Disruption of sncRNA Improves the Protective Efficacy of Outer Membrane Vesicles against Helicobacter pylori Infection in a Mouse Model

ABSTRACT

The outer membrane vesicles (OMVs) secreted by Helicobacter pylori contain various bacterial components, such as proteins, phospholipids, toxins, and nucleic acids, including small noncoding RNA (sncRNA), which have regulatory functions in cell envelope structure, metabolism, bacterial communication, biofilm formation, and virulence. We previously showed that knocking out sncRNAs sR-989262 and sR-2509025 at the cellular level increased interleukin 8 (IL-8) levels in mice exposed to OMVs. In this study, we show that immunization with ΔsR-989262 and ΔsR-2509025 OMVs intragastrically significantly increased immunoglobulin G (IgG) and secreted IgA levels in mice compared to wild-type OMVs and without weight changes, which indicated that sncRNA-deficient OMVs are relatively safe to immunize mice. The detection of IgG subtypes IgG1 and IgG2c showed that the sncRNA-deficient OMVs primarily stimulate the T helper 2 (Th2)-mediated immune response. Moreover, levels of the cytokines IL-4, IL-13, gamma interferon (IFN-γ), IL-12 (p40), IL-8, and IL-17 indicate that ΔsR-989262 and ΔsR-2509025 OMVs trigger the Th2-type immune response but primarily trigger a Th1-mediated and Th17-mediated immune response. These findings show that OMV-encapsulated sncRNA plays an important role in regulating the immune response in hosts infected by H. pylori at the animal level. Moreover, they show that knocking out of sR-989262 and sR-2509025 improves the immunogenicity and protective efficacy of OMVs, and this may be beneficial to the design of OMV-based H. pylori vaccines.

INTRODUCTION

Helicobacter pylori is a Gram-negative bacterium that specifically colonizes the human gastric mucosa (1). Persistent H. pylori infection can lead to gastritis, gastric ulcer, duodenal ulcer, and even gastric cancer (2). The different clinical characteristics of chronic H. pylori infection may be related to the interaction of the host’s microenvironment and bacterial infection. However, the precise immune regulatory mechanism of H. pylori causing host infection remains unclear. Therefore, a better understanding of the interaction between H. pylori and the host can facilitate the development of novel vaccine design strategies for preventing and treating H. pylori infections.

Outer membrane vesicles (OMVs) are a type of nanoparticle spontaneously produced by bacteria that are about 20 to 200 nm in size (3). Similar to other Gram-negative bacteria, H. pylori can secrete OMVs. Proteomics analyses have shown that OMVs are composed of proteins, nucleic acids, toxins, and phospholipids, similar to bacterial outer membranes. However, their structure is unique compared to the compounds on the surface of bacteria (4, 5). H. pylori OMVs play an important role in the interaction mechanism inducing chronic H. pylori infection and promoting bacterial survival in the host, including bacterial adhesion, nutrient acquisition, and interaction and communication between bacteria and host cells (6). In addition, OMVs can induce the expression and release of proinflammatory cytokines and chemokines to attract immune cells into the infection site, effectively eliminating bacterial infections (7, 8). Therefore, OMVs represent an effective antigen or adjuvant target for vaccine design. We have previously explored the potential of H. pylori OMVs as vaccines, finding that wild-type (wt) OMVs can provide effective protection against H. pylori infection in mice. Moreover, as an adjuvant, they can enhance stimulation of an immune response biased toward T helper 1 (Th1) and Th17, facilitating the elimination of H. pylori (9, 10). Nevertheless, the continued exploration of design strategies to enhance the immune-stimulating ability of OMVs is required to design more effective H. pylori vaccines.

The regulation of bacterial gene expression by small noncoding RNAs (sncRNAs) is a recently emerged research area. In many RNA-based regulatory mechanisms, sncRNAs have unique characteristics in many cellular processes. Moreover, they are key regulators of bacterial gene expression and the posttranscriptional expression of target genes by binding to their mRNA (11, 12). Intracellular sncRNAs have regulatory functions in H. pylori, including regulating cell membrane structure, metabolism, bacterial communication, biofilm formation, and virulence (13, 14). Studies have found that sncRNAs in H. pylori participate in the posttranscriptional translation of H. pylori pathogenic factors, such as cytotoxin-associated gene A (CagA), and regulate bacterial infection and adhesion (15). In our previous studies, we confirmed the presence of two sncRNAs in the H. pylori OMVs, sR-989262 and sR-2509025, which can be delivered by wt OMVs to reduce the secretion of interleukin 8 (IL-8) and help bacteria evade the host immune response; knocking out these sncRNAs increased IL-8 levels in response to OMVs at the cellular level (16). These two sncRNAs highlight the great potential in immune regulation and can facilitate the design of OMV-based H. pylori vaccines. Therefore, in this study, we assessed whether knocking out the sR-989262 and sR-2509025 sncRNAs encapsulated in OMVs affects their immune stimulation at the animal level. We provide corroboratory evidence that OMVs containing sncRNA play an important role in the immune regulation process of H. pylori at the animal level, and this provides a solid foundation for developing new OMV-based H. pylori vaccines.

RESULTS

ΔsncRNA OMVs are safe and stimulate IgG, vaginal IgA, and stomach IgA production.

To determine whether OMVs secreted by H. pylori missing sncRNAs sR-2509025 and sR-989262 are toxic to cells, we purified OMVs produced by wt and mutant strains and incubated RAW 264.7 cells with different concentrations of them for 24 h. The lipopolysaccharide (LPS) group was used as the positive control, the ΔsR-7497631 OMV group was used as the negative control, and the cell lysis group was used for quality control. The viability of macrophages treated with the purified OMVs from the different groups did not have apparent cytotoxic effects compared with LPS (Fig. 1B).

FIG 1.

(A) Animal immunization program. Mice (6 weeks old, 16 to 22 g) were divided into groups of five. Mice were immunized with OMVs by the intragastric route. Booster immunizations were implemented at week 4. Blood samples and vaginal secretions were collected repeatedly, and then mice were orally challenged with a lethal dose of H. pylori SS1 after booster immunization. The antibody concentration was measured 4 weeks after immunization. Animal experiments were performed twice, and the data were combined for analysis. (B) Safety evaluation of different OMVs at the cell level. RAW 264.7 macrophages were treated with different doses of ΔsR-2509025 and ΔsR-989262 OMVs, with the LPS group as a positive control, ΔsR-7497631 OMVs as a negative control, and the cell lysis group as a quality control group. The means were compared using the least significant difference test. A P value of <0.01 reflects statistical significance between the comparison groups.

In addition, the safety of ΔsncRNA OMVs compared to the control group in mice was assessed. No animal death or disease signs were observed in any group by the end of the observation period, and the physical activity, appetite, and general outward condition of mice were moderate. Assessment of percent change in body weight over 30 days after first immunizations showed an increase in body weight for all groups, and mean body weight in experimental groups did not differ significantly from that in the control group (P > 0.5) (Fig. 2A).

FIG 2.

(A) Percentages of body weight change of mice in various groups recorded daily 30 days after first immunization. (B to D) Concentrations of anti-HP OMP IgG (B) and secretory IgA (C) in serum and the concentrations of stomach IgA by anti-HP OMPs (D) in stomachs were measured by ELISA at the fourth and eighth week. The results of IgG analysis are expressed as means ± SD. The means were compared using the least significant difference test. P levels of <0.05 and <0.01 represent significant differences between the comparison groups.

The blood and vaginal secretions of mice were collected in the fourth and eighth weeks after their first immunization to determine the immune-stimulatory abilities of OMVs secreted by the ΔsR-2509025 and ΔsR-989262 strains. We found that serum IgG levels of mice immunized with ΔsR-989262 OMVs were higher than those of wt OMVs, ΔsR-2509025 OMVs, and controls after 4 weeks. However, by the eighth week after immunization, both ΔsR-989262 and ΔsR-2509025 OMVs showed a strong humoral immune response stimulation ability (Fig. 2B). Interestingly, we found that secretory IgA (sIgA) levels were significantly different in vaginal secretions and gastric mucosa of the ΔsR-2509025 OMVs group compared to the control group (P < 0.01). The ΔsR-2509025 OMVs group had higher IgA levels in vaginal secretions than the control group. However, while the ΔsR-989262 OMVs group appeared to have a higher IgA level, it was not significantly different. In addition, stomach sIgA levels in the ΔsR-989262 and ΔsR-2509025 OMVs groups differed significantly from the control group (P < 0.01) (Fig. 2C and D). Therefore, disrupting sR-989262 and sR-2509025 sncRNAs improved OMV immunogenicity and their ability to stimulate mucosal immunity.

ΔsncRNA OMVs induce a Th2-type immune response, based on levels of IgG subtypes.

We next explored the secretion levels of the two subtypes of IgG, IgG1 and IgG2c, that reflect the type of immune response to better understand the immuno-protective molecular mechanism of OMVs. Previous studies found that high IgG1 response levels, indicated by higher IgG1/IgG2a or IgG1/IgG2c ratios, are usually associated with Th2 responses in mice, while high IgG2c response levels, indicated by lower IgG1/IgG2a or IgG1/IgG2c ratios, are usually associated with Th1 responses (17). Based on the estimated IgG subtype distribution in mouse serum, we found that OMVs stimulated mice to produce higher levels of IgG1 and IgG2c compared to controls (P < 0.01). In addition, IgG1 and IgG2c levels showed an upward trend with increasing immunization time. However, IgG1 levels were higher than IgG2c levels (Fig. 3). Moreover, ΔsR-2509025 and ΔsR-989262 OMVs stimulated mice to produce IgG1 at levels that differed significantly from levels with wt and ΔsR-7497631 OMVs (Fig. 3A). However, IgG2c levels were significantly higher in the ΔsR-989262 OMVs group than in the other groups, but only in the eighth week after immunization (P < 0.01) (Fig. 3B). Therefore, we hypothesize that sncRNA disruption enhances the ability of OMVs to trigger an immune response, which is biased toward Th2.

FIG 3.

Concentrations of anti-Hp IgG1 (A) or anti-Hp IgG2c (B) in groups that received OMP as vaccine were determined by ELISA against OMPs isolated from H. pylori. Each group was comprised of 9 mice. The exact concentration of IgG1 and IgG2c subclass antibodies in serum samples of mice at 4 weeks and 8 weeks after immunization are shown. The means were compared using the least significant difference test. A P value of <0.01 reflects statistical significance between the comparison groups.

ΔsncRNA OMVs induce Th1- and Th17-type immune responses, based on levels of cytokines.

To further explore the type of immune response induced by sncRNA-deficient OMVs, cytokine levels in the supernatants of primary splenocytes harvested from mice and stimulated with OMP antigen were quantified. Gamma interferon (IFN-γ) and IL-12 are considered indicators of the Th1 polarization response, in which IFN-γ is secreted by Th1 cells and IL-12 promotes CD4⁺ cells to transform into Th1 cells (18). Similarly, IL-4 and IL-13 are indicators of Th2 polarization responses (19). We found that splenocytes produced higher cytokine levels in the OMV groups than in the control group (Fig. 4). IL-4 levels stimulated by ΔsncRNA OMVs were significantly lower than those of wt OMVs (P < 0.01). In addition, while IL-13 levels in the ΔsncRNA and wt OMV groups differed significantly from those of the control group (P < 0.01), they did not differ significantly from each other (Fig. 4B). Moreover, after ΔsR-2509025 OMV immunization, IFN-γ levels in mouse spleen cells were significantly higher than those in animals that received wt, ΔsR-989262, and ΔsR-7497631 OMVs (P < 0.01) (Fig. 4C). Furthermore, we found that ΔsR-2509025 and ΔsR-989262 OMVs induced greater IL-12 production than wt and ΔsR-7497631 OMVs (Fig. 4D). In addition, ΔsR-2509025 and ΔsR-989262 OMVs enhanced IL-8 secretion, consistent with our previous study (16) (Fig. 4E). Finally, the Th17-type immune response in mice immunized with ΔsR-2509025 and ΔsR-989262 OMVs was significantly enhanced compared with wt OMVs (Fig. 4F). Therefore, sncRNA disruption induces a Th2-type immune response, and unlike wt OMV, it also induces a Th1- and Th17-type immune response, which is more beneficial for clearing H. pylori infection.

FIG 4.

Cytokine production was detected in excised splenocytes from sacrificed mice at 8 weeks after first immunization. The concentrations of IL-4 (A), IL-13 (B), IFN-γ (C), IL-12 (p40) (D), IL-8 (E), and IL-17 (F) in groups that received H. pylori OMPs as stimulator were then subjected to cytokine-specific ELISAs. The results of cytokine production analyses are expressed as means ± SD (n = 18). Means were compared using the least significant difference test. P values of <0.05 and <0.01 reflect statistical significance between the comparison groups.

We also evaluated OMV immunization safety by assessing whether it induced mucosal inflammation. The levels of proinflammatory cytokines IL-6 and tumor necrosis factor alpha (TNF-α) were quantified with an enzyme-linked immunosorbent assay (ELISA) (20). We found that mucosal inflammation in mice immunized with wt and ΔsncRNA OMVs did not differ significantly from controls, and stomach IL-6 and TNF-α levels remained low in all groups (see Fig. S1 in the supplemental material).

ΔsncRNA OMVs reduce H. pylori infection.

The challenge protection experiment intuitively and effectively reflects the protective effect of OMVs against H. pylori infection after immunization in mice. The optical density (OD) of urease activity in stomach tissues obtained from the mice immunized with various OMVs was significantly lower than that in controls (Fig. 5A). Importantly, urease activity in the ΔsR-2509025 and ΔsR-989262 OMV groups was lower than in the wt OMV groups. In addition, we cultured H. pylori isolated from gastric tissue and measured the bacterial loading level of H. pylori. The colony count results were consistent with the urease activity results. The CFU counts of all OMVs groups were lower than the control group, and the ΔsR-2509025 and ΔsR-989262 OMV groups were significantly lower than wt and ΔsR-7497631 OMVs (P < 0.01) (Fig. 5B). Therefore, these tests show that sncRNA disruption inhibits bacterial loading in the stomach of C57BL/6 mice, indicating that ΔsncRNA OMVs can elicit effective immune protection against H. pylori infection in mice.

FIG 5.

OMVs as vaccine elicit protection against H. pylori SS1 infection. Stomach tissues were collected from immunized mice and washed in PBS buffer to determine the urease activity and for bacterial load quantitation. (A) Urease activity in stomach homogenates from mice with OMVs was assessed 4 weeks after the challenge infection. (B) H. pylori colony counts in stomach homogenates were quantified. The means were compared using the least significant difference test. P values of <0.05 and <0.01 reflect statistical significance between the comparison groups.

DISCUSSION

sncRNAs are a key component of the regulatory cascade, as they coordinate the expression of toxic genes in response to environmental or other changes. They directly act on either virulence genes or virulence gene regulatory genes, adapting their expression to stress and metabolic demands (21). Pathogenicity islands encoding sncRNAs IsrM play an important role in Salmonella invasion of epithelial cells, intracellular replication in macrophages, and virulence and colonization in mice (22). Sjöström and colleagues reported that RNA is one of the various bacterial components related to OMVs and plays an important role in the bacteria-host interactions mediated by them (23). Some studies have reported that sncRNAs exist in OMVs and play a role in bacteria-host communication (24, 25). One study on oral squamous cell carcinoma caused by Porphyromonas gingivalis found that its OMVs contained differently packaged small RNAs (sRNAs) that can potentially target host mRNA function (26). In addition, it was found that methionine tRNA was abundant in Pseudomonas aeruginosa OMVs, which reduced IL-8 secretion by primary human airway epithelial cells induced by LPS and OMVs. Moreover, sncRNAs attenuated OMV-induced mouse lung cytokine secretion and neutrophil infiltration (27). Therefore, these studies indicate that sncRNAs can act as mediators in host-pathogen interactions.

We previously identified 2,134 relatively enriched sncRNA sequences in H. pylori OMVs for function prediction analysis, of which 5 showing the greatest OMV enrichment were isolated as candidate regulators of LPS-mediated cytokine secretion. We found that two sncRNAs, sR-2509025 and sR-989262, could effectively reduce LPS-mediated IL-8 secretion, and OMVs deficient in these two sncRNAs enhanced IL-8 secretion in cells (16). These findings showed that H. pylori could reduce the host cell immune response via sncRNAs delivered by OMVs. However, the immune regulatory function of ΔsncRNA OMVs was only explored at the cellular level, and further studies were required to evaluate them at the animal level. Therefore, in this study, the immunogenicity of ΔsncRNA OMVs was assessed at the animal level, confirming that sncRNA disruption inhibits the host immune response and effectively improves OMV immune protection, providing new avenues for efficient OMV-based vaccine design.

The safety of OMVs as vaccines is essential. Our data showed that sncRNA deletion does not increase OMV cytotoxicity, and they are relatively safe for treating macrophages (Fig. 1B). In addition, daily physiological activities and weight growth of mice immunized with various OMV types did not differ from mice immunized with the phosphate-buffered saline (PBS) control, indicating that OMV immunization is safe (Fig. 2A). We also quantified the expression levels of cytokines IL-6 and TNF-α, which are indicators of mucosal inflammation (20). While TNF-α expression in gastric mucosa of mice immunized with OMVs was higher than in controls, its overall level was low (see Fig. S1B in the supplemental material). Therefore, our findings suggest that OMVs do not cause gastric mucosal inflammation, and OMV immunization in mice is safe.

We also measured the levels of specific IgG and secreted IgA from the serum of mice immunized with OMVs, and we found that ΔsR-2509025 and ΔsR-989262 OMVs stimulated mice to produce higher serum IgG, vaginal sIgA, and stomach sIgA levels than wt OMVs (Fig. 2). These antibody levels showed that sncRNA disruption in OMVs induces greater mucosal immunity and humoral immunity, which is important for vaccine development (28). sIgA is considered important for preventing H. pylori infection, because it can inhibit H. pylori binding to the surface of gastric epithelial cells and reduce its colonization (29). Guo et al. compared various immunization routes such as intranasal and intragastric for preventive or therapeutic H. pylori recombinant or multivalent epitope vaccines to reduce H. pylori colonization in mice, and they found that this protection is related to antigen-specific IgG, IgA, and mucosal sIgA antibody responses (30–32). Based on these findings, in this study we used the intragastric route to immunize mice, and our results also showed that intragastric immunization with OMVs can effectively elicit immune protection in mice. This finding indicates that intragastric immunization is the most appropriate and effective delivery strategy for future H. pylori OMV-based vaccines.

In humans, Th cell-related immune responses against H. pylori are strongly biased toward the Th1 type. IFN-γ, TNF-α, and IL-12 are the primary cytokines involved in inducing H. pylori-specific Th1 cells, recruiting macrophages to the infected site (33). Type 2 Th cells produce IL-4, IL-5, IL-10, and IL-13, responsible for strong antibody production, eosinophil activation, and inhibition of several macrophage functions, and thus providing phagocyte-independent protective responses (34). Th17 cells are the third subgroup of effector Th cells, and they can produce IL-17, IL-17F, and IL-22, which play important roles in host defense against extracellular pathogens and autoimmune disease pathogenesis (35).

Some studies have described the immune response produced by H. pylori-related vaccines. An oral multivalent epitope vaccine based on H. pylori urease subunit beta (UreB), adhesin A (HpaA), chloramphenicol acetyltransferase (CAT), and B subunit heat-labile toxin (LTB) can activate the Th1, Th2, and Th17 mixed immune response in mice (36). Similarly, immunization with a fusion protein of H. pylori flagellar hook-associated protein (FliD), UreB, vacuolating cytotoxin A (VacA), and CagL (rFUVL) in mice elicited a mixed Th1, Th2, and Th17 response and antigen-specific IgG2a and IgG1 production, cellular immune response, and gastric IgA production (37). Our data also showed that mice immunized with ΔsR-2509025 and sR-989262 OMVs produce mixed Th1, Th2, and Th17 immune responses (Fig. 3 and 4). In addition, IgG2c and IgG1 levels showed that ΔsR-2509025 and ΔsR-989262 OMVs induced a Th2-type immune response. Moreover, mouse spleen cytokine levels showed that ΔsR-2509025 and ΔsR-989262 OMVs are more capable of inducing Th1 and Th17 immune responses than are wt OMVs. While previous studies have reported that Th2- and Th17-biased immunity induced by H. pylori OMVs is effective in eliciting strong systemic and local immune responses, there is no evidence that the Th1 immune response is unimportant for protecting against H. pylori infection (10, 38, 39). Therefore, the available data suggest that these mixed Th1, Th2, and Th17 immune response types play an important role in OMV-mediated prevention of H. pylori infection. Further studies using gene knockout mouse infection models corresponding to the type of immune response are required to determine their relative importance.

In addition, there are other areas for improvement in this study. First, we only used female C57BL/6 mice to show that H. pylori OMVs missing sncRNAs can induce a stronger immune response and promote H. pylori elimination in the stomach. In the future, the protective efficacy must be evaluated in various animal models, such as Mongolian gerbils, which share gastric pathological features with humans infected with H. pylori, and larger animal hosts in nonhuman primate models, such as piglets and cats, to facilitate the development of H. pylori OMV vaccines (40, 41). Second, we found that the ΔsR-2509025 and ΔsR-989262 OMVs improved protection against H. pylori infection in mice. The next step will be to create a double knockout for these sncRNAs and explore whether its protective efficacy is greater than the individual deletions. Third, cytokine measurements were performed by stimulating the total cells of splenic lymphocytes. While our current research on cytokines has found that ΔsncRNA OMVs tend to cause Th1, Th2, and Th17 immune responses, it is necessary to clarify which immune cells secrete these cytokines to clarify the immune mechanisms. Kim et al. reported that the immune response induced by specific Th cells could provide a new direction for vaccine development against bacterial infections and bacteria-induced inflammatory diseases (42). Therefore, future studies could use flow cytometry to isolate these cells to better understand the immune mechanism mediated by H. pylori ΔsncRNA OMVs.

Conclusion.

H. pylori OMVs missing previously identified sncRNAs enhance the Th1-, Th2-, and Th17-type immune response, resulting in better humoral immunity. In addition, ΔsR-2509025 and ΔsR-989262 OMVs are more effective at clearing the colonization of H. pylori in the stomach than wt H. pylori OMVs. Therefore, our results identify novel strategies for developing H. pylori vaccines and lay the foundation for advancing our understanding of the role of OMVs in regulating the host immune response to H. pylori.

MATERIALS AND METHODS

Bacterial culture and OMV preparation.

The gerbil-adapted H. pylori 7.13 strain derived from clinical strain B128 was provided by Yong Xie at the First Affiliated Hospital of Nanchang University in Nanchang, China. H. pylori Sydney strain 1 (SS1) suspensions used for challenging were prepared from fresh exponential-phase cultures to ensure a high level of viable cells. The mutant strains used in this study (ΔsR-989262, ΔsR-2509025, and ΔsR-7497631) were constructed based on the H. pylori 7.13 strain as previously reported (16). All strains were cultured in Campylobacter agar base supplemented with 10% sheep blood (Difco, Detroit, MI, USA) at 37°C with 5% O₂, 10% CO₂, and 85% N₂.

OMVs from the H. pylori 7.13 strain were isolated via ultracentrifugation as previously described (8). Briefly, 500 mL of H. pylori 7.13 strain was cultured to logarithmic phase (optical density at 600 nm [OD₆₀₀] of 1; 48 to 72 h) and centrifuged at 4,500 × g and 4°C for 1 h to remove bacteria particles. Then, the supernatant was filtered twice with the cap of a 0.45-μm sterilized filter (Millipore; Billerica, MA, USA). The filtrate containing H. pylori OMVs was ultracentrifuged at 20,000 × g and 4°C for 2 h, and the OMV particles were washed with PBS (Mediatech, Manassas, VA, USA). The pelleted OMV particles were resuspended in OptiPrep density gradient medium (Sigma-Aldrich, St. Louis, MO, USA) in PBS and ultracentrifuged for 24 h at 100,000 × g and 4°C in a density gradient of 40%, 35%, 30%, and 20%. The OMV fractions were pooled, gently washed three times with PBS, dissolved in 1 mL PBS, and stored at −20°C. For antigen coating in the ELISA and antigen stimulation in cytokine detection, outer membrane proteins (OMPs) were isolated from the H. pylori 7.13 strain as previously described (43). The protein concentrations of obtained OMVs and OMPs were quantified using a bicinchoninic acid (BCA) assay kit (Thermo Fisher, Rockford, IL, USA) according to the manufacturer’s instructions.

Evaluation of OMV cytotoxicity at the cellular level.

The cytotoxicity of OMVs on macrophages was evaluated using RAW 264.7 cells in 24-well plates (5 × 10⁵ cells/well) inoculated with different OMV concentrations (3.125 to 100 μg/mL) as previously described (7). After 24 h, the supernatant was collected from each well and cytotoxicity was quantified by using a Multitox-Fluor multiplex cytotoxicity assay (Promega, Madison, WI, USA) according to the manufacturer’s instructions. Three independent biological replicates were performed for all experiments, and the data were collected for later analysis.

Ethics statement.

All animal experiments were performed in accordance with the guidelines of the Animal Welfare Law and the relevant regulations of Nanchang University (Nanchang, China; approval number NCDXYD-2019028). All animal procedures were approved by the Animal Welfare Committee of Nanchang University and followed the principles described in the guidelines for the care and use of laboratory animals. We made every effort to minimize animal pain throughout our experiments.

Animal experiments.

This study used female C57BL/6 mice (6 weeks old, 16 to 22 g body weight) obtained from the Laboratory Animal Science Center of Nanchang University (Nanchang, China). The experimental process is depicted in Fig. 1A. The mice were adapted to their new environment for a week before the experiment and then divided into several groups, each containing 18 mice, including a negative control group treated with PBS (100 μL). The mice were immunized intragastrically with 100 μg of OMVs suspended in 100 μL of PBS as the first immunization. All immunized groups used OMV as a single antigen without additional adjuvants. Orbital sinus puncture blood collection and vaginal secretion collection were performed in the fourth and eighth weeks after the first immunization. The soluble components in the serum and vaginal secretions were isolated by centrifugation. In the fourth week (30 days), the same antigen was administered intragastrically to boost the immunization. Eight weeks after the first immunization, 9 of the 18 mice in each group were sacrificed to extract primary spleen cells. The remaining 9 mice in each group were orally challenged with 10⁹ CFU of H. pylori SS1 strain in 20 μL PBS containing 0.01% gelatin (BSG buffer) and monitored until the 12th week, when they were sacrificed to collect splenocytes for cytokine detection. The gastric tissue of mice was used to quantify urease test and bacterial load. Two independent biological replicates were performed for all experiments, and data were collected for later analysis.

Assessment of OMV safety.

Body weight changes of immunized mice were monitored daily for 30 days after both the first and booster vaccinations with wt and ΔsncRNA OMVs and compared to the negative control (PBS only) group to assess their safety. General observations for safety were animal survival and general condition, behavior, and body weight change dynamics.

ELISA.

Antibody levels in mouse blood samples and vaginal secretions were quantified using quantitative ELISA as described previously (9). We used 1 μg OMP suspended in 100 μL of sodium carbonate buffer (pH 9.6) to antigen coat each well of a 96-well plate (Nalge Nunc Inc., Naperville, IL, USA) that was incubated overnight at 4°C. Purified mouse Ig isotype standards (IgG, IgG1, IgG2c, and IgA; BD Biosciences, Billerica, MA, USA) were prepared in triplicate and diluted twice (0.5 μg/μL). After three washes with PBS containing 0.1% Tween 20 (PBST), the plate was blocked with 2% bovine serum albumin (BSA) solution for 2 h at room temperature. Subsequently, 100 μL of each sample was added to the respective wells at different dilutions in triplicate, and the plate was incubated at room temperature for 1 h. After three washes with PBST, biotinylated goat anti-mouse antibodies IgG, IgG1, IgG2c, and IgA (Southern Biotechnology Inc., Birmingham, AL, USA) were added to each well. Then, streptococcal protease alkaline phosphatase conjugate (Southern Biotechnology) was added, and the substrate p-nitrophenyl phosphatase (Sigma-Aldrich) in diethanolamine buffer (pH 9.8) was used to develop the wells. Absorbance was measured at 405 nm on an automatic ELISA plate reader (model EL311SX; Biotek, Winooski, VT, USA). The final Ig isotype concentration of antibody samples was calculated separately for each antibody isotype using a standard curve.

Detection of cytokines in mouse spleen cells.

Mouse spleen cells were collected 4 weeks after the booster immunization and stimulated with 6 μg/mL OMP isolated from H. pylori 7.13 strain for 24 h as previously described (9). Then, the supernatant of the stimulated spleen cells was collected, and cytokine levels were quantified by ELISA. A 96-well plate was coated with monoclonal antibodies against IFN-γ, IL-4, IL-13, IL-17, IL-8, IL-12 (p40), IL-6, and TNF-α (BD Biosciences, Mountain View, CA, USA). Next, samples were blocked with PBS containing 1% BSA, added to wells in triplicate, and incubated overnight at 4°C. Then, the wells were washed and incubated with biotinylated monoclonal anti-IFN-γ, anti-IL-4, anti-IL-13, anti-IL-17, anti-IL-8, anti-IL-12 (p40), anti-IL-6, and anti-TNF-α antibodies (BD Biosciences, Billerica, MA, United States). Finally, horseradish peroxidase-labeled anti-biotin antibody (Vector Laboratories, Burlingame, CA, USA) was added to each well along with 3,3′,5,5′-tetramethylbenzidine (TMB; Moss Inc., Pasadena, CA, USA) to enhance the reaction, which was terminated with 0.5 M hydrochloric acid (HCl). A standard curve was created based on mouse recombinant (r) IFN-γ, IL-4, IL-13, IL-17, IL-8, IL-12 (p40), IL-6, and TNF-α to determine cytokine expression levels in spleen cells.

Determination of bacterial loading.

At 4 weeks after the oral H. pylori SS1 challenge, gastric tissues were collected to quantify bacterial load. A lower pathogen load indicates a stronger protective effect. First, the separated tissue was washed with precooled PBS and then transferred to preweighed test tubes containing 5 mL of brain heart infusion medium (Difco). Next, the tissue was reweighed to an accuracy of 0.0001 g, homogenized with a sterile homogenizer, and continuously inoculated onto a Campylobacter agar matrix (Difco) plate containing 10% sheep blood at dilutions of 1:10, 1:100, and 1:1,000. Then, the plates were incubated at 37°C under microaerobic conditions for 6 to 7 days. H. pylori colonies were identified based on the urease and oxidase reactions and wet morphology analysis.

Urease test.

Stomach specimens from each mouse were immersed in 0.5 mL of 0.8% sodium chloride (NaCl) solution to prepare a tissue homogenate for urease activity quantification (44). Briefly, 3 mL of urea broth (1 mg/mL glucose, 1 mg/mL peptone, 2 mg/mL monopotassium phosphate [KH₂PO₄], 5 mg/mL NaCl, and 1% urea) containing phenol red indicator was mixed with 100 μL of the tissue mixture and homogenized. Tissue homogenate containing PBS was used as the negative control. After incubating at 37°C for 4 h, the urease activity of each gastric tissue was quantified based on the OD₅₅₀ by using a UV/visible spectrophotometer.

Statistical analysis.

All ELISA experiments were performed in triplicate. The significance of differences in the average values between the experimental and control group was assessed using one- or two-way analysis of variance (ANOVA), followed by Tukey’s post hoc test. All data are expressed as means ± SD. All statistical analyses were performed using GraphPad Prism software v.6 (GraphPad Software Inc., San Diego, CA, USA).