You are here

Article Text

Article menu

Download PDFPDF

Helicobacter pylori
Helicobacter pylori-induced peptic ulcer disease is associated with inadequate regulatory T cell responses
  1. K Robinson1,2,
  2. R Kenefeck1,2,
  3. E L Pidgeon1,
  4. S Shakib1,
  5. S Patel1,2,
  6. R J Polson1,
  7. A M Zaitoun2,3,
  8. J C Atherton1,2
  1. 1
    Institute of Infection, Immunity & Inflammation, University of Nottingham, Centre for Biomolecular Sciences, Nottingham, UK
  2. 2
    Wolfson Digestive Diseases Centre, University of Nottingham, Queen’s Medical Centre, Nottingham, UK
  3. 3
    Department of Histopathology, University of Nottingham, Queen’s Medical Centre, Nottingham, UK
  1. Dr K Robinson, Centre for Biomolecular Sciences, University of Nottingham, Nottingham, NG7 2RD, UK; karen.robinson{at}nottingham.ac.uk

Abstract

Background and aims: Helicobacter pylori infection is the major cause of peptic ulceration and gastric adenocarcinoma. To address the hypothesis that the human acquired immune response to H pylori influences pathogenesis, we characterised the gastric T helper (Th) and regulatory T cell (Treg) response of infected patients.

Methods: The human gastric CD4+ T cell response of 28 donors who were infected with H pylori and 44 who were not infected was analysed using flow cytometry. The T cell associated mucosal cytokine response was analysed by real-time polymerase chain reaction assay of samples from 38 infected and 22 uninfected donors. Recombinant interleukin 10 (IL10) was added to co-cultures of H pylori and AGS cells and its suppressive effects upon inflammatory responses were measured.

Results: We found that the H pylori-specific response consists of both T helper 1 and 2 subsets with high levels of IL10-secreting Tregs. People with peptic ulcer disease had a 2.4-fold reduced CD4+CD25hiIL10+ Treg response (p = 0.05) but increased Th1 and Th2 responses (Th1: 3.2-fold, p = 0.038; Th2: 6.1-fold, p = 0.029) compared to those without ulcers. In vitro studies showed that IL10 inhibited IL8 expression and activation of nuclear factor kappa B induced by H pylori in gastric epithelial cells, and enhanced H pylori growth in a bacterial-cell co-culture model.

Conclusions: Together our data suggest that H pylori induces a regulatory T cell response, possibly contributing to its peaceful coexistence with the human host, and that ulcers occur when this regulatory response is inadequate.

https://doi.org/10.1136/gut.2007.137539

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Helicobacter pylori colonises the stomachs of approximately half the world’s population and is associated with peptic ulceration and gastric malignancy.1 In mice, protection and mucosal damage appear to be mediated by T helper 1 (Th1) cellular responses.2–4 Most infected humans develop asymptomatic gastritis but some develop severe inflammation and peptic ulceration.5 Regulatory T cells (Tregs) are thought to be important in suppressing deleterious immune and inflammatory responses.6–8 In H pylori-infected mice, Tregs suppress inflammation and promote persistent colonisation.9 10 Increased numbers of Tregs have been reported in the H pylori-infected human gastric mucosa.10 11 Most Treg cell types are CD4+ and express high levels of CD25. FOXP3 is expressed at a high level in CD25hiCD4+ Tregs,12 which may act by secreting suppressive cytokines, eg, IL1013 14 and transforming growth factor-β (TGF-β).15

    For many years it has been thought that H pylori is able to persist life-long in the gastric mucosa because of a downregulation of the bacteria-induced inflammatory response.1 2 Th1 and Treg responses have been demonstrated in the H pylori-infected human gastric mucosa using reverse transcriptase polymerase chain reaction (RT-PCR) and immunohistochemistry.11 16 Flow cytometry studies using mitogens to re-stimulate cytokine responses from gastric lymphocytes in vitro found predominantly Th1 cells,17 but similar experiments also found Th1 cells present in normal uninfected gastric tissue, indicating that gastric T cells are not all H pylori-specific.18 Several studies have found an epidemiological association between H pylori infection and protection from allergy or asthma.19–21 Recent data showed that a stronger protective effect against asthma was associated with H pylori strains expressing the virulence determinant CagA.22 Such inhibition of allergy and asthma is likely to be mediated by Tregs,23 raising the hypothesis that infection with CagA+ strains may elicit higher-level Treg responses. In view of this we aimed to stimulate cells with H pylori antigens for characterisation and quantification of the human T helper and Treg response, to compare the magnitude of the Treg response in patients with cagA+ and cagA− H pylori infections, and to investigate the effects of regulatory cytokines on gastric epithelial cell inflammatory responses. We found that Th1, Th2 and Treg cells were present in the infected gastric mucosa; the Treg cells were predominantly IL10+. Low frequencies of Tregs in the mucosa correlated with the occurrence of peptic ulcer disease. Although not significant, a higher level IL10+ Treg response was observed in individuals infected with cagA+ strains. In vitro co-culture experiments of H pylori with a human gastric epithelial cell line, AGS, showed that IL10 can inhibit IL8 expression, suppress activation of the transcription factor nuclear factor kappa B (NF-κB), and increase bacterial densities. Where human gastric tissues contained high levels of bacteria, high IL10 mRNA expression levels were found, further indicating a close relationship between the host response and outcome of infection.

    MATERIALS AND METHODS

    All reagents were obtained from Sigma–Aldrich (Poole, UK) unless stated otherwise.

    Volunteers and clinical materials

    Samples were donated by 82 patients (age range 19–84 years) undergoing upper gastro-intestinal endoscopy at the Queen’s Medical Centre, Nottingham. Patients underwent the procedure for a variety of indications, most commonly dyspepsia, but were otherwise healthy. Thirty-eight donors were H pylori positive (12 duodenal ulcer, five gastric ulcer, 21 no ulceration) and 44 were uninfected. The male to female ratios of these groups were 0.59 and 0.69, respectively; mean ages were 56.1 and 56.7 years. Mean ages of patients with duodenal ulcer, gastric ulcer and no ulceration were 58.1, 65.2 and 60.1 years, respectively. None of the patients were regularly taking non-steroidal anti-inflammatory drugs or had taken antibiotics or proton pump inhibitor drugs in the preceding 4 weeks.

    A series of biopsies was collected from the gastric antrum. One was placed directly into CLO test medium for determination of H pylori infection status by urease detection; the second was collected into iso-sensitest broth (Oxoid, Cambridge, UK)/10% glycerol for isolation and culture of H pylori; a third was placed in formalin for subsequent histopathological scoring; a fourth was placed in RNAlater solution and preserved for mRNA analysis. Four additional biopsies were placed into culture medium (RPMI1640/10% fetal calf serum (FCS)/100 U/ml penicillin G/100 μg/ml streptomycin sulfate/250 ng/ml amphotericin B) for fluorescence-activated cell sorting (FACS) analysis. Complete sets of eight biopsies were not available for all subjects. Biopsies for FACS analysis were collected from 28 infected and all 44 uninfected donors, whereas mRNA was available from the 38 infected donors but only from 22 of the uninfected donors. H pylori infection status was confirmed using the CLO test, culture and histology. Clinical endoscopy observations and disease severity were recorded. DNA was purified from H pylori isolates and PCR genotyping was carried out to ascertain cagA and vacA status:24 25 25/38 (65.8%) were cagA+ and 22/38 (57.9%) were vacA s1/m1+. We found 12/17 (70.6%) cagA+ and 9/17 (52.9%) vacA s1/m1+ strains in patients with PUD, compared with 15/21 (71.4%) cagA+ and 11/21 (52.4%) vacA s1/m1+ from patients without PUD.

    Cell isolation, culture, intracellular staining and flow cytometry

    Four gastric biopsies collected into culture medium (as described above) were rubbed through a sterile disposable 100 μm cell strainer (BD Biosciences, Oxford, UK), washed and resuspended in culture medium (as above) at 1×106 per ml. Typical recovery was 0.5×106 to 3×106 cells.

    H pylori whole-cell sonicated antigen from the SS1 strain26 was added to 5×105 cells to a final concentration of 25 μg/ml, as described previously.27 For positive control mitogenic stimulation, 20 ng/ml phorbol myristate acetate (PMA) and 1 μmol/l ionomycin were added. Negative control tubes received no antigens or mitogens. Brefeldin A (10 μg/ml) was added before incubation for 15 h. Intracellular staining was carried out as previously described27 using two combinations of monoclonal antibodies (mAbs): anti-CD45–phycoerythrin–cyanin 5.1 (PC5), anti-CD4–PE–Texas red (ECD), anti-IL4–PE and anti-interferon γ–fluorescein isothiocyanate (IFNγ–FITC; Beckman Coulter, High Wycombe, UK) or anti-CD45–PC5, anti-CD4–ECD, anti-CD25–FITC (Beckman Coulter) and anti-IL10–PE (Serotec, Oxford, UK). Data on 100 000 events were acquired using a Coulter EPICS Altra flow cytometer. Isotype controls were used to ensure that staining was specific. Analysis was performed using WinMDI version 2.8, based upon the percentage of leukocytes that were gated by CD45 staining. Example plots are shown in fig 1.

    Figure 1
    Figure 1 Flow cytometry dot plots of gastric biopsy cells from one Helicobacter pylori-positive donor without peptic ulcer disease (PUD). Cells were stained for CD45, CD4, interferon γ (IFNγ), interleukin 4 (IL4), IL10 and CD25 using monoclonal antibodies conjugated to the fluorochromes anti-CD45–phycoerythrin–cyanin 5.1 (PC5), anti-CD4–phycoerythrin–Texas red (ECD), fluorescein isothiocyanate (FITC) and phycoerythrin (PE). The intensity of staining is shown on a log10 scale, with side scatter (SS) on a linear scale, and the number of double-stained events is given in the right upper quadrant. From 100 000 events, 46 349 were CD45+ (in region 1) (A). Of these, there were 0.319% IFNγ+CD4+ (B), 0.483% IL4+CD4+ (C), 1.667% IL10+CD4+ (D), and 0.742% CD4+CD25hi (E). CD4+CD25hi cells were identified as those CD4+ cells with brighter CD25 staining than that of CD4− cells (E).

    Real-time polymerase chain reaction

    Total RNA was purified from gastric biopsies using an SV Total RNA Isolation System (Promega, Southampton, UK). cDNA was synthesised using SuperScript II reverse transcriptase (Invitrogen, Paisley, UK) with an oligo(dT) primer. RT-PCR was performed on a Rotor-Gene 3000 (Corbett Research, Cambridge, UK) using a DyNAmo HS SYBR green qPCR kit (GRI, Braintree, UK). Amplification was carried out over 45 cycles of 15 s at 95°C, 30 s at 61°C and 30 s at 72°C. No template controls were included in each run, and a melting curve analysis was performed. First stage RT-PCR samples, produced without reverse transcriptase were tested in parallel. The results were analysed according to the Pfaffl method.28 Relative gene expression levels were determined by normalising against GAPDH, comparing to a reference sample from an H pylori-negative donor to obtain a fold difference. The donor for this reference sample was selected at random. A commercial, pooled, human cDNA standard (BD Biosciences) was included as a positive control in all assays. Commercial primers were used for human IFNγ (SuperArray RT2, Tebu-Bio, Peterborough, UK), IL8 and IL10 (BioSource UK). Other PCR primer sequences are shown in table 1.

    View this table:
    Table 1 Primer sequences for real-time polymerase chain reaction analysis

    Effect of IL10 on IL8 release from gastric epithelial cells in response to H pylori

    AGS cells were grown in 24-well plates (Nunc, Loughborough, UK) in F12 Ham nutrient mixture (Invitrogen) supplemented with 10% FCS and 2 mmol/l l-glutamine in 5% CO2 at 37°C.25 Recombinant human IL10 (Serotec) was added to 0.5–100 ng/ml, based from a previous study demonstrating suppressive effects of this cytokine on neutrophils.30 Suspensions of the H pylori 60190 strain (cagPaI-positive, vacA s1/m1), shown previously to elicit an IL8 response from AGS cells in vitro, were added to the appropriate wells in quadruplicate at a multiplicity (MOI) of 100 bacteria/AGS cell and to give a total well volume of 0.5 ml before incubation for 7–24 h.31 IL8 was measured in culture supernatants using a PeliKine Compact human IL8 ELISA kit (Mast Group, Bootle, UK) as per the manufacturer’s instructions. The limit of sensitivity of the assay was 0.56 pg/ml.

    Effect of IL10 on H pylori-induced NF-κB activation in gastric epithelial cells

    AGS cells in 75 cm2 tissue culture flasks were exposed to the H pylori 60190 strain (MOI = 100) or medium, with and without 20 ng/ml rhIL10, for 1 h at 37°C. As a positive control, cells were incubated with 50 ng/ml recombinant human tumour necrosis factor-α (TNF-α; Autogen Bioclear UK, Calne, UK). The cells were scraped from the flasks, nuclear extracts were prepared, and equalised protein concentrations were tested in duplicate using an NF-κB/p65 ActivELISA kit (BioCarta Europe, Hamburg, Germany). Concentrations of p65 (ng/ml) were determined from a standard curve and the ratio of test to untreated controls samples was calculated.

    Effect of IL10 upon bacterial densities in co-cultures of H pylori with gastric epithelial cells

    AGS cells in 24-well plates were inoculated with H pylori as described above, cultured with and without 20 ng/ml rhIL10. Supernatants from six replicate wells were collected after incubation for 0, 2, 6 and 24 h. A 10-fold serial dilution series was prepared from each well in sterile phosphate-buffered saline (PBS) before applying 10 μl aliquots to blood agar plates. Bacterial densities were determined by colony counting after 4 days incubation at 37°C in 5% CO2.

    Histopathological grading of H pylori density

    Biopsy specimens were fixed in buffered formalin, embedded in paraffin, cut to 4 μm thickness, and stained with toluidine blue. The grading of H pylori density was carried out independently by a histopathologist, where 0  =  no H pylori, 1  =  mild, 2  =  moderate and 3  =  high density.

    Statistical analysis

    Statistical analyses and box-and-whisker plots were prepared using MINITAB Release 14 software. Boxes represent the first to third quartiles, the median values are shown as horizontal lines within the bars, and the whiskers depict the lowest and highest observations within 1.5 times the first to third interquartile range. Statistical tests of paired sets of data were carried out using the Wilcoxon signed rank test. For unpaired data, the Mann–Whitney U test was used. Correlations were examined using the Spearman’s rank correlation test. A significant difference was taken at p⩽0.05.

    RESULTS

    Gastric T helper cell response

    We incubated gastric biopsy cell suspensions with a bacterial lysate antigen and analysed the cytokine profiles of CD4+CD45+ cells using flow cytometry. Biopsies from H pylori-infected patients contained higher frequencies of CD45+ cells compared to uninfected tissues (p = 0.0002) (median total CD45+ counts per tissue sample: 9.14×104 and 3.65×104; CD45+ CD4+ counts: 1.71×104 and 4.75×103 from infected and uninfected samples, respectively). H pylori stimulation of cells from infected donors resulted in increased CD4+IFNγ+ (median increase 4-fold, p = 0.044), CD4+IL4+ (55-fold, p = 0.002) and CD4+IL10+ (4-fold, p = 0.033) events compared to unstimulated cells (fig 2A). No significant differences were observed when cells from uninfected donors were used (fig 2B and table 2), indicating that the responses were a result of antigen-specific re-stimulation of memory T cells. H pylori-stimulated cells from infected donors produced a mixed Th1/Th2 cytokine response, with a predominant CD4+IL10+ response.

    Figure 2
    Figure 2 Characterisation of gastric T helper subset responses to Helicobacter pylori. Frequencies of cytokine+ CD4 events among H pylori antigen-stimulated and unstimulated CD45+ cells from 28 infected (A) and 44 uninfected (B) patients. *Significant difference between stimulated and unstimulated cells (p<0.05). Comparison of IFNγ, IL4 and IL10 mRNA levels in tissue from 38 infected and 22 uninfected patients (C). *Significant difference between infected and uninfected tissue (p<0.05). Comparison of IFNγ, IL4 and IL10 mRNA levels in tissue from patients infected with cagA+ (n = 25) and cagA− (n = 13) H pylori strains (D). IFN, interferon; IL, interleukin.
    View this table:
    Table 2 The proportions of IFNγ-, IL4- and IL10-positive CD4 events after culture and stimulation of gastric leukocytes with Helicobacter pylori antigen or mitogen

    RT-PCR was carried out on biopsy tissues to confirm increased expression of these cytokines in the gastric mucosa. In infected compared to uninfected tissues, there were significantly increased levels of IFNγ (119-fold, p = 0.001), IL4 (231-fold, p = 0.001), and IL10 (6975-fold, p = 0.002) mRNA (fig 2C), confirming the predominant IL10 response. To test for associations with H pylori virulence factors, the data were re-stratified according to cagA and vacA status. The levels of IFNγ mRNA were approximately 4-fold lower in tissues infected with cagA+ strains compared to cagA− samples, but this was not statistically significant (p = 0.08). The levels of IL10 mRNA were approximately 5-fold higher in tissues infected with cagA+ strains compared to cagA− samples, but again this was not statistically significant (p = 0.06) (fig 2D). No differences according to the vacA type of the colonising strain were found.

    Because our results differed from previous studies where gastric T cells were stimulated with PMA and ionomycin,17 we compared the IFNγ, IL4 and IL10 responses to antigen and mitogen stimulation (table 2). PMA/ionomycin treatment stimulated cytokine production from cells of infected and uninfected donors, whereas only cells from infected donors responded to antigen stimulation. PMA/ionomycin induced an increased frequency of CD4+IFNγ+ events compared to antigen stimulation (p = 0.030), but no significant difference in the frequency of IL4+ or IL10+ cells was observed. The Th response to PMA did not reflect the antigen-specific response, indicating that the mitogen stimulated additional H pylori-unrelated Th1 cells.

    Regulatory T cell responses

    IL10 may be produced by Tregs in the infected gastric mucosa.11 Therefore we quantified CD4+CD25hi cells (median total counts per tissue: 5460 (infected) and 1006 (uninfected)) and CD4+CD25hiIL10+ cells (median total counts: 720 (infected) and 219 (uninfected)). We found 7-fold (p = 0.023) and 12-fold (p = 0.001), respectively, higher percentages of these cells in tissue from infected patients (fig 3A,B). Up to 95% of the CD4+CD25hi events from infected donors were IL10+, 36-fold greater than in cells from uninfected donors (p = 0.002) (fig 3C). Trends indicated slightly increased frequencies of Tregs in cagA+ compared to cagA− samples but this was not statistically significant (fig 3A–C). In the samples from infected donors 48.55% of the IL10+CD4+ events were CD25hi, whereas those from uninfected donors contained only 8.05% (p = 0.007), indicating the presence of other IL10-secreting CD4+ T cells in the gastric mucosa.

    Figure 3
    Figure 3 Regulatory T cells in the gastric mucosa of patients who were infected with H pylori and those who were uninfected. Whole gastric biopsy cell suspensions from 18 infected (11 cagA+) and 23 uninfected donors were cultured with H pylori sonicate for 16 h and the frequencies of CD4+CD25hi (A) and CD4+CD25hiIL10+ events (B) were determined. (C) Percentage of IL10+ events amongst CD4+CD25hi gastric CD45+ events. *Significant difference from uninfected (p<0.05). IL, interleukin.

    RT-PCR was used to confirm that Tregs were upregulated in infected gastric tissue (fig 4A). The median FOXP3 mRNA level was 37-fold higher in gastric tissue from infected patients, compared to uninfected donors (p = 0.0001). TGF-β1 mRNA levels were also significantly elevated (p = 0.03). We were interested to discover whether there were associations of FOXP3 and TGF-β1 expression with H pylori virulence factors. The frequencies of cagA+ and vacA s1/m1+ strains amongst individuals with and without PUD, however, were virtually identical making this unlikely. Indeed, when the RT-PCR data were re-stratified according to cagA status or vacA type of the colonising strain no significant differences in FOXP3 and TGF-β1 were found (not shown). There were no associations of IL10, TGF-β1 or FOXP3 expression with age or gender.

    Figure 4
    Figure 4 Real-time polymerase chain reaction assays for IL10, FOXP3 and TGF-β1 mRNA in gastric biopsy tissue. Data were normalised relative to GAPDH expression, and a sample from an uninfected donor was used as an internal comparator to provide a fold-difference. (A) Gastric biopsies from 38 infected and 22 uninfected donors. (B) Biopsies from infected donors, 17 with peptic ulcer disease (PUD) and 21 normal. *Significant difference (p⩽0.03). IL, interleukin; FOXP, forkhead box P3; GADPH, glyceraldehyde-3-phosphate dehydrogenase; TGF, transforming growth factor.

    Correlation of regulatory response with disease severity

    To test whether low-level Treg responses correlated with the development of PUD, RT-PCR was performed on biopsy specimens from infected donors, 17 with and 21 without PUD (defined as the presence of duodenal or gastric ulceration or erosions). A 5-fold reduction of FOXP3 mRNA was found in PUD samples compared to those without PUD (p = 0.005) (fig 4B). IL10 mRNA was reduced by 22-fold (p = 0.03) amongst the PUD samples, but there was no significant difference in the expression of TGF-β1. FOXP3 mRNA levels in H pylori-infected biopsies correlated with IL10 (rs = 0.707, p = 0.033), but there was no correlation between FOXP3 and TGF-β1 expression, indicating that a low IL10+ Treg response was associated with the incidence of ulceration. There were no associations of IL10, TGF-β1 or FOXP3 expression with gastric versus duodenal ulceration. cagA+ strains are usually found at higher frequencies in patients with PUD compared to those without PUD,32 indicating a possibility that some of the peptic ulcers in our study could have been unrelated to H pylori. Patients who were regularly taking non-steroidal anti-inflammatory drugs (a second major cause of peptic ulceration) were excluded from the study, but we cannot rule this out. We therefore re-examined the data excluding samples from those infected with cagA− strains. The analysis of FOXP3 mRNA between donors with (n = 12) and without PUD (n = 15) strengthened to 9-fold (p = 0.012).

    To confirm RT-PCR findings, Tregs were compared in gastric biopsies from 25 infected donors (13 with PUD, 12 without) by flow cytometry. Lower frequencies of CD4+CD25hi cells (4-fold; p = 0.046) and CD4+CD25hiIL10+ cells (2.5-fold; p = 0.02) were detected in the PUD samples (fig 5A,B). Reduced levels of IL10-secreting Tregs in the gastric mucosa were therefore associated with occurrence of PUD.

    Figure 5
    Figure 5 Association of peptic ulcer disease with low frequencies of regulatory T cells in the gastric mucosa of patients infected with Helicobacter pylori. Gastric biopsy suspensions from 21 infected donors (13 with peptic ulcer disease (PUD) and 12 normal) were cultured with H pylori sonicate. Frequencies of CD4+CD25hi (A) and CD4+CD25hiIL10+ events (B) were compared. *Significant difference (p<0.05). IL, interleukin.

    In support of these findings, we found inverse trends in the gastric Th responses of 21 infected patients for which we had sufficient cells to compile a full set of Th and Treg cellular data during the study (table 3). The frequencies of Th1 and Th2 cells in the gastric biopsies of patients without PUD were significantly lower than in those from patients with PUD (p = 0.038 and p = 0.029). Biopsies from the patients without PUD contained higher frequencies of IL10+ Tregs (p = 0.05). These results indicate that in the patients without PUD the balance of the response is shifted towards elevated Tregs and a reduced T helper response in the gastric mucosa.

    View this table:
    Table 3 Frequencies of regulatory T cells, (Tregs), T helper 1 and T helper 2 (Th1 and Th2) cells in the gastric mucosa of patients infected with Helicobacter pylori, and with or without peptic ulcer disease

    Cellular effects of elevated IL10 secretion

    Effect of IL10 on the IL8 response of gastric epithelial cells to H pylori in vitro and in vivo

    Since we found a prominent mucosal IL10-secreting T cell response, we investigated its suppressive effects upon H pylori-induced expression of the pro-inflammatory chemokine IL8 by gastric epithelial cells. A range of concentrations of recombinant human IL10 (from 0.5 to 100 ng/ml) was added to co-cultures of H pylori strain 60190 with AGS cells and IL8 was assayed after 24 h. A dose-dependent suppression of IL8 secretion was demonstrated (fig 6A) (p<0.05 in all cases). No effects were observed when 10 ng/ml TGF-β1 was added (data not shown). To test whether high IL10 responses are associated with reduced IL8 expression in vivo we re-assayed gastric biopsy mRNA samples for IL8 by RT-PCR. IL8 mRNA levels in the infected biopsies were 17-fold higher than uninfected biopsies (p = 0.0001). A weak inverse correlation was found between IL10 and IL8 mRNA levels but this was not statistically significant (rs = −0.490, p = 0.089).

    Figure 6
    Figure 6 Interleukin 10 (IL10)-mediated effects upon human gastric epithelial cells. AGS cells were co-cultured with Helicobacter pylori (Hp) strain 60190, and a range of concentrations of recombinant human IL10. (A) After 24 h, IL8 in culture supernatants from six experiments was assayed by enzyme-linked immunosorbant assay (ELISA). (B) Nuclear factor kappa B (NF-κB) p65 in AGS cell nuclear extracts was measured after 1 h of incubation with H pylori or medium (with or without 20 ng/ml IL10), or TNF-α (50 ng/ml) as a positive control. The ratio of NF-κB p65 concentration for each sample compared to untreated control preparations was calculated in three separate experiments and these are displayed together on one graph. (C) After 0, 2, 6 and 24 h, the number of colony-forming units/ml (cfu/ml) in six cultures with (solid bars) and without (open bars) 20 ng/ml IL10 was determined. Bars and error bars represent the mean and standard deviation. Significant difference compared with cells cultured in the absence of IL10 (*p<0.05; **p⩽0.002). (D) The density of H pylori on gastric biopsies from 38 infected donors was assessed by histological grading (1 = mild, 2 = moderate, 3 = high) and compared to IL10 mRNA levels.

    Effect of IL10 on H pylori-induced NF-κB activation in gastric epithelial cells in vitro

    Since H pylori-induced IL8 expression involves the transcription factor NF-κB,33 and IL10 can inhibit NF-κB activation,34 we hypothesised that this was the mechanism of IL8 suppression. AGS cells were treated for 1 h with either medium or 60190 strain bacteria in the presence or absence of 20 ng/ml rhIL10, before assaying for translocated nuclear NF-κB p65. Three replicate experiments were performed and the ratio of NF-κB p65 concentration in test samples to untreated controls was calculated and compared (fig 6B). Relative to untreated controls, treatment with 50 ng/ml TNF-α (p = 0.032) or H pylori (p = 0.045) induced significantly elevated nuclear NF-κB. The response to H pylori plus IL10, however, was significantly lower than observed with bacteria alone (p = 0.045), indicating that IL10 indeed inhibits H pylori-induced NF-κB activation.

    Effect of IL10 upon bacterial densities in co-cultures of H pylori with gastric epithelial cells

    Helicobacter colonisation densities are greater in mice with higher-level IL10 responses.35 We therefore hypothesised that a high human IL10 response would cause higher bacterial loads, and determined whether bacterial densities were influenced by 20 ng/ml rhIL10 in co-cultures of H pylori with AGS cells in vitro. Viable count assays revealed that IL10 induced a significantly increased bacterial density (p<0.002) (fig 6C), but TGF-β1 or TNF-α had no significant effects. From cultures of H pylori alone, it was determined that rIL10 had no direct effect upon bacterial growth, therefore IL10 may contribute to the persistence of H pylori infection by inducing bacterial growth-promoting effects upon epithelial cells.

    Association of IL10 responses with bacterial densities in vivo

    To test whether high IL10 responses are associated with higher colonisation levels in vivo, the density of H pylori on gastric biopsies from 38 infected donors was assessed by histological grading of tissue sections. The samples with the highest H pylori density grading (3 on a scale of 0–3) contained 6-fold higher levels of IL10 mRNA (n = 17; median 9.91; interquartile range (IQR) 4.23–67.46) than the samples with a grade of 1 (n = 10; median 1.62; IQR 0.00–15.89; p = 0.046). There was no significant difference between samples of grade 3 and grade of 2 (n = 11; median 8.96; IQR 0.00–11.53) (fig 6D). Samples from donors without PUD had significantly higher H pylori grades (p = 0.020), with 40.7% being graded 3. Only 5.9% of the PUD samples had a score of 3, confirming our hypothesis that an elevated Treg response is associated with a lack of peptic ulceration and higher bacterial densities.

    DISCUSSION

    Using flow cytometry and RT-PCR we found mixed H pylori-specific T helper subsets with a predominant CD4+IL10+ response in the human gastric antrum, whereas most previous investigations have reported a predominant Th1 response.17 36 Different methodologies can influence the results of T helper assays; eg, comparisons of IFNγ and IL4 concentrations do not account for the different potencies of these cytokines or their cellular sources. More accurate methods compare the frequencies of cytokine-expressing CD4+ cells.37 Flow cytometry studies of isolated and mitogen-stimulated gastric T cells reported a Th1-dominant response during infection;17 however, PMA downregulates CD4 expression,38 making it difficult to identify T helper cells. Th1 cells in the uninfected gastric mucosa can also be stimulated with PMA to produce IFNγ.18 Our data showed a highly Th1-skewed response following PMA stimulation and we argue that antigenic stimulation provides a more accurate analysis. Quiding-Järbrink et al.39 previously attempted to analyse antigen-stimulated gastric T cells but were unsuccessful, possibly because they cultured the cells for a shorter period. Orsini et al40 showed IL4 mRNA in the infected human gastric epithelium, supporting our finding that IL4 takes part in the mucosal immune response to H pylori, and work with human T cell clones showed that antigen-specific Th1 and Th0 cells are induced by H pylori infection.41

    Our assays used a lysate extract of the SS1 H pylori strain for antigenic re-stimulation of T cells as we had successfully used this in previous studies.27 This has an inactive cag pathogenicity island and cannot deliver CagA into cells, but this was unnecessary for antigenic T cell stimulation. We confirmed that CagA was present in our antigen preparation using western blotting (data not shown). A better option would have been to stimulate cells with a lysate from the donor’s own homologous strain, but since the T cells had to be isolated from biopsies immediately this was impossible. We have assayed the responses of previously frozen peripheral blood T cells to lysates prepared from the homologous colonising strain and also a variety of heterologous cagPaI+ and cagPaI− H pylori isolates. We observed only mimimal differences in the T helper cell cytokine responses and the trends did not change (not shown).

    Studies with mice suggest that a Th1 response is associated with gastric inflammation and is protective,4 42 whereas Th1-impaired transgenic mice have less gastric pathology, cannot clear their infections and have higher colonisation densities.43 Th2 and IL10+CD4+ Treg cells in the human gastric mucosa may inhibit protective mechanisms and permit the infection to persist. As shown previously,10 11 patients infected with H pylori had significantly elevated gastric CD4+CD25hi cells and FOXP3 mRNA. A high proportion of the CD4+CD25hi cells secreted IL10, indicating that it may be an important mechanism for Treg activity in the infected stomach. This is supported by reports that H hepaticus, which causes colitis in mice, elicits an IL10+ Treg response.44 Elevated levels of TGF-β1 mRNA were present in the infected gastric mucosa, but this was not correlated with disease severity. Contact-dependent Treg mechanisms were explored in preliminary experiments by measurement of CTLA-4 expression on CD4+CD25hi cells but no differences were found. Around half of the IL10-secreting CD4+ cells detected in the gastric mucosa were not CD25hi, and these cells could also play an important role in modulating inflammation. It has recently been shown that Th1 cells may secrete IL10 under some circumstances,45 and that Th17 cells (a recently described subset of CD4 cells associated with chronic inflammation46) may exert regulatory activity by secreting IL10.47 Further studies will be necessary to identify the non-CD25hi IL10+ cells in H pylori-infected gastric mucosal tissue.

    CD4+CD25+ cells modulate H pylori-induced gastric inflammation in mice,10 48 and suppress human T cell proliferative responses in vitro.49 Our data show that H pylori-induced disease is associated with low-level Treg responses and significantly higher frequencies of gastric Th1 and Th2 cells. We found a stronger association of PUD with low Treg response than that between virulence factor expression and disease. A possible explanation is that virulence factors independently drive inflammation and ulceration, or that the factors contribute to the Treg response. We hypothesise that the Treg response to infection determines the immune and inflammatory response, hence the likelihood of PUD, and may explain why severe disease outcomes arise in some cases.

    Several epidemiological studies have shown that H pylori infection is associated with protection from atopy and autoimmune diseases,19 20 22 50 51 implying a role for H pylori in the hygiene hypothesis.52 Similar protective effects are induced by other chronic infections such as intestinal worms, and mediated by Treg cells.53 Chen and Blaser22 found a stronger protective association between asthma and infection with cagA+ strains of H pylori, but no mechanism has yet been elucidated. We therefore hypothesised that higher-level Treg responses would be observed in patients with cagA+ infections. Higher levels of IL10 mRNA were found in cagA+ compared to cagA− tissue samples, and elevated frequencies of CD4+CD25hi cells were detected indicating that this was the case, but unfortunately these differences did not achieve statistical significance. We were surprised to find such similar proportions of cagA+ strains from patients with and without PUD, although this has been reported by others previously.54 55 Further studies with larger numbers of samples will be required to fully investigate this question. It seems counter-intuitive that cagA+ infections are associated with both increased Tregs and ulceration. Possibly there is a different balance of Treg sub-populations in the mucosa infected with cagA+ and cagA−; these could have different modes and levels of suppressive activity. Further studies will be necessary to characterise gastric Tregs in more detail.

    IL8 is known to play a major role in the human response to H pylori,1 33 therefore we investigated the suppression of H pylori-induced AGS cell IL8 expression by IL10. The most dramatic effects were observed with 20–100 ng/ml IL10, but a significant inhibition was also obtained with 0.5 ng/ml. In support of the in vitro experiments, we found a weak inverse relationship between IL10 and IL8 mRNA levels in gastric biopsies. The H pylori type IV secretion apparatus encoded by the cag pathogenicity island induces IL8 expression via activation of NF-κB.56 Assays confirmed that addition of IL10 to co-cultures of H pylori and AGS cells inhibited nuclear translocation of NF-κB, which is consistent with literature on suppressive mechanisms mediated by IL10.34 IL10 inhibition of NF-κB activation could therefore be a major mechanism for Treg-mediated activity in vivo, but future experiments will investigate its effect on other signalling pathways.

    IL10 influenced the growth of H pylori in co-culture with AGS cells, via indirect effects on the epithelial cells. This could be by inhibiting the expression of anti-bacterial effectors, such as defensins, or by inducing the production of factors beneficial for the growth of H pylori. RT-PCR assays found no effect of IL10 upon human β-defensin 3 expression (data not shown), but future array-based studies will more broadly investigate the effects of IL10. In support of the in vitro experiments, we found an association between IL10 and the grade of H pylori density in gastric biopsies. Increased colonisation densities were found in H pylori-infected mice with adoptively transferred CD4+CD25hi cells,48 indicating that Tregs inhibit protective immunity. Our data show that H pylori colonisation levels may also be influenced by the cytokine environment in the gastric epithelium. Mice with higher-level IL10 responses to H pylori have reduced inflammatory pathology and higher colonisation densities.35 H pylori densities in humans are not consistently correlated with peptic ulceration,57 supporting our hypothesis that IL10 protects from ulceration but prevents clearance, possibly by inhibiting anti-bacterial effectors (fig 7). Our observational data cannot indicate whether pre-existing high levels of Tregs influenced the susceptibility to H pylori infection, but our data and published work10 11 49 indicate that the Treg response is involved in the host–bacterial interaction.

    Figure 7
    Figure 7 Diagram of how the T cell response may influence inflammation, bacterial colonisation density and occurrence of H pylori-mediated disease. Helicobacter pylori stimulates gastric epithelial cells to elicit a pro-inflammatory response.58 The transcription factor nuclear factor kappa B (NF-κB) is activated through surface-bound (eg, toll-like receptors), or intracellular (eg, nucleotide-binding oligomerisation domain containing gene 1 (NOD1)) pattern recognition receptors. This leads to the expression of pro-inflammatory cytokines and chemokines which attract cells such as neutrophils and lymphocytes to the infected tissue, and contributes to increased tissue damage and ulceration. The expression of antibacterial effectors such as defensins is also controlled by NF-κB.59 In the presence of predominantly T helper 1 (Th1) cells, the inflammatory response is exacerbated leading to increased tissue damage and more efficient killing of bacteria via enhanced defensin expression and activation of bactericidal macrophages. With a predominantly regulatory T cell (Treg) response, we propose there to be suppressive effects on innate and adaptive immunity such that the activity of Th1 and inflammatory cells is reduced. Lower levels of pro-inflammatory factors are released and therefore the risk of ulceration and tissue damage is diminished. It is hypothesised that Treg-mediated suppression of inflammatory and epithelial cell bactericidal activity would result in increased bacterial densities and maintain prolonged colonisation.

    In conclusion, we found that the human gastric CD4+ cellular response to H pylori consists of mixed Th subsets and IL10-secreting cells including Tregs. We postulate that this is a mechanism involved in ineffective immune clearance and persistence of H pylori infections, where the human immune response to H pylori regulates the density of bacterial colonisation, the level of pathology and determines the severity of disease. We hypothesise that an efficient H pylori-specific human regulatory response is protective against peptic ulcer disease.

    Acknowledgments

    We thank Dr A Robins and Ms J Swales for their help with flow cytometry, and Dr M Bajaj-Elliot (Institute of Child Health, London) for her work on defensin expression.

    REFERENCES

    1. ↵
      1. Blaser MJ,
      2. Atherton JC
      . Helicobacter pylori persistence: biology and disease. J Clin Invest 2004;113:321–33.
      OpenUrlCrossRefPubMedWeb of Science
    2. ↵
      1. Blaser MJ
      . Hypotheses on the pathogenesis and natural history of Helicobacter pylori-induced inflammation. Gastroenterology 1992;102:720–7.
      OpenUrlPubMedWeb of Science
      1. Smythies LE,
      2. Waites KB,
      3. Lindsey JR,
      4. et al
      . Helicobacter pylori-induced mucosal inflammation is Th1 mediated and exacerbated in IL-4, but not IFN-gamma, gene-deficient mice. J Immunol 2000;165:1022–9.
      OpenUrlAbstract/FREE Full Text
    4. ↵
      1. Akhiani AA,
      2. Pappo J,
      3. Kabok Z,
      4. et al
      . Protection against Helicobacter pylori infection following immunization is IL-12-dependent and mediated by Th1 cells. J Immunol 2002;169:6977–84.
      OpenUrlAbstract/FREE Full Text
    5. ↵
      1. Mitani K,
      2. Tatsuta M,
      3. Iishi H,
      4. et al
      . Helicobacter pylori infection as a risk factor for gastric ulceration. Hepatogastroenterology 2004;51:309–12.
      OpenUrlPubMed
    6. ↵
      1. Paust S,
      2. Cantor H
      . Regulatory T cells and autoimmune disease. Immunol Rev 2005;204:195–207.
      OpenUrlCrossRefPubMedWeb of Science
      1. Hawrylowicz CM,
      2. O’Garra A
      . Potential role of interleukin-10-secreting regulatory T cells in allergy and asthma. Nat Rev Immunol 2005;5:271–83.
      OpenUrlCrossRefPubMedWeb of Science
    8. ↵
      1. Coombes JL,
      2. Robinson NJ,
      3. Maloy KJ,
      4. et al
      . Regulatory T cells and intestinal homeostasis. Immunol Rev 2005;204:184–94.
      OpenUrlCrossRefPubMedWeb of Science
    9. ↵
      1. Raghavan S,
      2. Suri-Payer E,
      3. Holmgren J
      . Antigen-specific in vitro suppression of murine Helicobacter pylori-reactive immunopathological T cells by CD4CD25 regulatory T cells. Scand J Immunol 2004;60:82–8.
      OpenUrlCrossRefPubMed
    10. ↵
      1. Rad R,
      2. Brenner L,
      3. Bauer S,
      4. et al
      . CD25+/Foxp3+ T cells regulate gastric inflammation and Helicobacter pylori colonization in vivo. Gastroenterology 2006;131:525–37.
      OpenUrlCrossRefPubMedWeb of Science
    11. ↵
      1. Lundgren A,
      2. Stromberg E,
      3. Sjoling A,
      4. et al
      . Mucosal FOXP3-expressing CD4+ CD25high regulatory T cells in Helicobacter pylori-infected patients. Infect Immun 2005;73:523–31.
      OpenUrlAbstract/FREE Full Text
    12. ↵
      1. Khattri R,
      2. Cox T,
      3. Yasayko SA,
      4. et al
      . An essential role for scurfin in CD4+CD25+ T regulatory cells. Nat Immunol 2003;4:337–42.
      OpenUrlCrossRefPubMedWeb of Science
    13. ↵
      1. Thompson C,
      2. Powrie F
      . Regulatory T cells. Curr Opin Pharmacol 2004;4:408–14.
      OpenUrlCrossRefPubMedWeb of Science
    14. ↵
      1. O’Garra A,
      2. Vieira PL,
      3. Vieira P,
      4. et al
      . IL-10-producing and naturally occurring CD4+ Tregs: limiting collateral damage. J Clin Invest 2004;114:1372–8.
      OpenUrlCrossRefPubMedWeb of Science
    15. ↵
      1. Weiner HL
      . Induction and mechanism of action of transforming growth factor-beta-secreting Th3 regulatory cells. Immunol Rev 2001;182:207–14.
      OpenUrlCrossRefPubMedWeb of Science
    16. ↵
      1. Lindholm C,
      2. Quiding-Järbrink M,
      3. Lonroth H,
      4. et al
      . Local cytokine response in Helicobacter pylori-infected subjects. Infect Immun 1998;66:5964–71.
      OpenUrlAbstract/FREE Full Text
    17. ↵
      1. Bamford KB,
      2. Fan X,
      3. Crowe SE,
      4. et al
      . Lymphocytes in the human gastric mucosa during Helicobacter pylori have a T helper cell 1 phenotype. Gastroenterology 1998;114:482–92.
      OpenUrlCrossRefPubMedWeb of Science
    18. ↵
      1. Itoh T,
      2. Wakatsuki Y,
      3. Yoshida M,
      4. et al
      . The vast majority of gastric T cells are polarized to produce T helper 1 type cytokines upon antigenic stimulation despite the absence of Helicobacter pylori infection. J Gastroenterol 1999;34:560–70.
      OpenUrlCrossRefPubMedWeb of Science
    19. ↵
      1. Kosunen TU,
      2. Hook-Nikanne J,
      3. Salomaa A,
      4. et al
      . Increase of allergen-specific immunoglobulin E antibodies from 1973 to 1994 in a Finnish population and a possible relationship to Helicobacter pylori infections. Clin Exp Allergy 2002;32:373–8.
      OpenUrlCrossRefPubMedWeb of Science
    20. ↵
      1. Matricardi PM,
      2. Rosmini F,
      3. Riondino S,
      4. et al
      . Exposure to foodborne and orofecal microbes versus airborne viruses in relation to atopy and allergic asthma: epidemiological study. BMJ 2000;320:412–7.
      OpenUrlAbstract/FREE Full Text
    21. ↵
      1. Pessi T,
      2. Virta M,
      3. Adjers K,
      4. et al
      . Genetic and environmental factors in the immunopathogenesis of atopy: interaction of Helicobacter pylori infection and IL4 genetics. Int Arch Allergy Immunol 2005;137:282–8.
      OpenUrlCrossRefPubMedWeb of Science
    22. ↵
      1. Chen Y,
      2. Blaser MJ
      . Inverse associations of Helicobacter pylori with asthma and allergy. Arch Intern Med 2007;167:821–7.
      OpenUrlCrossRefPubMedWeb of Science
    23. ↵
      1. Umetsu DT,
      2. DeKruyff RH
      . The regulation of allergy and asthma. Immunol Rev 2006;212:238–55.
      OpenUrlCrossRefPubMedWeb of Science
    24. ↵
      1. Aviles-Jimenez F,
      2. Letley DP,
      3. Gonzalez-Valencia G,
      4. et al
      . Evolution of the Helicobacter pylori vacuolating cytotoxin in a human stomach. J Bacteriol 2004;186:5182–5.
      OpenUrlAbstract/FREE Full Text
    25. ↵
      1. Zhang Y,
      2. Argent RH,
      3. Letley DP,
      4. et al
      . Tyrosine phosphorylation of CagA from Chinese Helicobacter pylori isolates in AGS gastric epithelial cells. J Clin Microbiol 2005;43:786–90.
      OpenUrlAbstract/FREE Full Text
    26. ↵
      1. Lee A,
      2. O’Rourke J,
      3. De Ungria MC,
      4. et al
      . A standardized mouse model of Helicobacter pylori infection: introducing the Sydney strain. Gastroenterology 1997;112:1386–97.
      OpenUrlCrossRefPubMedWeb of Science
    27. ↵
      1. Robinson K,
      2. Loughlin M,
      3. Potter R,
      4. et al
      . Host adaptation and immune modulation are mediated by homologous recombination in Helicobacter pylori. J of Infect Dis 2005;191:579–87.
      OpenUrlAbstract/FREE Full Text
    28. ↵
      1. Pfaffl MW
      . A new mathematical model for relative quantification in real-time RT-PCR. Nucleic Acids Res 2001;29:e45.
      OpenUrlAbstract/FREE Full Text
    29. ↵
      1. Walker MR,
      2. Kasprowicz DJ,
      3. Gersuk VH,
      4. et al
      . Induction of FoxP3 and acquisition of T regulatory activity by stimulated human CD4+CD25− T cells. J Clin Invest 2003;112:1437–43.
      OpenUrlCrossRefPubMedWeb of Science
    30. ↵
      1. Niiro H,
      2. Otsuka T,
      3. Izuhara K,
      4. et al
      . Regulation by interleukin-10 and interleukin-4 of cyclooxygenase-2 expression in human neutrophils. Blood 1997;89:1621–8.
      OpenUrlAbstract/FREE Full Text
    31. ↵
      1. Reyes-Leon A,
      2. Atherton JC,
      3. Argent RH,
      4. et al
      . Heterogeneity in the activity of Mexican Helicobacter pylori strains in gastric epithelial cells and its association with diversity in the cagA gene. Infect Immun 2007;75:3445–54.
      OpenUrlAbstract/FREE Full Text
    32. ↵
      1. Hamlet A,
      2. Thoreson AC,
      3. Nilsson O,
      4. et al
      . Duodenal Helicobacter pylori infection differs in cagA genotype between asymptomatic subjects and patients with duodenal ulcers. Gastroenterology 1999;116:259–68.
      OpenUrlCrossRefPubMedWeb of Science
    33. ↵
      1. Viala J,
      2. Chaput C,
      3. Boneca IG,
      4. et al
      . Nod1 responds to peptidoglycan delivered by the Helicobacter pylori cag pathogenicity island. Nat Immunol 2004;5:1166–74.
      OpenUrlCrossRefPubMedWeb of Science
    34. ↵
      1. Moore KW,
      2. de Waal Malefyt R,
      3. Coffman RL,
      4. et al
      . Interleukin-10 and the interleukin-10 receptor. Annu Rev Immunol 2001;19:683–765.
      OpenUrlCrossRefPubMedWeb of Science
    35. ↵
      1. Fox JG,
      2. Beck P,
      3. Dangler CA,
      4. et al
      . Concurrent enteric helminth infection modulates inflammation and gastric immune responses and reduces Helicobacter-induced gastric atrophy. Nat Med 2000;6:536–42.
      OpenUrlCrossRefPubMedWeb of Science
    36. ↵
      1. Sommer F,
      2. Faller G,
      3. Konturek P,
      4. et al
      . Antrum- and corpus mucosa-infiltrating CD4(+) lymphocytes in Helicobacter pylori gastritis display a Th1 phenotype. Infect Immun 1998;66:5543–6.
      OpenUrlAbstract/FREE Full Text
    37. ↵
      1. Maino VC,
      2. Picker LJ
      . Identification of functional subsets by flow cytometry: intracellular detection of cytokine expression. Cytometry 1998;34:207–15.
      OpenUrlCrossRefPubMedWeb of Science
    38. ↵
      1. Pelchen-Matthews A,
      2. Parsons IJ,
      3. Marsh M
      . Phorbol ester-induced downregulation of CD4 is a multistep process involving dissociation from p56lck, increased association with clathrin-coated pits, and altered endosomal sorting. J Exp Med 1993;178:1209–22.
      OpenUrlAbstract/FREE Full Text
    39. ↵
      1. Quiding-Järbrink M,
      2. Lundin BS,
      3. Lonroth H,
      4. et al
      . CD4+ and CD8+ T cell responses in Helicobacter pylori-infected individuals. Clin Exp Immunol 2001;123:81–7.
      OpenUrlCrossRefPubMed
    40. ↵
      1. Orsini B,
      2. Ottanelli B,
      3. Amedei A,
      4. et al
      . Helicobacter pylori cag pathogenicity island is associated with reduced expression of interleukin-4 (IL-4) mRNA and modulation of the IL-4delta2 mRNA isoform in human gastric mucosa. Infect Immun 2003;71:6664–7.
      OpenUrlAbstract/FREE Full Text
    41. ↵
      1. D’Elios MM,
      2. Amedei A,
      3. Del Prete G
      . Helicobacter pylori antigen-specific T-cell responses at gastric level in chronic gastritis, peptic ulcer, gastric cancer and low-grade mucosa-associated lymphoid tissue (MALT) lymphoma. Microbes Infect 2003;5:723–30.
      OpenUrlCrossRefPubMed
    42. ↵
      1. Garhart CA,
      2. Heinzel FP,
      3. Czinn SJ,
      4. et al
      . Vaccine-induced reduction of Helicobacter pylori colonization in mice is interleukin-12 dependent but gamma interferon and inducible nitric oxide synthase independent. Infect Immun 2003;71:910–21.
      OpenUrlAbstract/FREE Full Text
    43. ↵
      1. Akhiani AA,
      2. Schon K,
      3. Lycke N
      . Vaccine-induced immunity against Helicobacter pylori infection is impaired in IL-18-deficient mice. J Immunol 2004;173:3348–56.
      OpenUrlAbstract/FREE Full Text
    44. ↵
      1. Kullberg MC,
      2. Jankovic D,
      3. Gorelick PL,
      4. et al
      . Bacteria-triggered CD4(+) T regulatory cells suppress Helicobacter hepaticus-induced colitis. J Exp Med 2002;196:505–15.
      OpenUrlAbstract/FREE Full Text
    45. ↵
      1. O’Garra A,
      2. Vieira P
      . T(H)1 cells control themselves by producing interleukin-10. Nat Rev Immunol 2007;7:425–8.
      OpenUrlCrossRefPubMedWeb of Science
    46. ↵
      1. Chen Z,
      2. O’Shea JJ
      . Th17 cells: a new fate for differentiating helper T cells. Immunol Res (in press).
    47. ↵
      1. Fitzgerald DC,
      2. Zhang GX,
      3. El-Behi M,
      4. et al
      . Suppression of autoimmune inflammation of the central nervous system by interleukin 10 secreted by interleukin 27-stimulated T cells. Nat Immunol 2007;8:1372–9.
      OpenUrlCrossRefPubMedWeb of Science
    48. ↵
      1. Raghavan S,
      2. Fredriksson M,
      3. Svennerholm AM,
      4. et al
      . Absence of CD4+CD25+ regulatory T cells is associated with a loss of regulation leading to increased pathology in Helicobacter pylori-infected mice. Clin Exp Immunol 2003;132:393–400.
      OpenUrlCrossRefPubMedWeb of Science
    49. ↵
      1. Lundgren A,
      2. Suri-Payer E,
      3. Enarsson K,
      4. et al
      . Helicobacter pylori-specific CD4+ CD25high regulatory T cells suppress memory T-cell responses to H. pylori in infected individuals. Infect Immun 2003;71:1755–62.
      OpenUrlAbstract/FREE Full Text
    50. ↵
      1. Sawalha AH,
      2. Schmid WR,
      3. Binder SR,
      4. et al
      . Association between systemic lupus erythematosus and Helicobacter pylori seronegativity. J Rheumatol 2004;31:1546–50.
      OpenUrlAbstract/FREE Full Text
    51. ↵
      1. Li W,
      2. Minohara M,
      3. Su JJ,
      4. et al
      . Helicobacter pylori infection is a potential protective factor against conventional multiple sclerosis in the Japanese population. J Neuroimmunol 2007;184:227–31.
      OpenUrlCrossRefPubMedWeb of Science
    52. ↵
      1. von Mutius E
      . Allergies, infections and the hygiene hypothesis – the epidemiological evidence. Immunobiology 2007;212:433–9.
      OpenUrlCrossRefPubMedWeb of Science
    53. ↵
      1. Wilson MS,
      2. Maizels RM
      . Regulatory T cells induced by parasites and the modulation of allergic responses. Chem Immunol Allergy 2006;90:176–95.
      OpenUrlPubMed
    54. ↵
      1. Evans DG,
      2. Queiroz DM,
      3. Mendes EN,
      4. et al
      . Helicobacter pylori cagA status and s and m alleles of vacA in isolates from individuals with a variety of H. pylori-associated gastric diseases. J Clin Microbiol 1998;36:3435–7.
      OpenUrlAbstract/FREE Full Text
    55. ↵
      1. Go MF,
      2. Graham DY
      . Presence of the cagA gene in the majority of Helicobacter pylori strains is independent of whether the individual has duodenal ulcer or asymptomatic gastritis. Helicobacter 1996;1:107–11.
      OpenUrlPubMed
    56. ↵
      1. Fischer W,
      2. Puls J,
      3. Buhrdorf R,
      4. et al
      . Systematic mutagenesis of the Helicobacter pylori cag pathogenicity island: essential genes for CagA translocation in host cells and induction of interleukin-8. Mol Microbiol 2001;42:1337–48.
      OpenUrlCrossRefPubMedWeb of Science
    57. ↵
      1. Gunn MC,
      2. Stephens JC,
      3. Stewart JA,
      4. et al
      . The significance of cagA and vacA subtypes of Helicobacter pylori in the pathogenesis of inflammation and peptic ulceration. J Clin Pathol 1998;51:761–4.
      OpenUrlAbstract
    58. ↵
      1. Robinson K,
      2. Argent RH,
      3. Atherton JC
      . The inflammatory and immune response to Helicobacter pylori infection. Best Pract Res Clin Gastroenterol 2007;21:237–59.
      OpenUrlCrossRefPubMed
    59. ↵
      1. George JT,
      2. Boughan PK,
      3. Karageorgiou H,
      4. et al
      . Host anti-microbial response to Helicobacter pylori infection. Mol Immunol 2003;40:451–6.
      OpenUrlCrossRefPubMedWeb of Science

    Footnotes

    • Funding: This work was supported by the Medical Research Council (G0601170) and an award from the University of Nottingham.

    • Competing interests: None.

    • Ethics approval: The use of the patient samples for this study was approved by the Nottingham University Hospital Ethics Committee on 16 January 2001.

    Linked Articles

    • Digest
      Digest
      Robin Spiller Magnus Simren
      Gut 2008; 57 i-ii Published Online First: 12 Sep 2008.