ABSTRACT
In developing countries, Shigella is a primary cause of diarrhea in infants and young children. Although antibiotic therapy is an effective treatment for shigellosis, therapeutic options are narrowing due to the emergence of antibiotic resistance. Thus, preventive vaccination could become the most efficacious approach for controlling shigellosis. We have identified several conserved protein antigens that are shared by multiple Shigella serotypes and species. Among these, one antigen induced cross-protection against experimental shigellosis, and we have named it pan-Shigella surface protein 1 (PSSP-1). PSSP-1-induced protection requires a mucosal administration route and coadministration of an adjuvant. When PSSP-1 was administered intranasally, it induced cross-protection against Shigella flexneri serotypes 2a, 5a, and 6, Shigella boydii, Shigella sonnei, and Shigella dysenteriae serotype 1. Intradermally administered PSSP-1 induced strong serum antibody responses but failed to induce protection in the mouse lung pneumonia model. In contrast, intranasal administration elicited efficient local and systemic antibody responses and production of interleukin 17A and gamma interferon. Interestingly, blood samples from patients with recent-onset shigellosis showed variable but significant mucosal antibody responses to other conserved Shigella protein antigens but not to PSSP-1. We suggest that PSSP-1 is a promising antigen for a broadly protective vaccine against Shigella.
INTRODUCTION
Shigella is one of the major causes of diarrheal disease in infants and young children in developing countries (1). Approximately 1.1 million people die as a result of Shigella infections each year, and 60% of the deaths involve children under 5 years of age (2). In developed countries, 500,000 cases of shigellosis are reported each year by military personnel and travelers (3). The causative agents are four facultative intracellular Gram-negative Shigella species, i.e., Shigella flexneri, Shigella dysenteriae, Shigella sonnei, and Shigella boydii. More than 40 serotypes, as specified by the composition of the surface polysaccharide O antigen, have been identified.
Antibiotics can effectively treat shigellosis but, because of the emergence of antibiotic resistance, the World Health Organization has made the development of an effective Shigella vaccine a top priority. Population-based studies conducted by the International Vaccine Institute (IVI) in six Asian countries indicate that, while S. flexneri is the most commonly isolated species (except in Thailand), levels of different S. flexneri serotypes vary greatly among sites and even from year to year at a given outbreak site (4). Because a Shigella vaccine must include multiple serotypes and species, we tried to identify common Shigella antigens that could yield protection across species and serotypes. Using comparative analyses of whole-genome and virulence plasmid sequences, including publically available sequences from different species of Shigella, we recently identified a surface-expressed protein common to all or most species, termed pan-Shigella surface protein 1 (PSSP-1).
PSSP-1 is a C-terminal half-polypeptide of IcsP (5) that is expressed by more than 300 Shigella isolates (representing all species and serotypes) collected from field sites in Asia. It is not present among common intestinal commensal organisms. IcsP is an outer membrane protein, of the enterobacterial omptin family of proteases, that cleaves IcsA (VirG) and is involved in Shigella virulence (6–8). Shigella actin-based motility correlates with the amount of IcsA α-domain on the bacterial cell surface, which is modulated by IcsP through cleavage of IcsA between Arg758 and Arg759, removing the entire IcsA α-domain from the surface of the bacterium (9).
In this study, PSSP-1, when added to cholera toxin (CT) (10) or the double mutant (R192G/L211A) of heat-labile toxin (dmLT) of Escherichia coli (11), showed cross-protection against multiple Shigella species and serotypes when tested in a mouse pneumonia model. Clinical data from shigellosis patients showed rare antibody responses to PSSP-1. The natural hierarchical level of PSSP-1 seems to be lower than that of the other Shigella Ipa proteins included in this study. Based on data in this study, PSSP-1 may be a relevant antigen for a Shigella vaccine.
MATERIALS AND METHODS
Bacterial strains.S. flexneri serotype 2a strain 2457T (12), serotype 5a strain M90T (13), and serotype 6 (14), S. boydii (4), S. dysenteriae serotype 1 strain 10398 (15), and S. flexneri serotype 2a vaccine strain SC602 (16) were used in this study. For intranasal (i.n.) immunization or challenge experiments, an aliquot of bacteria stored at −80°C was thawed, streaked on a Bacto tryptic soy broth (BD, Sparks, MD) agar plate containing Congo red (Serva, Heidelberg, Germany), and incubated overnight at 37°C. One Congo red-binding colony was picked from the plate and cultured in Bacto tryptic soy broth at 37°C, with shaking, until the subculture reached an optical density at 600 nm (OD600) of ∼0.5 (∼2 × 108 CFU/ml). The bacteria were diluted to the optimal concentrations with phosphate-buffered saline (PBS). We confirmed the CFU values by serially diluting each bacterial challenge solution in PBS and then spreading appropriate dilutions on agar plates containing Congo red.
Cloning and purification of PSSP-1.A PCR-amplified fragment carrying PSSP-1 (IcsP amino acids 171 to 300; GenBank accession no. AAP78966.1) (17) was cloned between the EcoRI and XhoI sites of pET21-d and pET24-d (Novagen, Madison, WI) by using S. flexneri serotype 2a strain 2457T as the template. Primers used for PCR were as follows: forward, 5′-CCGGAATTCGGAGTGAAAACGGGGGGAGC-3′; reverse, 5′-CGGCGGCTCGAGCTAGTGGTGGTGGTGGTGGTGAATACTTGCACTATTTTT-3′. The bases that are underlined are the IcsP sequences. Recombinant PSSP-1 was expressed in E. coli BL21(DE3) and purified from inclusion bodies as His-tagged protein. The protein was purified using nickel-nitrilotriacetic acid (NTA) His·Bind resin (Novagen), according to the manufacturer's recommendations, and was tested for endotoxin (18). Endotoxin levels of the proteins were less than 100 endotoxin units (EU)/mg. The expressed PSSP-1 polypeptide was 14.8 kDa.
Construction of ΔicsP Shigella strain.The ΔicsP strain was constructed by disruption of the icsP gene through insertion of a suicide plasmid, as described previously (19). In brief, a 600-bp internal fragment of icsP was amplified using PCR with the following primers: forward, 5′-GCCGGCGAGCTCACTAACTATCCACTTTTCATA-3′; reverse, 5′-GCCGGCGTCGACACCTATTGTTCTTTCACCACT-3′. The resulting PCR product was inserted into the chloramphenicol resistance suicide plasmid pSW23OriT (20). The recombinant plasmid pSWicspTr was transferred by conjugation to S. flexneri serotype 2a strain 2457T (19).
Immunization and infection of mice.We used 6-week-old female BALB/c mice purchased from Charles River Laboratories (Orient Bio, Seongnam, South Korea). Mice were kept under specific-pathogen-free conditions at the IVI, and all animal experiments were performed with approval of the IVI Institutional Animal Care and Use Committee (protocol no. 2010-011). Mice were anesthetized by intramuscular injections with ketamine hydrochloride (Yuhan Co., Ltd., Seoul, South Korea) (0.1 mg/g of body weight) combined with xylazine hydrochloride (Rompun; Bayer Korea, Seoul, South Korea) (12.5 μg/g of body weight) and were immunized i.n. three times, at 2-week intervals, with PSSP-1 (20 μg) plus adjuvant, CT (2 μg), or dmLT (5 μg) (11). The total volume for intranasal immunization was 20 μl. One week after the last immunization, the mice were inoculated i.n. with different doses of bacteria as indicated. Survival and body weights were observed daily.
Enzyme-linked immunosorbent assay.PSSP-1-specific antibody titers were determined by an enzyme-linked immunosorbent assay (ELISA) using serum or bronchoalveolar lavage fluid (BALF) from each mouse, as described previously (18, 21). In 96-well plates, the coating concentration of PSSP-1 in each well was 200 ng/100 μl. The endpoint titer was expressed as the reciprocal log2 of dilutions showing values of 0.2 at a wavelength of 450 nm.
Enzyme-linked immunosorbent spot assay.On day 7 after the last PSSP-1 immunization, we obtained spleens and lungs from the mice. Single-cell suspensions were prepared as described previously (18). We coated 96-well nitrocellulose microplates (Millipore, Bedford, MA) with purified recombinant PSSP-1 (30 μg/ml) in PBS and performed an enzyme-linked immunosorbent spot (ELISPOT) assay as described previously (18). PSSP-1-specific IgG or IgA spots were developed with the addition of 3-amino-9-ethylcarbazole (AEC)–H2O2 chromogenic substrate (Sigma-Aldrich) and were counted with an ImmunoSpot analyzer (Cellular Technology, Cleveland, OH).
Detection of cytokine production.Single-cell suspensions from spleens or lymph nodes were cultured with PSSP-1 peptides, as described above, for 72 h at 37°C in an incubator with 5% CO2. The culture supernatants were stored at −80°C until cytokine analysis was performed. Cytokine levels in the supernatants were assessed by using the mouse Th1/Th2/Th17 cytokine standards of the BD cytometric bead array (CBA) kit (BD Biosciences, San Jose, CA), according to the manufacturer's recommendations.
Detection of antibody-secreting cells by ELISPOT assay in blood samples from patients with shigellosis.The study population for this analysis included 34 adult patients with diarrhea who were admitted to the Infectious Diseases and Beliaghata General Hospital (Kolkata, India); all had ≥3 stool emissions per day, with or without blood. Informed consent was obtained according to the protocol of the Institutional Ethical Committee. Enrolled subjects were monitored for 1 month, and blood specimens were collected at different times. Stool specimens were collected immediately after hospitalization, and the first blood samples were obtained at the preliminary study examination. The second blood specimens were obtained 5 to 6 days later. History of diarrheal illness was recorded for all enrolled subjects. All subjects received oral rehydration therapy and antimicrobial treatment, in accordance with the current guidelines of the Infectious Diseases and Beliaghata General Hospital.
Human whole-blood ELISPOT assay.We used the ELISPOT assay with whole blood, as described previously (22), to detect the numbers of antibody-secreting cells (ASCs) (IgA or IgG) per milliliter of blood in response to PSSP-1, IpaC (23), and IpaD (24).
Statistical analysis.Data are expressed as mean ± standard deviation (SD). Statistical significance between the individual groups was analyzed using the unpaired Student t test, with a threshold of P values of <0.05. A log rank (Mantel-Cox) test was used for comparison of survival rates between groups.
RESULTS
Selection of conserved Shigella-specific protein antigens based on protection against Shigella challenges.To identify conserved Shigella-specific protein antigens that could be used in a Shigella vaccine to target a broad range of Shigella species and serotypes, we performed a comparative analysis of published genomic sequences of various Shigella species and virulence plasmids using the PubMed BLAST search engine. We selected 13 different genes, encoding Shigella-specific surface antigen, motility-associated, or secreted or secretion-related proteins (see Table S1 in the supplemental material). Each was cloned by PCR into a His-tagged fusion protein expression vector by using genomic DNA from S. flexneri serotype 2a strain 2457T (12) as the template. Most recombinant proteins were expressed in E. coli and purified. For proteins that were difficult to purify from the E. coli expression system, truncated forms of the proteins were cloned instead and purified. Each recombinant protein was administered intranasally (i.n.) to mice three times, at 2-week intervals, using CT as an adjuvant. A homologous challenge study with S. flexneri serotype 2a strain 2457T was carried out, and we found that polypeptides corresponding to a region within IcsP (GenBank accession no. AAP78966.1) or SigA (GenBank accession no. AAP18272.1) could generate homologous protection upon S. flexneri serotype 2a strain 2457T challenge. We also found that mice immunized with recombinant IcsP polypeptide were protected against heterologous challenge with S. sonnei (see Table S1 in the supplemental material). We named the IcsP polypeptide PSSP-1. The sequence homology of PSSP-1 has been confirmed for more than 300 Shigella isolates, representing all four species and serotypes collected from field sites in Asia. PSSP-1 is not present in common intestinal commensal bacteria. PSSP-1 is 14.8 kDa and includes the sequence of IcsP, an outer membrane serine protease involved in bacterial motility (25) (Fig. 1).
Cloning and purification of PSSP-1. (A) Amino acid sequence of PSSP-1. Genomic DNA of S. flexneri serotype 2a strain 2457T was used as a template, and a PCR-amplified fragment carrying PSSP-1 (amino acids 171 to 300 of polypeptide IcsP [GenBank accession no. AAP78966.1]) (underlined sequence) was cloned into a protein expression vector. C-terminal His-tagged PSSP-1 was expressed in E. coli BL21(DE3) and purified from inclusion bodies. (B) SDS-PAGE result for purified recombinant PSSP-1. Purified PSSP-1 (10 μg) (right lane) was subjected to SDS-PAGE before staining with Coomassie blue. Left lane, molecular mass standards.
PSSP-1-induced homologous and heterologous protection against Shigella challenges.We tested protective efficacy conferred by PSSP-1 immunization against more species and serotypes of Shigella by using a mouse pneumonia model (26). While immunization with S. flexneri serotype 2a vaccine strain SC602 (16) generated homologous protection only against S. flexneri serotype 2a, i.n. PSSP-1 immunization with CT protected the animals against a number of Shigella species and serotypes, including S. flexneri serotypes 2a (12) and 5a (13), S. dysenteriae serotype 1 (15), and S. boydii (4), although the survival rates were less than 100% (Table 1). Without adjuvant, PSSP-1 alone did not protect mice (data not shown). To determine whether the cross-protection was totally due to PSSP-1, PSSP-1-immunized mice were challenged with the icsP gene-disrupted S. flexneri serotype 2a mutant strain (ΔicsP strain). None of the PSSP-1-immunized mice survived the ΔicsP strain challenge, while all SC602-immunized mice were completely protected (Table 1). These results suggest that PSSP-1 plays a critical role in protection against Shigella infection.
Protective efficacy of PSSP-1 in mouse pneumonia modela
We next evaluated the protective efficacy generated by PSSP-1 immunization when PSSP-1 was administered with dmLT instead of CT, since dmLT is known to be safer (27). Intranasal immunization with 20 μg of PSSP-1 plus 5 μg of dmLT as adjuvant also protected mice against heterologous species and serotypes of Shigella, including S. flexneri serotype 2a strain 2457T, S. flexneri serotype 6, and S. dysenteriae serotype 1 (Fig. 2), suggesting that dmLT, like CT, is an adjuvant that effectively promotes the induction of PSSP-1-specific immunity.
Survival curves after PSSP-1 immunization, showing protection of mice against homologous and heterologous Shigella challenges. Mice were immunized intranasally (i.n.) with PSSP-1 (20 μg) plus dmLT 3 times, at 2-week intervals, and challenged i.n. with virulent S. flexneri serotype 2a strain 2457T, S. flexneri serotype 6, or S. dysenteriae serotype 1 strain 10398 (all at 107 CFU). Survival was observed daily (n = 10 to 12 mice per group except that n = 5 for all subgroups in the S. flexneri serotype 2a strain 2457T challenge and the dmLT-alone and naive groups in the S. flexneri serotype 6 challenge). Results are representative of three independent experiments. *, statistically significant difference for PSSP-1 plus dmLT versus naive or dmLT alone.
PSSP-1-induced humoral and cellular immune responses in the presence of adjuvant (CT or dmLT).Next, we examined the immunogenicity of PSSP-1 in mice. Each mouse was immunized i.n. three times, at 2-week intervals, with PSSP-1 plus CT or dmLT as adjuvant. Following immunization, we examined the PSSP-1-specific antibody responses in the sera. The IgG responses were comparable in the two adjuvant groups. When the levels of IgG2a and IgG1 (two subclasses of IgG, representing Th1 and Th2 CD4+ T cell responses, respectively) were measured, PSSP-1-specific IgG1 levels were higher than IgG2a levels in all mice that received either CT or dmLT as adjuvant (Fig. 3A).
Systemic B cell and T cell immune responses to PSSP-1 in mice. (A) PSSP-1-specific IgG levels in serum. Mice were immunized intranasally 3 times, at 2-week intervals, with PSSP-1 protein (20 μg) plus either CT (2 μg) or dmLT (5 μg). Six days after the third immunization, serum samples were collected from individual mice and levels of IgG, IgG1, and IgG2a in response to recombinant PSSP-1 were determined by ELISA (n = 5 for CT and n = 8 for dmLT). n.s., not significant. (B) Analysis of cytokines secreted from splenocytes of BALB/c mice immunized with PSSP-1 plus CT, 2 weeks after the third immunization with PSSP-1 plus adjuvant. After sacrifice, isolated spleen cells were stimulated with pooled whole PSSP-1 peptides (25 amino acids) for 72 h at 37°C in 5% CO2. Cytokines from culture supernatants were analyzed using the BD mouse Th1/Th2/Th17 CBA kit, according to the manufacturer's protocol. The values obtained were subtracted from unstimulated splenocyte data for individual mice (n = 3 or 4). M2e, ectodomain of matrix protein 2 of human influenza virus type A (SLLTEVETPIRNEWGCRCNGSSD) (used as control). Data are means ± SD and are representative of two experiments.
We next examined whether PSSP-1 immunization could elicit cytokine responses. Two weeks after the final booster immunization with CT, splenocytes harvested from immunized mice were stimulated with pooled 25-mer peptides that overlapped 14 amino acids in PSSP-1, and the levels of gamma interferon (IFN-γ), interleukin 17A (IL-17A), IL-2, IL-6, IL-4, and tumor necrosis factor (TNF) in cell culture supernatants were analyzed. We observed that levels of all cytokines, including IFN-γ and IL-17A, which are widely known to be involved in generating protection against Shigella infections (28, 29), were increased by PSSP-1 immunization (Fig. 3B). Mice that received PSSP-1 with dmLT as adjuvant demonstrated comparable results (data not shown). Collectively, these results suggest that PSSP-1 immunization, with the use of an appropriate adjuvant, effectively induces systemic humoral immune responses as well as Th1, Th2, and Th17 cellular immune responses.
PSSP-1-conferred protection when administered mucosally, but not systemically, in a lung pneumonia model.We evaluated the route-dependent protective efficacy generated by PSSP-1 immunization. Mice were immunized i.n. or intradermally (i.d.) with dmLT-adjuvanted PSSP-1. Mice immunized i.n. were protected against S. flexneri serotype 2a strain 2457T or S. dysenteriae serotype 1, although the survival rates were below 100%. The i.d. route did not confer protection against either homologous or heterologous Shigella challenge (Fig. 4A).
Data showing better protection against Shigella infections with mucosal administration of PSSP-1 than with systemic administration in a mouse lung model. (A) Mice immunized with PSSP-1 (20 μg i.n. or 40 μg i.d.) plus dmLT (5 μg i.n. or 0.5 μg i.d.) 3 times, at 2-week intervals, were challenged with lethal doses of S. dysenteriae serotype 1 strain 101398 (5 × 106 CFU) or S. flexneri serotype 2a strain 2457T (5 × 107 CFU). Survival was observed daily. Results are representative of two independent experiments (n = 5 or 6 mice per group). (B) At 7 days after the last immunization with PSSP-1 plus dmLT, serum and BALF samples were obtained and PSSP-1-specific immunoglobulin titers were determined by ELISA (n = 4 per group). B.D., below detection level. (C) At 7 days after the second immunization with PSSP-1 plus dmLT, spleen and cervical lymph node cells were prepared, and anti-PSSP-1 IgG- or IgA-producing B cells were detected with an ELISPOT assay. The numbers of antibody-secreting cells (ASC) per 106 cells are shown (n = 4). Results are representative of three independent experiments. (D) At 14 days after the third immunization with PSSP-1 plus dmLT, single cells were prepared from spleens and cervical lymph nodes (CLN) from the mice. Cells were stimulated for 72 h with a PSSP-1 whole-peptide pool or influenza virus matrix protein 2 ectodomain (M2e), and culture supernatants were analyzed for secretion of IFN-γ and IL-17A (n = 5 per group). Data are means ± SD and are representative of two independent experiments. *, P < 0.05; **, P < 0.01; n.s., not significant.
Antibody responses to PSSP-1 were also evaluated following immunization by different routes. PSSP-1-specific IgG levels in blood and bronchoalveolar lavage fluid (BALF) samples were comparable with i.n. and i.d. immunizations. However, PSSP-1-specific IgA levels were below detection levels and near zero in serum and BALF samples, respectively, after i.d. immunization. In contrast, PSSP-1-specific IgA titers were high in both serum and BALF samples after i.n. administration (Fig. 4B).
Of note, ELISPOT assay data demonstrated that PSSP-1-specific IgG- or IgA-secreting B cells were most abundant in the lungs of mice immunized i.n. but were only minimally detected in the lungs of mice immunized i.d. In contrast, the spleens of mice immunized i.d. had numerous PSSP-1-specific IgG-secreting B cells but not IgA-secreting B cells (Fig. 4C). Our data demonstrate that mucosal immunization is more effective than systemic immunization in inducing mucosal B cell immune responses.
We also evaluated the IFN-γ and IL-17A levels in response to PSSP-1 immunization via different routes. Spleens and cervical lymph nodes were isolated from mice immunized i.n. or i.d. with PSSP-1 plus dmLT as adjuvant and were stimulated with a PSSP-1 peptide pool for 72 h. Following peptide stimulation, IFN-γ and IL-17A levels in cell culture supernatants were analyzed. We found higher levels of IFN-γ and IL-17A in the culture supernatants of spleen cells isolated from mice immunized i.n. than in those from mice immunized i.d. High levels of IL-17A were detected from both spleens and cervical lymph nodes from mice immunized i.n. These results suggest that cellular immune responses are strongly induced only when PSSP-1 is administered i.n. with dmLT. Taken together, our data indicate that i.n. delivery of PSSP-1 produces better protective immunity in this Shigella pulmonary infection model.
Low levels of blood antibody-secreting cell responses to PSSP-1 in patients with recent-onset shigellosis.To evaluate whether natural Shigella infections in human patients can induce mucosal antibody responses to PSSP-1 (which is possibly related to protective immunity), we examined the development of PSSP-1-specific gut-homing (α4β7) B cell immunity (30, 31) in human patients with shigellosis. Positive shigellosis was confirmed by culture of stool samples and by PCR findings positive for the ipaH gene (32). Of 34 subjects studied, 24 were microbiologically and PCR positive for shigellae. The ASC counts for Shigella antigens PSSP-1, IpaC (23), and IpaD (24) were determined 5 or 6 days after the first diarrheal symptoms. The cutoff value for ASC positivity was defined as ≥10 spots per milliliter of blood after subtraction of the value for the PBS control for each sample. Ten spots per milliliter represents at least 1 spot/well of the ELISPOT assay plate for each antigen. Because one person among the 24 patients with shigellosis withdrew consent, blood samples from 23 subjects were analyzed. The subjects had many fewer gut-homing PSSP-1-specific IgA (3/23 subjects) or IgG (1/23 subjects) ASCs than IpaC-specific IgA (6/23 subjects) or IgG (19/23 subjects) ASCs or IpaD-specific IgA (8/23 subjects) or IgG (16/23 subjects) ASCs (Table 2). Details of positive and negative shigellosis cases are shown in Tables S2 and S3 in the supplemental material. Since only a few subjects mounted ASC responses to the PSSP-1 protein antigen during natural infections, PSSP-1 immunization may provide cross-protective immunity in humans who previously had shigellosis.
Blood antibody-secreting cell responses (α4β7, IgA, and IgG) to Shigella antigens in patients with recent-onset shigellosisa
DISCUSSION
In this study, we found that mucosal immunization with the common Shigella outer membrane protein PSSP-1 could induce broadly protective immunity against Shigella. Initially, 13 genes were identified as Shigella specific by comparative genomic analyses. Among them, two polypeptides, encoded by partial sigA and icsP genes, showed protection against homologous Shigella challenges in a mouse pneumonia model (see Table S1 in the supplemental material). Of note, the partial IcsP polypeptide PSSP-1 protected the mice against heterologous Shigella challenges. The failure of PSSP-1-immunized mice to survive challenge with an icsP gene-disrupted Shigella strain (ΔicsP strain) confirms that PSSP-1-mediated immune responses play an important protective role against Shigella infections (Table 1). SC602 also has IcsP on the outer membrane, but it failed to provide cross-protection against heterologous Shigella challenges, probably due to the decreased magnitude of immune responses in the presence of large amounts of lipopolysaccharide (LPS). Shigella vaccine development targeting LPS (an important component of the outer membrane of Gram-negative bacteria) alone has not been successful (33). LPS contains three chemically linked components, i.e., lipid A, core polysaccharide, and O antigen (34). Among the three, the O antigen polysaccharide chain is known to be one of the main virulence factors for shigellae and also the major target for innate immunity (35, 36). However, the structural variability of the O antigen polysaccharide chain among serotypes makes it difficult to utilize serotype-specific LPS as a broadly protective agent in a Shigella vaccine. Also, polysaccharide induces a T cell-independent antibody response and has poorer immunological memory than the T cell-dependent antigens. Thus, the best way to prevent shigellosis would be to use conserved proteins that could provide broad protection against different species and serotypes.
Subunit polypeptide vaccines are considered to be safer than live attenuated vaccines, but they require adjuvant to confer protective immunity. CT produced by Vibrio cholerae and heat-labile toxin (LT) produced by enterotoxigenic E. coli are very strong mucosal adjuvants but, due to their toxicity, are not approved for human use. dmLT was designed to reduce LT toxicity and is safe for animals and humans (27). Thus, we investigated dmLT as an adjuvant for use with PSSP-1. As shown in Fig. 2 and 3, dmLT enhances systemic and mucosal antibody and T cell responses, as indicated by Th1, Th2, and Th17 cytokine production in response to PSSP-1. Also, dmLT-adjuvanted PSSP-1 protected mice against challenges with multiple Shigella serotypes.
To test whether the route of PSSP-1 immunization is important for the induction of protective immunity, we compared the i.n. and i.d. routes. Although i.d. immunizations with PSSP-1 could induce higher levels of PSSP-1-specific IgG in blood than i.n. immunizations, mice had no protection against either S. dysenteriae serotype 1 or S. flexneri serotype 2a strain 2457T challenge. Intranasal immunization with PSSP-1 plus dmLT as the adjuvant resulted in high levels of PSSP-1-specific IgA in lungs and protection. We found higher levels of IFN-γ and IL-17A in response to PSSP-1 after i.n. immunizations than after i.d. immunizations. These results indicate that mucosal immunity induced by PSSP-1 plays an important role in protective immunity against Shigella infection. The combination of PSSP-1 and other Shigella proteins or novel mucosal adjuvants is under consideration for better protection against Shigella infection.
According to a recent publication by Tran et al. (37), IcsP is preferentially concentrated at the new pole of nonseptating cells and at the septum of dividing cells and is masked by LPS O antigens. This is consistent with our data showing that PSSP-1-specific antibodies could not bind IcsP on the bacterial surface of wild-type S. flexneri serotype 2a strain 2457T, while the same polyclonal antibodies could bind the LPS mutant Shigella strain, which has only one unit of O antigen (38) (data not shown). This suggests that PSSP-1-specific antibodies may react to or neutralize shigellae during certain stages of the dividing cycle, when less LPS is expressed on the surface of the bacteria at the gut site. Here, we found that mucosal immunity to PSSP-1 is important for protection against Shigella infections. Detailed mechanisms that are involved in PSSP-1-induced protection need to be further elucidated. When the ASCs in blood samples from patients with recent-onset shigellosis were analyzed, the hierarchical level of mucosal antibody responses to PSSP-1 seemed to be lower than those of other Shigella Ipa proteins (Table 2). Due to our inability to obtain blood samples from patients after discharge from the hospital, we could not measure PSSP-1-specific antibody levels later. However, early antibody responses to PSSP-1 were rare in shigellosis cases, and thus it is assumed that people with naturally acquired shigellosis have much less B cell memory for PSSP-1. Therefore, vaccination with PSSP-1 can provide protective immunity to vaccinees against repeated Shigella infections.
Because there is no murine gut model of shigellosis, the mouse pneumonia model has been used to test Shigella vaccine candidates (28). Invasions of lung epithelial cells occur with i.n. Shigella infections but do not mimic the gut environment in which Shigella infections evoke pathology in humans. We have also evaluated the guinea pig rectocolitis model of shigellosis, using intrarectal administration of shigellae (39). As guinea pigs aged (weights of >300 g), however, they ceased to respond to intrarectal inoculation of virulent Shigella, which made it difficult to test the protective efficacy of Shigella vaccine candidates with long-term immunization. Further improvements of the protocols will be required to evaluate novel Shigella subunit vaccine candidates using this model. In conclusion, our study provides evidence that mucosal immunization with PSSP-1 induces cross-protective immunity against Shigella infections and therefore PSSP-1 can be utilized as a promising universal Shigella vaccine candidate.
ACKNOWLEDGMENTS
This work was supported by grants from the National Research Foundation of Korea (grants NRF-2013K1A2A1058637 and NRF-2013K1A2A1058645), PATH (PSSP-1 SUBVACS against Dysentery Preclinical and Early Clinical Development), and the governments of the Republic of Korea and Sweden. D.W.K. was supported by grant 2012R1A2A2A01009741 from the National Research Foundation of Korea.
We have no financial conflicts of interest.
FOOTNOTES
- Received 13 October 2014.
- Returned for modification 2 December 2014.
- Accepted 24 January 2015.
- Accepted manuscript posted online 4 February 2015.
Supplemental material for this article may be found at http://dx.doi.org/10.1128/CVI.00661-14.
- Copyright © 2015, American Society for Microbiology. All Rights Reserved.
REFERENCES
