C 2005) Digestive Diseases and Sciences, Vol. 50, No. 6 (June 2005), pp. 1078–1086 ( DOI: 10.1007/s10620-005-2708-4
Infection with Helicobacter pylori Affects All Major Secretory Cell Populations in the Human Antrum JEROEN H. B. VAN DE BOVENKAMP, PhD, ANITA M. KORTELAND-VAN MALE, BSc, ¨ HANS A. BULLER, MD, PhD, ALEXANDRA W. C. EINERHAND, PhD, and JAN DEKKER, PhD
We have investigated how gastric H. pylori infection affects antrum secretory cell types by studying the expression of secretory proteins in antrum epithelium. Antrum biopsy specimens were prospectively collected from 102 individuals (49 H. pylori-infected). Immunohistochemistry was performed for secretory mucins (MUC5AC, MUC5B, MUC6), Trefoil factor family (TFF)-peptides (TFF1, TFF2), endocrine peptides (gastrin, chromogranin A), and proliferating cells (Ki-67). Protein expression was quantified morphometrically. H. pylori infection was significantly correlated to mucosal inflammation and to epithelial atrophy and proliferation. In H. pylori-infected patients the number of proliferating cells increased significantly, and the zone of proliferating cells shifted toward the surface epithelium of the antral glands. Infection was correlated with decreased MUC5AC, TFF1, and TFF2 expression and increased MUC6 and MUC5B expression. Endocrine cells expressing chromagranin A and gastrin shifted toward the surface epithelium of the antral glands in H. pylori-infected patients. H. pylori infection and concomitant inflammation induced increased epithelial proliferation and triggered coordinate deregulation of secretory cell populations in the antrum. In particular, infection led to a coordinated increase in cells expressing MUC6 and MUC5B at the expense of MUC5AC-producing cells. KEY WORDS: H. pylori; inflammation; mucin; proliferation; stomach; TFF-peptides.
Since the discovery of Helicobacter pylori (1, 2), chronic infection with this bacterium is identified as the major etiologic factor in gastritis, gastric ulcers, gastric atrophy, and gastric carcinoma (3, 4). Following gastric infection by H. pylori both proliferation and apoptosis in the epithelium increase (5, 6), altering the turnover of the epithelium. Gastric epithelial turnover is a dynamic process, characterized by continuous cell apoptosis, which is normally balanced by cell proliferation and concomitant differentiation of daughter cells in to various, mostly secretory, cell Manuscript received February 15, 2004; accepted July 2, 2004. From the Laboratory of Pediatrics and Sophia Children’s Hospital, Erasmus MC, Rotterdam, The Netherlands. Address for reprint requests: Jan Dekker, PhD, Rudolf Magnus Institute of Neuroscience, Universiteitsweg 100, Room STR5-203, 3584 CG Utrecht, The Netherlands;
[email protected].
1078
types. Disturbance of the balance between apoptosis and proliferation could interfere with normal mucosal homeostasis and could lead to shifts in the various cell populations, altering the functionality of the gastric epithelium. In this study we investigated changes in secretory cell populations of the human antrum, by studying three separate groups of secretory proteins of the antrum in relation to gastric H. pylori infection. The first group consists of secretory mucins that are expressed in the stomach (MUC5AC, MUC5B, and MUC6). Secretory, gel-forming mucin is the most important structural component of the mucus gel layer and plays an important role in the protection of the underlying epithelium (7). Of the gastric mucins, MUC5AC is thought to play a role in the adhesion of H. pylori to the gastric epithelium (8–10). The second group of secretory proteins consists of Trefoil factor Digestive Diseases and Sciences, Vol. 50, No. 6 (June 2005)
C 2005 Springer Science+Business Media, Inc. 0163-2116/05/0600-1078/0
H. pylori AFFECTS SECRETORY CELLS
family (TFF)-peptides (formerly P-domain peptides or Trefoil factor (11), which are often expressed in close association with secretory mucins (12). In the human stomach two TFF-peptides have been identified: TFF1 (formerly pS2) and TFF2 (formerly hSP). TFF-peptides very likely play a role in gastrointestinal defense and repair (13). The third group of secretory proteins comprises endocrine cell markers, gastrin and chromogranin A (CGA). Gastrin is produced as a prohormone by G-cells located within the gastric antrum. The most important role of gastrin is the regulation of gastric acid secretion. In the human stomach, enterochromaffin-like (ECL) cells of the corpus have been identified as the main source for CGA expression (14). Besides ECL cells, CGA expression has been demonstrated in antral G-cells (15). CGA is thought to influence granule stability, prohormone processing, and peptide sorting into the regulated secretory pathway (see Hocker et al. (16) and references therein). As the fourth group of epithelial cells we investigated the effect of H. pylori infection on proliferating cells, which were detected using a specific antibody to Ki-67. This nuclear protein is expressed in proliferating cells (G1, S, G2, and M phase) but not in resting cells (G0 phase) (17). Taking onto account these marker proteins we were able to identify all cells secreting these major secretory products of the human antrum. Thus, we were able to study any coordinate changes in expression of the secretory proteins, in particular, any shifts in the population producing these secretory proteins. As most studies so far have concentrated on just one or a few of the many products of the antrum, the underlying consequences of H. pylori infection for the various cell populations may be not revealed. In this study we demonstrate the close correlations of H. pylori infection and altered proliferation with the expression of the various secretory proteins of the antrum and shifts in the cell populations producing these proteins. From this, it appears that H. pylori induces a coordinate rearrangement of all secretory cell types in the infected antrum.
MATERIALS AND METHODS Patients and Tissue. Biopsy specimens of the antrum of 102 patients were collected per endoscopy, as part of a prospective study on the occurrence of H. pylori in adult patients with chronic upper abdominal complaints. Patients were asked to abstain from antiacids drugs for at least 1 week prior to endoscopy. None of the patients received specific anti-H. pylori therapy prior to examination. Biopsy specimens were collected at the Academic Medical Center in Amsterdam, by permission of the Medical Ethics Committee and with informed consent of the patients. Two biopsy specimens were taken from the antrum per patient, immediately fixed in phosphate-buffered saline (PBS)-buffered 4% (wt/vol) paraformaldehyde solution for 4 hr, and then processed into paraffin blocks according to standard procedures. Clinical data were recovered from standardized pathologist’s records, based on at least two separately collected biopsy specimens from closely adjacent gastric mucosa, using the criteria of the Sydney classification (18). H. pylori infection was recorded based on standard histological staining and microbiology. H. pylori infection was confirmed in each patient using anti-H. pylori antibodies, as described previously (8). Active inflammation was recorded semiquantitatively: 0, no active inflammation; 1, mild active inflammation; 2, moderately active inflammation; 3, marked active inflammation. Atrophy was also recorded semiquantitatively: 0, absent; 1, mild; 2, moderate; 3, marked. Immunohistochemistry. Polyclonal antibodies raised in rabbits were used: anti-MUC5B, anti-MUC6, anti-TFF1, antiTFF2, antigastrin, and anti-CGA (Table 1). In addition, mouse monoclonal antibodies were used: anti-MUC5AC and anti-Ki-67 (Table 1). Tissue sections were deparaffinized through three changes of xylene and then rehydrated through a series of decreasing concentrations of ethanol solution to distilled water. Endogenous peroxidase activity was inactivated in 3% (vol/vol) hydrogen peroxide in PBS for 30 min and washed in PBS for 5 min. Antigen retrieval was performed by heating the sections for 10 min at 100◦ C in 10 mM citrate buffer, pH 6.0, and then left to cool at room temperature for 20 min. Sections were washed three times for 5 min in PBS and incubated with 1% (wt/vol) blocking agent (Boehringer, Mannheim, Germany) in PBS for 30 min. Primary antibodies were diluted in PBS (Table 1) and incubated with the tissues for 16 hr at 4◦ C. Slides were washed three times for 5 min in PBS, followed by incubation with biotinylated secondary antibodies and avidin–biotin peroxidase complex (Vectastain ABCkit, Vector Laboratories, Burlingame, UK) according to the manufacturer’s protocol. Staining was performed using 0.5 mg/ml 3,3 -diaminobenzidine/0.03% (vol/vol)
TABLE 1. CHARACTERISTICS OF ANTIBODIES AND DILUTIONS USED FOR IMMUNOHISTOCHEMICAL METHODS Name in this study Anti-MUC5AC Anti-MUC5B Anti-MUC6 Anti-TFF1 Anti-TFF2 Anti-Gastrin Anti-CGA Anti-Ki-67 ∗ Variable
Original name in reference
Epitope
Dilution
Reference
45M1 Anti-BGBM M6.1 Anti-pS2 Anti-hSP A0568 A0430 MiB-1
Peptide domain Deglycosylated polypeptide Peptide in VNTR∗ domain C-terminal peptide C-terminal peptide Gastrin-17 20-kD C-terminal peptide Peptide
1:50 1:3000 1:200 1:4000 1:2000 1:2000 1:1000 1:1000
39 40 41 42 43 DAKO, Glostrup, DK DAKO, Glostrup, DK 17
number of tandem repeats.
Digestive Diseases and Sciences, Vol. 50, No. 6 (June 2005)
1079
VAN DE BOVENKAMP ET AL. hydrogen peroxide in imidazole (30 mM)/EDTA (1 mM), pH 7.0. Sections were rinsed in water, dehydrated through a series of increasing concentrations of ethanol solutions, and mounted under coverslips. Control stainings were performed, leaving each of the primary antibodies out of the procedure, resulting in absence of staining. Semiquantitative Scoring of Immunohistological Staining. The extent of expression of each protein was determined semiquantitatively on serial sections. Two independent observers, who were blinded to the clinical status of the patients, determined the expression of each protein. The surface-to-gland axis and the expression patterns of the respective proteins were determined in at least two locations within each tissue section. In the case of disagreement among the observers the average scores were taken. Biopsy specimens were cut perpendicular to the mucosal surface, and only those sections were assessed where the whole surface-to-gland axis was fully visible. We used a relative scale for the protein expression along this axis, since, due to either hyperproliferation or atrophy in particular specimens, the surface-to-gland axis varied in absolute length. The microscopic images of the sections of the biopsy specimens were projected through a CCD videocamera onto a monitor, the total surfaceto-gland axis was measured manually, and then this axis was divided into 10 regions of equal length. Expression of the individual proteins was assessed by identifying which of the 10 regions along the surface-to-gland axis expressed the respective proteins (Figure 1). Note that the intensity of staining was not accounted for in these measurements; the absence or presence of a protein was scored per defined region. The number of regions containing positively stained cells, as well as their position along the pit–gland axis, was taken as a measure of the extent of
Fig 1. Schematic representation of the quantitation of protein expression in the antrum. The scheme represents three serial sections of an antrum biopsy specimen, indicating the typical protein expression patterns of MUC5AC, MUC6, and gastrin (in black). Microscopic images of these sections were projected through a CCD camera onto a monitor. The total surface-to-gland axis was measured manually using a ruler, and this axis was divided into 10 regions of equal length (regions 1–10), with region 1 representing the surface epithelium and region 10 representing the deepest glandular structures by our definition. Expression of the individual proteins was assessed by identifying the defined regions in which the respective proteins were expressed. In our example MUC5AC is expressed in regions 1–4, and MUC6 is expressed in regions 5–10, whereas gastrin is expressed in regions 7–9. The number of regions staining positively along the surface-to-gland axis was taken as a measure of the extent of expression; in our example, MUC5AC, MUC6, and gastrin score 4, 6, and 3 regions, respectively.
1080
expression (Figure 1). Region 1 denotes by definition the most superficial region along the axis, whereas region 10 corresponds to the deepest parts of the glands. Statistics. The χ 2 test or Fisher’s exact test was used to analyze differences in frequencies. Spearman’s correlation test was used to analyze correlations between parameters. Statistical significance was defined at P < 0.05.
RESULTS Pathology. Antrum biopsies were prospectively collected from 102 patients, of whom 49 appeared to suffer from H. pylori infection at the time of endoscopy. We analyzed the correlations between H. pylori infection and the extent of inflammation, atrophy, and proliferation in the antrum mucosa. Proliferation was detected using an antiKi-67 antibody, and expression of Ki-67 was quantified as the number of regions showing positive expression along the surface-to-gland axis, whereas inflammation and atrophy were scored according to the Sidney classification. Figure 2 shows the comparison of H. pylori-infected and noninfected individuals. There was a strongly positive and highly significant correlation between H. pylori infection and active inflammation of the mucosa (Figure 2A). H. pylori infection also correlated statistically significantly with mucosal atrophy and cell proliferation in the epithelium (Figures 2B and C). The latter correlations with H. pylori infection were less strong than with inflammation, implying that also a number of noninfected
Fig 2. H. pylori infection is correlated with increased mucosal inflammation, epithelial atrophy, and proliferation. Active mucosal inflammation (A) and mucosal atrophy (B) were scored according to Sydney classifications. The presence of H. pylori was scored as present or absent. Epithelial proliferation was determined by immunohistochemical staining for Ki-67 (C). The number of regions along the surface-to-gland axis containing Ki-67-positive cells (indicated on the X axis) was taken as a measure of proliferation (Figure 1). The percentage of H. pylori-infected patients is indicated in black; the percentage of H. pylori-negative individuals is indicated in white. The number of patients in each group is indicated above each bar. Spearman’s correlation test was used to determine significant correlations. H. pylori infection was positively correlated with inflammation (ρ = 0.907, P < 0.0005), atrophy (ρ = 0.189, P < 0.05), and proliferation (ρ = 0.358, P < 0.0005). Digestive Diseases and Sciences, Vol. 50, No. 6 (June 2005)
H. pylori AFFECTS SECRETORY CELLS
Fig 3. Active mucosal inflammation correlates with increased epithelial proliferation and atrophy. Active mucosal inflammation and mucosal atrophy were scored according to Sydney classifications. Epithelial proliferation was determined by immunohistochemical staining for Ki-67 (A). The number of regions along the surface-to-gland axis containing Ki-67-positive cells (indicated on the X axis) was taken as a measure of proliferation (see Materials and Methods). Atrophy (B) was scored as absent, mild, or moderate. The extent of inflammation is indicated: gray, no inflammation; cross-hatched, mild inflammation; white, moderate inflammation; and black, marked inflammation. The number of patients in each group is indicated above each bar. Spearman’s correlation test was used to determine significant correlations. Inflammation was positively and statistically significantly correlated with proliferation (A: ρ = 0.332, P < 0.001) and atrophy (B: ρ = 0.235, P = 0.02).
individuals showed elevated atrophy and proliferation scores. The tissue of these patients, however, showed no other pathology. Independent of H. pylori infection, the correlations of mucosal inflammation with epithelial proliferation and mucosal atrophy were determined (Figure 3), and both appeared positive and statistically significant. Expression of Secretory Proteins and Ki-67 During H. pylori Infection. Sections of the biopsy specimens of the 102 patients were stained for the presence of MUC5AC, MUC5B, MUC6, TFF1, TFF2, gastrin, and CGA. Expression of each protein was identified in each patient. Figure 4 shows representative examples of an H. pylori-infected and a noninfected individual. MUC5AC was expressed in the surface epithelium, in the epithelium of the gastric pits, and, to a variable extent, in the glandular structures. In contrast, both MUC5B and MUC6 were only expressed in the glands, and never at the surface, in a manner complementary to the MUC5AC expression pattern. TFF1 was expressed in the surface and pit epithelium and, to a variable extent, deeper into the glands. TFF2 was expressed in the entire epithelium in nearly all patients. Gastrin and CGA were expressed in Digestive Diseases and Sciences, Vol. 50, No. 6 (June 2005)
Fig 4. Protein expression in the antrum epithelium as determined by immunohistochemistry. Representative examples of a noninfected individual (A–H) and of an H. pylori-infected patient (I–P). Expression was assessed by immunohistochemistry of MUC5AC (A and I), MUC6 (B and J), MUC5B (C and K), gastrin (D and L), CGA (E and M; arrowheads in M indicate positive cells, higher-magnification inset), TFF1 (F and N), TFF2 (G and O), and Ki-67 (H and P; arrowheads in H indicate positive cells, higher-magnification inset). Micrographs A–P are at the same magnification. Insets in H and M: bar = 100 µm.
quite a number of cells in the glands. Analysis of the number of cells and their morphology in consecutive sections makes it likely that these two polypeptides were expressed at least partly within the same cells. The cells expressing gastrin or CGA were usually found in the same regions, as defined by our localization method. However, CGA was in general expressed in slightly fewer cells compared to gastrin. These cell types are described separately here, as gastrin- and CGA-expressing cells, respectively. Semiquantification of Protein Expression During H. pylori Infection. When reviewing the sections of all patients it became highly probable that the expression patterns of the secretory proteins as expressed by the various cell types shifted in a coordinate manner along the surface-to-gland axis. Also, the position of the proliferating cells, as detected by Ki-67 staining, seemed to follow
1081
VAN DE BOVENKAMP ET AL.
Fig 5. Expression of secretory proteins and Ki-67 along the surface-to-gland axis of the antrum. As explained under Materials and Methods (Figure 1), the surface-to-gland axis of each section was divided into 10 equal regions. Expression of the indicated secretory proteins and of Ki-67, visualized as in Figure 4, was assessed per region. Region 1 denotes by definition the most superficial region along the axis, whereas region 10 corresponds to the deepest parts of the glands. Distinction was made between noninfected biopsy specimens (n = 53; white bars) and H. pylori-infected specimens (n = 49; black bars). Presence of a protein per region is expressed as the percentage of patients showing positive staining in a particular region. For example, in the upper-left panel, in region 5, 100% of noninfected patients showed expression of MUC5AC, whereas only 58% of H. pylori-infected patients expressed MUC5AC. Statistically significant differences in protein expression per region were assessed by χ 2 test: ‡P < 0.05; †P < 0.005; *P < 0.0005.
this pattern (Figure 4). Therefore, we localized and quantified the extent of expression of each protein along the surface-to-gland axis (Figure 5). In the lower regions of the surface-to-gland axis (regions 4–9), there was less frequent expression of MUC5AC in H. pylori-infected individuals than in noninfected patients. At the same time, expression of both MUC6 and MUC5B was more frequently found in these regions (i.e., regions 4–8 for both MUC6 and MUC5B) of the surface-to-gland axis in H. pylori-infected individuals. TFF1 was found along all regions of the surface-to-gland axis in more that 90% of the noninfected group (Figure 5). However, TFF1 was found significantly less frequently in all regions, except for region 1, in the H. pylori-infected group. TFF2 showed the most widespread expression, and was only significantly less frequently found in regions 2–4 of the H. pylori-infected group, but even in these regions expression was still found in at least 80% of the patients (Figure 5).
1082
Gastrin, CGA, and Ki-67 showed peaks of expression along the surface-to-gland axis. As noted, gastrin and CGA showed overlapping expression patterns. In noninfected individuals, the mean peaks of both gastrin and CGA expression were found in region 7 (Figure 5). The mean expression of Ki-67 was found just above the peak in gastrin and CGA expresion in region 6 in noninfected individuals (Figure 5). Although the regions of expression of these three markers usually overlapped in individual patients, this characteristic distribution of gastrin, CGA, and Ki-67 along the surface-to-gland axis was very consistently found in each patient. Remarkably, each of these proteins showed a mean expression peak closer to the surface in H. pylori-infected patients compared to noninfected patients. Thus, gastrin, CGA, and Ki-67 in the infected tissue showed mean peaks of expression in regions 6, 6, and 5, respectively (Figure 5). Interestingly, the relative positions of expression of these three markers in infected patients remained very similar to those in Digestive Diseases and Sciences, Vol. 50, No. 6 (June 2005)
H. pylori AFFECTS SECRETORY CELLS TABLE 2. CORRELATIONS BETWEEN EXTENT OF EXPRESSION OF SECRETORY PROTEINS AND H. pylori-RELATED PATHOLOGY MUC5AC MUC5B MUC6 H. pylori Inflammation Atrophy
TFF1
TFF2
−0.667∗∗ 0.619∗∗ 0.691∗∗ −0.572∗∗ −0.392∗∗ −0.688∗∗ 0.602∗∗ 0.714∗∗ −0.594∗∗ −0.447∗∗ −0.246∗
Note. The presence of H. pylori was scored as either absent or present, whereas inflammation was scored semiquantitatively based on the Sydney classifications (see Material and Methods). Expression of secretory proteins was determined as the number of regions along the surface-to-gland axis in patients in whom expression was found (see Figure 1). The extent of expression of gastrin and CGA did not correlate significantly with H. pylori infection, inflammation, or atrophy and, therefore, is not shown here. *P < 0.05; **P < 0.0005.
noninfected individuals: from surface to glands—Ki-67, CGA, and gastrin in overlapping regions. Coordinated Protein Expression. From the analysis shown in Figure 5 it seemed evident that the expression of secretory proteins within the antrum epithelium shifted coordinately, depending on the H. pylori-related pathology. We analyzed the extent of expression of the individual secretory proteins within the antrum epithelium and related these to H. pylori infection and the degrees of inflammation within the individual biopsy specimens (Table 2). It appeared that the extent of expression of MUC5AC, TFF1, and TFF2 was negatively affected by both increasing infection and inflammation, whereas the areas within the epithelium showing MUC5B and MUC6 expression were positively correlated with both increasing infection and inflammation (Table 2). Neither infection nor inflammation affected the extent of expression of gastrin or CGA. Interestingly, the correlations of the individual secretory proteins were equally strong with H. pylori infection as with active inflammation. DISCUSSION Our aim was to investigate the effects of H. pylori infection on the expression of secretory proteins in the antrum epithelium and thereby get a closer view of the major secretory cell populations within the epithelium. We have identified H. pylori as a very likely inducer of coordinated shifts in secretory cell populations in the human antrum. We found positive correlations between H. pylori infection and active inflammation, mucosal atrophy, and proliferation. Previous studies also showed increased epithelial proliferation and apoptosis in H. pylori-infected individuals in combination with altered differentiation patterns of epithelial cells, whereas after eradication therapy epithelial homeostasis and differentiation patterns return to normal (19–23). The mechanism by which H. pylori induces apoptosis, increased proliferation, and deranged Digestive Diseases and Sciences, Vol. 50, No. 6 (June 2005)
cellular differentiation in gastric epithelial cells remains unclear, and was beyond the scope of the present study. However, taking our and previously published results together, it seems evident that H. pylori infection plays a direct causal role in gastric inflammation, whereas changes in epithelial proliferation and apoptosis and occurrence of epithelial atrophy are most likely the effects of inflammatory processes within the gastric mucosa, which are induced by H. pylori. The expression of marker proteins is discussed for individual markers below. MUC5AC was expressed in a manner complementary to MUC6 both in H. pylori-infected and noninfected individuals. In all patients MUC5AC-expressing cells formed a seemingly uninterrupted sheet of cells on the surface of the epithelium, extending to a variable degree into the gastric pit and gland epithelium. MUC6-expressing cells were localized to the glandular structures and never found in the surface epithelium. With respect to the localization of MUC5AC- and MUC6-producing cells, this reciprocal mucin expression was briefly mentioned in one of our earlier studies (23) and is in accordance with the results of Ho and coworkers (24). In agreement with a recent study, we demonstrated that MUC5B expression was also found in the antrum of H. pylori-infected and noninfected individuals (25). As MUC5B and MUC6 were always expressed in virtually all cells in the lower part of the glands, it seems likely that these cells constitute one cell type that coexpresses both mucins. In H. pylori-infected patients there were fewer MUC5AC-expressing cells in the middle and lower regions of the surface-to-gland axis, while MUC5B and MUC6 expression was more extended in these regions compared to that in noninfected individuals. The decreased MUC5AC expression in H. pylori-infected patients is consistent with the findings of Byrd et al., who studied gastric-type mucin expression in 29 H. pyloripositive patients (26). Although we found a similar shift of cells expressing MUC5AC toward the surface epithelium and an increased number of MUC6-expressing cells, we never detected MUC6-positive cells in the surface epithelium of H. pylori-infected patients. TFF1 and -2 were expressed in the antrum of all individuals. TFF1 expression was generally localized to the surface and pit epithelium and, to some extent, into the deeper glands, while TFF2 expression was found throughout the entire epithelium. Within the gastrointestinal tract, the stomach is considered the major source of TFF1 and TFF2 expression (27, 28). It is known that TFF-peptides are commonly expressed in association with secretory mucins (29), as is also evident from the present study. The major difference we observed between H. pylori-infected and noninfected patients, with respect to the extent of
1083
VAN DE BOVENKAMP ET AL.
TFF-peptide expression, was that TFF1 expression was found significantly less frequently in regions deeper in the glands in the H. pylori-infected group. TFF1 is usually expressed in close association with MUC5AC in the antrum (24, 30). MUC5AC and TFF1 expression are thus mainly expressed in the same antral cell type, and consequently, as this cell type is diminished in H. pyloriinfected individuals, these two important secretory proteins decrease coordinately. TFF2 expression is evidently associated with both MUC5AC- and MUC5B/MUC6expressing cells, and as a result, the overall expression of TFF2 is hardly affected by shifts in the population sizes of MUC5AC- and MUC5B/MUC6-expressing cell types, as occur in H. pylori-infected individuals. The endocrine cell markers, gastrin and CGA, were primarily colocalized to the same cells, which are presumably all G-cells. Gastrin is by definition expressed by antral G-cells (31), and virtually all of these cells were also stained for CGA, as has been demonstrated by others (15, 32, 33). Occasionally in our study, by comparison of adjacent sections, CGA expression was found in cells not stained for gastrin, but the identity of this small subset of CGA-positive cells was not investigated further. In noninfected individuals, gastrin- and CGA-expressing cells were localized in a rather discrete band in the lower part of the surface-to-gland axis, whereas in H. pyloriinfected patients the band of cells expressing gastrin and CGA was displaced toward the surface epithelium. Within these bands these cells are scattered within the epithelium, and do not form continuous sheets of cells like, e.g., MUCproducing cells. However, the number of these CGA- and gastrin-expresing cells, i.e., the alleged G-cells, did not change noticeably during H. pylori- infection, indicating that the location of this cell type was changed, but not the general size of the G-cell population. The cells in which the proliferation marker Ki-67 was detected were present in a rather discrete band, which was found low in the antral glands in noninfected individuals, yet slightly above the position of the G-cells. However, in H. pylori-infected patients the number of Ki-67expressing cells was significantly increased compared to that in noninfected individuals, and these cells were detected in a much broader band which was much closer to the surface epithelium than in noninfected individuals. Increased proliferation with an upward shift of proliferating cells toward the surface of the gastric epithelium during H. pylori infection was also described by Anti et al. (34). Increased proliferation within the epithelium in response to H. pylori infection may be the result of increased gastrin production (35). However, our study does not indicate an increase in the number of gastrin-producing cells, notwithstanding a possible higher output of gastrin per cell.
1084
Taking the expression profiles of MUC5AC, MUC5B, MUC6, gastrin, CGA, and Ki-67 together, we demonstrated a highly coordinated change of protein expression in the H. pylori-infected antrum epithelium. Given the shift in the position of proliferating cells, toward the surface of the antrum, it seems evident that changes in the relative size of the different populations of the predominant cell types underlie these observed changes in protein expression. The possible factors that could induce such shifts in the size of the cell populations are not known, but a clue regarding a possible mechanism comes from the work of Van den Brink et al. (36). They demonstrated in the gastric corpus that Sonic hedgehog (Shh), which acts as a morphostat, regulates gastric gland morphogenesis. Shh controls the expression of at least three other factors important for epithelial differentiation. In mice, inhibition of Shh markedly enhanced gastric epithelial proliferation and changed the gastric pit–gland asymmetry. We suggest that Shh is a candidate polarizing signal in the maintenance of this pit–gland asymmetry in the adult stomach. In the antrum, a similar morphostat has not yet been identified: Shh is absent in the human antrum (36). But when present, the expression of such a morphogen could be changed in H. pylori-infected epithelium and, consequently, result in coordinate redistribution of cell types along the surfaceto-gland axis. In conclusion, we have demonstrated that H. pylori infection of the antrum resulted in a coordinated deregulation of secretory protein expression. The changes in expression of the secretory proteins suggest that H. pylori infection is responsible for changes in (i) the sizes of cell populations within the antral epithelium and (ii) antral functions that are related to these secretory (glyco-) proteins. The first, in conjunction with the increased epithelial proliferation, is most likely indicative of changed epithelial dynamics, and may be mediated by the inflammation within the H. pylori-infected mucosa. Changes in function can be inferred from the large differences in the numbers of cells exerting those functions. In particular, defensive modalities may be changed. The production of secretory mucins seems altered, which may result in a mucus gel layer of different composition and properties. As the mucus gel layer is thinner in H. pylori-infected individuals (37, 38), increased penetration of potentially damaging substances toward the mucosa may be implicated to increase gastritis. Although less salient, changes in TFF-peptide expression may change the capacity for epithelial repair. In particular, the decrease in the number of cells expressing TFF1 may indicate a potential deficit in epithelial response toward inflamed conditions in H. pylori-infected tissue. Most interestingly, we found indications that MUC5AC constitutes a major receptor Digestive Diseases and Sciences, Vol. 50, No. 6 (June 2005)
H. pylori AFFECTS SECRETORY CELLS
for H. pylori, as bacteria were shown to colocalize in situ with MUC5AC and with MUC5AC-producing cells in the human stomach (8). Moreover, recently we demonstrated that Lewis B carbohydrate structures on MUC5AC are the most likely targets for adherence of H. pylori to the gastric epithelium (9). Thus, H. pylori induces a reduction in the number of its main target cells within the epithelium and, thereby, possibly reduces the production of one of its main receptors. This could indicate that the reduction in the number of MUC5AC-producing cells is a possible mechanism of the gastric epithelium to try to dispel the bacterium. ACKNOWLEDGMENTS This research was financially supported by The Netherlands Digestive Diseases Foundation, Irene Foundation, Jan Dekker/Ludgardine Bouman Foundation, and Foundation “De Drie Lichten.” Dr. Gijs van den Brink and Dr. Kristien Tytgat are gratefully acknowledged for their efforts in collecting the biopsies for this study. The authors thank the providers of the antibodies: Dr. G. Offner (anti-BGBM), Dr. C. de Bol´os (anti-M6.1), and Dr. A. Giraud (anti-pS2 and anti-hSP).
11.
12. 13. 14.
15.
16.
17.
18.
REFERENCES 19. 1. Marshall BJ: Unidentified curved bacilli on gastric epithelium in active chronic gastritis [letter]. Lancet 1:1273–1275, 1983 2. Warren JR: Unidentified curved bacilli on gastric epithelium in active chronic gastritis [letter]. Lancet 1:1273, 1983 3. Marshall BJ, Warren JR: Unidentified curved bacilli in the stomach of patients with gastritis and peptic ulceration. Lancet 1:1311–1315, 1984 4. Parsonnet J, Friedman GD, Vandersteen DP, Chang Y, Vogelman JH, Orentreich N, Sibley RK: Helicobacter pylori infection and the risk of gastric carcinoma. N Engl J Med 325:1127–1131, 1991 5. Yamaguchi T, Nakajima N, Kuwayama H, Ito Y, Iwasaki A, Arakawa Y: Gastric epithelial cell proliferation and apoptosis in Helicobacter pylori-infected mice. Aliment Pharmacol Ther 14 (Suppl 1):68–73, 2000 6. Moss SF, Sordillo EM, Abdalla AM, Makarov V, Hanzely Z, PerezPerez GI, Blaser MJ, Holt PR: Increased gastric epithelial cell apoptosis associated with colonization with cagA+ Helicobacter pylori strains. Cancer Res 61:1406–1411, 2001 7. Van Klinken BJ, Dekker J, Buller HA, Einerhand AWC: Mucin gene structure and expression: protection vs. adhesion. Am J Physiol 269:G613–G627, 1995 8. Van den Brink GR, Tytgat KM, Van Der Hulst RW, Van Der Loos CM, Einerhand AW, Buller HA, Dekker J: H pylori colocalises with MUC5AC in the human stomach. Gut 46:601–607, 2000 9. Van de Bovenkamp JHB, Mahdavi J, Korteland-Van Male AM, B¨uller HA, Einerhand AWC, Boren T, Dekker J: The MUC5AC glycoprotein is the primary receptor for Helicobacter pylori in the human stomach. Helicobacter 8:521–532, 2003 10. Teixeira A, David L, Reis CA, Costa J, Sobrinho-Simoes M: Expression of mucins (MUC1, MUC2, MUC5AC, and MUC6) and Digestive Diseases and Sciences, Vol. 50, No. 6 (June 2005)
20.
21.
22.
23.
24.
25.
26.
type 1 Lewis antigens in cases with and without Helicobacter pylori colonization in metaplastic glands of the human stomach. J Pathol 197:37–43, 2002 Wright NA, Hoffmann W, Otto WR, Rio MC, Thim L: Rolling in the clover: trefoil factor family (TFF)-domain peptides, cell migration and cancer. FEBS Lett 408:121–123, 1997 Wong WM, Poulsom R, Wright NA: Trefoil peptides. Gut 44:890– 895, 1999 Hoffmann W, Jagla W, Wiede A: Molecular medicine of TFFpeptides: from gut to brain. Histol Histopathol 16:319–334, 2001 Dockray GJ, Varro A, Dimaline R: Gastric endocrine cells: gene expression, processing, and targeting of active products. Physiol Rev 76:767–798, 1996 Wiedenmann B, Huttner WB: Synaptophysin and chromogranins/ secretogranins—Widespread constituents of distinct types of neuroendocrine vesicles and new tools in tumor diagnosis. Virchows Arch B Cell Pthol Incl Mol Pathol 58:95–121, 1989 Hocker M, Cramer T, O’Connor DT, Rosewicz S, Wiedenmann B, Wang TC: Neuroendocrine-specific and gastrin-dependent expression of a chromogranin A-luciferase fusion gene in transgenic mice. Gastroenterology 121:43–55, 2001 Gerdes J, Lemke H, Baisch H, Wacker HH, Schwab U, Stein H: Cell cycle analysis of a cell proliferation-associated human nuclear antigen defined by the monoclonal antibody Ki-67. J Immunol 133:1710–1715, 1984 Dixon MF, Genta RM, Yardley JH, Correa P: Classification and grading of gastritis. The updated Sydney System. International Workshop on the Histopathology of Gastritis, Houston 1994. Am J Surg Pathol 20:1161–1181, 1996 Moss SF, Calam J, Agarwal B, Wang S, Holt PR: Induction of gastric epithelial apoptosis by Helicobacter pylori. Gut 38:498–501, 1996 Jones NL, Shannon PT, Cutz E, Yeger H, Sherman PM: Increase in proliferation and apoptosis of gastric epithelial cells early in the natural history of Helicobacter pylori infection. Am J Pathol 151:1695– 1703, 1997 Fraser AG, Sim R, Sankey EA, Dhillon AP, Pounder RE: Effect of eradication of Helicobacter pylori on gastric epithelial cell proliferation. Aliment Pharmacol Ther 8:167–173, 1994 Syder AJ, Guruge JL, Li Q, Hu Y, Oleksiewicz CM, Lorenz RG, Karam SM, Falk PG, Gordon JI: Helicobacter pylori attaches to NeuAcα2,3Galβ1,4 glycoconjugates produced in the stomach of transgenic mice lacking parietal cells. Mol. Cell 3:263–274, 1999 Mahdavi J, Sonden B, Hurtig M, Olfat FO, Forsberg U, Roche N, Angstrom J, Larsson T, Teneberg S, Karlsson KA, Altraja S, Wadstrom T, Kersulyte D, Berg DE, Dubois A, Petersson C, Magnusson KE, Norberg T, Lindh F, Lundskog BB, Arnqvist A, Hammerstrom L, Boren T: Helicobacter pylori SabA adhesin in persistent infection and chronic inflammation. Science 297:579–578, 2002 Ho SB, Roberton AM, Shekels LL, Lyftogt CT, Niehans GA, Toribara NW: Expression cloning of gastric mucin complementary DNA and localization of mucin gene expression. Gastroenterology 109:735–747, 1995 Longman RJ, Douthwaite J, Sylvester PA, Poulsom R, Corfield AP, Thomas MG, Wright NA: Coordinated localisation of mucins and trefoil peptides in the ulcer associated cell lineage and the gastrointestinal mucosa. Gut 47:792–800, 2000 Byrd JC, Yan P, Sternberg L, Yunker CK, Scheiman JM, Bresalier RS: Aberrant expression of gland-type gastric mucin in the surface epithelium of Helicobacter pylori-infected patients. Gastroenterology 113:455–464, 1997
1085
VAN DE BOVENKAMP ET AL. 27. Hanby AM, Poulsom R, Elia G, Singh S, Longcroft JM, Wright NA: The expression of the trefoil peptides pS2 and human spasmolytic polypeptide (hSP) in ‘gastric metaplasia’ of the proximal duodenum: implications for the nature of ‘gastric metaplasia.’ J Pathol 169:355– 360, 1993 28. Hanby AM, Poulsom R, Singh S, Elia G, Jeffery RE, Wright NA: Spasmolytic polypeptide is a major antral peptide: distribution of the trefoil peptides human spasmolytic polypeptide and pS2 in the stomach. Gastroenterology 105:1110–1116, 1993 29. Sands BE, Podolsky DK: The trefoil peptide family. Annu Rev Physiol 58:253–273, 1996 30. Rio MC, Bellocq JP, Daniel JY, Tomasetto C, Lathe R, Chenard MP, Batzenschlager A, Chambon P: Breast cancer-associated pS2 protein: synthesis and secretion by normal stomach mucosa. Science 241:705–708, 1988 31. McColl KE, Gillen D, El-Omar E: The role of gastrin in ulcer pathogenesis. Bailliere’s Best Pract Res Clin Gastroenterol 14:13–26, 2000 32. Norlen P, Curry WJ, Bjorkqvist M, Maule A, Cunningham RT, Hogg RB, Harriott P, Johnston CF, Hutton JC, Hakanson R: Cell-specific processing of chromogranin A in endocrine cells of the rat stomach. J Histochem Cytochem 49:9–18, 2001 33. Van den Brink GR, ten Kate FJ, Ponsioen CY, Rive MM, Tytgat GN, van Deventer SJ, Peppelenbosch MP: Expression and activation of NF-kappa B in the antrum of the human stomach. J Immunol 164:3353–3359, 2000 34. Anti M, Armuzzi A, Gasbarrini A, Gasbarrini G: Importance of changes in epithelial cell turnover during Helicobacter pylori infection in gastric carcinogenesis. Gut 43 (Suppl 1):S27–S32, 1998 35. Johnson LR: Regulation of gastrointestinal mucosal growth. Physiol Rev 68:456–502, 1988
1086
36. Van den Brink GR, Hardwick JC, Tytgat GN, Brink MA, Ten Kate FJ, Van Deventer SJ, Peppelenbosch MP: Sonic hedgehog regulates gastric gland morphogenesis in man and mouse. Gastroenterology 121:317–328, 2001 37. Newton JL, Jordan N, Oliver L, Strugala V, Pearson J, James OF, Allen A: Helicobacter pylori in vivo causes structural changes in the adherent gastric mucus layer but barrier thickness is not compromised. Gut 43:470–475, 1998 38. Newton JL, Jordan N, Pearson J, Williams GV, Allen A, James OF: The adherent gastric antral and duodenal mucus gel layer thins with advancing age in subjects infected with Helicobacter pylori. Gerontology 46:153–157, 2000 39. Bara J, Chastre E, Mahiou J, Singh RL, Forgue-Lafitte ME, Hollande E, Godeau F: Gastric M1 mucin, an early oncofetal marker of colon carcinogenesis, is encoded by the MUC5AC gene. Int J Cancer 75:767–773, 1998 40. Van Klinken BJ, Dekker J, van Gool SA, van Marle J, Buller HA, Einerhand AW: MUC5B is the prominent mucin in human gallbladder and is also expressed in a subset of colonic goblet cells. Am J Physiol 274:G871–G878, 1998 41. De Bolos C, Garrido M, Real FX: MUC6 apomucin shows a distinct normal tissue distribution that correlates with Lewis antigen expression in the human stomach. Gastroenterology 109:723–734, 1995 42. Rio MC, Chenard MP, Wolf C, Marcellin L, Tomasetto C, Lathe R, Bellocq JP, Chambon P: Induction of pS2 and hSP genes as markers of mucosal ulceration of the digestive tract. Gastroenterology 100:375–379, 1991 43. Srivatsa G, Giraud AS, Ulaganathan M, Yeomans ND, Dow C, Nicoll AJ: Billary epithelial trefoil peptide expression is increased in biliary diseases. Histopathology 40:261–268, 2002
Digestive Diseases and Sciences, Vol. 50, No. 6 (June 2005)
