Feifei She, Department of Medical Microbiology, Fujian Medical University, Fuzhou 350004, Fujian, China. Tel: +86 591 83569309; email: firstname.lastname@example.org
Development of an effective vaccine for controlling H. pylori-associated infection, which is present in about half the people in the world, is a priority. The H. pylori outer inflammatory protein (oipA) has been demonstrated to be a potential antigen for a vaccine. In the present study, use of oipA gene encoded construct (poipA) for C57BL/6 mice vaccination was investigated. Whether co-delivery of IL-2 gene encoded construct (pIL-2) and B subunit heat-labile toxin of Escherichia coli gene encoded construct (pLTB) can modulate the immune response and enhance DNA vaccine efficacy was also explored. Our results demonstrated that poipA administered intradermally (‘gene gun’ immunization) promoted a strong Th2 immune response, whereas co-delivery of either pIL-2 or pLTB adjuvant elicited a Th1-biased immune response. PoipA administered with both pIL-2 and pLTB adjuvants promoted a strong Th1 immune response. Regardless of the different immune responses promoted by the various vaccination regimes, all immunized mice had smaller bacterial loads after H. pylori challenge than did PBS negative and pVAX1 mock controls. Co-delivery of adjuvant(s) enhances poipA DNA vaccine efficacy by shifting the immune response from being Th2 to being Th1-biased, which results in a greater reduction in bacterial load after H. pylori challenge. Both prophylactic and therapeutic vaccination can achieve sterile immunity in some subjects.
B subunit heat-labile toxin gene encoded pVAX1 construct
Helicobacter pylori outer membrane inflammatory protein gene encoded pVAX1 construct
Helicobacter pylori is a spiral bacterium that colonizes the gastric mucosa of more than half the world's population. Infection with H. pylori is associated with the development of chronic active gastritis, peptic ulcer and gastric cancer (1). The current management of H. pylori infection relies on antibiotic therapy. This strategy has many drawbacks, including treatment failure due to emergence of bacterial resistance, poor patient compliance, side effects of the antibiotics and high cost of treatment. Two significant drawbacks of antibiotic therapy are the failure to prevent re-infection and the increase in bacterial resistance (2). Therefore, there is an urgent need to identify alternative approaches to prevention and treatment of H. pylori infection.
Some researchers have explored and tested vaccination as a strategy for combating H. pylori and found it to be effective (3–7). A number of candidate antigens, including H. pylori urease A/B subunits, catalase, neutrophil activating protein, and heat shock proteins, administered either as a single antigen or co-delivered with various adjuvants, have been reported to induce protective effects in prophylactic and therapeutic studies (5,8,9).
Recently, researchers identified a novel virulence factor: outer inflammatory protein, designated as oipA, which is encoded by an inflammation-related gene located approximately 100 kb from the Cag pathogenicity island on the chromosome (10). It correlates significantly with the clinical outcome of H. pylori infection and has been demonstrated to be a potent protective antigen (11,12).
A priority in H. pylori vaccine development has been to clarify the types of immune responses that are effective in reducing the bacterial load of H. pylori. Previous studies have demonstrated that a Th2 immune response is required for prevention of H. pylori infection. However, a Th1-biased immune response is required for eradication of H. pylori colonization. Hence, an ideal vaccine against H. pylori would mainly trigger Th1-biased immune responses. Studies that have attempted to shift the immune response after DNA vaccination from a Th2 to a Th1 type have emphasized the role of adjuvants in modulating the type of immune response (13–15).
DNA vaccination is an important immunization strategy that has many of the characteristics of an ideal vaccine, including the elicitation of broad immune responses and long-lasting immunity, simplicity and cost-benefit efficacy (6). However, because DNA vaccines tend to trigger a Th2-biased response, they are unattractive as anti-H. pylori vaccines, which must efficiently prime a Th1 immune response(7). In the present study, we aimed to construct an oipA-encoding DNA vaccine and to deliver it with mouse IL-2 gene and LTB gene encoding plasmids as adjuvants. We also evaluated the efficacy of the DNA vaccine against H. pylori challenge in C57BL/6 mice.
MATERIALS AND METHODS
Five-week old female C57BL/6 mice were purchased from Shanghai Laboratory Animal Center (Shanghai, China). All animals were maintained under specific pathogen-free conditions at the Fujian Institute of Medical Sciences (Fuzhou, China). Mice were housed in micro-isolator cages and provided with autoclaved food, water, and bedding. Mice were fed and watered ad libitum. All animal experiments were approved by the Animal Ethical and Experimental Committee of the Fujian Medical University.
Helicobacter pylori strain
Mouse-adapted H. pylori strain SS1 was sub-cultured on 5% sheep blood Columbia agar with Dent's H. pylori selective supplement (Oxoid, Cambridge, UK), and grown at 37°C in a 5% CO2 microaerophilic incubator for 4–5 days, as described previously (16). Then H. pylori was incubated in Brucella broth supplemented with 10% (v/v) heat-inactivated FBS (Invitrogen, Carlsbad, CA, USA), and placed on a rotor shaker at 35°C in 5% CO2 in an incubator. The liquid culture was harvested at mid-log phase (OD600 0.4–0.6) and pelleted by centrifugation. For H. pylori challenge, the pellets were re-suspended in sterile PBS, and adjusted to 108 CFU in 0.1 mL of PBS per dose. Quantitative culture of H. pylori and histopathological evaluation of gastric tissues demonstrated that this challenge dose resulted in nearly 100% infection rates in C57BL/6 mice.
Construction of recombinant plasmids expressing outer inflammatory protein, interleukin-2, and B subunit heat-labile toxin
Helicobacter pylori SS1 and enterotoxigenic E. coli H44815 (purchased from the National Institute for the Control of Pharmaceutical and Biological Products, Beijing, China) genomic DNA were prepared by QIAamp DNA Mini Kit (Qiagen, Valencia, CA, USA) as described by the manufacturer. DNA was dissolved in 100 μL distilled water, and used directly as a template for the PCR. The plasmid containing mouse IL-2 gene was a gift from Professor Kongwei (Jilin University, China). The primers for oipA, IL-2 and LTB were designed with primer3 software. For oipA, the primers used were F: 5′-AACTGCAGATGAAAAAAGCTCTCTTAC-3′, and R: 5′-CCGCTCGAGATGTTTGTTTTTAAAG-3′. For IL-2, the primers used were F: 5′-AACTGCAGATGTACAGCATGCAGCTCGC-3′, and R: 5′-CCGCTCGAGTTATTGAGGGCTTGTTG-3′. For LTB, the primers used were F: 5′-AACTGCAGATGAATAAAAGTAAAATGTTATG-3′, and R: 5′-CCGCTCGAGCTAGTTTTCCATACTGATTGCCGC-3′. The forward primers all contained a Pst I restrictive endonuclease site (underlined), and the reverse primers all contained a Xho I restrictive endonuclease site (underlined). The PCR mixture (50 μL) consisted of 10 mM Tris–HCl (pH 8.3), 50 mM KCl, 4.0 mM MgCl2, 200 mM deoxynucleoside triphosphates, 1.0 Unit of pfu DNA polymerase (Stratagene, La Jolla, CA, USA), 20 pmol of each primer, and 1 μL of DNA template. A typical PCR program was comprised of 2 min at 95°C for pre-denaturizing, then 30 cycles of 30 sec at 95°C, 45 sec at 55°C, and 90 sec at 72°C, followed by 10 min at 72°C for extension. The PCR product lengths were 924 bp, 510 bp and 375 bp, respectively. The PCR products were separated using a QIAquick Gel Extraction Kit (Qiagen). The purified DNA fragments were digested with Pst I and Xho I and inserted into the Pst I and Xho I sites of pVAX1 (Invitrogen, San Diego, CA, USA). The pVAX1 containing no insert was used as a control (mock control). All plasmids were transformed into E. coli DH5α host strain and screened with kanamycin (50 μg/mL) containing LB agar plates. Kanamycin-resistance clones were confirmed by PCR, Pst I and Xho I restrictive endonuclease digestion. The sequences were detected using an Applied Biosystems model automatic sequencer. The constructs were designated as poipA, pIL-2 and pLTB, respectively.
DNA vaccine preparation
For preparation of DNA vaccine, the plasmid containing stains were cultured in LB broth and harvested at mid-log phase. The pellets were harvested by centrifugation. The plasmids were extracted with an Endofree plasmid Mega kit (Qiagen) as described by the manufacturer. The plasmid concentrations were determined by ultraviolet spectrophotometer and adjusted to 1 mg/mL with endotoxin-free water.
Immunization and challenge
For immunization, anaesthetized mice were immunized intradermally three times, at 2 week intervals, in shaved abdominal skin with three non-overlapping shots of plasmid-coated gold particles using a helium-driven gene gun (Bio-Rad, Hercules, CA, USA) with a discharge pressure of 400 psi. At every administration, each mouse received 3 μg poipA. pIL-2 or/and pLTB plasmids were co-precipitated onto gold particles as adjuvants and 1 μg/shot administered. The negative controls received 5 μL PBS intradermally, whereas the mock control group received three intradermal immunizations of 3 μg pVAX1. Four weeks after the last immunization, the mice were inoculated intragastrically at 2-day intervals with three doses of 108 CFU of H. pylori SS1 bacterial suspension in 0.1 mL brucella broth, as described previously (16).
Antigen-specific antibody assays
For the H. pylori-specific antibody assay, blood was withdrawn from an orbital vein of each mouse 4 weeks after final immunization. The serum IgG1 and IgG2a titers were determined by indirect ELISA. Briefly, U-bottomed, 96-well ELISA plates (eBioscience, San Diego, CA, USA) were coated with 5 μg/well oipA recombinant protein in PBS overnight at 4°C. The plates were washed with 0.05% Tween20 PBS and then blocked with 2% goat serum PBS for 2 hr at 37°C. Sera were added at an initial dilution of 1:100 in duplicate, with 1:5 serial dilutions performed in 2% goat serum PBS. All plates were incubated for 1 hr at 37°C and then washed with 0.05% Tween20 PBS five times. A 1:500 dilution of horseradish peroxidase-labeled goat anti-mouse IgG1 and IgG2a (Santa Cruz Biotechnology, Santa Cruz, CA, USA) was added to the plates for 1 hr at 37°C. All plates were again washed, then developed with 3,3′,5,5′-tetramethylbenzidine (Sigma-Aldrich, Louis, MO, USA) for 15 min, and stopped with 2 M HCl. The optical density of each well was determined at 450 nm on a microplate reader (Bio-Tek, Winooski, VT, USA).
Quantitative culture of Helicobacter pylori from gastric tissue
The mice were killed 4 weeks after H. pylori challenge and their stomachs and spleens sterilely collected. The stomachs were divided in half longitudinally, placed in 500 μL of Brucella broth, weighed, and homogenized using a sterile ground-glass pestle. Serial ten-fold dilutions were plated on H. pylori selective agar medium and incubated at 35°C in a 5% CO2 microaerophilic incubator for 5 days. Identification of H. pylori was based on its small translucent gray colonies, Gram-negative curved or comma-shaped rods, and positive reactions in oxidase, catalase and urease tests. The CFUs per gram of stomach were calculated by enumerating colonies, adjusting for the dilution, and dividing by the tissue weight.
Antigen specific cytokines assay
Tissues were ground through a screen mesh. Red blood cells from the spleen samples were lysed with ACK lysis buffer (Biosource, Camarillo, CA, USA). Two million splenocytes per plate were incubated with 30 μg/mL of H. pylori sonicate in complete RPMI 1640 containing 5% FBS, 2 mM glutamine, penicillin, streptomycin, nonessential amino acids, sodium pyruvate and 10 mM HEPES (all from Invitrogen/Gibco BRL, Grand Island, NY, USA) for 24 hr at 37°C. Following incubation, the supernatants were collected and concentrations of cytokines (IFN-γ, IL-2, IL-4, IL-10, and IL-12) determined according to the manufacturer's instructions using a mouse cytokines detection system (BD Biosciences, San Jose, CA, USA). The optical density of each well was read and data were analyzed by Biotek software (Bio-Tek) using five parameter logistics.
One way ANOVA and the Student-Newman-Keuls test were performed using SigmaPlot (Systat Software, San Jose, CA, USA) to compare statistically significant differences among groups. Differences between groups were considered statistically significant at P < 0.05.
In the immunized mice, H. pylori oipA-specific IgG1 and IgG2a isotypes were determined from sera after immunization. IgG1 titer and IgG2a titers were low in the PBS negative and pVAX1 mock controls, these showed no statistically significant difference (P < 0.05). In the poipA immunized mice, although IgG2a was slightly higher than in the mock controls (P < 0.05), IgG1 was significantly higher than in the mock controls (P < 0.001). The ratio of IgG2a/IgG1 was less than that of the mock controls, indicating a strong Th2 immune response. In the groups receiving poipA co-delivered with either pIL-2 or pLTB, both IgG1 and IgG2a were significantly higher than in the mock control group (P < 0.001), but the ratios of IgG2a/IgG1 were both greater than those of the mock controls, indicating a Th1-biased immune response. In poipA co-delivery of pIL-2 and pLTB group, both of them were significantly higher than that of the mock control group (P < 0.001). The ratios of IgG2a/IgG1 were significantly greater than that of the mock control and other groups, which indicating a strong Th1 immune response (Fig. 1). Altogether, these results suggest that poipA co-delivered with pIL-2 and/or pLTB promotes a Th1-biased immune response.
Vaccine induced antigen specific cytokine responses after challenge
To confirm the Th1 polarizing effects of immunization with poipA co-delivered with pIL-2 and/or pLTB adjuvants, we assessed production of antigen specific cytokines after challenge. Although all the immunized mice showed more IFN-γ and IL-4 secretion by splenocytes than the PBS negative and pVAX1 mock controls (P < 0.05), only the mice immunized with poipA co-delivered with pIL-2 and/or pLTB secreted significantly more IFN-γ than the pVAX1 mock controls (P < 0.001). Meanwhile, in the groups in which poipA was co-delivered with either pIL-2 or pLTB, secretion of IFN-γ by splenocytes was significantly greater than in the poipA alone group (P < 0.001). In the group in which poipA was co-delivered with both pIL-2 and pLTB, IFN-γ secretion by splenocytes was as much as three times that of the poipA alone group (P < 0.001). Conversely, IL-4 secretion by splenocytes from poipA mice was significantly greater than in the PBS control groups (P < 0.001) and other groups (P < 0.05) (as shown in Fig. 2a).
To evaluate inflammation and immunity, we also assessed IL-2, IL-10 and IL-12 production by immunized mice following H. pylori challenge. In the groups in which poipA was administered with either pIL-2 or pLTB, IL-2 concentrations were significantly greater than in the mock control (P < 0.001) and poipA groups (P < 0.001). In the group in which poipA was administered with both pIL-2 and pLTB, the immunized mice showed an approximately four-fold increase in IL-2 secretion as compared to that of the mock group (P < 0.001) (Fig. 2b). Production of IL-12 was significantly increased in all immunized mice compared to the mock control group (P < 0.05). In the groups in which poipA was administered with either pIL-2 or pLTB, IL-12 was significantly greater than in the poipA group (P < 0.05), whereas in the group in which poipA was administered with both pIL-2 and pLTB, immunized mice showed a two-fold increase in IL-12 secretion compared to the pVAX1 group (P < 0.001) (Fig. 2c). In the poipA group and the groups in which poipA was administered with both pIL-2 and pLTB, splenocyte production of IL-10 were significantly greater in the immunized mice than in the control groups and other groups (P < 0.001) (as shown in Fig. 2D).
Protection against infection of the stomach with Helicobacter pylori SS1
Having determined antibodies and cytokines responses, we next examined whether immunization with poipA co-delivery of pIL-2 and/or pLTB would protect against oral H. pylori challenge or not. Four weeks after last immunization, immunized mice were challenged with three doses of 108 CFU of H. pylori SS1. In the PBS negative control and pVAX1 mock control groups, H. pylori remained at high density in the mouse stomachs. Mice immunized with poipA had significant drops in bacterial load compared to the PBS negative and pVAX1 mock control groups (P < 0.001). In the groups in which poipA was administered with either pIL-2 or pLTB, the mice had an almost 2-log decrease in bacterial load compared to that of the mock controls (P < 0.001). Mice in the group in which poipA was administered with both pIL-2 and pLTB showed an almost 4-log decrease in bacteria load compared to that of the mock control group (P < 0.001). Furthermore, two mice had no detectable H. pylori with both the quantitative culture method and PCR-based detection of H. pylori urease C and 16S rRNA genes. These data demonstrate that immunization with poipA co-delivered with pIL-2 and pLTB improves protection against H. pylori challenge and can sometimes achieve sterile immunity (Fig. 3).
In the present study we demonstrated that an H. pylori oipA encoding construct is capable of inducing humoral and cellular responses in immunized mice. The antibody response profiles elicited by the DNA vaccine alone administered intradermally (the gene gun method) showed that it produced a Th2 immune response, while co-delivery of IL-2 and LTB gene encoding constructs promoted a Th1-biased immune response. Regardless of the different immune responses promoted by the various combined vaccination protocols, all immunized mice had reduced bacteria loads after H. pylori challenge. pIL-2 and pLTB enhanced the efficacy of the poipA DNA vaccine by shifting the immune response from a Th2 to a Th1 type, the latter playing an important role in both control and eradication after H. pylori challenge.
Although different vaccination protocols in various animal models of infection and in human trials have been studied (17), the protective mechanism against H. pylori infection is still unclear. Data obtained with transgenic mice suggest that major histocompatibility complex-II restricted CD4-positive T cells may play a fundamental role in protection (18, 19). In addition, it has been proposed that a shift from the Th1 cytokine response induced by H. pylori infection to a Th2 response is a protective mechanism (20). However, recent studies have contradicted the Th2 paradigm and demonstrated the importance of the Th1 response in effective protection following vaccination against H. pylori (21–23). Blanchard et al. have also proposed a role for regulatory T cell in H. pylori vaccination and suggested that the site of T cell activation influences protection (24). Therefore, the definite protective mechanism against H. pylori challenge still needs to be explored. Nevertheless our results show that either poipA alone, or poipA co-delivered with pIL-2 or/and pLTB, promotes protective immunity, via Th2 and Th1-biased immune responses, respectively. However, mice administered poipA co-delivered with pIL-2 or/and pLTB have a significantly reduced bacterial load. In our preliminary study, in which H. pylori-infected mice were vaccinated intradermally with poipA, co-delivered with pIL-2 or/and pLTB, similar results were achieved by therapeutic vaccination (data not shown). Therefore, poipA co-delivered with pIL-2 and pLTB could be a promising candidate for the development of prophylactic and therapeutic vaccines against H. pylori.
Several factors contribute to the immune responses to DNA vaccine, including the properties of the gene encoding the antigen, the route and dose of administration, and co-delivery of adjuvants. Todoroki et al. demonstrated that H. pylori heat shock proteins encoding pcDNA3.1-HspA construct confer less protection against the bacterium than does the pcDNA3.1-HspB construct (3). Though the pcDNA 3.1-HspA construct induced higher IgG and IgA responses, the pcDNA 3.1-HspB construct resulted in a Th0-like response and greatly reduced inflammation in infected mice. In our previous study, intramuscular injection of a high dose (100 microgram per dose) of oipA-encoding pVAX1-oipA construct stimulated Th1-biased responses (unpublished data). However, intradermal delivery of nanogram amounts (about 5 microgram per dose) of the same construct via gene gun favors priming of Th2-biased responses. This is in agreement with the findings of Lin et al. (25). Because DNA vaccination tends to prime Th2 responses preferentially, it has seemed relatively unattractive as a H. pylori vaccine, for which priming of Th1 immune responses is the main requirement.
In order to overcome the inferior, low magnitude, immune response elicited by DNA vaccination, researchers have explored the role of various adjuvants in modulating and enhancing the immune response primed by immunization. The CT from Vibrio cholerae and LT from enterotoxigenic strains of E.coli are potent mucosal adjuvants. It has been found that CT induces a Th2 response (26), whereas LT induces a Th1-biased immune response (27, 28). The B subunits of these toxins are less potent than the active toxins. However, they have practical potential as carrier molecules, having shown less toxicity in both experimental animals and human immunization (29–31). Research has also shown that co-administration of cytokine proteins or cytokine gene-encoded plasmids enhances comprehensive humoral and cellular immune responses (32–35). Recently, synthetic CpGs, which prime a Th1 immune response and have shown some encouraging results in H. pylori vaccines, have been approved for use as adjuvants in humans (21, 23, 36–38). In this study, we used pIL-2 and pLTB as adjuvants in poipA DNA vaccination. Our results indicate that co-delivery of adjuvants modulates the type and magnitude of immune responses. Furthermore, we found that two of the immunized mice developed sterile immunity, demonstrating that effective adjuvants can eradicate the organisms after H. pylori challenge.
The ability of DNA vaccines to induce production of cytokines depends on the use of adjuvants. Our data demonstrate that mice immunized with poipA co-delivered with pIL-2 and/or pLTB have significant increases in IFN-γ secretion with no significant changes in IL-4 secretion, indicated elicitation of a Th1-biased immune type. All mice in all immunization groups had significantly increased IL-2, IL-10, and IL-12 secretion. However, the secretion of IL-2, IL-10, and IL-12 is not specifically associated with either Th1 or Th2 immune responses. Recent studies have also claimed that IL-12 might be an important mediator against H. pylori infection. Akhiani et al. demonstrated that IL-12 is necessary for protection against H. pylori infection in IL-12 knockout mice, which are unable to induce protective immune responses upon challenge (39). Taylor et al. also demonstrated that increases in IL-12 secretion correlate well with long-term protection (40). In addition, IL-10 has proven to be an important regulator of the inflammatory response to H. pylori. Akhiani et al. studied IL-10-deficient mice that had developed severe hyperplastic gastritis (41). They proposed that expression of IL-10 may be an anti-inflammatory marker for predicting the severity of post-immunization gastritis in vaccine studies. The relationship between IL-10 and post-immunization gastritis and the histopathology primed by H. pylori vaccination should be investigated.
pVAX1, the first Food and Drug Administration-approved vector used safely in animal and human DNA vaccines, was constructed by modifying the pcDNA3.1 vector. In the present study, we demonstrated the protective ability of DNA vaccine encoding H. pylori oipA using the pVAX1 plasmid. In addition, we also constructed IL-2 and LTB gene encoding plasmids separately. We did not use a bicistronic vector to construct the DNA vaccine, because we believed that preparing DNA vaccine and adjuvants separately would provide more flexibility. For example, in a prime and boost immunization strategy, an antigen can be administered alone or co-administered with adjuvants. Multiple antigens can be constructed separately and co-delivered with the same adjuvants. To our knowledge, this is the first study showing that a combination of IL-2 and LTB gene encoding constructs shifts immune responses and enhances the efficacy of DNA vaccine. In future studies, various ‘cocktails’ and proportions of DNA vaccines and adjuvants and various routes of administration should be investigated to achieve the most effective protection against H. pylori.
We thank Professor Fu-Chou Cheng for critically reading this manuscript. This work was supported by the Key Sci-Tech Research Foundation of Fujian Province of China (No.2009Y0024), the Natural Science Foundation of Fujian Province of China (No. 2009J01145), the Research Foundation of Education Bureau of Fujian Province (No. JA08098), and the Key Program of Scientific Research of Fujian Medical University of China (No.09ZD018).