- Email: [email protected]
Sustained increase in gastric antral epithelial cell proliferation despite cure of Helicobacter pylori infection
THE AMERICAN JOURNAL OF GASTROENTEROLOGY © 2000 by Am. Coll. of Gastroenterology Published by Elsevier Science Inc.
Vol. 95, No. 4, 2000 ISSN 0002-9270/00/$20.00 PII S0002-9270(00)00724-3
Sustained Increase in Gastric Antral Epithelial Cell Proliferation Despite Cure of Helicobacter pylori Infection Hala M. T. El-Zimaity, M.D., David Y. Graham, M.D., M.A.C.G., Robert M. Genta, M.D., F.A.C.G., and Juan Lechago, M.D., Ph.D. Departments of Pathology and Medicine, Veterans Affairs Medical Center, The Methodist Hospital, and Baylor College of Medicine, Houston, Texas
OBJECTIVE: Studies of the effect of Helicobacter pylori treatment on gastric mucosa proliferation have yielded inconsistent results. We compared gastric mucosa cell proliferation posttherapy and in uninfected controls. METHODS: Biopsies were obtained from patients with H. pylori infection before treatment and at intervals for up to 33 months. Epithelial cell proliferation was determined using Ki-67 immunostaining. The labeling index (LI) is the proportion of positively labeled cells with respect to the total number of cells. The proliferative index was calculated by multiplying the labeling index (LI) and the proliferation zone PZ (PZ ⫽ length of the area between the uppermost and lowest labeled cells). RESULTS: The study included 27 patients with H. pylori gastritis and 35 controls. Epithelial cell proliferation (LI) was greater with H. pylori infection than without in both the antrum and corpus (65 ⫾ 5 vs 91 ⫾ 8 in the antrum and 44 ⫾ 4 vs 72 ⫾ 8 in the corpus, for uninfected controls vs H. pylori gastritis, respectively) (p ⫽ 0.0001). In the antrum there was no significant decrease in epithelial cell proliferation after cure of the H. pylori infection despite follow-up for ⬎2 yr (labeling index ⫽ 83 ⫾ 10). In contrast, epithelial cell proliferation decreased in the corpus and became similar to that in controls after 7–13 months. CONCLUSIONS: Patients with H. pylori infection have sustained high epithelial cell proliferation in the antrum compared to that in uninfected subjects. A continued increase in proliferation in the antrum after cure of H. pylori infection suggests continuing damage. (Am J Gastroenterol 2000;95: 930 –935. © 2000 by Am. Coll. of Gastroenterology)
INTRODUCTION A wealth of data links Helicobacter pylori (H. pylori) infection with chronic atrophic gastritis and gastric adenocarcinoma (1). Gastric cancer is thought to be a multifactorial disease (2). Suggested mechanisms for carcinogenesis in H. pylori infection include the presence of mutagenic oxygen radicals produced by the inflammatory response to the in-
fection, reduction in mucosal antioxidants, and induction of a hyperproliferative state (3, 4). Studies of the effect of H. pylori infection on gastric mucosa proliferation and the results of its eradication on mucosal proliferation have yielded conflicting results (5–11). Although there is general agreement that H. pylori causes a hyperproliferative state of the gastric epithelium (7–10), not all authors agree that it is reversible after cure of the infection (5). Differences between studies can be attributed both to the convoluted shape of the gastric antral crypts rendering reproducible data difficult in nonoriented sections (12) and to differences in the duration of studies (8 –11). Our study was extended to almost 3 yr, and both the presence of proliferating cells and the length of the proliferating compartment were taken into account in assessing the proliferative index.
MATERIALS AND METHODS Patients Cases for evaluation of the effect of treatment of the H. pylori infection on mucosal proliferation were chosen from among individuals 1) who had three gastric biopsy specimens (two antral and one corpus) obtained from predetermined sites on the gastric lesser and greater curvature, 2) whose H. pylori infection had been successfully treated, and 3) for whom formalin-fixed, paraffinembedded tissues were available in the GI Mucosa Pathology laboratory. Jumbo forceps were used to obtain biopsies and each tissue sample was placed in a separate bottle containing 10% buffered formalin. At 2 and 4 wk after the end of antibiotic therapy, endoscopy with biopsy was repeated. Each successfully cured patient was asked to return for endoscopy at 6, 12, and 24 months for up to 3 yr. Controls were selected from a large group of normal volunteers who had participated in a study of the effect of H. pylori on the gastric mucosal histology. Gastric mucosal biopsies from the same predetermined sites were also taken.
AJG – April, 2000
Gastric Epithelial Cell Proliferation Increase
931
Table 1. Proliferation Indices for the Body Group
Median LI
p Value
Median PZ
Median PI
p Value
Control Day 0 2 wk 1–2 months 7–13 months 18–33 months
40 61 55 37 38 37
0.004 0.05 0.9 0.9 0.4
8 18 10 14 14 5
358 1117 529 424 528 233
⬍0.001 0.06 0.07 0.14 0.3
To isolate the group or groups that differ from the others, a multiple comparison procedure was used (the Mann-Whitney rank sum test) to compare all groups versus the control group. LI ⫽ labeling index; R ⫽ length of area between the uppermost and lowest labeled cells; PI ⫽ LI ⫻ PZ.
H. pylori Status All cases had been evaluated according to our routine protocol for H. pylori assessment (13), which includes a biopsy from the greater curve of the distal antrum and the midcorpus as well as a biopsy from the gastric angle. This approach has proven to provide close to a 100% accuracy in identifying H. pylori status (13, 14). In most cases, a lesser curve antral biopsy and a greater curve corpus biopsy were also taken for culture. The combination of negative culture, histology, and rapid urease tests at follow-up confirmed cure of the infection. Histology All specimens were fixed in 10% buffered formalin, routinely processed, and embedded in paraffin. Sections were cut at 5 m and stained with the Genta stain (15). The biopsy specimens measured on the glass slide ranged from 5 to 8 mm in length and measured approximately 1 mm in thickness. All biopsies were well oriented, contained surface mucosa and crypts, and extended to the muscularis mucosae. A visual analog scale from 0 (absent/normal) to 5 (maximal intensity) (16) was used by one pathologist (H.E.Z.) to grade the following histological features: H. pylori, intestinal metaplasia, neutrophils, and mononuclear cells. Immunohistochemistry and Cellular Proliferation Epithelial cell proliferation was determined immunohistochemically using the monoclonal antibody MIB1 (Amac, Westbrook, ME) against the nuclear proliferation antigen Ki-67. Tissues were placed in a microwave oven at high power in 0.1 mol/L citrate buffer pH 6.5, and treated four times for 5 min each. This was followed by a 2-h incubation at room temperature with the MIB1 antibody diluted 1:50. The immunoperoxidase reaction was developed using the
biotin-streptavidin immunoperoxidase method (Dako, Santa Barbara, CA). Two parameters were used to quantify the proliferative status of the gastric epithelium. The labeling index (LI) was calculated by counting the positively stained nuclei in four 20⫻ fields within the proliferating compartment and expressed as the ratio of the labeled nuclei divided by the total number of epithelial cell nuclei in the proliferating compartment. The proliferative zone (PZ) was defined as the length of the crypt comprised between the highest and the lowest labeled nucleus. Areas with intestinal metaplasia were excluded. Biopsies that did not include surface epithelium and muscularis mucosae were not used for the determination of the proliferative index. The proliferative index (PI) was obtained by multiplying the labeling index by the proliferative zone (PI ⫽ LI ⫻ PZ). Statistical Analysis All scores were entered into a database and analyzed using SigmaStat 2.03 (SPSS, Chicago, IL). All data were tested for normality of distribution. Pairwise comparisons were done using Kruskal-Wallis analysis of variance or, when appropriate, the Dunn’s method was used. Data are presented as mean ⫾ SEM. Statistical significance of differences and relationships was determined by p values of ⬍0.05.
RESULTS Patients Gastric biopsies from 27 patients (five women and 22 men, median age 54, range 29 –71 yr) with H. pylori infection and
Table 2. Proliferation Indices for the Antrum Group
Median LI
p Value
Median PZ
Median PI
p Value
Control Day 0 2 wk 1–2 months 7–13 months 18–33 months
64 75 101 83 85 72
0.04 0.005 0.05 ⬍0.001 0.07
12 21 19 21 21 21
742 1364 1701 1728 1795 1344
⬍0.01 ⬍0.001 ⬍0.001 ⬍0.001 ⬍0.008
To isolate the group or groups that differ from the others, a multiple comparison procedure was used (the Mann-Whitney rank sum test) to compare all groups versus the control group. LI ⫽ labeling index; R ⫽ length of area between the uppermost and lowest labeled cells; PI ⫽ LI ⫻ PZ.
932
El-Zimaity et al.
AJG – Vol. 95, No. 4, 2000
Figure 1. The mean proliferative index is shown for control (uninfected control subjects) and for patients with H. pylori infection both before (zero time) and after treatment of the H. pylori infection. The dotted line shows the mean proliferative index for the control group. (A) Gastric corpus. The mean proliferative index fell progressively after cure of the H. pylori infection and reached control levels by 13 months. (B) Gastric antrum. The proliferative index remained significantly greater after treatment of H. pylori infection compared to that in noninfected control subjects at all points during follow-up for up to 33 months.
Figure 2. MIB-1 immunostaining of the gastric mucosa expressing Ki-67 in the nuclei of the glandular necks (⫻200). (A) Normal gastric corpus. (B) Normal gastric antrum.
AJG – April, 2000
Gastric Epithelial Cell Proliferation Increase
933
Figure 3. MIB-1 immunostaining in active gastritis showing an increased number of Ki-67–positive nuclei and extension to superficial and deep areas of the mucosa (⫻200). (A) Gastric corpus. (B) Gastric antrum.
from 35 normal noninfected volunteers were studied. Each had gastric mucosal biopsies taken from the antrum and corpus. A total of 264 biopsies were available for the 27 patients; 76 biopsies were examined for the controls (41 antrum, 35 body). Not all controls had biopsies from all sites. Data are summarized in Tables 1 and 2.
At initial presentation, on a visual analog scale of 0 –5, the median for mononuclear cell infiltration was 3 in both the antrum and body. At 7–13 months posttherapy, the antrum still had a median of 3, but the body’s median for mononuclear cells dropped to 1.
Epithelial Cell Proliferation Pretreatment The LI was significantly higher in subjects with active H. pylori infection than in noninfected subjects. At initial presentation (pretreatment), the LI in the gastric corpus was 72 ⫾ 8 versus 44 ⫾ 4 for the noninfected control subjects (p ⫽ 0.0001) (Fig. 1A and Table 1). Likewise, in the antrum, the LI of H. pylori-infected subjects was 91 ⫾ 8 versus 65 ⫾ 5 for the noninfected controls (p ⫽ 0.0001) (Fig. 1B and Table 2). Specimens from 10, 16, 22, and 13 patients were available at 2 wk, 1–2 months, 7–13 months, and 18 –33 months posttherapy. After successful cure of the infection, the LI decreased in the corpus (45 ⫾ 6), becoming similar to that of uninfected individuals by 5–11 months posttherapy (Fig. 1A, Table 1). By contrast, there was no significant change in the PI in the antrum, even when follow-up was as long as 33 months after eradication (83 ⫾ 10) (Fig. 1B, Table 2).
DISCUSSION It is generally accepted that H. pylori infection increases gastric mucosal cell proliferation (4, 17, 18). This has been documented using proliferation markers such as the nucleolar organizing regions and the proliferating cell nuclear antigen. These markers have also been reported to decline significantly after cure of the H. pylori infection (10, 11). Most previous studies have been short (2– 4 wk posttherapy) and have been limited to the antrum (8 –11). Whereas most available data support the notion that H. pylori causes a reversible hyperproliferative state of the gastric epithelium (7–10), not all authors agree (5), and long-term studies have been lacking (5–11). We used MIB-1 antibody to detect an epitope of Ki-67, a reliable indicator of proliferating cells in gastric mucosal biopsies (Figs. 2A, B and 3A, B). The nuclear Ki-67 antigen
934
El-Zimaity et al.
is present in all stages of the cell cycle except G0 (19, 20). We used jumbo forceps to obtain large specimens that could be oriented correctly to provide a full-thickness view of the proliferative compartment from both the antrum and corpus of the stomach. We were also able to examine biopsies obtained before treatment and for up to 33 months after cure of the infection. Our study confirmed that the proliferation rate of the gastric epithelium is increased in the presence of H. pylori. Epithelial cell proliferation declined in the gastric corpus posttherapy but remained increased in the antrum even after 1–3 yr of follow-up. The fact that the increase in epithelial cell proliferation may be long-lived suggests that, in addition to the presence of the infection, other mechanisms come into play. The inflammatory reaction to H. pylori is slow to resolve completely. We have previously shown that an increase in the number of mast cells persists in the gastric mucosa after successful treatment of the infection (21). As H. pyloriassociated gastritis is usually more severe in the antrum than in the corpus, the sustained increase in mucosal proliferation may reflect the slower rate of recovery of chronic inflammation with continued stimulation by cytokines in the antrum (22, 23). Less clear is the reason why, in the absence of the pathogen, the inflammatory reaction does not subside for a long time after eradication. A possible explanation is that there may be continuous antigenic stimulation of the inflammatory process in the gastric mucosa caused by an autoimmune phenomenon (24). Previous studies have shown that monoclonal antibodies against H. pylori cross-react with the cells of the human gastric mucosa. Kirchner et al. (25) showed that antigastric autoantibodies, in particular the anticanalicular type, were significantly correlated with serum antibodies against H. pylori and with gastric atrophy. Such cross-reactivity was confirmed by adsorption of serum with H. pylori, which removed most of the reactivity (26). Those results suggest the possibility of molecular mimicry between H. pylori and the host tissue (25–31). Hyperproliferation may be initiated by an appropriate immune response to an enteric pathogen and may become sustained or slow to resolve because of molecular mimicry based on cross-reactivity to the gastric epithelium. Based on our results, for molecular mimicry to be an important mechanism in this case, it must be directed against a target present in the antrum and must be either not present or present to a negligible degree in the corpus. Thus, anticanalicular antibodies are unlikely candidates. Alternatively, mutations may have already accumulated, which would explain the sustained hyperproliferation. Whatever the mechanism, sustained hyperproliferation for up to 3 yr after eradication of the infection suggests that cure of the infection does not result in short-term, complete healing of gastritis and cessation of progression of mucosal injury. Further work will be necessary to assess when and if complete disappearance of inflammation and hyperproliferation indeed takes place after H. pylori eradication. Other
AJG – Vol. 95, No. 4, 2000
areas of research include the potential role of this persistent proliferative activity in the gastric mucosa in diffuse variant of gastric adenocarcinoma, in the absence of chronic atrophic gastritis and/or of H. pylori infection. Reprint requests and correspondence: Hala M.T. El-Zimaity, M.D., Director, Gastrointestinal Mucosa Pathology Laboratory, RM 3A352, VA Medical Center (111-D), 2002 Holcombe Boulevard, Houston, TX 77030. Received July 15, 1999; accepted Dec. 1, 1999.
REFERENCES 1. Schistosomes, liver flukes, and Helicobacter pylori. IARC Working Group on the Evaluation of Carcinogenic Risks to Humans. Lyon, France, 7–14 June 1994. IARC Monogr Eval Carcinog Risks Hum 1994;61:1–241. 2. Correa P. Clinical implications of recent developments in gastric cancer: Pathology and epidemiology. Semin Oncol 1985;12:2–10. 3. Correa P. Helicobacter pylori and gastric carcinogenesis. Am J Surg Pathol 1995;19(suppl 1):S37– 43. 4. Preston-Martin S, Pike MC, Ross RK, et al. Increased cell division as a cause of human cancer. Cancer Res 1990;50: 7415–21. 5. Chow KW, Bank S, Ahn J, et al. Helicobacter pylori infection does not increase gastric antrum mucosal cell proliferation. Am J Gastroenterol 1995;90:64 – 6. 6. Ricci V, Ciacci C, Zarrilli R, et al. Effect of Helicobacter pylori on gastric epithelial cell migration and proliferation in vitro: Role of VacA and CagA. Infect Immun 1996;64:2829 –33. 7. Fan XG, Kelleher D, Fan XJ, et al. Helicobacter pylori increases proliferation of gastric epithelial cells. Gut 1996;38: 19 –22. 8. Cahill RJ, Xia H, Kilgallen C, et al. Effect of eradication of Helicobacter pylori infection on gastric epithelial cell proliferation. Dig Dis Sci 1995;40:1627–31. 9. Lynch DA, Mapstone NP, Clarke AM, et al. Cell proliferation in the gastric corpus in Helicobacter pylori associated gastritis and after gastric resection. Gut 1995;36:351–3. 10. Brenes F, Ruiz B, Correa P, et al. Helicobacter pylori causes hyperproliferation of the gastric epithelium: Pre- and posteradication indices of proliferating cell nuclear antigen. Am J Gastroenterol 1993;88:1870 –5. 11. Correa P, Ruiz B, Shi TY, et al. Helicobacter pylori and nucleolar organizer regions in the gastric antral mucosa. Am J Clin Pathol 1994;101:656 – 60. 12. Havard TJ, Sarsfield P, Wotherspoon AC, et al. Increased gastric epithelial cell proliferation in Helicobacter pylori associated follicular gastritis. J Clin Pathol 1996;49:68 –71. 13. El-Zimaity HM, Al-Assi MT, Genta RM, et al. Confirmation of successful therapy of Helicobacter pylori infection: Number and site of biopsies or a rapid urease test. Am J Gastroenterol 1995;90:1962– 4. 14. El-Zimaity HMT, Graham DY. Evaluation of gastric mucosal biopsy site and number for identification of Helicobacter pylori or intestinal metaplasia: Role of the Sydney System. Hum Pathol 1999;30:72–7. 15. Genta RM, Robason GO, Graham DY. Simultaneous visualization of Helicobacter pylori and gastric morphology: A new stain. Hum Pathol 1994;25:221– 6. 16. El-Zimaity HM, Graham DY, Al-Assi MT, et al. Interobserver variation in the histopathological assessment of Helicobacter pylori gastritis (see comments). Hum Pathol 1996;27:35– 41.
AJG – April, 2000
17. Correa P. How does Helicobacter pylori infection increase gastric cancer risk? Eur J Gastroenterol Hepatol 1994;6: 1117– 8. 18. Ohyama S, Yonemura Y, Miyazaki I. Proliferative activity and malignancy in human gastric cancers. Cancer 1992;69:314 – 21. 19. Sawhney N, Hall PA. Ki67—Structure, funtion, and new antibodies. J Pathol 1992;168:161–2. 20. Lynch DA, Clarke AM, Jackson P, et al. Comparison of labelling by bromodeoxyuridine, MIB-1, and proliferating cell nuclear antigen in gastric mucosal biopsy specimens. J Clin Pathol 1994;47:122–5. 21. Nakajima S, Krishnan B, Ota H, et al. Mast cell involvement in gastritis with or without Helicobacter pylori infection. Gastroenterology 1997;113:746 –54. 22. Yamaoka Y, Kita M, Kodama T, et al. Induction of various cytokines and development of severe mucosal inflammation by cagA gene positive Helicobacter pylori strains (see comments). Gut 1997;41:442–51. 23. Karttunen RA, Karttunen TJ, Yousfi MM, et al. Expression of mRNA for interferon-gamma, interleukin-10, and interleukin-12 (p40) in normal gastric mucosa and in mucosa infected with Helicobacter pylori. Scand J Gastroenterol 1997;32:22–7. 24. Bamford KB, Hunt R, Muller M, et al. Gastric T cells and H. pylori: Regulation of pathogenesis and prevention. In: Ernst PB, Michetti P, Smith PD, eds. The immunobiology of H.
Gastric Epithelial Cell Proliferation Increase
25. 26.
27. 28. 29.
30.
31.
935
pylori: From pathogenesis to prevention. New York: Lippincott-Raven, 1997:227–35. Kirchner T, Steininger H, Faller G. Immunopathology of Helicobacter pylori gastritis. Digestion 1997;58:14 – 6. Negrini R, Savio A, Poiesi C, et al. Antigenic mimicry between Helicobacter pylori and gastric mucosa in the pathogenesis of body atrophic gastritis. Gastroenterology 1996;111: 655– 65. Telford JL, Covacci A, Rappuoli R, et al. Immunobiology of Helicobacter pylori infection. Curr Opin Immunol 1997;9: 498 –503. Negrini R, Lisato L, Zanella I, et al. Helicobacter pylori infection induces antibodies cross-reacting with human gastric mucosa. Gastroenterology 1991;101:437– 45. Faller G, Steininger H, Eck M, et al. Antigastric autoantibodies in Helicobacter pylori gastritis: Prevalence, in-situ binding sites and clues for clinical relevance. Virchows Arch 1996; 427:483– 6. Appelmelk BJ, Simoons-Smit I, Negrini R, et al. Potential role of molecular mimicry between Helicobacter pylori lipopolysaccharide and host Lewis blood group antigens in autoimmunity. Infect Immun 1996;64:2031– 40. Faller G, Steininger H, Kranzlein J, et al. Antigastric autoantibodies in Helicobacter pylori infection: Implications of histological and clinical parameters of gastritis. Gut 1997;41: 619 –23.
Vol. 95, No. 4, 2000 ISSN 0002-9270/00/$20.00 PII S0002-9270(00)00724-3
Sustained Increase in Gastric Antral Epithelial Cell Proliferation Despite Cure of Helicobacter pylori Infection Hala M. T. El-Zimaity, M.D., David Y. Graham, M.D., M.A.C.G., Robert M. Genta, M.D., F.A.C.G., and Juan Lechago, M.D., Ph.D. Departments of Pathology and Medicine, Veterans Affairs Medical Center, The Methodist Hospital, and Baylor College of Medicine, Houston, Texas
OBJECTIVE: Studies of the effect of Helicobacter pylori treatment on gastric mucosa proliferation have yielded inconsistent results. We compared gastric mucosa cell proliferation posttherapy and in uninfected controls. METHODS: Biopsies were obtained from patients with H. pylori infection before treatment and at intervals for up to 33 months. Epithelial cell proliferation was determined using Ki-67 immunostaining. The labeling index (LI) is the proportion of positively labeled cells with respect to the total number of cells. The proliferative index was calculated by multiplying the labeling index (LI) and the proliferation zone PZ (PZ ⫽ length of the area between the uppermost and lowest labeled cells). RESULTS: The study included 27 patients with H. pylori gastritis and 35 controls. Epithelial cell proliferation (LI) was greater with H. pylori infection than without in both the antrum and corpus (65 ⫾ 5 vs 91 ⫾ 8 in the antrum and 44 ⫾ 4 vs 72 ⫾ 8 in the corpus, for uninfected controls vs H. pylori gastritis, respectively) (p ⫽ 0.0001). In the antrum there was no significant decrease in epithelial cell proliferation after cure of the H. pylori infection despite follow-up for ⬎2 yr (labeling index ⫽ 83 ⫾ 10). In contrast, epithelial cell proliferation decreased in the corpus and became similar to that in controls after 7–13 months. CONCLUSIONS: Patients with H. pylori infection have sustained high epithelial cell proliferation in the antrum compared to that in uninfected subjects. A continued increase in proliferation in the antrum after cure of H. pylori infection suggests continuing damage. (Am J Gastroenterol 2000;95: 930 –935. © 2000 by Am. Coll. of Gastroenterology)
INTRODUCTION A wealth of data links Helicobacter pylori (H. pylori) infection with chronic atrophic gastritis and gastric adenocarcinoma (1). Gastric cancer is thought to be a multifactorial disease (2). Suggested mechanisms for carcinogenesis in H. pylori infection include the presence of mutagenic oxygen radicals produced by the inflammatory response to the in-
fection, reduction in mucosal antioxidants, and induction of a hyperproliferative state (3, 4). Studies of the effect of H. pylori infection on gastric mucosa proliferation and the results of its eradication on mucosal proliferation have yielded conflicting results (5–11). Although there is general agreement that H. pylori causes a hyperproliferative state of the gastric epithelium (7–10), not all authors agree that it is reversible after cure of the infection (5). Differences between studies can be attributed both to the convoluted shape of the gastric antral crypts rendering reproducible data difficult in nonoriented sections (12) and to differences in the duration of studies (8 –11). Our study was extended to almost 3 yr, and both the presence of proliferating cells and the length of the proliferating compartment were taken into account in assessing the proliferative index.
MATERIALS AND METHODS Patients Cases for evaluation of the effect of treatment of the H. pylori infection on mucosal proliferation were chosen from among individuals 1) who had three gastric biopsy specimens (two antral and one corpus) obtained from predetermined sites on the gastric lesser and greater curvature, 2) whose H. pylori infection had been successfully treated, and 3) for whom formalin-fixed, paraffinembedded tissues were available in the GI Mucosa Pathology laboratory. Jumbo forceps were used to obtain biopsies and each tissue sample was placed in a separate bottle containing 10% buffered formalin. At 2 and 4 wk after the end of antibiotic therapy, endoscopy with biopsy was repeated. Each successfully cured patient was asked to return for endoscopy at 6, 12, and 24 months for up to 3 yr. Controls were selected from a large group of normal volunteers who had participated in a study of the effect of H. pylori on the gastric mucosal histology. Gastric mucosal biopsies from the same predetermined sites were also taken.
AJG – April, 2000
Gastric Epithelial Cell Proliferation Increase
931
Table 1. Proliferation Indices for the Body Group
Median LI
p Value
Median PZ
Median PI
p Value
Control Day 0 2 wk 1–2 months 7–13 months 18–33 months
40 61 55 37 38 37
0.004 0.05 0.9 0.9 0.4
8 18 10 14 14 5
358 1117 529 424 528 233
⬍0.001 0.06 0.07 0.14 0.3
To isolate the group or groups that differ from the others, a multiple comparison procedure was used (the Mann-Whitney rank sum test) to compare all groups versus the control group. LI ⫽ labeling index; R ⫽ length of area between the uppermost and lowest labeled cells; PI ⫽ LI ⫻ PZ.
H. pylori Status All cases had been evaluated according to our routine protocol for H. pylori assessment (13), which includes a biopsy from the greater curve of the distal antrum and the midcorpus as well as a biopsy from the gastric angle. This approach has proven to provide close to a 100% accuracy in identifying H. pylori status (13, 14). In most cases, a lesser curve antral biopsy and a greater curve corpus biopsy were also taken for culture. The combination of negative culture, histology, and rapid urease tests at follow-up confirmed cure of the infection. Histology All specimens were fixed in 10% buffered formalin, routinely processed, and embedded in paraffin. Sections were cut at 5 m and stained with the Genta stain (15). The biopsy specimens measured on the glass slide ranged from 5 to 8 mm in length and measured approximately 1 mm in thickness. All biopsies were well oriented, contained surface mucosa and crypts, and extended to the muscularis mucosae. A visual analog scale from 0 (absent/normal) to 5 (maximal intensity) (16) was used by one pathologist (H.E.Z.) to grade the following histological features: H. pylori, intestinal metaplasia, neutrophils, and mononuclear cells. Immunohistochemistry and Cellular Proliferation Epithelial cell proliferation was determined immunohistochemically using the monoclonal antibody MIB1 (Amac, Westbrook, ME) against the nuclear proliferation antigen Ki-67. Tissues were placed in a microwave oven at high power in 0.1 mol/L citrate buffer pH 6.5, and treated four times for 5 min each. This was followed by a 2-h incubation at room temperature with the MIB1 antibody diluted 1:50. The immunoperoxidase reaction was developed using the
biotin-streptavidin immunoperoxidase method (Dako, Santa Barbara, CA). Two parameters were used to quantify the proliferative status of the gastric epithelium. The labeling index (LI) was calculated by counting the positively stained nuclei in four 20⫻ fields within the proliferating compartment and expressed as the ratio of the labeled nuclei divided by the total number of epithelial cell nuclei in the proliferating compartment. The proliferative zone (PZ) was defined as the length of the crypt comprised between the highest and the lowest labeled nucleus. Areas with intestinal metaplasia were excluded. Biopsies that did not include surface epithelium and muscularis mucosae were not used for the determination of the proliferative index. The proliferative index (PI) was obtained by multiplying the labeling index by the proliferative zone (PI ⫽ LI ⫻ PZ). Statistical Analysis All scores were entered into a database and analyzed using SigmaStat 2.03 (SPSS, Chicago, IL). All data were tested for normality of distribution. Pairwise comparisons were done using Kruskal-Wallis analysis of variance or, when appropriate, the Dunn’s method was used. Data are presented as mean ⫾ SEM. Statistical significance of differences and relationships was determined by p values of ⬍0.05.
RESULTS Patients Gastric biopsies from 27 patients (five women and 22 men, median age 54, range 29 –71 yr) with H. pylori infection and
Table 2. Proliferation Indices for the Antrum Group
Median LI
p Value
Median PZ
Median PI
p Value
Control Day 0 2 wk 1–2 months 7–13 months 18–33 months
64 75 101 83 85 72
0.04 0.005 0.05 ⬍0.001 0.07
12 21 19 21 21 21
742 1364 1701 1728 1795 1344
⬍0.01 ⬍0.001 ⬍0.001 ⬍0.001 ⬍0.008
To isolate the group or groups that differ from the others, a multiple comparison procedure was used (the Mann-Whitney rank sum test) to compare all groups versus the control group. LI ⫽ labeling index; R ⫽ length of area between the uppermost and lowest labeled cells; PI ⫽ LI ⫻ PZ.
932
El-Zimaity et al.
AJG – Vol. 95, No. 4, 2000
Figure 1. The mean proliferative index is shown for control (uninfected control subjects) and for patients with H. pylori infection both before (zero time) and after treatment of the H. pylori infection. The dotted line shows the mean proliferative index for the control group. (A) Gastric corpus. The mean proliferative index fell progressively after cure of the H. pylori infection and reached control levels by 13 months. (B) Gastric antrum. The proliferative index remained significantly greater after treatment of H. pylori infection compared to that in noninfected control subjects at all points during follow-up for up to 33 months.
Figure 2. MIB-1 immunostaining of the gastric mucosa expressing Ki-67 in the nuclei of the glandular necks (⫻200). (A) Normal gastric corpus. (B) Normal gastric antrum.
AJG – April, 2000
Gastric Epithelial Cell Proliferation Increase
933
Figure 3. MIB-1 immunostaining in active gastritis showing an increased number of Ki-67–positive nuclei and extension to superficial and deep areas of the mucosa (⫻200). (A) Gastric corpus. (B) Gastric antrum.
from 35 normal noninfected volunteers were studied. Each had gastric mucosal biopsies taken from the antrum and corpus. A total of 264 biopsies were available for the 27 patients; 76 biopsies were examined for the controls (41 antrum, 35 body). Not all controls had biopsies from all sites. Data are summarized in Tables 1 and 2.
At initial presentation, on a visual analog scale of 0 –5, the median for mononuclear cell infiltration was 3 in both the antrum and body. At 7–13 months posttherapy, the antrum still had a median of 3, but the body’s median for mononuclear cells dropped to 1.
Epithelial Cell Proliferation Pretreatment The LI was significantly higher in subjects with active H. pylori infection than in noninfected subjects. At initial presentation (pretreatment), the LI in the gastric corpus was 72 ⫾ 8 versus 44 ⫾ 4 for the noninfected control subjects (p ⫽ 0.0001) (Fig. 1A and Table 1). Likewise, in the antrum, the LI of H. pylori-infected subjects was 91 ⫾ 8 versus 65 ⫾ 5 for the noninfected controls (p ⫽ 0.0001) (Fig. 1B and Table 2). Specimens from 10, 16, 22, and 13 patients were available at 2 wk, 1–2 months, 7–13 months, and 18 –33 months posttherapy. After successful cure of the infection, the LI decreased in the corpus (45 ⫾ 6), becoming similar to that of uninfected individuals by 5–11 months posttherapy (Fig. 1A, Table 1). By contrast, there was no significant change in the PI in the antrum, even when follow-up was as long as 33 months after eradication (83 ⫾ 10) (Fig. 1B, Table 2).
DISCUSSION It is generally accepted that H. pylori infection increases gastric mucosal cell proliferation (4, 17, 18). This has been documented using proliferation markers such as the nucleolar organizing regions and the proliferating cell nuclear antigen. These markers have also been reported to decline significantly after cure of the H. pylori infection (10, 11). Most previous studies have been short (2– 4 wk posttherapy) and have been limited to the antrum (8 –11). Whereas most available data support the notion that H. pylori causes a reversible hyperproliferative state of the gastric epithelium (7–10), not all authors agree (5), and long-term studies have been lacking (5–11). We used MIB-1 antibody to detect an epitope of Ki-67, a reliable indicator of proliferating cells in gastric mucosal biopsies (Figs. 2A, B and 3A, B). The nuclear Ki-67 antigen
934
El-Zimaity et al.
is present in all stages of the cell cycle except G0 (19, 20). We used jumbo forceps to obtain large specimens that could be oriented correctly to provide a full-thickness view of the proliferative compartment from both the antrum and corpus of the stomach. We were also able to examine biopsies obtained before treatment and for up to 33 months after cure of the infection. Our study confirmed that the proliferation rate of the gastric epithelium is increased in the presence of H. pylori. Epithelial cell proliferation declined in the gastric corpus posttherapy but remained increased in the antrum even after 1–3 yr of follow-up. The fact that the increase in epithelial cell proliferation may be long-lived suggests that, in addition to the presence of the infection, other mechanisms come into play. The inflammatory reaction to H. pylori is slow to resolve completely. We have previously shown that an increase in the number of mast cells persists in the gastric mucosa after successful treatment of the infection (21). As H. pyloriassociated gastritis is usually more severe in the antrum than in the corpus, the sustained increase in mucosal proliferation may reflect the slower rate of recovery of chronic inflammation with continued stimulation by cytokines in the antrum (22, 23). Less clear is the reason why, in the absence of the pathogen, the inflammatory reaction does not subside for a long time after eradication. A possible explanation is that there may be continuous antigenic stimulation of the inflammatory process in the gastric mucosa caused by an autoimmune phenomenon (24). Previous studies have shown that monoclonal antibodies against H. pylori cross-react with the cells of the human gastric mucosa. Kirchner et al. (25) showed that antigastric autoantibodies, in particular the anticanalicular type, were significantly correlated with serum antibodies against H. pylori and with gastric atrophy. Such cross-reactivity was confirmed by adsorption of serum with H. pylori, which removed most of the reactivity (26). Those results suggest the possibility of molecular mimicry between H. pylori and the host tissue (25–31). Hyperproliferation may be initiated by an appropriate immune response to an enteric pathogen and may become sustained or slow to resolve because of molecular mimicry based on cross-reactivity to the gastric epithelium. Based on our results, for molecular mimicry to be an important mechanism in this case, it must be directed against a target present in the antrum and must be either not present or present to a negligible degree in the corpus. Thus, anticanalicular antibodies are unlikely candidates. Alternatively, mutations may have already accumulated, which would explain the sustained hyperproliferation. Whatever the mechanism, sustained hyperproliferation for up to 3 yr after eradication of the infection suggests that cure of the infection does not result in short-term, complete healing of gastritis and cessation of progression of mucosal injury. Further work will be necessary to assess when and if complete disappearance of inflammation and hyperproliferation indeed takes place after H. pylori eradication. Other
AJG – Vol. 95, No. 4, 2000
areas of research include the potential role of this persistent proliferative activity in the gastric mucosa in diffuse variant of gastric adenocarcinoma, in the absence of chronic atrophic gastritis and/or of H. pylori infection. Reprint requests and correspondence: Hala M.T. El-Zimaity, M.D., Director, Gastrointestinal Mucosa Pathology Laboratory, RM 3A352, VA Medical Center (111-D), 2002 Holcombe Boulevard, Houston, TX 77030. Received July 15, 1999; accepted Dec. 1, 1999.
REFERENCES 1. Schistosomes, liver flukes, and Helicobacter pylori. IARC Working Group on the Evaluation of Carcinogenic Risks to Humans. Lyon, France, 7–14 June 1994. IARC Monogr Eval Carcinog Risks Hum 1994;61:1–241. 2. Correa P. Clinical implications of recent developments in gastric cancer: Pathology and epidemiology. Semin Oncol 1985;12:2–10. 3. Correa P. Helicobacter pylori and gastric carcinogenesis. Am J Surg Pathol 1995;19(suppl 1):S37– 43. 4. Preston-Martin S, Pike MC, Ross RK, et al. Increased cell division as a cause of human cancer. Cancer Res 1990;50: 7415–21. 5. Chow KW, Bank S, Ahn J, et al. Helicobacter pylori infection does not increase gastric antrum mucosal cell proliferation. Am J Gastroenterol 1995;90:64 – 6. 6. Ricci V, Ciacci C, Zarrilli R, et al. Effect of Helicobacter pylori on gastric epithelial cell migration and proliferation in vitro: Role of VacA and CagA. Infect Immun 1996;64:2829 –33. 7. Fan XG, Kelleher D, Fan XJ, et al. Helicobacter pylori increases proliferation of gastric epithelial cells. Gut 1996;38: 19 –22. 8. Cahill RJ, Xia H, Kilgallen C, et al. Effect of eradication of Helicobacter pylori infection on gastric epithelial cell proliferation. Dig Dis Sci 1995;40:1627–31. 9. Lynch DA, Mapstone NP, Clarke AM, et al. Cell proliferation in the gastric corpus in Helicobacter pylori associated gastritis and after gastric resection. Gut 1995;36:351–3. 10. Brenes F, Ruiz B, Correa P, et al. Helicobacter pylori causes hyperproliferation of the gastric epithelium: Pre- and posteradication indices of proliferating cell nuclear antigen. Am J Gastroenterol 1993;88:1870 –5. 11. Correa P, Ruiz B, Shi TY, et al. Helicobacter pylori and nucleolar organizer regions in the gastric antral mucosa. Am J Clin Pathol 1994;101:656 – 60. 12. Havard TJ, Sarsfield P, Wotherspoon AC, et al. Increased gastric epithelial cell proliferation in Helicobacter pylori associated follicular gastritis. J Clin Pathol 1996;49:68 –71. 13. El-Zimaity HM, Al-Assi MT, Genta RM, et al. Confirmation of successful therapy of Helicobacter pylori infection: Number and site of biopsies or a rapid urease test. Am J Gastroenterol 1995;90:1962– 4. 14. El-Zimaity HMT, Graham DY. Evaluation of gastric mucosal biopsy site and number for identification of Helicobacter pylori or intestinal metaplasia: Role of the Sydney System. Hum Pathol 1999;30:72–7. 15. Genta RM, Robason GO, Graham DY. Simultaneous visualization of Helicobacter pylori and gastric morphology: A new stain. Hum Pathol 1994;25:221– 6. 16. El-Zimaity HM, Graham DY, Al-Assi MT, et al. Interobserver variation in the histopathological assessment of Helicobacter pylori gastritis (see comments). Hum Pathol 1996;27:35– 41.
AJG – April, 2000
17. Correa P. How does Helicobacter pylori infection increase gastric cancer risk? Eur J Gastroenterol Hepatol 1994;6: 1117– 8. 18. Ohyama S, Yonemura Y, Miyazaki I. Proliferative activity and malignancy in human gastric cancers. Cancer 1992;69:314 – 21. 19. Sawhney N, Hall PA. Ki67—Structure, funtion, and new antibodies. J Pathol 1992;168:161–2. 20. Lynch DA, Clarke AM, Jackson P, et al. Comparison of labelling by bromodeoxyuridine, MIB-1, and proliferating cell nuclear antigen in gastric mucosal biopsy specimens. J Clin Pathol 1994;47:122–5. 21. Nakajima S, Krishnan B, Ota H, et al. Mast cell involvement in gastritis with or without Helicobacter pylori infection. Gastroenterology 1997;113:746 –54. 22. Yamaoka Y, Kita M, Kodama T, et al. Induction of various cytokines and development of severe mucosal inflammation by cagA gene positive Helicobacter pylori strains (see comments). Gut 1997;41:442–51. 23. Karttunen RA, Karttunen TJ, Yousfi MM, et al. Expression of mRNA for interferon-gamma, interleukin-10, and interleukin-12 (p40) in normal gastric mucosa and in mucosa infected with Helicobacter pylori. Scand J Gastroenterol 1997;32:22–7. 24. Bamford KB, Hunt R, Muller M, et al. Gastric T cells and H. pylori: Regulation of pathogenesis and prevention. In: Ernst PB, Michetti P, Smith PD, eds. The immunobiology of H.
Gastric Epithelial Cell Proliferation Increase
25. 26.
27. 28. 29.
30.
31.
935
pylori: From pathogenesis to prevention. New York: Lippincott-Raven, 1997:227–35. Kirchner T, Steininger H, Faller G. Immunopathology of Helicobacter pylori gastritis. Digestion 1997;58:14 – 6. Negrini R, Savio A, Poiesi C, et al. Antigenic mimicry between Helicobacter pylori and gastric mucosa in the pathogenesis of body atrophic gastritis. Gastroenterology 1996;111: 655– 65. Telford JL, Covacci A, Rappuoli R, et al. Immunobiology of Helicobacter pylori infection. Curr Opin Immunol 1997;9: 498 –503. Negrini R, Lisato L, Zanella I, et al. Helicobacter pylori infection induces antibodies cross-reacting with human gastric mucosa. Gastroenterology 1991;101:437– 45. Faller G, Steininger H, Eck M, et al. Antigastric autoantibodies in Helicobacter pylori gastritis: Prevalence, in-situ binding sites and clues for clinical relevance. Virchows Arch 1996; 427:483– 6. Appelmelk BJ, Simoons-Smit I, Negrini R, et al. Potential role of molecular mimicry between Helicobacter pylori lipopolysaccharide and host Lewis blood group antigens in autoimmunity. Infect Immun 1996;64:2031– 40. Faller G, Steininger H, Kranzlein J, et al. Antigastric autoantibodies in Helicobacter pylori infection: Implications of histological and clinical parameters of gastritis. Gut 1997;41: 619 –23.