Therapeutic Application of an Extract of Helicobacter pylori Ameliorates the Development of Allergic Airway Disease

Asthma is one of the most prevalent chronic diseases worldwide (1), and a further increase in asthma prevalence is expected (2). Asthma is a heterogeneous syndrome that is characterized by airway obstruction, hyperresponsiveness, and inflammation (3). As part of the heterogeneity of this syndrome, patients can be assigned to certain phenotypes based on clinical or immunological criteria (4). One of the most prominent clinical phenotypes is allergic asthma, which is characterized by the induction of inflammation following exposure of sensitized patients to allergens. Indeed, the immunological pathophysiology underlying allergic asthma is well described and involves complex interactions of epithelial cells, dendritic cells (DCs), mast cells, T cells, B cells, and eosinophilic granulocytes (3, 5, 6). So far, only symptomatic treatment is available for these patients.

The increase in the prevalence of asthma has been linked to changes in environmental exposure, such as tobacco smoke and traffic emission, as well as to changes in microbial exposure (7, 8). High hygienic standards seem to predispose to the development of allergies, such as allergic asthma (9), which was later postulated as the hygiene hypothesis. It is now argued that changes in exposure to microorganisms (10), as well as other factors associated with a Western lifestyle, such as diet, antibiotic use, and cesarean sections (11), are related to the altered composition of our microbiota, which has been implicated in the loss of tolerance to allergens. An example of a microorganism that is found less frequently in the human microbiota today is the pathobiont H. pylori (12). H. pylori is a Gram-negative spirochete that resides in the stomach (13) and is known for its association with gastritis, gastric ulcer, and gastric cancer (14). Interestingly, multiple epidemiological studies have reported a protective effect of infection with H. pylori on the development of asthma (15, 16).

In addition, an increasing number of experimental studies showed that, especially, neonatal infection with H. pylori or application of an extract of the bacteria from neonatal age on reduced allergic airway disease (17–19). Further mechanistic studies showed that protection induced by H. pylori involved induction of tolerogenic DCs and regulatory T cells (Tregs) via IL-18– and IL-10–dependent mechanisms. However, the protective effects of H. pylori have been exclusively assessed in prophylactic models, investigating animals infected or treated before sensitization.

Therefore, in this study, the therapeutic effect of an H. pylori extract was assessed in sensitized animals that were treated just prior to airway challenge. Treatment with H. pylori extract just prior to airway challenge was found to inhibit airway inflammation and remodeling, and changes in the DC compartment were observed. Importantly, this effect was observed in two models of allergic airway disease. These data show the therapeutic potential of H. pylori components for the treatment of allergic asthma.

Eight- to twelve-week-old female C57BL/6 mice were purchased from Charles River (‘s-Hertogenbosch, the Netherlands). Animals were housed under specific pathogen–free conditions in individually ventilated cages, with free access to food and water, at the animal facilities of the Leiden University Medical Center. All experiments were approved by the Ethics Committee for Animal Experimentation of the University of Leiden (Dieren Experimentele Commissie 12246) and were conducted under strict governmental and international guidelines in accordance with EU Directive 2010/63/EU.

An allergic reaction to OVA in the lungs of mice was induced by i.p. injection with 10 μg of OVA (InvivoGen, San Diego, CA) suspended in 1 mg of aluminum hydroxide (alum) (Imject alum adjuvant; Pierce, Thermo Fisher Scientific, Waltham, MA) in 500 μl of PBS on days 0 and 7, followed by three aerosol challenges with 1% (w/v) OVA (albumin, from chicken egg white, grade III; Sigma-Aldrich, Zwijndrecht, the Netherlands) in PBS using an ultrasonic nebulizer (NE-U17; Omron, Osaka, Japan) for 25 min on days 14, 15, and 16 (Fig. 1A). Control mice were injected i.p. with PBS and challenged with 1% OVA in PBS. Mice were treated i.p. with 200 μg of H. pylori extract in 100 μl of PBS or 100 μl of PBS as a control on days 13, 14, and 15. When mice were treated and challenged on the same day, they were challenged 1 h after the treatment. Mice were analyzed 1 d after the last challenge. House dust mite (HDM) allergic airway inflammation (AAI) was induced by one intranasal (i.n.) sensitization with 1 μg of HDM (Greer, Lenoir, NC) in 50 μl of PBS on day 0, followed by five i.n. challenges with 10 μg of HDM in 50 μl of PBS on days 7, 8, 9, 10, and 11 (Fig. 1B). Intranasal administration was performed under isoflurane anesthesia (3%, 0.8 l/min). Control mice were sensitized with 1 μg of HDM and challenged with PBS. On days 6, 7, 9, and 11, H. pylori extract (200 μg) in 100 μl of PBS or 100 μl of PBS as a control was applied i.p. These mice were analyzed on day 13. After euthanasia with an overdose of sodium pentobarbital (Euthasol; AST Farma, Oudewater, the Netherlands), blood was collected from the vena cava, and the lungs of mice were lavaged three times with 1 ml of PBS via a tracheal cannula and then were perfused with PBS via the heart. The left lobe of the lungs was collected for flow cytometry, whereas the right lobe was inflated with 3.9% formaldehyde (Sigma-Aldrich) in PBS to study the histology.

Oxidase, catalase, and urease production by H. pylori premouse SS1 strain grown on H. pylori culture plates (bioMérieux, Benelux B.V., Zaltbommel, the Netherlands) was confirmed before bacteria were transferred to Brucella broth supplemented with 10% FBS (Life Technologies, Thermo Fisher Scientific). Bacteria were grown under microaerophilic conditions (6% O₂, 7.1% CO₂, 7.1% H₂, and 79.7% N₂) (Anoxomat; Mart Microbiology, Drachten, the Netherlands) in a shaking incubator (New Brunswick Scientific Europe B.V., Nijmegen, the Netherlands) at 185 rpm and 37°C overnight. The following day, the culture was expanded twice by adding two parts medium to one part culture. After overnight culturing, the bacteria were harvested by centrifugation. Gram staining was performed to check the culture. Pellets of multiple cultures were pooled, washed in PBS, and freeze-thawed in nitrogen. The bacteria were lysed using a French pressure cell press (Cell Pressure Homogenizer; Stansted Fluid Power, Harlow, U.K.) at 3000 bar. The product was centrifuged, and the supernatant was filtered through a 0.2-μm filter. Protein content was determined by bicinchoninic acid (Pierce BCA Protein Assay Kit; Thermo Fisher Scientific) and adjusted to the desired concentration using PBS. Approximately 30 mg of extract was obtained from 8 l of bacterial culture. Toxicity of the extract was ruled out by testing in various in vitro assays (unpublished observations, Y. van Wijck).

OVA/HDM-specific IgE and IgG1 Abs in the serum were measured by ELISA. Nunc MaxiSorp plates (Sigma-Aldrich) were coated with 25 μg/ml OVA (grade III; Sigma-Aldrich) or 50 μg/ml HDM in 0.1 M sodium carbonate buffer (pH 9.6). Nonspecific binding plates were blocked with 1% (w/v) BSA (Sigma-Aldrich) in PBS. Samples and an anti-OVA IgE standard (Bio-Rad, Veenendaal, the Netherlands) were incubated, and OVA/HDM-specific Abs were detected by biotin-labeled anti-mouse IgE (clone 23G3; SouthernBiotech, Birmingham, AL) and biotin-labeled anti-mouse IgG1 (clone A85-1; BD Pharmingen, Becton Dickinson B.V., Breda, the Netherlands), followed by streptavidin coupled to HRP (Sanquin, Amsterdam, the Netherlands) and a 3,3′,5,5′-tetramethylbenzidine reaction. Plates were measured on a Bio-Rad iMark microplate reader at 450 nm. For anti-OVA IgE, the detection limits were determined by the upper and lower concentration of the linear part of the standard curve. Values higher or lower than these values were set to the detection limits.

Cells in bronchoalveolar lavage fluid (BALF) were counted using Türk’s solution (Merck, Schiphol-Rijk, the Netherlands). A single-cell suspension of the lung tissue cells was obtained by dissecting the tissue into small pieces, followed by Collagenase/DNase (Collagenase type I of Clostridium histolyticum [Calbiochem, San Diego, CA]; DNase I [Sigma-Aldrich]) treatment for 45 min, dispersion through a 70-μm cell strainer (BD Falcon, Becton Dickinson B.V.), and lysis of the RBCs. Lung tissue cells were counted on a Countess automatic cell counter (Invitrogen, Thermo Fisher Scientific). Next, cells in BALF and lung tissue were washed in PBS, stained with LIVE/DEAD aqua stain (Invitrogen, Thermo Fisher Scientific), washed, and fixed with 2% formaldehyde (Merck) in PBS. The following Abs were used to distinguish cell populations in BALF by flow cytometry: Ly-6G and Ly-6C-FITC (Gr-1) (RB6-8C5; BD Biosciences, Becton Dickinson B.V.), Siglec-F–PE (E50-2440; BD Biosciences, Becton Dickinson B.V.), CD3–PerCP–eFluor 710 (17A2; eBioscience, Thermo Fisher Scientific), CD11b–PE–Cy7 (M1/70; eBioscience, Thermo Fisher Scientific), CD45R–allophycocyanin–eFluor 780 (B220) (RA3-6B2; eBioscience, Thermo Fisher Scientific), and CD11c-V450 (HL3; BD Biosciences, Becton Dickinson B.V.). To analyze DC populations in the lungs, the following Abs were used: CD3-FITC (17A2; eBioscience, Thermo Fisher Scientific), CD19-FITC (MB19-1; eBioscience, Thermo Fisher Scientific), NK1.1-FITC (PK136; eBioscience, Thermo Fisher Scientific) and Ly-6G and Ly-6C-FITC (Gr-1) (RB6-8C5; BD Biosciences, Becton Dickinson B.V.) in a dump channel, Siglec-F–PE (E50-2440; BD Biosciences, Becton Dickinson B.V.), CD64-PerCP/Cy5.5 (FcγRI) (X54-5/7.1; BioLegend, San Diego, CA), CD11b–PE–Cy7 (M1/70; eBioscience, Thermo Fisher Scientific), CD103–Alexa Fluor 647 (2E7; BioLegend), MHC class II (MHCII) (I-A/I-E)–allophycocyanin–eFluor 780 (M5/114.15.2; eBioscience, Thermo Fisher Scientific), and CD11c-V450 (HL3; BD Biosciences, Becton Dickinson B.V.). Ab mixes were supplemented with mouse FcγRII/III-binding inhibitor (2.4G2; Bioceros, Utrecht, the Netherlands). Cells were measured on a FACSCanto II (BD Biosciences, Becton Dickinson B.V.), and analysis was performed using FlowJo (v7.6.5) software (TreeStar, Ashland, OR). Gating strategies can be found in Supplemental Figs. 1 and 2. Gates were placed according to unstained and Fluorescence Minus One samples.

The formaldehyde-inflated tissue was fixed for 48 h at room temperature. Four-micrometer paraffin sections were stained with H&E (Klinipath, Duiven, the Netherlands). Peribronchial inflammation, as examined by H&E staining, was scored on a scale from 0 to 4 by two investigators who were blinded to the identity of the slides. Alcian blue–periodic acid–Schiff (PAS) (both from Sigma-Aldrich) staining was performed to quantify the extent of goblet cell metaplasia. For analysis, digital images of Alcian blue–PAS–stained lung sections were taken at 100× magnification on an Olympus BX41 microscope (Olympus Nederland B.V., Leiderdorp, the Netherlands) and analyzed with ImageJ software. Briefly, epithelial regions were selected and separated into red, green, and blue channels. The color threshold was manually set in green channeled images in such a way that the magenta mucin content matched the color threshold by comparing the original Alcian blue–PAS–stained images simultaneously. These images were converted into binary images, and the area fraction of mucin density (%) was calculated using “summarize” and “analyze particle” options.

Mice were sensitized to OVA by two i.p. injections with 10 μg of OVA (InvivoGen) suspended in 1 mg of alum (Pierce, Thermo Fisher Scientific) in 500 μl of PBS on days 0 and 7. To analyze the effect of the extract on processing of allergen by DCs, mice were treated i.p. with 200 μg of extract in 100 μl of PBS on days 13 and 14. An hour after the second treatment, mice were challenged i.n. with DQ OVA (40 μg in 50 μl of PBS) (Thermo Fisher Scientific), a self-quenched form of OVA that emits light only when it is processed by proteases. Processing was analyzed 24 h after the challenge. Single-cell suspensions of the lung and mediastinal lymph node (mLN) were stained with Aqua and fixed in formaldehyde. Subsequently, cells were stained for CD19–allophycocyanin–Cy7 (1D3; BD Pharmingen, Becton Dickinson B.V.), CD11c-V450 (HL3; BD Biosciences, Becton Dickinson B.V.), and MHCII (I-A/I-E)–allophycocyanin (M5/114.15.2; eBioscience, Thermo Fisher Scientific) in the presence of mouse FcγRII/III-binding inhibitor (2.4G2; Bioceros) and analyzed by flow cytometry.

Bone marrow was isolated from C57BL/6 mice. Cells were cultured in RPMI 1640 GlutaMAX (Thermo Fisher Scientific) supplemented with 100 U/ml penicillin (Sigma-Aldrich), 100 μg/ml streptomycin (Sigma-Aldrich), 5 × 10⁻⁵ M 2-ME (Sigma-Aldrich), 5% heat-inactivated FBS (Bodinco, Alkmaar, the Netherlands), and 20 ng/ml GM-CSF (PeproTech, London, U.K.). The medium was refreshed on days 3 and 6. On day 7, cells were replated, rested for 3 h, and stimulated with 20 μg/ml nonlabeled OVA (InvivoGen) or 1 μg/ml DQ OVA (Thermo Fisher Scientific) in the presence or absence of 100 μg/ml H. pylori extract for 24 h. Cells were stained with Aqua, labeled with CD11c-V450 (HL3; BD Biosciences, Becton Dickinson B.V.) and MHCII (I-A/I-E)–allophycocyanin (M5/114.15.2; eBioscience, Thermo Fisher Scientific) in the presence of mouse FcγRII/III-binding inhibitor (2.4G2; Bioceros), and analyzed by flow cytometry.

Graphs were designed and statistical analysis was performed in GraphPad Prism 7.02 (GraphPad Software, La Jolla, CA). Data are shown as means ± SEM. Differences were considered significantly different at p values < 0.05, as determined by ANOVA and the Dunnett multiple-comparison analysis.

The development of AAI was assessed in OVA-sensitized mice, which received H. pylori extract by i.p. injection and then were challenged with OVA aerosols (experimental setup in Fig. 1A). Sensitization was confirmed by increased serum levels of allergen-specific IgE and IgG1 (Fig. 2). Sensitized animals treated with H. pylori extract exhibited levels of allergen-specific IgE and IgG1 comparable to the sensitized and challenged controls. Sensitized and challenged animals showed increased numbers of total cells and eosinophils in flow cytometry analysis of BALF (gating Supplemental Fig. 1) compared with animals that were challenged but not sensitized (Fig. 3A, 3B). Treatment with H. pylori extract significantly decreased the total cell count, eosinophil count, and percentage of eosinophils in BALF (Fig. 3). This anti-inflammatory effect was also apparent when analyzing lung tissue. Indeed, sensitized and challenged animals showed increased airway inflammation (Fig. 4A, 4B), as well as goblet cell metaplasia (Fig. 4A, 4C), compared with animals that were only challenged. Peribronchial and perivascular inflammation in allergic mice was significantly reduced upon treatment with H. pylori extract. Furthermore, mucus-producing cells in the airways were decreased in mice treated with H. pylori extract.

DCs play a pivotal role in the development and maintenance of allergic airway disease. DC subtypes were assessed in the lung by flow cytometry based on previously published methods (Supplemental Fig. 2). As expected, sensitization and challenge resulted in an increase in the number of DCs (CD11c⁺MHCII⁺) in the lung (Fig. 5A). Interestingly, the overall number of DCs was decreased in animals treated with H. pylori extract. In addition, treatment with H. pylori extract altered the composition of DC subtypes in the lung. The ratio of DCs with a proinflammatory phenotype to DCs with an anti-inflammatory phenotype was reduced after treatment with the extract (Fig. 5B). Animals treated with H. pylori extract showed a reduction in the ratio of proinflammatory monocyte-derived CD11b⁺CD64⁺ DCs to conventional CD11b⁺CD64⁻ DCs (Fig. 5C).

To further examine the effect of an extract of H. pylori on allergic airway disease, a model with a human relevant allergen and a route of allergen exposure that is more comparable to human exposure was used. Mice were sensitized to HDM by i.n. application and received H. pylori extract by i.p. injections prior to and during the allergen challenges (Fig. 1B). In contrast to the findings in the OVA/alum model, allergen-specific IgE and IgG1 serum levels were significantly reduced following treatment with the extract in this model (Fig. 6). A trend toward a lower total cell count in BALF was observed after treatment (Fig. 7A). Flow cytometry analysis showed a significant reduction in the absolute number and percentage of eosinophils in BALF following treatment with H. pylori extract (Fig. 7B, 7C). In addition, allergen-induced lung tissue inflammation and mucus-producing cells in the airways were significantly reduced upon treatment with H. pylori extract (Fig. 8). Interestingly, treatment with H. pylori extract resulted in a trend toward decreased DCs in the lung tissue (Fig. 9A) and a shift in DC phenotype, with a significant effect on the ratio of CD11b⁺/CD103⁺ DCs and a trend toward a reduced ratio of CD11b⁺CD64⁺/CD11b⁺CD64⁻ DCs (Fig. 9B, 9C).

To further assess whether the observed changes in the DC compartment following treatment with H. pylori extract affected DC function, we analyzed Ag processing by DCs. Mice sensitized to OVA were treated with H. pylori extract and subsequently challenged once with DQ OVA by i.n. application. Detection of DQ OVA by FACS is only possible after Ag processing, because proteases cleave the quenching molecules from the DQ OVA, resulting in fluorescence. Treatment with H. pylori extract significantly reduced the number of fluorescent DCs in the lung (Fig. 10A, 10B). In the mLN, a trend toward a decrease in fluorescent DCs was observed (Fig. 10C). To analyze whether the effect of the extract on processing by DCs could be due to a direct effect on DCs, bone marrow–derived DCs (BMDCs) were isolated and exposed to DQ OVA in the presence or absence of H. pylori extract for 24 h. In this in vitro culture, exposure of DCs to H. pylori extract resulted in a significant reduction in fluorescence compared with DCs that were only exposed to DQ OVA (Fig. 10D, 10E).

Previous studies have shown that infection with H. pylori, as well as exposure to an extract of H. pylori from especially neonatal age on, prior to sensitization, reduces the development of allergic airway disease (17–19). The aim of this study was to further examine whether administration of H. pylori extract after sensitization, and only prior to airway challenge, was also effective in reducing the hallmarks of allergic airway disease, thereby modeling a more therapeutic application. Sensitization leads to an increase in Ag-specific allergen capture and processing by various APCs in the airway mucosa at challenge, facilitated by allergen-specific Abs (21). Furthermore, there is increased DC precursor recruitment but also DC proliferation (22) and rapid T cell activation at challenge (21). Therefore, therapeutic interventions to reduce AAI at this stage are more challenging compared with preventive strategies, and many compounds that have been effective in prophylactic approaches failed to reduce allergic airway disease when used in therapeutic models. This study shows that systemic treatment of sensitized mice with an extract of H. pylori before allergen exposure decreases the development of AAI. This was observed in a model of systemic sensitization following i.p. injection of the allergen OVA, as well as in a model of sensitization following HDM allergen administration into the lung. The OVA/alum model is a well-known model that has contributed to the understanding of the mechanisms of AAI (23). In contrast, HDM is a human-relevant allergen that is applied i.n. without an adjuvant for sensitization or challenge, which is a more common route than the i.p. injection used for sensitization in the OVA/alum model (24, 25).

It is now well established that exposure to microorganisms is an important factor in shaping the immune system and also is important for the susceptibility to develop allergies and asthma (26). This holds true for the exposure to exogenous bacteria, and different research groups have shown that exposure to bacteria prior to or starting at the time of sensitization reduces the development of allergic disease in various models (27–30). There is also increasing evidence that microbiota contribute to the susceptibility to develop an allergic disease, and changes in the composition of the microbiota have been linked to the increased prevalence of allergies and asthma (31, 32). Indeed, it has been demonstrated that the composition and diversity of gut and lung microbiota influence the susceptibility for the development of allergic airway disease (33, 34). As one specific component of the human microbiome, H. pylori (the dominant bacterial species in the gastric mucosa) has received increasing attention with regard to its inverse association with asthma (15). Several previous studies have demonstrated that infection with (but also an extract of) H. pylori has a protective effect on the development of allergen-induced airway disease (17–19). The present study adds novel insights, showing the efficacy of the therapeutic use of an extract of H. pylori. Systemic application of H. pylori extract significantly reduced AAI, as well as goblet cell metaplasia, and these effects were apparent in two models of allergic airway disease. This is of interest, because these findings suggest a potential therapeutic effectiveness of this extract.

Gnotobiotic mice would make it possible to analyze the effects of H. pylori on AAI in the absence of other microorganisms. However, this makes it harder to translate the effects to humans, which have a large microbiome, and it is not clear whether H. pylori affects the microbiome (35, 36). In this study, an extract of the bacteria is used, and the treatment is only for a few days; therefore, it is unlikely that the extract affects AAI by affecting the microbiome. The protective effect of infection or H. pylori extract for the development of allergic airway disease has been linked to tolerogenic reprogramming of DCs by the bacteria (18, 19). Therefore, we also analyzed the composition of DC subsets in this study. Interestingly, application of H. pylori extract resulted in a reduction in CD11b⁺CD64⁺ proinflammatory monocyte-derived DCs. This subset was not analyzed in the preventative setting. It is likely that the reduction in proinflammatory DCs is associated with an increase in tolerogenic DCs. Hoffmann et al. (37) described CD103⁺ DCs and plasmacytoid DCs as pulmonary DCs that can induce tolerance in allergic asthma. In the preventative setting, BATF3-dependent CD103⁺ DCs were indispensable for the protective effect of H. pylori infection or extract treatment on allergic airway disease (19). A decrease in the ratio of CD11b⁺/CD103⁺ DCs was also observed in our more therapeutic setting. Previous studies in animals showed that plasmacytoid DCs were also actively attracted to the lungs of allergic mice after H. pylori infection or extract treatment (19), but we did not analyze that subset in our more therapeutic experiments. Based on these data, it was hypothesized that inhibition of processing of Ag by DCs might be a mechanism by which treatment with an extract of H. pylori leads to reduced AAI. Indeed, our present in vitro experiments show that treatment with H. pylori extract reduced the Ag-processing capacity of BMDCs. This was detectable in vitro, as well as in vivo, because a reduced capacity of DCs in the lung (and to a lesser extent also in the mLN) to process allergen was observed in mice treated with H. pylori extract. Not only was the absolute number of DCs that processed Ag reduced (which could also be explained by the overall reduction in DCs in the lungs and mLN), there was a trend toward a reduction in the fraction of DCs that processed Ag and in the amount of Ag processed by the DCs in mice treated with the extract (unpublished observations, Y. van Wijck). It is tempting to speculate that H. pylori might reduce Ag processing by DCs as a way to promote bacterial persistence in the host. It is unclear which components of the extract affect the DCs and by which mechanisms, as well as whether the same effect on Ag processing can be seen in the preventative setting. It would be interesting to analyze whether DC function is affected by vacuolating cytotoxin A, γ-glutamyl transpeptidase, and urease, which were shown to be of importance in the preventative setting (19, 38, 39). Similar to the effects of a helminth cathelicidin-like protein on macrophages (40), inhibition of endolysosomal acidification might be a mechanism by which DC processing is reduced by H. pylori extract. Furthermore, previous reports show reduced uptake of particles by monocyte-derived DCs when preincubated with H. pylori (41) and inhibition of Ag presentation by these DCs after H. pylori incubation (42). However, in the current study, reduced uptake of Ag, as assessed by adding fluorescently labeled Ag to BMDCs treated with the extract, was not detected (unpublished observations, Y. van Wijck), which reduces the likelihood that the effects on processing are caused by reduced uptake.

Interestingly, the suppression of allergic airway disease was achieved by systemic application of the H. pylori extract, clearly showing a systemic effect of this therapy. It can be speculated that exposure to this bacterial extract shapes the immune system to a more tolerogenic response. However, this effect seems to be independent of the mere exposure to LPS, because previous studies have shown that Escherichia coli or Salmonella extract was not effective in reducing the development of allergic airway disease (19). Following i.p. application, H. pylori extract is exposed to cells in the peritoneal cavity, which have been shown to travel to the mLN and, therefore, could exert a suppressive function on the lungs. Indeed, Kool et al. (43) have shown that the mLN is involved in the reaction to an Ag following i.p. injection, even before the spleen and nondraining lymph nodes were fully involved. Therefore, i.p. injection of the extract seems to be a suitable route for inhibiting immune responses in the lungs. This was the case when sensitization was induced by i.p. injection of OVA, as well as in the HDM model, in which sensitization was induced by application of the allergen into the lung.

Lung function tests are often performed in clinical trials to assess the effectiveness of asthma medication. These tests should be evaluated in future studies. It would also be interesting to test the extract in a two-phase challenge model of AAI, in which sensitized and challenged mice are treated before a second challenge. Such repeated exposure models are more comparable to human disease, as patients with asthma are also repeatedly exposed to inhaled allergens.

In summary, an extract of H. pylori is effective in reducing the development of AAI in sensitized mice when used in a therapeutic fashion. A reduction in inflammation in BALF and tissue was observed in the OVA/alum and HDM models. The current data suggest that systemic application of H. pylori extract leads to induction of DCs with a less inflammatory phenotype and reduced Ag-processing ability. Therefore, the previously published prophylactic application can now be extended to prophylactic and therapeutic applications. Future studies are needed to assess whether this extract might be used in patients who are sensitized to allergens to prevent symptoms upon allergen exposure.

We thank Annemarie van Schadewijk, Padmini Khedoe, Anne van der Does, Simone Hermans, and Ruben van den Oever for technical assistance. Mouse FcγRII/III-binding inhibitor (2.4G2) was a kind gift of Louis Boon (Bioceros).

The authors have no financial conflicts of interest.