Increased frequency of activated T-cells in the Helicobacter pylori-infected antrum and duodenum

Transcription

1 FEMS Immunology and Medical Microbiology 36 (2003) 159^168 Increased frequency of activated T-cells in the Helicobacter pylori-infected antrum and duodenum E. Stro«mberg a, A. Lundgren a, A. Edebo b, S. Lundin a, A.-M. Svennerholm a, C. Lindholm a; a Abstract Department of Medical Microbiology and Immunology, Go«teborg University Vaccine Research Institute (GUVAX), Go«teborg University, Box 435, Go«teborg, Sweden b Department of Surgery, Go«teborg University, Go«teborg, Sweden Received 4 July 2002; received in revised form 16 October 2002; accepted 2 December 2002 First published online 6 February 2003 Helicobacter pylori colonize the human stomach and duodenum. The infection has been shown to induce a strong T-cell response in the stomach, whereas the response within the duodenum has been poorly characterized. Furthermore, it remains to be elucidated whether the T-cell response may contribute to ulcer formation in the host. In this study, the frequency of different T-cell subsets, their degree of activation and expression of co-stimulatory receptors in biopsies from the duodenum as well as the antrum were studied by immunohistochemistry and flow cytometry. It was also evaluated whether there are differences in the T-cell responses between duodenal ulcer patients and asymptomatic carriers that might explain why only 10^15% of the infected subjects develop duodenal ulcers. The frequencies of CD4+, CD8+ and CD45RO+, i.e. memory T-cells, were significantly increased in the antrum, and the number of CD25+ cells was considerably higher in both the antrum and duodenum of duodenal ulcer patients and asymptomatic carriers as compared to uninfected individuals. Interestingly, the levels of immunosuppressive CTLA-4+ cells were significantly higher in the duodenum of duodenal ulcer patients, as compared to the asymptomatic carriers. H. pylori cause activation of T-cells in the duodenum as well as in the stomach. Our observation of higher levels of CTLA-4+ cells in the duodenum of duodenal ulcer patients than in the asymptomatic carriers suggests that a suppressive T-cell response may be related to the development of duodenal ulcers. ß 2003 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. Keywords: Helicobacter pylori; T-cell; Human; Metaplasia; Duodenum; Antrum 1. Introduction Helicobacter pylori colonize the human stomach and areas of duodenal gastric metaplasia (DGM) in the duodenum [1,2]. Although about half of the world s population is infected with H. pylori, only a minority develops any of the symptoms that are associated with the infection, i.e. peptic ulcers, gastric lymphoma or adenocarcinoma [3]. The H. pylori infection triggers a strong immune response with in ltration of neutrophils, B- and T-cells into the gastric mucosa. However, despite inducing both local [4^8] and systemic [8^10] immune responses, the infection usually persists for decades. * Corresponding author. Tel.: +46 (31) ; Fax: +46 (31) address: catharina.lindholm@microbio.gu.se (C. Lindholm). The T-cell response is believed to play an important role in H. pylori infection and has been implicated in both protection against disease and the damaging e ects of the infection. In the human stomach, the T-cell response to H. pylori has been shown to be predominantly a T-helper type 1 (Th1) response with increased levels of interferon-q, but not of interleukin (IL)-4 [11^14]. So far, no di erences in B-cell [4,9] or T-cell responses [13,15] between duodenal ulcer (DU) patients and asymptomatic (AS) carriers have been identi ed that may explain why only 10^15% of the infected individuals develop duodenal ulcers. However, all previous studies have been limited to the immune responses in the blood or in the stomach rather than at the site of most ulcers, i.e. the duodenum. It has been demonstrated that DU patients have a higher bacterial density in the duodenal mucosa as compared to AS carriers [16] and that H. pylori strains isolated from the duodenum of DU patients express certain virulence / 03 / $22.00 ß 2003 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. doi: /s (03)

2 160 E. Stro«mberg et al. / FEMS Immunology and Medical Microbiology 36 (2003) 159^168 factors to a higher extent than strains isolated from AS carriers [17]. In addition, we have recently shown that the epithelial cytokine levels in the duodenum are signi cantly lower in DU patients than in AS carriers [18], suggesting that the duodenal immune response may be of importance for the development of duodenal ulcers. Characterization of the T-cells in ltrating the mucosa of individuals with di erent outcomes of H. pylori infection may give an indication of the importance of the T-cell response in H. pylori-associated diseases. For example, analysis of the expression of cell surface markers such as cluster of di erentiation (CD) 25 (the K-chain of the IL-2 receptor), the activation marker CD69, and the memory T-cell marker CD45RO may give an indication of the type of T-cell response that is induced in the H. pylori-infected mucosa. Furthermore, expression of the co-stimulatory receptors CD28 and cytotoxic T-lymphocyte antigen-4 (CTLA-4) is of major importance for the nature of the T-cell response. Thus, binding of CD28 to its ligands, B7.1 and B7.2 on the antigen-presenting cell (APC), is necessary for e ective activation of the T-cell [19], whereas ligation of CTLA-4 with the same ligands inhibits T-cell activation [20^23] and cytokine production [24]. In recent years a subpopulation of CD4+CD25+ cells has been identi ed as an important immunoregulatory cell population [25] and CTLA-4 expressed on these cells can actively down-regulate other immune cells [26^28]. Despite the fact that most peptic ulcers occur in the duodenal bulb, there have previously not been any reports on the nature of the T-cell response in the H. pylori-infected duodenum. In this study we have analyzed whether the T-cell responses in the duodenum and the antrum differ between H. pylori-infected AS carriers and DU patients with regard to frequencies of di erent subsets of in ltrating T-cells, their activation or expression of co-stimulatory receptors, which in turn might explain the di erent clinical outcomes of the infection. 2. Materials and methods 2.1. Subjects In total, 10 DU patients (mean age 47 years, range 24^ 75 years, six men and four women), 10 H. pylori-infected AS carriers (mean age 50 years, range 26^78 years, eight men and two women) and 10 uninfected controls (mean age 42 years, range 26^61 years, three men and seven women) were included in the study. From the rst ve AS carriers and ve DU patients, at least one biopsy from normal duodenal mucosa and one from an area of DGM were collected for comparison. In continued experiments, biopsies with an unde ned degree of DGM were collected from the other ve AS carriers and ve DU patients. Biopsies from the antrum were collected from all subjects. The uninfected individuals and the AS carriers had no history of gastrointestinal disease or symptoms. None of the H. pylori-infected subjects had taken any antibiotic medication prior to sampling of duodenal biopsies, or any anti-secretory medication for at least 10 days prior to the endoscopy. The study was approved by the Human Research Ethics Committee of the Medical Faculty, Go«- teborg University, Go«teborg, Sweden, and informed consent was obtained from all subjects included in the study Collection and handling of specimens Biopsies from the antrum and duodenum were collected during endoscopy after intake of an anti-foaming agent (Minifoam 0 ). In ve H. pylori-infected AS carriers and in ve DU patients, normal and metaplastic mucosa was identi ed by staining the duodenal bulb with methylene blue, as previously described [29,30]. In order to ensure that at least one biopsy would be from an area of DGM, an average of 10 biopsies were collected from the duodenal bulb of these subjects. Two adjacent biopsies were collected from each area of normal or metaplastic mucosa, respectively, and one of each type cut in half by means of a sterile scalpel. One of the halves was snapfrozen in OCT compound (Tissue-Tek; Miles, Elkhart, IN, USA) and used for immunohistochemical staining, and the remaining half was embedded in para n, stained with hematoxylin and examined for histopathology. The second biopsy was put in 1 ml physiological saline and used for culturing of H. pylori. In the remaining 10 infected individuals (i.e. ve AS carriers and ve DU patients), an average of six biopsies were collected from the duodenum and snap-frozen for subsequent immunohistochemical staining. In all subjects, an average of ve biopsies were collected from the antrum. One biopsy was immediately placed in 1 ml physiological saline for subsequent culturing, one was used for histopathological evaluation, and the remaining biopsies were frozen for immunohistochemical staining. Blood samples were collected from all participants and screened for H. pylori-speci c antibodies [31] Diagnosis of infection H. pylori infection was con rmed by serology and positive culture for H. pylori. All uninfected individuals included in the study were shown to be negative in both serology and culture for H. pylori. For culture, one antral and one duodenal biopsy was placed in 1 ml physiological saline and, after homogenization of biopsies, subsequently cultured on 8.5% horse blood Colombia Iso and Colombia IsoA agar plates. Plates were screened for H. pylori-like colonies after 3 days of incubation at 37 Cunder microaerobic conditions (10% CO 2,5%O 2, 85% N 2 ) and a dotblot assay using the H. pylori-speci c monoclonal antibody (mab) HP30-1:1.6 [32] was used to identify H. pylori colonies.

3 E. Stro«mberg et al. / FEMS Immunology and Medical Microbiology 36 (2003) 159^ Histopathology All samples were examined by an experienced histopathologist and evaluated for the active and chronic in ammation score, which were both graded according to the Sydney system [33]. In mucosa stained with methylene blue for the detection of gastric metaplasia, biopsies were analyzed for presence and extent of DGM, which was graded 0%, 1^20%, 20^50%, or more than 50% of the total epithelial area in the section Immunohistochemical staining of cell surface markers Immunohistochemical staining was performed on 8 Wm thick frozen sections mounted on glass slides (SuperFrost/ Plus; Menzel-Gla«ser, Braunschweig, Germany). Phosphate-bu ered saline (PBS) was used as washing solution, except for the CTLA-4 staining, which was performed with PBS^0.1% saponin to allow intracellular detection. Sections were incubated with mabs against di erent cell surface markers for 1 h at room temperature. The optimal concentrations were determined in initial experiments to be 5 Wg ml 31 for anti-cd8, 19 Wg ml 31 for anti-cd25, 30 Wg ml 31 for CD45RO, 20 Wg ml 31 for anti-cd4 (all IgG1 and purchased from Dako, Denmark), 20 Wg ml 31 for anti-cd69 (IgG1), 25 Wg ml 31 for anti-cd28 (IgG1), and 6 Wg ml 31 for CTLA-4 (IgG2a) (all purchased from Pharmingen, San Diego, CA, USA). In addition, an isotype-matched mouse IgG1 or IgG2a primary mab (Dako, Denmark) was included in each experiment. Sections were washed twice in PBS and subsequently incubated with a secondary antibody. Sections stained for CD4, CD8, CD45RO and the negative control were incubated with a horseradish peroxidase (HRP)-coupled rabbit anti-mouse antibody (P0260, Dako, Denmark), diluted 1:40 in PBS with 5% human AB þ serum for 20 min. Signals for CD25, CD69, CD28 and CTLA-4 were enhanced by using an indirect three-step method where, after 30 min incubation with a biotinylated goat anti-mouse IgG antibody (Caltag Laboratories, South San Francisco, CA, USA) diluted 1:300, sections were incubated with an avidin^biotin^ HRP complex (Vectastain, ABC-HRP kit; Vector Laboratories, Burlingame, CA, USA) according to the manufacturer s instructions. Tissue sections were developed with the chromogen 3,3P-diaminobenzidine (DAB) (Vector Laboratories) and H 2 O 2 as substrate, counterstained with Mayer s hematoxylin, dehydrated with ethanol and xylene and subsequently mounted with Mountex (Histolab, Go«teborg, Sweden). The numbers of positively stained mononuclear cells (MNC) in the lamina propria and intraepithelial lymphocytes (IEL) in the entire section were calculated using a Leica microscope at 200U magni cation. The total area of the sections and epithelium, respectively, was determined using Leica Qwin software (Leica, Germany). The number of positive cells was expressed as MNCand IEL per mm 2 tissue and epithelium, respectively Isolation of lamina propria lymphocytes Lamina propria MNCwere isolated from biopsies collected from the antrum and duodenum of two AS carriers and two uninfected volunteers. The epithelial cells as well as the intraepithelial lymphocytes were removed from the biopsies by incubation in 1 mm dithiothreitol for 30 min at 37 C, followed by 3 times 45 min incubation in 1 mm EDTA, dissolved in Hanks balanced salt solution without calcium or magnesium, on a magnetic stirrer at 37 C. Lamina propria lymphocytes were subsequently isolated from the remaining tissue by incubation in collagenase/ DNase solution (100 U ml 31 collagenase, Sigma C-0255, and 0.1 mg ml 31 DNase, Sigma D-5025) at 37 Cfor 2 h on a magnetic stirrer and the cell suspension was subsequently ltered through a mesh Flow cytometric characterization of cells The cells isolated from the lamina propria compartment were analyzed for di erent T-cell subsets by ow cytometry. Lamina propria MNCwere labelled with anti-cd3 PerCP, anti-cd25 FITC (Becton-Dickinson, San Jose, CA, USA) and anti-cd69-apc (Becton-Dickinson Pharmingen, San Diego, CA, USA) antibodies diluted in PBS supplemented with 1 mm EDTA, 0.05% sodium azide, 0.1% bovine serum albumin and 1% human AB þ serum for 30 min on ice. The cells were washed and xed in formaldehyde before analysis on a FACSCalibur (Becton-Dickinson) Statistical evaluation Statistical signi cance of di erences between normal and metaplastic biopsies within a subject group was evaluated with the Wilcoxon match pairs test, and di erences between AS carriers, DU patients and uninfected individuals were calculated with the Mann^Whitney test. P values were de ned as signi cant. 3. Results 3.1. Increased frequency of CD25+ cells in the duodenum of H. pylori-infected individuals The frequencies of di erent T-cell subsets, i.e. cells expressing CD4, CD8, CD25, the activation marker CD69, and the memory marker CD45RO, were determined by immunohistochemical staining of biopsies from the duodenum of H. pylori-infected DU patients, AS carriers and uninfected controls (Fig. 1). Since H. pylori only colonize the duodenum in areas of gastric metaplasia, the characteristics of the in ltrating T-cells in biopsies from these areas and from normal duodenal mucosa from the same subject were initially analyzed in parallel. However, in

4 162 E. Stro«mberg et al. / FEMS Immunology and Medical Microbiology 36 (2003) 159^168 Fig. 1. Light microphotographs showing immunohistochemical detection of cell surface markers in tissue sections from antrum (A^D) and duodenum (E,F). Staining of CD4+ cells in biopsies from a DU patient (A) and an uninfected control (B), CD8+ staining in biopsies from a DU patient (C) and an uninfected control (D), and staining of CD25+ cells in biopsies from an AS carrier (E) and an uninfected control (F). Arrows indicate examples of positive cells. Magni cation U100.

5 E. Stro«mberg et al. / FEMS Immunology and Medical Microbiology 36 (2003) 159^ Fig. 2. Immunohistochemical detection of lamina propria MNCpopulations in the duodenum of AS carriers, DU patients and uninfected controls (H.p.3). Each dot represents a biopsy from one individual and horizontal lines represent the median for each study group. initial experiments when normal and metaplastic biopsies from the duodenum of DU patients and AS carriers were compared, the frequencies of the di erent T-cell subsets analyzed were found to be similar (data not shown). Therefore, biopsies from the two di erent mucosa types were not analyzed separately in subsequent experiments. In the lamina propria of the duodenal biopsies, the frequencies of cells expressing CD4 and CD8, as well as CD45RO, i.e. memory cells, were similar in H. pylori-infected AS carriers, DU patients and uninfected subjects (Fig. 2). However, signi cantly higher numbers of CD25+ cells were observed in both the H. pylori-infected AS carriers (P ) and the DU patients (P ), as compared to the uninfected controls (Fig. 2). The di erent T-cell subsets CD4+, CD8+, CD45RO+ and CD69+ could all be detected in the intraepithelial compartment of the duodenum (Table 1). The majority of the IEL were CD8+. No di erences in frequency of IEL with di erent surface markers could be detected between infected and uninfected subjects, or between AS carriers and DU patients Increased frequency and activation of T-cells in the H. pylori-infected antrum In the antrum, the frequencies of CD4+, CD8+ and CD45RO+ cells were increased in the lamina propria of both the H. pylori-infected AS carriers and DU patients, Table 1 Comparison of IEL populations in biopsies from the antrum and duodenum of H. pylori-infected AS carriers, DU patients, and uninfected individuals (H.p.3) Surface marker Mean number of positive cells mm 32 epithelium Duodenum Antrum AS (n = 10) DU (n = 10) H.p.3 (n = 10) AS (n = 10) DU (n = 10) H.p.3 (n = 10) CD4 16(6^144) 32(5^118) 21(4^97) 24(6^69) 46(25^176)* 17(0^52) CD8 145(53^644) 203(106^302) 215(105^414) 209(10^491) 269(57^1541) 108(9^516) CD45RO 22(8^197) 22(0^64) 23(3^114) 22(2^507) 71(7^528)* 10(0^89) CD69 78(0^479) 68(5^246) 67(0^294) 112(26^332) 124(36^283)* 31(0^188) CD25 0(0^7) 0(0^28) 0(0^0) 0(0^13) 0(0^6) 0(0^2) CD28 0(0^14) 0(0^33) 0(0^16) 0(0^17) 0(0^84) 0(0^2) CTLA-4 0(0^11) 2(0^100) 0(0^32) 4(0^28) 10(0^63) 0(0^31) Values are given as medians (interquartile ranges). *P , DU vs. H.p.3.

6 164 E. Stro«mberg et al. / FEMS Immunology and Medical Microbiology 36 (2003) 159^168 as compared to the uninfected subjects, although the differences reached statistical signi cance only for CD4+ (P ) and CD45RO+ (P ) cells in the DU patients (Fig. 3). Furthermore, the numbers of activated T-cells, i.e. cells expressing CD69, were higher in the lamina propria of the antrum of both AS carriers (P ) and DU patients (P ) than the uninfected controls. Also the frequency of CD25+ cells was increased in the AS carriers (P ) and the DU patients (P ) as compared to the uninfected individuals, where almost no CD25+ cells were detected (Fig. 3). When comparing the frequencies of the di erent T-cell subsets and the degree of activation between H. pylori-infected DU patients and AS carriers, there was a tendency for higher numbers of CD69+ and CD45RO+ memory cells in the DU patients. Among the IEL there was a signi cant increase of CD4+ (P ), CD45RO+ (P ) and CD69+ (P ) cells in the DU patients, and an increase, although not signi cant, of CD8+, CD45RO+ and CD69+ cells in the AS carriers, as compared to the uninfected individuals (Table 1) The CD25+ and CD69+ cells in the antrum and duodenum are mainly T-cells In order to determine the percentages of CD25+ and CD69+ cells that were T-cells, cells isolated from the antral and duodenal mucosa of two H. pylori-infected AS carriers and two uninfected subjects were analyzed for co-expression of the T-cell marker CD3 by ow cytometry. The majority of the CD25+ cells in the lamina propria of the duodenum (mean 78%) as well as the antrum (mean 87%) were found to be positive for CD3 (Fig. 4). Also the majority of the CD69+ cells in the duodenum and antrum were found to be T-cells (mean 89% and 88%, respectively) (Fig. 4). The proportions of CD3+ cells expressing CD25 and CD69 in the duodenum were found to be 6% and 84%, respectively. In the antrum, 12% of the CD3+ cells were CD25+ and 77% CD69+. This agrees well with the levels observed in the immunohistochemical stainings Increased numbers of CTLA-4+ cells in the H. pylori-infected duodenum Co-stimulatory molecules are necessary in order to get an e ective activation of T-cells. CD28 expressed on the T-cell promotes T-cell activation, whereas ligation of CTLA-4 with its ligands B7.1 and B7.2 attenuates the T-cell activation. In initial experiments, expression of CD28 and CTLA-4 in both normal and metaplastic duodenal mucosa was evaluated by immunohistochemistry. However, since no major di erences in the number of cells positive for the cell surface markers could be detected Fig. 3. Immunohistochemical detection of lamina propria MNCpopulations in the antrum of AS carriers, DU patients and uninfected controls (H.p.3). Each dot represents a biopsy from one individual and horizontal lines represent the median for each study group.

7 E. Stro«mberg et al. / FEMS Immunology and Medical Microbiology 36 (2003) 159^ Fig. 4. Flow cytometric analysis of CD3 expression on lamina propria cells from the duodenum and the antrum of an AS carrier. The graphs shown are from one representative experiment. Lamina propria cells expressing CD25 or CD69 were gated as indicated and the percentages of CD3+ T-cells within each of these populations were determined and are shown as numbers inside the gates. Total number of gated events s Fig. 5. Immunohistochemical detection of CD28 and CTLA-4 in the duodenum (A) and antrum (B) of AS carriers, DU patients and uninfected controls (H.p.3). Each dot represents a biopsy from one individual and horizontal lines represent the median for each study group.

8 166 E. Stro«mberg et al. / FEMS Immunology and Medical Microbiology 36 (2003) 159^168 between the two mucosa types (data not shown), duodenal biopsies were analyzed without staining for the presence of gastric metaplasia in subsequent experiments. The frequency of cells expressing CD28 in the lamina propria of the duodenum was similar in all three study groups. In contrast, the number of CTLA-4+ cells, i.e. possibly down-regulatory cells, was signi cantly higher in the duodenum of DU patients as compared to both the AS carriers (P ) and uninfected subjects (P ) (Fig. 5). Thus, the ratio of CTLA-4+ to CD28+ cells (CTLA-4/CD28) was increased in the duodenum of DU patients (median value 0.92) as compared to the AS carriers (median value 0.29), indicating predominantly downregulatory signals in the DU patients. In the antrum, the frequency of CD28+ cells in the lamina propria was found to be up-regulated in both the DU patients (P ) and the AS carriers (P ), as compared to uninfected controls (Fig. 5). Also the frequency of CTLA-4+ cells was increased in the antral lamina propria of both DU patients (P ) and AS carriers (P ), as compared to uninfected controls (Fig. 5). In the antrum, the median ratio of CTLA-4+ to CD28+ cells was found to be very similar in the AS carriers (0.63) and the DU patients (0.55). Only low numbers of CD28+ and CTLA-4+ cells were present in the intraepithelial compartment of the antrum and the duodenum (Table 1), and no signi cant di erences between the study groups were found. 4. Discussion H. pylori has been shown to trigger a strong immune response in the stomach, but much less is known about the duodenal immune response against the bacterium. We have recently shown that the cytokine levels in the epithelium of the duodenal mucosa are signi cantly lower in DU patients, as compared to both AS carriers and uninfected controls [18]. In the present study we have compared the in ltrating T-cell populations in the duodenum as well as in the antrum of H. pylori-infected DU patients and AS carriers to evaluate whether there also are di erences in the local T-cell responses between DU patients and AS carriers that could be associated with the development of duodenal ulcers. The numbers of CD4+ and CD8+ cells were found to be similar in both the lamina propria and the epithelium of the duodenum of H. pylori-infected DU patients, AS carriers and uninfected subjects. However, we found that the frequency of cells positive for CD25+, i.e. the K-chain of the IL-2 receptor, was increased in the duodenum of both AS carriers and DU patients, as compared to uninfected subjects. Interestingly, the number of cells expressing the down-regulatory receptor CTLA-4 was found to be signi cantly higher in the duodenal lamina propria of DU patients as compared to AS carriers and uninfected individuals. In recent years the existence of an immunoregulatory CD4+CD25+ T-cell subset has become evident [25]. It is still not clear how this T-cell population exerts the immunosuppression, but it has been suggested to be mediated by transforming growth factor (TGF) L and CTLA-4 [27,28,34], which both are important down-regulators of T-cell activation. Ligation of CTLA-4 on the activated T-cell with the ligands B7.1 and B7.2 on the APCresults in attenuation of T-cell activation [20^23]. The co-stimulatory receptor CD28 on the other hand, which interacts with the same ligands as CTLA-4, instead provides a strong signal for T-cell activation and the induction of cytokine production [19]. Our observation of increased numbers of CTLA-4+ cells in the duodenal mucosa of DU patients, together with our previous ndings of a more intense TGF-L staining in the lamina propria of the duodenum of DU patients than of AS carriers [18], leads us to speculate that these cells, at least in part, may be regulatory T-cells. Considering that we have found that DU patients have signi cantly lower epithelial cytokine responses in the duodenum than AS carriers [18], it seems likely that the development of duodenal ulcers at least in part may be explained by a down-regulated local immune response. Furthermore, depletion of regulatory T-cells has recently been demonstrated in an experimental mouse model of H. pylori infection to increase the gastritis and decrease the bacterial load in the stomach (S. Raghavan et al., unpublished). Thus, increased numbers of regulatory T-cells and a down-regulated epithelial cytokine response in the duodenum may lead to an increased H. pylori colonization. Indeed, we have previously shown that DU patients have, as a mean, a 20-fold higher bacterial density in the duodenum than AS carriers [16]. As a consequence, increased concentrations of harmful bacterial products, e.g. di erent enzymes and toxins, may be produced locally in the duodenum and subsequently give rise to duodenal ulcers. Whether the increased frequency of CTLA-4+ cells in the duodenum of DU patients is due to bacterial or genetic host factors, or a combination of both, remains to be determined. However, di erences between H. pylori strains infecting the duodenal mucosa of DU patients and AS carriers [17], as well as genetic di erences between individuals with and without symptoms have been described [35,36]. Since H. pylori only colonize the duodenum in areas of gastric metaplasia, we compared the frequency of the different T-cell subsets in normal and metaplastic duodenal mucosa in initial experiments. However, the frequencies of all T-cell subsets studied were comparable in normal and metaplastic duodenal mucosa. This might be explained by released bacterial factors, e.g. urease and other enzymes or toxins, which may also a ect the surrounding tissue. In the antrum, increased numbers of CD4+ and CD45RO+ were observed in the lamina propria of both

9 E. Stro«mberg et al. / FEMS Immunology and Medical Microbiology 36 (2003) 159^ AS carriers and DU patients, indicating the presence of memory T-cells in the infected stomach. This is consistent with results from previous studies, comparing the T-cell responses in individuals with H. pylori-induced gastritis and uninfected subjects [8,37^39]. The frequencies of CD69+, i.e. activated cells, as well as CD25+ cells, were found to be up-regulated in the lamina propria of the antrum of both the AS carriers and the DU patients, as compared to the healthy controls, which agrees well with a previous study comparing H. pylori-positive and -negative gastritis [37]. In our study, no di erences between AS carriers and DU patients with regard to frequencies of the di erent T-cell subsets in the antrum were observed. A previous study, using ow cytometry on isolated gastric cells, has reported increased frequencies of CD25+ cells in the stomach of DU patients, as compared to non-ulcer dyspeptic patients [40]. However, in contrast to our study no AS carriers or uninfected individuals were included in the study by Ihan et al. [40]. The vast majority of the CD25+ and CD69+ cells were con rmed to be T-cells, as demonstrated by ow cytometric analysis of their co-expression of the T-cell marker CD3. The numbers of IEL positive for CD4, CD45RO and CD69 were found to be higher in the antrum of DU patients, as compared to the uninfected controls. This supports previous ndings of an increased in ltration of CD4+ IEL in mice infected with H. pylori [41]. Also the number of CTLA-4+ cells was increased in the antrum of DU patients and AS carriers. However, in contrast to the duodenum, the frequency of CTLA-4+ cells was found to be similar in the antrum of DU patients and AS carriers. In conclusion, H. pylori cause activation of T-cells in the duodenal, as well as the antral mucosa. Our observation of an increased frequency of down-regulating T-cells, together with our previous ndings of decreased epithelial cytokine levels in the duodenum of DU patients, suggests that a locally down-regulated immune response may at least in part be related to the development of duodenal ulcers. Acknowledgements This study was nancially supported by grants from Astra Research Center, Boston, MA, USA, the Swedish Medical Research Council (16X-09084), the Swedish Medical Society, the Royal Society of Arts and Sciences in Go«teborg, and Stiftelsen Sigurd och Elsa Goljes minne. References [1] Steer, H.W. (1984) Surface morphology of the gastroduodenal mucosa in duodenal ulceration. Gut 25, 1203^1210. [2] Wyatt, J.I., Rathbone, B.J., Dixon, M.F. and Heatley, R.V. (1987) Campylobacter pyloridis and acid induced gastric metaplasia in the pathogenesis of duodenitis. J. Clin. Pathol. 40, 841^848. [3] Kuipers, E.J., Thijs, J.C. and Festen, H.P. (1995) The prevalence of Helicobacter pylori in peptic ulcer disease. Aliment. Pharmacol. Ther. 9, 59^69. [4] Mattsson, A., Quiding-Jarbrink, M., Lonroth, H., Hamlet, A., Ahlstedt, I. and Svennerholm, A. (1998) Antibody-secreting cells in the stomachs of symptomatic and asymptomatic Helicobacter pylori-infected subjects. Infect. Immun. 66, 2705^2712. [5] Lindholm, C., Quiding-Jarbrink, M., Lonroth, H. and Svennerholm, A.M. (2001) Induction of chemokine and cytokine responses by Helicobacter pylori in human stomach explants. Scand. J. Gastroenterol. 36, 1022^1029. [6] D Elios, M.M., Manghetti, M., De Carli, M., Costa, F., Baldari, C.T., Burroni, D., Telford, J.L., Romagnani, S. and Del Prete, G. (1997) T helper 1 e ector cells speci c for Helicobacter pylori in the gastric antrum of patients with peptic ulcer disease. J. Immunol. 158, 962^967. [7] Di Tommaso, A., Xiang, Z., Bugnoli, M., Pileri, P., Figura, N., Bayeli, P.F., Rappuoli, R., Abrignani, S. and De Magistris, M.T. (1995) Helicobacter pylori-speci c CD4+ T-cell clones from peripheral blood and gastric biopsies. Infect. Immun. 63, 1102^1106. [8] Fan, X.J., Chua, A., Shahi, C.N., McDevitt, J., Keeling, P.W. and Kelleher, D. (1994) Gastric T lymphocyte responses to Helicobacter pylori in patients with H. pylori colonisation. Gut 35, 1379^1384. [9] Mattsson, A., Tinnert, A., Hamlet, A., Lonroth, H., Bolin, I. and Svennerholm, A.M. (1998) Speci c antibodies in sera and gastric aspirates of symptomatic and asymptomatic Helicobacter pylori-infected subjects. Clin. Diagn. Lab. Immunol. 5, 288^293. [10] Sharma, S.A., Miller, G.G., Perez-Perez, G.I., Gupta, R.S. and Blaser, M.J. (1994) Humoral and cellular immune recognition of Helicobacter pylori proteins are not concordant. Clin. Exp. Immunol. 97, 126^132. [11] Karttunen, R., Karttunen, T., Ekre, H.P. and MacDonald, T.T. (1995) Interferon gamma and interleukin 4 secreting cells in the gastric antrum in Helicobacter pylori positive and negative gastritis. Gut 36, 341^345. [12] Bamford, K.B., Fan, X., Crowe, S.E., Leary, J.F., Gourley, W.K., Luthra, G.K., Brooks, E.G., Graham, D.Y., Reyes, V.E. and Ernst, P.B. (1998) Lymphocytes in the human gastric mucosa during Helicobacter pylori have a T helper cell 1 phenotype. Gastroenterology 114, 482^492. [13] Lindholm, C., Quiding-Jarbrink, M., Lonroth, H., Hamlet, A. and Svennerholm, A.M. (1998) Local cytokine response in Helicobacter pylori-infected subjects. Infect. Immun. 66, 5964^5971. [14] Sommer, F., Faller, G., Konturek, P., Kirchner, T., Hahn, E.G., Zeus, J., Rollingho, M. and Loho, M. (1998) Antrum- and corpus mucosa-in ltrating CD4(+) lymphocytes in Helicobacter pylori gastritis display a Th1 phenotype. Infect. Immun. 66, 5543^5546. [15] Quiding-Jarbrink, M., Lundin, B.S., Lonroth, H. and Svennerholm, A.M. (2001) CD4+ and CD8+ T cell responses in Helicobacter pyloriinfected individuals. Clin. Exp. Immunol. 123, 81^87. [16] Hamlet, A., Thoreson, A.C., Nilsson, O., Svennerholm, A.M. and Olbe, L. (1999) Duodenal Helicobacter pylori infection di ers in caga genotype between asymptomatic subjects and patients with duodenal ulcers. Gastroenterology 116, 259^268. [17] Thoreson, A.C., Hamlet, A., Celik, J., Bystrom, M., Nystrom, S., Olbe, L. and Svennerholm, A.M. (2000) Di erences in surface-exposed antigen expression between Helicobacter pylori strains isolated from duodenal ulcer patients and from asymptomatic subjects. J. Clin. Microbiol. 38, 3436^3441. [18] Stro«mberg, E., Edebo, A., Svennerholm, A.M. and Lindholm, C. (2002) Decreased epithelial cytokine responses in the duodenal mucosa of Helicobacter pylori-infected duodenal ulcer patients. Clin. Diagn. Lab. Immunol. 10, 116^124. [19] June, C.H., Bluestone, J.A., Nadler, L.M. and Thompson, C.B. (1994) The B7 and CD28 receptor families. Immunol. Today 15, 321^331. [20] Krummel, M.F. and Allison, J.P. (1996) CTLA-4 engagement inhib-

10 168 E. Stro«mberg et al. / FEMS Immunology and Medical Microbiology 36 (2003) 159^168 its IL-2 accumulation and cell cycle progression upon activation of resting T cells. J. Exp. Med. 183, 2533^2540. [21] Perez, V.L., Van Parijs, L., Biuckians, A., Zheng, X.X., Strom, T.B. and Abbas, A.K. (1997) Induction of peripheral T cell tolerance in vivo requires CTLA-4 engagement. Immunity 6, 411^417. [22] Walunas, T.L., Bakker, C.Y. and Bluestone, J.A. (1996) CTLA-4 ligation blocks CD28-dependent T cell activation. J. Exp. Med. 183, 2541^2550. [23] Thompson, C.B. and Allison, J.P. (1997) The emerging role of CTLA-4 as an immune attenuator. Immunity 7, 445^450. [24] Alegre, M.L., Shiels, H., Thompson, C.B. and Gajewski, T.F. (1998) Expression and function of CTLA-4 in Th1 and Th2 cells. J. Immunol. 161, 3347^3356. [25] Sakaguchi, S., Sakaguchi, N., Asano, M., Itoh, M. and Toda, M. (1995) Immunologic self-tolerance maintained by activated T cells expressing IL-2 receptor alpha-chains (CD25). Breakdown of a single mechanism of self-tolerance causes various autoimmune diseases. J. Immunol. 155, 1151^1164. [26] Takahashi, T., Tagami, T., Yamazaki, S., Uede, T., Shimizu, J., Sakaguchi, N., Mak, T.W. and Sakaguchi, S. (2000) Immunologic self-tolerance maintained by CD25(+)CD4(+) regulatory T cells constitutively expressing cytotoxic T lymphocyte-associated antigen 4. J. Exp. Med. 192, 303^310. [27] Read, S., Malmstrom, V. and Powrie, F. (2000) Cytotoxic T lymphocyte-associated antigen 4 plays an essential role in the function of CD25(+)CD4(+) regulatory cells that control intestinal in ammation. J. Exp. Med. 192, 295^302. [28] Chen, W., Jin, W. and Wahl, S.M. (1998) Engagement of cytotoxic T lymphocyte-associated antigen 4 (CTLA-4) induces transforming growth factor beta (TGF-beta) production by murine CD4(+) T cells. J. Exp. Med. 188, 1849^1857. [29] Mertz, H., Kovacs, T., Thronson, M. and Weinstein, W. (1998) Gastric metaplasia of the duodenum: identi cation by an endoscopic selective mucosal staining technique. Gastrointest. Endosc. 48, 32^38. [30] Futami, H., Takashima, M., Furuta, T., Hanai, H. and Kaneko, E. (1999) Relationship between Helicobacter pylori infection and gastric metaplasia in the duodenal bulb in the pathogenesis of duodenal ulcer. J. Gastroenterol. Hepatol. 14, 114^119. [31] Hamlet, A.K., Erlandsson, K.I., Olbe, L., Svennerholm, A.M., Backman, V.E. and Pettersson, A.B. (1995) A simple, rapid, and highly reliable capsule-based 14Curea breath test for diagnosis of Helicobacter pylori infection. Scand. J. Gastroenterol. 30, 1058^1063. [32] Bolin, I., Lonroth, H. and Svennerholm, A.M. (1995) Identi cation of Helicobacter pylori by immunological dot blot method based on reaction of a species-speci c monoclonal antibody with a surfaceexposed protein. J. Clin. Microbiol. 33, 381^384. [33] Price, A.B. (1991) The Sydney System: histological division. J. Gastroenterol. Hepatol. 6, 209^222. [34] Nakamura, K., Kitani, A. and Strober, W. (2001) Cell contact-dependent immunosuppression by CD4(+)CD25(+) regulatory T cells is mediated by cell surface-bound transforming growth factor beta. J. Exp. Med. 194, 629^644. [35] Garcia-Gonzalez, M.A., Lanas, A., Santolaria, S., Crusius, J.B., Serrano, M.T. and Pena, A.S. (2001) The polymorphic IL-1B and IL- 1RN genes in the aetiopathogenesis of peptic ulcer. Clin. Exp. Immunol. 125, 368^375. [36] Lanas, A., Garcia-Gonzalez, M.A., Santolaria, S., Crusius, J.B., Serrano, M.T., Benito, R. and Pena, A.S. (2001) TNF and LTA gene polymorphisms reveal di erent risk in gastric and duodenal ulcer patients. Genes Immun. 2, 415^421. [37] Seifarth, C., Funk, A., Reich, K., Dahne, I., Classen, M. and Deusch, K. (1995) Selective increase of CD4+ and CD25+ T cells but not of gamma delta T cells in H. pylori associated gastritis. Adv. Exp. Med. Biol. 371B, 931^934. [38] Hatz, R.A., Meimarakis, G., Bayerdor er, E., Stolte, M., Kirchner, T. and Enders, G. (1996) Characterization of lymphocytic in ltrates in Helicobacter pylori-associated gastritis. Scand. J. Gastroenterol. 31, 222^228. [39] Agnihotri, N., Bhasin, D.K., Vohra, H., Ray, P., Singh, K. and Ganguly, N.K. (1998) Characterization of lymphocytic subsets and cytokine production in gastric biopsy samples from Helicobacter pylori patients. Scand. J. Gastroenterol. 33, 704^709. [40] Ihan, A., Tepez, B., Kavcic, I. and Gubina, M. (1996) Il-2 receptor expression on gastric mucosa T lymphocytes is enhanced in duodenal ulcer patients compared with non-ulcer dyspeptic patients. Hepatogastroenterology 43, 1665^1670. [41] Ibraghimov, A., Kabok, Z., Garcia, G., Castriotta, L., Torrey, D. and Pappo, J. (2000) Gastric CD4+ intraepithelial lymphocytes: A novel mucosa-associated lymphoid population induced in mice by Helicobacter pylori infection. Mucosal Immunol. Sept/Dec, 9^12.