Gastrin Increases Murine Intestinal Crypt Regeneration Following Injury

GASTROENTEROLOGY 2006;130:1169 –1180

BASIC–ALIMENTARY TRACT Gastrin Increases Murine Intestinal Crypt Regeneration Following Injury PENELOPE D. OTTEWELL,* CARRIE A. DUCKWORTH,* ANDREA VARRO,‡ ROD DIMALINE,‡ TIMOTHY C. WANG,§ ALASTAIR J.M. WATSON,* GRAHAM J. DOCKRAY,‡ and D. MARK PRITCHARD* Divisions of *Gastroenterology and ‡Physiology, University of Liverpool, Liverpool, UK; and §Columbia University, New York, New York

Background & Aims: A number of growth factors affect the regeneration of intestinal epithelia following injury, but the effects of amidated gastrin have not previously been assessed. We therefore investigated the effects of gastrin on intestinal regeneration following a range of stimuli. Methods: Intestinal crypt regeneration was assessed in transgenic mice overexpressing amidated gastrin (INS-GAS) and mice in which hypergastrinemia was induced using omeprazole, following ␥-radiation, 5-fluorouracil, and dextran sulphate sodium (DSS). Abundance of the CCK-2 receptor was assessed in intestinal epithelia and IEC-6 intestinal epithelial cells following ␥-radiation. Results: Four days following 14 Gy ␥-radiation, or 2 injections of 400 mg/kg 5-fluorouracil, INSGAS mice exhibited significantly increased small intestinal and colonic crypt survival compared with their wildtype counterparts (FVB/N). INS-GAS mice treated with 3% DSS for 5 days showed less weight loss and increased colonic crypt regeneration at 8 days compared with FVB/N. Increased small intestinal and colonic crypt survival was also demonstrated following ␥-radiation in FVB/N mice rendered hypergastrinemic using omeprazole. The increased crypt survival in INS-GAS mice following 14 Gy ␥-radiation was inhibited by administration of a CCK-2 receptor antagonist (YF476). Increased abundance of the CCK-2 receptor was demonstrated in intestinal epithelia following 14 Gy ␥-radiation by Western blotting and immunohistochemistry. Similarly, increased CCK-2 receptor mRNA abundance and increased 125I-gastrin binding was demonstrated in IEC-6 cells following 4 Gy ␥-radiation. Conclusions: Hypergastrinemia increases regeneration of intestinal epithelia following diverse forms of injury. Induction of the CCK-2 receptor in damaged epithelium confers potential for protection against injury by administration of gastrin.

growth factors to reduce the severity of mucositis induced by cancer therapy and some agents are currently in human clinical trials.1 Murine models have been particularly useful for studies of the effects of growth factors on radiation and chemotherapeutic drug-induced intestinal mucositis. The intestinal epithelium is constantly renewed from stem cells located near the bottom of small intestinal and colonic crypts (reviewed in Potten et al2 and Booth and Potten3). Following administration of a damage inducing stimulus such as ␥-radiation, some cells near the bottom of intestinal crypts die by apoptosis (reviewed in Watson and Pritchard4). If all crypt cells die, the crypt is reproductively sterilized and disappears within 48 hours. However if 1 or more “clonogenic” cell survives the insult, it rapidly proliferates to regenerate the crypt within 72–96 hours and subsequently the tissue heals by clonal expansion. The number of crypts that survive and regenerate following a cytotoxic insult correlates well with severity of symptoms and survival in animal models. A number of growth factors, such as keratinocyte growth factor, transforming growth factor-␤, and interleukin-11 have been shown to affect crypt regeneration in murine intestinal epithelium following ␥-radiation.5– 8 Growth factors have been postulated to affect the process of crypt regeneration in a number of ways. For example they may alter the number of clonogenic cells, the susceptibility of clonogenic cells to cell death, the time taken to start regeneration, or the rate of proliferation in regenerating crypts.7 Gastrin is a hormone secreted from G-cells in the antrum of the stomach. It has important functions in

adiotherapy and cancer chemotherapy often cause the side effect of intestinal mucositis, manifest clinically as diarrhea and weight loss. Over recent years there has been considerable interest in the use of various

Abbreviations used in this paper: 5-FU, 5-fluorouracil; DSS, dextran sulfate sodium; CCK, cholecystokinin. © 2006 by the American Gastroenterological Association Institute 0016-5085/06/$32.00 doi:10.1053/j.gastro.2005.12.033

R

1170

OTTEWELL ET AL

regulating acid secretion, proliferation, and differentiation in the gastric mucosa (reviewed in Dockray et al9). Gastrin has also been shown to enhance the healing of experimentally induced ulcers in rat stomach.10 Although precursor forms of gastrin, such as progastrin and glycine-extended gastrin exert well described effects in the normal murine colon,11–13 fully processed amidated gastrin exerts few effects upon normal intestinal epithelia as the CCK-2 receptor is not usually expressed in this tissue.14,15 We have now assessed whether amidated gastrin acts as a growth factor influencing the healing of intestinal epithelia following induction of injury. We demonstrate that amidated gastrin significantly increases crypt regeneration within intestinal epithelia following a variety of forms of injury and show that this occurs as a result of signaling via CCK-2 receptors in the intestine. This suggests a novel function of gastrin in protecting the distal intestinal epithelium from injury.

Materials and Methods Animals Mice used were INS-GAS and their wild-type counterparts FVB/N. INS-GAS mice contain a transgene consisting of 0.4 kb of the insulin promoter upstream of the human gastrin coding sequence. This results in the overexpression of gastrin in pancreatic ␤-cells and elevated serum concentrations of human amidated gastrin.16 FVB/N mice were obtained from B⫹K (Hull, UK). We also assessed hGAS (hg⫹/⫹) mice, which express increased serum concentrations of human progastrin,11 G⫺/⫺hg⫹/⫹ mice (which express human progastrin but no forms of murine gastrin) and G⫺/⫺hg⫺/⫺ mice (which express no gastrin).13 To investigate the importance of basal concentrations of gastrin we also compared gastrin knockout mice17 and their wild-type counterparts (C57BL/6). Male mice, 10 –12 weeks old, were used in most experiments, with a minimum of 6 animals in each experimental group. Female mice were used for the DSS experiments because of fighting between male mice when caged together for the 11-day time course of this experiment. Mice were fed a commercially prepared pelleted diet and allowed water ad libitum. All animals were maintained in a conventional, nonspecific pathogen free, mouse facility on a 12:12-hour light– dark cycle, and all experiments were conducted during the day time. Experiments were performed with home office approval.

Irradiation Mice were exposed to whole-body ␥-irradiation using a source at a dose rate of 2.6 Gy/min. All irradiation treatments were begun between 09.00 and 10.00. Animals were sacrificed 96 hours after irradiation. Three hours prior to sacrifice, animals were injected with 0.02 mg/kg vincristine sulphate (Sigma, Poole, UK) IP to facilitate detection of regenerating crypts.

137Cs

GASTROENTEROLOGY Vol. 130, No. 4

To assess whether the effects observed were specific to hypergastrinemia, FVB/N mice were rendered hypergastrinemic by gavage with 75 mg/kg omeprazole (AstraZeneca, Luton, UK) suspended in 0.25% methylcellulose (Sigma). Effects of gastrin acting via the CCK-2 receptor were blocked by IP injection of 10 ␮mol/kg YF476 (a gift from Yamanouchi [Osaka, Japan]) dissolved in polyethylene glycol 300 (Sigma). Omeprazole or YF476 was given 24 hours prior to exposure to 14 Gy ␥-radiation and daily up until sacrifice.

5-Fluorouracil–Induced Enteritis 5-Fluorouracil (5-FU; 400 mg/kg; Sigma), dissolved in 20% DMSO/80% saline was administered by IP injection at 10.00 and again at 16.00. Mice were sacrificed 3 or 4 days posttreatment. Three hours prior to sacrifice, animals were injected with 0.02 mg/kg vincristine sulphate (Sigma) IP.

Dextran Sulfate Sodium–Induced Colitis Female FVB/N and INS-GAS mice were given 3% DSS in the drinking water for 5 days and then water ad libitum. Mice were weighed daily and were inspected for diarrhea. Groups of mice were humanely killed at days 8 and 11 and the intestines were removed, processed, and scored as described below.

Tissue Preparation and Scoring Following sacrifice, intestines were removed and fixed in Carnoy’s solution, then embedded in paraffin wax. Transverse sections (3–5 ␮m) of small intestine and colon were prepared and stained with H&E as previously described.12 Ten transverse sections per small intestine or colon were scored for numbers of surviving crypts.18 A surviving crypt was defined as containing ⱖ10 adjacent healthy looking epithelial cells and a lumen. The widths (at the widest point) of 15 surviving crypts were also measured to allow size correction.18 Such a correction factor adjusts for the probability of overscoring larger regenerating crypts or underscoring smaller ones. Data are presented as percentage of surviving crypts (⫾ standard deviation [SD]) compared with control following correction for crypt width. For 5-FU–treated samples, the number of cells per hemicrypt was also scored in 10 separate small intestinal crypts and 10 midcolonic crypts per mouse. This is an alternative technique for assessing the toxic effects of 5-FU and reflects the previous observation that 5-FU in the current dosing regime causes decreases in the cell number and height of both small intestinal and colonic crypts.19 Data are presented as percentage of number of cells per hemicrypt (⫾ SD) following treatment compared with control as previously described.19

Western Blotting Intestinal epithelial cells were prepared by using a modified Weiser technique.20 Intestines were excised and incised along their length to expose the epithelial surface. After washing in PBS the intestines were immersed in Weiser solution for 45 minutes.20 The contents were shaken vigor-

April 2006

ously to separate crypts and villi from the muscle layer and the epithelial cell population was pelleted by centrifugation. These crypts and villi were used to prepare protein lysates for Western blotting. Small intestinal and colonic epithelial cells were lysed and homogenized in lysis buffer (0.06 mol/L Tris-HCl, pH 6.8, 10% glycerol, 2% SDS, 5% ␤-mercaptoethanol). Forty micrograms of protein were electrophoresed on 8% SDS-polyacrylamide gels followed by transfer onto nitrocellulose membrane (Protran, Schleicher & Schuell, Germany). Nonspecific antibody binding was blocked by incubating the membrane in 1% nonfat milk in PBS–Tween-20 prior to incubation with rabbit polyclonal CCK-2 receptor antibody (a gift from Prof. S. Watson, University of Nottingham, UK) used at a dilution of 1:100 overnight at 4 °C. This antibody was raised to amino acids 5–21 in the aminoterminal extracellular part of the human CCK-2 receptor21 and has previously been used for Western blotting,21 functional studies,22,23 and immunohistochemistry.24 The secondary antibody was HRP-conjugated anti-rabbit from DakoCytomation (Cambridge, UK). Membranes were developed using Supersignal (PIERCE, Tattenhall, UK) and chemiluminescence was detected using a Fluor-S molecular imager (BIO-RAD, Hertfordshire, UK). A mouse monoclonal anti-pan actin antibody (ab5) (Neomarkers, Freemont, CA) was used as a loading control. Western blots using IEC-6 cells were performed in exactly the same way except that only 20 ␮g protein was loaded per lane. Thirteen of the sequence of 16 amino acids in human CCK-2 to which the anti–CCK-2 antibody was raised are identical in the rat protein and 12 are identical in the mouse protein.

Immunohistochemistry Tissue sections were prepared as described earlier except that tissues were fixed in 4% formaldehyde in normal saline rather than Carnoy’s fixative. Antigen retrieval was performed by microwaving in 10 mmol/L citric acid buffer (pH 6) for 10 minutes. The primary antibody was the same anti–CCK-2 receptor antibody at a dilution of 1:150, and an anti-rabbit biotinylated secondary antibody (DakoCytomation, Cambridge, UK) was used at a dilution of 1:200 as previously described.24 An ABC (Vector Labs, Peterborough, UK) amplification step was carried out before the biotin signal was developed with 1-3-diaminobenzene (Sigma).

Analysis of Plasma Gastrin Concentrations Radioimmunoassay for amidated gastrin was performed on serum samples using antibody L2 as previously described.11 This antibody reacts with both mouse and human amidated gastrin but does not cross-react with glycineextended gastrin or progastrin.25 Human gastrin was used as the standard in this assay.

Cell Line The nontransformed neonatal rat intestinal epithelial cell line IEC-6 (provided by Dr C. Booth, Epistem Ltd, Manchester, UK) was cultured in DMEM (Sigma) supplemented with 10% fetal calf serum (Gibco, Paisley, UK), 2

MURINE INTESTINAL CRYPT REGENERATION

1171

mmol/L L-glutamine, and 1% penicillin-streptomycin at 37 °C in a water-saturated atmosphere of 95% air and 5% CO2.

Detection of CCK-2 Receptor mRNA in IEC6 Cells At various times following exposure to 4 Gy ␥-radiation, IEC-6 cells were harvested and stored in RNA Later (Ambion, Austin, TX). RNA was isolated using an RNeasy (QIAGEN, Sussex, UK) kit according to the manufacturer’s protocols. RNA was treated with RNase-free DNase (QIAGEN) for 15 minutes prior to cDNA synthesis. First-strand cDNA was synthesized using ABgene Reverse iT first-strand synthesis kit (ABgene, Surrey, UK) according to the manufacturer’s protocols. One microliter of cDNA was added to 0.1 ␮mol/L (outer) forward primer, 5=-GTGAAAATGACAGCGAGAC-3=, 0.1 ␮mol/L (outer) reverse primer, 5=-GGAGGGGGTAGGAGGAT-3= (synthesized by Genosys, Sigma) and 18 ␮L PCR MasterMix (ABgene). DNA was amplified using an OmniGene PCR thermo cycler (Hybaid, Middlesex, UK) and the following conditions: DNA was denatured for 10 minutes at 95 °C before 30 cycles of 95 °C for 15 seconds (denaturation), 56 °C for 15 seconds (annealing), 72 °C for 30 seconds (elongation), followed by a final elongation step of 72 °C for 5 minutes. One microliter of PCR product was added to 0.1 ␮mol/L (inner) forward primer, 5=AGCTGGGGAAGACAGTGAT-3=, 0.1 ␮mol/L (inner) reverse primer 5=-GGGGTTGACACAAGCAGA-3= and 18 ␮L of PCR MasterMix. DNA was amplified as described above except 40 cycles instead of 30 were used for DNA amplification. The product was electrophoresed on a 1.5% agarose gel and bands were visualized using a Molecular FX imager (BIORAD, Hertfordshire, UK).

CCK-2 Receptor Binding IEC-6 cells were seeded in full media overnight, exposed to 4 Gy ␥-radiation and then incubated for 24 hours at 37 °C. Media was removed from all flasks and cells washed 3⫻ in PBS. One milliliter of HANK’s salt solution (Sigma) supplemented with 10 mmol/L HEPES (Sigma) and 125I-labeled G17 was added to all flasks; to control irradiated flask 0.1 nmol to 1 ␮mol/L “cold” G17 (Bachem, St. Helens, UK) was added for competition. Cells were incubated overnight at 4 °C to allow gastrin/receptor binding and washed 3⫻ in PBS. One milliliter of 0.1 mol/L NaOH was added and incubated at 60 °C for 15 minutes. These experiments were performed at 4 °C to prevent internalization of the ligand–receptor complex. A RIASTAR (PACKARD, Berks) machine was used to record total 125I counts per minute. 3HTdR

Proliferation Assay

IEC-6 cells were plated out at a density of 50,000 cells per flask in full media. Twenty-four hours following seeding, cells were exposed to 4 Gy ␥-radiation. Immediately following exposure to ␥-radiation cells were washed 3⫻ in PBS and media was replaced with serum free DMEM. Twenty-four hours following incubation in serum-free DMEM, the media

1172

OTTEWELL ET AL

was changed and flasks were dosed with 10 nmol/L G17 (Bachem) for 48 hours prior to measuring 3HTdR incorporation as previously described.26 Five microliters of 3HTdR (NEN Life Science Products, Zaventem, Belgium) was added to each flask and incubated at 37 °C in 5% CO2 for 2 hours. Media was removed and cells washed 3⫻ in PBS. One milliliter of 5% trichloroacetic acid was added and incubated for 20 minutes at 4 °C to precipitate DNA. Cells were then washed 2⫻ in absolute ethanol before being incubated in 1 mL 0.5 mol/L NaOH at 60 °C for 1 hour. Two hundred microliters of sample was added to 10 mL scintillation fluid and total tritiated thymidine incorporation (counts per minute) were recorded using a TRI-CARB 1900 TR liquid scintillation analyzer (PACKARD).

Statistical Analyses Differences between genotypes were assessed by Student t-test assuming unequal variance of the groups being tested. A level of P ⬍ .05 was interpreted as significant.

Results Intestinal Crypt Regeneration Is Increased in INS-GAS Mice Following ␥-Radiation Ninety-six hours following exposure to 12 or 14 Gy ␥-radiation INS-GAS mice exhibited significantly increased small intestinal and colonic crypt survival compared with wild-type (FVB/N) mice (P ⬍ .05 for each tissue at each radiation dose) (Figures 1 and 2). The differences were most pronounced after 14 Gy ␥-radiation; thus, this dose was used for subsequent experiments. By contrast, significant differences in small intestinal or colonic crypt survival were not demonstrated between either hGAS and FVB/N mice, or between G⫺/⫺hg⫹/⫹ and G⫺/⫺hg⫺/⫺ mice 96 hours following 12 Gy ␥-radiation (data not shown), suggesting that the responses exhibited in INS-GAS mice are specific to amidated gastrin and that increased serum concentrations of human progastrin do not cause the same effects. In addition, no significant differences in small intestinal or colonic crypt survival were demonstrated between gastrin-knockout and C57BL/6 mice 96 hours following 12 Gy ␥-radiation (data not shown), suggesting that basal plasma and tissue concentrations of gastrin have little effect on intestinal crypt regeneration following induction of injury. Intestinal Crypt Regeneration Is Increased in FVB/N Mice Rendered Hypergastrinemic Using Omeprazole Treatment of wild-type FVB/N mice with 75 mg/kg of the proton pump inhibitor omeprazole daily by gavage significantly elevated serum gastrin levels from

GASTROENTEROLOGY Vol. 130, No. 4

76.7 ⫾ 46.4 pmol/L to 170.1 ⫾ 74.8 pmol/L (Figure 3C). The serum gastrin concentration in omeprazole treated FVB/N mice was, however, significantly less than observed in INS-GAS mice (326.0 ⫾ 167.7 pmol/L; P ⬍ .05) (Figure 3C). Significantly increased colonic (P ⬍ .001) and small intestinal crypt regeneration (P ⬍ .001) was observed 96 hours following 14 Gy ␥-radiation in omeprazole-treated mice compared to FVB/N (Figures 3A and B), although the responses were significantly less than in INS-GAS mice (P ⬍ .05 for each tissue) (Figures 3A and B), correlating with the less pronounced hypergastrinemia in these mice. Treatment of Hypergastrinemic Mice With a CCK-2 Receptor Antagonist Reduces Intestinal Crypt Regeneration Following ␥Radiation To investigate whether gastrin signals via the CCK-2 receptor to increase intestinal crypt survival in hypergastrinemic mice, INS-GAS mice and omeprazoletreated FVB/N mice were injected with the specific CCK-2 receptor antagonist, YF476 prior to exposure to 14 Gy ␥-radiation and daily up until sacrifice. Treatment of INS-GAS mice with YF476 resulted in a significant decrease in both colonic (P ⬍ .01) (Figure 4A) and small intestinal (P ⬍ .01) (Figure 4B) crypt survival 96 hours after 14 Gy ␥-radiation. Similarly, treatment with YF476 of FVB/N mice rendered hypergastrinemic by omeprazole resulted in significantly decreased colonic (P ⬍ .001) (Figure 4A) and small intestinal (P ⬍ .001) (Figure 4B) crypt survival compared with vehicle treated mice 96 hours after 14 Gy ␥-radiation. The Severity of 5-Fluorouracil–Induced Enteritis and Dextran Sulfate Sodium–Induced Colitis Is Reduced in Hypergastrinemic INS-GAS Mice To assess whether induction of hypergastrinemia also reduced the severity of inflammatory conditions of the intestine, we used 2 murine models of intestinal inflammation. The chemotherapeutic drug 5-FU induces intestinal mucositis when administered in the regime of 2 injections of 400 mg/kg 5-FU given 6 hours apart.19 Following this regime there is destruction of crypt integrity in both small intestine and colon at 72 and 96 hours manifesting as crypt shortening. At 72 and 96 hours following 5-FU administration, the percentage of cells per hemicrypt (relative to untreated controls) in both colon (Figure 5A) and small intestine (Figure 5B) was significantly increased in INS-GAS mice compared with wild-type FVB/N (P ⬍ .05 for each tissue at each

April 2006

MURINE INTESTINAL CRYPT REGENERATION

1173

Figure 1. Photomicrographs showing small intestine at low power (A–D), small intestine at high power (E–H), colon at low power (I–L), and colon at high power (M–P) from control FVB/N (A, E, I, M), control INS-GAS (B, F, J, N), FVB/N 96 hours after 12 Gy ␥-radiation (C, G, K, O), INS-GAS 96 hours after 12 Gy ␥-radiation (D, H, L, P) mice.

timepoint). Similarly, small intestinal crypt survival was significantly increased in INS-GAS mice relative to FVB/N (P ⬍ .05) 96 hours after treatment with 5-FU (Figure 5C). 5-FU at this dose does not cause sterilization

of colonic crypts; hence colonic crypt survival was not measured following administration of this drug. DSS is widely used to induce colitis in mice. We administered 3% DSS for 5 days and then mice were

1174

OTTEWELL ET AL

GASTROENTEROLOGY Vol. 130, No. 4

active bands of approximately 70 and 40 kDa.21 The antibody has previously been shown to specifically detect the CCK-2 receptor in CCK-2–transfected NIH3T3 cells and gastrointestinal cell lines21,27 and to displace radiolabeled gastrin and inhibit gastrin-stimulated effects in several studies.22,23,28 Protein extracted from the AGS-GR cell line, a human gastric carcinoma cell line that has been stably transfected with the CCK-2 recep-

Figure 2. Colonic (A) and small intestinal crypt survival (B) 96 hours following 8, 10, 12, or 14 Gy ␥-radiation in male wild-type FVB/N (solid line) and INS-GAS (dotted line) mice. Mean ⫾ standard deviation; differences were significant in both tissues at 12 and 14 Gy (P ⬍ .05).

given water and allowed to recover. INS-GAS mice continued to display significant hypergastrinemia after this treatment regime (Figure 6C). INS-GAS mice showed significantly less weight loss at days 8 and 9 following commencement of DSS (P ⬍ .05) (Figure 6A). Crypt survival in the distal colon was also significantly increased in INS-GAS mice relative to FVB/N 8 days following commencement of DSS (P ⬍ .05) (Figure 6B). Both FVB/N and INS-GAS mice showed complete recovery of colonic epithelium 11 days after commencement of DSS (Figure 6B). Induction of the CCK-2 Receptor in Murine Intestinal Epithelial Cells Following ␥Radiation To investigate whether gastrin signaled directly via CCK-2 receptors located in intestinal epithelia, we measured abundance of the receptor both before and after exposure to 14 Gy ␥-radiation. Western blotting with the CCK-2 receptor antibody used in these studies has previously been reported to show 2 specific immunore-

Figure 3. Colonic (A) and small intestinal (B) crypt survival and serum amidated gastrin concentrations (C) from wild-type (FVB/N) (black), INS-GAS (grey) and FVB/N treated daily with 75 mg/kg omeprazole (white) mice 96 hours after 14 Gy ␥-radiation. Mean ⫾ standard deviation. *P ⬍ .05 compared to FVB/N.

April 2006

MURINE INTESTINAL CRYPT REGENERATION

1175

hours following 4 Gy ␥-radiation, CCK-2 receptor mRNA was detected by nested PCR (Figure 8A). Increased abundance of CCK-2 receptor was also demonstrated by Western blotting 24 hours after 4 Gy ␥radiation (Figure 8B). In addition, 24 hours after 4 Gy ␥-radiation, IEC-6 cells exhibited binding of 125I-G17,

Figure 4. Colonic (A) and small intestinal (B) crypt survival from male omeprazole treated FVB/N mice ⫹ vehicle (black), omeprazole treated FVB/N mice ⫹ YF476 (white), INS-GAS mice ⫹ vehicle (grey) and INS-GAS mice ⫹ YF476 (white) 96 hours following 14 Gy ␥-radiation. Mean ⫾ standard deviation. *P ⬍ .05 compared to equivalent mice without YF476.

tor,29 showed bands of similar mass as well as an unidentified band at approximately 60 kDa, whereas no signal was detected in the parental AGS cell line, which acted as a negative control (Figure 7A). Increased abundance of CCK-2 receptor protein was observed in small intestinal and colonic epithelial cells at increasing times after irradiation, with maximum abundance in both tissues 96 hours after 14 Gy ␥-radiation (Figure 7A). Immunohistochemistry showed that CCK-2 receptor protein was not detected in nonirradiated colonic or small intestinal tissue (Figures 7B and D). However 96 hours following 14 Gy ␥-radiation, abundant expression of CCK-2 receptor protein was found specifically within the regenerating crypts of both the colon and small intestine (Figures 7C and E). CCK-2 Receptor Expression Is Induced in IEC-6 Cells Following ␥-Radiation, Leading to Increased Proliferation After Addition of Gastrin-17 No CCK-2 receptor mRNA was detected by nested PCR in untreated IEC-6 cells. However 24 –96

Figure 5. Percentage of surviving cells per hemicrypt in colon (A) and small intestine (B) 72 and 96 hours following administration of 2⫻ 400 mg/kg 5-FU. (C) Small intestinal crypt survival 96 hours following administration of 2⫻ 400 mg/kg 5-FU from wild-type male FVB/N (black) and INS-GAS (grey) mice. Mean ⫾ standard deviation. *P ⬍ .05 compared to FVB/N.

1176

OTTEWELL ET AL

GASTROENTEROLOGY Vol. 130, No. 4

trin-17 for 48 hours following 4 Gy irradiation caused a significant increase in proliferation (Figure 8C). These experiments suggest that irradiation of the nontransformed rat intestinal cell line, IEC-6 leads to increased abundance of the CCK-2 receptor and that this has functional consequences on addition of gastrin-17.

Discussion

Figure 6. Percent weight change (A), crypt survival in distal colon at days 8 and 11 (B), and serum amidated gastrin concentration at day 8 (C) in FVB/N (black/solid line) and INS-GAS (grey/dotted line) female mice given 3% DSS in the drinking water for 5 days and then allowed to recover. Mean ⫾ standard deviation. *P ⬍ .05 compared to FVB/N.

which was displaced by unlabelled G17 with an IC50 of approximately 1 nmol/L. There was no binding of 125IG17 to control cells (Figure 8C). The data suggest that the receptor detected by PCR and Western blotting is functional. Administration of 4 Gy ␥-radiation caused a significant suppression of cellular proliferation in IEC-6 cells as measured by tritiated thymidine incorporation (Figure 8C). However, treatment with 10 nmol/L gas-

We have demonstrated increased small intestinal and colonic crypt regeneration in hypergastrinemic INSGAS mice compared with wild-type (FVB/N) animals following ␥-radiation, 5-FU, and DSS and have also demonstrated increased crypt survival following ␥-radiation in FVB/N mice rendered pharmacologically hypergastrinemic using omeprazole. The increase in crypt regeneration was greater in INS-GAS mice than omeprazole-treated FVB/N mice, correlating with higher serum amidated gastrin concentrations in the former animals. The serum concentrations of gastrin observed in omeprazole-treated mice were similar to those previously documented by others.30,31 These responses were not observed in mice which express high serum concentrations of progastrin rather than amidated gastrin, suggesting that the response is specific to the latter. Although increased basal colonic proliferation has been observed in progastrin-overexpressing (hGAS) mice,11,12 this seems to have little effect on the capacity of the tissue to regenerate following high doses of radiation. The lack of effect of progastrin may reflect absence of the putative progastrin receptor in the clonogenic cell compartment and hence in regenerating intestinal crypts. We present 2 lines of evidence that suggest that gastrin exerts these effects by signaling via the CCK-2 receptor. First, administration of a specific CCK-2 receptor antagonist reversed the effects of hypergastrinemia on intestinal crypt regeneration in mice that transgenically overexpress gastrin and mice rendered hypergastrinemic using omeprazole. Second, we have shown increased abundance of the CCK-2 receptor in intestinal epithelia and IEC-6 intestinal epithelial cells following ␥-radiation. Previous experiments have suggested only limited roles for amidated gastrin in the distal intestine. Although initial experiments reported slightly increased proliferation in the colonic crypts of INS-GAS mice compared with their wild-type FVB/N counterparts,11 this was not confirmed in 2 subsequent studies.12,32 In addition, we previously demonstrated no significant differences between INS-GAS and wild-type FVB/N mice in rates of small intestinal or colonic apoptosis and mitosis up to 24 hours following 1 or 8 Gy ␥-radiation.12

April 2006

MURINE INTESTINAL CRYPT REGENERATION

1177

Figure 7. (A) Western blot analysis of CCK-2 receptor and pan-actin expression in small intestinal and colonic epithelium from FVB/N mice in the unirradiated state and at various times following 14 Gy ␥-radiation. AGS cells were used as a negative control and AGS cells stably transfected with the CCK-2 receptor (AGS-GR) cells were used as a positive control. Murine samples were pooled from 4 mice per timepoint. (B) Photomicrographs of immunohistochemistry for the CCK-2 receptor in colonic (B and C) and small intestinal (D and E) crypts from FVB/N mice in the resting state (B and D) and 96 hours after 14 Gy ␥-radiation (C and E).

1178

OTTEWELL ET AL

GASTROENTEROLOGY Vol. 130, No. 4

Figure 8. (A) Nested RT-PCR analysis of IEC-6 cells for CCK-2 receptor mRNA and GAPDH mRNA at various times following 4 Gy ␥-radiation. St, rat stomach positive control. (B) Western blot analysis of CCK-2 and pan-actin in IEC-6 cells in the untreated state and 24 hours after 4 Gy ␥-radiation. (C) Gastrin/receptor binding in IEC-6 cells (expressed as percent of maximum binding after irradiation) 24 hours following exposure to 4 Gy ␥-radiation. 125I-labeled G17 competed with 0.1-100 nmol “cold” G17 for binding to the CCK-2 receptor (n ⫽ 1 in triplicate). (C) 3H-thymidine incorporation (cpm) in unirradiated IEC-6 cells (black) or following exposure to 4 Gy ␥-radiation (grey) following 48 hours in serum free media. White bars show cells similarly treated following administration of 10 nmol/L gastrin-17 in serum-free media for 48 hours (n ⫽ 4 in triplicate). *P ⬍ .05 compared to irradiated cells without addition of gastrin.

Other investigators have, however, suggested that amidated gastrin can exert effects on the murine intestine. For example, omeprazole-induced hypergastrinemia has been shown to increase the number of small intestinal adenomas in APCMin⫺/⫹ mice.31 The paucity of effects of amidated gastrin in the normal intestine is likely to reflect the observation that the CCK-2 receptor is expressed only weakly (if at all) in this tissue. Although CCK-2 receptor mRNA has been detected by RT-PCR in a proportion of human colorectal cancers33–35 and in some specimens of normal mouse36 and human33 colon, CCK-2 receptor transcripts were not detected by Northern blot, suggesting that mRNA is present at low abundance. In our studies, Western blot analysis and immunohistochemistry confirmed that in normal murine small intestine and colon the expression of CCK-2 receptor protein was below the limit of detection. However, CCK-2 protein was detected in both tissues following irradiation

and immunohistochemistry, suggesting that the receptor was located specifically in regenerating crypt epithelial cells (clonogenic cells and their immediate daughters). These data suggest that following damage by high doses of ␥-radiation, CCK-2 receptor expression is increased in the intestinal epithelia of mice. In the presence of hypergastrinemia, gastrin signals via this receptor to cause increased intestinal crypt survival. This observation is supported by experiments using the nontransformed rat intestinal epithelial cell line IEC-6. Following ␥-radiation, CCK-2 transcripts were detected by nested PCR, 125I-gastrin binding was increased, and 10 nmol/L gastrin-17 increased cellular proliferation. Other investigators have also been unable to demonstrate expression of the CCK-2 receptor in this cell line prior to irradiation, but have demonstrated increased proliferation upon addition of both gastrin-17 and progastrin.37,38 It is therefore possible that some of the effects of gastrins on IEC-6 cells occur as a result of signaling via novel gastrin

April 2006

receptors in addition to the CCK-2 signaling suggested by current experiments. There are accumulating data to suggest that the expression of the CCK-2 receptor is increased in response to injury or inflammation within gastrointestinal epithelia. For example, there is a rapid, specific increase in expression of the CCK-2 receptor at the margin of cryoulcers induced in rat stomach, resulting in increased wound healing following induction of hypergastrinemia.39 Additionally, we previously documented a 3-fold increased expression of the CCK-2 receptor in the epithelium of Barrett esophagus compared with esophageal epithelium from unaffected individuals and also increased proliferation of mucosal biopsies from patients with Barrett esophagus following addition of gastrin.26 Our current data suggest that induction of hypergastrinemia might also be a useful therapeutic maneuver for inflammatory disorders of the distal intestine, such as ulcerative colitis and Crohn’s disease as well as radiotherapy and chemotherapy-induced enteritis. Two small studies have indicated a beneficial effect of omeprazole in inflammatory bowel disease.40,41 It is hypothesized that induction of hypergastrinemia in patients with these conditions may promote epithelial cellular proliferation and hence healing of the epithelium. In conclusion, we have shown that both transgenic overexpression of gastrin and hypergastrinemia induced by omeprazole result in increased crypt survival in murine intestinal epithelia following ␥-radiation, 5-FU, and DSS. We have also demonstrated that the CCK-2 receptor antagonist YF476 reverses these effects and that the CCK-2 receptor is expressed in regenerating crypts of both small intestinal and colonic mucosa following high doses of ␥-radiation. These data suggest that the CCK-2 receptor is upregulated in intestinal epithelia following injury. Gastrin is able to promote healing by signaling via the CCK-2 receptor to increase epithelial regeneration. These observations suggest a novel role for amidated gastrin in protection against small intestinal and colonic injury in vivo and may have implications for the treatment of inflammatory conditions of the intestine and radiation enteritis.

References 1. Xian CJ. Roles of growth factors in chemotherapy-induced intestinal mucosal damage repair. Curr Pharm Biotechnol 2003;4: 260 –269. 2. Potten CS, Booth C, Pritchard DM. The intestinal epithelial stem cell: the mucosal governor. Int J Exp Pathol 1997;78:219 –243. 3. Booth C, Potten CS. Gut instincts: thoughts on intestinal epithelial stem cells. J Clin Invest 2000;105:1493–1499. 4. Watson AJ, Pritchard DM. Lessons from genetically engineered animal models. VII. Apoptosis in intestinal epithelium: lessons

MURINE INTESTINAL CRYPT REGENERATION

5.

6.

7. 8.

9. 10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

1179

from transgenic and knockout mice. Am J Physiol Gastrointest Liver Physiol 2000;278:G1–G5. Booth D, Haley JD, Bruskin AM, Potten CS. Transforming growth factor-B3 protects murine small intestinal crypt stem cells and animal survival after irradiation, possibly by reducing stem-cell cycling. Int J Cancer 2000;86:53–59. Farrell CL, Bready JV, Rex KL, Chen JN, DiPalma CR, Whitcomb KL, Yin S, Hill DC, Wiemann B, Starnes CO, Havill AM, Lu ZN, Aukerman SL, Pierce GF, Thomason A, Potten CS, Ulich TR, Lacey DL. Keratinocyte growth factor protects mice from chemotherapy and radiation-induced gastrointestinal injury and mortality. Cancer Res 1998;58:933–939. Booth D, Potten CS. Protection against mucosal injury by growth factors and cytokines. J Natl Cancer Inst Monogr 2001;16 –20. Potten CS. Interleukin-11 protects the clonogenic stem cells in murine small-intestinal crypts from impairment of their reproductive capacity by radiation. Int J Cancer 1995;62:356 –361. Dockray G, Dimaline R, Varro A. Gastrin: old hormone, new functions. Pflugers Arch 2005;449:344 –355. Schmassmann A, Reubi JC. Cholecystokinin-B/gastrin receptors enhance wound healing in the rat gastric mucosa. J Clin Invest 2000;106:1021–1029. Wang TC, Koh TJ, Varro A, Cahill RJ, Dangler CA, Fox JG, Dockray GJ. Processing and proliferative effects of human progastrin in transgenic mice. J Clin Invest 1996;98:1918 –1929. Ottewell PD, Watson AJ, Wang TC, Varro A, Dockray GJ, Pritchard DM. Progastrin stimulates murine colonic epithelial mitosis after DNA damage. Gastroenterology 2003;124:1348 –1357. Ottewell PD, Varro A, Dockray GJ, Kirton CM, Watson AJ, Wang TC, Dimaline R, Pritchard DM. COOH-terminal 26-amino acid residues of progastrin are sufficient for stimulation of mitosis in murine colonic epithelium in vivo. Am J Physiol Gastrointest Liver Physiol 2005;288:G541–G549. Biagini P, Monges G, Vuaroqueaux V, Parriaux D, Cantaloube JF, De Micco P. The human gastrin/cholecystokinin receptors: type B and type C expression in colonic tumors and cell lines. Life Sci 1997;61:1009 –1018. Lay JM, Jenkins C, Friis-Hansen L, Samuelson LC. Structure and developmental expression of the mouse CCK-B receptor gene. Biochem Biophys Res Commun 2000;272:837– 842. 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. 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. Booth C, Booth D, Williamson S, Demchyshyn LL, Potten CS. Teduglutide ([Gly2]GLP-2) protects small intestinal stem cells from radiation damage. Cell Prolif 2004;37:385– 400. Pritchard DM, Potten CS, Hickman JA. The relationships between p53-dependent apoptosis, inhibition of proliferation, and 5-fluorouracil-induced histopathology in murine intestinal epithelia. Cancer Res 1998;58:5453–5465. Flint N, Cove FL, Evans GS. A low-temperature method for the isolation of small-intestinal epithelium along the crypt-villus axis. Biochem J 1991;280:331–334. Watson SA, Clarke PA, Smith AM, Varro A, Michaeli D, Grimes S, Caplin M, Hardcastle JD. Expression of CCKB/gastrin receptor isoforms in gastro-intestinal tumour cells. Int J Cancer 1998;77: 572–577. McWilliams DF, Grimes S, Watson SA. Antibodies raised against the extracellular tail of the CCKB/gastrin receptor inhibit gastrinstimulated signalling. Regul Pept 2001;99:157–161.

1180

OTTEWELL ET AL

23. Watson SA, Clarke PA, Morris TM, Caplin ME. Antiserum raised against an epitope of the cholecystokinin B/gastrin receptor inhibits hepatic invasion of a human colon tumor. Cancer Res 2000;60:5902–5907. 24. 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. 25. Dockray GJ, Hamer C, Evans D, Varro A, Dimaline R. The secretory kinetics of the G cell in omeprazole-treated rats. Gastroenterology 1991;100:1187–1194. 26. Haigh CR, Attwood SE, Thompson DG, Jankowski JA, Kirton CM, Pritchard DM, Varro A, Dimaline R. Gastrin induces proliferation in Barrett’s metaplasia through activation of the CCK2 receptor. Gastroenterology 2003;124:615– 625. 27. Caplin ME, Clarke P, Grimes S, Dhillon AP, Khan K, Savage K, Lewin J, Michaeli D, Pounder RE, Watson SA. Demonstration of new sites of expression of the CCK-B/gastrin receptor in pancreatic acinar AR42J cells using immunoelectron microscopy. Regul Pept 1999;84:81– 89. 28. Stubbs M, Khan K, Watson SA, Savage K, Dhillon AP, Caplin ME. Endocytosis of anti-CCK-B/gastrin receptor antibody and effect on hepatoma cell lines. J Histochem Cytochem 2002;50:1213– 1217. 29. Varro A, Noble PJ, Wroblewski LE, Bishop L, Dockray GJ. Gastrincholecystokinin(B) receptor expression in AGS cells is associated with direct inhibition and indirect stimulation of cell proliferation via paracrine activation of the epidermal growth factor receptor. Gut 2002;50:827– 833. 30. Zavros Y, Rieder G, Ferguson A, Samuelson LC, Merchant JL. Hypergastrinemia in response to gastric inflammation suppresses somatostatin. Am J Physiol Gastrointest Liver Physiol 2002;282:G175–G183. 31. Watson SA, Smith AM. Hypergastrinemia promotes adenoma progression in the APC(Min-/⫹) mouse model of familial adenomatous polyposis. Cancer Res 2001;61:625– 631. 32. Singh P, Velasco M, Given R, Wargovich M, Varro A, Wang TC. Mice overexpressing progastrin are predisposed for developing aberrant colonic crypt foci in response to AOM. Am J Physiol Gastrointest Liver Physiol 2000;278:G390 –G399. 33. Biagini P, Monges G, Vuaroqueaux V, Parriaux D, Cantaloube JF, De Micco P. The human gastrin/cholecystokinin receptors: type B

GASTROENTEROLOGY Vol. 130, No. 4

34.

35.

36.

37.

38.

39.

40. 41.

and type C expression in colonic tumors and cell lines. Life Sci 1997;61:1009 –1018. Clerc P, Dufresne M, Saillan C, Chastre E, Andre T, Escrieut C, Kennedy K, Vaysse N, Gespach C, Fourmy D. Differential expression of the CCK-A and CCK-B/gastrin receptor genes in human cancers of the esophagus, stomach and colon. Int J Cancer 1997;72:931–936. Monstein HJ, Nylander AG, Salehi A, Chen D, Lundquist I, Hakanson R. Cholecystokinin-A and cholecystokinin-B/gastrin receptor mRNA expression in the gastrointestinal tract and pancreas of the rat and man. A polymerase chain reaction study. Scand J Gastroenterol 1996;31:383–390. Lay JM, Jenkins C, Friis-Hansen L, Samuelson LC. Structure and developmental expression of the mouse CCK-B receptor gene. Biochem Biophys Res Commun 2000;272:837– 842. Singh P, Narayan S, Adiga RB. Phosphorylation of pp62 and pp54 src-like proteins in a rat intestinal cell line in response to gastrin. Am J Physiol 1994;267:G235–G244. Singh P, Lu X, Cobb S, Miller BT, Tarasova N, Varro A, Owlia A. Progastrin1-80 stimulates growth of intestinal epithelial cells in vitro via high-affinity binding sites. Am J Physiol Gastrointest Liver Physiol 2003;284:G328 –G339. Schmassmann A, Reubi JC. Cholecystokinin-B/gastrin receptors enhance wound healing in the rat gastric mucosa. J Clin Invest 2000;106:1021–1029. Dickinson JB. Is omeprazole helpful in inflammatory bowel disease? J Clin Gastroenterol 1994;18:317–319. Guslandi M, Tittobello A. Symptomatic response to omeprazole in inflammatory bowel disease. J Clin Gastroenterol 1996;22:159 – 160.

Received April 26, 2005. Accepted December 14, 2005. Address requests for reprints to: Dr D.M. Pritchard, Division of Gastroenterology, 5th Floor UCD Building, University of Liverpool, Daulby Street, Liverpool, L69 3GA, UK. e-mail: [email protected]; fax: 44 151 794 6825 Funded by North West Cancer Research Fund, National Association of Colitis and Crohn’s Disease, Royal Liverpool and Broadgreen University Hospitals NHS Trust R and D fund, Medical Research Council. DMP is a Wellcome Trust Advanced Clinical Fellow. The authors thank C. McLean for gastrin radioimmunoassay.