally immunized with 1 × 10⁹ CFU of serovar Typhimurium strain χ11475 harboring different yadC expression plasmids. Mice were challenged s.c. or i.n. with 2300 CFU or 3100 CFU of virulent Y. pestis CO92, respectively, at 4 weeks after second immunization. Mortality was monitored for 15 days after challenge. (A) Mice were challenged s.c. with 2300 CFU of virulent Y. pestis CO92 (~230 LD₅₀). The χ11475(pYA5047) and χ11475(pYA5048) vaccination groups were significantly different from the pYA3620 controls (**, P < 0.01). (B) Mice were challenged i.n. with 3100 CFU of virulent Y. pestis CO92 (~31 LD₅₀). There were no significant differences between groups (P > 0.05). The experiments were performed twice.

4. Discussion

As two new members of the Oca family of proteins, YadB and YadC in Y. pestis have the ability to form trimers and increase invasion of epithelioid cells. The yadBC is unique to Y. pestis and Y. pseudotuberculosis, which implicates this operon is important in the more highly disseminated character of these yersiniae compared to Y. enterocolitica, and acquisition of this operon could represent one important step in the evolution of Y. pestis as a flea-borne pathogen. Forman et al. indicated that the virulence of the ΔyadBC Y. pestis is the same as the parent Y. pestis strain in pneumonic plague, whether F1 is present or not. When the subcutaneous route was used, the ΔyadBC mutant and ΔyadC single mutant were only slightly attenuated (Forman, et al., 2008, 2013). Murphy et al. also reported that mice immunized with a GST-YadC_137-409 fusion protein were afforded partial protections against 10,000 CFU of Y. pestis CO99-3015.S7 (5 LD₅₀) by the intravenous route challenge or 3,000 CFU of Y. pestis CO99-3015.S9 (6 LD₅₀) by the intranasal route challenge (Murphy, et al., 2007), but no data about protective efficacy against subcutaneous challenge was provided. Our results showed that YadC810 protein immunization could not provide any protection against subcutaneous and intranasal challenge of virulent Y. pestis CO92. The explanations for this contrary result: 1) the higher challenge dose we used; 2) the YadC protein (aa 32-551) used for immunization in our studies was different with the YadC protein (aa 137-422) used by Murphy et al (Murphy, et al., 2007). The variation of amino acid sequences might change the configuration of the YadC protein in the two cases. Our data showed that YadC810, χ11475(pYA5047) (yadC), χ11475(pYA5048) (yadC810) or χ11475(pYA5067) (yadBC) all elicited similar levels of anti-YadC IgG2a, also YadC810 induced a higher titer of IgG1 than the live Salmonella vaccines did (Fig. 5). Researches indicated that Alum adjuvant provokes a strong Th2 response, but is rather ineffective against pathogens that require Th1-cell-mediated immunity (Mannhalter, et al., 1985, Kool, et al., 2008). Immunization with YadC810 protein emulsified with Alum adjuvant induced significantly Th2-biased immune response (Fig. 3 A). Thus, the humoral immune response bias elicited by protein immunization might not provide enough protection against plague challenge using high doses of virulent Y. pestis. The reasons are uncertain about why the immunization of YadC810 protein primed a similar level of IgG2a as χ11475(pYA5047) (yadC) or χ11475(pYA5048) (yadC810) did, but could not provide similar protection as χ11475(pYA5047) (yadC) or χ11475(pYA5048) (yadC810) did.

Unlike Alum adjuvant, live Salmonella vectors with surface-exposed antigens such as LPS and flagella which are potent inducers of innate immunity and then adaptive immunity (Neutra & Kozlowski, 2006). As previous studies shown, live attenuated Salmonella as vectors often tuned the immune responses to Th1 bias (Pascual, et al., 1999, Pashine, et al., 1999, Li, et al., 2008, Xin, et al., 2008) or a balanced Th1/Th2 mixture (Li, et al., 2009, Shi, et al., 2010). Our results also showed that χ11475(pYA5047) (yadC), χ11475(pYA5048) (yadC810) or χ11475(pYA5067) (yadBC) all elicited a balanced Th1/Th2 mixture immune response to YadC-T (Fig. 5 B, C, D). Immunization by recombinant Salmonella delivering YadC and YadC810 provides partial protection against subcutaneous challenge (Fig. 6 A), while immunization with χ11475 delivering YadB and YadC simultaneously cannot provide any protection against either bubonic or pneumonic plagues. Currently, the reasons about why immunization with χ11475(pYA5067) simultaneously delivering YadB and YadC fail to provide protection are unclear.

Our long-term goal is to develop a live Salmonella vaccine strain expressing multiple Yersinia antigens to protect against plague. We have demonstrated that a single attenuated S. Typhimurium strain expressing truncated lcrV or psn can protect mice against challenge with virulent Y. pestis (Branger, et al., 2007, Branger, et al., 2010). Therefore, the results reported here will provide basis for combination of multiple antigens (including LcrV, Psn or YadC) to develop more efficacious vaccines against plague.