INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

Helicobacter pylori is recognized as the most common cause of chronic active gastritis (3, 22). This bacterium is also an important pathogenic factor in peptic ulcer disease (23, 36). Treatment of H. pylori infection markedly reduces the risk of duodenal ulcer recurrence and improves associated gastric inflammatory changes (5, 13, 23, 33). Histological improvement of gastric inflammation occurs immediately following H. pylori eradication (24, 29, 34). However, the resolutions of the mucosal infiltrations with neutrophils and mononuclear cells are different: neutrophils disappear within several weeks, whereas the regression of mononuclear cells occurs gradually and may not be complete until more than 6 months after H. pylori eradication (24, 34). Since colonization by H. pylori is known to induce several cytokines regulating accumulation and cellular activation of neutrophils and mononuclear cells (2, 6, 11, 12, 25, 26, 37), these differences in resolution of the cellular infiltrates after eradication may be associated with variation in the expression of relevant cytokines in the gastric mucosa.

Among the cytokines produced in the gastric mucosa of persons colonized by H. pylori, interleukin-8 (IL-8) is known to be specifically involved in recruiting and activating neutrophils (32). In contrast, IL-6 has broad biological effects on mononuclear cells, including differentiation of lymphocytes and activation of macrophages (18). As such, we hypothesized that after H. pylori eradication, declines in mucosal IL-8 or IL-6 production are related to the resolution of infiltrating neutrophils and mononuclear cells, respectively. In the present study, we addressed this hypothesis in the setting of H. pylori eradication by using sequential examinations of mucosal IL-8 and IL-6 activity and histological indices of resolution of H. pylori-associated gastritis. To more directly examine the relationship between H. pylori and mucosal cytokine activity, we also analyzed the changes in IL-8 and IL-6 activity induced by addition of H. pylori to cultures of mucosal tissues obtained after H. pylori eradication and from uninfected controls.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Subjects. Eighty-one duodenal ulcer patients with H. pylori colonization were enrolled in the present study. Of these, 61 had active duodenal ulcers and 20 had ulcer scarring. Patients who had taken nonsteroidal anti-inflammatory drugs, proton pump inhibitors, or antibiotics during the preceding 3 months were excluded from the study. For controls, we included 20 asymptomatic H. pylori-negative subjects who underwent endoscopy as part of a routine medical checkup (Table 1). Informed consent was obtained from all patients in accordance with the Helsinki Declaration.

Endoscopic biopsies. All endoscopies, performed with an Olympus (Tokyo, Japan) Q20 or Q200 endoscope, were done under local anesthesia. At the time of endoscopy, five to seven biopsy specimens were obtained from adjacent areas of the gastric antrum, one each for bacterial culture of H. pylori, urease activity (Campylobacter-like organism [CLO] test; Delta West, Bentley, Australia), and routine histological examination (hematoxylin-eosin and Giemsa stains) and two to four for in vitro organ cultures for IL-8 and IL-6 measurement. All biopsy specimens were obtained from an area of endoscopically intact mucosa distant from any focal lesions, such as erosions.

Assessment of H. pylori status. H. pylori status was determined by bacterial culture, CLO test, identification of the organism in tissue sections stained with Giemsa, and ¹³C-urea breath test. Before treatment, H. pylori infection was diagnosed in all patients by positive results with at least two of the methods. Eradication was defined as the absence of H. pylori in all four assays at 6 months after therapy directed against H. pylori.

Eradication therapy and follow-up. The patients were randomly assigned to be treated with one of two protocols for eradication of H. pylori. The first regimen (regimen I) consisted of omeprazole, 20 mg once daily, and amoxicillin, 750 mg twice daily, for 2 weeks. The second regimen (regimen II) consisted of omeprazole, 20 mg twice daily, metronidazole, 500 mg twice daily, and clarithromycin, 200 mg thrice daily, for 1 week. Endoscopy was repeated at 1 and 6 months after the end of eradication therapy. For 20 patients, endoscopy also was repeated 18 months after the end of eradication therapy. Those patients who continued to have active duodenal ulcers at 1 month after eradication therapy received a full dose of a histamine H₂-receptor antagonist for the ensuing 4 months. The patients were otherwise monitored without receiving any additional medication after the end of eradication therapy.

Histology. According to the Sydney system (28), an increase in lymphocytes and plasma cells in the lamina propria categorizes the gastritis as chronic, and activity in the context of chronic gastritis refers to the density of neutrophils in the lamina propria, gastric pits, and surface epithelium. We assessed inflammation (mononuclear cell infiltration) and activity (neutrophil infiltration) on a scale of four grades, 0, 1, 2, and 3 (corresponding to none, mild, moderate, and severe, respectively, in the Sydney system), by using hematoxylin-eosin-stained tissue sections. In addition, the numbers of neutrophils and mononuclear cells present in the lamina propria were counted in five high-power (×200 magnification) fields. A semiquantitative assessment of H. pylori density was made by classifying the numbers of bacteria into four grades (0 to 3), by using the serial sections stained immunohistochemically with polyclonal anti-H. pylori antibody, as described previously (2, 30). All histological evaluations were made without knowledge of H. pylori status and without clinical or experimental results.

Organ culture. Mucosal biopsy specimens were introduced into culture inserts (Falcon, Oxnard, Calif.) in six-well polystyrene plates (Falcon) containing RPMI 1640 medium with 5% heat-inactivated fetal calf serum, HEPES buffer, 100 U of penicillin G per ml, and 100 µg of streptomycin per ml (1.0 ml of medium per 10 mg of tissue) in a 5% CO₂ incubator for 24 h (2). In some experiments, mucosal biopsy tissues were cultured on a culture insert placed over the wells containing the culture medium without antibiotics, in the presence of 10⁸ H. pylori bacteria (a cagA-positive strain, ATCC 43504, and a cagA-negative clinical isolate, NUHP498). At the end of the incubation, the culture supernatant was collected from each well, filter sterilized, and kept at 70°C until assayed for IL-8 and IL-6 levels. The biopsy tissues were recovered from the culture insert and homogenized in 1.0 ml of 3.3 mM CaCl₂. The total protein in the homogenate was assayed by a modification of the Lowry method (27).

IL-8 and IL-6 assays. Levels of IL-8 and IL-6 in the culture supernatant were assayed in duplicate by using enzyme-linked immunosorbent assay kits specific for IL-8 and IL-6 (TFB, Tokyo, Japan) in accordance with the manufacturer's instructions. In these assays, the lower limits of detection were 3.0 pg/ml for IL-8 and 4.0 pg/ml for IL-6. The amounts of IL-8 and IL-6 in the organ cultures were expressed as nanograms per milligram of biopsy specimen protein.

Statistical analysis. Statistical analysis was performed by ², Mann-Whitney U, paired t, or Spearman rank test depending on the data set of concern. A P value of less than 0.05 was considered to be statistically significant.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

H. pylori eradication. Fifty-one duodenal ulcer patients with H. pylori infection were treated with regimen I, and eradication of H. pylori was achieved in 14 patients (27%). For the other 30 patients treated with regimen II, eradication was obtained in 25 patients (83%) (Table 1). One month after the end of therapy, ulcer healing was confirmed endoscopically in 36 of 42 patients (86%) treated with regimen I and in 18 of 19 patients (95%) treated with regimen II. For both regimens, there were no significant differences in the eradication rates between the patients who had active ulcers and those who had ulcer scarring before treatment. In total, eradication was successful in 39 patients and failed in 42 patients, providing two large groups in which to study the effect of eradication on histology and cytokine levels.

Histology. We compared the grades of histological antral gastritis in the initial and follow-up endoscopic biopsy specimens from patients who achieved or failed H. pylori eradication (Fig. 1). In the patients in whom H. pylori was eradicated, there was a rapid decrease in the grades of activity and inflammation after therapy. At 1 month following therapy, neutrophil infiltration was scored as grade 0 in all patients except for three, who were scored as grade 1. Resolution of mononuclear cell infiltration occurred more gradually (Fig. 1). In contrast, in the patients in whom H. pylori was not eradicated, there were much smaller or no decreases in the grades of activity (neutrophil infiltration) and inflammation (mononuclear cell infiltration) at 1 and 6 months after therapy. In addition to histological grading of the antral gastritis, we performed a more direct quantitative analysis of histological changes in which neutrophil and mononuclear cell densities in the lamina propria were examined (Table 2). In patients in whom H. pylori was eradicated, the neutrophil density decreased rapidly to the range observed in control (H. pylori-negative) subjects. The mononuclear cell density showed progressive and significant decreases after eradication therapy; however, even 6 months after eradication therapy, it remained higher in patients in whom H. pylori was eradicated than in the control subjects. In the patients who failed to achieve H. pylori eradication, decreases in the density of neutrophils and mononuclear cells were more modest than those in the patients with H. pylori eradication.

View larger version (35K): [in this window] [in a new window]   FIG. 1.   Histological analysis of activity defined by density of polymorphonuclear leukocytes (A), inflammation defined by density of mononuclear cells (B), and H. pylori density (C) in duodenal ulcer patients in whom H. pylori was or was not eradicated. Before, before treatment; 1M, 1 month after treatment; 6M, 6 months after treatment. *, P < 0.05 compared to value before treatment (2 test).

Cytokine levels. IL-8 and IL-6 levels in organ cultures of antral biopsy tissues were determined before and after therapy (Fig. 2). In those patients in whom H. pylori was eradicated, there was a substantial decrease in IL-8 levels at 1 month after the end of therapy, with levels approximating those observed in control subjects, which persisted for at least 18 months. IL-6 activity decreased more gradually after H. pylori therapy and normalization was observed only at 6 months; these decreased levels persisted for the next 12 months. In the patients who failed H. pylori eradication therapy, both IL-8 and IL-6 activity decreased only slightly or not at all, with levels remaining significantly higher than those in control subjects.

View larger version (14K): [in this window] [in a new window]   FIG. 2.   IL-8 (A) and IL-6 (B) levels in 24-h organ cultures of mucosal tissues in H. pylori-infected duodenal ulcer patients in whom eradication therapy was attempted and in uninfected controls. Data are stratified by whether eradication occurred (n = 39) or not (n = 42). Before, before treatment; 1M, 1 month after treatment; 6M, 6 months after treatment; 18M, 18 months after treatment. For the 18M group, samples were obtained from only 20 patients. Bars indicate median values for each group. In statistical analyses by paired t tests, all the following comparisons are significant (P < 0.05): for panel A, a versus b, a versus c, a versus d, e versus f, f versus i, g versus i, and h versus i; for panel B, a versus b, a versus c, b versus c, a versus d, b versus d, b versus i, f versus i, g versus i, and h versus i.

Addition of H. pylori to mucosal organ culture. To address the relationship between the presence of H. pylori colonization and mucosal cytokine activity more directly, either one of two H. pylori strains, ATCC 43504 (cagA positive) and NUHP498 (cagA negative), was added to cultures of mucosal biopsy tissues obtained 6 months after the end of therapy from patients in whom H. pylori was eradicated, and the cytokine levels in these cultures were compared with those in biopsy tissue cultures from H. pylori-negative control subjects (Fig. 3). The addition of these strains induced a significant increase in IL-8 activity in both the control subjects and in the patients in whom H. pylori was eradicated, indicating a nonspecific effect. In contrast, addition of H. pylori yielded a significant increase in IL-6 activity only in the formerly colonized group, indicating a specific effect.

View larger version (11K): [in this window] [in a new window]   FIG. 3.   Cytokine levels in 24-h organ cultures of mucosal tissues cocultured with a cagA-positive H. pylori strain ATCC 43504, or a cagA-negative strain, NUHP498. Tissues are from uninfected controls and from duodenal ulcer patients in whom H. pylori was eradicated. The specimens were obtained from the patients 6 months after eradication therapy. Bars indicate median values for each group. In statistical analyses by the Mann-Whitney U test, all the following comparisons are significant: for panel A, a versus c (P < 0.001), a versus e (P < 0.001), b versus d (P < 0.001), and b versus f (P < 0.001); for panel B, a versus c (P < 0.001) and a versus e (P < 0.001).

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

H. pylori is the major recognized etiological agent inducing gastric inflammatory responses (3, 10, 22). These responses can be considered to have two components: an acute inflammatory response characterized by intraepithelial and interstitial neutrophilic infiltration and chronic inflammatory responses associated with increased numbers of mononuclear cells in the lamina propria, including lymphocytes, monocytes/macrophages, and plasma cells (3). These two responses may be regulated differentially following induction of cytokines involved in the inflammatory cascade, including tumor necrosis factor alpha (TNF-), IL-1, IL-6, and IL-8 (2, 6, 11, 26). IL-8 is specifically implicated in the pathogenesis of infectious and inflammatory conditions associated with neutrophil infiltration because of its potent chemotactic and stimulatory activity for neutrophils (15, 21, 31). In contrast, IL-6 may be involved in chronic inflammatory changes through its broad effects on growth and differentiation of mononuclear cells, including T and B lymphocytes and macrophages (14, 16, 19, 20). We have previously reported that in biopsy specimens from persons colonized with H. pylori, IL-8 is mainly present in gastric epithelial cells and macrophages (2). As assessed by immunofluorescence microscopy, we have also found that IL-6 is present chiefly in macrophages (data not shown). In the present study, by evaluating the effects of H. pylori eradication on IL-8 and IL-6 levels and the resolution of H. pylori-associated gastritis, we hoped to better understand the interrelationships between histological gastritis and mucosal cytokine activity in the context of H. pylori colonization.

With H. pylori eradication, the substantial decreases in the Sydney system-derived grades of activity and inflammation were verified by more quantitative analysis of the actual numbers of neutrophils and mononuclear cells present in the lamina propria. The observations that neutrophilic infiltration had completely disappeared by 1 month after the end of successful therapy and that mononuclear cell infiltration resolved more slowly confirm earlier studies in Western populations that histological resolution of neutrophilic and mononuclear cell infiltration occurs differentially after H. pylori eradication (24, 34).

The most significant observation in the present study is that there was a dichotomy in the changes of IL-8 and IL-6 levels in the organ cultures of mucosal tissues after H. pylori eradication. The IL-8 normalization occurred rapidly within 1 month after eradication therapy, whereas the IL-6 levels normalized more slowly. Gastric epithelial cells are the producers of IL-8 in H. pylori infection (2, 7, 9), and we have reported that IL-8 levels are low in persons with low levels of gastric inflammation (2). Thus, removal of H. pylori and the subsequent turnover of the gastric epithelium may have caused the rapid IL-8 normalization. Consistent with the anticipated roles of these cytokines, the IL-8 levels showed a strong positive correlation with the numbers of neutrophils in tissue sections, while IL-6 levels correlated strongly with those of mononuclear cells. Moss et al. found a rapid and marked reduction in both tissue IL-8 mRNA and the numbers of infiltrating neutrophils after H. pylori eradication, which differed from the more modest fall in TNF- gene expression and only partial resolution of the chronic inflammatory cell infiltrates (25). Taken together, these observations suggest that the difference in the resolution of inflammatory cells after eradication may reflect the kinetics of two types of cytokines with different leukocyte specificities (IL-8 versus IL-6 and TNF-). However, to understand the mechanisms involved in the persistence of chronic inflammatory cell infiltrates after H. pylori eradication, other cytokines should also be examined (8), since we now show that compared with the mononuclear cell density in H. pylori-negative controls, that in patients 6 months after eradication therapy remained higher, despite IL-6 normalization.

That the addition of H. pylori to cultures of mucosal tissues induced significant increases in IL-8 levels in both patients who had achieved H. pylori eradication and in controls indicates a direct innate (nonimmune) response to H. pylori cells, regardless of prior exposure to H. pylori. Such an innate response is also consistent with the rapid fall in tissue IL-8 levels and neutrophils following H. pylori eradication, the immediate stimulus no longer being present. In contrast, the mounting of an IL-6 response after H. pylori challenge in the biopsy specimens from patients in whom the organism was eradicated, but not by those from the uninfected control subjects indicates the possibility that there is a persistent specific immune response in hosts in whom the organism was eradicated, lasting at least 6 months. Considering the preferential effects of IL-6 on B-cell-mediated responses (18), that IL-6 can be induced by rechallenge with H. pylori for at least 6 months after successful eradication may explain at least in part why very few patients are colonized with H. pylori after eradication therapy (4, 35).

Interestingly, the patients who failed to eradicate H. pylori also showed decreases in the grades of activity and inflammation and in H. pylori density, although these phenomena were much less prominent than in those who achieved eradication. However, these were transient responses, since by 6 months a trend toward return to the pretreatment levels emerged. Previous studies have shown a close relationship between the density of H. pylori, the histological severity of gastritis, and the mucosal IL-8 activity in the gastric mucosa (1, 2, 30). In total, the findings for these suppression-without-eradication subjects are consistent with a highly regulated bidirectional interaction between H. pylori populations in the mucus layer and inflammatory mediators in tissue, as hypothesized previously (17).

In conclusion, eradication of H. pylori leads to significant improvement in both acute and chronic inflammatory responses in the gastric mucosa. The resolution of the acute inflammatory response occurs promptly, with a rapid decrease in mucosal IL-8 activity, while chronic inflammatory changes resolve slowly, with a gradual decrease in IL-6 activity. These studies illustrate the interplay between innate and immune responses to H. pylori in the development of gastric inflammation. Further work is needed to fully understand the determinants of the slowly resolving chronic inflammatory response after H. pylori eradication.