- Email: [email protected]
Gastric Inflammation, Metaplasia, and Tumor Development in Gastrin–Deficient Mice
GASTROENTEROLOGY 2006;131:246 –258
MICROARRAYS AND OTHER NEW TECHNOLOGIES Gastric Inflammation, Metaplasia, and Tumor Development in Gastrin–Deficient Mice LENNART FRIIS–HANSEN,* KLAUS RIENECK,‡ HANS–OLOF NILSSON,§ TORKEL WADSTRÖM,§ and JENS F. REHFELD* *Department of Clinical Biochemistry and ‡Institute for Inflammation Research, Rigshospitalet, University of Copenhagen, Copenhagen, Denmark; and §Department of Microbiology, University of Lund, Lund, Sweden
Background & Aims: Gastrin deficiency and proton pump inhibitor treatment cause achlorhydria, which predisposes to disease. To elucidate the underlying molecular biology, we examined the changes in gastric gene expression in both types of achlorhydria. We also explored the associated changes in the gastric microflora and the long-term consequences of gastrin-deficient achlorhydria. Methods: Expression profiles were generated from gastric RNA from wild-type mice, gastrin knockout (KO) mice, gastrin KO mice after 1 week of gastrin infusion, and wild-type mice treated for 1 month with a proton pump inhibitor. The results were confirmed using real-time polymerase chain reaction and immunohistochemistry. Selective media were used to characterize the gastric microflora. Results: The number of gastric bacteria was increased in both gastrin KO and PPItreated mice. The expression profiles revealed activation of immune defense genes, interferon-regulated response genes, and intestinal metaplasia of the gastric mucosa. In young gastrin-deficient mice, gastrin infusions reversed the changes. Over time, the changes accumulated, became irreversible, and progressed into metaplasia and polyp development. Finally, the study showed that gastrin regulated the expression of genes encoding extracellular matrix proteins. Conclusions: Independently of gastrin, achlorhydria is associated with gastric bacterial overgrowth and intestinal gene expression patterns and is associated with predisposition to disease. Gastrin is therefore essential for prevention of gastric disease, mainly through control of acid secretion but to a lesser extent also through control of gastric gene expression. The gastrin-deficient mouse serves as a useful new model for gastric metaplasia and neoplasia.
astrin stimulates gastric acid secretion via enterochromaffin-like cells1 and to a lesser degree through direct activation of the parietal cells.2– 4 Gastrin regulates gene expression in enterochromaffin-like,5,6 parietal,7 and other gastric cells8 (for review, see Rehfeld9 and
G
Dockray et al10). Long-term hypergastrinemia results in hypertrophy and hyperplasia of parietal and enterochromaffin-like cells11 and promotes development of gastric cancer.12 Despite the association between hypergastrinemia and neoplasia, antrectomized hypogastrinemic patients also have an increased risk of cancer in the gastric remnant.13–15 To examine the molecular biology of diseases due to lack of gastrin, mice deficient for gastrin16,17 or its receptor18,19 have been developed. They are achlorhydric due to enterochromaffin-like and parietal cell dysfunction, whereas the remaining mucosal growth is largely unaffected.16 –19 The defects in acid secretion are reversed by infusion of exogenous gastrin.17,20,21 Furthermore, gastrin knockout (KO) mice aged 4 –5 months developed gastric bacterial overgrowth, signs of intestinal metaplasia,22 and tumors.23 This study examines the early changes in gene expression due to lack of gastrin and its reversibility after gastrin infusion. The changes in gene expression were compared with those in achlorhydric but hypergastrinemic proton pump inhibitor (PPI)-treated mice.
Materials and Methods Mice and RNA Preparation Gastrin KO mice, backcrossed 7 times to C57BL6/J through both male and female germline, aged 9 –12 weeks were used.17 After removal of the forestomach, RNA was extracted using TRIzol (Invitrogen, Carlsbad, CA) from wildtype mice, gastrin KO mice, gastrin KO mice after 1 week of gastrin replacement (10 nmol · h⫺1 · kg⫺1) using minipumps (Alzet 2001; Alza Corp, Palo Alto, CA), and wild-type mice given a PPI (omeprazole 400 mol · kg⫺1) once daily for 1 Abbreviations used in this paper: KO, knockout; PPI, proton pump inhibitor. © 2006 by the American Gastroenterological Association Institute 0016-5085/06/$32.00 doi:10.1053/j.gastro.2006.04.031
July 2006
INTESTINAL METAPLASIA AND TUMORS IN GASTRIN KO MICE
Table 1. Characteristics for the Antisera Used for Immunocytochemistry Dilution Antigen Cdx2
1:10
Best5
1:5000
Schlafen3/4
1:3000
ISG-15
1:2000
CD45
1:4000
IGF-BP2
1:2000
Syndecan 4
1:3000
NOV
1:2000
Code and source AM392-5M, Biogenex, San Ramon, CA Professor Skerry, Heslington, United Kingdom sc-8903, Santa Cruz, Paso Robles, CA Professor Hass, Milwaukee, WI sc-1178, Santa Cruz, Paso Robles, CA sc-6001, Santa Cruz, Paso Robles, CA sc-9497, Santa Cruz, Paso Robles, CA Professor Perbal, Paris, France
Secondary antibodies with conjugates Envision⫹ (horseradish Prediluted peroxidase conjugated) Tertlary antibodies with conjugates Cy3-conjugated anti– 1:400 horseradish peroxidase
247
pressed as mean fold change with P values using the following settings: minimal difference between baseline and experiment was 50, expression levels ⬍20 were assigned the value 20, and P ⬍ .05 was adjusted so that the 90th percentile false discovery rate was ⬍5%. Expressed sequence tags (ESTs) were compared with GenBank database sequences using BLAST.25
Quantitative Polymerase Chain Reaction The GeneChip results were confirmed by real-time polymerase chain reaction with the Lightcycler (Roche, Mannheim, Germany) using the LightCycler - DNA Master SYBR Green Reaction kit (Roche) as described.20 Each experiment consisted of one negative control, 4 standards, and 7– 8 complementary DNA samples from each group of mice. The messenger RNA (mRNA) expressions were normalized to the glyceraldehyde-3-phosphate dehydrogenase expression in each complementary DNA sample. Statistical analysis was performed using Student t tests. The primer sequences are available on request.
Dako, Glostrup, Denmark
Gastric pH and Gastrin Measurements Jackson, West Grove, PA
month (n ⫽ 20 for each genotype). Long-term alterations were examined in 1-year-old and 1½-year-old mice.
Gene Expression Profiling Probes were generated from 4 pools of total RNA from each of the 4 groups of mice (5 per group) as described by the manufacturer (Affymetrix, Santa Clara, CA). The probe quality was checked on test arrays hybridization to GeneChips Mu74A, B and C, version 2 (Affymetrix).
Data Analysis The microarray data were analyzed using the dChip perfect match-mismatch model (software version 1.3; supplied by Dr Li, Department of Biostatistics, Harvard School of Public Health, Boston, MA).24 The expression in the wild-type mice was compared with that in gastrin KO mice, gastrin KO mice receiving gastrin infusion, and wild-type mice treated with omeprazole. The changes in gene expression were ex-
The gastrin concentration in plasma was measured by radioimmunoassay.26 Following pyloric ligation for 1 hour, pH was measured in the accumulated juice with a PHM 85 pH-meter (Radiometer, Copenhagen, Denmark).27
Immunohistochemistry For gastric histologic examination of mice (9 –11 weeks, 1 and 1½–2 years) stomachs were fixed overnight in 4% paraformaldehyde (pH 7) and embedded in paraffin (6 – 8 mice per time point per genotype). Immunohistochemistry was performed as described20 and using the antibodies and detection systems described in Table 1. H&E and periodic acid– Schiff/Alcian staining was also performed.
Microbiology Semiquantitative analysis of enterococci and lactobacilli per stomach (corpus/fundus and antrum) was performed by smearing half of a mouse stomach on bile-esculin agar and the other half on MRS agar. After 48 hours of aerobic incubation at 37°C, the number of colonies was counted. Single colonies were identified using the api-test (BioMéreux, Marcy, France) (n ⫽ 6 – 8 mice, aged 12–16 weeks from each of the 4 groups).
Table 2. Plasma Gastrin Concentrations and Gastric pH Characteristics of the Examined Mice Wild-type Plasma-gastrin (pmol/L) pH
42 ⫾ 14 3.5 ⫾ 0.2
Gastrin KO detecteda
Not 6.2 ⫾ 0.3a
Gastrin KO ⫹ gastrin infusion
Omeprazole-treated wild-type
415 ⫾ 3.6 ⫾ 0.2b
98 ⫾ 34a 5.7 ⫾ 0.4a
56a
NOTE. Plasma gastrin concentrations were measured using a radioimmunoassay in wild-type mice, gastrin KO mice, gastrin KO mice after 1 week of gastrin subcutaneous infusion, and omeprazole treated wild-type mice (see Materials and Methods). To measure the gastric acidity, the mice were anesthetized with intraperitoneal 2,2,2-tribromoethanol (Sigma-Aldrich Corp, St Louis, MO), the abdominal cavity opened, and the esophagus and pylorus ligated for 1 hour. n ⫽ 8 –9. aP ⬍ .05, bP ⬎ .05 (NS).
248
FRIIS–HANSEN ET AL
GASTROENTEROLOGY Vol. 131, No. 1
Results Gastrin and Gastric Acidity Gastrin KO and omeprazole-treated, hypergastrinemic mice were achlorhydric, whereas the KO mice that received gastrin infusion had normal gastric acidity despite hypergastrinemia (Table 2). Microbiology The achlorhydria due to lack of gastrin was associated with a partly reversible overgrowth of lactobacilli and enterococci. PPI treatment did not cause the same degree of bacterial overgrowth (Figure 1). Gastric Gene Expression Profiling General observations. Hierarchical cluster analysis separated each of the 4 groups. The 2 groups of KO mice were more closely related to each other than to the wild-type mice (Figure 2). The expression of chromogranin A and histidine decarboxylase (known to be regulated by gastrin5,6) was low in the gastrin KO mice, restored to wild-type levels after gastrin infusion, and increased in the omeprazole-treated mice (Table 3). Gastric somatostatin expression was unchanged, which is probably because this experiment only examined the total gastric expression and not fundic and antral expression independently. Up-regulated genes. Despite normal gastric morphology in young gastrin KO mice (Figure 3A and B), 59 genes were up-regulated in KO mice and their expression was normalized following gastrin infusion (Ta-
Figure 1. Semiquantitative analysis of the load of gastric enterococci and lactobacilli. The type and number of gastric bacteria were determined in wild-type, gastrin KO, omeprazole-treated wild-type, and gastrin KO mice after 1 week of gastrin infusion. All mice were 12–16 weeks old. Half of the stomach was smeared on bile-esculin agar and the other half on MRS agar. After 48 hours of aerobic incubation at 37°C, the number of colonies was counted. The data are given as 109 colonies per stomach. Single colonies were identified using the api-test (BioMéreux). The mice were housed in a barrier mouse facility with 12-hour light/dark cycles. The mouse colony was monitored according to the FELASA guidelines,67 and during the experiments the colony tested negative for all monitored viruses and bacteria except Helicobacter bilis. n ⫽ 6 – 8 in each group. *P ⬍ .05.
Figure 2. Hierarchical cluster analysis of gastric gene expression profiles from wild-type, gastrin KO, gastrin KO ⫹ gastrin, and omeprazole-treated wild-type mice. The 2 groups of wild-type mice were more closely related to each other than they were to the 2 groups of gastrin KO mice. Exogenous gastrin infusion restores acid secretion but does not normalize gene expression. Similarly, even though omeprazoletreated mice are achlorhydric, their gastric gene expression is still closer to wild-type mice than to the achlorhydric gastrin KO mice. (Gastric RNAs from 5 mice aged 12–16 weeks were pooled and subsequently analyzed using the Affymetrix u74 GeneChips.)
July 2006
INTESTINAL METAPLASIA AND TUMORS IN GASTRIN KO MICE
249
Table 3. Gastric Gene Expression Down-regulated due to Gastrin Deficiency Fold change
Gene Solute carrier family 4 (anion exchanger) member 2 Aquaporin 4 Insulin-like growth factor binding protein 2 Nephroblastoma overexpressed gene Cytochrome c oxidase, subunit VIIa 1 LIM only 2 Ankyrin 3, epithelial Small inducible cytokine A27 Syndecan 4 TYRO protein tyrosine kinase binding protein Hepcidin antimicrobial peptide Chromogranin A Histidine decarboxylase Gastrin
Accession no.
Wild-type expression
Wild-type ⫹ omeprazole
KO ⫹ gastrin
KO
P value
Function
J04036 U88623
1131 67
⫺1.3 ⫺1.3
1.0 1.4
1.6 3.5
.0004 .002
HCO3-transport Water channel
X81580
615
⫺1.4
⫺1.0
1.8
.002
Hormone
Y09257
171
⫺1.4
1.2
1.2
.004
Hormone
AF037370 M64360 L40631 AW124975 D89571
452 194 344 199 765
⫺1.4 ⫺1.4 ⫺1.5 ⫺1.5 ⫺1.5
⫺1.0 ⫺1.1 ⫺1.2 ⫺1.2 ⫺1.6
1.4 1.1 1.3 ⫺1.0 ⫺1.1
.01 .006 .004 .001 .0002
Apoptosis Angiogenesis Cytoskeletal Secreted Extracellular matrix
AF024637
280
⫺1.6
1.0
⫺1.2
.004
Unknown
AI255961 M64278 X57437 U34293
721 626 160 2999
⫺1.8 ⫺2.1 ⫺3.2 ⫺35.8
⫺1.4 1.3 1.8 ⫺34.5 Wild-type ⫽ 1.0
1.1 2.0 2.2 1.1
.00005 .006 .002 ⬍.00001
Iron metabolism, antimicrobial Chaparone Generation of histamine Hormone
NOTE. Down-regulated gene expression was identified through gastric expression profiling of wild-type mice, gastrin KO mice, gastrin KO mice, after gastrin infusion for 1 week, and omeprazole-treated wild-type mice. From each group of mice, 4 expression profiles were generated by pooling equal amounts of gastric RNA from 5 mice. Expression indexes for wild-type mice and fold changes to these using wild-type as baseline are given. t test was used to determine the level of significance between gene expression in wild-type and gastrin KO mice.
bles 4 and 5. Most of the genes were also activated during omeprazole treatment and could be divided into 2 groups: (1) immune defense genes that are expressed by the immune system itself or expressed in enterocytes after activation by cytokines and interferon (Table 4) and (2) genes predominantly expressed by the intestines (Table 5). Immune system and defense genes. The source of immunoglobulin synthesis was lymphocytes that infiltrated the mucosa of KO mice but not of wild-type mice (Figure 3D and F and Table 4). Furthermore, the products of 3 of the induced genes (Best5, Schlafen, and ISG15) were localized to the mucosal cells and parietal cells (Figure 4B, F, and H). These genes are known to be regulated by interferon, and the expression of interferon gamma was increased in the gastrin KO mice (Table 6). Best5 was only marginally activated in gastrin KO mice (Figure 4B and C) but abundant although not uniformly expressed in the foveolar hyperplastic polyps (Figure 4D). Gastrin infusion in KO mice reduced the expression of most of the “immune system and defense genes,” but not to the level of expression seen in wild-type mice. Achlorhydria after PPI treatment induced the same response as in gastrin KO mice but at a lower level (Table 4).
Intestinal and other genes. These genes are predominantly expressed in the intestine and only occasionally in the stomach (Table 5). The expression increased ⬃1.5-3–fold in gastrin KO mice, which is less than the group of immune and defense genes (Table 4). One week of gastrin infusion suppressed but did not normalize the expression of “intestinal genes.” Omeprazole treatment also activated the expression of these genes, although the level of expression was lower than in gastrin KO mice (Table 5). These results were confirmed by quantitative polymerase chain reaction (Table 6). The transcription factors Cdx1 and Cdx2 are involved in intestinal metaplasia28 –30 and control the expression of many intestinal genes. While the Cdx2 expression was increased in the gastrin KO mice and normalized after gastrin infusion, Cdx1 expression was increased only in the 1-year-old mice but not 12- to 16-week-old gastrin KO mice (Table 7).
Age-Induced Gastric Changes Despite normal gastric morphology at 12–16 weeks (Figure 3A and B), major metaplastic changes, including accumulation of mucins within the oxyntic mucosa, were present in the gastrin KO mice after 12 months (Figure 3C and D). Infiltration of lymphocytes
250
FRIIS–HANSEN ET AL
GASTROENTEROLOGY Vol. 131, No. 1
Figure 3. Intestinal metaplasia in gastrin-deficient mice. (A and B) Periodic acid–Schiff/Alcian staining of the stomach showed that there are no major differences between young (12–16 weeks old) gastrin KO and wildtype mice. However, at 1 year of age, there are increasing signs of metaplasia with the accumulation of neutral and acid mucins. (D and I) Furthermore, lymphocytes infiltrated both the fundic and the antral mucosa (arrow). At 1½–2 years of age, there is increasing accumulation of mucins within the gastric glands, (H and I) development of cyst changes, and (J) ultimately the development of resulting polyp/tumor development. Immunohistochemical identification of lymphocytes is shown in Supplementary Figure 1 (see supplemental material online at www.gastrojournal. org). Bar ⫽ 50 m.
was found in both fundic and antral mucosa in many gastrin KO mice (Figure 3D and Supplementary Figure 1; see supplemental material online at www.gastrojournal.org). In gastrin KO mice, Cdx2 was expressed within
the metaplastic lesions (Figure 5C–E). Wild-type mice, in contrast, had duodenal but not gastric Cdx2 expression (Figure 5A, B, and F). During the next 6 –12 months, the intestinal metaplasia progressed and mice
July 2006
INTESTINAL METAPLASIA AND TUMORS IN GASTRIN KO MICE
251
Table 4. Changes due to Achlorhydria in Gastric Expression of Immunedefense Genes Fold change
Gene Interferon-induced protein with tetratricopeptide repeats 1 Immunoglobulin heavy chain 4 (serum IgG1) Best-5 Schlafen 3 ISG15 Immunoglobulin joining chain Molecule possessing ankyrin repeats induced by lipopolysaccharide (MAIL) Schlafen 4 Polymeric immunoglobulin receptor Immunoglobulin heavy chain Interferon-induced protein with tetratricopeptide repeats 3 Interferon-inducible guanosine triphosphatase Small inducible cytokine A6 Interferon regulatory factor 7
Wild-type ⫹ omeprazole
P value
5.5 5.9 2.1 1.8 1.1 4.2
3.8 4.8 5.0 1.1 ⫺1.0 3.6
.002 .008 .002 .002 .001 .002
3.7 2.4 2.3 2.2
2.4 ⫺1.0 1.7 2.6
⫺1.5 ⫺1.1 1.7 2.9
.001 .002 .000 .022
1.8 1.8 1.7 1.6
1.0 1.1 1.3 1.1 Wild-type ⫽ 1.0
1.2 1.2 1.4 1.8
.013 .017 .001 .003
Accession no.
Wild-type expression
KO
U43084 X88902 AA204579 AF099974 AV152244 M90766
9 15 87 21 22 36
16.0 4.9 4.8 4.4 4.3 3.8
AW060549 AF099977 AB001489 AF042798
219 173 404 49
U43086 AA914345 M58004 U73037
50 27 148 177
KO ⫹ gastrin
NOTE. Expression profiles were generated from wild-type mice, gastrin KO mice, gastrin KO mice, after gastrin infusion for 1 week, and omeprazole-treated wild-type mice. Wild-type expression index was 1.0, and fold changes to these are given. t test was used to determine the level of significance between gene expression in wild-type and gastrin KO mice. Selected genes are shown.
Table 5. Activation of Nonimmune Response Genes in Gastrin KO Mice, Display of Selected Genes Fold change
Gene Zymogen granule membrane protein 16 Human mRNA for Immunoglobulin G Fc binding protein Cysteine-rich intestinal protein Polymeric immunoglobulin receptor Chloride channel calcium activated 3 ␣ 1-microglobulin/bikunin Villin Melanoma differentiation associated protein 5 Arginine-rich, mutated in early stage tumors Crumbs related ELAV-like 1 (Hu antigen R) Claudin 7 Crp-ductin
Accession Wild-type no. expression KO
KO ⫹ gastrin
Wild-type ⫹ omeprazole
Intestinal P value expression
Function
AI593999
559
3.1
⫺1.1
1.1
.004
*
Unknown
AI465965
603
2.7
1.6
1.5
.000
*
Immune modulation
M13018 AB001489
838 404
2.4 2.3
2.0 1.7
1.3 1.7
.003 .000
* *
Immune modulation Immunoglobulin A transport
AV373378
2054
2.1
1.5
1.4
.000
*
Chloride channel
X68680 M98454 AA959954
53 311 145
2.1 2.0 1.8
1.7 1.9 1.3
2.0 1.2 1.1
.003 .005 .000
*
AW122364
482
1.5
1.2
1.4
.004
AI845823 U65735 AF087825 U37438
336 292 95 567
1.4 1.4 1.8 1.8
1.2 1.2 1.4 1.5
1.3 1.6 1.1 1.3
.001 .006 .000 .000
*
Anti-inflammatory role Microvilli turnover RNA metabolism and apoptosis RNA binding Tight junctions RNA stability Tight junction Immune defense and epithelial functions
Wild-type ⫽ 1.0 NOTE. Several of the genes are primarily expressed in the intestine. Their presence reflects molecular signs of intestinal metaplasia. The genes were identified through gastric expression profiling of wild-type mice, gastrin KO mice, gastrin KO mice after gastrin infusion for 1 week, and omeprazole-treated wild-type mice. From each group of mice, 4 expression profiles were generated by pooling equal amounts of gastric RNA from 5 mice. Expression index for wild-type mice and fold changes to these using wild-type mice as baseline are given. t test was used to determine the level of significance between gene expression in wild-type and gastrin KO mice.
252
FRIIS–HANSEN ET AL
GASTROENTEROLOGY Vol. 131, No. 1
Figure 4. Gastric mucosal expression of inflammatory response genes in gastrin KO mice. Expression of Best5 is not detectable in (A) fundic mucosal cells in wild-type mice; expression is activated in (B) fundic cells but not (C) antral cells. (D) However, in antral polyps in the gastrin KO mice, there is an abundant expression (the arrows indicate clusters of cells with abundant Best5 expression). Schlafen4 is not detectable in (E) wild-type but present in (F) the nucleus of mucosal cells in the gastrin KO mice (the arrows point to typical cells). ISG15 is not detectable in (G) wild-type but present in (H) the nucleus of mucosal cells in gastrin KO mice (some of these are indicated by the arrows). The mice in A–C and E–H were 12–16 months old except in D, where they were 1½ years old.
aged 1½–2 years developed cysts within the glands (Figure 3H). The antral mucosa proliferated (Figure 3I), and most of the mice (6 of 8 mice) developed foveolar hyperplasia and polyps (Figure 3J). There was ulceration, Table 6. Gastric Gene Expression Activated in Mice (Aged 12–16 Weeks) by Achlorhydria and Gastrin Gene
Wild-type
Interferon gamma mRNA 1.0 ⫾ 0.1 Best5 mRNA 1.00 ⫾ 0.1 ISG-15 mRNA 1.0 ⫾ 0.1 Schl4 mRNA 1.0 ⫾ 0.1 Cdx1 mRNA 1.0 ⫾ 0.1 Cdx2 mRNA 1.0 ⫾ 0.1 IgG-FC1 mRNA 1.0 ⫾ 0.1 Muclin mRNA 1.0 ⫾ 0.1 MUC2 mRNA 1.0 ⫾ 0.1 SPRP mRNA 1.0 ⫾ 0.2 Ankyrin mRNA 1.0 ⫾ 0.1 Heparin-binding epidermal growth factor mRNA 1.0 ⫾ 0.1 Insulin-like growth factor binding protein mRNA 1.00 ⫾ 0.14 NOV mRNA 1.0 ⫾ 0.2 Syndecan 4 mRNA 1.00 ⫾ 0.14 Hepcidin mRNA 1.0 ⫾ 0.2
Gastrin KO
Gastrin KO ⫹ gastrin
4.5 ⫾ 1.9a 4.7 ⫾ 0.9a 4.5 ⫾ 0.9a 3.5 ⫾ 0.9a 1.9 ⫾ 0.4a 1.9 ⫾ 0.4a 3.3 ⫾ 0.5a 4.6 ⫾ 0.7a 1.6 ⫾ 0.2a 3.1 ⫾ 0.5a 0.48 ⫾ 0.08a
2.1 ⫾ 1.3b 1.4 ⫾ 0.2b 1.1 ⫾ 0.3b 1.5 ⫾ 0.2b 1.3 ⫾ 0.07b 1.3 ⫾ 0.07b 1.3 ⫾ 0.2b 2.8 ⫾ 0.6a 1.3 ⫾ 0.1b 2.0 ⫾ 0.4b 1.0 ⫾ 0.2b
0.7 ⫾ 0.2b
0.7 ⫾ 0.2b
0.57 ⫾ 0.09a 1.75 ⫾ 0.49a 0.55 ⫾ 0.07a 1.2 ⫾ 0.1b 0.57 ⫾ 0.09a 1.75 ⫾ 0.49a 0.4 ⫾ 0.1a 1.3 ⫾ 0.2b
NOTE. Quantitative polymerase chain reaction analysis of gene expression activated in the gastrin KO mice and normalized/reduced after gastrin infusion. Mean ⫾ SEM of expression in wild-type given in arbitrary units (n ⫽ 8) in gastrin KO mice (n ⫽ 8) and gastrin KO mice that have received 1 week of gastrin infusion subcutaneously (n ⫽ 9). See Materials and Methods. aP ⬍ .05; bP ⬎ .05 (NS).
intestinal metaplasia, and Cdx2 expression within these polyps (Figure 5C–E). The morphologic changes were paralleled by increased expression of intestinal genes, growth factors, interferon gamma, and inflammatory response genes (Table 7). Down-regulated genes. Thirteen genes were down-regulated in gastrin KO mice and normalized or up-regulated following gastrin infusion in gastrin KO mice (Table 3). The PPI treatment resulted in mild hypergastrinemia only. Consistently, the expression of these genes was only slightly up-regulated (Tables 6 and 7). Immunohistochemistry localized the expression of IGFBP2, syndecan 4, and NOV to the parietal cells and to the mucosal cells (Figure 6). The expression of other members of the C-C knot (CCN) or IGF binding protein (IGFBP) families was unchanged (Table 7). Two other genes important for growth were also down-regulated: the extracellular matrix protein syndecan 4 and the LIM domain transcription regulator LIM only 2 (Lmo2). Additionally, 2 secreted peptides were down-regulated in KO mice: hepcidin, an antimicrobial peptide involved in iron metabolism, and CCL27, a chemokine.31,32 Genes Not Affected by Gastrin Deficiency Gastrin has been reported to regulate the expression of genes encoding regenerating protein and epidermal growth factor family, genes encoding proteins belonging to the acid secretory machinery, and genes encoding mucins and trefoil factors. However, none of these were differentially regulated in young gastrin KO mice (Table 8). Instead, their expression increased in
July 2006
INTESTINAL METAPLASIA AND TUMORS IN GASTRIN KO MICE
253
Table 7. Changes in Gastric Gene Expression in Aging Gastrin KO Mice Gene
Wild-type at 12 Weeks
Gastrin KO at 12 Weeks
Wild-type at 52 Weeks
Gastrin KO at 52 Weeks
Interferon gamma mRNA Cdx1 mRNA Cdx2 mRNA Transforming growth factor ␣ mRNA Heparin-binding epidermal growth factor mRNA Betacellulin mRNA Amphiregulin mRNA Trefoil factor 1 mRNA Trefoil factor 2 mRNA IgG-FC mRNA Muclin mRNA MUC2 mRNA SPRP mRNA Best5 mRNA ISG-15 mRNA Schl4 mRNA
1.0 ⫾ 0.1 1.0 ⫾ 0.1 1.0 ⫾ 0.2 1.0 ⫾ 0.1 1.0 ⫾ 0.1 1.0 ⫾ 0.1 1.0 ⫾ 0.1 1.0 ⫾ 0.1 1.0 ⫾ 0.1 1.0 ⫾ 0.1 1.0 ⫾ 0.1 1.0 ⫾ 0.1 1.0 ⫾ 0.6 1.0 ⫾ 0.1 1.0 ⫾ 0.1 1.0 ⫾ 0.1
4.5 ⫾ 1.9a 1.2 ⫾ 0.3b 1.7 ⫾ 0.2a 1.4 ⫾ 0.3b 1.0 ⫾ 0.2b 1.2 ⫾ 0.2b 2.5 ⫾ 0.6a 5.8 ⫾ 1.3a 1.6 ⫾ 0.2a 5.8 ⫾ 1.3a 2.4 ⫾ 0.6a 1.2 ⫾ 0.1b 3.1 ⫾ 1.4a 2.0 ⫾ 0.5a 2.4 ⫾ 0.3a 2.4 ⫾ 0.6a
2.1 ⫾ 1.3b 1.1 ⫾ 0.1b 0.9 ⫾ 0.1b 0.9 ⫾ 0.1b 0.7 ⫾ 0.1a 1.8 ⫾ 0.2a 1.1 ⫾ 0.1b 1.4 ⫾ 0.2b 1.4 ⫾ 0.3b 1.6 ⫾ 0.3b 2.1 ⫾ 0.3a 1.5 ⫾ 0.2a 1.2 ⫾ 0.4b 0.9 ⫾ 0.1b 1.1 ⫾ 0.1b 1.4 ⫾ 0.2b
11.7 ⫾ 3.6a 2.8 ⫾ 0.3a 2.3 ⫾ 0.1a 1.6 ⫾ 0.2a 2.1 ⫾ 0.3a 2.1 ⫾ 0.3a 3.6 ⫾ 0.8a 6.6 ⫾ 1.3a 4.4 ⫾ 1.0a 6.6 ⫾ 1.3a 3.9 ⫾ 0.3a 4.4 ⫾ 1.0a 2.4 ⫾ 1.2a 2.4 ⫾ 0.6a 2.6 ⫾ 0.8a 6.1 ⫾ 2.8a
NOTE. Real-time polymerase chain reaction quantitation of gastric gene expression (arbitrary units) in wild-type mice (n ⫽ 8), gastrin KO mice aged 12–16 weeks (n ⫽ 8), and gastrin KO mice aged 52 weeks (n ⫽ 9) (see Materials and Methods). aP ⬍ .05; bP ⬎ .05 (NS).
parallel with the metaplastic changes in the old gastrin KO mice (Table 7).
Discussion Hypergastrinemia is well known to cause gastric or duodenum ulcer and ECLoma.9,10,33 This study shows that lack of gastrin also predisposes to gastric disease through achlorhydria. Achlorhydria due to lack of gastrin or PPI treatment resulted in specific changes in gastric gene expression. These changes were more prominent in the gastrin KO mice than in PPI-treated wild-type mice and correlated with the greater reduction of acid secretion due to gastrin deficiency than PPI treatment. Exogenous gastrin restored gastric acidity (as shown by this study, FriisHansen et al,17 and Chen et al20) and reversed the changes in gene expression in young mice. However, progressive and irreversible intestinal metaplastic changes develop in aging gastrin KO mice and result in development of tumors (as shown by this study and Zavros et al23). The gastrin-deficient mouse may therefore serve as a new model for how Helicobacter-negative achlorhydria predispose to gastric cancer. The gastric mucosal expression of genes normally expressed in the intestines indicates that the molecular adaptation to achlorhydria is an early event. Activation of the transcription factor Cdx2 could be implicated because its expression is inversely correlated with luminal acidity in the duodenum.34 This may also explain the partial reversibility after gastrin infusion that normalizes acid secretion. The progressive morphologic abnormali-
ties were associated with further expression of Cdx2 and activation of Cdx1 expression, a late step in intestinal metaplasia and a sign of malignant transformation.35 Acidic conditions are also needed to maintain the gastric mucosa through the transcription of sonic hedgehog (Shh).36 Loss of Shh leads to intestinal transformation of the stomach.37 Furthermore, other achlorhydric mouse models also develop intestinal metaplasia; the hypergastrinemic mice deficient for the ␣ subunit of the gastric H,K–adenosine triphosphatase38 and the gp130 mutant mice develop achlorhydria and hypogastrinemia during the period they develop gastric cancer.39 Together, the studies demonstrate that achlorhydria independently of gastrin expression predisposes to intestinal metaplasia. Achlorhydria also reduces gastric elimination of bacteria22 and results in bacterial overgrowth in gastrin KO mice (this study and Zavros et al40). Earlier studies have shown that gastric infection activates the secretion of cytokines and antimicrobial peptides such as -defensins 1 and 2.41 In this study, achlorhydric gastritis and bacterial overgrowth also activated epithelial expression of other genes activated by interferons (for instance, Best-542 and ISG-1543). Gastric,41 intestinal,44,45 and pulmonary epithelial cells46 also activate the expression of genes belonging to the innate immune defense in the presence of infections. Thus, the immune response in the stomach following bacterial overgrowth may reflect a general epithelial cell response to infectious agents. The bacterial overgrowth in gastrin KO mice could promote gastric carcinogenesis through the generation of carcinogenic substances.47,48 The enterococci in the stomach convert nitrate to N-nitrosamines, which pre-
254
FRIIS–HANSEN ET AL
GASTROENTEROLOGY Vol. 131, No. 1
Figure 5. Cdx2 expression in wild-type and KO gastric mucosa. (A and B) Cdx2 expression was absent in wild-type fundic and antral mucosa. However, in the gastrin KO mice, Cdx2 expression was found in (C and D) metaplastic fundic mucosa as well as (E) some parts of the antral foveolar polyps that developed in 1½-year-old gastrin KO mice. (F) The abundant Cdx2 expression in the wild-type duodenum. The tissues in A–D and F were from 1-year-old mice, whereas the tissue in E is from 1/12-year-old mice. Bar ⫽ 100 m.
Figure 6. Immunocytochemical localization of IGF-BP2, syndecan 4, and NOV in wild-type gastric mucosa. Immunocytochemistry was used to localize 3 of the genes whose expression was down-regulated in the gastrin KO mice. IGF-BP2 was expressed in few cells (arrow) in the base of the glands. Syndecan 4 was detected in the extracellular space at low levels throughout the mucosa; areas with stronger staining are indicated by the arrows. NOV was expressed by few cells in each gland (arrows). All mice were 12–16 weeks old.
July 2006
INTESTINAL METAPLASIA AND TUMORS IN GASTRIN KO MICE
255
Table 8. Analysis of the Expression of Genes Belonging to the Acid Secretory Machinery or Involved In Mucus Formation in Gastrin-Deficient and Hypochlorhydric Mice. Fold change
Gene Acid secretion ATPase, H⫹/K⫹ transporting, ␣ polypeptide ATPase, H⫹/K⫹ transporting,  polypeptide Ezrin LIM and SH3 protein 1 (LASP-1) Trefoil factors Trefoil factor 1 Trefoil factor 2 (spasmolytic protein 1) Trefoil factor 3 (intestinal) Mucin family Mucin 1, transmembrane Mucin 1, transmembrane Mucin 3, intestinal Mucin 5, subtypes A and C CCN family Nephroblastoma overexpressed gene WNT1 inducible signaling pathway protein 1 Cysteine-rich protein 61 WNT1 inducible signaling pathway protein 2 Insulin-like growth factor binding protein family Insulin-like growth factor binding protein 1 Insulin-like growth factor binding protein 3 Insulin-like growth factor binding protein 4 Insulin-like growth factor binding protein 5 Insulin-like growth factor binding protein 6 Insulin-like growth factor binding protein 7 EGF family Heparin-binding epidermal growth factor–like growth factor Transforming growth factor ␣ Cripto Amphiregulin Epidermal growth factor Betacellulin Reg protein family Regenerating islet-derived 1 Regenerating islet-derived 2 Regenerating islet-derived 3 ␣ Regenerating islet-derived 3  Regenerating islet-derived 3 ␥
Accession no.
Wild-type ⫹ omeprazole
Wild-type expression
KO
KO ⫹ gastrin
U17282
7204
⫺1.2
⫺1.0
1.1
.00
M80251 X60671 AW122780
6588 1413 1610
⫺1.2 1.0 1.0
⫺1.0 1.1 1.1
1.1 1.2 1.1
.01 .88 .47
Z21858
8889
1.1
⫺1.1
1.0
.18
U78770 D38410
6439 807
⫺1.0 1.0
1.0 1.0
⫺1.0 1.0
.78 .82
M84683 U16175 AF027131 AJ010792
2209 912 458 3935
1.1 ⫺1.1 ⫺1.2 1.0
1.1 ⫺1.0 ⫺1.2 1.1
1.1 1.1 1.1 ⫺1.0
.05 .22 .02 .78
Y09257
171
⫺1.2
1.2
1.2
.19
AF100777 M32490
130 97
⫺1.1 ⫺1.1
⫺1.2 ⫺1.0
⫺1.1 1.2
.38 .50
AF100778
166
1.2
1.1
1.2
.30
X81579
150
1.1
1.0
1.5
.41
X81581
172
1.1
1.0
1.1
.51
X76066
256
1.6
1.8
1.9
.004
L12447
542
⫺1.2
⫺1.1
1.2
.04
X81584
267
⫺1.0
⫺1.0
1.1
.59
AB012886
855
1.0
1.0
1.8
.98
L07264 M92420 U57720 L41352 V00741 L08394
217 81 290 116 70 42
⫺1.1 1.1 ⫺1.0 1.4 ⫺2.4 1.7
1.0 1.3 ⫺1.1 1.4 ⫺1.9 1.4
1.2 1.1 ⫺1.0 ⫺1.1 ⫺1.9 1.6
.51 .31 .57 .02 .03 .02
D14010 D14011 D63357
341 194 222 Not on the chip 173
⫺1.1 ⫺1.6 ⫺1.0
⫺1.1 ⫺1.3 1.0
⫺1.0 ⫺1.4 ⫺1.1
.57 .04 .95
1.1 Wild-type ⫽ 1.0
⫺1.2
.46
D63362
1.1
P value
NOTE. Gene expression profiling of wild-type mice, gastrin KO mice, gastrin KO mice after gastrin infusion for 1 week, and omeprazole-treated wild-type mice was performed. From each group of mice, 4 expression profiles were generated by pooling equal amounts of gastric RNA from 5 mice. Expression index for wild-type mice and fold changes to these using wild-type mice as baseline are given. t test was used to determine the level of significance between gene expression in wild-type and gastrin KO mice.
256
FRIIS–HANSEN ET AL
dispose to cancer.49,50 In contrast, Helicobacter does not have this capability.51 Increased production of N-nitrosamines51,52 is often seen in permanent achlorhydria due to either atrophic gastritis or antrectomy and predisposes to gastric cancer.13–15,53,54 In contrast, long-term PPI treatment is not associated with increased formation of N-nitrosamines55 and is considered safe.56 However, the initial dose (20 – 40 mg omeprazole daily in humans) was initially chosen to avoid severe hypergastrinemia and did not completely block acid secretion. The introduction of newer, more efficient drugs demands a repetition of these studies, preferably with longer follow-up.57 Taken together, complete inhibition of acid secretion favors bacterial overgrowth55,58 that activates an immune response that alters gene expression. Furthermore, carcinogenic substances can also be generated.47,48 Antibiotic treatment of young gastrin KO mice reversed the fundic and antral hyperplasia.40 However, it is presently not known whether sterilization of the gastric juice will also prevent the metaplasia. In contrast to the major changes in gene expression due to achlorhydria and inflammation, the expression of only few genes depended on gastrin. Two of these (NOV and syndecan 4) encode matrix proteins that modulate extracellular signaling59,60 and are important for cell maturation, differentiation, migration, and adhesion. Altered synthesis of extracellular matrix proteins has been associated with gastric disease in Foxl1 KO mice.61 Gastrin-controlled modulation of the extracellular matrix could be a mechanism behind gastrin-controlled growth.62,63 The expression of growth factor known to be activated by hypergastrinemia (trefoil factor 1, transforming growth factor ␣, and heparin-binding epidermal growth factor)8,64,65 were unaffected in young gastrin KO mice. Furthermore, the gastric expression of these increased in parallel with the morphologic abnormalities in the gastrin KO mice (this study and Franic et al66) and is thus controlled by factors other than gastrin. In conclusion, the most important action of gastrin is to maintain gastric homeostasis through control of acid secretion because only few genes operate under direct gastrin control. Achlorhydria is associated with an intestinal expression profile and activation of immune defense genes. Achlorhydria also predisposes to gastric bacterial overgrowth with bacteria that have the capacity to form carcinogenic N-nitrosamines.
Appendix Supplemetary data Supplemetary data associated with this article can be found, in the online version, at doi:10.1053/j.gastro. 2006.04.031
GASTROENTEROLOGY Vol. 131, No. 1
References 1. Håkanson R, Chen D, Sundler F. The ECL cells. In: Johnson LR, ed. Physiology of the gastrointestinal tract. 3rd ed. New York, NY: Raven, 1994:1171–1184. 2. Tömmerås K, Hammer P, Sundler F, Borch K, Mårdh S, Cabero JL. Immunolocalization of cholecystokinin-2 receptors in rat gastric mucosa. Scand J Gastroenterol 2002;37:1017–1024. 3. Tömmerås K, Bakke I, Sandvik AK, Larsson E, Waldum HL. Rat parietal cells express CCK(2) receptor mRNA: gene expression analysis of single cells isolated by laser-assisted microdissection. Biochem Biophys Res Commun 2002;297:335–340. 4. Soll AH, Berglindh T. Receptors that regulate gastric acid-secretory function. In: Johnson L, ed. Physiology of the gastrointestinal tract. 3rd ed. New York, NY: Raven, 1994:1139 –1169. 5. Dimaline R, Sandvik AK. Histidine decarboxylase gene expression in rat fundus is regulated by gastrin. FEBS Lett 1991;281:20 –22. 6. Dimaline R, Evans D, Forster ER, Sandvik AK, Dockray GJ. Control of gastric corpus chromogranin A messenger RNA abundance in the rat. Am J Physiol 1993;264:G583–G588. 7. Swarovsky B, Eissele R, Gross M, Kortner G, Koop H, Arnold R, Simon B. Expression of the gastric H⫹/K(⫹)-ATPase and histidine decarboxylase during omeprazole treatment. Digestion 1994;55:97–102. 8. Khan ZE, Wang TC, Cui G, Chi AL, Dimaline R. Transcriptional regulation of the human trefoil factor, TFF1, by gastrin. Gastroenterology 2003;125:510 –521. 9. Rehfeld JF. The new biology of gastrointestinal hormones. Physiol Rev 1998;78:1087–1088. 10. Dockray GJ, Varro A, Dimaline R, Wang T. The gastrins: their production and biological activities. Annu Rev Physiol 2001;63: 119 –139. 11. Koh TJ, Chen D. Gastrin as a growth factor in the gastrointestinal tract. Regul Pept 2000;93:37– 44. 12. Wang TC, Dangler CA, Chen D, Goldenring JR, Koh T, Raychowdhury R, Coffey RJ, Ito S, Varro A, Dockray GJ, Fox JG. Synergistic interaction between hypergastrinemia and Helicobacter infection in a mouse model of gastric cancer. Gastroenterology 2000;118:36 – 47. 13. Lundegardh G, Adami HO, Helmick C, Zack M, Meirik O. Stomach cancer after partial gastrectomy for benign ulcer disease. N Engl J Med 1988;319:195–200. 14. Caygill CP, Hill MJ, Kirkham JS, Northfield TC. Mortality from gastric cancer following gastric surgery for peptic ulcer. Lancet 1986;1:929 –931. 15. Viste A, Björnestad E, Opheim P, Skarstein A, Thunold J, Hartveit F, Eide GE, Eide TJ, Söreide O. Risk of carcinoma following gastric operations for benign disease. A historical cohort study of 3470 patients. Lancet 1986;2:502–505. 16. Koh TJ, Goldenring JR, Ito S, Mashimo H, Kopin AS, Varro A, Dockray GJ, Wang TC. Gastrin deficiency results in altered gastric differentiation and decreased colonic proliferation in mice. Gastroenterology 1997;113:1015–1025. 17. Friis-Hansen L, Sundler F, Li Y, Gillespie PJ, Greenson JL, Owyang C, Rehfeld JF, Samuelson LC. Impaired gastric acid secretion in gastrin deficient mice. Am J Physiol 1998;274:G561–G568. 18. Langhans N, Rindi G, Chiu M, Rehfeld JF, Ardman B, Beiborn M, Kopin AS. Abnormal gastric histology and decreased acid production in cholecystokinin-B/gastrin receptor deficient mice. Gastroenterology 1997;112:280 –286. 19. Nagata A, Ito M, Iwata N, Kuno J, Takano H, Minowa O, Chihara K, Matsui T, Noda T. G-protein-coupled cholecystokinin-B/gastrin receptors are responsible for physiological cell growth of the stomach in vivo. Proc Natl Acad Sci U S A 1996;93:11825– 11830. 20. Chen D, Zhao C-M, Håkanson R, Samuelson LC, Rehfeld JF, Friis-Hansen L. Altered control of gastric acid secretion in gastrin-
July 2006
21.
22.
23.
24.
25. 26.
27.
28.
29.
30.
31.
32. 33. 34.
35.
36.
37.
38.
cholecystokinin double mutant mice. Gastroenterology 2004; 126:476 – 487. Chen D, Zhao CM, Dockray GJ, Varro A, Van Hoek A, Sinclair NF, Wang TC, Koh TJ. Glycine-extended gastrin synergizes with gastrin 17 to stimulate acid secretion in gastrin-deficient mice. Gastroenterology 2000;119:756 –765. Sun FJ, Kaur S, Ziemer D, Banerjee S, Samuelson LC, De Lisle RC. Decreased gastric bacterial killing and up-regulation of protective genes in small intestine in gastrin-deficient mouse. Dig Dis Sci 2003;48:976 –985. Zavros Y, Eaton KA, Kang W, Rathinavelu S, Katukuri V, Kao JY, Samuelson LC, Merchant JL. Chronic gastritis in the hypochlorhydric gastrin-deficient mouse progresses to adenocarcinoma. Oncogene 2005;24:2354 –2366. Li C, Wong WH. Model-based analysis of oligonucleotide arrays: expression index computation and outlier detection. Proc Natl Acad Sci U S A 2001;98:31–36. Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ. Basic local alignment search tool. J Mol Biol 1990;215:403– 410. Rehfeld JF, Stadil F, Rubin B. Production and evaluation of antibodies for the radioimmunoassay of gastrin. Scand J Clin Lab Invest 1972;30:221–232. Shay H, Sun DCH, Gruenstein M. A quantitative method for measuring spontaneous gastric acid secretion in the rat. Gastroenterology 1954;26:906 –913. Mizoshita T, Inada K, Tsukamoto T, Kodera Y, Yamamura Y, Hirai T, Kato T, Joh T, Itoh M, Tatematsu M. Expression of Cdx1 and Cdx2 mRNAs and relevance of this expression to differentiation in human gastrointestinal mucosa—with special emphasis on participation in intestinal metaplasia of the human stomach. Gastric Cancer 2001;4:185–191. Bai YQ, Yamamoto H, Akiyama Y, Tanaka H, Takizawa T, Koike M, Kenji Yagi O, Saitoh K, Takeshita K, Iwai T, Yuasa Y. Ectopic expression of homeodomain protein CDX2 in intestinal metaplasia and carcinomas of the stomach. Cancer Lett 2002;176:47– 55. Silberg DG, Sullivan J, Kang E, Swain GP, Moffett J, Sund NJ, Sackett SD, Kaestner KH. Cdx2 ectopic expression induces gastric intestinal metaplasia in transgenic mice. Gastroenterology 2002;122:689 – 696. Krause A, Neitz S, Magert HJ, Schulz A, Forssmann WG, SchulzKnappe P, Adermann K. LEAP-1, a novel highly disulfide-bonded human peptide, exhibits antimicrobial activity. FEBS Lett 2000; 480:147–150. Nibbs RJ, Graham GJ. CCL27/PESKY: a novel paradigm for chemokine function. Expert Opin Biol Ther 2003;3:15–22. Rozengurt E, Walsh JH. Gastrin, CCK, signaling, and cancer. Annu Rev Physiol 2001;63:49 –76. Faller G, Dimmler A, Rau T, Spaderna S, Hlubek F, Jung A, Kirchner T, Brabletz T. Evidence for acid-induced loss of Cdx2 expression in duodenal gastric metaplasia. J Pathol 2004;203: 904 –908. Eda A, Osawa H, Yanaka I, Satoh K, Mutoh H, Kihira K, Sugano K. Expression of homeobox gene CDX2 precedes that of CDX1 during the progression of intestinal metaplasia. J Gastroenterol 2002;37:94 –100. Dimmler A, Brabletz T, Hlubek F, Hafner M, Rau T, Kirchner T, Faller G. Transcription of sonic hedgehog, a potential factor for gastric morphogenesis and gastric mucosa maintenance, is upregulated in acidic conditions. Lab Invest 2003;83:1829 –1837. Ramalho-Santos M, Melton DA, McMahon AP. Hedgehog signals regulate multiple aspects of gastrointestinal development. Development 2000;127:2763–2772. Spicer Z, Miller ML, Andringa A, Riddle TM, Duffy JJ, Doetschman T, Shull GE. Stomachs of mice lacking the gastric H,K-ATPase alpha-subunit have achlorhydria, abnormal parietal cells, and ciliated metaplasia. J Biol Chem 2000;275:21555–21565.
INTESTINAL METAPLASIA AND TUMORS IN GASTRIN KO MICE
257
39. Judd LM, Alderman BM, Howlett M, Shulkes A, Dow C, Moverley J, Grail D, Jenkins BJ, Ernst M, Giraud AS. Gastric cancer development in mice lacking the SHP2 binding site on the IL-6 family co-receptor gp130. Gastroenterology 2004;126:196 –207. 40. Zavros Y, Rieder G, Ferguson A, Samuelson LC, Merchant JL. Genetic or chemical hypochlorhydria is associated with inflammation that modulates parietal and G-cell populations in mice. Gastroenterology 2002;122:119 –133. 41. Bajaj-Elliott M, Fedeli P, Smith GV, Domizio P, Maher L, Ali RS, Quinn AG, Farthing MJ. Modulation of host antimicrobial peptide (beta-defensins 1 and 2) expression during gastritis. Gut 2002; 51:356 –361. 42. Grewal TS, Genever PG, Brabbs AC, Birch M, Skerry TM. Best5: a novel interferon-inducible gene expressed during bone formation. FASEB J 2000;14:523–531. 43. Reid LE, Brasnett AH, Gilbert CS, Porter AC, Gewert DR, Stark GR, Kerr IM. A single DNA response element can confer inducibility by both alpha- and gamma-interferons. Proc Natl Acad Sci U S A 1989;86:840 – 844. 44. Seydel KB, Li E, Swanson PE, Stanley SL Jr. Human intestinal epithelial cells produce proinflammatory cytokines in response to infection in a SCID mouse-human intestinal xenograft model of amebiasis. Infect Immun 1997;65:1631–1639. 45. Eckmann L, Smith JR, Housley MP, Dwinell MB, Kagnoff MF. Analysis by high density cDNA arrays of altered gene expression in human intestinal epithelial cells in response to infection with the invasive enteric bacteria Salmonella. J Biol Chem 2000;275: 14084 –14094. 46. Ichikawa JK, Norris A, Bangera MG, Geiss GK, van’t Wout AB, Bumgarner RE, Lory S. Interaction of pseudomonas aeruginosa with epithelial cells: identification of differentially regulated genes by expression microarray analysis of human cDNAs. Proc Natl Acad Sci U S A 2000;97:9659 –9664. 47. Reed PI, Smith PL, Haines K, House FR, Walters CL. Effect of cimetidine on gastric juice N-nitrosamine concentration. Lancet 1981;2:553–556. 48. Stockbrugger RW, Cotton PB, Eugenides N, Bartholomew BA, Hill MJ, Walters CL. Intragastric nitrites, nitrosamines, and bacterial overgrowth during cimetidine treatment. Gut 1982;23:1048 – 1054. 49. Risch HA, Jain M, Choi NW, Fodor JG, Pfeiffer CJ, Howe GR, Harrison LW, Craib KJ, Miller AB. Dietary factors and the incidence of cancer of the stomach. Am J Epidemiol 1985;122:947– 959. 50. Gonzalez CA, Riboli E, Badosa J, Batiste E, Cardona T, Pita S, Sanz JM, Torrent M, Agudo A. Nutritional factors and gastric cancer in Spain. Am J Epidemiol 1994;139:466 – 473. 51. Vermeer IT, Gerrits MM, Moonen EJ, Engels LG, Dallinga JW, Kleinjans JC, van Maanen JM, Kuipers EJ, Kusters JG. Helicobacter pylori does not mediate the formation of carcinogenic Nnitrosamines. Helicobacter 2002;7:163–169. 52. Schlag P, Bockler R, Peter M. Nitrite and nitrosamines in gastric juice: risk factors for gastric cancer? Scand J Gastroenterol 1982;17:145–150. 53. Meining A, Morgner A, Miehlke S, Bayerdorffer E, Stolte M. Atrophy-metaplasia-dysplasia-carcinoma sequence in the stomach: a reality or merely an hypothesis? Best Pract Res Clin Gastroenterol 2001;15:983–998. 54. Thompson JC, Fender HR, Watson LC, Villar HV. The effects on gastrin and gastric secretion of five current operations for duodenal ulcer. Ann Surg 1976;183:599 – 608. 55. Yeomans ND, Brimblecombe RW, Elder J, Heatley RV, Misiewicz JJ, Northfield TC, Pottage A. Effects of acid suppression on microbial flora of upper gut. Dig Dis Sci 1995;40:81S–95S. 56. Lamberts R, Brunner G, Solcia E. Effects of very long (up to 10 years) proton pump blockade on human gastric mucosa. Digestion 2001;64:205–213.
258
FRIIS–HANSEN ET AL
57. Gillen D, McColl KE. Problems associated with the clinical use of proton pump inhibitors. Pharmacol Toxicol 2001;89:281–286. 58. Howden CW, Hunt RH. Relationship between gastric secretion and infection. Gut 1987;28:96 –107. 59. Shimo T, Gentili C, Iwamoto M, Wu C, Koyama E, Pacifici M. Indian hedgehog and syndecans-3 coregulate chondrocyte proliferation and function during chick limb skeletogenesis. Dev Dyn 2004;229:607– 617. 60. Brigstock DR. The CCN family: a new stimulus package. J Endocrinol 2003;178:169 –175. 61. Perreault N, Katz JP, Sackett SD, Kaestner KH. Foxl1 controls the Wnt/beta-catenin pathway by modulating the expression of proteoglycans in the gut. J Biol Chem 2001;276:43328 – 43333. 62. Tömmerås K, Chen Y, Rhedin M, Cabero JL, Mårdh S. Proliferation and differentiation of cells from explants of fetal rat stomach. Acta Physiol Scand 1997;159:155–161. 63. Tömmerås K, Cabero JL, Mårdh S. Expression of extracellular matrix proteins in the fetal rat gastric mucosa. Anat Embryol (Berl) 2000;201:149 –156. 64. Kinoshita Y, Ishihara S. Mechanism of gastric mucosal proliferation induced by gastrin. J Gastroenterol Hepatol 2000; 15(Suppl):D7–D11.
GASTROENTEROLOGY Vol. 131, No. 1
65. Fukui H, Kinoshita Y, Maekawa T, Okada A, Waki S, Hassan S, Okamoto H, Chiba T. Regenerating gene protein may mediate gastric mucosal proliferation induced by hypergastrinemia in rats. Gastroenterology 1998;115:1483–1493. 66. Franic TV, Judd LM, Nguyen NV, Samuelson LC, Loveland KL, Giraud AS, Gleeson PA, Van Driel IR. Growth factors associated with gastric mucosal hypertrophy in autoimmune gastritis. Am J Physiol Gastrointest Liver Physiol 2004;287:G910 –G918. 67. Nicklas W, Baneux P, Boot R, Decelle T, Deeny AA, Fumanelli M, Illgen-Wilcke B. Recommendations for the health monitoring of rodent and rabbit colonies in breeding and experimental units. Lab Anim 2002;36:20 – 42.
Received April 13, 2005. Accepted March 30, 2006. Address requests for reprints to: Lennart Friis-Hansen, MD, Department of Clinical Biochemistry, KB-3014, Rigshospitalet, 9 Blegdamsvej, DK-2100 Copenhagen, Denmark. e-mail: [email protected]; fax: (45) 3545-46-40. Supported by the Novo Nordic Foundation (to L.F.-H.) and the Swedish Research Council (64-04723 to T.W.). The authors thank Bo Lindberg for skillful technical assistance.
MICROARRAYS AND OTHER NEW TECHNOLOGIES Gastric Inflammation, Metaplasia, and Tumor Development in Gastrin–Deficient Mice LENNART FRIIS–HANSEN,* KLAUS RIENECK,‡ HANS–OLOF NILSSON,§ TORKEL WADSTRÖM,§ and JENS F. REHFELD* *Department of Clinical Biochemistry and ‡Institute for Inflammation Research, Rigshospitalet, University of Copenhagen, Copenhagen, Denmark; and §Department of Microbiology, University of Lund, Lund, Sweden
Background & Aims: Gastrin deficiency and proton pump inhibitor treatment cause achlorhydria, which predisposes to disease. To elucidate the underlying molecular biology, we examined the changes in gastric gene expression in both types of achlorhydria. We also explored the associated changes in the gastric microflora and the long-term consequences of gastrin-deficient achlorhydria. Methods: Expression profiles were generated from gastric RNA from wild-type mice, gastrin knockout (KO) mice, gastrin KO mice after 1 week of gastrin infusion, and wild-type mice treated for 1 month with a proton pump inhibitor. The results were confirmed using real-time polymerase chain reaction and immunohistochemistry. Selective media were used to characterize the gastric microflora. Results: The number of gastric bacteria was increased in both gastrin KO and PPItreated mice. The expression profiles revealed activation of immune defense genes, interferon-regulated response genes, and intestinal metaplasia of the gastric mucosa. In young gastrin-deficient mice, gastrin infusions reversed the changes. Over time, the changes accumulated, became irreversible, and progressed into metaplasia and polyp development. Finally, the study showed that gastrin regulated the expression of genes encoding extracellular matrix proteins. Conclusions: Independently of gastrin, achlorhydria is associated with gastric bacterial overgrowth and intestinal gene expression patterns and is associated with predisposition to disease. Gastrin is therefore essential for prevention of gastric disease, mainly through control of acid secretion but to a lesser extent also through control of gastric gene expression. The gastrin-deficient mouse serves as a useful new model for gastric metaplasia and neoplasia.
astrin stimulates gastric acid secretion via enterochromaffin-like cells1 and to a lesser degree through direct activation of the parietal cells.2– 4 Gastrin regulates gene expression in enterochromaffin-like,5,6 parietal,7 and other gastric cells8 (for review, see Rehfeld9 and
G
Dockray et al10). Long-term hypergastrinemia results in hypertrophy and hyperplasia of parietal and enterochromaffin-like cells11 and promotes development of gastric cancer.12 Despite the association between hypergastrinemia and neoplasia, antrectomized hypogastrinemic patients also have an increased risk of cancer in the gastric remnant.13–15 To examine the molecular biology of diseases due to lack of gastrin, mice deficient for gastrin16,17 or its receptor18,19 have been developed. They are achlorhydric due to enterochromaffin-like and parietal cell dysfunction, whereas the remaining mucosal growth is largely unaffected.16 –19 The defects in acid secretion are reversed by infusion of exogenous gastrin.17,20,21 Furthermore, gastrin knockout (KO) mice aged 4 –5 months developed gastric bacterial overgrowth, signs of intestinal metaplasia,22 and tumors.23 This study examines the early changes in gene expression due to lack of gastrin and its reversibility after gastrin infusion. The changes in gene expression were compared with those in achlorhydric but hypergastrinemic proton pump inhibitor (PPI)-treated mice.
Materials and Methods Mice and RNA Preparation Gastrin KO mice, backcrossed 7 times to C57BL6/J through both male and female germline, aged 9 –12 weeks were used.17 After removal of the forestomach, RNA was extracted using TRIzol (Invitrogen, Carlsbad, CA) from wildtype mice, gastrin KO mice, gastrin KO mice after 1 week of gastrin replacement (10 nmol · h⫺1 · kg⫺1) using minipumps (Alzet 2001; Alza Corp, Palo Alto, CA), and wild-type mice given a PPI (omeprazole 400 mol · kg⫺1) once daily for 1 Abbreviations used in this paper: KO, knockout; PPI, proton pump inhibitor. © 2006 by the American Gastroenterological Association Institute 0016-5085/06/$32.00 doi:10.1053/j.gastro.2006.04.031
July 2006
INTESTINAL METAPLASIA AND TUMORS IN GASTRIN KO MICE
Table 1. Characteristics for the Antisera Used for Immunocytochemistry Dilution Antigen Cdx2
1:10
Best5
1:5000
Schlafen3/4
1:3000
ISG-15
1:2000
CD45
1:4000
IGF-BP2
1:2000
Syndecan 4
1:3000
NOV
1:2000
Code and source AM392-5M, Biogenex, San Ramon, CA Professor Skerry, Heslington, United Kingdom sc-8903, Santa Cruz, Paso Robles, CA Professor Hass, Milwaukee, WI sc-1178, Santa Cruz, Paso Robles, CA sc-6001, Santa Cruz, Paso Robles, CA sc-9497, Santa Cruz, Paso Robles, CA Professor Perbal, Paris, France
Secondary antibodies with conjugates Envision⫹ (horseradish Prediluted peroxidase conjugated) Tertlary antibodies with conjugates Cy3-conjugated anti– 1:400 horseradish peroxidase
247
pressed as mean fold change with P values using the following settings: minimal difference between baseline and experiment was 50, expression levels ⬍20 were assigned the value 20, and P ⬍ .05 was adjusted so that the 90th percentile false discovery rate was ⬍5%. Expressed sequence tags (ESTs) were compared with GenBank database sequences using BLAST.25
Quantitative Polymerase Chain Reaction The GeneChip results were confirmed by real-time polymerase chain reaction with the Lightcycler (Roche, Mannheim, Germany) using the LightCycler - DNA Master SYBR Green Reaction kit (Roche) as described.20 Each experiment consisted of one negative control, 4 standards, and 7– 8 complementary DNA samples from each group of mice. The messenger RNA (mRNA) expressions were normalized to the glyceraldehyde-3-phosphate dehydrogenase expression in each complementary DNA sample. Statistical analysis was performed using Student t tests. The primer sequences are available on request.
Dako, Glostrup, Denmark
Gastric pH and Gastrin Measurements Jackson, West Grove, PA
month (n ⫽ 20 for each genotype). Long-term alterations were examined in 1-year-old and 1½-year-old mice.
Gene Expression Profiling Probes were generated from 4 pools of total RNA from each of the 4 groups of mice (5 per group) as described by the manufacturer (Affymetrix, Santa Clara, CA). The probe quality was checked on test arrays hybridization to GeneChips Mu74A, B and C, version 2 (Affymetrix).
Data Analysis The microarray data were analyzed using the dChip perfect match-mismatch model (software version 1.3; supplied by Dr Li, Department of Biostatistics, Harvard School of Public Health, Boston, MA).24 The expression in the wild-type mice was compared with that in gastrin KO mice, gastrin KO mice receiving gastrin infusion, and wild-type mice treated with omeprazole. The changes in gene expression were ex-
The gastrin concentration in plasma was measured by radioimmunoassay.26 Following pyloric ligation for 1 hour, pH was measured in the accumulated juice with a PHM 85 pH-meter (Radiometer, Copenhagen, Denmark).27
Immunohistochemistry For gastric histologic examination of mice (9 –11 weeks, 1 and 1½–2 years) stomachs were fixed overnight in 4% paraformaldehyde (pH 7) and embedded in paraffin (6 – 8 mice per time point per genotype). Immunohistochemistry was performed as described20 and using the antibodies and detection systems described in Table 1. H&E and periodic acid– Schiff/Alcian staining was also performed.
Microbiology Semiquantitative analysis of enterococci and lactobacilli per stomach (corpus/fundus and antrum) was performed by smearing half of a mouse stomach on bile-esculin agar and the other half on MRS agar. After 48 hours of aerobic incubation at 37°C, the number of colonies was counted. Single colonies were identified using the api-test (BioMéreux, Marcy, France) (n ⫽ 6 – 8 mice, aged 12–16 weeks from each of the 4 groups).
Table 2. Plasma Gastrin Concentrations and Gastric pH Characteristics of the Examined Mice Wild-type Plasma-gastrin (pmol/L) pH
42 ⫾ 14 3.5 ⫾ 0.2
Gastrin KO detecteda
Not 6.2 ⫾ 0.3a
Gastrin KO ⫹ gastrin infusion
Omeprazole-treated wild-type
415 ⫾ 3.6 ⫾ 0.2b
98 ⫾ 34a 5.7 ⫾ 0.4a
56a
NOTE. Plasma gastrin concentrations were measured using a radioimmunoassay in wild-type mice, gastrin KO mice, gastrin KO mice after 1 week of gastrin subcutaneous infusion, and omeprazole treated wild-type mice (see Materials and Methods). To measure the gastric acidity, the mice were anesthetized with intraperitoneal 2,2,2-tribromoethanol (Sigma-Aldrich Corp, St Louis, MO), the abdominal cavity opened, and the esophagus and pylorus ligated for 1 hour. n ⫽ 8 –9. aP ⬍ .05, bP ⬎ .05 (NS).
248
FRIIS–HANSEN ET AL
GASTROENTEROLOGY Vol. 131, No. 1
Results Gastrin and Gastric Acidity Gastrin KO and omeprazole-treated, hypergastrinemic mice were achlorhydric, whereas the KO mice that received gastrin infusion had normal gastric acidity despite hypergastrinemia (Table 2). Microbiology The achlorhydria due to lack of gastrin was associated with a partly reversible overgrowth of lactobacilli and enterococci. PPI treatment did not cause the same degree of bacterial overgrowth (Figure 1). Gastric Gene Expression Profiling General observations. Hierarchical cluster analysis separated each of the 4 groups. The 2 groups of KO mice were more closely related to each other than to the wild-type mice (Figure 2). The expression of chromogranin A and histidine decarboxylase (known to be regulated by gastrin5,6) was low in the gastrin KO mice, restored to wild-type levels after gastrin infusion, and increased in the omeprazole-treated mice (Table 3). Gastric somatostatin expression was unchanged, which is probably because this experiment only examined the total gastric expression and not fundic and antral expression independently. Up-regulated genes. Despite normal gastric morphology in young gastrin KO mice (Figure 3A and B), 59 genes were up-regulated in KO mice and their expression was normalized following gastrin infusion (Ta-
Figure 1. Semiquantitative analysis of the load of gastric enterococci and lactobacilli. The type and number of gastric bacteria were determined in wild-type, gastrin KO, omeprazole-treated wild-type, and gastrin KO mice after 1 week of gastrin infusion. All mice were 12–16 weeks old. Half of the stomach was smeared on bile-esculin agar and the other half on MRS agar. After 48 hours of aerobic incubation at 37°C, the number of colonies was counted. The data are given as 109 colonies per stomach. Single colonies were identified using the api-test (BioMéreux). The mice were housed in a barrier mouse facility with 12-hour light/dark cycles. The mouse colony was monitored according to the FELASA guidelines,67 and during the experiments the colony tested negative for all monitored viruses and bacteria except Helicobacter bilis. n ⫽ 6 – 8 in each group. *P ⬍ .05.
Figure 2. Hierarchical cluster analysis of gastric gene expression profiles from wild-type, gastrin KO, gastrin KO ⫹ gastrin, and omeprazole-treated wild-type mice. The 2 groups of wild-type mice were more closely related to each other than they were to the 2 groups of gastrin KO mice. Exogenous gastrin infusion restores acid secretion but does not normalize gene expression. Similarly, even though omeprazoletreated mice are achlorhydric, their gastric gene expression is still closer to wild-type mice than to the achlorhydric gastrin KO mice. (Gastric RNAs from 5 mice aged 12–16 weeks were pooled and subsequently analyzed using the Affymetrix u74 GeneChips.)
July 2006
INTESTINAL METAPLASIA AND TUMORS IN GASTRIN KO MICE
249
Table 3. Gastric Gene Expression Down-regulated due to Gastrin Deficiency Fold change
Gene Solute carrier family 4 (anion exchanger) member 2 Aquaporin 4 Insulin-like growth factor binding protein 2 Nephroblastoma overexpressed gene Cytochrome c oxidase, subunit VIIa 1 LIM only 2 Ankyrin 3, epithelial Small inducible cytokine A27 Syndecan 4 TYRO protein tyrosine kinase binding protein Hepcidin antimicrobial peptide Chromogranin A Histidine decarboxylase Gastrin
Accession no.
Wild-type expression
Wild-type ⫹ omeprazole
KO ⫹ gastrin
KO
P value
Function
J04036 U88623
1131 67
⫺1.3 ⫺1.3
1.0 1.4
1.6 3.5
.0004 .002
HCO3-transport Water channel
X81580
615
⫺1.4
⫺1.0
1.8
.002
Hormone
Y09257
171
⫺1.4
1.2
1.2
.004
Hormone
AF037370 M64360 L40631 AW124975 D89571
452 194 344 199 765
⫺1.4 ⫺1.4 ⫺1.5 ⫺1.5 ⫺1.5
⫺1.0 ⫺1.1 ⫺1.2 ⫺1.2 ⫺1.6
1.4 1.1 1.3 ⫺1.0 ⫺1.1
.01 .006 .004 .001 .0002
Apoptosis Angiogenesis Cytoskeletal Secreted Extracellular matrix
AF024637
280
⫺1.6
1.0
⫺1.2
.004
Unknown
AI255961 M64278 X57437 U34293
721 626 160 2999
⫺1.8 ⫺2.1 ⫺3.2 ⫺35.8
⫺1.4 1.3 1.8 ⫺34.5 Wild-type ⫽ 1.0
1.1 2.0 2.2 1.1
.00005 .006 .002 ⬍.00001
Iron metabolism, antimicrobial Chaparone Generation of histamine Hormone
NOTE. Down-regulated gene expression was identified through gastric expression profiling of wild-type mice, gastrin KO mice, gastrin KO mice, after gastrin infusion for 1 week, and omeprazole-treated wild-type mice. From each group of mice, 4 expression profiles were generated by pooling equal amounts of gastric RNA from 5 mice. Expression indexes for wild-type mice and fold changes to these using wild-type as baseline are given. t test was used to determine the level of significance between gene expression in wild-type and gastrin KO mice.
bles 4 and 5. Most of the genes were also activated during omeprazole treatment and could be divided into 2 groups: (1) immune defense genes that are expressed by the immune system itself or expressed in enterocytes after activation by cytokines and interferon (Table 4) and (2) genes predominantly expressed by the intestines (Table 5). Immune system and defense genes. The source of immunoglobulin synthesis was lymphocytes that infiltrated the mucosa of KO mice but not of wild-type mice (Figure 3D and F and Table 4). Furthermore, the products of 3 of the induced genes (Best5, Schlafen, and ISG15) were localized to the mucosal cells and parietal cells (Figure 4B, F, and H). These genes are known to be regulated by interferon, and the expression of interferon gamma was increased in the gastrin KO mice (Table 6). Best5 was only marginally activated in gastrin KO mice (Figure 4B and C) but abundant although not uniformly expressed in the foveolar hyperplastic polyps (Figure 4D). Gastrin infusion in KO mice reduced the expression of most of the “immune system and defense genes,” but not to the level of expression seen in wild-type mice. Achlorhydria after PPI treatment induced the same response as in gastrin KO mice but at a lower level (Table 4).
Intestinal and other genes. These genes are predominantly expressed in the intestine and only occasionally in the stomach (Table 5). The expression increased ⬃1.5-3–fold in gastrin KO mice, which is less than the group of immune and defense genes (Table 4). One week of gastrin infusion suppressed but did not normalize the expression of “intestinal genes.” Omeprazole treatment also activated the expression of these genes, although the level of expression was lower than in gastrin KO mice (Table 5). These results were confirmed by quantitative polymerase chain reaction (Table 6). The transcription factors Cdx1 and Cdx2 are involved in intestinal metaplasia28 –30 and control the expression of many intestinal genes. While the Cdx2 expression was increased in the gastrin KO mice and normalized after gastrin infusion, Cdx1 expression was increased only in the 1-year-old mice but not 12- to 16-week-old gastrin KO mice (Table 7).
Age-Induced Gastric Changes Despite normal gastric morphology at 12–16 weeks (Figure 3A and B), major metaplastic changes, including accumulation of mucins within the oxyntic mucosa, were present in the gastrin KO mice after 12 months (Figure 3C and D). Infiltration of lymphocytes
250
FRIIS–HANSEN ET AL
GASTROENTEROLOGY Vol. 131, No. 1
Figure 3. Intestinal metaplasia in gastrin-deficient mice. (A and B) Periodic acid–Schiff/Alcian staining of the stomach showed that there are no major differences between young (12–16 weeks old) gastrin KO and wildtype mice. However, at 1 year of age, there are increasing signs of metaplasia with the accumulation of neutral and acid mucins. (D and I) Furthermore, lymphocytes infiltrated both the fundic and the antral mucosa (arrow). At 1½–2 years of age, there is increasing accumulation of mucins within the gastric glands, (H and I) development of cyst changes, and (J) ultimately the development of resulting polyp/tumor development. Immunohistochemical identification of lymphocytes is shown in Supplementary Figure 1 (see supplemental material online at www.gastrojournal. org). Bar ⫽ 50 m.
was found in both fundic and antral mucosa in many gastrin KO mice (Figure 3D and Supplementary Figure 1; see supplemental material online at www.gastrojournal.org). In gastrin KO mice, Cdx2 was expressed within
the metaplastic lesions (Figure 5C–E). Wild-type mice, in contrast, had duodenal but not gastric Cdx2 expression (Figure 5A, B, and F). During the next 6 –12 months, the intestinal metaplasia progressed and mice
July 2006
INTESTINAL METAPLASIA AND TUMORS IN GASTRIN KO MICE
251
Table 4. Changes due to Achlorhydria in Gastric Expression of Immunedefense Genes Fold change
Gene Interferon-induced protein with tetratricopeptide repeats 1 Immunoglobulin heavy chain 4 (serum IgG1) Best-5 Schlafen 3 ISG15 Immunoglobulin joining chain Molecule possessing ankyrin repeats induced by lipopolysaccharide (MAIL) Schlafen 4 Polymeric immunoglobulin receptor Immunoglobulin heavy chain Interferon-induced protein with tetratricopeptide repeats 3 Interferon-inducible guanosine triphosphatase Small inducible cytokine A6 Interferon regulatory factor 7
Wild-type ⫹ omeprazole
P value
5.5 5.9 2.1 1.8 1.1 4.2
3.8 4.8 5.0 1.1 ⫺1.0 3.6
.002 .008 .002 .002 .001 .002
3.7 2.4 2.3 2.2
2.4 ⫺1.0 1.7 2.6
⫺1.5 ⫺1.1 1.7 2.9
.001 .002 .000 .022
1.8 1.8 1.7 1.6
1.0 1.1 1.3 1.1 Wild-type ⫽ 1.0
1.2 1.2 1.4 1.8
.013 .017 .001 .003
Accession no.
Wild-type expression
KO
U43084 X88902 AA204579 AF099974 AV152244 M90766
9 15 87 21 22 36
16.0 4.9 4.8 4.4 4.3 3.8
AW060549 AF099977 AB001489 AF042798
219 173 404 49
U43086 AA914345 M58004 U73037
50 27 148 177
KO ⫹ gastrin
NOTE. Expression profiles were generated from wild-type mice, gastrin KO mice, gastrin KO mice, after gastrin infusion for 1 week, and omeprazole-treated wild-type mice. Wild-type expression index was 1.0, and fold changes to these are given. t test was used to determine the level of significance between gene expression in wild-type and gastrin KO mice. Selected genes are shown.
Table 5. Activation of Nonimmune Response Genes in Gastrin KO Mice, Display of Selected Genes Fold change
Gene Zymogen granule membrane protein 16 Human mRNA for Immunoglobulin G Fc binding protein Cysteine-rich intestinal protein Polymeric immunoglobulin receptor Chloride channel calcium activated 3 ␣ 1-microglobulin/bikunin Villin Melanoma differentiation associated protein 5 Arginine-rich, mutated in early stage tumors Crumbs related ELAV-like 1 (Hu antigen R) Claudin 7 Crp-ductin
Accession Wild-type no. expression KO
KO ⫹ gastrin
Wild-type ⫹ omeprazole
Intestinal P value expression
Function
AI593999
559
3.1
⫺1.1
1.1
.004
*
Unknown
AI465965
603
2.7
1.6
1.5
.000
*
Immune modulation
M13018 AB001489
838 404
2.4 2.3
2.0 1.7
1.3 1.7
.003 .000
* *
Immune modulation Immunoglobulin A transport
AV373378
2054
2.1
1.5
1.4
.000
*
Chloride channel
X68680 M98454 AA959954
53 311 145
2.1 2.0 1.8
1.7 1.9 1.3
2.0 1.2 1.1
.003 .005 .000
*
AW122364
482
1.5
1.2
1.4
.004
AI845823 U65735 AF087825 U37438
336 292 95 567
1.4 1.4 1.8 1.8
1.2 1.2 1.4 1.5
1.3 1.6 1.1 1.3
.001 .006 .000 .000
*
Anti-inflammatory role Microvilli turnover RNA metabolism and apoptosis RNA binding Tight junctions RNA stability Tight junction Immune defense and epithelial functions
Wild-type ⫽ 1.0 NOTE. Several of the genes are primarily expressed in the intestine. Their presence reflects molecular signs of intestinal metaplasia. The genes were identified through gastric expression profiling of wild-type mice, gastrin KO mice, gastrin KO mice after gastrin infusion for 1 week, and omeprazole-treated wild-type mice. From each group of mice, 4 expression profiles were generated by pooling equal amounts of gastric RNA from 5 mice. Expression index for wild-type mice and fold changes to these using wild-type mice as baseline are given. t test was used to determine the level of significance between gene expression in wild-type and gastrin KO mice.
252
FRIIS–HANSEN ET AL
GASTROENTEROLOGY Vol. 131, No. 1
Figure 4. Gastric mucosal expression of inflammatory response genes in gastrin KO mice. Expression of Best5 is not detectable in (A) fundic mucosal cells in wild-type mice; expression is activated in (B) fundic cells but not (C) antral cells. (D) However, in antral polyps in the gastrin KO mice, there is an abundant expression (the arrows indicate clusters of cells with abundant Best5 expression). Schlafen4 is not detectable in (E) wild-type but present in (F) the nucleus of mucosal cells in the gastrin KO mice (the arrows point to typical cells). ISG15 is not detectable in (G) wild-type but present in (H) the nucleus of mucosal cells in gastrin KO mice (some of these are indicated by the arrows). The mice in A–C and E–H were 12–16 months old except in D, where they were 1½ years old.
aged 1½–2 years developed cysts within the glands (Figure 3H). The antral mucosa proliferated (Figure 3I), and most of the mice (6 of 8 mice) developed foveolar hyperplasia and polyps (Figure 3J). There was ulceration, Table 6. Gastric Gene Expression Activated in Mice (Aged 12–16 Weeks) by Achlorhydria and Gastrin Gene
Wild-type
Interferon gamma mRNA 1.0 ⫾ 0.1 Best5 mRNA 1.00 ⫾ 0.1 ISG-15 mRNA 1.0 ⫾ 0.1 Schl4 mRNA 1.0 ⫾ 0.1 Cdx1 mRNA 1.0 ⫾ 0.1 Cdx2 mRNA 1.0 ⫾ 0.1 IgG-FC1 mRNA 1.0 ⫾ 0.1 Muclin mRNA 1.0 ⫾ 0.1 MUC2 mRNA 1.0 ⫾ 0.1 SPRP mRNA 1.0 ⫾ 0.2 Ankyrin mRNA 1.0 ⫾ 0.1 Heparin-binding epidermal growth factor mRNA 1.0 ⫾ 0.1 Insulin-like growth factor binding protein mRNA 1.00 ⫾ 0.14 NOV mRNA 1.0 ⫾ 0.2 Syndecan 4 mRNA 1.00 ⫾ 0.14 Hepcidin mRNA 1.0 ⫾ 0.2
Gastrin KO
Gastrin KO ⫹ gastrin
4.5 ⫾ 1.9a 4.7 ⫾ 0.9a 4.5 ⫾ 0.9a 3.5 ⫾ 0.9a 1.9 ⫾ 0.4a 1.9 ⫾ 0.4a 3.3 ⫾ 0.5a 4.6 ⫾ 0.7a 1.6 ⫾ 0.2a 3.1 ⫾ 0.5a 0.48 ⫾ 0.08a
2.1 ⫾ 1.3b 1.4 ⫾ 0.2b 1.1 ⫾ 0.3b 1.5 ⫾ 0.2b 1.3 ⫾ 0.07b 1.3 ⫾ 0.07b 1.3 ⫾ 0.2b 2.8 ⫾ 0.6a 1.3 ⫾ 0.1b 2.0 ⫾ 0.4b 1.0 ⫾ 0.2b
0.7 ⫾ 0.2b
0.7 ⫾ 0.2b
0.57 ⫾ 0.09a 1.75 ⫾ 0.49a 0.55 ⫾ 0.07a 1.2 ⫾ 0.1b 0.57 ⫾ 0.09a 1.75 ⫾ 0.49a 0.4 ⫾ 0.1a 1.3 ⫾ 0.2b
NOTE. Quantitative polymerase chain reaction analysis of gene expression activated in the gastrin KO mice and normalized/reduced after gastrin infusion. Mean ⫾ SEM of expression in wild-type given in arbitrary units (n ⫽ 8) in gastrin KO mice (n ⫽ 8) and gastrin KO mice that have received 1 week of gastrin infusion subcutaneously (n ⫽ 9). See Materials and Methods. aP ⬍ .05; bP ⬎ .05 (NS).
intestinal metaplasia, and Cdx2 expression within these polyps (Figure 5C–E). The morphologic changes were paralleled by increased expression of intestinal genes, growth factors, interferon gamma, and inflammatory response genes (Table 7). Down-regulated genes. Thirteen genes were down-regulated in gastrin KO mice and normalized or up-regulated following gastrin infusion in gastrin KO mice (Table 3). The PPI treatment resulted in mild hypergastrinemia only. Consistently, the expression of these genes was only slightly up-regulated (Tables 6 and 7). Immunohistochemistry localized the expression of IGFBP2, syndecan 4, and NOV to the parietal cells and to the mucosal cells (Figure 6). The expression of other members of the C-C knot (CCN) or IGF binding protein (IGFBP) families was unchanged (Table 7). Two other genes important for growth were also down-regulated: the extracellular matrix protein syndecan 4 and the LIM domain transcription regulator LIM only 2 (Lmo2). Additionally, 2 secreted peptides were down-regulated in KO mice: hepcidin, an antimicrobial peptide involved in iron metabolism, and CCL27, a chemokine.31,32 Genes Not Affected by Gastrin Deficiency Gastrin has been reported to regulate the expression of genes encoding regenerating protein and epidermal growth factor family, genes encoding proteins belonging to the acid secretory machinery, and genes encoding mucins and trefoil factors. However, none of these were differentially regulated in young gastrin KO mice (Table 8). Instead, their expression increased in
July 2006
INTESTINAL METAPLASIA AND TUMORS IN GASTRIN KO MICE
253
Table 7. Changes in Gastric Gene Expression in Aging Gastrin KO Mice Gene
Wild-type at 12 Weeks
Gastrin KO at 12 Weeks
Wild-type at 52 Weeks
Gastrin KO at 52 Weeks
Interferon gamma mRNA Cdx1 mRNA Cdx2 mRNA Transforming growth factor ␣ mRNA Heparin-binding epidermal growth factor mRNA Betacellulin mRNA Amphiregulin mRNA Trefoil factor 1 mRNA Trefoil factor 2 mRNA IgG-FC mRNA Muclin mRNA MUC2 mRNA SPRP mRNA Best5 mRNA ISG-15 mRNA Schl4 mRNA
1.0 ⫾ 0.1 1.0 ⫾ 0.1 1.0 ⫾ 0.2 1.0 ⫾ 0.1 1.0 ⫾ 0.1 1.0 ⫾ 0.1 1.0 ⫾ 0.1 1.0 ⫾ 0.1 1.0 ⫾ 0.1 1.0 ⫾ 0.1 1.0 ⫾ 0.1 1.0 ⫾ 0.1 1.0 ⫾ 0.6 1.0 ⫾ 0.1 1.0 ⫾ 0.1 1.0 ⫾ 0.1
4.5 ⫾ 1.9a 1.2 ⫾ 0.3b 1.7 ⫾ 0.2a 1.4 ⫾ 0.3b 1.0 ⫾ 0.2b 1.2 ⫾ 0.2b 2.5 ⫾ 0.6a 5.8 ⫾ 1.3a 1.6 ⫾ 0.2a 5.8 ⫾ 1.3a 2.4 ⫾ 0.6a 1.2 ⫾ 0.1b 3.1 ⫾ 1.4a 2.0 ⫾ 0.5a 2.4 ⫾ 0.3a 2.4 ⫾ 0.6a
2.1 ⫾ 1.3b 1.1 ⫾ 0.1b 0.9 ⫾ 0.1b 0.9 ⫾ 0.1b 0.7 ⫾ 0.1a 1.8 ⫾ 0.2a 1.1 ⫾ 0.1b 1.4 ⫾ 0.2b 1.4 ⫾ 0.3b 1.6 ⫾ 0.3b 2.1 ⫾ 0.3a 1.5 ⫾ 0.2a 1.2 ⫾ 0.4b 0.9 ⫾ 0.1b 1.1 ⫾ 0.1b 1.4 ⫾ 0.2b
11.7 ⫾ 3.6a 2.8 ⫾ 0.3a 2.3 ⫾ 0.1a 1.6 ⫾ 0.2a 2.1 ⫾ 0.3a 2.1 ⫾ 0.3a 3.6 ⫾ 0.8a 6.6 ⫾ 1.3a 4.4 ⫾ 1.0a 6.6 ⫾ 1.3a 3.9 ⫾ 0.3a 4.4 ⫾ 1.0a 2.4 ⫾ 1.2a 2.4 ⫾ 0.6a 2.6 ⫾ 0.8a 6.1 ⫾ 2.8a
NOTE. Real-time polymerase chain reaction quantitation of gastric gene expression (arbitrary units) in wild-type mice (n ⫽ 8), gastrin KO mice aged 12–16 weeks (n ⫽ 8), and gastrin KO mice aged 52 weeks (n ⫽ 9) (see Materials and Methods). aP ⬍ .05; bP ⬎ .05 (NS).
parallel with the metaplastic changes in the old gastrin KO mice (Table 7).
Discussion Hypergastrinemia is well known to cause gastric or duodenum ulcer and ECLoma.9,10,33 This study shows that lack of gastrin also predisposes to gastric disease through achlorhydria. Achlorhydria due to lack of gastrin or PPI treatment resulted in specific changes in gastric gene expression. These changes were more prominent in the gastrin KO mice than in PPI-treated wild-type mice and correlated with the greater reduction of acid secretion due to gastrin deficiency than PPI treatment. Exogenous gastrin restored gastric acidity (as shown by this study, FriisHansen et al,17 and Chen et al20) and reversed the changes in gene expression in young mice. However, progressive and irreversible intestinal metaplastic changes develop in aging gastrin KO mice and result in development of tumors (as shown by this study and Zavros et al23). The gastrin-deficient mouse may therefore serve as a new model for how Helicobacter-negative achlorhydria predispose to gastric cancer. The gastric mucosal expression of genes normally expressed in the intestines indicates that the molecular adaptation to achlorhydria is an early event. Activation of the transcription factor Cdx2 could be implicated because its expression is inversely correlated with luminal acidity in the duodenum.34 This may also explain the partial reversibility after gastrin infusion that normalizes acid secretion. The progressive morphologic abnormali-
ties were associated with further expression of Cdx2 and activation of Cdx1 expression, a late step in intestinal metaplasia and a sign of malignant transformation.35 Acidic conditions are also needed to maintain the gastric mucosa through the transcription of sonic hedgehog (Shh).36 Loss of Shh leads to intestinal transformation of the stomach.37 Furthermore, other achlorhydric mouse models also develop intestinal metaplasia; the hypergastrinemic mice deficient for the ␣ subunit of the gastric H,K–adenosine triphosphatase38 and the gp130 mutant mice develop achlorhydria and hypogastrinemia during the period they develop gastric cancer.39 Together, the studies demonstrate that achlorhydria independently of gastrin expression predisposes to intestinal metaplasia. Achlorhydria also reduces gastric elimination of bacteria22 and results in bacterial overgrowth in gastrin KO mice (this study and Zavros et al40). Earlier studies have shown that gastric infection activates the secretion of cytokines and antimicrobial peptides such as -defensins 1 and 2.41 In this study, achlorhydric gastritis and bacterial overgrowth also activated epithelial expression of other genes activated by interferons (for instance, Best-542 and ISG-1543). Gastric,41 intestinal,44,45 and pulmonary epithelial cells46 also activate the expression of genes belonging to the innate immune defense in the presence of infections. Thus, the immune response in the stomach following bacterial overgrowth may reflect a general epithelial cell response to infectious agents. The bacterial overgrowth in gastrin KO mice could promote gastric carcinogenesis through the generation of carcinogenic substances.47,48 The enterococci in the stomach convert nitrate to N-nitrosamines, which pre-
254
FRIIS–HANSEN ET AL
GASTROENTEROLOGY Vol. 131, No. 1
Figure 5. Cdx2 expression in wild-type and KO gastric mucosa. (A and B) Cdx2 expression was absent in wild-type fundic and antral mucosa. However, in the gastrin KO mice, Cdx2 expression was found in (C and D) metaplastic fundic mucosa as well as (E) some parts of the antral foveolar polyps that developed in 1½-year-old gastrin KO mice. (F) The abundant Cdx2 expression in the wild-type duodenum. The tissues in A–D and F were from 1-year-old mice, whereas the tissue in E is from 1/12-year-old mice. Bar ⫽ 100 m.
Figure 6. Immunocytochemical localization of IGF-BP2, syndecan 4, and NOV in wild-type gastric mucosa. Immunocytochemistry was used to localize 3 of the genes whose expression was down-regulated in the gastrin KO mice. IGF-BP2 was expressed in few cells (arrow) in the base of the glands. Syndecan 4 was detected in the extracellular space at low levels throughout the mucosa; areas with stronger staining are indicated by the arrows. NOV was expressed by few cells in each gland (arrows). All mice were 12–16 weeks old.
July 2006
INTESTINAL METAPLASIA AND TUMORS IN GASTRIN KO MICE
255
Table 8. Analysis of the Expression of Genes Belonging to the Acid Secretory Machinery or Involved In Mucus Formation in Gastrin-Deficient and Hypochlorhydric Mice. Fold change
Gene Acid secretion ATPase, H⫹/K⫹ transporting, ␣ polypeptide ATPase, H⫹/K⫹ transporting,  polypeptide Ezrin LIM and SH3 protein 1 (LASP-1) Trefoil factors Trefoil factor 1 Trefoil factor 2 (spasmolytic protein 1) Trefoil factor 3 (intestinal) Mucin family Mucin 1, transmembrane Mucin 1, transmembrane Mucin 3, intestinal Mucin 5, subtypes A and C CCN family Nephroblastoma overexpressed gene WNT1 inducible signaling pathway protein 1 Cysteine-rich protein 61 WNT1 inducible signaling pathway protein 2 Insulin-like growth factor binding protein family Insulin-like growth factor binding protein 1 Insulin-like growth factor binding protein 3 Insulin-like growth factor binding protein 4 Insulin-like growth factor binding protein 5 Insulin-like growth factor binding protein 6 Insulin-like growth factor binding protein 7 EGF family Heparin-binding epidermal growth factor–like growth factor Transforming growth factor ␣ Cripto Amphiregulin Epidermal growth factor Betacellulin Reg protein family Regenerating islet-derived 1 Regenerating islet-derived 2 Regenerating islet-derived 3 ␣ Regenerating islet-derived 3  Regenerating islet-derived 3 ␥
Accession no.
Wild-type ⫹ omeprazole
Wild-type expression
KO
KO ⫹ gastrin
U17282
7204
⫺1.2
⫺1.0
1.1
.00
M80251 X60671 AW122780
6588 1413 1610
⫺1.2 1.0 1.0
⫺1.0 1.1 1.1
1.1 1.2 1.1
.01 .88 .47
Z21858
8889
1.1
⫺1.1
1.0
.18
U78770 D38410
6439 807
⫺1.0 1.0
1.0 1.0
⫺1.0 1.0
.78 .82
M84683 U16175 AF027131 AJ010792
2209 912 458 3935
1.1 ⫺1.1 ⫺1.2 1.0
1.1 ⫺1.0 ⫺1.2 1.1
1.1 1.1 1.1 ⫺1.0
.05 .22 .02 .78
Y09257
171
⫺1.2
1.2
1.2
.19
AF100777 M32490
130 97
⫺1.1 ⫺1.1
⫺1.2 ⫺1.0
⫺1.1 1.2
.38 .50
AF100778
166
1.2
1.1
1.2
.30
X81579
150
1.1
1.0
1.5
.41
X81581
172
1.1
1.0
1.1
.51
X76066
256
1.6
1.8
1.9
.004
L12447
542
⫺1.2
⫺1.1
1.2
.04
X81584
267
⫺1.0
⫺1.0
1.1
.59
AB012886
855
1.0
1.0
1.8
.98
L07264 M92420 U57720 L41352 V00741 L08394
217 81 290 116 70 42
⫺1.1 1.1 ⫺1.0 1.4 ⫺2.4 1.7
1.0 1.3 ⫺1.1 1.4 ⫺1.9 1.4
1.2 1.1 ⫺1.0 ⫺1.1 ⫺1.9 1.6
.51 .31 .57 .02 .03 .02
D14010 D14011 D63357
341 194 222 Not on the chip 173
⫺1.1 ⫺1.6 ⫺1.0
⫺1.1 ⫺1.3 1.0
⫺1.0 ⫺1.4 ⫺1.1
.57 .04 .95
1.1 Wild-type ⫽ 1.0
⫺1.2
.46
D63362
1.1
P value
NOTE. Gene expression profiling of wild-type mice, gastrin KO mice, gastrin KO mice after gastrin infusion for 1 week, and omeprazole-treated wild-type mice was performed. From each group of mice, 4 expression profiles were generated by pooling equal amounts of gastric RNA from 5 mice. Expression index for wild-type mice and fold changes to these using wild-type mice as baseline are given. t test was used to determine the level of significance between gene expression in wild-type and gastrin KO mice.
256
FRIIS–HANSEN ET AL
dispose to cancer.49,50 In contrast, Helicobacter does not have this capability.51 Increased production of N-nitrosamines51,52 is often seen in permanent achlorhydria due to either atrophic gastritis or antrectomy and predisposes to gastric cancer.13–15,53,54 In contrast, long-term PPI treatment is not associated with increased formation of N-nitrosamines55 and is considered safe.56 However, the initial dose (20 – 40 mg omeprazole daily in humans) was initially chosen to avoid severe hypergastrinemia and did not completely block acid secretion. The introduction of newer, more efficient drugs demands a repetition of these studies, preferably with longer follow-up.57 Taken together, complete inhibition of acid secretion favors bacterial overgrowth55,58 that activates an immune response that alters gene expression. Furthermore, carcinogenic substances can also be generated.47,48 Antibiotic treatment of young gastrin KO mice reversed the fundic and antral hyperplasia.40 However, it is presently not known whether sterilization of the gastric juice will also prevent the metaplasia. In contrast to the major changes in gene expression due to achlorhydria and inflammation, the expression of only few genes depended on gastrin. Two of these (NOV and syndecan 4) encode matrix proteins that modulate extracellular signaling59,60 and are important for cell maturation, differentiation, migration, and adhesion. Altered synthesis of extracellular matrix proteins has been associated with gastric disease in Foxl1 KO mice.61 Gastrin-controlled modulation of the extracellular matrix could be a mechanism behind gastrin-controlled growth.62,63 The expression of growth factor known to be activated by hypergastrinemia (trefoil factor 1, transforming growth factor ␣, and heparin-binding epidermal growth factor)8,64,65 were unaffected in young gastrin KO mice. Furthermore, the gastric expression of these increased in parallel with the morphologic abnormalities in the gastrin KO mice (this study and Franic et al66) and is thus controlled by factors other than gastrin. In conclusion, the most important action of gastrin is to maintain gastric homeostasis through control of acid secretion because only few genes operate under direct gastrin control. Achlorhydria is associated with an intestinal expression profile and activation of immune defense genes. Achlorhydria also predisposes to gastric bacterial overgrowth with bacteria that have the capacity to form carcinogenic N-nitrosamines.
Appendix Supplemetary data Supplemetary data associated with this article can be found, in the online version, at doi:10.1053/j.gastro. 2006.04.031
GASTROENTEROLOGY Vol. 131, No. 1
References 1. Håkanson R, Chen D, Sundler F. The ECL cells. In: Johnson LR, ed. Physiology of the gastrointestinal tract. 3rd ed. New York, NY: Raven, 1994:1171–1184. 2. Tömmerås K, Hammer P, Sundler F, Borch K, Mårdh S, Cabero JL. Immunolocalization of cholecystokinin-2 receptors in rat gastric mucosa. Scand J Gastroenterol 2002;37:1017–1024. 3. Tömmerås K, Bakke I, Sandvik AK, Larsson E, Waldum HL. Rat parietal cells express CCK(2) receptor mRNA: gene expression analysis of single cells isolated by laser-assisted microdissection. Biochem Biophys Res Commun 2002;297:335–340. 4. Soll AH, Berglindh T. Receptors that regulate gastric acid-secretory function. In: Johnson L, ed. Physiology of the gastrointestinal tract. 3rd ed. New York, NY: Raven, 1994:1139 –1169. 5. Dimaline R, Sandvik AK. Histidine decarboxylase gene expression in rat fundus is regulated by gastrin. FEBS Lett 1991;281:20 –22. 6. Dimaline R, Evans D, Forster ER, Sandvik AK, Dockray GJ. Control of gastric corpus chromogranin A messenger RNA abundance in the rat. Am J Physiol 1993;264:G583–G588. 7. Swarovsky B, Eissele R, Gross M, Kortner G, Koop H, Arnold R, Simon B. Expression of the gastric H⫹/K(⫹)-ATPase and histidine decarboxylase during omeprazole treatment. Digestion 1994;55:97–102. 8. Khan ZE, Wang TC, Cui G, Chi AL, Dimaline R. Transcriptional regulation of the human trefoil factor, TFF1, by gastrin. Gastroenterology 2003;125:510 –521. 9. Rehfeld JF. The new biology of gastrointestinal hormones. Physiol Rev 1998;78:1087–1088. 10. Dockray GJ, Varro A, Dimaline R, Wang T. The gastrins: their production and biological activities. Annu Rev Physiol 2001;63: 119 –139. 11. Koh TJ, Chen D. Gastrin as a growth factor in the gastrointestinal tract. Regul Pept 2000;93:37– 44. 12. Wang TC, Dangler CA, Chen D, Goldenring JR, Koh T, Raychowdhury R, Coffey RJ, Ito S, Varro A, Dockray GJ, Fox JG. Synergistic interaction between hypergastrinemia and Helicobacter infection in a mouse model of gastric cancer. Gastroenterology 2000;118:36 – 47. 13. Lundegardh G, Adami HO, Helmick C, Zack M, Meirik O. Stomach cancer after partial gastrectomy for benign ulcer disease. N Engl J Med 1988;319:195–200. 14. Caygill CP, Hill MJ, Kirkham JS, Northfield TC. Mortality from gastric cancer following gastric surgery for peptic ulcer. Lancet 1986;1:929 –931. 15. Viste A, Björnestad E, Opheim P, Skarstein A, Thunold J, Hartveit F, Eide GE, Eide TJ, Söreide O. Risk of carcinoma following gastric operations for benign disease. A historical cohort study of 3470 patients. Lancet 1986;2:502–505. 16. Koh TJ, Goldenring JR, Ito S, Mashimo H, Kopin AS, Varro A, Dockray GJ, Wang TC. Gastrin deficiency results in altered gastric differentiation and decreased colonic proliferation in mice. Gastroenterology 1997;113:1015–1025. 17. Friis-Hansen L, Sundler F, Li Y, Gillespie PJ, Greenson JL, Owyang C, Rehfeld JF, Samuelson LC. Impaired gastric acid secretion in gastrin deficient mice. Am J Physiol 1998;274:G561–G568. 18. Langhans N, Rindi G, Chiu M, Rehfeld JF, Ardman B, Beiborn M, Kopin AS. Abnormal gastric histology and decreased acid production in cholecystokinin-B/gastrin receptor deficient mice. Gastroenterology 1997;112:280 –286. 19. Nagata A, Ito M, Iwata N, Kuno J, Takano H, Minowa O, Chihara K, Matsui T, Noda T. G-protein-coupled cholecystokinin-B/gastrin receptors are responsible for physiological cell growth of the stomach in vivo. Proc Natl Acad Sci U S A 1996;93:11825– 11830. 20. Chen D, Zhao C-M, Håkanson R, Samuelson LC, Rehfeld JF, Friis-Hansen L. Altered control of gastric acid secretion in gastrin-
July 2006
21.
22.
23.
24.
25. 26.
27.
28.
29.
30.
31.
32. 33. 34.
35.
36.
37.
38.
cholecystokinin double mutant mice. Gastroenterology 2004; 126:476 – 487. Chen D, Zhao CM, Dockray GJ, Varro A, Van Hoek A, Sinclair NF, Wang TC, Koh TJ. Glycine-extended gastrin synergizes with gastrin 17 to stimulate acid secretion in gastrin-deficient mice. Gastroenterology 2000;119:756 –765. Sun FJ, Kaur S, Ziemer D, Banerjee S, Samuelson LC, De Lisle RC. Decreased gastric bacterial killing and up-regulation of protective genes in small intestine in gastrin-deficient mouse. Dig Dis Sci 2003;48:976 –985. Zavros Y, Eaton KA, Kang W, Rathinavelu S, Katukuri V, Kao JY, Samuelson LC, Merchant JL. Chronic gastritis in the hypochlorhydric gastrin-deficient mouse progresses to adenocarcinoma. Oncogene 2005;24:2354 –2366. Li C, Wong WH. Model-based analysis of oligonucleotide arrays: expression index computation and outlier detection. Proc Natl Acad Sci U S A 2001;98:31–36. Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ. Basic local alignment search tool. J Mol Biol 1990;215:403– 410. Rehfeld JF, Stadil F, Rubin B. Production and evaluation of antibodies for the radioimmunoassay of gastrin. Scand J Clin Lab Invest 1972;30:221–232. Shay H, Sun DCH, Gruenstein M. A quantitative method for measuring spontaneous gastric acid secretion in the rat. Gastroenterology 1954;26:906 –913. Mizoshita T, Inada K, Tsukamoto T, Kodera Y, Yamamura Y, Hirai T, Kato T, Joh T, Itoh M, Tatematsu M. Expression of Cdx1 and Cdx2 mRNAs and relevance of this expression to differentiation in human gastrointestinal mucosa—with special emphasis on participation in intestinal metaplasia of the human stomach. Gastric Cancer 2001;4:185–191. Bai YQ, Yamamoto H, Akiyama Y, Tanaka H, Takizawa T, Koike M, Kenji Yagi O, Saitoh K, Takeshita K, Iwai T, Yuasa Y. Ectopic expression of homeodomain protein CDX2 in intestinal metaplasia and carcinomas of the stomach. Cancer Lett 2002;176:47– 55. Silberg DG, Sullivan J, Kang E, Swain GP, Moffett J, Sund NJ, Sackett SD, Kaestner KH. Cdx2 ectopic expression induces gastric intestinal metaplasia in transgenic mice. Gastroenterology 2002;122:689 – 696. Krause A, Neitz S, Magert HJ, Schulz A, Forssmann WG, SchulzKnappe P, Adermann K. LEAP-1, a novel highly disulfide-bonded human peptide, exhibits antimicrobial activity. FEBS Lett 2000; 480:147–150. Nibbs RJ, Graham GJ. CCL27/PESKY: a novel paradigm for chemokine function. Expert Opin Biol Ther 2003;3:15–22. Rozengurt E, Walsh JH. Gastrin, CCK, signaling, and cancer. Annu Rev Physiol 2001;63:49 –76. Faller G, Dimmler A, Rau T, Spaderna S, Hlubek F, Jung A, Kirchner T, Brabletz T. Evidence for acid-induced loss of Cdx2 expression in duodenal gastric metaplasia. J Pathol 2004;203: 904 –908. Eda A, Osawa H, Yanaka I, Satoh K, Mutoh H, Kihira K, Sugano K. Expression of homeobox gene CDX2 precedes that of CDX1 during the progression of intestinal metaplasia. J Gastroenterol 2002;37:94 –100. Dimmler A, Brabletz T, Hlubek F, Hafner M, Rau T, Kirchner T, Faller G. Transcription of sonic hedgehog, a potential factor for gastric morphogenesis and gastric mucosa maintenance, is upregulated in acidic conditions. Lab Invest 2003;83:1829 –1837. Ramalho-Santos M, Melton DA, McMahon AP. Hedgehog signals regulate multiple aspects of gastrointestinal development. Development 2000;127:2763–2772. Spicer Z, Miller ML, Andringa A, Riddle TM, Duffy JJ, Doetschman T, Shull GE. Stomachs of mice lacking the gastric H,K-ATPase alpha-subunit have achlorhydria, abnormal parietal cells, and ciliated metaplasia. J Biol Chem 2000;275:21555–21565.
INTESTINAL METAPLASIA AND TUMORS IN GASTRIN KO MICE
257
39. Judd LM, Alderman BM, Howlett M, Shulkes A, Dow C, Moverley J, Grail D, Jenkins BJ, Ernst M, Giraud AS. Gastric cancer development in mice lacking the SHP2 binding site on the IL-6 family co-receptor gp130. Gastroenterology 2004;126:196 –207. 40. Zavros Y, Rieder G, Ferguson A, Samuelson LC, Merchant JL. Genetic or chemical hypochlorhydria is associated with inflammation that modulates parietal and G-cell populations in mice. Gastroenterology 2002;122:119 –133. 41. Bajaj-Elliott M, Fedeli P, Smith GV, Domizio P, Maher L, Ali RS, Quinn AG, Farthing MJ. Modulation of host antimicrobial peptide (beta-defensins 1 and 2) expression during gastritis. Gut 2002; 51:356 –361. 42. Grewal TS, Genever PG, Brabbs AC, Birch M, Skerry TM. Best5: a novel interferon-inducible gene expressed during bone formation. FASEB J 2000;14:523–531. 43. Reid LE, Brasnett AH, Gilbert CS, Porter AC, Gewert DR, Stark GR, Kerr IM. A single DNA response element can confer inducibility by both alpha- and gamma-interferons. Proc Natl Acad Sci U S A 1989;86:840 – 844. 44. Seydel KB, Li E, Swanson PE, Stanley SL Jr. Human intestinal epithelial cells produce proinflammatory cytokines in response to infection in a SCID mouse-human intestinal xenograft model of amebiasis. Infect Immun 1997;65:1631–1639. 45. Eckmann L, Smith JR, Housley MP, Dwinell MB, Kagnoff MF. Analysis by high density cDNA arrays of altered gene expression in human intestinal epithelial cells in response to infection with the invasive enteric bacteria Salmonella. J Biol Chem 2000;275: 14084 –14094. 46. Ichikawa JK, Norris A, Bangera MG, Geiss GK, van’t Wout AB, Bumgarner RE, Lory S. Interaction of pseudomonas aeruginosa with epithelial cells: identification of differentially regulated genes by expression microarray analysis of human cDNAs. Proc Natl Acad Sci U S A 2000;97:9659 –9664. 47. Reed PI, Smith PL, Haines K, House FR, Walters CL. Effect of cimetidine on gastric juice N-nitrosamine concentration. Lancet 1981;2:553–556. 48. Stockbrugger RW, Cotton PB, Eugenides N, Bartholomew BA, Hill MJ, Walters CL. Intragastric nitrites, nitrosamines, and bacterial overgrowth during cimetidine treatment. Gut 1982;23:1048 – 1054. 49. Risch HA, Jain M, Choi NW, Fodor JG, Pfeiffer CJ, Howe GR, Harrison LW, Craib KJ, Miller AB. Dietary factors and the incidence of cancer of the stomach. Am J Epidemiol 1985;122:947– 959. 50. Gonzalez CA, Riboli E, Badosa J, Batiste E, Cardona T, Pita S, Sanz JM, Torrent M, Agudo A. Nutritional factors and gastric cancer in Spain. Am J Epidemiol 1994;139:466 – 473. 51. Vermeer IT, Gerrits MM, Moonen EJ, Engels LG, Dallinga JW, Kleinjans JC, van Maanen JM, Kuipers EJ, Kusters JG. Helicobacter pylori does not mediate the formation of carcinogenic Nnitrosamines. Helicobacter 2002;7:163–169. 52. Schlag P, Bockler R, Peter M. Nitrite and nitrosamines in gastric juice: risk factors for gastric cancer? Scand J Gastroenterol 1982;17:145–150. 53. Meining A, Morgner A, Miehlke S, Bayerdorffer E, Stolte M. Atrophy-metaplasia-dysplasia-carcinoma sequence in the stomach: a reality or merely an hypothesis? Best Pract Res Clin Gastroenterol 2001;15:983–998. 54. Thompson JC, Fender HR, Watson LC, Villar HV. The effects on gastrin and gastric secretion of five current operations for duodenal ulcer. Ann Surg 1976;183:599 – 608. 55. Yeomans ND, Brimblecombe RW, Elder J, Heatley RV, Misiewicz JJ, Northfield TC, Pottage A. Effects of acid suppression on microbial flora of upper gut. Dig Dis Sci 1995;40:81S–95S. 56. Lamberts R, Brunner G, Solcia E. Effects of very long (up to 10 years) proton pump blockade on human gastric mucosa. Digestion 2001;64:205–213.
258
FRIIS–HANSEN ET AL
57. Gillen D, McColl KE. Problems associated with the clinical use of proton pump inhibitors. Pharmacol Toxicol 2001;89:281–286. 58. Howden CW, Hunt RH. Relationship between gastric secretion and infection. Gut 1987;28:96 –107. 59. Shimo T, Gentili C, Iwamoto M, Wu C, Koyama E, Pacifici M. Indian hedgehog and syndecans-3 coregulate chondrocyte proliferation and function during chick limb skeletogenesis. Dev Dyn 2004;229:607– 617. 60. Brigstock DR. The CCN family: a new stimulus package. J Endocrinol 2003;178:169 –175. 61. Perreault N, Katz JP, Sackett SD, Kaestner KH. Foxl1 controls the Wnt/beta-catenin pathway by modulating the expression of proteoglycans in the gut. J Biol Chem 2001;276:43328 – 43333. 62. Tömmerås K, Chen Y, Rhedin M, Cabero JL, Mårdh S. Proliferation and differentiation of cells from explants of fetal rat stomach. Acta Physiol Scand 1997;159:155–161. 63. Tömmerås K, Cabero JL, Mårdh S. Expression of extracellular matrix proteins in the fetal rat gastric mucosa. Anat Embryol (Berl) 2000;201:149 –156. 64. Kinoshita Y, Ishihara S. Mechanism of gastric mucosal proliferation induced by gastrin. J Gastroenterol Hepatol 2000; 15(Suppl):D7–D11.
GASTROENTEROLOGY Vol. 131, No. 1
65. Fukui H, Kinoshita Y, Maekawa T, Okada A, Waki S, Hassan S, Okamoto H, Chiba T. Regenerating gene protein may mediate gastric mucosal proliferation induced by hypergastrinemia in rats. Gastroenterology 1998;115:1483–1493. 66. Franic TV, Judd LM, Nguyen NV, Samuelson LC, Loveland KL, Giraud AS, Gleeson PA, Van Driel IR. Growth factors associated with gastric mucosal hypertrophy in autoimmune gastritis. Am J Physiol Gastrointest Liver Physiol 2004;287:G910 –G918. 67. Nicklas W, Baneux P, Boot R, Decelle T, Deeny AA, Fumanelli M, Illgen-Wilcke B. Recommendations for the health monitoring of rodent and rabbit colonies in breeding and experimental units. Lab Anim 2002;36:20 – 42.
Received April 13, 2005. Accepted March 30, 2006. Address requests for reprints to: Lennart Friis-Hansen, MD, Department of Clinical Biochemistry, KB-3014, Rigshospitalet, 9 Blegdamsvej, DK-2100 Copenhagen, Denmark. e-mail: [email protected]; fax: (45) 3545-46-40. Supported by the Novo Nordic Foundation (to L.F.-H.) and the Swedish Research Council (64-04723 to T.W.). The authors thank Bo Lindberg for skillful technical assistance.