American Journal of Gastroenterology ISSN 0002-9270C© 2004 by Am. Coll. of Gastroenterology doi: 10.1111/j.1572-0241.2004.30291.xPublished by Blackwell Publishing

Effect of Helicobacter pylori Infection on GhrelinExpression in Human Gastric MucosaAtsushi Tatsuguchi, M.D., Kazumasa Miyake, M.D., Katya Gudis, M.S., Seiji Futagami, M.D.,Taku Tsukui, M.D., Ken Wada, M.D., Teruyuki Kishida, M.D., Yuh Fukuda, M.D., Yuichi Sugisaki, M.D.,and Choitsu Sakamoto, M.D.Third Department of Internal Medicine; Department of Pathology; Division of Surgical Pathology, NipponMedical School Hospital, Tokyo, Japan

OBJECTIVES: One of the counter-effects of Helicobacter pylori eradication therapy is subsequent obesity. Ghrelinis a recently discovered growth hormone releasing peptide. This endogenous secretagogueincreases appetite and facilitates fat storage. The majority of circulating ghrelin is produced in thegastric mucosa. Therefore, we aimed at investigating changes in ghrelin immunoreactivity in gastricmucosa tissues of patients infected with H. pylori.

METHODS: Sixty-one patients with H. pylori infection (25 cases each of duodenal and gastric ulcer, and 11cases of gastritis) and 22 healthy controls without H. pylori infection were included in the study. H.pylori-infected patients received standard proton pump-based triple therapy followed by histologicalexamination and 13C-urea breath test to confirm H. pylori eradication. H. pylori was eradicated in 50out of 61 patients. Biopsy specimens were obtained from antrum and corpus before and 3 monthsfollowing eradication. Ghrelin expression was evaluated immunohistochemically with an anti-ghrelinantibody, and the number of ghrelin-positive cells determined per 1 mm2 of the lamina propriamucosa.

RESULTS: There was no relationship between ghrelin immunoreactivity and body weight or body mass indexfor healthy controls. The number of ghrelin-positive cells was significantly lower for H. pylori-infectedpatients than for healthy controls. However, the ghrelin-positive cell number increased significantlyfollowing H. pylori eradication without significant change in severity of atrophy.

CONCLUSIONS: These data indicated that H. pylori infection affected ghrelin expression. After H. pylori eradication,gastric tissue ghrelin concentration increased significantly. This could lead to the increased appetiteand weight gain seen following H. pylori eradication.

(Am J Gastroenterol 2004;99:2121–2127)

INTRODUCTION

Helicobacter pylori is the major etiologic agent for chronicactive gastritis and also plays a crucial role in gastric andduodenal ulcer disease, as well as in gastric carcinoma (1).The eradication of H. pylori infection improves histologicalgastritis, decreases the rate of recurrence of gastric and duo-denal ulcer disease, and may prevent the occurrence of gastriccarcinoma (2–4). It has been established that following erad-ication therapy, appetite increases not only in patients whoseulcer or gastritis has been eliminated (5, 6), but also in healthysubjects with intact mucosa, implying that factors other thanelimination of pathology are involved in stimulating appetite(7).

Ghrelin is a recently discovered growth hormone (GH)-releasing peptide, that stimulates GH release, increases ap-petite, and facilitates fat storage, and may influence glucosemetabolism (8–12). The majority of circulating ghrelin isproduced in the gastric mucosa (8, 13, 14). However, regula-

tion of gastric ghrelin secretion is poorly understood. There-fore, we hypothesize that increase in body weight followingH. pylori eradication is associated with changes in ghrelinexpression, and thus aimed our retrospective study at investi-gating changes in ghrelin immunoreactivity in gastric mucosatissues of patients infected with H. pylori.

MATERIALS AND METHODS

SubjectsSixty-one outpatients (30 males, 31 females; mean age 47.4yr; range, 20–73 yr) with either H. pylori-positive active duo-denal ulcer (n = 25), gastric ulcer (n = 25), or gastritis (n =11) diagnosed endoscopically were registered in this study.Twenty-two healthy controls without H. pylori infection werealso included in the study. All patients gave detailed informedconsent in accordance with the Helsinki Declaration, andall were recruited between July 1999 and November 2000.H. pylori infected patients were treated with triple therapy

2121

2122 Tatsuguchi et al.

consisting of a daily regimen of 30 mg lansoprazole (b.i.d),750 mg amoxicillin b.i.d, and 200 mg clarithromycin b.i.d for7 days followed by an additional 2–4-wk-long 30 mg lanso-prazole treatment. H. pylori was eradicated in 50 out of 61patients.

EndoscopyTo prevent the effects of diurnal variations in hormones, en-doscopy was performed after overnight fasting, between 0900and 1200 h. Two specimens were taken from intact mucosain the gastric antrum 2 cm proximal to the pylorus, and onefrom the middle body of the greater curvature. One of theantrum specimens was used for a rapid urease test (CLOtest: Delta West, Bentley, Australia) and the other one forhistology. A second endoscopy with a sampling of the spec-imens was performed 3 months after the end of treatment.Patients infected with H. pylori were defined by positive re-sults in at least two of the three methods, i.e., a 13C-ureabreath test (UBT), histology, and a rapid urease test. 13C-UBT was performed following standard methods. 13C-ureaat 100 mg was dissolved in 100 ml of water. The test so-lution was ingested while in the sitting position, followedimmediately by mouth rinsing. The patient was subsequentlyplaced in the left decubitus position for 5 min and then in thesitting position for 15 min. Breath samples at baseline and20 min after dosing were collected, and analyzed using anisotope-selected nondispersive infrared spectrometer; UBiT-IR200 (Otsuka Electronics Co., Hirakata, Japan). The cut-offvalue was set at 2.5%. Eradication was defined as negativemicrobiologic and UBT tests >8 wk after ending therapy.

HistologyBiopsy samples from the greater curvature of the antrum andthe corpus were treated as usual, and activity (neutrophil in-filtration) and inflammation (mononuclear cell infiltration)assessed by the updated Sydney system (15). In brief, epithe-lial inflammation was defined to be positive when any of thefollowing three criteria were fulfilled: an aggregate of morethan five neutrophils within the epithelium, more than tenneutrophils infiltrating a pit or gland circumferentially, and/oran inflammatory exudate with more than five neutrophils inthe lumen (pit abscess). Neutrophil infiltration in the lam-ina propria was scored from grade 0–3 as suggested by theSydney system. The presence of H. pylori in the biopsy ma-terial was determined histologically using the Giemsa stain.This evaluation was made by a single well-trained patholo-gist (Y.F.) without knowledge of the endoscopic diagnosis orH. pylori status.

Immunohistochemical AnalysisFormalin-fixed, paraffin-embedded tissues were used in thisstudy. Tissue sections were rehydrated and immersed in a0.3% H2O2-methanol solution for 30 min to block endoge-nous peroxidase activity. To enhance immunoreactivity ofghrelin, antigens were retrieved by microwaving 10 minin target retrieval solution (citrate buffer, pH 6.1; DAKO).

Sections were incubated with 10% normal goat serum for20 min to block nonspecific binding of the secondary anti-body, and then incubated overnight with the affinity-purifiedgoat polyclonal antibody raised against a peptide mappingwith internal region of ghrelin of human origin (C-18, SantaCruz Biotechnology Inc., Santa Cruz, CA) (16) at a dilu-tion rate 1:400 at 4◦C. Antibody binding was detected by theavidin-biotin-peroxidase complex method (Vectastatin ABCkit; Vector Laboratories, Burlingame, CA) and visualizedwith a diaminobenzidine substrate kit (Vector) according tomanufacturer’s instructions. Nuclei were counterstained withMayer’s hematoxylin.

Negative control immunohistochemical procedures in-cluded omission of the primary antibody and replacementof the primary antibody with normal goat IgG. To demon-strate specificity, antihuman ghrelin antibody incubated withhuman ghrelin control peptide (10 nmol/ml; Santa CruzBiotechnology) for 1 h at room temperature was used forstaining procedures in some experiments. The number ofghrelin-positive cells were visually counted and determinedper 1 mm2 of the lamina propria mucosa.

Statistical AnalysisData are expressed as mean ± SE. The significance ofdifferences between values in various groups of patientswas evaluated using the Mann-Whitney U test. The num-ber of ghrelin-positive cells/mm2 among groups was com-pared using the Wilcoxon’s signed rank test. The correlationsamong inflammation, activity, atrophy, and H. pylori scorewith ghrelin immunoreactivity were evaluated by Spearmanrank correlation test. A value of p < 0.05 was consideredstatistically significant.

RESULTS

Clinical Characteristics of SubjectsH. pylori positive patients who subsequently responded toeradication did not differ from controls concerning age(45.3 ± 2.8 vs 42.8 ± 1.8), sex (male/female; 26/24 vs 11/11),height (166.7 ± 1.9 vs 165.4 ± 1.8 cm), body weight (58.8 ±2.6 vs 60.1 ± 2.4 kg), and body mass index (21.5 ± 0.8 vs21.9 ± 0.8 kg/m2).

Localization of Ghrelin in the Normal StomachIn H. pylori-free gastric mucosa, ghrelin-immunoreactivecells were abundant in oxyntic glands, from the neck to theirbase, but were scarce in pyloric glands (Fig. 1A). The ghrelin-positive cells in the stomach were small and round or spindle-shaped. We used corpus biopsy specimens to evaluate ghrelinexpression, as ghrelin concentration was highest in the oxyn-tic gland.

Although ghrelin-positive cells were always present in nor-mal oxyntic mucosa, there were wide fluctuations in theirnumber. There was no relationship in the healthy controls

H. pylori and Ghrelin in Gastric Mucosa 2123

Figure 1. Immunohistochemical reactivity for ghrelin in the human gastric fundic mucosa infected without H. pylori (A) or with H. pylori(B). (A) Ghrelin-immunoreactive cells were abundant from the neck to the base of the fundic gland area in gastric mucosa infected withoutH. pylori. The ghrelin cells in the stomach were small and round shaped or spindle shaped. (B) In the gastric mucosa infected with H. pylori,staining for ghrelin was localized in the same cells as in the mucosa without H. pylori, but less frequently in these cells. (C) Negative control.

between ghrelin immunoreactivity and any other parameter:gender, age, body weight, height, or body mass index.

Localization of Ghrelin in the H. pylori Infected StomachIn the H. pylori infected gastric mucosa, the distribution pat-tern of ghrelin was similar to that in the H. pylori-free mucosa(Fig. 1B). However, there were fewer ghrelin-expressing cellsin H. pylori infected patients than in healthy controls (14.2 ±1.6 and 72.0 ± 10.0, respectively, Mann-Whitney U-test; p <

0.001) (Fig. 2). Among all parameters tested in patients in-fected with H. pylori, an inverse correlation was seen betweenghrelin immunoreactivity and inflammation/activity grade(r = −0.465, p < 0.001; r = −0.420, p = 0.025, respec-tively). In H. pylori infected patients, there was an inverse

Figure 2. Comparison of ghrelin expression between normal sub-jects and H. pylori positive patients. The number of cells showingpositive ghrelin was lower in H. pylori infected patients than innormal subjects (14.2 ± 1.6 and 72.0 ± 10.0, respectively, Mann-Whitney U-test; p < 0.001).

correlation between ghrelin immunoreactivity and metapla-sia grade (r = −0.275, p = 0.048), but not atrophy/H. pylorigrade (Fig. 3).

Effect of H. pylori Eradication on GhrelinImmunoreactivity in the Oxyntic MucosaEither inflammation or activity significantly improved aftersuccessful H. pylori eradication (Table 1). In contrast, nochange in either atrophy or metaplasia grade was observed.Ghrelin-positive cells number/mm2 significantly increased,from 14.2 ± 1.6 to 26.1 ± 2.2 in the 50 patients who un-derwent successful H. pylori eradication therapy (p < 0.001,Wilcoxon’s signed rank test) (Fig. 4A). However, there wasno difference in the number of ghrelin-positive cells beforeand after eradication therapy in the 11 patients who failed torespond to H. pylori eradication therapy (Fig. 4B).

DISCUSSION

Our results show significantly lower ghrelin expression lev-els in the H. pylori positive versus the H. pylori negativegroups, and subsequent increases in ghrelin expression fol-lowing eradication therapy. In addition, we found a significantinverse correlation between ghrelin expression levels and in-flammation/activity scores in the H. pylori-infected gastricmucosa.

The majority of circulating ghrelin is produced in the gas-tric mucosa (8, 13, 14). Studies indicate that ghrelin is local-ized in X/A-like cells, one of several endocrine cell types inthe human oxyntic gland mucosa, and one whose secretoryproduct has yet to be described (17). Furthermore, we wereprevented from properly evaluating the number of X/A-likecells before or after H. pylori eradication, as these cells aredifficult to identify under the light microscope. However, sev-eral lines of evidence suggest that these endocrine, or otherglandular, cells are not likely to change radically in numberbefore and after H. pylori eradication. First, the number andlocalization of neuroendocrine cells have been shown not to

2124 Tatsuguchi et al.

Figure 3. Relationship between ghrelin expression and scores of Sydney system parameters. Among all parameters tested in patientsinfected with H. pylori, ghrelin immunoreactivity was correlated with degree of inflammation, degree of activity, that increased as eachscore decreased.

change in gastric mucosa with or without H. pylori infection(18). Second, a number of studies have shown that gastricglandular atrophy is unlikely to recover even after a year fol-lowing successful H. pylori eradication, whether evaluatedendoscopically or histologically (19, 20). In spite of this lackof recovery in atrophy of the gastric mucosa, we saw in-creased ghrelin immunoreactivity 3 months after H. pylorieradication. Furthermore, we found no correlation betweenghrelin immunoreactivity and atrophy scores in H. pylori-infected gastric mucosa, suggesting that changes in ghrelinimmunoreactivity in the gastric mucosa is not a simple reflec-tion of glandular atrophy. Third, ghrelin immunoreactivitieswere found to inversely correlate with inflammation scores inthe H. pylori-infected gastric mucosa. Taken together, theseresults suggest that ghrelin suppression is regulated by signalsakin to those involved in the inflammatory processes in H. py-lori gastritis. It is well known that H. pylori infection inducesthe infiltration of inflammatory cells into the lamina propriaof the gastric mucosa. Chemokines involved in the infiltrationof leukocytes into the mucosa, such as interleukin (IL)-8, and

Table 1. Average Score of Sydney System Parameters before andafter H. pylori Eradication

Parameter Before After p value

Inflammation 2.19 ± 0.09 1.07 ± 0.04 <0.001Activity 2.30 ± 0.11 0.06 ± 0.06 <0.001Atrophy 0.13 ± 0.06 0.11 ± 0.04 NSMetaplasia 0.09 ± 0.06 0.11 ± 0.06 NSH. pylori 2.41 ± 0.09 0 <0.001

proinflammatory cytokines like IL-1β, have also been shownto contribute to the progression of inflammatory processes inthe mucosa (21, 22). In fact, bacterial lipopolysaccharide orIL-1β has been shown to decrease both ghrelin levels and foodintake in rodents (23, 24). To clarify whether such inflamma-tory cytokines and bacterial toxin are in fact involved in theregulation of ghrelin expression in the gastric mucosa in hu-mans, further investigation is needed. Gokcel et al. reportedthat plasma ghrelin levels between H. pylori negative and pos-itive groups did not differ significantly (25). Although theydid not describe the degree of inflammation and activity in H.pylori positive patients, they are thought to be of low degreebecause patients with gastrointestinal disease were excluded.They examined the genuine effect of H. pylori on ghrelin ex-pression and concluded that H. pylori has no effect on plasmaghrelin concentration. As we have shown in this study, therewas a negative relationship between ghrelin expression andinflammation/activity scores. Ghrelin expression may be af-fected by various cytokines and chemokines produced frominflammatory cells, rather than by H. pylori itself.

Suzuki et al. examined levels of stomach and plasma ghre-lin in Mongolian gerbils with H. pylori colonization of thegastric mucosa (26). They found that the number of ghrelinimmunoreactive cells decreased in the stomach after H. py-lori colonization, inconsistent with our data. Although ghre-lin mRNA and protein levels significantly decreased, plasmaghrelin levels somewhat increased after H. pylori coloniza-tion. Stomach tissue ghrelin does not always correspond tothe circulating levels. It has been reported that 8 h of milk re-striction significantly decreased the ghrelin concentration in

H. pylori and Ghrelin in Gastric Mucosa 2125

Figure 4. Comparison of ghrelin expression between before and after H. pylori eradication therapy. (A) Ghrelin-positive cells numberincreased significantly following successful H. pylori eradication (14.2 ± 1.6 and 26.1 ± 2.2, respectively, Wilcoxon signed-ranks test; p <0.001). (B) There was no difference in the number of ghrelin-positive cells before and after eradication therapy in 11 patients who failed torespond to eradicate H. pylori.

the stomachs of 1-wk-old rats, while increasing plasma ghre-lin levels (27). In GH-treated rats, plasma ghrelin levels werereduced approximately 40%, while stomach ghrelin peptidestores were unaffected by increased GH levels. In contrast,in aged rats, stomach ghrelin expression and peptide storesand plasma ghrelin levels increased significantly when com-pared to young rats (28). A possible explanation for thisphenomenon is that tissues other than stomach, such as duo-denum, contribute to a certain amount of circulating ghre-lin; considering that plasma ghrelin levels in gastrectomizedpatients still remain about one-third of those in normal sub-jects. Another explanation is that ghrelin release may be regu-lated at both translational and posttranslational levels. Furtherstudy is required to address this issue.

Increases in ghrelin immunoreactivity 3 months after H.pylori eradication is well-consistent with a recent reportshowing that plasma ghrelin levels significantly increased6 wk after cure of H. pylori infection (29). The majority ofcirculating ghrelin has been shown to be produced in the gas-tric mucosa (8, 13, 14). Thus, our data support that increasesin plasma ghrelin levels is actually due to increased ghrelinexpression in the gastric mucosa, following H. pylori eradica-tion. Plasma ghrelin levels, mainly of gastric origin, have beenshown to be suppressed by ingestion or gastric delivery of nu-trients (14, 30). Thus, it is possible that ghrelin expression inthe gastric mucosa, and its secretion into the bloodstream,is controlled by a variety of factors regulating physiologicfunctions in the stomach with the possible involvement ofgastrin or acid. Although gastrin levels were not recordedbefore and after H. pylori eradication, plasma gastrin levelshave been shown to decrease following H. pylori eradicationwith a concomitant recovery of acid secretion levels (31). Infact, plasma ghrelin levels and intragastric acidity have beenshown to increase in parallel after cure of H. pylori infec-tion (29). Ghrelin stimulates gastric acid secretion in rodents,

whether administered centrally or peripherally (32, 33). Allthese data suggest that plasma ghrelin levels are closely tiedto acid secretion. Accordingly, prolonged suppression of gas-tric acid secretion, commonly seen in patients with chronicatrophic gastritis, might affect ghrelin expression in the gas-tric mucosa. Prescribing lansoprazole could have affected thenumber of immunoreactive ghrelin-secreting cells. To ex-clude the effect of suppression of gastric acid secretion onghrelin expression, patients were divided into two groups, H.pylori-eradicated and uneradicated patients. However, therewas no difference in the number of ghrelin-positive cells be-fore and after eradication therapy in patients who failed torespond to H. pylori eradication therapy, compared to pa-tients who underwent successful therapy. In addition, it hasbeen reported in rats that the serum ghrelin concentration wasunaffected by 1-wk treatment with omeprazole. The samestudy has shown that omeprazole treatment for 10 wk didnot change the level of ghrelin mRNA in the oxyntic mucosa(13). Finally it has been reported that omeprazole-inducedhypergastrinaemia was associated with enlargement of theoxyntic area and with hyperplasia of enterochromaffin-likecells but not to be associated with A-like cells, the cells be-lieved to stain positive for ghrelin (34). Taken together, weconcluded that recovery from ghrelin suppression is mainlya result of inhibition of signals akin to those involved in theinflammatory processes in H. pylori gastritis.

It is well known that successful eradication of H. pylori canoften lead to improved appetite followed by increased bodyweight (5–7). Ghrelin is a GH-releasing peptide; this endoge-nous secretagogue stimulates GH release, increases appetite,and facilitates fat storage, and may also influence glucosemetabolism (8–12). In rodents, ghrelin administration intra-peritoneally or intraventricularly has been shown to inducea marked increase in food intake with consequent increasein body weight (9, 11). Thus, it is reasonable to speculate

2126 Tatsuguchi et al.

that increased ghrelin immunoreactivity following H. pylorieradication, closely linked to increased plasma ghrelin levelsas shown in a recent study (29), may stimulate nutritionalparameters leading to obesity in humans. In addition, studieshave shown that H. pylori eradication induces GERD, a phe-nomenon that could be explained not only by the recovery ofgastric acid secretion, but also by increased body weight thathas been connected to esophageal motility dysfunction (5).If ghrelin does in fact stimulate gastric-acid secretion as hasbeen shown in rodents, increased ghrelin expression after H.pylori eradication may be a key factor involved in the mech-anism leading to GERD. However, we must emphasize thatthe present study only shows changes in ghrelin immunore-activity before and after eradication of H. pylori: the ghrelinantibody used is not known to distinguish active from latentghrelin forms. Thus, it is not clear that the immunoreactivitymeasured by this antibody reflects ghrelin activity. Likewise,it is not clear that the immunoreactivity in the gastric mucosarepresents overall function of ghrelin, linking it to appetiteand body weight increases, in particular to obesity. Obesity isassociated with decreased circulating levels of ghrelin (35).Murray and Emmanuel suggested that it is not physiologi-cally feasible for a hyper-ghrelinaemic state after eradicationto result in increased appetite, because obesity is associatedwith decreased circulating levels of ghrelin (36). The physi-ological response would be for resulting weight loss to leadto a compensatory increase in ghrelin, increased appetite,and regained weight. However, this mechanism may be al-tered by the presence of H. pylori, resulting in slimness andhypo-ghrelinaemia to occur together. Another explanation isthat the hypo-ghrelinaemia seen in obese patients is due tocompensatory hyperinsulinemia associated with insulin re-sistance, which is generally a manifestation of obesity (37).It would be of interest to see whether H. pylori eradicatedpatients are insulin resistant. Therefore, further investigationis clearly required to understand the role of ghrelin expressedin the gastric mucosa with or without H. pylori infection.

Reprint requests and correspondence: Atsushi Tatsuguchi, M.D.,Third Department of Internal Medicine, Nippon Medical School,1-1-5, Sendagi, Bunkyo-ku, Tokyo 113-8603, Japan.

Received November 5, 2003; accepted June 9, 2004.

REFERENCES

1. Marshall BJ, McGechie DB, Francis GJ, et al. Pyloriccampylobacter serology. Lancet 1984;2:281.

2. Marshall BJ, Goodwin CS, Warren JR, et al. Prospec-tive double-blind trial of duodenal ulcer relapse aftereradication of Campylobacter pylori. Lancet 1988;2:1437–42.

3. Graham DY, Lew GM, Klein PD, et al. Effect of treatment ofHelicobacter pylori infection on the long-term recurrence ofgastric or duodenal ulcer. A randomized, controlled study.Ann Intern Med 1992;116:705–8.

4. Uemura N, Okamoto S, Yamamoto S, et al. Helicobacterpylori infection and the development of gastric cancer. NEngl J Med 2001;345:784–9.

5. Labenz J, Blum AL, Bayerdorffer E, et al. Curing Helicobac-ter pylori infection in patients with duodenal ulcer may pro-voke reflux esophagitis. Gastroenterology 1997;112:1442–7.

6. Furuta T, Shirai N, Xiao F, et al. Effect of Helicobacterpylori infection and its eradication on nutrition. AlimentPharmacol Ther 2002;16:799–806.

7. Azuma T, Suto H, Ito Y, et al. Eradication of Helicobacterpylori infection induces an increase in body mass index.Aliment Pharmacol Ther 2002;16:240–4.

8. Kojima M, Hosoda H, Date Y, et al. Ghrelin is a growth-hormone-releasing acylated peptide from stomach. Nature1999;402:656–60.

9. Nakazato M, Murakami N, Date Y, et al. A role for ghrelinin the central regulation of feeding. Nature 2001;409:194–8.

10. Cummings DE, Weigle DS, Frayo RS, et al. Plasma ghre-lin levels after diet-induced weight loss or gastric bypasssurgery. N Engl J Med 2002;346:1623–30.

11. Wren AM, Small CJ, Ward HL, et al. The novel hypothala-mic peptide ghrelin stimulates food intake and growth hor-mone secretion. Endocrinology 2000;141:4325–8.

12. Tschop M, Smiley DL, Heiman ML. Ghrelin induces adi-posity in rodents. Nature 2000;407:908–13.

13. Dornonville de la Cour C, Bjorkqvist M, Sandvik AK,et al. A-like cells in the rat stomach contain ghrelin and donot operate under gastrin control. Regul Pept 2001;99:141–50.

14. Ariyasu H, Takaya K, Tagami T, et al. Stomach is a majorsource of circulating ghrelin, and feeding state determinesplasma ghrelin-like immunoreactivity levels in humans. JClin Endocrinol Metab 2001;86:4753–8.

15. Dixon MF, Genta RM, Yardley JH, et al. Classification andgrading of gastritis. The updated Sydney System. Interna-tional Workshop on the Histopathology of Gastritis, Houston1994. Am J Surg Pathol 1996;20:1161–81.

16. Caminos JE, Nogueiras R, Blanco M, et al. Cellular distri-bution and regulation of ghrelin mRNA in the rat pituitarygland. Endocrinology 2003;144:5089–97.

17. Date Y, Kojima M, Hosoda H, et al. Ghrelin, a novel growthhormone-releasing acylated peptide, is synthesized in a dis-tinct endocrine cell type in the gastrointestinal tracts of ratsand humans. Endocrinology 2000;141:4255–61.

18. Tzaneva M, Julianov A. Chromogranin A-, somatostatin-and serotonin- containing endocrine cells in the corporalgastric mucosa of patients with Helicobacter pylori associ-ated chronic gastritis. Endocr Regul 1999;33:79–82.

19. van der Hulst RW, van der Ende A, Dekker FW, et al. Effectof Helicobacter pylori eradication on gastritis in relation tocagA: A prospective 1-year follow-up study. Gastroenterol-ogy 1997;113:25–30.

20. Rollan A, Giancaspero R, Fuster F, et al. The long-termreinfection rate and the course of duodenal ulcer disease aftereradication of Helicobacter pylori in a developing country.Am J Gastroenterol 2000;95:50–6.

21. Crabtree JE, Peichl P, Wyatt JI, et al. Gastric interleukin-8and IgA IL-8 autoantibodies in Helicobacter pylori infec-tion. Scand J Immunol 1993;37:65–70.

22. Noach LA, Bosma NB, Jansen J, et al. Mucosal tumor necro-sis factor-alpha, interleukin-1 beta, and interleukin-8 pro-duction in patients with Helicobacter pylori infection. ScandJ Gastroenterol 1994;29:425–9.

23. Basa NR, Wang L, Arteaga JR, et al. Bacterial lipopolysac-charide shifts fasted plasma ghrelin to postprandial levels inrats. Neurosci Lett 2003;343:25–8.

24. Asakawa A, Inui A, Kaga T, et al. Ghrelin is an appetite-stimulatory signal from stomach with structural resem-blance to motilin. Gastroenterology 2001;120:337–45.

H. pylori and Ghrelin in Gastric Mucosa 2127

25. Gokcel A, Gumurdulu Y, Kayaselcuk F, et al. Helicobacterpylori has no effect on plasma ghrelin levels. Eur J En-docrinol 2003;148:423–6.

26. Suzuki H, Masaoka T, Hosoda H, et al. Helicobacter pyloriinfection modifies gastric and plasma ghrelin dynamics inMongolian gerbils. Gut 2004;53:187–94.

27. Hayashida T, Nakahara K, Mondal MS, et al. Ghrelin inneonatal rats: Distribution in stomach and its possible role.J Endocrinol 2002;173:239–45.

28. Qi X, Reed J, Englander EW, et al. Evidence that growth hor-mone exerts a feedback effect on stomach ghrelin productionand secretion. Exp Biol Med (Maywood) 2003;228:1028–32.

29. Nwokolo CU, Freshwater DA, O’Hare P, et al. Plasma ghre-lin following cure of Helicobacter pylori. Gut 2003;52:637–40.

30. Tschop M, Wawarta R, Riepl RL, et al. Post-prandial de-crease of circulating human ghrelin levels. J Endocrinol In-vest 2001;24:RC19–21.

31. Iijima K, Ohara S, Sekine H, et al. Changes in gastricacid secretion assayed by endoscopic gastrin test before

and after Helicobacter pylori eradication. Gut 2000;46:20–6.

32. Date Y, Nakazato M, Murakami N, et al. Ghrelin acts in thecentral nervous system to stimulate gastric acid secretion.Biochem Biophys Res Commun 2001;280:904–7.

33. Masuda Y, Tanaka T, Inomata N, et al. Ghrelin stimulatesgastric acid secretion and motility in rats. Biochem BiophysRes Commun 2000;276:905–8.

34. Cui GL, Syversen U, Zhao CM, et al. Long-term omepra-zole treatment suppresses body weight gain and bonemineralization in young male rats. Scand J Gastroenterol2001;36:1011–5.

35. Tschop M, Weyer C, Tataranni PA, et al. Circulatingghrelin levels are decreased in human obesity. Diabetes2001;50:707–9.

36. Murray CD, Emmanuel AV. Ghrelin and Helicobacter py-lori. Gut 2004;53:315 (author reply).

37. McLaughlin T, Abbasi F, Lamendola C, et al. Plasma ghre-lin concentrations are decreased in insulin-resistant obeseadults relative to equally obese insulin-sensitive controls. JClin Endocrinol Metab 2004;89:1630–5.