Effects of Helicobacter pylori infection and long-term proton pump inhibitor use on enterochromaffin-like cells.

BACKGROUND: Excessive release of gastrin leads to hypertrophy and hyperplasia of enterochromaffin-like cells (ECL) and prolonged stimulation of these cells causes functional impairment. The purpose of this study was to investigate the effect of Helicobacter pylori (H. pylori) infection and long-term proton pump inhibitors (PPI) use on ECL cells.
METHODS: Fifteen patients who underwent endoscopy because of dyspeptic symptoms were enrolled in the present study. Biopsies were taken from corpus and antrum and existence of H. pylori was investigated with culture, cytology and CLOtest. The patients were divided into 3 groups. Group-A: H. pylori-negative, never treated previously with PPI; Group-B: H. pylori-positive, never treated previously with PPI; and group-C: H. pylori-negative and continuously treated with PPI for more than 6 months before the subject recruitment period. The features of ECL cell in oxyntic glands were examined with electron microscopy on biopsy specimens.
RESULTS: ECL cells were completely normal in Group A. In group B, moderate hyperplasia and vacuolization was seen in ECL cells. In group C, ECL cell hyperplasia was observed and vacuoles with greater amounts of granules in enlarged vesicles were found more intensely in cytoplasm.
CONCLUSION: The use of PPI for a long period of time and presence of H. pylori infection are risk factors for ECL hyperplasia.

Introduction

The epithelium of the mammalian stomach consists of parietal cells, chief cells, mucous cells, and endocrine cells. So-called enterochromaffin-like cells (ECL) are predominant among the endocrine cells [1]. They respond to gastrin and play an important role in the regulation of gastric acid secretion by mobilizing histamine to stimulate the parietal cells via H2-receptor activation [2].

The ECL cells display a characteristic ultrastructure with numerous, fairly large secretory vesicles, and a few electron-dense granules and small clear microvesicles.

Mobilization of histamine is associated with ultrastructural changes in the ECL cells, most notably the loss of secretory vesicles, which represents the presumed storage site of histamine [1,3,4]. In response to sustained hypergastrinemia, the ECL cells increase in size and number and start to develop vacuoles and lipofuscin bodies in their cytoplasm [5]. It has been suggested that both vacuoles and lipofuscin bodies are products of the process of crinophagy (autodigestion of secretory organelles), which allows overstimulated ECL cells to cope with excess secretory material [3-5].

Hypergastrinemia associated with long-term proton pump inhibitor (PPI) therapy may lead to ECL cell hyperplasia [6-9]. However, atrophic gastritis of corpus is also known to be associated with ECL cell hyperplasia and severity of atrophic gastritis is correlated with the grade of hyperplasia [10]. Chronic Helicobacter pylori (H. pylori) infection can also affect ECL cells via mucosal pro-inflammatory cytokines such as tumor necrosis factor (TNF)-α or interleukin (IL)-1β elevated by H. pylori, and causes inhibition of histamine synthesis and secretion which can predispose to hypochlorhydria and hypergastrinemia. Additionally, continuous long-term PPI therapy has been found to be associated with progression of chronic atrophic gastritis and ECL cell hyperplasia in patients infected by H. pylori [8,9,11,12].

The aim of the present study was to examine structural changes in ECL cells with electron microscopy in the biopsy samples of gastric corpus mucosa in patients who were H. pylori negative, treated with PPI for a long time because of reflux disease and H. pylori-positive, not treated with PPI.

Patients and Methods

Fifteen patients undergoing endoscopic examination because of dyspeptic symptoms were enrolled in the study. The presence of H. pylori was sought on biopsies by CLOtest, culture and cytology. Nine patients were male and 6 were female. Mean age was 36.2 (20-56) years. The patients were divided into 3 groups. Group A: Five H. pylori-negative patients, who never previously received PPI therapy; Group B: Five H. pylori-positive patients, who never received PPI treatment; Group C consisted of 5 patients who were using long-term (> 6 months) PPI due to gastroesophageal reflux disease and were H. pylori-negative. The patients underwent repeat endoscopy and 4 biopsies were obtained from corpus for electron microscopy examination.

Corpus biopsy samples were fixed in buffered 10% formaldehyde and they were submitted to the same pathologist. After fixing for at least 24 h, the samples were dehydrated in alcohol and carefully oriented before embedding in paraffin. Sections from each corpus biopsy sample were stained by the Grimelius silver impregnation method to identify argyrophil cells. Biopsy samples were cut into 1 mm3 blocks and fixed in 2.5% gluteraldehyde for 2 h at 4 °C then post fixed in 2% osmium tetraoxide for 2 h at 4 °C. The sections were then hydrated with ethanol and embedded in epoxy resin. Ultra-thin sections were prepared and double stained with uranium acetate and lead citrate, the sections were observed through transmission electron microscope (Leo 906 E transmission electron microscope, 80 kVOberkochen- Germany).

ECL cell properties of oxyntic glands were investigated by transmission Jeol 100 electron microscopy. The ECL cells were identified by their characteristic ultrastructure [4-13]. Granules were defined as cytoplasmic membrane-enclosed organelles (with a diameter of 25–200 nm), displaying an electron-dense core and a thin, electron-lucent halo between the membrane and the dense core, the diameter of the dense core representing more than 50% of the diameter of the entire organelles.

Vesicles were membrane-enclosed, electron-lucent organelles, sometimes possessing a small, often eccentrically located dense core, which was less than 50% of the diameter of the vesicle. The lipofuscin bodies were identified by their high electron density and irregular shape. The ECL cells contained few lipofuscin bodies, numerous vesicles and vacuoles (Fig. 1). The cell numbers and cytoplasmic changes in all three groups were evaluated qualitatively. Fasting serum gastrin concentrations (normal range; 5.6-29.3 pg/mL) were measured in 5 mm3 blood samples drawn via venous route.

Figure 1

Enterochromaffin-like (ECL) cell and cytoplasmic organelles. (a) normal ECL cell, (b) granules, (c) vacuoles, (d) lipofuscin

Results

Regarding the electron microscopic examination in Group A, hyperplasia was not observed in ECL cells. While small numbers of vacuoles were detected in the cytoplasm, no histamine storage was observed. In Group B, moderate increase in the number of ECL cells was observed. There was also increase in the number of cytoplasmic vacuoles. In Group C, patients who were H. pylori-negative with a history of long-term PPI usage, ECL cell hyperplasia was more prominent compared to Group B and wide-large vacuoles and lipofuscin pigments were detected in the cytoplasm (Fig. 2).

Figure 2

(a) Helicobacter pylori (H. pylori) and proton pump inhibitor (PPI)-negative group. Enterochromaffin-like (ECL)hyperplasia was not observed; only a small number of vacuoles were seen in the cytoplasm; no histamine storage was observed; (b) The H. pylori-positive and PPI-negative group. Histamine-containing granule accumulation within vesicles and vacuoles. (c-d) PPI-positive group. Hyperplasia, vacuoles, lipofuscin and granules in ECL cells

Mean serum gastrin levels in group A and group B were detected as 18.42 pg/mL (10-28) and 27.62 pg/mL (14-36) respectively. Among patients in Group C who received PPI treatment for more than 6 months, serum gastrin level was detected as 125.67 pg/mL (68-267).

Discussion

ECL cells can be visualized by light microscopy, immunohistochemistry, and the electron microscopy (morphometry) technique [4,14]. Zhao et al identified characteristic ultrastructure of ECL cells with the electron microscopy technique [4,13]. The ECL cells display a characteristic ultrastructure with numerous, fairly large secretory vesicles, and a few electron-dense granules and small clear microvesicles [4]. The granules and vesicles in the ECL cells have been classified in electron-dense granules and electron-lucent secretory vesicles and microvesicles [15]. The granules are defined as membrane-enclosed, dense-cored organelles with a diameter of 25–200 nm, displaying a thin, electron-lucent halo between the membrane and the dense core, the diameter of the dense core being more than 50% of the diameter of the entire organelles. Vesicles are membrane-enclosed, electron-lucent organelles without a dense core or with a small, often eccentrically located dense core, the diameter of the dense core being less than 50% of the diameter of the organelle. Based on their profile size, vesicles belong to one of the three populations: 1. secretory vesicles with a diameter of 125–500 nm (with or without visible dense core); 2. vacuoles (seen in ECL cells of omeprazole-treated rats only) with a diameter of at least 500 nm (with one or more dense cores); and 3. clear, electron-lucent microvesicles (without dense core) with a diameter of 25–125 nm [15]. In the present study, ECL cells were examined by electron microscopy and a qualitative method was used during the examination. This is a limitation for our study. In the group with H. pylori-negative patients who had never received PPI treatment previously, the number of ECL cells was normal and hyperplasia was not observed in ECL cells.

In patients treated with omeprazole, serum gastrin levels may increase up to 4-fold; this is accompanied by development of diffuse, linear, and/or micronodular ECL cell hyperplasia. Nishi et al administered high-dose omeprazole to patients with Barrett’s esophagus for 2 years, and investigated changes in gastric ECL cells using endoscopic biopsy specimens to clarify the etiology of hyperplasia of the parietal cells [16]. In another study, Rindi et al determined the effects of 5-year treatment with rabeprazole or omeprazole on the gastric mucosa. Two hundred and forty-three patients received rabeprazole (20 mg or 10 mg) or omeprazole (20 mg) once daily for up to 5 years, for gastro-esophageal reflux disease and 51% of the patients completed the whole 5 year period. ECL cell hyperplasia occurred in a minority of patients, and was associated with serum gastrin concentrations. No ECL cell dysplasia or tumors were observed [8]. In the present study, in H. pylori-negative patients who received PPI treatment for more than six months, there was a distinct hyperplasia in ECL cells and the numbers of granules, vesicles, vacuoles and inclusions were significantly greater with electron microscopy.

In the presence of H. pylori infection, acid suppression results in more marked gastrin increase and a higher risk of developing ECL cell hyperplasia than in non-infected subjects [12,17-20]. Infection with H. pylori may stimulate the production of IL-1 and nitric oxide, which are recognized inhibitors of gastric acid secretion. H. pylori-positive subjects under acid suppression also have elevated gastrin levels, which are 30-50% greater than in H. pylori-negative subjects. The reason seems to be that gastrin stimulates both ECL cell histamine secretion and proliferation [21]. Gastric carcinoids are rare neoplasms, and usually are seen only in the stomach of patients with severe atrophic gastritis, pernicious anemia or Zollinger Ellison syndrome [22-24]. However, other than a single case report of carcinoid in the apparent absence of atrophic gastritis or H. pylori infection, rigorous studies in adults have not shown carcinoids in patients treated with acid-blocking agents [22,25,26]. These studies showed that ECL cell hyperplasia in the context of patients treated with long-term PPIs is a benign change. It occurs with greater prevalence in PPI-treated patients who are H. pylori-positive, but it also occurs in individuals with chronic H. pylori antral gastritis who were never treated with acid-blocking agents [10,22,26,27].

There has been no evidence of the relationship between H. pylori infection and gastric carcinoids in humans. Hovewer, gastric cancer and gastric carcinoids have been reported in H. pylori-infected Mongolian gerbils [28,29]. Honda et al showed that gastric cancers appeared in 2 of 5 animals infected with H. pylori 18 months after H. pylori inoculation. Gastric cancers were similarly observed in 5 of 10 infected animals, 24 months after infection [30]. In another study, Kawaga et al evaluated serum gastrin and histopathological examination of the stomach at 6, 12, 18, and 24 months after H. pylori inoculation in Mongolian gerbils. They observed gastric carcinoids in 5 of 10 infected animals at 24 months after infection [29].

In the present study, the ECL cell hyperplasia was moderately increased in the H. pylori-positive group compared to the H. pylori-negative group, which received no PPI treatment. The serum gastrin levels were found higher than in H. pylori-negative patients. However, the increase both in ECL cell hyperplasia and serum gastrin levels was not as high in the group using long-term PPI.

During long-term acid suppression, gastric fundic ECL cell hyperplasia may develop, especially in the presence of H. pylori infection. Acid-suppressive therapy alters the H. pylori gastritis, transforming the antrum-predominant pattern into a body-predominant pattern. Eissele et al investigated 42 patients with reflux esophagitis or peptic ulcer disease, who were treated with lansoprazole for up to 5 years. They found an increased prevalence of linear/micronodular ECL cell hyperplasia from 6.7% to 54.5% in the H. pylori-positive patients, compared with an increase from 0% to 7.7% in the H. pylori-negative subjects [31]. In the present study, ECL cells were not evaluated in patients with H. pylori-positive and long-term acid suppressive therapy groups. This is a limitation of our study.

Chromogranin A (CgA) is a well-recognized marker of neuroendocrine neoplasia and is released into the circulation from the ECL cell population of the stomach [32-36]. The release of CgA increases during profound gastric acid inhibition, possibly reflecting the trophic effect of gastrin on the ECL cells. Sanduleanu et al evaluated serum CgA levels in 230 dyspeptic patients. The CgA levels were found to correlate with hypergastrinemia, duration of acid inhibition, H. pylori infection, body gland atrophy and ECL cell hyperplasia. They showed that in subjects on long-term acid inhibition, serum CgA was equally sensitive but more specific than serum gastrin for the detection of ECL cell hyperplasia [37]. In the present study, serum CgA levels were not measured in all study groups.

To conclude, in the present study an increased density of ECL cells was observed in the H. pylori-positive group and patients treated with long-term acid suppressive therapy. H. pylori may be an important factor for the progression of fundic gastritis and the development of argyrophil cell hyperplasia during long-term treatment with PPI. For this reason, if PPI therapy is needed for a long time, H. pylori infection must be eradicated.

Summary Box

What is already known:

The use of proton pump inhibitors for a long period and presence of H. pylori infection are risk factors for enterochromaffin-like cell hyperplasia

Long-term PPI therapy has been is also associated with progression of chronic atrophic gastritis and ECL cell hyperplasia in patients infected by Helicobacter pylori

Hypergastrinemia which is caused by long-term proton pump inhibitor therapy may lead to ECL cell hyperplasia

Chronic Helicobacter pylori infection can also affect ECL cells via mucosal pro-inflammatory cytokines such as tumor necrosis factor -α or interleukin-1β.

These cytokines cause inhibition of histamine synthesis and secretion which can predispose to hypochlorhydria and hypergastrinemia

What the new findings are:

The changing of enterochromaffin like cells in long-term users of proton pump inhibitors and Helicobacter pylori in patients can be shown with electron microscopic examination

Helicobacter pylori infection is an important factor for the progression of fundic gastritis and the development of argyrophil cell hyperplasia during long-term PPI therapy

For this reason, if PPI therapy is needed for a long time, Helicobacter pylori infection must be eradicated