|Home | About | Journals | Submit | Contact Us | Français|
Th17-driven immune responses contribute to the pathogenesis of many chronic inflammatory diseases. In this study, we investigated the role of IL-17 signaling in chronic gastric inflammation induced by Helicobacter pylori, a Gram-negative bacterium that persistently colonizes the human stomach. Wild-type C57BL/6 mice and mice lacking IL-17 receptor A (IL-17RA-/-) were orogastrically infected with H. pylori. Differences in bacterial colonization density and gastric inflammation were not apparent at 1 month post-infection, but by 3 months post-infection, H. pylori colonization density was higher and mononuclear gastric inflammation more severe in infected IL-17RA-/- mice than in infected wild-type mice. A striking feature was a marked increase in gastric B cells, plasma cells, and lymphoid follicles, along with enhanced H. pylori-specific serum antibody responses, in infected IL-17RA-/- mice. Fewer gastric neutrophils and lower levels of neutrophil-recruiting chemokines were detected in infected IL-17RA-/- mice than in infected wild-type mice. Gastric IL-17a and IL-21 transcript levels were significantly higher in infected IL-17RA-/- mice than in infected wild-type mice or uninfected mice, which suggested that a negative feedback loop was impaired in the IL-17RA-/- mice. These results underscore an important role of IL-17RA signaling in regulating B cell recruitment. In contrast to many chronic inflammatory diseases in which IL-17RA signaling promotes an inflammatory response, IL-17RA signaling downregulates the chronic mononuclear inflammation elicited by H. pylori infection.
Helicobacter pylori persistently colonizes the stomachs of approximately half of the global human population. A hallmark of H. pylori infection is a gastric mucosal inflammatory response, termed superficial gastritis (1, 2). In most persons, H. pylori infection and gastritis persist for decades without any adverse effects. However, the presence of H. pylori increases the risk for development of duodenal ulcer disease, gastric ulcer disease, non-cardia gastric adenocarcinoma, and B-cell malignancies such as gastric mucosa associated lymphoid tumors (MALT lymphomas) and high grade lymphomas [reviewed in (3, 4)].
Previous studies have shown that H. pylori infection of humans and experimental infection of rodents typically results in a Th1 immune response (5-9). Recently, several studies have reported that IL-17a is expressed in the stomachs of H. pylori-infected humans and experimentally infected mice, which suggests that a Th17 response may also be elicited (10-13). However, the roles of IL-17 and Th17 responses in gastric immunopathology and control of H. pylori proliferation have not been fully elucidated.
The Th17 lineage develops in a pathway independent from Th1 and Th2 differentiation (14). A hallmark of Th17 cells is the production of IL-17, a pro-inflammatory cytokine. Th17 cells produce IL-17a and IL-17f homodimers, as well as IL-17a/f heterodimers. These cytokines bind to multimeric IL-17 receptors comprised of two IL-17RA subunits and one IL-17RC subunit (15). Additional IL-17 receptors also have been described, including IL-17RB, IL-17RD and IL-17RE [reviewed in (16)]. The functional roles of these additional receptors are not yet completely understood. IL-17 can induce an inflammatory response by signaling a variety of cell types (including epithelial cells, endothelial cells, and fibroblasts) to express IL-8 (or KC, MIP2 and LIX in mouse), IL-1β, TNF-α and IL-6 [reviewed in (16)].
Th17 cells are known to have an important role in a growing list of immune-mediated diseases, including inflammatory bowel disease, experimental autoimmune encephalopathy (EAE), and collagen-induced arthritis (CIA) [reviewed in (17-19)]. In addition, in several animal models of infectious diseases, the IL-23/IL-17 axis promotes cell migration to the site of infection to kill microorganisms and activates the bactericidal activity of macrophages [reviewed in (20)]. The host's ability to control the proliferation of Klebsiella pneumonia, Citrobacter rodentium, Mycoplasma pneumonia, Bordetella pertussis, Pseudomonas aeruginosa, Porphyromonas gingivalis, Escherichia coli, Listeria monocytogenes, Helicobacter felis, and Salmonella enterica is at least partially dependent on IL-23 and IL-17 (21-30).
In previous studies, IL-17RA-/- mice have been used to study the role of IL-17 and Th17 responses in various infections and inflammatory conditions (including colitis, synovitis, arthritis, and allergic asthma) (21, 26, 31, 32-38). In this study, we used IL-17RA-/- mice to investigate a potential role of IL-17 signaling during H. pylori infection.
Male and female IL-17RA-/- deficient mice on a C57BL/6 background were obtained from Amgen Inc. (Thousand Oaks, CA) for the establishment of a breeding colony. The IL-17RA-/- mice have a targeted deletion of exons 4-11 in the IL-17RA locus on mouse chromosome 6 (39). C57BL/6 mice (Taconic, Germantown, NY) were used as controls. Helicobacter-free male mice, 8-10 weeks old, were used in all experiments. The Vanderbilt University Institutional Animal Care and Use Committee approved all animal protocols used in this study. The IL-17RA-/- breeding pairs tested negative for intestinal Helicobacter species. Feces from sentinel mice housed in the same room were routinely tested by PCR for intestinal Helicobacter, pinworms, mouse parvovirus and several other murine pathogens, and consistently tested negative for each of these infections.
A mouse-passaged derivative of H. pylori strain SS1 was used in all experiments. Bacteria were grown on trypticase soy agar (TSA) plates containing 5% sheep blood. Alternatively, bacteria were grown in Brucella broth containing 5% heat-inactivated fetal bovine serum (FBS) and 10 μg/ml vancomycin. Cultures were grown at 37°C in either room air supplemented with 5% CO2, or under microaerobic conditions generated by a CampyPak Plus* Hydrogen + CO2 with Integral Palladium Catalyst (BD).
One day prior to infection of mice, H. pylori were inoculated into liquid medium and were cultured for 18 hours under microaerobic conditions, as described above. Mice were orogastrically inoculated with a suspension of 5×108 CFU H. pylori (in 0.5 ml of Brucella broth) twice over 5 days.
The stomach was removed from each mouse by excising between the esophagus and the duodenum. The forestomach (nonglandular portion) was removed from the glandular stomach and discarded. The glandular stomach was opened, rinsed gently in cold PBS, and cut into three longitudinal strips that were used for bacterial culture, RNA analysis, and histology. For culturing of H. pylori from the stomach, gastric tissue was placed into Brucella broth-10% FBS for immediate processing. Gastric tissue was stored in RNALater solution for subsequent RNA isolation. A longitudinal strip from the greater curvature of the stomach was excised and placed in 10% normal buffered formalin for 24 hours, embedded in paraffin and processed routinely for hematoxylin and eosin (H&E) staining. Indices of inflammation and injury were scored by a single pathologist (KW) who was blinded to the identity of the mice. Acute and chronic inflammation in the gastric antrum and corpus were graded on a 0-3 scale (40-42). Acute inflammation was graded based on density of neutrophils and chronic inflammation was graded based on the density of lamina propria mononuclear cell infiltration independent of lymphoid follicles. The density of lymphoid follicles were graded on a scale of 0-2 (0 absent, 1 low grade, and 2 high grade). Plasma cells were scored on a 0-2 scale, with 1 being defined as scattered plasma cells, not in clusters, and 2 defined as band-like infiltrates of plasma cells.
Gastric tissue was homogenized using a disposable pestle (Kimble-Kontes, Vineland, NJ). Serial dilutions of the homogenate were plated on trypticase soy agar plates containing 5% sheep blood, 10 μg/ml nalidixic acid, 100 μg/ml vancomycin, 2 μg/ml amphotericin, 200 μg/ml bacitracin, and 2500 U/ml polymyxin. After 5 to 7 days of culture under microaerobic conditions, H. pylori colonies were counted.
RNA was isolated from the stomach using the TRIZOL isolation protocol (Invitrogen, Carlsbad, CA) with slight modifications. Stomach tissue was homogenized in 1 ml of TRIZOL reagent and then two chloroform extractions were performed. Following an isopropanol precipitation, the RNA was washed with 70% ethanol and treated with RNase Inhibitor (Applied Biosystems, Foster City, CA) for 45 minutes. Following resuspension of the RNA at 65° C for 15 minutes, RNA preparations were further purified using the Qiagen RNA isolation kit and were treated with RNase-free DNase as directed by the manufacturer (Qiagen Inc., Valencia, CA). RNA was reverse transcribed using the High Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Foster City, CA). For real time rtPCR, we used the relative gene expression method (43). Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) served as the normalizer, and tissue from uninfected mouse stomachs served as the calibrator. All cDNA samples were analyzed in triplicate, along with “no reverse transcriptase” controls, using Applied Biosystems 7000 or Step One Plus real time PCR instruments. Levels of cytokine expression are indicated as “relative units”, based on comparison of tissue from H. pylori-infected mice with tissue from uninfected mice (calibrator tissue) (43). Primer and probe sets were purchased as Taqman Gene Expression Assays from Applied Biosystems.
To analyze gastric cellular infiltrates, whole mouse stomachs were harvested and processed. As a first step in the analysis, gastric epithelial cells were disrupted and released by incubating mouse stomachs in Hanks buffered saline solution (HBSS) containing 10% FBS, 15 mM HEPES, 5 mM EDTA and 0.014% DTT for one hour at 37°C. The released epithelial cells were discarded, and stomachs were then digested for 30 min at 37°C in RPMI containing 10% FBS and 1 mg/ml collagenase A. The suspension was passed through a 70 μm cell strainer (BD). Cells were harvested by centrifugation, washed, and then live cells were counted by using a hemocytometer and trypan blue exclusion staining. The samples were stained with 2 μg/ml anti-CD4, anti-CD8, and anti-CD3, or 1.5 μg/ml anti-Gr1, 2 μg/ml anti-CD11b and anti-B220 (BD) in a volume of 100μl FACS buffer (PBS, pH 7.4, containing 0.1% sodium azide, 0.1% BSA, and 20% mouse serum). Cells were washed, resuspended in FACS buffer, and analyzed on a BD LSR II flow cytometer (Beckon Dickinson). The numbers of each cell type in a sample were calculated as described previously (13).
Primary gastric epithelial cells were isolated using a modification of a previously published protocol (44). The glandular portion of the stomach was opened via dissection, and after washing the stomach in cold PBS, the tissue was placed in 0.04% sodium hypochlorite for 15 minutes at room temperature. The stomach was then washed in PBS and placed in PBS containing 0.5 mM DTT and 3 mM EDTA for 1.5 hours at room temperature. The EDTA/DTT solution was then replaced with PBS and the tissue was disrupted using manual shaking. Cells in the supernatant were collected by centrifugation, resuspended in Ham's F-12 medium containing 5% FBS and penicillin/streptomycin, and plated into collagen coated 24 well plates (BD). After 24 hours in culture, non-adherent cells were removed. Cells were then washed with PBS and fresh medium was added. Primary gastric epithelial cells isolated from uninfected wild-type mice and IL-17RA-/- mice were stimulated with IL-17 (10 ng/ml) (R&D Systems, Minneapolis, MN). After 24 hours, the medium overlying epithelial cells was collected, centrifuged, and the supernatant filtered. Supernatants were analyzed for expression of GM-CSF, IL-1β, IL-6, MIP2, KC by using the Searchlight System (Pierce Biotechnology, Woburn, MA) or for LIX, KC, and MIP2 by using Milliplex technology (Millipore, Billerica, MA). These are ELISA-based systems in which capture antibodies are spotted in arrays within microplate wells, which allows multiple cytokines to be detected simultaneously using chemiluminescent methodology.
Mouse serum antibody responses to H. pylori were analyzed by enzyme-linked immunosorbent assay (ELISA). To generate an antigen preparation for use in ELISA, H. pylori strain SS1 was cultured overnight in liquid medium, and then the bacteria were collected by centrifugation at 2100g and washed with cold PBS. Subsequently, the bacterial suspension was sonicated. The sonicate was then centrifuged at 10,000g, and the soluble fraction was used for subsequent experiments. Immulon 2HB plates (Thermo Fisher Scientific, Pittsburgh, PA) were coated with 10μg/ml of the H. pylori antigen preparation overnight at 4°C. Plates were washed with 0.05% Tween 20-PBS and blocked with 5% skim milk-PBS for 1 hour at 37°C. Following incubation with mouse serum samples, plates were washed and detection antibodies were added [anti-mouse IgG conjugated to horseradish peroxidase (HRP) (Pierce, Rockford, IL), goat anti-mouse IgG1-HRP, goat anti-mouse IgG2a-HRP, or goat anti-mouse IgA-HRP (Santa Cruz Biotechnology, Santa Cruz, CA)]. Following another series of washes, H. pylori-specific antibodies were detected using 1-Step Ultra TMB ELISA substrate (Pierce). Titers are expressed as the reciprocal of the maximum dilution of sera that yielded an optical density (450 nm) three times higher than the background signal generated by sera from uninfected mice.
Four to ten mice per group per time point were used for all of the studies. To compare results obtained with different groups of mice, statistical analysis was performed using one-way ANOVA, followed by a Student-Neuman-Keuls post-hoc test. For analyses of bacterial numbers and cell numbers, the data were normalized by log transformation prior to statistical analysis. Histology scores for wild-type and IL-17RA-/- mice were compared using the Mann-Whitney U-test.
To investigate a potential role of IL-17RA signaling in H. pylori-host interactions, wild-type C57BL/6 mice and IL-17RA-/- mice were orogastrically infected with H. pylori. Mice were sacrificed at serial timepoints and bacterial colonization of the stomach was quantified as described in Methods. At 1 month post-infection, there was no significant difference in the density of H. pylori colonization in the two populations of mice (Fig. 1A). By 3 months post-infection, significantly higher numbers of H. pylori were cultured from the stomachs of IL-17RA-/- mice than from stomachs of wild-type mice, and this difference also was observed at 6 months post-challenge (Fig. 1A). Despite the higher burden of H. pylori in the stomachs of the IL-17RA-/- mice, these mice did not exhibit weight loss or other physical signs of illness during an observation period of 6 months post-infection.
The severity of gastric inflammation in H. pylori-infected IL-17RA-/- and wild-type mice was examined at 1, 3, and 6 months post-infection. Total gastric inflammation was analyzed using a scoring system that evaluates both acute inflammation and chronic inflammation in the gastric corpus and antrum of each animal. At one month post-infection, there was no significant difference in the severity of total gastric inflammation in the two populations of infected mice (Fig. 1B). At 3 and 6 months post-infection, however, the severity of total gastric inflammation was significantly greater in infected IL-17RA-/- mice compared to wild type mice (Fig. 1B). Both antral and corpus chronic (mononuclear) inflammation scores were significantly higher in infected IL-17RA-/- mice than in wild-type mice at 3 months and 6 months post-infection (Tables I and andII).II). At 3 months post-infection, plasma cell infiltration of the antrum was detected in all of the H. pylori-infected IL-17RA-/- mice, but not in the infected wild-type mice (Table I). At 6 months post-infection, gastric lymphoid follicles with germinal centers were detected in nearly all of the infected IL-17RA-/- mice, but not in the infected wild-type mice (Table II). In addition to the results shown in Tables I and andII,II, wild-type and IL-17RA-/- mice were experimentally infected with H. pylori in 3 other independent experiments, and the IL-17RA-/- mice consistently exhibited increased severity of chronic inflammation at 3 months post-infection compared to wild-type mice (data not shown). There were no significant differences in acute (polymorphonuclear) gastric inflammation scores when comparing the infected wild-type and IL-17RA-/- mice (Tables I and andII).II). At 3 and 5 months of age, uninfected IL-17RA-/- mice and uninfected wild-type mice did not exhibit detectable gastric inflammation (data not shown).
Figure 2 illustrates representative histology of the transition zone between the gastric corpus and the antrum, a region of the mouse stomach that typically exhibits relatively severe inflammation in response to H. pylori. By 3 months post-infection, infected IL-17RA-/- mice had increased gastric mononuclear cell infiltration compared to infected wild-type mice, and at 6 months post-infection, thickening of the gastric mucosa (gastric mucosal hyperplasia) was observed in infected IL-17RA-/- mice (Fig. 2). Of great interest, a focus of gastric adenocarcinoma invasive into the submucosa was detected in one of the infected IL-17RA-/- mice at 6 months post-infection (data not shown).
A hallmark of H. pylori is its ability to persistently colonize the stomach, and therefore, all of the experiments described above focused on analysis of bacterial density and inflammation in the setting of chronic infection. We also analyzed these parameters in the setting of acute infection. Wild-type mice and IL-17RA-/- mice were infected with H. pylori, and bacterial density and gastric histology were analyzed 3 days later. The density of H. pylori infection was relatively low at this early timepoint compared to later timepoints, but there was no significant difference in the density of H. pylori colonization in the two populations of mice (data not shown). Consistent with our previously published studies of acute H. pylori infection in wild-type mice (13), minimal gastric inflammation was detected at this early timepoint, and there was no significant difference in the severity of inflammation in the two groups of H. pylori-infected mice (data not shown).
The histologic analysis shown in Figure 2 demonstrated that there were significant differences in the chronic gastric mucosal inflammatory responses in H. pylori-infected IL-17RA-/- mice compared to infected wild-type mice. To investigate differences in the gastric mucosal cytokine milieu that might contribute to these differences, we analyzed the gastric expression of several proinflammatory cytokines and CXC neutrophil chemoattractant chemokines that are considered to be Th17 signature proteins [reviewed in (16)]. Real time rtPCR was performed on RNA isolated from gastric tissue of H. pylori-infected wild-type mice and IL-17RA-/- mice at 3 and 6 months post-infection. KC, LIX, and GM-CSF were expressed at significantly lower levels in the stomachs of H. pylori-infected IL-17RA-/- mice than in stomachs of infected wild-type mice (Fig. 3A and 3B), which suggests that IL-17 is an important factor driving expression of these cytokines and chemokines in the mouse stomach. At the 3 month timepoint, there was no significant difference in levels of IL1-β and IL-6, but at the 6-month time point, IL-1β and IL-6 were expressed at higher levels in H. pylori-infected IL-17RA-/- mice than in wild-type mice (Fig. 3A). There were no significant differences in the gastric transcript levels of these cytokines or chemokines in the two groups of uninfected mice (data not shown).
We next tested the hypothesis that gastric epithelial cells might be one source of the Th17-signature proinflammatory cytokines and chemokines detected in the mouse stomach. Primary gastric epithelial cells were isolated from wild-type mice and IL-17RA-/- mice, and then were stimulated with recombinant IL-17a (rIL-17a) or left unstimulated. Supernatants were collected, and concentrations of selected Th17 signature proteins were quantified by multiplex ELISA. In response to stimulation with rIL-17a, wild-type gastric epithelial cells secreted significantly higher levels of KC (CXCL1), LIX (CXCL5), MIP2 (CXCL2), and GM-CSF compared to IL-17-stimulated gastric epithelial cells derived from IL-17RA-/- mice (Fig. 4). These results are consistent with the pattern of gastric cytokine expression detected in vivo (Fig. 3).
In further studies, we specifically analyzed gastric neutrophil infiltration in H. pylori-infected mice. At 3 months post infection, significantly fewer gastric neutrophils (Gr1+, CD11b+ cells) were present in the stomachs of the IL-17RA-/- mice than in wild-type mice (Fig. 5). The failure to detect this difference in neutrophil density in an analysis of histological sections (Table I) may be attributable to the fact that neutrophils represent a minority of the gastric inflammatory cells identified by histologic assessment and are intimately admixed with mononuclear inflammatory cells; in this setting, light microscopy may be insufficiently sensitive to detect a difference in the neutrophil populations of the two groups. There was no significant difference in the number of gastric macrophages (CD11b+, Gr1-, F4/80+ cells) in H. pylori-infected IL-17RA-/- mice compared to infected wild-type mice (data not shown). As expected based on previous studies of IL-17RA-/- mice (39), there were significantly fewer neutrophils present in the spleens of H. pylori-infected IL-17RA-/- mice than wild-type mice at 1, 3 and 6 months post infection; the numbers (or percentages) of macrophages present in the spleens of H. pylori infected IL-17RA-/- mice and wild-type mice were not significantly different (data not shown).
Next we analyzed CD4+ and CD8+ T cells in the mouse stomachs. Significantly higher numbers of gastric CD4+ T cells were detected in H. pylori-infected IL-17RA-/- mice than in wild-type mice at 3 months post-infection (Fig. 6A). There was no significant difference in the number of CD8+ T cells (Fig. 6A). To define the type of T helper responses present in the stomachs of infected mice, we analyzed the expression of several signature cytokines by real time rtPCR. Th1 and Th2 cytokines (IFNγ and IL-4, respectively) were expressed at similar levels in infected IL-17RA-/- mice and infected wild-type mice (Fig. 6B). To examine cytokines known to be produced by Th17 cells, we analyzed gastric expression of IL-17a, IL-17f, IL-21, and IL-22 (Fig. 6C). IL-17a and IL-21 were expressed at significantly higher levels in the stomachs of infected IL-17RA-/- mice than in infected wild-type mice (Fig. 6C), and levels of IL-17a were significantly higher in H. pylori-infected IL-17RA-/- mice than in uninfected IL-17RA-/- mice (Fig. 6D). This suggests that a negative feedback loop might be impaired in the IL-17RA-/- mice. In contrast to the pattern observed with expression of IL-17a and IL-21, there was no significant difference in gastric expression of IL-17f in infected IL-17RA-/- mice compared to infected wild-type mice (Fig. 6C).
To further analyze the mononuclear cells observed infiltrating the mouse stomachs following H. pylori infection (Tables I and andIIII and Fig. 2), gastric B cells were enumerated using flow cytometry. At 3 months post-infection, B cells (B220+ cells) were significantly more abundant in the stomachs of H. pylori-infected IL-17RA-/- mice than in infected wild-type mice (Fig. 7A). Based on these data, in conjunction with the visualization of plasma cells and lymphoid follicles in histologic analysis (Table I and andII;II; Fig. 7D), we hypothesized that there might be enhanced production of anti-H. pylori antibodies in the H. pylori-infected IL-17RA-/- mice. As shown in Fig. 7B, the levels of H. pylori-specific serum IgG and IgA antibodies were significantly higher in the IL-17RA-/- mice compared to wild-type control mice at 3 months post-infection. We analyzed the expression of several B cell-recruiting chemokines, including two that are known to be upregulated in the stomachs of H. pylori-infected humans (45), using real time rtPCR on RNA isolated from the mouse stomachs. Expression of CCL28 and CXCL12 did not correlate with the observed recruitment of B cells in IL-17RA-/- mice (Fig. 7C). However, CXCL13, also known as B lymphocyte chemoattractant (BLC) and a mediator of B cell organization in lymphoid tissues and the gut, was expressed at significantly higher levels in H. pylori-infected IL-17RA-/- mice than in infected wild-type mice (Fig. 7C).
In this study, we investigated a potential role of IL-17 signaling in the chronic gastric inflammation that is induced by H. pylori infection. One of the findings was that IL-17 signaling through an intact IL-17RA contributes to the control of H. pylori proliferation in the stomach (Fig. 1). This observation is consistent with previous reports indicating that IL-17RA signaling is required for optimal host defense against several other pathogenic bacteria, including Klebsiella pneumoniae, Pseudomonas aeruginosa, Streptococcus pneumoniae, and Porphyromanas gingivalis (21, 26, 32, 46). In contrast to many previous studies that analyzed the role of IL-17RA signaling in acute bacterial infections, the current study analyzed the role of IL-17RA signaling in a chronic bacterial infection. The insights gained from this study are potentially relevant to understanding the role of IL-17RA signaling in other chronic inflammatory conditions.
A striking finding in the current study was that H. pylori-infected IL-17RA-/- mice exhibited increased severity of mononuclear gastric inflammation compared to infected wild-type mice. The mononuclear gastric inflammation in infected IL-17RA-/- mice was largely attributable to the recruitment of B cells, and gastric lymphoid follicles with germinal centers were detected in these mice at 6 months post-infection (Fig. 7 and Tables I and andII).II). We also detected an increased number of gastric plasma cells and higher levels of specific serum antibodies to H. pylori in infected IL-17RA-/- mice than in infected wild-type mice. The strong humoral antibody response to H. pylori elicited in IL-17RA-/- mice clearly was not sufficient to eradicate the bacteria (Fig. 1). Similarly, a large body of literature indicates that humoral immune responses typically fail to eradicate H. pylori infection in humans or in experimentally infected animals (1, 2). Potentially antibodies are ineffective in the gastric environment due to the low pH, or they may fail to facilitate opsonization due to limited entry of phagocytic cells into the gastric mucus layer. In murine models, T cell immunity rather than humoral immunity appears to be required for protection (47, 48). One previous study reported that μMT (-/-) mice (B lymphocyte deficient) harbored decreased numbers of H. pylori compared to wild-type mice, and suggested that the humoral immune response might enhance H. pylori colonization (49). Therefore, it is of interest that in the current study we also observed a correlation between strong humoral anti-H. pylori immune responses and increased H. pylori colonization of the stomach.
This is the first report of B cell-predominant chronic inflammation in IL-17RA-/- mice. Since B cell recruitment has not been detected in other studies in which IL-17RA-/- mice were infected with various bacterial pathogens, and gastric B cell infiltration was not observed in uninfected IL-17RA-/- mice, this suggests that the B cell recruitment observed in H. pylori-infected IL-17RA-/- mice is driven by the presence of H. pylori. The mechanisms underlying the striking recruitment of B cells into the gastric mucosa of H. pylori-infected IL-17RA-/- mice are not yet entirely understood, but several scenarios can be considered. One possibility is that elevated levels of a B cell chemokine such as CXCL13 contributes to the recruitment of B cells (Fig. 7). Another possibility is that B cells in wild-type and IL-17RA-/- mice might differ in responsiveness to chemotactic signals. In support of this hypothesis, a recent study reported that, in the presence of intact IL-17 signaling, B cells exhibit attenuated responsiveness to chemokines such as CXCL12 and CXCL13 (50). The increased plasma cells and germinal center formation observed in the IL-17RA-/- mice might also be attributable to increased expression of IL-21 (Fig. 6C). IL-21 has been shown to play a role in antibody production in humans and mice by inducing differentiation of B cells into plasma cells (51, 52).
In comparison to H. pylori-infected wild-type mice, H. pylori-infected IL-17RA-/- mice had significantly fewer gastric neutrophils and reduced levels of gastric CXC neutrophil chemoattractant chemokines (KC and LIX). Neutrophil deficiencies have been detected in IL-17RA-/- mice compared to wild-type mice in several other models of inflammation. For example, neutrophil recruitment to the lungs and airways was impaired in IL-17RA-/- mice in models of allergic asthma and Klebsiella pneumoniae pulmonary infection (21, 37); neutrophil recruitment to the cornea was transiently impaired in IL-17RA-/- mice in response to HSV-1 infection (35); and neutrophil infiltration was reduced in IL-17RA-/- mice compared to wild-type mice in a model of acute trinitrobenzenesulfonic acid (TNBS)- induced colitis (36). The reduced or delayed neutrophil recruitment observed in IL-17RA-/- mice has been attributed to impaired IL-17 signaling and a consequent reduction in the production of neutrophil-chemoattractant CXC chemokines.
Recently it was reported that H. pylori-infected IL-17-/- mice had lower levels of gastric neutrophil infiltration compared to what was observed in infected wild-type mice (53). Thus, both IL-17RA-/- mice and IL-17-/- mice have impaired neutrophil responses to H. pylori compared to the responses observed in wild-type mice. Several previous studies have demonstrated that neutrophils contribute to the control of H. pylori proliferation (54-56), and therefore, a paucity of gastric neutrophils in infected IL-17RA-/- mice may account for the observed increase in bacterial colonization density in these mice. Interestingly, levels of H. pylori colonization were reported to be significantly lower in the IL-17-/- mice compared to wild-type mice (53), a finding that contrasts the increased levels of H. pylori colonization detected in IL-17RA-/- mice in the current study. The reasons for this difference are not clear at present.
As shown in Fig. 6, IL-17a was expressed at very high levels in the stomachs of infected IL-17RA-/- mice compared to infected wild-type mice. The increase in IL-17a levels in infected IL-17RA-/- mice was dependent on the presence of H. pylori, since gastric IL-17a levels were markedly higher in infected IL-17RA-/- mice than in uninfected IL-17RA-/- mice (Fig. 6D). One explanation for these findings may be the loss of a negative feedback loop; for example, in wild-type mice, IL-17a may signal through IL-17RA to downregulate IL-17a expression. Support for this hypothesis comes from a recent study that reported higher levels of IL-17a production by T cell populations from IL-17RA-/- mice than by T cell populations from wild-type mice, when the T cell populations were cultured under conditions known to induce Th17 cell differentiation (57). An additional factor accounting for the high IL-17a levels may be that increased levels of IL-21 expression (Fig. 6C) drive Th17 cell differentation and IL-17a expression (58, 59).
In summary, this study demonstrates an important role of IL-17RA signaling in regulating H. pylori-induced gastritis, a model of inflammation induced by a chronic bacterial infection. The results of this current study highlight the importance of IL-17RA signaling in regulating B cell migration and function. In contrast to many chronic inflammatory diseases in which IL-17RA signaling promotes an inflammatory response, IL-17RA signaling downregulates the chronic mononuclear inflammation elicited by H. pylori infection. Such downregulation may be an important factor that allows H. pylori to persistently colonize the human gastric environment.
We thank Amgen for providing IL-17RA-/- mice, and we thank Keith Wilson and Mark Boothby for helpful discussions.
1This work was supported by the Medical Research Service of the Department of Veterans Affairs and NIH grants R01AI39657, R01AI068009, and ROI DK58587. The VMC Flow Cytometry Shared Resource is supported by the Vanderbilt Ingram Cancer Center (P30 CA68485) and the Vanderbilt Digestive Disease Research Center (P30 DK058404). Research Histology Core Services were supported by the Vanderbilt Digestive Disease Research Center.