Differences in peripheral blood lymphocyte phenotypes between Helicobacter pylori-positive children and adults with duodenal ulcer

Transcription

1 ORIGINAL ARTICLE /j x Differences in peripheral blood lymphocyte phenotypes between Helicobacter pylori-positive children and adults with duodenal ulcer T. Figueiredo Soares 1, G. Aguiar Rocha 1, A. M. Camargos Rocha 1, R. Corrêa-Oliveira 2, O. A. Martins- Filho 3, A. S. Teles Carvalho 4, P. F. Souto Bittencourt 5, C. Afonso Oliveira 1, A. M. M. Ferreira Nogueira 6, M. M. D. Álvares Cabral 6, A. M. Caetano Faria 7 and D. M. M. Queiroz 1 1 Laboratory of Research in Bacteriology, Faculdade de Medicina UFMG, 2 Laboratory of Cellular and Molecular Immunology, 3 Laboratory of Chagas Disease, CPqRR FIOCRUZ, 4 Department of Pediatrics, Faculdade de Medicina UFMG, 5 Service of Endoscopy, Hospital Felício Rocho, 6 Department of Pathology, Faculdade de Medicina UFMG and 7 Department of Biochemistry and Immunology, ICB UFMG, Belo Horizonte, Brazil ABSTRACT The immunological mechanisms involved in the development of duodenal ulcer, especially in childhood, are unclear. Helicobacter pylori-positive children and adults, with and without duodenal ulcer, were therefore compared with respect to CD4 + T-cells, and CD8 + T-cells, B-cells and B1a-cells, as well as cell activation (CD4 + HLA-DR + and CD8 + HLA-DR + ) and co-stimulatory (CD4 + CD28 + and CD8 + CD28 + ) markers, in peripheral blood. Children with and without duodenal ulcer differed significantly. In particular, there was a phenotypic change in CD8 + T-cells from children with ulcer that involved a 200% increase in the number of CD8 + HLA-DR + cells mm 3 and a decrease of 34.2% in the number of CD8 + CD28 + cells mm 3. This phenotype of chronically activated memory CD8 + T-cells, which has also been observed in patients with AIDS and tuberculosis, is associated with disease severity and progression. A lower frequency of B1a-cells was also observed in the group of children with ulcer. Conversely, no difference between infected adults with and without ulcer was observed, but the percentage of CD4 + HLA-DR + cells was lower in adults with ulcer, suggesting that a downregulated immune response may play a role in the development of duodenal ulcer in adults. Gastric inflammation correlated positively with CD4 + and chronically activated CD4 + T-cells in children and adults without duodenal ulcer, respectively. These results suggest that there are differences in the immunophenotyping profile between H. pylori-positive children and adults with duodenal ulcer, indicating the possibility of distinct immune mechanisms in the development of the disease according to age. Keywords Duodenal ulcer, gastric inflammation, Helicobacter pylori, immune response, immunophenotyping profile, T-cells Original Submission: 6 September 2006; Revised Submission: 17 February 2007; Accepted: 2 June 2007 Clin Microbiol Infect 2007; 13: INTRODUCTION The initial isolation and characterisation of Helicobacter pylori is attributed to Warren and Marshall [1], who provided evidence that this organism causes gastritis and peptic ulcer. H. pylori Corresponding author and reprint requests: D. M. M. Queiroz, Laboratory of Research in Bacteriology, Faculdade de Medicina, Universidade Federal de Minas Gerais, Av. Alfredo Balena , CEP , Belo Horizonte, Brazil dqueiroz@medicina.ufmg.br infects >50% of the world s population, is acquired predominantly in childhood, and persists throughout life unless treated [2]. Most infected individuals remain asymptomatic, but 15 20% of H. pylori-positive individuals will develop peptic ulcer, gastric carcinoma or mucosa-associated lymphoid tissue lymphoma [3]. Although duodenal ulcer (DU) is less common in children, primary DU in childhood is associated frequently with H. pylori infection [4,5]. The reasons for the uncommon early onset of duodenal peptic ulcer are unknown, but may be Journal Compilation Ó 2007 European Society of Clinical Microbiology and Infectious Diseases

2 1084 Clinical Microbiology and Infection, Volume 13 Number 11, November 2007 associated with an immune response of the host linked to age [6], rather than with the shorter duration of the infection in childhood. Indeed, there are marked differences in the gastric inflammatory response to H. pylori infection between children and adults with DU [7]. Although gastric inflammation is less intense in children than in adults, it is not essentially confined to the antral region, as occurs in adults, but spreads out beyond the antrum and involves the gastric corpus. Most studies evaluating the immune response to H. pylori infection have focused on adult populations [8 10]. Recently, it has been demonstrated that H. pylori infection in children without DU, in contrast to adults, does not increase the number of CD4 + T-cells in the blood, or the percentage of activation and co-stimulatory T-cell markers [6]. Since the immune response may contribute to the outcome of H. pylori infection, the present study investigated CD4 + T-cells (CD4 + CD3 + ) and CD8 + T-cells (CD8 + CD3 + ), B- cells (CD19 + CD3 ) and B1a-cells (CD19 + CD5 + ), as well as cell activation (CD4 + HLA-DR + and CD8 + HLA-DR + ) and co-stimulatory (CD4 + CD28 + and CD8 + CD28 + ) markers, in peripheral blood from H. pylori-infected children with and without DU, and compared the results with those obtained from adults. In addition, lymphocyte phenotypes were correlated with gastric inflammation scores. PATIENTS AND METHODS The study was approved by the Ethics Committee of Universidade Federal de Minas Gerais, Brazil. Signed informed consent was obtained from the participating adults as well as from the children (whenever possible) and their parents. Between January 2000 and July 2001, 25 H. pylori-positive children (15 boys; mean age 10.2 years, range 4 17 years; seven children with DU) were studied prospectively. Between May 2001 and July 2001, 21 H. pylori-positive adults (five males, mean age 40.5 years, range years, seven with DU) were also studied. These patients had been examined using gastroduodeno-endoscopy by the Gastrointestinal Service of the University Hospital Universidade Federal de Minas Gerais and Hospital Felício Rocho, Belo Horizonte, Brazil, to investigate the origin of symptoms related to the upper gastrointestinal tract. None of the patients had received antimicrobial drugs, anti-cholinergic or anti-inflammatory agents, proton pump inhibitors, or H 2 -receptor antagonists for 30 days before endoscopy. Patients who presented with complications, e.g., gastric perforation or haemorrhage, an anatomical obstacle preventing endoscopy, a history of surgery, bleeding conditions, or evidence of other clinical symptoms, were not included in the study. None of the patients was infected by Trypanosoma cruzi, as demonstrated by negative results in both a haemagglutination test (Hemacruzi; biomérieux, Rio de Janeiro, Brazil) and an indirect immunofluorescence test (IMUNOCOM Chagas; Wama Diagnóstica, São Carlos, Brazil). Intestinal parasitic infections were excluded by using Kato s, merthiolate iodine formalin and Baermann Moraes tests to examine stool samples collected before the endoscopy. The patients did not have other infectious or immunological disorders, and all were residents of Minas Gerais state, with the same ethnic background. During endoscopy, biopsy specimens were obtained from the antral and oxyntic gastric mucosa for microbiological study and histological evaluation. Two venous blood samples were also obtained from each patient: an EDTA-whole venous blood for blood cell counting and immunophenotyping, and a sample without anticoagulant for the T. cruzi tests. The serum was separated, divided into aliquots and stored at )20 C before testing. Biopsy samples were fixed in formalin 10% v v and paraffin-embedded; 5-lm sections stained with haematoxylin eosin for histological analysis were examined by two independent pathologists who were unaware of their origin. The mucosa were analysed in terms of the degree of inflammation (mononuclear cells) and activity (neutrophils), with scoring according to the revised Sydney system [11], i.e., none (0), mild (1), moderate (2) and marked (3). H. pylori status was assessed by means of culture, preformed urease test, carbol fuchsin-stained smears and [ 13 C]urea breath tests [12]. The patients were considered to be H. pylori-positive if the culture was positive, or if at least two of the other tests were positive. Peripheral blood lymphocyte phenotypes were analysed by immunofluorescence as described previously [6]. The results were expressed as absolute numbers mm 3 for CD4 + CD3 +, CD8 + CD3 + and CD19 + CD3 cell sets, based on the total number of peripheral blood lymphocytes, and as percentages for HLA-DR + - and CD28 + -expressing CD4 + and CD8 + T-cells (CD4 + HLA-DR + CD4 + cells 100; CD8 + HLA- DR + CD8 + cells 100; CD4 + CD28 + CD4 + cells 100; and CD8 + CD28 + CD8 + cells 100), and also for CD5 + -expressing CD19 + cells (CD19 + CD5 + CD19 + cells 100). The results were analysed using SPSS software v.10.0 (SPSS Inc., Chicago, IL, USA). T- and B-cell absolute numbers mm 3 of peripheral blood were presented as the median. Activation and co-stimulatory markers were presented as a median percentage of CD4 + CD28 +, CD4 + HLA- DR +, CD8 + CD28 +, CD8 + HLA-DR + and CD19 + CD5 + cells. The gastric inflammatory data were correlated with the absolute number mm 3 of T- and B-cells, and with activation and co-stimulatory T-cell percentages, scored using a threepoint system based on the reference values for each cell population [6]. This system was as follows: for CD4 + CD28 + : 0, <95%; 1, 95 96%; 2, 97 98%; and 3, >99% of the marked cells; for CD4 + HLA-DR + :0,<5%; 1, 5 10%; 2, 11 15%; and 3, >15% of the marked cells; for CD8 + CD28 + : 0, <25%; 1, 25 50%; 2, 51 80%; and 3, >80% of the marked cells; and for CD8 + HLA-DR + : 0, <20%; 1, 20 40%; 2, 41 70%; and 3, >70% of the marked cells. In order to minimise potential bias caused by sample size, the data were analysed using rank tests for small samples. Pearson s correlation coefficient (r) was used to evaluate correlations. The level of significance was set at p 0.05.

3 Figueiredo Soares et al. Lymphocyte phenotyping and duodenal ulcer 1085 RESULTS Children The gastritis activity (neutrophils) was greater in the antral mucosa of children with DU than in those without (p 0.05). No other difference was observed in the gastric histology of children (p >0.4) (Table 1). Neither atrophy nor intestinal metaplasia was seen in the childrens gastric mucosa. No difference was observed in the number of CD4 + CD3 +,CD8 + CD3 + and CD19 + CD3 cells between H. pylori-infected children with and without DU (Fig. S1; see Supplementary material). Also, the percentage of CD4 + CD28 + and CD4 + HLA-DR + cells did not differ between the two groups. However, a lower percentage of CD8 + CD28 + (p 0.05) and a higher percentage of CD8 + HLA-DR + (p 0.04) cells were observed in children with DU compared to those without. A lower percentage (p 0.02) of CD19 + CD5 + cells was also seen in the former group (Fig. S1). In order to evaluate the impact of these differences, the changes in the number of CD8 + CD28 +, CD8 + HLA-DR + and CD19 + CD5 + cells mm 3 between children with and without DU were calculated, and the differences remained significant (p <0.05) (Fig. 1). In the group of children without DU, the number of CD4 + T-cells was correlated significantly with the intensity (mononuclear cells) (p 0.01; r 0.71) and activity (neutrophils) (p 0.004; r 0.71) of the antral gastritis. A positive correlation was also observed between the number of B-cells and the intensity of gastritis in the antral mucosa (p 0.03; r 0.61). No correlation was observed in the group with DU (p >0.3). Table 1. Histological comparison of the gastric mucosa of Helicobacter pylori-positive children and adults with and without duodenal ulcer (median score according to the revised Sydney system: 0, none; 1, mild; 2, moderate; 3, marked) HP-positive patients Antral mucosa MN cells (intensity) PMN cells (activity) Oxyntic mucosa MN cells (intensity) PMN cells (activity) With duodenal ulcer Children 2 (1 3) 2 (1 3) 1 (0 2) 1 (0 1) Adults 2 (2 3) 2 (2 3) 1 (1 3) 1 (0 3) Without duodenal ulcer Children 2 (1 3) 1 (0 3) 1 (0 2) 1 (0 1) Adults 2 (2 3) 2 (2 3) 2 (1 3) 1 (0 3) HP, H. pylori; MN, mononuclear; PMN, polymorphonuclear. Cell number/mm Adults CD8 + /CD28 + CD8 + /HLA-DR + CD19 + /CD5 + p 0.04 p 0.02 p 0.04 Fig. 1. Changes in the median number of peripheral blood cells between Helicobacter pylori-positive children with and without duodenal ulcer. CD8 + CD28 +, range cells mm 3 and cells mm 3, respectively; CD8 + HLA-DR +, range cells mm 3 and cells mm 3, respectively; and CD19 + CD5 +, range cells mm 3 and cells mm 3, respectively. Statistical treatment: Mann Whitney U-test. The grey bar indicates patients with duodenal ulcer; the white bar indicates patients without duodenal ulcer. No difference was observed between adults with and without DU in terms of the intensity of the antral (p 0.6) and oxyntic (p 0.4) gastritis, or in the activity of the antral (p 0.4) and oxyntic (p 0.3) gastritis (Table 1). Atrophy was observed in the antral (n = 3) and oxyntic (n = 2) mucosa of patients without DU. Intestinal metaplasia in both gastric mucosa was observed in one patient. No pre-cancerous lesions were observed in the gastric mucosa of the group of patients with DU. The percentage of activated CD4 + T-cells was significantly lower in patients with DU than in those without (Fig. S1; p 0.03). When the impact of this difference was evaluated in terms of the number of activated CD4 + T-cells mm 3, the results remained significant (88 cells mm 3 in patients without DU; 33 cells mm 3 in patients with DU; p 0.02). A tendency towards a decrease in the percentage of CD8 + cells expressing both CD28 + (p 0.1) and HLA-DR + (p 0.1) was also observed in the group with DU, but this did not reach statistical significance. The number of T- and B-cells, and the percentage of B-cells expressing CD5, did not differ between infected adults with and without DU (Fig. S1). In adults without DU, the degree of inflammation (mononuclear cells) in the antrum correlated

4 1086 Clinical Microbiology and Infection, Volume 13 Number 11, November 2007 positively with the number of CD4 + HLA-DR + cells (p 0.05; r 0.87), and negatively with the number of CD4 + CD28 + cells (p 0.03; r )0.91). No other correlation was observed in this group or the group of patients with DU. Comparison between children and adults When children and adults without DU were compared, the intensity and activity of gastritis, both in the antral (p and p 0.002, respectively) and oxyntic (p and p 0.002, respectively) mucosa, were greater in adults than in children (Table 1). The groups did not differ in other histological parameters (p >0.2). A higher number of B-cells (p 0.01) and a higher percentage of CD19 + CD5 + cells (p 0.003) were seen in children (Fig. S1), but no other differences between the groups were identified. In the group of patients with DU, the intensity and activity of gastritis did not differ between adults and children. The number of CD8 + CD3 + cells (p 0.03) and the percentage of activated CD4 + (p 0.03) and CD8 + (p 0.01) T-cells were higher in children than in adults. Furthermore, the percentage of CD8 + CD28 + cells was lower in children than in adults (p 0.05). Otherwise, the percentage of CD19 + CD5 + cells did not differ between children and adults (p 0.7) (Fig. S1). DISCUSSION In adults, H. pylori infection increases the number of CD4 + T-cells and induces activation of both CD4 + and CD8 + T-cells [6,8 10]. Conversely, infection in children without DU affects neither the number of CD4 + and CD8 + T-cells nor the percentage of activation of the T-cells and co-stimulatory markers [6]. This more regulated immune response may explain, at least in part, why H. pylori infection is not complicated by DU in most infected children. Since the immune response may contribute to the outcome of H. pylori infection, the present study compared children and adults with and without DU. The most notable finding was a phenotypic change in children with DU that included a three-fold increase in the number of CD8 + T-cells expressing the activation marker HLA-DR +, and a decrease of c. 34% in the number of CD8 + CD28 + cells. CD28 is a cell surface molecule that is critically involved in T-cell co-stimulation, leading to increased expression of interleukin-2, interleukin-2 receptor and genes for cell survival [13]. At the time of birth, almost all CD8 + T-cells express the CD28 marker [14], but this is lost over time, resulting in a low percentage of CD8 + CD28 + cells in the elderly [15]. The loss of the CD28 marker is accompanied by acquisition of distinct functional activities [16,17]. CD8 + CD28 cells display the characteristics of activated memory lymphocytes with a greater cytotoxic ability [17,18]. Thus, a large fraction of peripheral CD8 + T-cells from children with DU (>60.0% are CD8 + CD28 ) display the phenotype of chronically activated memory T-cells. In agreement with this observation, the number of CD8 + T-cells expressing the HLA-DR activation marker increased by 200% in children with the disease. A decrease in CD8 + CD28 + cells and an increase in CD8 + HLA-DR + cells have also been reported for cases of infection with human immunodeficiency virus, and have been linked to immune dysfunction and progression to AIDS in both children [19] and adults [20]. The same changes, although to a lesser degree, have been observed in patients with active pulmonary tuberculosis [20], and a similar pattern has been demonstrated for the cardiac form of Chagas disease, in which a decreased frequency of CD8 + CD28 + cells is observed in peripheral blood [21] and activated CD8 + T-cells are found in the inflammatory infiltrate of cardiac lesions [22]. In the present study, since none of the patients had T. cruzi infection, AIDS or tuberculosis, the chronically activated cytotoxic CD8 + T-cells observed in the group of children with DU seemed to be associated with this severe clinical manifestation of H. pylori infection in childhood. Another difference observed in children with DU was a decrease of 61.5% in the number of CD19 + cells expressing CD5 +. B-cells expressing the CD5 marker on their surface (B1-cells) secrete large amounts of natural antibodies in a T-cellindependent manner, and are accepted as having a clear role in the early defence mechanisms against bacteria and viruses [23,24]. They have also been regarded as part of a primitive mechanism that bridges innate and adaptive immunity in certain specialised microenvironments, e.g., in the gut mucosa [23,24]. Since it has been suggested that B1-cell-derived natural IgA antibodies have an important role in host defence mechanisms against bacteria at the mucosa surface, the

5 Figueiredo Soares et al. Lymphocyte phenotyping and duodenal ulcer 1087 H. pylori load would be increased at the duodenal mucosa of children with a low percentage of B1- cells, with a consequent increased production of enzymes, cytotoxins and other noxious bacterial products that would predispose to ulcer formation. Given that the number of B-cells did not differ significantly between children with and without DU, the shift from CD19 + CD5 + to CD19 + CD5 cells observed in the former group may be explained by a proportional increase in the memory B-cells responsible for adaptive B immunity. This may explain, at least in part, the fact that serum levels of specific IgG against H. pylori are higher in H. pylori-positive children with DU than in those without DU [25]. As mentioned above, H. pylori infection leads to CD4 + T-cell activation in adults [6], but patients with DU have not been evaluated previously. In the present study, a lower percentage of activated CD4 + T-cells was observed in adults with DU than in those without DU. These results, together with those of Strömberg et al. [26] that showed low levels of pro-inflammatory cytokines in the duodenal mucosa of adults with DU, suggest that a down-regulated immune response occurs in these patients, which may lead to a higher bacterial load at the duodenal mucosa that would predispose to ulceration. The present results, showing a positive correlation between the degree of gastritis and the number of peripheral blood CD4 + T-cells in children without DU, are similar to those of Krauss-Etschmann et al. [27] and Lopes et al. [28], who observed correlations between gastritis scores and CD4 + T-cell numbers in the gastric mucosa of infected children. These findings show that peripheral blood, at least in part, mirrors the gastric mucosa in terms of the immune response to H. pylori. Thus, the positive correlations observed between the number of mononuclear cells in the gastric mucosa of adults without DU and the number of peripheral blood CD4 + HLA- DR + and CD4 + CD28 cells suggests an increase in the number of chronically activated CD4 + T-cells in the gastric mucosa of H. pylori-positive adults. In interpreting the data of the present study, a potential limitation that could be a source of bias is the small number of children. The low prevalence of DU in childhood, ranging from four to six new cases per year, or 1:2500 paediatric admissions in large paediatric centres [29,30], including this service [31], means that this problem is difficult to solve, but it was minimised as much as possible by analysing the data with rank tests for small samples. Overall, the results of the present study demonstrated that there are differences in the immunophenotyping profile between H. pylori-positive children and adults with DU, suggesting distinct immune mechanisms in the development of the disease according to age. ACKNOWLEDGEMENTS This work was supported by CNPq, FAPEMIG, PRPq UFMG, and FIOCRUZ, Brazil. SUPPLEMENTARY MATERIAL The following supplementary material is available for this article online at well-synergy.com: Fig. S1. Immunophenotyping profile of peripheral blood lymphocytes from Helicobacter pyloripositive children and adults. REFERENCES 1. Warren RJ, Marshall BJ. Unidentified curved bacilli in gastric epithelium in active chronic gastritis. Lancet 1983; i: Megraud F. Epidemiology of Helicobacter pylori infection. Gastroenterol Clin North Am 1993; 22: Suerbaum S, Michetti P. Helicobacter pylori infection. N Engl J Med 2002; 347: Carvalho AST, Queiroz DMM, Mendes EN et al. Helicobacter pylori (HP) infection in a large series of children with peptic ulcer: a prospective study. Gut 1997; 41 (suppl 1): A Nogueira AMF, Soares PM, Cabral MMDA, Carvalho AST, Mendes CC. Helicobacter pylori infection and gastroduodenal diseases in 513 Brazilian children. Gut 2000; 47 (suppl 1): A Soares TF, Rocha GA, Rocha AMC et al. Phenotypic study of peripheral blood lymphocytes and humoral immune response in Helicobacter pylori infection according to age. Scand J Immunol 2005; 62: Queiroz DMM, Rocha GA, Mendes EN et al. Differences in distribution and severity of Helicobacter pylori gastritis in children and adults with duodenal ulcer diseases. J Pediatr Gastroenterol Nutr 1991; 12: D Elios MM, Manghetti M, De Carli M et al. T helper 1 effector cells specific for Helicobacter pylori in the gastric antrum of patients with peptic ulcer disease. J Immunol 1997; 158: Ohara T, Arakawa T, Higuchi K et al. Overexpression of co-stimulatory molecules in peripheral mononuclear cells of Helicobacter pylori-positive peptic ulcer patients: possible difference in host responsiveness compared with nonulcer patients. Eur J Gastroenterol Hepatol 2001; 13:

6 1088 Clinical Microbiology and Infection, Volume 13 Number 11, November Fan XJ, Chua A, Shaahi CN, McDevitt J, Keeling D, Kelleger D. Gastric T lymphocyte responses to Helicobacter pylori in patients with H. pylori colonization. Gut 1994; 35: Dixon MF, Genta RM, Yardley JH, Correa P. Classification and grading of gastritis. The updated Sydney System International Workshop on the Histopathology and Gastritis, Houston, Ann J Surg 1996; 20: Queiroz DMM, Guerra JB, Rocha GA et al. IL1B and IL1RN polymorphic genes and Helicobacter pylori caga strains decrease the risk of reflux esophagitis. Gastroenterology 2004; 127: Bois LH, Minn AJ, Noel PJ et al. CD28 co-stimulation can promote T cell survival by enhancing the expression of Bcl-X L. Immunity 1995; 3: Jennings C, Rich K, Siegel JN et al. A phenotypic study of CD8 + lymphocyte subsets in infants using three-color flow cytometry. Clin Immunol Immunopathol 1994; 71: Goronzy JJ, Weyand CM. Ageing, autoimmunity and arthritis: T-cell senescence and contraction of T-cell repertoire diversity catalysts of autoimmunity and chronic inflammation. Arthritis Res Ther 2003; 5: Hamann D, Baars PA, Rep MH et al. Phenotypic and functional separation of memory and effector human CD8+ T cells. J Exp Med 1997; 186: Azuma M, Lanier LL. The role of CD28 co-stimulation in the generation of cytotoxic T lymphocytes. Curr Top Microbiol Immunol 1995; 198: Chamberlain WD, Falta MT, Kotzin BL. Functional subsets within clonally expanded CD8 + memory T cells in elderly humans. Clin Immunol 2000; 94: Paul ME, Shearer WT, Kozinetz CA et al. Comparison of CD8 + T-cell subsets in HIV-infected rapid progressor children versus non-rapid progressor children. J Allergy Clin Immunol 2001; 108: Hertoche T, Wajja A, Ntambi L et al. T cell activation, apoptosis and cytokine dysregulation in the (co)pathogenesis of HIV and pulmonary tuberculosis (TB). Clin Exp Immunol 2000; 122: Dutra WO, Martins-Filho OA, Cancado JR et al. Chagasic patients lack CD28 expression on many of their circulating T lymphocytes. Scand J Immunol 1996; 43: Cunha-Neto E, Rizzo LV, Albuquerque F et al. Cytokine production profile of heart-infiltrating T cells in Chagas disease cardiomyopathy. Braz J Med Biol Res 1996; 31: Martin F, Kearney JF. B1 cells: similarities and differences with other B cell subsets. Curr Opin Immunol 2001; 13: Fagarasan S, Honjo T. T-independent immune response: new aspects of B cell biology. Science 2000; 290: Oliveira AMR, Rocha GA, Queiroz DMM et al. Evaluation of an enzyme-linked immunosorbent assay for the diagnosis of Helicobacter pylori infection in children from different age groups with and without duodenal ulcer. J Pediatr Gastroenterol Nutr 1999; 28: Strömberg E, Edebo A, Lundin BS et al. Down-regulation of epithelial IL-8 responses in Helicobacter pylori-infected duodenal ulcer patients depends on host factors, rather than bacterial factors. Clin Exp Immunol 2005; 140: Krauss-Etschmann S, Gruber R, Plikat K et al. Increase of antigen-presenting cells in the gastric mucosa of Helicobacter pylori-infected children. Helicobacter 2005; 10: Lopes AI, Victorino RMM, Palha AM, Ruivo J, Fernandes A. Mucosal lymphocyte subsets and HLA-DR antigen expression in paediatric Helicobacter pylori-associated gastritis. Clin Exp Immunol 2000; 145: Bujanover Y, Reif S, Yahav J. Helicobacter pylori and peptic disease in the pediatric patient. Pediatr Clin North Am 1996; 43: Kawakami E, Machado RS, Fonseca JA et al. Clinical and histological features of duodenal ulcer in children and adolescents. J Pediatr (Rio J) 2004; 80: Carvalho AST. Úlcera péptica (peptic ulcer). J Pediatr 2000; 76 (suppl 2): S127 S134.