M cells targeted H. pylori antigen SAM-FAdE displayed on bacterium-like particles induce protective immunity

Abstract

Background

Helicobacter pylori (H. pylori), a specific bacterium capable of surviving in the acidic environment of the stomach, has been recognized as a group of causative agents of gastric cancer. Therefore, the development of mucosal vaccines against H. pylori is expected to provide an important direction for the treatment of chronic gastritis and the prevention of gastric cancer.

Methods and results

In this study, we used bacteria-like particles (BLPs) obtained by treating Lactic acid bacteria (L. lactis) with hot acid, and successfully displayed the M cell-targeted H. pylori multi-epitope purified antigen SAM-FAdE, with 90% display efficiency. In addition, BLPs-SAM-FAdE can effectively target M ​​cell models and M cells of mouse Peyer’s patches (PPs) through oral immunization, promote the transport of particulate vaccines to dendritic cells (BMDCs) and stimulate their maturation, significantly increased proportion of plasma cells and germinal centers B cells. This indicates that the vaccination can induce notable antigen-specific mucosal immune responses (production of sIgA), CD4⁺ T cell responses (Th1/Th2/Th17) and humoral immune responses (production of serum IgG). Furthermore, oral BLPs-SAM-FAdE dramatically reduced the H. pylori adhesion and specific 16S rRNA expression of H. pylori in gastric mucosal tissue, protecting gastric tissue from damage.

Conclusion

BLPs-SAM-FAdE can significantly reduce the adhesion of H. pylori in gastric mucosal tissue and inhibit gastritis and gastric damage caused by H. pylori infection.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12951-025-03111-9.

Introduction

Helicobacter pylori (H. pylori) is bacteria commonly seen in humans. It can cause a number of upper gastrointestinal disorders, such as indigestion, upper abdominal pain or discomfort, and other symptoms [1]. H. pylori infection is a global health issue, particularly in areas with high infection rates, such as China, and poses a major threat to public health [2]. The discovery of H. pylori was a watershed moment in medicine in the 20th century, transforming people’s perceptions and treatments of upper gastrointestinal illnesses [3]. First-line treatment for H. pylori infection usually consists of a combination of proton pump inhibitors (PPIs) and antibiotics, with a treatment period of 14 days [4]. Although growing socioeconomic levels and improved therapies have slowed the spread of infection, H. pylori remains one of the most frequent human bacterial infections, affecting about half of the global population [5]. As a result, diagnosing, treating, and preventing H. pylori infection remains a critical public health concern.

In the human immune system, the mucosal barrier is the first line of protection against infections. The mucosal immune system is an essential component of the body’s immune system, guarding it from microbial invasion using a variety of sophisticated methods [6]. Microfold cells (M cells) serve as “sentinels” on the mucosal surface, successfully collecting antigens via their distinct morphological structure and transport processes [7]. This procedure is critical for generating adaptive immune responses because M cells may directly transport trapped antigens to immune cells in the submucosa (dendritic cells and macrophages), triggering a particular immune response [8, 9]. Designing vaccine carriers that target M cells can greatly improve vaccine immunogenicity and effectiveness, propelling M cells to the forefront of vaccine research [10]. The National Institutes of Health (NIH) in the United States is developing an oral rotavirus vaccine that targets M cells [11]. Using genetic engineering techniques, rotavirus antigens are displayed on bacterial-like particles (BLPs) that can be specifically taken up by M cells, thereby activating mucosal immune responses. The International AIDS Vaccine Initiative (IAVI) is developing an HIV vaccine based on targeting M cells. By utilizing specific adjuvants and delivery systems, such as virus-like particles (VLPs), HIV antigens are delivered to M cells to enhance mucosal and systemic immune responses [12]. The World Health Organization (WHO) is researching the use of M cell characteristics to develop a nasal spray influenza vaccine. By employing M cell-targeted delivery systems, such as nanoparticles, influenza antigens are delivered to M cells to boost local mucosal immunity and systemic immune responses [13]. Researchers at the University of Phayao are utilizing M cell characteristics to develop a vaccine for porcine reproductive and respiratory syndrome virus (PRRSV) [14]. They use a specific delivery system (polylactic acid nanoparticles) for nasal administration to enhance the antigen uptake efficiency of inactivated PRRSV loaded with LTB and DDA, activating immune responses to combat PRRS virus invasion.

With the advancement of vaccination technology, the hunt for safe, effective, and widely acceptable vaccine carriers has been a primary research emphasis. In recent years, there has been substantial development in employing lactic acid bacteria (L. lactis) as oral vaccination carriers [15]. Similarly, the use of bacterial-like particles (BLPs) generated from L. lactis has piqued researchers’ interest. BLPs are generated using a heat-acid process that forms a peptidoglycan (PGN) shell, onto which particular pathogen antigens are bonded, resulting in particle vaccinations of 1–2 μm size [16]. These particle vaccinations can resemble pathogen structures, stimulating the host’s immune response while avoiding the safety hazards associated with live germs [17]. By expressing influenza virus surface antigens (such as hemagglutinin and neuraminidase) on BLPs, it is feasible to imitate the shape of viral particles, eliciting a significant immune response [18]. A zoonotic virus known as Rift Valley fever virus (RVFV) is spread by mosquitoes and can seriously infect both humans and ruminants. By acting as carriers of the RVFV Gn head protein, it stimulates the immune system to prevent serious outbreaks brought on by the Rift Valley fever virus [19]. Furthermore, the vaccine stimulates mice to produce high levels of peptide-specific antibodies and FMDV-specific neutralizing antibodies when a soluble foot-and-mouth disease virus (FMDV) target fusion protein is displayed on BLPs [20]. It also increases spleen lymphocyte proliferation and Th1-type cytokine secretion [21].

This work developed a M cell-targeted H. pylori vaccine using the mucosal delivery system of BLPs. By optimizing conditions, we constructed a high-quality BLPs delivery system, presented the M cell targeting H. pylori multi-epitope antigen (FAdE), and assessed the targeting and immune activation effects. In addition, under the action of M cell targeting peptide (Mtp), the mechanism of pathogenic microorganisms activating the body’s mucosal immune system was simulated to explore the immunogenicity and protective effect of the vaccine in H. pylori mouse models (Scheme 1). This work offers a fresh approach to overcoming the drawbacks of conventional antibiotic therapies and slowing the emergence of drug resistance. Mucosal immune activation strengthens the body’s defense against H. pylori, lowering the risk of infection and illness. Furthermore, developing novel technology and approaches for developing vaccines against other mucosal pathogens has significant scientific and public health implications.

Scheme 1.

Schematic illustration of immunity by oral delivery of BLPs into PPs. (A) Preparation of BLPs-SAM-FAdE; (B) BLPs-SAM-FAdE enhance both the resistance of vaccinations against gastric insults and the delivery of BLPs-SAM-FAdE into PPs through M cells, which promote robust mucosal immune responses that can positively enhanced against H. pylori

Materials and methods

Strains and mouse

Escherichia coli DE3 and L. lactis NZ9000 are maintained in this laboratory. E. coli DE3 was routinely cultured in Luria-Bertani (LB) medium at 37 °C, while NZ9000 was amplified and cultured in M17 medium supplemented with 0.5% glucose at 30 °C. H. pylori SS1 was cultured on 90 mm Columbia blood agar plate containing H. pylori additive (HB8646a, Qingdao Haibo) under microaerobic conditions for 2–3 days. The BALB/c male mice used in the experiments, aged 5–6 weeks, were purchased from Beijing Huafukang Biotechnology, and housed in a SPF experimental animal center. They were provided with a standard diet and drinking water. After one week of acclimatization, the relevant experiments were conducted.

Preparation of BLPs-SAM-FAdE

To obtain BLPs, we first washed the precipitated L. lactis obtained from amplification culture with sterile water, then re-suspended them in 0.1 M HCl and heated at 100 °C for 30 min to form the precipitate. Afterward, we washed the precipitate three times with sterile PBS and re-suspended it, adjusting the BLPs concentration to 1 U/mL for later use. To obtain the M cell-targeting recombinant protein SAM-FAdE, we first amplified the E. coli DE3 containing the pCZN1-SAM-FAdE plasmid, and when the OD600 reached 0.6–0.8, we induced expression for 4 h using 0.2 mM IPTG, followed by purification of the recombinant protein SAM-FAdE using a Ni column. To prepare the particle vaccine BLPs-SAM-FAdE, we mixed 1 U of BLPs with 80 µg of purified antigen SAM-FAdE at room temperature for 2 h, and then washed to remove unbound antigen.

SDS-PAGE and Western blot

Collect the bacterial cells after amplification and induction, the supernatant after ultrasonic disruption, and the bacterial cells for SAS-PAGE analysis of protein expression. Then, use the AKTA protein purification system (GE Healthcare, USA) to obtain SAM-FAdE. The concentration of the recombinant proteins SAM-FAdE is determined using a BCA protein assay kit and protein densitometry analysis. In brief, the obtained samples are mixed with loading buffer and denatured at 98 °C for 10 min. The proteins are then subjected to vertical electrophoresis at a constant current of 150 V for 1 h (Bio-Rad, USA). After staining with Bradford, the gel is decolorized until clear bands are visible. For Western blotting, the steps include membrane transfer, blocking, incubation with anti-His (Proteintech, China), and incubation with the secondary antibody. Finally, protein bands are analyzed using an exposure system.

Characteristic analysis of BLPs

To obtain qualified BLPs for the preparation of particulate vaccines, the morphology changes of the NZ9000, BLPs, and BLPs-SAM-FAdE were observed using a transmission electron microscope (HT7800, Japan). Meanwhile, the samples were diluted and analyzed for changes in particle size using a Malvern laser particle size analyzer (Mastersizer 2000, UK).

Binding of BLPs and purified antigens

Take 100 µL of BLPs and BLPs-SAM-FAdE respectively, incubate with FITC-labeled anti-His antibody for 2 h, then wash to collect the precipitate, re-suspend in 1 mL PBS and through the flow cytometer (C6, USA) to analyze its fluorescent expression. In addition, we also analyzed the binding of BLPs to purified antigens by indirect immunofluorescence. BLPs and BLPs-SAM-FAdE were evenly spread on lysine-treated slides, air-dried, and blocked with 3% BSA-PBS for 30 min. After washing twice with PBS, the slides were incubated with anti-FAdE mouse polyclonal antibody (1:100) prepared in our laboratory for 60 min at room temperature. After washing three times with PBS, the slides were incubated with FITC-labeled goat anti-mouse IgG (1:200, Proteintech, China) for 60 min in the dark. The slides were washed three times with PBS, and fluorescence was observed under a laser confocal microscope (ZEISS, Germany).

Construction and verification of M cell model

According to the method of Kerneis et al. [22], the M cell model was constructed using co-culture of Caco-2 cells and Raji B cells. We added 5 × 10⁵ Caco-2 cells to a 3 μm Transwell chamber, changed the medium every other day, and collected the culture medium in the upper chamber for later use until they were fully differentiated (14 days). Next, Raji B cells (10⁵) were added to the basolateral side of the Transwell chamber and cultured for 4 days. At the same time, no Raji B cells were added as a control. Order to verify whether the M cell model was successfully constructed [23], the membrane of the Transwell chamber was first observed using a scanning electron microscope (S-3400 N, Japan). At the same time, the changes in ALP activity on days 4, 8, 12, 14, 16, 18, and 20 were analyzed using an alkaline phosphatase (ALP) detection kit (Beyotime, China).

After successfully constructing the model, we validated the M cell targeting of BLPs-SAM-FAdE in vitro by adding green fluorescently labeled BLPs-SAM-FAdE and BLPs to both mono-culture and co-culture systems and incubating them in the dark for 6 h. After washing, 1 mL of 4% paraformaldehyde was added to fixation, and targeting was observed using a fluorescence microscope (OLYMPUS, Japan). Additionally, we validated the targeting of the particulate vaccine in vivo using an ileal loop experiment. After thoroughly cleaning a segment of the mouse ileum, one end was tightly tied off, and BLPs-SAM-FAdE and FAdE were separately infused from the other end before being sealed. After a 6-hour reaction, the tissue was frozen and sectioned. BLPs-SAM-FAdE and FAdE were identified using Alexa Fluor^®647-labeled anti-His antibodies, while FITC-labeled GP2 antibodies were used to identify PPs M cells. Finally, the fluorescent signals were observed using a confocal microscope (ZEISS, Germany).

Finally, we analyzed the biological distribution of the BLPs-SAM-FAdE vaccine using in vivo imaging system (IVIS). Six-week-old BALB/c mice were fasted with water and food for 24 h and then gavaged with specific volumes of AlexFluor^®488-FAdE and AlexFluor^®488-BLPs-SAM-FAdE, respectively. After 24 h of free movement, the mice were euthanized, and the stomach and intestinal tissues were dissected separately to observe the fluorescence distribution using IVIS (IVIS Lumina III, USA).

Immunization and sample collection

As shown in Fig. 1A, to evaluate the effect of BLPs-SAM-FAdE in clearing H. pylori, 24 male BALB/c mice aged 6–8 weeks were randomly divided into 4 groups (6 mice per group). One group was given normal drinking water as a control (Control group), while the other three groups were gavage with a H. pylori suspension (1 × 10⁹ CFUs/mL) at 300 µL per mouse every other day, for a total of 4 gavage sessions to establish an H. pylori mouse infection model. To verify the successful construction of the model, we randomly took the stomach tissues of 2 mice at the 4th week after the final infection, and used the rapid urease detection assay, H. pylori colony formation assay, and RT-qPCR to verify whether the mice were successfully infected with H. pylori. One month after the gavage, the three groups of mice were respectively gavage with PBS (H. pylori group), a combination antibiotic treatment (metronidazole + amoxicillin + omeprazole, Antibiotics group), and BLPs-SAM-FAdE (1 U BLPs with 80 µg antigen, BLPs-SAM-FAdE group) once a week, and the treatment was continued for 4 weeks. Ten days after the last immunization, the gastric tissues of the mice were collected to evaluate the effect of H. pylori clearance.

Fig. 1.

(A) Schematic diagram of time points of BLPs-SAM-FAdE treatment of H. pylori (Hp) infection model; (B) RT-qPCR detects expression of H. pylori 16S rRNA in gastric tissue; (C) H. pylori quantitative culture statistical chart; (D) Stomach statistical chart of gastritis score of HE staining; (E) HE staining of gastric tissue (×200)

Furthermore, to investigate the specific mechanism by which BLPs-SAM-FAdE induces an immune response, healthy male BALB/c mice aged 6–8 weeks were randomly divided into 4 groups and gavage with PBS, BLPs (1 U), the monovalent antigen SAM-FAdE (80 µg), and BLPs-SAM-FAdE (1 U BLPs with 80 µg antigen) once a week for 4 consecutive weeks. In the sixth week, gastrointestinal lavage fluid and feces were collected to evaluate the production of antigen-specific sIgA. Blood was collected via the orbital vein to separate serum for detecting antigen-specific IgG and subtype. Additionally, spleens and lymph nodes were collected to isolate lymphocytes for analyzing the activation of germinal centers, plasma cells, and T cell responses induced by the BLPs-SAM-FAdE.

Ability of BLPs-SAM-FAdE to activate BMDCs in vitro

BMDCs were obtained from the femurs and tibias of 8-week-old BALB/c mice and cultured in RPMI 1640 medium containing 5% FBS, 1% antibiotics, 20 ng/mL GM-CSF, and 20 ng/mL IL-4 for 7 days, with medium changes on days 3 and 6. To evaluate the induction capability of BLPs-SAM-FAdE on BMDCs, the BMDCs were seeded at a density of 1 × 10⁵ cells per well and co-cultured with BLPs, SAM-FAdE single antigen, and BLPs-SAM-FAdE for 24 h. BMDCs treated with LPS served as a positive control, and untreated BMDCs served as a negative control. Finally, the cells were collected and incubated with PE anti-mouse CD11c, FITC anti-mouse MHC II, APC anti-mouse CD40, Super Bright™ 436 anti-mouse CD80, and Super Bright™ 600 anti-mouse CD86 (Invitrogen) monoclonal antibodies for 40 min. The expression of surface markers or co-stimulatory molecules was detected using flow cytometry. Additionally, the secretion of IL-1β, IL-12p70, IL-4, and IL-6 in the supernatant was measured using an ELISA kit (Fine Test^®, China) according to the manufacturer’s instructions.

FAdE-specific IgG, IgG₁, IgG_2a and sIgA ELISA

In summary, 100 µL of SAM-FAdE antigen coating solution was added to each well of the ELISA plate at a final concentration of 2 µg/mL and incubated overnight at 4 °C. The next day, the plate was washed twice and blocked with 5% BSA-PBS for 2 h, followed by four washes before adding samples. Gradients of diluted serum or samples such as gastrointestinal lavage fluid and feces were added to the ELISA plate and incubated for 1 h, followed by three washes and subsequent incubation with HRP-conjugated sheep anti-mouse IgG, IgG₁, IgG_2a, and sIgA (Abcam, USA) for 1 h. After incubation, the ELISA plate was washed 3–4 times and developed with 100 µL TMB for 30 min. Finally, 50 µL of 2 M H₂SO₄ was added to stop the reaction, and the absorbance was measured at 450 nm.

FAdE-specific IFN-γ/IL-4/IL-17 T-cell

To assess the induction of FAdE-specific CD4⁺ T cell responses following oral immunization, splenocytes were extracted from mice and stimulated with SAM-FAdE. The cells were then collected for intracellular cytokine staining, using the following antibodies: BV421-anti mouse CD3, FITC-anti mouse CD4, PE-anti mouse IFN-γ, Percp cy5.5-anti mouse IL-4, and APC-anti mouse IL-17A (Biolegend). Additionally, cytokine secretion levels of IFN-γ, IL-4, and IL-17A in the supernatant were measured using an ELISA kit (Fine Test^®, China). Furthermore, we evaluated the activation levels of Th1, Th2, and Th17 corresponding to the secretion of IFN-γ, IL-4, and IL-17A using the ELISPOT assay according to the manufacturer’s instructions (Mabtech).

GC B cell and plasma cell staining

The mesenteric lymph nodes (MLN) of mice were immersed in sterile PBS, mechanically ground, and filtered through a 200-mesh filter to remove debris. The activation ratios of germinal center B cells and plasma cells were evaluated using the following antibodies: FITC-anti mouse CD45R/B220 (Invitrogen), PE-anti mouse CD95 (Fas), APC-anti mouse GL-7, and BV421-anti mouse CD138 (Biolegend). Samples were analyzed using a FACS Celesta flow cytometer (BD, USA) and data were processed with FlowJo 10.8.1 software. The activation status of germinal centers in the spleen was assessed through tissue immunofluorescence analysis. Stomach tissues from mice orally administered PBS, BLPs, SAM-FAdE, and BLPs-SAM-FAdE were embedded, sectioned into 10 μm thick slices, fixed, and blocked, then stained with the following antibodies: BV510-anti mouse CD4 (BD Biosciences), FITC-anti mouse B220 (Invitrogen), and Rhodamine-PNA (Vector). After mounting, observations were made using a confocal microscope (ZEISS, Germany).

Effects of BLPs-SAM-FAdE treatment on H. pylori infection

10 days after the final immunization, mouse gastric tissues were isolated and divided into three parts: one part was used for quantitative culture of H. pylori [24], another for RT-qPCR detection of H. pylori-specific 16S rRNA expression, and the last part for histopathological analysis. The primer sequences used were as follows: H. pylori 16S rRNA (Forward: CTCATTGCGAAGGCGACCT, Reverse: TCTAATCCTGTTTGCTCCCCA) and internal control 18S rRNA (Forward: GCAATTATTCCCCATGAACG, Reverse: GGCCTCACTAAACCATCCAA). According to quantitative culture of H. pylori, the gastric tissue was homogenized in 0.5 mL of PBS. After diluting it in the following ratios: 1:10, 1:100, and 1:1000, 100 µL was plated onto 90 mm Columbia blood agar plate containing H. pylori additive under microaerobic conditions for 2–3 days prior to a colony count being conducted. The colony-forming units per gram of stomach tissue (CFU/g): H. pylori colonization density was equal to bacteria colony count × dilution / gastric weight. For HE staining, gastric tissues were fixed in 10% formaldehyde, dehydrated through a graded ethanol series, embedded in paraffin, sectioned, and stained with hematoxylin-eosin for microscopic observation. The pathological scoring of gastric mucosal damage was based on the following scoring criteria Table 1 [25].

Table 1.

Histological scoring criteria

Score	Histological criteria
0	The epithelial layer structure is intact, with no obvious damage
1	The epithelial layer structure is disorganized, accompanied by slight deformation of epithelial cells.
2	The structure of the superficial lamina propria is disorganized or 1/3 of the glandular structure is missing.
3	The structure of the intermediate lamina propria is disorganized or 2/3 of the glandular structure is missing.
4	The lower lamina propria is eroded or more than 2/3 of the glandular structure is missing.

Statistical analysis

Unless otherwise specified, data are shown as the mean ± SD and each experiment was repeated two or three times. Data were analyzed by the two-tailed unpaired t-test or one-way ANOVA with Tukey’s post hoc analysis using GraphPad Prism.

Results

Bioinformatics software screening for M-cell targeting peptide Mtp

The modeling program trRosetta predicted the structure of Mtp, which is composed of CPE30, Co1, and CKS-9 (Fig S1, Additional file). Following that, the prediction model’s quality was evaluated, and it was discovered that the outcomes of VERIFY 3D and PROVE were subpar, which was in line with the absence of Mtp homologous proteins. The protein prediction model of Mtp was shown to be quite accurate based on the improved detection findings of ERRAT, WHATCHECK, and PROCHECK (Fig S2, Additional file). Lastly, protein-protein docking was carried out using Z-Dock to examine how Mtp bound to the receptor proteins C5aR and Claudin4 (Fig S3 left, Additional file). According to Additional file Fig S3 right, the binding surface residues demonstrated that Mtp has a strong binding affinity for C5aR and Claudin4. After that, we constructed and verified the pCZN1-SAM-FAdE plasmid (Fig S4-5, Additional file). The SAM-FAdE contains four main components (Fig S6, Additional file): the M cell-targeting peptide Mtp, the cA binding domain that associates with the BLPs, the 6*His lable and the multi-epitope antigen FAdE from H. pylori (which contains four adhesion factors: Urease, CagL, HpaA, and Lpp20) [26].

Identification of SAM-FAdE protein on BLPs

To obtain the purified antigen SAM-FAdE, the antigen SAM-FAdE was expressed using a prokaryotic inducible expression system and subsequently purified through a Ni column. The results showed (Fig. 2A-B) that the SAM-FAdE induced by IPTG and ultrasonically disrupted was concentrated in the precipitate which expressed in the form of inclusion bodies. Therefore, the inclusion body was denatured and purified to obtain SAM-FAdE with a purity greater than 90%. Then the BCA protein detection method and grayscale analysis (Fig. 2C) were used to find that the concentration was 0.6 mg/mL. The Western blot showed that the SAM-FAdE could be specifically recognized by the 6*His antibody, and the protein structure was not damaged. The molecular weight is approximately 98 kDa (Fig. 2D). Then, we observed the morphology of BLPs obtained by 0.1 M HCL heating treatment through transmission electron microscopy (Fig. 2E). The results showed that compared with the live L. lactis NZ9000, the cytoplasm of BLPs was uniform and loose, covered by an obvious layer of PGN. In addition, the particle sizes of both NZ9000 and BLPs remained between 1000 and 2000 nm (Fig. 2F). Finally, the antigens bound to the surface of BLPs were identified. The SDS-PAGE results showed that the molecular weight of the antigen bound to the surface of BLPs was consistent with the molecular weight of the recombinant antigen SAM-FAdE (Fig. 2G). At the same time, TEM, flow cytometry, and immunofluorescence all showed that the SAM-FAdE was successfully loaded on the surface of BLPs, and the loading efficiency was as high as over 90% (Fig. 2H-J).

Fig. 2.

(A) Protein induced expression (1, pCZN1 empty vector induction, 2, before induction, 3, after induction, 4, supernatant after sonication, 5, precipitation after sonication); (B) Protein purification (1, Sample after fragmentation, 2, flow-through sample, 3–4, elution sample); (C) Protein concentration analysis (1, 0.5 mg/mL BSA, 2, purified protein SAM-FAdE); (D) Protein identification (6*His antibody); (E) TEM image of NZ9000 and BLPs (×40000); (F) Particle size of NZ9000 and BLPs; (G) SDS-PAGE analysis of the preparation of BLPs-SAM-FAdE; (H) TEM of BLPs-SAM-FAdE (×60000); (I) Flow cytometry detection of the binding efficiency of SAM-FAdE and BLPs; (J) Laser confocal analysis of the binding efficiency of SAM-FAdE and BLPs (×1000)

BLPs-SAM-FAdE targets M cells through the action of targeting peptide Mtp

To confirm the formation of M cell model, the morphological characteristics of M cells were observed through SEM, and the dynamic changes of ALP activity in the upper chamber during the formation of M-like cells were detected. As can be seen from the results (Fig. 3A-B), in the co-cultured cells, as some Caco2 cells transformed into M cells, short and irregular microvilli could be observed on their apical surface, while the mono-cultured cells showed regular microvilli are neatly arranged on the apical surface of Caco2 cells. In addition, ALP activity continued to increase with the prolongation of culture time, reaching a peak after 16 days and remaining unchanged. In contrast, the ALP activity in the upper chamber of the co-culture model formed a significant difference after adding Raji B cells on day 14, and the ALP activity continued to decrease as time went on (Fig. 3B). This is related to the formation of loose microvilli after the formation of M cells. After confirming that the M cell model was successfully constructed, we used this model to verify the M cell targeting of BLPs-SAM-FAdE in vitro (Fig. 3C). Green fluorescently labeled BLPs-SAM-FAdE and BLPs was added to co-culture and mono-culture respectively and incubated in the dark for 6 h. Under a fluorescence microscope, obvious green fluorescence was observed in co-culture, but no fluorescence signal was observed in mono-culture or BLPs. Therefore, under the action of the targeting peptide Mtp, the BLPs-SAM-FAdE targets the M cell model. Finally, the ileal closed-loop experiment was also used to verify the targeting of the BLPs-SAM-FAdE in vivo (Fig. 3D). It was observed that the BLPs-SAM-FAdE containing the targeting peptide overlapped with the M cells labeled by the GP2 antibody, but the multi-epitope antigen FAdE was unable to interact with the M cells. Lastly, a IVIS was used to investigate the biological distribution of BLPs-SAM-FAdE in the gastrointestinal tract after 24 h (Fig. 3E). The results unmistakably demonstrated that the fluorescence intensity in the gastric and intestinal tissues of the BLPs-SAM-FAdE group was higher than that of the SAM-FAdE group, and that the ileum segment was the primary location of fluorescence. This is because BLPs have the capacity to protect against the degradation of recombinant antigens by gastric acid and Mtp can interact with the M cells.

Fig. 3.

(A) SEM image of the M-like cells (×5000); (B) Analysis of ALP activity of the M cell model during co-culture; (C) Fluorescence microscope to verify the in vitro targeting of BLPs-SAM-FAdE (×200); (D) Laser confocal verification of BLPs-SAM-FAdE targeting in vivo (×100); (E) Biological distribution of BLPs-SAM-FAdE vaccine 24 h after gavage was observed by IVIS (Left: SAM-FAdE group, right: BLPs-SAM-FAdE group)

BLPs-SAM-FAdE promotes the maturation and activation of BMDCs in vitro

Given that DCs are key to the initiation of innate and adaptive immune responses, we studied the effect of BLPs-SAM-FAdE on the activation of BMDCs. In short, cells extracted from bone marrow were successfully differentiated into DCs (CD11c⁺) through the combined action of GM-CSF and IL-4, and then treated with LPS, SAM-FAdE antigen and BLPs-SAM-FAdE respectively, and cell surface marker expression was analyzed by flow cytometry. Studies have shown that LPS treatment can increase the expression of DC costimulatory molecules and induce the activation of DCs, so it was used as a positive control [27]. The results are shown in Fig. 4A-B. Compared with the control group, SAM-FAdE single antigen and BLPs-SAM-FAdE can significantly increase the expression of MHC II, CD40, CD80 and CD86. In addition, compared with LPS, BLPs-SAM-FAdE has more advantages in increasing the expression of CD40. Finally, since the cytokines IL-1β, IL-12p70, IL-4, and IL-6 secreted by DCs correspond to the activation of Th1, Th2, and Th17 respectively, these cytokines are essential for inducing specific CD4⁺ T lymphocyte responses. ELISA results showed (Fig. 4C-F) that SAM-FAdE single antigen and BLPs-SAM-FAdE could enhance the secretion of cytokines IL-1β, IL-12p70, IL-4, and IL-6 by DCs.

Fig. 4.

(A) Flow cytometry detection of BMDCs activation markers MHC II, CD40, CD80, CD86; (B) Statistical chart of CD11c⁺MHC II⁺, CD11c⁺CD40⁺, CD11c⁺CD80⁺, CD11c⁺CD86⁺; ELISA detection of BMDCs Cytokines secreted after activation, IL-1β (C); IL-12p70 (D); IL-4 (E); IL-6 (F)

BLPs-SAM-FAdE induce specific IgG, IgG subclasses and sIgA

To analyze the effect of BLPs-SAM-FAdE on B cell responses and humoral immune responses, mice were orally administered BLPs, single antigen SAM-FAdE and BLPs-SAM-FAdE, once every other week, and immunized 4 times. At the 6th week, mouse serum was collected to detect the antigen-specific IgG and its subtypes, gastrointestinal lavage fluid and feces were collected to detect antigen-specific sIgA, and mesenteric lymph nodes (MLNs) were isolated to analyze the levels of activated germinal center B cells and plasma cells (Fig. 5A). Germinal centers are located in secondary lymphoid tissue lymph nodes and secondary follicles of the spleen. They are areas where B cells proliferate, differentiate and increase antibody affinity. Germinal center (GC) B cells have been recognized for giving rise to high-quality and robust antibody responses [28]. To determine the formation of GC B cells, we evaluated the formation of GC B cells in MLNs 2 weeks after finally dosing using flow cytometry (Fig. 5B), while the proportion of GC B cells in the spleen was assessed using tissue immunofluorescence (Fig. 5C). Orally administered BLPs, single antigen SAM-FAdE and BLPs-SAM-FAdE can both induce GC B cell responses. BLPs, as a safe adjuvant, can induce immune responses. Compared with single antigen SAM-FAdE, the effect of BLPs-SAM-FAdE is more obvious with the support of adjuvants. Similarly, BLPs-SAM-FAdE was able to induce a more stable plasma cell production ratio (Fig. 5D). In addition, ELISA was used to detect the antigen-specific serum IgG (Fig. 5E) and its subtypes IgG₁ and IgG_2a (Fig. 5F-G). The results were consistent with expectations. The IgG, IgG₁ and IgG_2a antibody levels of the BLPs-SAM-FAdE group were significantly higher than those in the PBS group and BLPs group. Compared with the PBS group, only the total IgG antibody level in the single-antigen SAM-FAdE group increased. In addition, compared with the single-antigen SAM-FAdE group, the levels of IgG₁ and IgG_2a in the BLPs-SAM-FAdE group were significantly higher. sIgA is crucial for clearing pathogens in mucosal sites. Therefore, the key to oral vaccines is to induce the body to produce high levels of sIgA antibodies to completely eliminate H. pylori colonization sites. We measured the titers of antigen-specific sIgA produced in gastric mucosa, intestinal mucosa and feces respectively (Fig. 5H-J). Although the sIgA produced by the single antigen SAM-FAdE group was slightly higher than that of the PBS group and BLPs group, it did not have statistical significance. The BLPs-SAM-FAdE group, under the protection of the carrier, transported the antigen to the immune effector site and induced significant sIgA. Another important aspect of our concern is whether mice can develop anti-H. pylori antibodies following BLPs-SAM-FAdE immunization. The results of ELISA and Western blot (Fig S7, Additional file) confirmed that the antiserum of BLPs-SAM-FAdE can specifically recognize the virulence factors of H. pylori (UreA, UreB, Lpp20, HpaA, CagL), and can also produce IgG and sIgA that specifically bind to H. pylori lysate (Fig S8, Additional file).

Fig. 5.

(A) Schematic diagram of oral immunization time points; (B) Flow cytometry detection of the proportion of GC B cells (B220⁺GL-7⁺FAS⁺) in mesenteric lymph nodes; (C) Immunofluorescence detection of spleen germinal center activation (×200) ; (D) Flow cytometry detects the proportion of plasma cells (B220⁺CD138⁺) in mesenteric lymph nodes; ELISA detects the antibody titers of antigen-specific IgG (E), IgG₁ (F), and IgG_2a (G) in serum; ELISA detects gastric antibody titers to antigen-specific sIgA in gastric lavage fluid (H), intestinal lavage fluid (I) and feces lavage fluid (J)

BLPs-SAM-FAdE induces antigen-specific Th1/Th2/Th17 immune response

Next, we evaluated the induction of helper T cell responses by BLPs-SAM-FAdE at week 6 (Fig. 6F). Through the flow cytometry results (Fig. 6A-B), it was found that BLPs-SAM-FAdE can produce significant antigen-specific Th1/Th2/Th17 immune responses. The proportions of the three antigen-specific Th cells were 0.77%, 0.69% and 0.53%. The single antigen SAM-FAdE group could only induce Th1 response, and the cell proportion was only 0.26%. The ELISPOT results (Fig. 6D-E) were consistent with the flow cytometry results, but the Th17 response was dominant in the single antigen SAM-FAdE group. In addition, ELISA detected the three cytokines (IFN-γ, IL-4, IL-17A) secreted by splenic lymphocytes (Fig. 6C). The results of the BLPs-SAM-FAdE group were consistent with the flow cytometry and ELISPOT. But the secretion of IL-4 in the single-antigen SAM-FAdE group was more obvious. The difference of the single-antigen SAM-FAdE group may be caused by the absorption effect of the single antigen in the mice.

Fig. 6.

(A) Flow cytometry gate plot of intracellular (IFN-γ, IL-4, IL-17A) staining; (B) CD4⁺IFN-γ⁺, CD4⁺IL-4⁺and CD4⁺IL-17A⁺ statistical chart; (C) ELISA detection of secreted cytokines (IFN-γ, IL-4, IL-17A) levels in spleen lymphocyte supernatant; (D) Representative spleen lymphocyte ELISPOT spot diagram; (E) IFN-γ/IL-4/IL-17A-producing T cells spot statistical diagram; (F) Schematic diagram of oral immunization and sample collection time points

The efficiency of BLPs-SAM-FAdE in clearing H. pylori from gastric tissue of infected mice

The H. pylori mouse infection model was constructed according to Fig. 1A. Firstly, we verified the model was successfully constructed 4 weeks after the last infection (Fig S9-11, Additional file) and then treated with BLPs-SAM-FAdE, while antibiotic treatment served as a control. The results show (Fig. 1B, C), whether it is H. pylori colony formation experiment or qPCR detection of H. pylori-specific 16S rRNA in gastric tissue, compared with the H. pylori infection group, the group given BLPs-SAM-FAdE and the triple combination of antibiotics can significantly reduce H. pylori in the gastric tissue of mice. At the same time, the qPCR results showed that the therapeutic effect of oral BLPs-SAM-FAdE was better than that of the oral antibiotic group (p<0.001). In addition, the mucosal damage of mouse gastric tissue was evaluated (Fig. 1D, E). The results showed that the gastric tissue of H. pylori-infected mice showed obvious damage, and the glandular structure basically disappeared, while BLPs-SAM-FAdE treatment and antibiotic treatment can significantly alleviate the damage to the gastric mucosa and have a certain repair effect, thus avoiding the development of gastritis and gastric ulcer to a certain extent.

Discussion

H. pylori is a spiral-shaped Gram-negative bacteria which could invade the human stomach. It was identified by Australian academics Marshall and Warren in the 1980s [29]. Chronic gastritis, peptic ulcers, gastric cancer, and gastric mucosa-related lymphoid tissue lymphoma are the gastrointestinal disorders most closely connected with this. H. pylori was classified as a clear carcinogen in the United States Food and Drug Administration’s 15th edition of the Carcinogen Report, which was published in 2021 (https://ntp.niehs.nih.gov/go/roc15) [30]. H. pylori is one of the bacteria with the highest infection rates in the globe. It is 50.8% in underdeveloped nations, while my country has an infection incidence of around 44.2% [31]. Faced with such a dire scenario, our project team has spent the last several years doing extensive research and development of oral H. pylori vaccines, with the goal of eliminating H. pylori at the point of invasion by stimulating the mucosal immune response. The genetic heterogeneity of H. pylori strains presents a significant challenge for vaccine development. This heterogeneity implies that a single antigen may not effectively cover all strains, thereby limiting the vaccine’s broad applicability. Consequently, the multi-epitope antigen FAdE was chosen for vaccine antigen selection based on these considerations. The FAdE antigen comprises four key adhesins of H. pylori: urease, Lpp20, HpaA, and CagL. The selected fragments include T-helper and B-cell epitopes (UreA_27-90, UreA_183-225, UreB_327-385, Lpp20_58-97, HpaA_52-194, and CagL_21-139) as well as a known Lpp20 mimotope (Lpp20^M). These epitopes are relatively conserved among different H. pylori strains. By combining multiple epitopes, the FAdE antigen is designed to provide broader immune coverage, effectively addressing the genetic heterogeneity of H. pylori [26]. Many studies have addressed BLPs’ safety and biocompatibility for mucosal vaccines, as well as their capacity to generate substantial cross-immune protection [32].

BLPs have unique adjuvant characteristics and contribute to protective immunity. The key to presenting antigens via BLPs is that a particular element on the antigen can attach to the L. lactis PGN shell and transport the antigen to the immunological activation site [16]. PA is now the most often utilized anchor in the BLP system, and BLPs may be used to show foreign antigens by bridging PA [33]. PA is generated from the C-terminus of the lactic acid bacterium cell wall hydrolase AcmA and is made up of three repetitive lysine motifs (LysM) [34]. As a consequence, particle vaccine was fused with PA and incubated with BLPs via high-affinity non-covalent binding [35]. In this study, we first used the E. coli inducible expression system to obtain the M cell-targeted H. pylori multi-epitope antigen SAM-FAdE fused with PA, with a purity of up to 90% and a concentration of 0.6 mg/mL. At the same time, a comprehensive comparison was made between 10% TCA and 0.1 M HCL to heat-treat L. lactis. The BLPs obtained after treatment with 0.1 M HCL were selected. Their morphology and particle size were consistent with the live NZ9000, indicating that the hot acid is only to remove the L. lactis’s proteins and nucleic acids. And obtained dead L. lactis - BLPs, maintain the character and size of the L. lactis cell wall. Finally, under the action of PA, SAM-FAdE displayed on the BLPs.

In recent years, BLPs delivery systems have been applied to a variety of pathogens, including human papillomavirus (HPV) [36], porcine circovirus type 2 (PCV2) [37], and foot-and-mouth disease virus (FMDV) [20], and can prevent the pathogens invasion through different immune methods. In this study, BLPs conveying H. pylori multi-epitope antigens eliminated H. pylori colonized in the gastric mucosa by inducing antigen-specific CD4⁺ T cell responses, mucosal immunity, and production of antibodies. Since the germinal center is the hub of B cell proliferation and differentiation, plasma cell production, and antibody class switching [28], we measured the intensity of the B cell response in addition to the level of plasma cell production two weeks after the fourth immunization. When it was established that BLPs-SAM-FAdE could significantly enhance the GC B cell response, we found antigen-specific IgG antibodies in the serum together with their subtypes (IgG1, IgG2a). The BLPs-SAM-FAdE group was able to discover the greatest titers of antigen-specific IgG, IgG1, and IgG2a when compared to the control group. These findings, together with the others, suggest that the vaccination can generate humoral immune responses.

BLPs-SAM-FAdE stimulated mucosal immune responses more than SAM-FAdE protein, according to our study, which compares specific sIgA levels in gastric lavage fluid, intestinal lavage fluid, and feces. These findings show that BLPs are a useful mucosal vaccine delivery vehicle. There are currently studies using BLPs carriers to deliver corresponding pathogenic microbial antigens through mucosal inoculation (oral, nasal drops) to prevent microbial infections, such as BLPs-PspA (Streptococcus pneumoniae) [38], ENV-BLPs (HIV virus) [39], H1N1 split-BLPs (influenza virus) [40], can enhance specific sIgA antibody levels.

A growing body of research has demonstrated that T cell-dominated cellular immunity, particularly that of CD4⁺ T cells, is essential for eliminating and evading H. pylori and is also involved in the onset and progression of digestive disorders [41]. Th1 predominantly secretes TNF-α and IFN-γ, which can activate NK cells and macrophages to increase H. pylori’s capacity to be killed. On the other hand, inflammation and injury to the stomach mucosa can readily result from a strong Th1 response [42]. Th2 primarily secretes cytokines such as IL-4, IL-5, and IL-13 to encourage B cell activation and antibody production. Th2 protects the stomach mucosa from damage by regulating immune responses and preventing overly aggressive inflammatory reactions. Furthermore, the increased Th2 response aids in the neutralization of H. pylori-produced toxins [43]. Th17 predominantly produces IL-17. In the early stages of H. pylori infection, Th17 promotes epithelial cell defense mechanisms, which improves H. pylori clearance [44]. Tregs secrete inhibitory cytokines like IL-10 and TGF-β to control the immune response and protect the stomach mucosa from injury. CD4⁺ T cell subtypes interact in diverse ways and maintain dynamic balances [45]. The balance of Th1, Th2, Th17, and Treg cells is crucial for immune response control. Once this equilibrium is broken, H. pylori may enter, causing the illness to continue. In this investigation, we employed flow cytometry intracellular factor staining, ELISPOT, and ELISA to assess the CD4⁺ T cell response induced by BLPs-SAM-FAdE. Before triggering T cell responses, DCs’ capacity to deliver antigens is critical for inducing humoral and cellular immunity. DCs are the most important antigen-presenting cells, with the greatest potential to activate early T cells. Studies shown that BLPs activate and promote the maturation of neonatal and adult mouse bone marrow-derived CD11c⁺ cells in vitro, resulting in increased expression of mature cell surface markers (CD80, CD86, CD40, and MHC-II) and secretion of pro-inflammatory cytokines (IL-12p70, TNF-a, IL-10, IL-6, IFN-γ, and MCP-1) [32]. In this work, we investigated the capacity of SAM-FAdE and BLPs-SAM-FAdE to induce BMDC maturation in vitro. Our findings revealed that, as compared to SAM-FAdE, BLPs-SAM-FAdE could considerably stimulate BMDC maturation, with MHC II, CD86, CD80, and CD40 greatly up-regulated, indicating higher antigen presentation and costimulatory activities. In addition, given that IL-12 and IL-1β are crucial for the early polarization and sustained activity of Th1 cells [46], we detected the levels of IL-1β, IL-12p70, IL-4, and IL-6 produced in the supernatants of BMDCs co-cultured with SAM-FAdE and BLPs-SAM-FAdE. These correspond to the early polarization of Th1, Th2, and Th17, respectively. Our findings demonstrate that BLPs-SAM-FAdE stimulates Th1, Th2, and Th17 in vivo, dynamically balancing defense against H. pylori infection.

The most favorable component of this work is the use of M cell-targeting peptides to help the H. pylori multi-epitope antigen FAdE arrive reach immune cells in the M cell pocket via the BLP delivery method. This pathway is required for naïve B cells to develop into plasma cells that produce sIgA. The unique physiological structure of M cells, which consists of dispersed microvilli, permits antigens to come into close contact with the top surface of M cells, boosting granular antigen absorption efficiency [47]. On the other hand, its basolateral invagination permits antigen-presenting cells and lymphocytes to accumulate near the lumen, reducing antigen transit time [48]. Based on these two benefits, directly targeting antigens to M cells is not only a novel technique for improving antigen utilization, but it is also an important strategy for inducing mucosal immunity and producing large levels of sIgA. M cells are found in MALT in numerous parts of the human; thus, they have been exploited to develop a range of mucosal vaccinations. In this study, our M cell-targeted oral vaccine delivery system based on BLPs presented FAdE (BLPs-SAM-FAdE) was validated not only in vitro by constructing a M cell model, but also in vivo through ileal closed-loop experiments, and the results showed that BLPs-SAM-FAdE can clearly target M cells. The targeting peptide Mtp is essential to this system. Following bio-confidence analysis and literature screening, the M-cell targeting peptide Mtp was created. Its primary components are CKS-9, Co1, and CPE30 ligands, which are joined by a Linker. This Linker has the capacity to bind selectively to the M-cell surface receptor (C5aR, Claudin4), enabling the BLPs-SAM-FAdE to precisely identify and target M cells while preventing uptake by other non-target cells. To date, M cell-targeted oral delivery systems based on PLA and PLGA particles have been used for a variety of vaccines, including hepatitis B surface antigen (HBsAg) and ovalbumin (OVA) [49, 50]. In addition, studies have shown that the use of UEA-1-labeled PLGA lipid nanoparticles containing the Toll-like receptor (TLR) agonist monophosphoryl lipid A (MPL) facilitates the delivery of oral vaccines [51]. UEA-1 and MPL-modified PLGA liposomes were transported to M cells and captured by mucosal dendritic cells (DCs). After in vivo vaccination, OVA-coated PLGA lipid globules stimulate the secretion of potent mucosal IgA and serum IgG during oral formulation. Studies have also shown that Salmonella typhimurium can specifically target M cells to provide excellent immune effects, and has been successfully used as a vaccine vector against H. pylori, Listeria, Anthrax and Tuberculosis [52–55]. In this study, we constructed a safer and biocompatible BLPs delivery system to target M cells. Its particle size and source are undoubtedly more advantageous.

This study has several limitations. While the results observed for this vaccine in the mouse model provide valuable insights, the applicability of these findings to human cells or patient plasma warrants careful consideration. The gastric structure of mice differs significantly from that of humans, including variations in the thickness of the gastric mucosa, the volume of gastric acid secretion, and the patterns of gastric peristalsis. These differences may lead to variations in the processes of H. pylori infection, pathological changes, and immune responses, which may not be directly comparable between animals and humans. We also recognize the critical need for increased data from human samples. However, due to constraints related to time, resources, and ethical considerations, the translation of animal experiment results into human applications presents numerous challenges. Consequently, we were unable to incorporate human sample data in our current study. Thus, investigating the effects of H. pylori vaccines in humans remains a significant challenge for researchers, necessitating the collection of extensive human sample data to advance research in this field.

Conclusion

In summary, this study successfully developed M cell-targeted H. pylori multi-epitope antigen FAdE based on BLPs carrier oral mucosal vaccine. Regarding the carrier, BLPs are derived from L. lactis and meet food-grade standards. Additionally, they consist of ‘dead bacteria,’ which are unaffected by their own genetic material and exhibit potential biocompatibility. Another advantage of BLPs is that exogenous antigens loaded on their surface demonstrate greater stability than free antigens, making them less susceptible to degradation by gastric acid and various enzymes. Under the influence of the PA binding domain, the H. pylori antigen FAdE is displayed on the surface of BLPs with an efficiency of up to 90% and exhibits strong immunogenicity. In vivo studies have demonstrated that BLPs-SAM-FAdE can activate antigen-specific immune responses, effectively eliminating H. pylori colonization in the gastric mucosa by inducing significant mucosal immunity, which leads to the production of specific sIgA. These findings indicate that BLPs serve as promising mucosal carriers capable of efficiently delivering H. pylori antigen to M cells, thereby initiating mucosal immune responses. Additionally, their ease of promotion and application, along with their high safety profile, suggest that this approach could represent a novel method for mucosal vaccine development.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

F.Z. conducted the statistical analysis and drafted the manuscript. L.G. and K.L. critically revised and finalized the manuscript. J.C., Z.Z., J.W., L.N., Y.Q. and G.Z. performed the data analysis and interpretation. F.Z., L.G. and K.L. reviewed and edited the manuscript. F.Z., J.C. and J.W. designed the study and performed all experiments. All authors have approved the final version of the manuscript.

Funding

This work was financially supported by the National Natural Science Foundation of China (32070930, 82160497, 82360711), Key R & D Plan Project of Ningxia Autonomous Region (2020BFG02012), Natural Science Foundation of Ningxia (2022AAC02034, 2023AAC03155, and 2023AAC03207), Open subject of Ningxia Key Laboratory of Clinical Pathogenic Microbiology (MKLG-2024-02), Director Fund of Ningxia Key Laboratory of Craniocerebral Disease (LNZR202304), Ningxia Youth Top Talent Training Project and Leading Talents in Scientific and Technological Innovation in Ningxia.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

This study was approved by the Animal Experimentation Ethics Committee of Ningxia Medical University (IACUU number: NYLAC-2022-169).

Consent for publication

All authors read and approve the final manuscript.

Competing interests

The authors declare no competing interests.