Based on a consensus approach which combines NN-align, SMM-align, and combinatorial library methods, potential immunodominant CD4⁺ T cell epitopes binding to human leukocyte antigen (HLA) class II molecules were screened from UreB by online T Cell Epitope Prediction Tools (IEDB Analysis Resource, http://tools.iedb.org/main/tcell/). The predicted CD4⁺ T cell epitopes with percentile rank ≤ 1 are considered high affinity and used as the candidate epitopes for vaccine design (Supplementary Data Sheet 1). At the same time, based on sequence characteristics of the UreB antigen using amino acid scales and Hidden Markov Model (HMM), potential linear B cell epitopes was screened by online B Cell Epitope Prediction Tools (IEDB Analysis Resource, http://tools.iedb.org/main/bcell/). An output threshold of 0.9 was used (sensitivity = 0.25, specificity = 0.91) for identification of linear B cell epitopes. This threshold was chosen to provide a high level of certainty for predicted B-cell epitopes from UreB. The predicted B cell epitopes are as shown in Supplementary Data Sheet 2. In addition, the known epitopes of UreB were also obtained from Immune Epitope Database (IEDB, http://www.iedb.org/home_v3.php), as shown in Supplementary Data Sheet 3. The final epitope-rich regions of UreB were determined by overall consideration on the predicted and known epitopes. Finally, the predicted epitope-rich regions (UreB_158–251 and UreB_321–385) and the other four B or Th cell epitopes [UreA_27–53 (Rizos et al., 2003), UreA_183–203 (Hifumi et al., 2006), HpaA_132–141 (Chaturvedi et al., 2001), and HSP60_189–203 (Yamaguchi et al., 2000)] were used to construct multivalent epitope-based vaccine. The reasonable combination of mucosal adjuvant CTB, Th1-type immune adjuvant NAP, the linkers (KK, GS, GGG, DPRVPSS), the tandem copies of Th or B cell epitopes (UreA_27–53, UreA_183–203, HpaA_132–141, and HSP60_189–203) and two predicted UreB epitope-rich regions (UreB_158–251 and UreB_321–385) were determined by modeling and prediction using RANKPEP, molecular operating environment (MOE), and DNAstar software.

To construct the vector expressing the multivalent epitope-based vaccine, a DNA fragment named WAE encoding NAP, multiple copies of selected B and Th cell epitopes (UreA_27–53, UreA_183–203, HpaA_132–141, and HSP60_189–203) and the UreB epitope-rich regions (UreB_158–251 and UreB_321–385) were synthesized and cloned into vector pETC containing cholera toxin B subunit (CTB) gene, generating the plasmid pETCWAE. The plasmid pETCWAE was transformed into E. coli BL21 for expression of the fusion protein CWAE. The CWAE protein was purified by Ni²⁺-NTA affinity chromatography according to according to the manufacturer's instructions.

Purified CWAE and CTB-UE, various H. pylori antigens (UreA, UreB, HpaA, HSP60, and NAP; Linc-Bio, Shanghai, China) were applied to 15% SDS-PAGE and transferred onto polyvinylidene difluoride membrane (PVDF, Millipore). Rabbit anti-H. pylori polyclonal antibody (Rabbit anti-Hp PcAb, Abace biology, Beijing, China) was used as primary antibody for CWAE and CTB-UE. Mice anti-CWAE polyclonal antibody (prepared by our laboratory) was for various H. pylori antigens (UreA, UreB, HpaA, Hsp60, and NAP). After washing, the membrane was then incubated with HRP-Goat Anti-Rabbit IgG (Proteintech) or HRP-Goat Anti-Mouse IgG (Proteintech). The positive signals were monitored using HRP-DAB Chromogenic Substrate Kit (Tiangen Biotech) according to manufacturer's reagent instructions.

All animal experiments were approved by the Animal Ethical and Experimental Committee of Ningxia Medical University. Immunological properties of CWAE vaccine were characterized in BALB/c mice model. SPF BALB/c mice (male, 5–6 weeks old; n = 6) were immunized with 100 μg of the purified CWAE, CTB-UE, or CTB (Absin Bioscience Inc., Shanghai) by abdominal multipoint subcutaneous injection with Freund's adjuvant 3 times at 1 week's interval. The antigens in PBS without adjuvant were for the last booster immunization. Anti-serum were separated on the fifth day after the last booster and were used for detection of antibody level and specificity.

Mongolian gerbils (male, 6 weeks old) were used for evaluating therapeutic effect of the CWAE vaccine. The experimental design of therapeutic vaccination was shown in Figure 1. Firstly, Mongolian gerbils were infected with H. pylori SS1 (10⁹ CFUs) using intubation, four times within the span of two weeks. For therapeutic vaccination, H. pylori-infected Mongolian gerbils were randomized into four groups (n = 7). Then H. pylori-infected Mongolian gerbils were vaccinated intra-gastrically with 100 μg of CWAE, CTB-UE or Urease in 500 μl aluminum hydroxide adjuvant for four times at 1-week interval. The infected Mongolian gerbils were also immunized with PBS, as control, using the same method. Two weeks after the final vaccination, Mongolian gerbils were sacrificed for evaluation of H. pylori infection.

Serum antibody levels were measured by ELISA. ELISA plates were coated with 0.5 μg/well of large antigen (Urease, UreA, UreB, HSP60, HpaA, NAP, or H. pylori lysates) or with 1 μg/well of synthetic epitope peptide (UreA_183–203, UreB_321–339, UreB_158–172, UreB_181–195, UreB_211–225, UreB_349–363, HpaA_132–141, and HSP60_189–203). To measure the serum IgG, blood was collected immediately before sacrifice. The serum was isolated and serially diluted before assay. The titer was determined as the reciprocal of the last dilution, with an OD >2 times the negative specimens (normal mouse serum). To detect secretory IgA (sIgA), the supernatants from the homogenized stomach tissue, intestinal tissue or feces were collected and diluted 1:5 in PBS for analysis of sIgA.

The ability of the CTB component in CWAE to bind to its cellular receptor was assessed by GM1-ELISA as previously described (Areas et al., 2004). Briefly, ELISA plates were coated with 1 μg/well GM1 ganglioside or BSA for 24 h. After washing, ELISA plates were blocked by incubating with 5% (m/V) skim milk for 2 h. The CWAE, CTB-UE, CTB or UreB proteins (100 μg/ml) were then added to ELISA plates and incubated for 2 h. After that, a proper dilution of anti-CTB polyclonal antibody (Biomade Technology) was added to the plates and incubated for 1 h. After washing, HRP-conjugated goat anti-mouse IgG (Jackson ImmunoResearch, USA) was added to the plate and incubated for 1 h. Substrate tetramethylbenzidine (TMB, Tiangen Biotech) was then added and incubated for 10 min. The absorbance was measured at 450 nm.

A protein A sepharose column (BioVision) was used to purify mouse IgG in the antiserum. H. pyloti urease were incubated with 50 μl serial dilutions of purified antiserum IgG (0–30 μg/ml) in ELISA plates for 12 h at 4°C. After that, The ELISA plates were incubated with 50 μl of 50 mM phosphate buffer (pH 6.8) containing 0.02% phenol red, 500 mM urea and 0.1 mM dithiothreitol (DTT) at 37°C. Color development was measured at 550 nm at 30 min intervals over a period of 3 h. Percentage inhibition was determined by the following equation: [(activity without antiserum − activity with antiserum)/(activity without antiserum) × 100].

Two weeks after the final therapeutic vaccination, mice were killed for determination of the H. pylori colonization in the stomachs. Briefly, the stomach was dissected into two tissue fragments along the lesser curvature. One fragment was homogenized in 2 ml of PBS by using a tissue homogenizer. Serial 10-fold dilutions of the homogenate were plated on H. pylori selective plate (QingDao Hopebio Technology) supplemented with 7% goat blood, trimethoprim (5 μg/ml), polymixin B (5 μg/ml), and vancomycin (10 μg/ml) under microaerobic conditions. After 4–6 days culture, colonies were counted and the number of Colony-Forming Units (CFU) per stomach was calculated. The degree of H. pylori colonization in the mouse stomach was also measured by rapid urease test. Briefly, the antral portion of the stomach was immediately immersed in 500 μl of sodium phosphate buffer containing 500 mM urea, 0.02 % phenol red, and 0.1 mM DTT. The stomach sample was incubated at 37°C for 3 h. The supernatant of specimens was used for quantification at 550 nm (A550).

One strip of stomach tissue was cut out and fixed with formalin. Then, the stomach tissue was embedded in paraffin and stained with hematoxylin and eosin (HE). For evaluation of gastritis, the slides were “blinded” and the extent of gastritis was graded as follows: 0, none; 1, a few leukocytes scattered in the deep mucosa; 2, moderate numbers of leukocytes in the deep to mid mucosa and occasional neutrophils in the gastric glands (microabscesses); 3, dense infiltrates in the deep to mid mucosa, a few microabscesses; and 4, dense, diffuse infiltrates throughout the lamina propria and into the submucosa, frequent microabscesses. The stomach tissue sections were also assessed for the presence of H. pylori infection by immunohistochemical (IHC) staining using polyclonal anti-H. pylori antibody (Linc-Bio, Shanghai, China) and polymer-HRP based detection system (BioGenex).

Splenic lymphocytes were isolated with lymphocyte separation medium (Dakewe Biotechnology Company) and cultured (2 × 10⁵ cells/well) with synthetic peptides of the Th epitopes (UreB_229–251, UreA_27–53, or UreB_373–385), NAP, Urease or H. pylori lysates (5 μg/ml) in plates at 37°C for 72 h. Then, 10 μl of the CCK-8 solution (Dojindo Molecular Technologies Inc) was added into plates for 4 h. The results were expressed as stimulation indices (SI), defined as the index of lymphocyte proliferation according to formula: SI = the absorbance 450 value of stimulated cultures/the absorbance 450 value of negative control cultures. To determine cytokine production, culture supernatants from splenic lymphocytes stimulated by H. pylori lysates for 72 h were collected to assay for IL-4, IFN-γ, and IL-17 by using ELISA kits (Shanghai Jiang Lai Biotechnology Co. Ltd., China) according to the manufacturer's instructions.

All data were analyzed with the GraphPad Prism 5 software and expressed as mean ± SD. Statistical significance was tested using Student's t-test. p < 0.05 was considered as statistically significant (^*p < 0.05, ^**p < 0.01, ^***p < 0.001; ns, not significant).

An univalent epitope-based vaccine CTB-UE against H. pylori urease composed of molecular adjuvants CTB and tandem copies of urease B and Th cell epitopes was constructed in previous study (Guo et al., 2014), as shown in Figure 2B. In order to obtain a vaccine with better protective effect, a multivalent epitope-based vaccine named CWAE (GenBank access number. MF402943) against Urease (Ure), NAP, HSP60, and HpaA was constructed. The structure diagram of multivalent epitope vaccine CWAE is shown in Figure 2A. The epitope-rich regions UreB_158–251 and UreB_321–385 in CWAE were obtained by using online T Cell Epitope Prediction Tools and B Cell Epitope Prediction Tools. The epitope-rich region UreB_158–251 was found to contain four known Th or B cell epitopes [UreB_158–172 and UreB_181–195 (Qiu et al., 2010), UreB_211–225 (Li et al., 2008), UreB_229–251 (Shi et al., 2007)]. The epitope-rich region UreB_321–385 was also found to contains three known Th or B cell epitopes [UreB_327–334 (Hirota et al., 2001), UreB_349–363 (Qiu et al., 2010), and UreB_373–385 (Yang et al., 2013)]. In addition, the other four B or Th cell epitopes UreA_27–53 (Rizos et al., 2003), UreA_183–203 (Hifumi et al., 2006), HpaA_132–141 (Chaturvedi et al., 2001), and HSP60_189–203 (Yamaguchi et al., 2000) were also used as the components of multivalent vaccine. The theoretically optimal combination of cholera toxin B subunit (CTB), NAP, linkers, the tandem copies of the selected epitopes, and epitope-rich regions of UreB was determined by modeling and prediction using RANKPEP, MOE, and DNAstar software. CTB and NAP were selected as intra-molecular adjuvants, and linker (KK, GS, GGG, DPRVPSS) was used as a spacer between epitope tandem. In order to construct the recombinant expression vector pETCWAE containing the fusion gene CWAE, a synthetic WAE gene was synthesized and subcloned into pETC, as shown in Supplementary Figure 1a. The recombinant vectors pETC and pETCWAE was identified by enzyme restriction (Supplementary Figures 1b,c) and confirmed by DNA sequencing (data not shown).

The CWAE protein (about 70 KD) was largely expressed in inclusion body of E. coli BL21(DE3)/pETCWAE. Based on the SDS-PAGE, the content of CWAE in the inclusion bodies was about 60% of the total protein (Figure 3A; Lanes 3 and 5). After purification by Ni²⁺-NTA affinity chromatography, the purity of the CWAE protein was 95.8% as analyzed by SDS-PAGE (Figure 3A; Lanes 6 and 7) and computer scan. Besides, the CTB-UE protein was also mainly expressed in inclusion body (Figure 4B; Lane 2, 3, and 5) and had a relative high level of expression. The purity of the CWAE protein was 96.2% as analyzed by SDS-PAGE (Figure 3B; Lane 6, 7, and 8) and computer scan. The antigenic characteristics of CWAE and CTB-UE was identified by Western blot and GM1-ELSIA. Both CWAE and CTB-UE protein could be recognized by rabbit anti-H. pylori polyclonal antibody (Figure 3Ci). Similarly, the polyclonal antibody induced by the CWAE protein could react with H. pylori UreA, UreB, HpaA, Hsp60, and NAP (Figure 3Cii). The adjuvanticity of CTB component was analyzed by GM1-ELSIA. When GM1 was used as the coating protein, both CWAE and CTB-UE were able to bind GM1 (Figure 3D), but the ability of CWAE and CTB-UE to combine GM1 decreased compared with the positive control CTB. In addition, UreB could not bind GM1.

Mice were killed for determination of antibody level and specificity. Antibodies specific for large antigens (H. pylori urease, UreA, UreB, HSP60, HpaA, and NAP) were measured by ELISA. CWAE and CTB-UE could induce similar levels of antibodies specific to H. pylori urease, UreA, or UreB (Figures 4A–C). In addition, the CWAE vaccine could induce antibodies specific for HSP60 (Figure 5D), NAP (Figure 4E), and HpaA (Figure 4F), but CTB-UE could not induce these specific antibodies, which confirmed that CWAE vaccine could mount antibody responses against various virulence factors and adhesion factors of H. pylori.

To examine the epitope-specific antibodies in mice induced by the CWAE vaccine, the B cell epitope peptides (UreA_183–203, UreB_321–339, UreB_158–172, UreB_181–195, UreB_211–225, UreB_349–363, HpaA_132–141, and HSP60_189–203) in the CWAE vaccine were synthesized and tested. The results indicated that both CWAE and CTB-UE vaccine were capable of generating similar levels of antibodies directed specifically against the epitope peptides UreA_183–203 (Figure 5A) and UreB_321–339 (Figure 5B). In addition, only CWAE vaccine could induce high levels of antibodies against UreB_158–172 (Figure 5C), UreB_181–195 (Figure 5D), UreB_211–225 (Figure 5E), UreB_349–363 (Figure 5F), HpaA_132–141 (Figure 5G), and HSP60_189–203 (Figure 5H). These results indicate that the B cell epitopes in CWAE vaccine retained their immunologic functions. To further confirm the effects of antibodies induced by CWAE and CTB-UE on H. pylori urease activity, a urease neutralization assay was performed. Natural H. pylori urease was incubated with a serial dilution of IgG induced by CWAE, CTB-UE, or rCTB. IgG antibodies induced by CWAE or CTB-UE could inhibit H. pylori urease activity dose-dependently, and the inhibitory effect of specific IgG from mice immunized with CWAE was stronger (Figure 5I). However, the IgG induced by CTB had no obvious inhibition. This result indicates that CWAE vaccine could induce higher levels of neutralizing antibodies against H. pylori urease than the whole urease antigen, which may be due to the B cell epitope UreA_183–203, UreB_321–339, UreB_158–172, UreB_181–195, UreB_349–363, and UreB_211–225 in CWAE vaccine.

Since the CWAE vaccine showed good immunogenicity and immunological specificity, we next investigated whether oral immunization with CWAE vaccine could reduce the H. pylori load in the stomachs of Mongolian gerbils infected with H. pylori, and show better therapeutic effect than CTB-UE and Urease. Mongolian gerbils (already infected with H. pylori) were orally immunized with CWAE, urease or PBS. The H. pylori colonization in the stomach was analyzed by quantitative culture. The result showed that oral therapeutic immunization with CWAE, CTB-UE, or urease dramatically decreased the H. pylori loads in the stomachs of H. pylori-infected Mongolian gerbils compared with those using oral therapeutic immunization with PBS (Figure 6A). More importantly, the CWAE vaccine had a better reduction of bacterial burden than CTB-UE and Urease.

H. pylori can release large amounts of urease into the stomach. Urease decomposes urea into ammonia, which neutralizes gastric acid. Therefore, H. pylori urease is critical for H. pylori survival in the stomach. The results of rapid urease test showed that oral therapeutic immunization with CWAE, CTB-UE or urease could dramatically reduce the urease activity in the stomach, and the CWAE vaccine had a better treatment effect in decreasing urease activity in the stomach than Urease (Figure 6B). Unfortunately, there were not significant differences in urease activity between CWAE and CTB-UE vaccine, though the urease activity in the stomachs of Mongolian gerbils immunized with CWAE was lower than that of Mongolian gerbils immunized with CTB-UE.

Therapeutic effect of the CWAE vaccine was also analyzed by histopathological analysis of stomach tissue. High levels of leukocytes and neutrophils were found in the stomachs from H. pylori-infected Mongolian gerbils immunized with vehicle control PBS. In contrast, inflammation was weakened in the stomachs from H. pylori-infected Mongolian gerbils immunized with CWAE, CTB-UE, or Urease. Typical histological findings of gastric mucosa are shown in Figure 7A, and the results of histological scoring showed that the scoring grades of Mongolian gerbils immunized with CWAE were significantly lower than those of Mongolian gerbils immunized with Urease or CTB-UE (Figure 7B). Besides, Figure 7C show representative IHC results obtained with biopsies from mice immunized with CWAE, CTB-UE, Ure, and PBS, respectively. The IHC substantially confirmed the histopathologic observations.

Antibody responses in serum, gastric and intestinal mucus, and feces were analyzed in Mongolian gerbils by ELISA after therapeutic vaccination. Oral immunization with CWAE, CTB-UE, or Urease significantly raised the levels of serum IgG and IgA against H. pylori lysates compared with oral immunization with PBS. In addition, the CWAE vaccine could induce higher levels of IgG and IgA specific for H. pylori lysates than Urease and CTB-UE (Figures 8A,B), which may be due to NAP component and the B cell epitopes from HSP60 and HpaA. Secretory IgA (sIgA) antibodies in gastric mucus, intestinal mucus and feces was also measured. A modest level of sIgA was found in extracts from the gastric mucus, intestinal mucus, or feces. Oral immunization with the CWAE, CTB-UE, or Urease remarkably increased the levels of sIgA antibodies against H. pylori lysates. Moreover, the CWAE vaccine could induce higher levels of sIgA specific for H. pylori lysates than Urease and CTB-UE (Figure 8C).

To determine the ability of CWAE to elicit lymphocyte specific responses for H. pylori, splenic lymphocytes after therapeutic immunization with CWAE, CTB-UE, Urease, or PBS were stimulated with NAP, Urease, or H. pylori lysates (Figure 9A). Splenic lymphocytes from Mongolian gerbils immunized with CWAE, displayed significantly high proliferation after stimulation with NAP, Urease or H. pylori lysates, compared with lymphocytes from Mongolian gerbils immunized with PBS. However, lymphocytes from Mongolian gerbils, immunized with Urease or CTB-UE after stimulation with H. pylori NAP, had no significant proliferation compared with cells from Mongolian gerbils immunized with PBS. These results showed that the multivalent vaccine CWAE could induce lymphocyte responses against various H. pylori antigens, and NAP component in CWAE kept the function as molecular adjuvant.

Cytokines IL-4, IFN-γ, and IL-17 in the supernatants of splenic lymphocyte cultures were measured by ELSIA after stimulation with H. pylori lysates. The splenic lymphocytes from Mongolian gerbils immunized with CWAE, CTB-UE, or H. pylori urease produced significantly high levels of IL-4, IFN-γ, and IL-17 cytokines. In addition, the CWAE vaccine could induce higher levels of IFN-γ (Figure 9B), IL-4 (Figure 9C), and IL-17 (Figure 9D) cytokines than CTB-UE, indicating that the CWAE vaccine induced comprehensive T cell responses of various types which were stronger than CTB-UE. However, there were no significant differences in the levels of IFN-γ, IL-4, and IL-17 cytokines between the CWAE- and Ure-immunized Mongolian gerbils.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.