Bacterial Strains and Plasmids.

Bacteria strains and plasmids utilized in this study are described in Table 1. S. Typhimurium and E. coli strains containing Asd⁺ plasmids were grown at 37 °C in Luria-Bertani (LB) broth or on LB agar. For the growth of strains with asdA gene deletions, diaminopimelic acid (DAP; 50 μg/mL) was added to the growth medium. S. pneumoniae strains D39 (serotype 2), WU2 (serotype 3), STREP5 (serotype 5), EF6796 (serotype 6A), EMC9V (serotype 9V), STREP14 (serotype 14), and OREP18C (serotype 18C) were grown on Trypticase soy agar plates supplemented with 5% sheep blood or in Todd-Hewitt broth supplemented with 0.5% yeast extract (THY medium) at 37 °C in an anaerobic jar system (GasPak system, BBL).

Table 1.

Bacterial strains and plasmids used in this study

Strains or plasmids	Description	Source
S. pneumoniae strains
D39	Wild-type S. pneumoniae CPS2 strain	Laboratory collection
WU2	Wild-type S. pneumoniae CPS3 strain	Lababoratory collection
STREP5	Wild-type S. pneumoniae CPS5 strain	BEI Resources
EF6796	Wild-type S. pneumoniae CPS6A strain	Laboratory collection
EMC9V	Wild-type S. pneumoniae CPS9V strain	BEI Resources
STREP14	Wild-type S. pneumoniae CPS14 strain	BEI Resources
OREP18C	Wild-type S. pneumoniae CPS18C strain	BEI Resources
E. coli strain
χ6097	∆(lac-pro) rpsL ∆asdA4 ∆(zhf-2::Tn10) thi ϕ80dlacZ∆M15	K-12/F−
Salmonella enterica serovar Typhimurium strains
χ3761	Wild-type UK-1	Laboratory collection
χ11316	∆pabA1516 ∆pabB232 ∆asdA16 ∆araBAD23 ∆relA198::araC PBAD lacI TT ∆wbaP45	χ9241 (42)
Recombinant plasmids
pG8R184	Asd+ vector, pSC101 ori, Kan+	Laboratory stock
pG8R307	pG8R184 - S. pneumoniae CPS2 operon (wchA-rmlD), pSC101 ori	Present study
pG8R308	pG8R184 - S. pneumoniae CPS3 (ugd-pgm), pSC101 ori	Present study
pG8R309	pG8R184 - S. pneumoniae CPS5 (wcil-fnlC), pSC101 ori	Present study
pG8R310	pG8R184 - S. pneumoniae CPS6A (wchA-rmlD), pSC101 ori	Present study
pG8R311	pG8R184 - S. pneumoniae CPS9V (wchA-wcjE), pSC101 ori	Present study
pG8R312	pG8R184 - S. pneumoniae CPS14 (wchA-Irp), pSC101 ori	Present study
pG8R313	pG8R184 - S. pneumoniae CPS18C (wchA-rmlD), pSC101 ori	Present study
Strains for immunization
RASV-CPS2	χ11316 electroporated with pG8R307
RASV-CPS3	χ11316 electroporated with pG8R308
RASV-CPS5	χ11316 electroporated with pG8R309
RASV-CPS6A	χ11316 electroporated with pG8R310
RASV-CPS9V	χ11316 electroporated with pG8R311
RASV-CPS14	χ11316 electroporated with pG8R312
RASV-CPS18C	χ11316 electroporated with pG8R313
RASV-184	χ11316 electroporated with pG8R184

Cloning of S. pneumoniae CPS Genes into pG8R184.

Genomic DNA from S. pneumoniae strains was isolated using the GenElute Bacterial Genomic DNA Kit (Sigma-Aldrich). All capsule loci were amplified from the genomic DNA using the PrimeSTAR GXL DNA Polymerase (Takara Bio) with primers listed in SI Appendix, Table S1. For each capsule locus, multiple primers were designed to ligate fragments amplified from S. pneumoniae genomic DNA with pG8R184 to create plasmids pG8R307 (CPS2), pG8R308 (CPS3), pG8R309 (CPS5), pG8R310 (CPS6A), pG8R311 (CPS9V), pG8R312 (CPS14), and pG8R313 (CPS18C). The resulting amplicons from each capsular type were ligated to pG8R184 by gene assembly using NEBuilder HiFi DNA Assembly Master Mix (New England Biolabs). The products were transformed into E. coli strain χ6097 by electroporation for propagation of plasmids. The plasmids were then confirmed by PCR, sequencing, and CPS production, before electroporated into S. Typhimurium strain χ11316. Both E. coli and S. Typhimurium transformants were detected on LB agar plates lacking DAP.

Detection of Capsule Expression by Western Blotting.

S. Typhimurium strains with or without recombinant plasmids, grown overnight at 37 °C, were diluted in fresh LB media to an OD₆₀₀ of 0.03. All strains were OD₆₀₀ matched (0.8 to 0.9). The LPS of the S. Typhimurium transformants and wild-type S. Typhimurium χ3761 were separated by SDS/PAGE (49). Samples were then transferred onto nitrocellulose membranes and immunologically detected with 1:1,000 dilutions of serotype-specific rabbit anti-CPS antibodies (Statens Serum Institute) for identification of CPS synthesis. Salmonella O Group B Antiserum (BD) was used for identification of S. Typhimurium native O-antigen at a 1:1,000 dilution.

Animal Immunization and Challenge.

The animal experiments were performed in accordance with the recommendations of the Animal Welfare Act and US Government Principles for the Utilization and Care of Vertebrate Animals Used in Testing, Research, and Training. The protocols were approved by the University of Florida Institutional Animal Care and Use Committee and Southwest University Institutional Animal Care and Use Committee.

For immunization, groups of 7-wk-old female BALB/c mice (Charles River Laboratories) were orally vaccinated with strains of RASV-CPS2 (χ11316 with pG8R307), RASV-CPS3 (χ11316 with pG8R308), RASV-CPS5 (χ11316 with pG8R309), RASV-CPS6A (χ11316 with pG8R310), RASV-CPS9V (χ11316 with pG8R311), RASV-CPS14 (χ11316 with pG8R312), RASV-CPS18C (χ11316 with pG8R313), and RASV-184 (χ11316 with pG8R184) at a dose of 1 × 10⁹ CFU in 20 μL of buffered saline with gelatin (BSG) on day 0. The same dose of the same strain was given 2 wk later (on day 14). Mice orally administrated with 20 μL of BSG were included as the negative control. For the positive control group, the PPV23 (Chengdu Institute of Biological Products) was diluted at 1:10 and 100 μL per mouse was intraperitoneally administered on day 0 and day 14. Three mice in each group were killed for the assay of cytokine expression on day 21. Blood was collected on day 28 (2 wk after secondary immunization) via mandibular vein puncture. Vaginal secretion samples were collected as described previously (47). For RASV-CPS2 (χ11316 with pG8R307) and RASV-CPS3 (χ11316 with pG8R308) immunized groups, mice (n = 10) were challenged on day 35 (3 wk after secondary immunization) intraperitoneally with 2 × 10⁴ CFU (150 times the LD₅₀) of D39 and 2 × 10⁴ CFU (100 times the LD₅₀) of WU2 in 100 μL BSG, respectively.

Antigen Preparation and ELISA.

Purified pneumococcal polysaccharide powders of types 2, 3, 5, 6A, 9V, 14, and 18C were purchased from the American Type Culture Collection (ATCC). S. Typhimurium LPS were extracted using the LPS extraction kit (Boca Scientific) from χ3761. Both of the pneumococcal polysaccharides and the S. Typhimurium LPS were coated on the 96-well ELISA plates (100 ng per well). Serum antibodies and vaginal secretion IgA (SIgA) against CPSs and S. Typhimurium LPS were measured using ELISA, as described previously (47).

Cell Stimulation and Flow Cytometry.

Single-cell preparations of spleen were obtained as previously described (47). Splenocytes were stimulated with each type of the CPSs or with the LPS of χ3761 (50 ng/2 × 10⁶ cells) in the presence of Protein Transport Inhibitor Mixture (eBioscience). The cells were then incubated for 8 h at 37 °C with 5% CO₂. Before staining, dead cells were excluded using Zombie Red (Biolegend) and mouse FcR were blocked with the rat anti-mouse CD16/CD32 monoclonal antibody (eBioscience). For surface staining, cells were incubated with fluorophore-conjugated antibodies on ice for 30 min. The eBioscience Intracellular Fixation and Permeabilization Buffer Set was used for intracellular staining. Cells were run on a BD LSRFortessa cell analyzer and analyzed with FlowJo software (TreeStar).

For detection of cytokine expression in CD4⁺ T cells at 21 d after primary immunization, the following antibodies purchased from Biolegend were used, including rat anti-mouse CD3-Alexa Fluor 700, CD4-FITC, IFN-γ–APC/Cy7, APC/Cy7-conjugated rat IgG1κ (isotype), IL-4–Alexa Fluor 647, Alexa Fluor 647-conjugated rat IgG1κ (isotype). Data are presented as described in the figure legends.

Complement Deposition Assays.

Serum samples collected from each group were heat-inactivated and pooled. Each strain of S. pneumoniae (10⁷ CFU) was resuspended in 90 μL of PBS containing 1% bovine serum albumin (BSA) (Sigma-Aldrich) and mixed with 10 μL of inactivated pooled serum from the corresponding group for 40 min at 37 °C. Treated S. pneumoniae strain cells were then washed twice in PBS and incubated with 100 μL of PBS-1% BSA containing 20 μL of baby rabbit complement (BRC, Sigma-Aldrich) for 30 min at 37 °C. FITC conjugated goat anti-rabbit complement polyclonal antibody (1% final concentration) (MP Biomedical Cappel) was used for staining of the bacteria for 45 min on ice. The samples were fixed with 2% paraformaldehyde solution and analyzed by flow cytometry on the BD LSRFortessa cell analyzer.

Opsonophagocytic Assay.

The opsonophagocytic assay/activity (OPA) was performed mainly as previously described (50). Briefly, differentiation of HL-60 cells (CCL240, ATCC) into granulocytes were carried out in RPMI-1640 supplemented with 1% l-glutamine, 10% fetal calf serum (FCS), and 100 mM N,N- dimethylformamide (DMF) at a cell density of 2 × 10⁵ cells/mL, in a 5% CO₂ incubator at 37 °C. HL-60 cells from day 4, 6, or 7 postdifferentiation were used as the effector cells for OPA. Cell viability greater than 90% was assessed by Trypan blue exclusion. Inactivated pooled sera from each group were serially diluted in twofold steps in the Hanks’ buffer with Ca²⁺, Mg²⁺, and 0.1% gelatin. The diluted serum samples were then incubated with the corresponding S. pneumoniae strain (∼1,000 CFU per well) for 15 min at 37 °C in a 5% CO₂ incubator. Following this incubation, BRC (final concentration: 12.5%; Sigma-Aldrich) and differentiated HL-60 cells (4 × 10⁵ cells) were added. The reaction mixtures were incubated at 37 °C for 45 min with rotation at 220 rpm. After incubation, viable counts of S. pneumoniae strains were determined by dilution plating on blood agar plates. Reaction mixtures containing all the test reagents except serum samples were included as negative controls. The percentage of bacterial killing was calculated by the equation [1 − (number of colonies presented in tested sample/number of colonies presented in negative control)] × 100. All the samples were tested and plated in duplicate. The mean of four independent CFU counts for each sample was used to calculate the percentage of killing.

Statistical Analyses.

The GraphPad Prism 5 (Graph Software) was used to perform statistical analyses. One-way or two-way ANOVA followed by Bonferroni’s test was used for comparisons among groups, depending on the experimental design. All data except for animal survival were expressed as means with deviations. Animal survival analysis was performed using the Kaplan–Meier method followed by the log-rank (Mantel–Cox) test. The levels of significance were set at 0.05, 0.01, and 0.001.

RASV Strains Were Established to Synthesize CPSs of S. pneumoniae.

The sequences of the cps locus have been reported for all 90 serotypes of S. pneumoniae (3). Most of the serotypes (except for types 3 and 37) utilize the wzy-dependent pathway for capsule assembly, which is similar to the mechanism of O-antigen synthesis in S. Typhimurium (7). In this study, we cloned and introduced seven serotypes of S. pneumoniae cps sequences encoding between two and six structural components into the RASV strain (Fig. 1A), to confirm if the CPSs from gram-positive S. pneumoniae could be expressed and ligated to Salmonella lipid A-core receptor. Sequences from the initial transferase gene to the last gene upstream of 3′ transposase gene in the cps locus of serotypes 2, 3, 5, 6A, 9V, 14, and 18C were cloned into the vector pG8R184, respectively (Fig. 1B). Each of the resulting constructs was then transformed into the RASV strain χ11316, which had been modified to facilitate the synthesis of heterologous polysaccharides by introduction of a ΔwbaP mutation (42).

Fig. 1.

S. pneumoniae CPS repeat units and construction of plasmids for CPS biosynthesis. (A) The capsule repeat unit of each serotype is exhibited to be linked with the reducing end sugar on the right-hand side. Monosaccharides are depicted using different colored shapes according to the graphic symbols. (B) Plasmids designed and constructed for CPS synthesis. The vector backbone used for cloning CPS gene clusters is a balanced lethal vector pG8R184, which contains low copy ori of pSC101, Asd⁺, and Kan-resistance marker and P_trc promoter with T1T2 terminator. Sequences of each pG8R184 derivative include the capsule coding genes from one of the seven serotypes, pSC101 ori, Asd⁺, Kan-resistance marker, and T1T2 terminator. The diagrams of CPS structures and CPS biosynthesis genes are modified from Bentley et al. (3). “P#” designations refer to primer names used to construct the plasmids, and the purposes of primers are listed in SI Appendix, Table S1.

To test the expression and polymerization of S. pneumoniae CPSs in the RASV system, polysaccharides from plasmid-bearing χ11316 derivatives RASV-CPS2 (pG8R307), RASV-CPS3 (pG8R308), RASV-CPS5 (pG8R309), RASV-CPS6A (pG8R310), RASV-CPS9V (pG8R311), RASV-CPS14 (pG8R312), RASV-CPS18C (pG8R313), along with empty plasmid control RASV-184 (pG8R184), were extracted and detected by Western blotting with CPS-specific antibodies. All the reconstituted polysaccharides could be detected by antisera against S. pneumoniae CPS, as indicated in Fig. 2 and SI Appendix, Fig. S2. RASV-CPS 2, RASV-CPS3, and RASV-CPS14 showed clearly ladder-like patterns, when detected with antibodies against CPS2, -3, and -14. RASV-CPS5 produced only CPS5-specific high molecular weight (HMW) polysaccharides, while RASV-CPS9V and RASV-CPS18C expressed CPSs with low molecular weight (LMW) (SI Appendix, Fig. S1). The recombinant CPSs synthesized by RASV-CPS6A contained both LMW and medium molecular weight polysaccharides. However, there were also LMW signals detected in the wild-type Salmonella strain χ3761 in the CPS6A detection assay, which indicated the cross-reaction of the antisera against S. pneumoniae CPS6A.

Fig. 2.

Immunoblot detection of S. pneumoniae CPS synthesis in RASV strains. Each plasmid carrying the capsule coding genes of one serotype was electroporated into the RASV strain χ11316. The transfected RASV strains were lysed and separated by 12% SDS/PAGE. The CPS profiles of serotype 2 (A), 3 (B), 6A (C), and 14 (D) were then detected using serotype-specific anti-CPS antibodies. The S. Typhimurium O-antigen was also identified by Western blot using the rabbit polyclonal Salmonella group B antisera. The χ11316 strain containing empty plasmid (RASV-CPS184) were included as the control sample. Blots of CPS profiles shown are representative of three independent experiments.

RASV-CPSs Strains Induced Significant CD4⁺ T Cell Immune Responses Against S. pneumoniae CPSs.

Flow cytometry assays were performed to investigate the CPS-specific CD4⁺ T-cell responses elicited by the RASV-CPS strains. Splenocytes were collected from mice at day 21 (7 d after secondary immunization) and stimulated with the type-specific CPSs for 8 h. After stimulation, cells were gated and analyzed as described in SI Appendix, Fig. S2. CD4⁺ T-cells from mice immunized with the corresponding RASV-CPS strains could stimulate significantly increased IFN-γ compared to those from PPV23-immunized mice (Fig. 3A and SI Appendix, Fig. S3A). In contrast, CD4⁺ T cells from PPV23-immunized mice induced more interleukin-4 (IL-4) than most of the RASV-CPS strains (Fig. 3B and SI Appendix, Fig. S3B). However, the IL-4 production levels showed no statistically difference between CD4⁺ T-cells from RASV-CPS18C–immunized mice and those from the PPV23 group when stimulated with the CPS18C.

Fig. 3.

Cytokine expression in CD4⁺ T cells stimulated with S. pneumoniae CPSs ex vivo. Mice splenocytes were isolated on day 21 (7 d after secondary immunization) and stimulated with the corresponding serotype-specific CPS for 8 h. Fluorochrome-conjugated antibodies, including rat anti-mouse CD3-Alexa Fluor 700, CD4-FITC, IFN-γ–APC/Cy7, and IL-4–Alexa Fluor 647 were used for cell staining and analysis with flow cytometry. The cytokine stimulation index (SI) was expressed relative to the percentage of IFN-γ⁺ (A) or IL-4⁺ (B) cell subsets in the BSG-treated group. Data from two independent experiments were pooled and analyzed (n = 6). Mean ± SD *P < 0.05, **P < 0.01, ***P < 0.001, ns = not significant, two-way ANOVA and Bonferroni’s multiple comparison test.

RASV-CPS Strains Elicited CPS-Specific Humoral Immune Responses.

Groups of mice were vaccinated with the seven RASV-CPS strains using a two-dose schedule. The PPV23-, RASV-CPS184– (empty plasmid control), and BSG-treated mice were included as control groups. All the RASV-CPS strains stimulated similar or higher levels of serum anti-CPS IgG compared to the PPV23-immunized group (Fig. 4A). Moreover, the IgA levels in the vaginal secretions were significantly increased in the RASV-CPS–immunized groups (except for RASV-CPS9V and RASV-CPS18C) (Fig. 4B). The antibody response to RASV-CPS strains was dominated by IgG2a (Fig. 4C). In contrast, the PPV23 induced significantly higher levels of IgG1 production (Fig. 4D). The levels of type-specific anti-CPS antibodies varied among groups. Lower concentrations of serum type-specific IgG were induced by RASV-CPS5, RASV-CPS9V, and RASV-CPS18C compared to other RASV-CPSs.

Fig. 4.

Humoral immune responses against S. pneumoniae CPSs. Serum (A, C, and D) and mucosal (B) antibody concentrations were measured by ELISA on day 28 (14 d after secondary immunization) (n = 12). The RASV-CPS strains stimulated comparable or higher levels of serum IgG, Ig2a, and mucosal IgA compared to the PPV23 group. Data were represented as mean ± SD *P < 0.05, **P < 0.01, ***P < 0.001, ns = not significant. Data were analyzed by using two-way ANOVA with Bonferroni’s multiple comparison test.

Serum Antibody Mediated Complement Deposition and OPA.

The complement system plays a vital role for host immunity to S. pneumoniae, especially in opsonophagocytosis of S. pneumoniae (5152–53). Serum antibodies mediated opsonization of S. pneumoniae activates the complement system, depositing C3 complement on the pneumococcal surface, which enables phagocytosis of the bacteria by phagocytic cells. Serum samples from mice immunized with RASV-CPS strains deposited similar or higher levels of C3 on the corresponding S. pneumoniae strains, compared to sera from PPV23-immunized mice (Fig. 5). Whereas the ability of sera to deposit C3 was reduced in RASV-184 (empty plasmid control) vaccinated mice, no C3 deposition was detected on bacteria for sera from BSG-treated group.

Fig. 5.

Complement deposition assays. Representative data of flow plots for antibody deposition on S. pneumoniae in 10% pooled antiserum from each group of RASV-CPSs immunized mice. PPV23, RASV-184 (empty plasmid control), and BSG vaccinated serum were included as controls.

OPA is suggested to be a good indicator of protection against pneumococcal infection (54, 55). All the RASV-CPSs elicited opsonizing antibodies that mediated the killing of pneumococci in a titer-dependent manner. Data in Fig. 6 indicated that sera from mice vaccinated with RASV-CPS2, RASV-CPS3, RASV-CPS5, RASV-CPS6A, and RASV-CPS14 showed increased or equivalent OPA compared to those from PPV23 treated mice. However, serum antibodies from RASV-CPS9V and RASV-CPS18C showed weaker responses than the PPV23 antibodies.

Fig. 6.

OPA in RASV-CPSs vaccinated mice. (A–G) OPAs were carried out using differentiated HL-60 cells and incubated with bacteria preopsonized RASV-CPS–specific antibodies or antibodies from control groups. Viable bacteria are counted after incubation at 37 °C for 45 min. Tested samples containing no serum (only bacteria, cells and complement) were included as negative control. The percentage of bacterial killing was determined relative to the negative control (contained only bacteria, cells and complement). Data shown are presented as mean ± SD.

Protection of Mice Vaccinated with RASV-CPS2 or RASV-CPS3 against Challenge of Wild-Type S. pneumoniae Strains.

To determine the protective efficacy of vaccinations with RASV-CPS2 and RASV-CPS3, survival experiments were performed using sepsis models of pneumococcal infection. The RASV-CPS2 and RASV-CPS3 vaccinated mice were challenged intraperitoneally with wild-type S. pneumoniae strains D39 (type 2) and WU (type 3), respectively. As shown in Fig. 7, immunization with either of these RASV-CPS strains resulted in significant protection against pneumococcal challenge compared to the RASV-CPS184– (empty plasmid control) and BSG-treated groups. However, there was no remarkable difference in protection between RASV-CPSs and PPV23-vaccinated mice.

Fig. 7.

Survival of mice challenged with wild-type S. pneumoniae strains after immunization with RASV-CPSs. Female BALB/c mice were immunized orally twice at 2-wk intervals with 1 × 10⁹ CFU of RASV-CPS2 (A) or RASV-CPS2 (B). Mice were challenged intraperitoneally with 2 × 10⁴ CFU of D39 (A) or 2 × 10⁴ CFU of WU2 (B) at 3 wk after the secondary immunization. Mice intraperitoneally immunized with PPV23 were included as positive control. The RASV-184–immunized or BSG-treated mice were used as negative control groups. Survival analysis was performed using the Kaplan–Meier method with the log-rank (Mantel–Cox) test. ***P < 0.001 compared to BSG, ^###P < 0.001 compared to RASV-184.