European Journal of Clinical Investigation (2001) 31, 812±820

CCK-B/gastrin receptors in human colorectal cancer

F. Schmitz², J.-M. Otte*, H. U. Stechele*, B. Reimann*, T. Banasiewicz³, U. R. FoÈlsch*,
W. E. Schmidt² and K.-H. Herzig*
*
Department of Medicine, Christian-Albrechts-University Kiel, Germany, ²Department of Medicine I, Ruhr-University
Bochum, Germany, ³Department of Surgery, University of Poznan/Poland

Abstract Background Mature amidated gastrin (G17 amide) mediates its effects in the gastrointestinal
tract by activating G protein-coupled CCK-B/gastrin receptors. Although trophic actions of
gastrin on the gastric mucosa have been well-established, the effect of G17 amide, progastrin
and intermediates to colon neoplasia in humans is controversial. While epidemiological
evidence from patients with elevated serum gastrin levels related to pernicious anaemia does not
support an increased risk for colon cancer [1], a recent study suggests that prolonged
hypergastrinaemia is associated with an increased risk for colon cancer [2]. The extent to which
trophic actions of gastrin in colorectal cancer are mediated by functional gastrin receptors
remains to be defined. The aim of the present study was to determine CCK-B/gastrin receptor
expression, structure, and function in 79 patients with colon cancer.
Materials and methods CCK-B/gastrin receptor cDNAs were isolated from 79 human
colorectal cancer specimens and 15 control tissues, subcloned into the eukaryotic expression
vector pCR3.1 and subjected to DNA sequence analysis. Wild-type and mutant cDNAs were
transiently expressed in COS-7 cells to determine ligand affinities by 125I-labelled CCK-8S
competition binding. Activation of the MAP kinase signalling cascade by G17 amide was
determined in transfected Colo 320 cells expressing the wild-type or mutant CCK-B/gastrin
receptors. Clonal expansion of single cells was quantified in transfected Colo 320 cells.
Results Gastrin mRNA is expressed in 44% of colorectal cancers and in 13% of control tissues.
CCK-B/gastrin receptor mRNA is expressed in 38% of colorectal cancers and 13% of normal
colonic tissue. Co-expression of gastrin and CCK-B/gastrin receptor message is significantly
increased in colorectal cancer specimens (32% vs. 0%). There is no correlation between CCK-B/
gastrin receptor expression and disease stage or histological grading. DNA sequence analysis
revealed one spontaneous CCK-B/gastrin receptor mutation within the third intracellular loop
with an exchange of valine-287 for phenylalanine. Pharmacological characterisation of the
287
V ! F CCK-B/gastrin receptor reveals wild-type affinities for G17 amide, glycine-extended
gastrin, CCK-8S and L-365,260. Mutation 287V ! F is associated with a loss of gastrin-induced
MAPK p44/p42 signalling in Colo 320 cells while clonal expansion from single cells is increased by
53´1 ^ 15´9% when compared to Colo 320 cells expressing wild-type CCK-B/gastrin receptors.
Conclusions Structural alterations of CCK-B/gastrin receptors may account for increased
growth-promoting effects of amidated gastrins in colorectal cancer.
Keywords CCK-B/gastrin receptor, colon cancer, gastrin, G protein-coupled receptor.
Eur J Clin Invest 2001; 31(9): 812±820

Department of Medicine, Christian-Albrechts-University Kiel, Introduction

Germany (J.-M. Otte, H. U. Stechele, B. Reimann, U. R. FoÈlsch,
K.-H. Herzig); Department of Medicine I, Ruhr-University
Bochum, Germany (F. Schmitz, W. E. Schmidt); Department of
Surgery, University of Poznan/Poland (T. Banasiewicz).
Correspondence to: Frank Schmitz, M.D., Department of
Medicine I, Ruhr-University Bochum, Gudrunstrasse 56,
D-44791 Bochum, Germany. Tel.: 1 49-234-509-2378; fax:
1 49-234-509-2309; e-mail: frank.schmitz@uni-bochum.de
Received 31 January 2001; accepted 13 May 2001
Colorectal cancer is the second leading cause of cancer
related deaths in most Western industrialised countries
affecting both genders with similar frequency [3]. The
pathogenesis of colon cancer has been characterised as a
transition from premalignant lesions via polyp formation
to malignant disease involving a series of somatic muta-
tions including the activation of the ras oncogene and the
inactivation of the tumour suppressor gene p53 [4±7].

Q 2001 Blackwell Science Ltd


The regulatory peptide gastrin, which primarily functions
as an acid secretagogue by activating CCK-B/gastrin
receptors on distinct cell types within the gastric mucosa,
has been claimed to exhibit trophic actions on the colonic
epithelium in vivo and in vitro [2,8]. It remains as yet largely
undefined how gastrin interferes with the progression from
premalignant colonic crypt lesions to invasive cancer but
substantial evidence has accumulated that autocrine growth
loops may at least in part account for its mitogenic potency
[9]. This finding is supported by extensive studies in various
colon cancer cell lines which are sensitive to mature gastrin
with respect to proliferation [8,10].
The extent to which mature gastrin is expressed in
colorectal cancer is unclear. Reports in the literature as
reviewed by Baldwin and Shulkes [11] range between 0 and
100% for C-terminally amidated gastrin as assessed by
radioimmunoassay in colorectal cancer specimens. A puta-
tive mechanism by which gastrin gene expression is regulated
in colorectal cancer, however, has been recently proposed by
Nakata and co-workers [12]. Oncogenic ras was postulated to
induce gastrin gene expression through activation of the Raf±
MEK±ERK signal transduction pathway.
Epidemiological evidence from studies in patients with
elevated serum gastrin levels from sources other than colon
provide conflicting data for gastrin as a trophic factor for the
colonic epithelium. Patients suffering from the Zollinger±
Ellison syndrome with endogenous hypergastrinaemia have
thus far not been identified as being at increased risk for
colorectal cancer [13]. The same state of knowledge can be
cited for patients with pernicious anaemia [14].
In contrast to the epidemiological evidence against trophic
actions of gastrin on the colon, Thorburn and co-workers
recently found a strong association of hypergastrinaemia with
increased risk of colorectal carcinoma in a nested case±
control study among 128 992 subscribers to a Californian
health maintenance programme [2]. Adding to the disparity
of findings concerning gastrin as a growth factor was their
observation that the relative risk for cancer in persons with
high gastrin levels further increased when early onset cases
were excluded to rule out autocrine hormone production by
pre-existing but as yet undetected tumours or polyps.
Animal studies in mice lacking mature gastrin through
targeted disruption of the murine gastrin gene further
support that gastrin may exhibit trophic effects on the normal
colonic mucosa in vivo [15,16]. Gastrin-deficient mice had a
significantly decreased rate of proliferation throughout the
entire colon when compared to littermate controls as assessed
by bromodeoxyuridine labelling of DNA in vivo.
Molecular cloning of mutant G protein-coupled receptors
has led to the identification of disease-causing mutations in
humans. Mutations identified to date either activate or
decrease receptor function [17]. Disease-causing mutations
of CCK-B/gastrin receptors have not been identified
although the cDNA encoding the CCK-B/gastrin receptor
from the rodent Mastomy natalensis has been utilised to
explore the genetic predisposition of this species to develop
gastric carcinoid tumours. In search for a CCK-B/gastrin-
receptor-mediated, and thus hypergastrinaemia-related,
mode of action, Schaffer et al. [18] noted higher basal inositol
Q 2001 Blackwell Science Ltd, European Journal of Clinical Investigation, 31, 812±820
CCK-B/gastrin receptors and colorectal cancer 813
phosphate levels in COS-7 cells transiently expressing the
Mastomy cDNA when compared to cells expressing the
human wild-type. The difference in basal signalling activity
was attributed to a species difference of four amino acid
residues within the sixth transmembrane receptor domain.
We therefore aimed for the identification of molecular as
yet unidentified isoforms of CCK-B/gastrin receptors in
human colorectal cancer specimens with particular
growth-promoting properties due to an acquired mutation
in the process of malignant transformation.
Methods
All chemicals were reagent grade and commercially available.
125
I-labelled CCK-8S (specific radioactivity 2200 Ci m-
mol21) was purchased from Amersham-Buchler, Braunsch-
weig, Germany. G17amide and sulphated CCK octapeptide
(CCK-8S) were from Bachem, Heidelberg, Germany. The
CCK-B/gastrin receptor antagonist L-365,260 was kindly
provided by Dr L. Iverson, ML Laboratories, Liverpool,
UK. Enzymes including restriction endonucleases for sub-
cloning purposes, Taq DNA polymerase and Superscript IIe
reverse transcriptase were from Gibco, Eggenstein, Germany.
Patients
Patients included in the study were undergoing either
lower gastrointestinal endoscopy for diagnostic reasons or
segmental large bowel resection for previously diagnosed
colorectal cancer at the University Medical Centres in
Kiel, Germany and Poznan, Poland. All patients gave their
informed consent to participitate in the study. The study
was approved by the ethical boards of the Universities of
Kiel and Poznan under full consideration of the declara-
tion on human rights from Helsinki.
Patient characteristics are given in Table 1. The mean age
of patients with colorectal cancer was 59´5 ^ 14´8 years.
Histological classification revealed adenocarcinomas in 89%
Table 1 Clinical characteristics of patients
n 79
Age (years) 59´5 ^ 14´8
range 22±83
Sex (male/female) 45/34
Tumour Stage (n)
Dukes A 6 (8%)
Dukes B 31 (39%)
Dukes C 37 (47%)
Dukes D 5 (6%)
Tumour Grading (n)
Grade 1 38 (48%)
Grade 2 21 (27%)
Grade 3 20 (25%)
m ˆ males. f ˆ females. Age is given as means ^ SD.
814 F. Schmitz et al.
of cases, mucinous carcinoma in 6% and signet ring cell
carcinoma in 5% of cases. Fifteen control cases were
included. Fifty-three per cent of all cancer patients were
staged according to Dukes class C and D indicating advanced
tumour progression. Tumours were well-differentiated
(grade 1) in 38 cases, moderately differentiated (grade 2) in
21 cases, and undifferentiated (grade 3) in 20 cases.
Anti-CCK-B/gastrin receptor antiserum
A specific antiserum directed against a synthetic peptide
comprising the N-terminal sequence of the human CCK-B/
gastrin receptor (MELLKLNRSVQGTGPGPGA) was
raised in rabbit. The synthetic peptide was linked to keyhole
limpet haemocyanin serving as carrier protein by esterifica-
tion to m-maleimidobenzoyl-N-hydroxysuccinimide. Anti-
sera from various time-points following immunization were
tested for immunoreactivity against the immunizing peptide
using an enzyme-linked immunosorbent assay. In brief,
microtitre plates were coated overnight at room temperature
with the immunizing peptide in 100 mL phosphate-buffered
saline (PBS) at a concentration of 2 mg mL21. Plates were
then washed three times with washing buffer (PBS/0´05%
Tween-20) and incubated with 150 mL blocking solution
[PBS/1% bovine serum albumin (BSA)] for 1 h at 37 8C and
subsequently washed again three times in washing buffer.
Then, 100 mL of diluted antisera collected 4±16 weeks after
immunization was added to each well and incubated for 2 h at
378C. Following three washes in washing buffer, 100 mL
alkaline phosphatase-labelled anti-rabbit immunoglobulin G
(IgG) from goat was added. After an incubation of 2 h at
378C, plates were washed three times and 100 mL alkaline
phosphatase substrate ABTSe (Boehringer Mannheim,
Germany) was dispensed into each well. Antigen-bound
anti-CCK-B/gastrin receptor antibody was detected color-
imetrically in a microtitre plate reader at 405±410 nm
wavelength. Preimmune antisera from individual rabbits
were included as negative controls.
The specificity of the anti-CCK-B/gastrin receptor
antiserum was also confirmed by Western blot analysis of
cell lysates from transiently transfected COS-7 cells
expressing either the human CCK-B/gastrin receptor or
the structurally related but N-terminally differing CCK-A
receptor to rule out nonspecific cross-reactivity.
Immunohistochemistry
Localisation of CCK-B/gastrin receptor expressing cells
within colorectal cancer specimens was performed using the
DAB kit from Vector Laboratories according to the manu-
facturer's recommendations. Sections of snap frozen tissue
were air dried, incubated in acetone at 2 20 8C for 10 min and
further treated with blocking solution, primary antibody
(dilution 1 : 400) and horseradish peroxidase-conjugated
secondary antibody (1 : 1000). Antigen-bound peroxidase
activity was visualised using 3,3 0 -diaminobenzidine as chro-
mogen. Negative controls included tissue staining with
Q 2001 Blackwell Science Ltd, European Journal of Clinical Investigation, 31, 812±820
preimmune antisera collected from identical rabbits directly
before administration of the antigen. Lamina propria cells
were defined by their appearance under a microscope in serial
tumour sections which were stained with haematoxylin and
eosin and antibodies directed against CD3 and CD20.
Reverse transcription-polymerase chain reaction
(RT-PCR)
All experiments were performed under RNase-free condi-
tions using glassware washed with diethylpyrocarbonate-
treated water (Sigma Chemicals, Munich, Germany) and
baked at 180 8C for 12 h. RNA from human colorectal
cancer specimens was isolated with the RNeasy total RNA
extraction kit (Qiagen, Hilden, Germany) according to the
manufacturer's recommendation. First-strand cDNA synth-
esis was accomplished by RTof 1 mg purified total RNA. The
RT was primed with 500 ng oligo-dT (Promega, Madison,
WI) and carried out in a total volume of 50 mL with 200 U
SuperScript IIe reverse transcriptase (GibcoBRL, Eggen-
stein, Germany) at 42 8C for 50 min followed by 10 min
incubation at 70 8C. CCK receptor mRNA expression was
determined by PCR using 100±500 ng first strand cDNA
and gene-specific primers at a final concentration of 0´2 mM in
a total volume of 50 mL PCR Supermixe (Gibco, Eggen-
stein, Germany). The primer sequences for the detection of
CCK-B/gastrin receptor cDNA were 5 0 -ATGGAGCTGC-
TAAAGCTGAACCGG-3 0 (sense) and 5 0 -AACATTCG-
CACCACGCGCTTC-3 0 (antisense). Primer sequences for
the detection of gastrin-34 were 5 0 -AGCTGGGACCC-
CAGGGT-3 0 (sense) and 5 0 -GAAGTCCATCCATCCA-
TAGG-3 0 (antisense). Primer sequences for the detection of
the DCCK-B/gastrin receptor were from McWilliams et al.
[19]. PCR conditions were chosen as follows. Samples were
initially denatured at 94 8C for 3 min, cycle conditions
consisted of denaturation at 94 8C for 45 s, annealing at
60 8C for 45 s, and extension at 72 8C for 45 s for a total of 35
repeats followed by a single terminal extension at 72 8C for
7 min. Six microlitres of the PCR reaction were subjected to
an 8% polyacrylamide gel electrophoresis and subsequent
staining with ethidium bromide (Sigma Chemicals). The
specificity of the RT-PCR was confirmed by direct sequen-
cing of purified PCR products.
As controls for the RNA purification, specific primers
for the human glyceryl aldehyde-3-phosphate dehydro-
genase (GAPDH) gene were used (Clontech, Heidelberg,
Germany). Primer sequences were GAPDH 1, 5 0 -TGAA
GGTCGGAGTCAACGGATTTGGT-3 0 and GAPDH 2,
5 0 -CATGTGGGCCATGAGGTCCACCAC-3 0 .
Radioligand binding studies
COS-7 cells (1´5  106) were plated in 10-cm culture dishes
(Nunc, Karlsruhe, Germany) and grown overnight in
Dulbecco's minimal essential medium (DMEM)/10% fetal
calf serum (FCS) at 37 8C. The cells were transfected [20]
with 5±7 mg of the eukaryotic expression vector pCR 3.1

(Invitrogen, Groningen, the Netherlands) containing the
cDNAs encoding either the wild-type human or mutant
CCK-B/gastrin receptor. The following day, cells were split
into 24-well dishes (0´5  10425´0  104 cells/well). After an
additional 24 h, competition binding experiments were per-
formed in Hank's balanced salt solution supplemented with
25 mM HEPES (pH 7´3), 0´2% bovine serum albumin and
0´15 mM phenylmethylsulphonylfluoride. Twenty picomoles
125
I-labelled CCK-8S (Amersham-Buchler, Braunschweig,
Germany) were used as radioligand. After incubation for
80 min at 37 8C, cell monolayers were washed three times with
Hank's balanced salt solution, hydrolysed in 1 M NaOH, and
bound radioactivity was quantified. Unlabelled competitors
which were tested included the peptide agonists CCK-8S,
gastrin-17 amide (Bachem, Heidelberg, Germany), the non-
peptide antagonists L-365,260 (ML Laboratories, Liverpool,
UK), and glycine-extended gastrin (Auspep, Parkville Victoria,
Australia). All half-maximally inhibitory concentrations (IC50)
reported represent data obtained from at least three indepen-
dent experiments, each analysed using computerized nonlinear
curve fitting (INPLOT 4´0, GraphPad, San Diego, CA).
Phosphorylation of ERK-1 and ERK-2
Analysis of the phosphorylation status of p44 and p42 MAP
kinases (ERK-1 and ERK-2) was accomplished with a
monoclonal antibody directed against p44/42 MAP kinase
from New England Biolabs, Frankfurt, Germany. Studies
were performed in transiently transfected Colo 320 cells.
Proteins were extracted by incubation for 10 min at 4 8C in
lysis buffer (10 mM Tris±HCl, 5 mM EDTA, 50 mM NaCl,
30 mM sodium pyrophsophate, 50 mM NaF, 100 mM
Na3VO4, and 1% Triton X-100, pH 7´6) supplemented
with protease inhibitor (1 mM phenylmethylsulphonylfluor-
ide). Protein concentrations were measured by Bio-Rad
protein assay reagent (Bio-Rad, Hercules, CA). One hundred
microgrammes of total protein were loaded per lane and
separated under reducing conditions on an 8% sodium
dodecylsulphate gel. Following gel electrophoresis, proteins
were transferred to nitrocellulose membranes by wet electro-
phoretic transfer. Membranes were blocked for 1 h in 5%
nonfat dry milk, washed and then incubated for an additional
hour with the monoclonal antibody against phosphorylated
p44 and p42 MAP kinase. Membranes were washed four
times and incubated with the secondary antibody for 1 h
followed by four additional washes. Immunoreactive bands
were visualised using the enhanced chemoluminescence
(ECL) method and exposure to a Hybond-N film (Amer-
sham, Little Chalfont, UK). The relative expression of
phosphorylated p44 and p42 protein was quantified by
densitometric scanning of the Hybond-N films.
Clonal expansion of Colo 320 cells expressing the
human CCK-B/gastrin receptor wild-type or
287
V ! F CCK-B/gastrin receptor
Colo 320 cells were transiently transfected with either the
Q 2001 Blackwell Science Ltd, European Journal of Clinical Investigation, 31, 812±820
CCK-B/gastrin receptors and colorectal cancer 815
wild-type or 287V ! F mutant CCK-B/gastrin receptor
cDNA using the Effectenee reagent (Qiagen, Hilden,
Germany) according to the manufacturer's instructions.
Following transfection, cells were seeded at very low
density to obtain single cell clones (5000 cells per 35-mm
culture dish). Cell clones were incubated in DMEM
containing 5% FCS for up to 72 h. Cells per clone were
counted at 24, 48 and 72 h post-transfection. To confirm
identical transfection efficiencies in Colo 320 cells, CCK-
B/gastrin receptor expression was tested by 125I-labelled
CCK-8S binding as described before.
Results
CCK-B/gastrin receptor expression in patients with
colorectal cancer and human colon cancer cell lines
Histological grading of colorectal cancer specimens and
clinical disease stages are given in Table 1. CCK-B/gastrin
receptor transcripts were detected in 30 out of 79 (38%)
colon cancer specimens (Table 2). CCK-B/gastrin recep-
tor mRNA expression did not correlate with the disease
stage or histological grading. In control specimens from
normal colorectal mucosa, CCK-B/gastrin receptor
mRNA was detected in two out of 15 samples tested.
PCR products were subjected to dideoxy sequence analysis
and compared to CCK-B/gastrin receptor sequences as
deposited in GenBank. All CCK-B/gastrin receptors ampli-
fied from colorectal cancers were wild-type sequence as
deposited in GenBank under accession number L08112with
the exception of one cDNA encoding a single-nucleotide
mutation resulting in an exchange of valine-287 for
phenylalanine (Fig. 1). Additional amplifications for the
DCCK-BR isoform as described by Miyake [21] or insertion
of intron IV as first described by Hellmich et al. [22] did not
reveal a single carcinoma expressing the N-terminally
truncated CCK-B/gastrin receptor mutant or CCK-B/
gastrin receptor splice variant which retains intron IV.
Gastrin and CCK-B/gastrin receptor mRNA
expression
Gastrin mRNA was detected in 35 out of 79 (44%)
Table 2 CCK-B/gastrin receptor mRNA expression in colon
cancer and normal tissue
Cancer (n ˆ 79) Normal
GI GII GIII tissue
(n ˆ 38) (n ˆ 21) (n ˆ 20) (n ˆ 15)
mRNA expression 34% 19% 65% 13%
(13/38) (4/21) (13/20) (2/15)
Total 38% 13%
(30/79) (2/15)
816 F. Schmitz et al.

Figure 1 Schematic illustration of mutant

CCK-B/gastrin receptor. Molecular cloning
of CCK-B/gastrin receptor cDNAs from
colorectal cancer specimens yielded a single-
point mutation in a Dukes C/grade III
tumour with an exchange of valine-287 for
phenylalanine. The point mutation projects
into the putative third intracellular receptor
loop.

tumours vs. two out of 15 (13%) control tissues (Table 3).
Gastrin expression was higher in grade I tumours (55%)
than in grade II (38%) or grade III (30%). Expression of
CCK-B/gastrin receptors and gastrin gene is shown in
Fig. 2. Thirty-two per cent of colorectal cancers expressed
both CCK-B/gastrin receptors and gastrin ligand. In
normal colonic mucosa co-expression of receptor and
ligand was not detected.
CCK-B/gastrin receptors were detected using an epi-
tope-specific anti-CCK-B/gastrin receptor antiserum from
rabbit. CCK-B/gastrin receptors mapped to cancer cells as
well as to polymorphonuclear lamina propria cells (LPC)
within the cancer specimen (Fig. 3). Staining for CCK-B/
gastrin receptor immunoreactivity correlated with CCK-B/
gastrin receptor mRNA expression in 12 out of 15
tumours tested. In three tumour samples, there was no
immunoreactivity for CCK-B/gastrin receptors although
mRNA encoding the CCK-B/gastrin receptor was detect-
able by RT-PCR.
287
Characterisation of mutant V ! F and wild-type
CCK-B/gastrin receptors
radioligand was not achieved with nonamidated glycine-
extended gastrin-17 in concentrations as high as 1 mM.
Clonal expansion of transfected Colo 320 cells was
accelerated in cells expressing the 287V ! F CCK-B/gastrin
receptor (Fig. 5). Non-transfected Colo 320 cells and Colo
320 cells expressing the human CCK-B/gastrin receptor
wild-type exhibited identical expansion kinetics (10´8 ^ 1´1
vs. 11´3 ^ 1´3 cells per clone) while clones of Colo 320
cells with the 287V ! F mutation within the CCK-B/gastrin
receptor had a significantly increased cell number
(53´1 ^ 15´9%) within 72 h. At day 3 after transfection,
single cell clones had expanded to 17´3 ^ 1´8 cells in Colo
320 cells expressing the 287V ! F CCK-B/gastrin receptor
while Colo 320 cells expressing the CCK-B/gastrin receptor
wild-type consisted of 11´3 ^ 1´3 cells per clone (P , 0´05
by Student's t-test).
Gastrin-17 stimulates phosphorylation of p44 and p42
MAP kinase in Colo 320 cells transiently expressing the
human CCK-B/gastrin receptor wild-type cDNA. Max-
imal stimulation was induced with 10 nM gastrin-17 within
5 min. Colo 320 expressing the 287V ! F CCK-B/gastrin
receptor did not phosphorylate p44 or p42 MAP kinase
following stimulation with gastrin-17 (Fig. 6).

Wild-type and mutant 287V ! F CCK-B/gastrin receptors

were transiently transfected into COS-7 cells. 125I-labelled
CCK-8S binding was assessed in the presence of increas-
ing concentrations of unlabelled competitor. Half-maximal
displacement of 125I-labelled CCK-8S is shown in Fig. 4.
Both CCK-B/gastrin receptors exhibited wild-type affi-
nities for gastrin-17 amide, CCK-8S and the specific
antagonist L-365,260. Half-maximal displacement of

Table 3 Gastrin mRNA expression in colon cancer and normal

tissue

Cancer (n ˆ 79) Normal

GI GII GIII tissue
(n ˆ 38) (n ˆ 21) (n ˆ 20) (n ˆ 15)

mRNA expression 55% 38% 30% 13%

(21/38) (8/21) (6/20) (2/15)
Figure 2 Co-expression of gastrin and CCK-B/gastrin receptor
Total 44% 13%
mRNA. Expression of gastrin and CCK-B/gastrin receptors was
(35/79) (2/15)
assessed by RT-PCR as described in Methods.

Q 2001 Blackwell Science Ltd, European Journal of Clinical Investigation, 31, 812±820
 CCK-B/gastrin receptors and colorectal cancer 817

Figure 3 Determination of the site of CCK-B/gastrin receptor

expression. CCK-B/gastrin-receptor-expressing cells were stained
as described in Methods. CCK-B/gastrin receptor expression
maps to cancer cells and to mononuclear cells with the tumour
stroma (!).

Discussion

Colorectal cancers arise from single crypt lesions and

progress to adenomatous polyps and later to invasive
carcinomas [5,7]. An early and central event in the majority
of cancers is the inherited or acquired mutation of the
adenomatous polyposis coli gene [6] followed later by an
accumulating series of somatic mutations. These include
mutations in the K-ras proto-oncogene and loss-of-function
mutations in the p53 tumour suppressor gene [4]. Besides
genetic alterations, several growth factors, including ami-
dated and nonamidated forms of gastrin and their
receptors, have been claimed to be of critical importance
for tumour progression [9,11,23±25]. Although nonami-
dated forms of gastrin are believed to mediate their effects
in colon cancer through distinct but as yet unidentified
receptors, amidated gastrins may mediate their growth
promoting effects via different receptor subtypes [24].
In the present study, we demonstrate CCK-B/gastrin
receptor mRNA expression in 38% of human colorectal
cancers and coexpression of receptor and ligand in 32% of
tumours. Co-expression of receptor and ligand is a
prerequisite for autocrine growth stimulation as inactiva-
tion of gastrin with antisense gastrin cDNA suppressed the
Figure 4 Radioligand binding studies. The human CCK-B/
gastrin receptor wild-type and 287V ! F mutant cDNAs were
subcloned into the eukaryotic expression vector pCR3.1 and
transiently expressed in COS-7 cells. IC50 values were calculated
as described in Methods.
proliferative potential in human colon cancer cell lines
Colo 320 and HCT 116 significantly [9]. We also
determined the cellular sites of expression within color-
ectal cancer sections. Interestingly, CCK-B/gastrin recep-
tor expression mapped to carcinoma cells but also to
polymorphonuclear (PMN) cells within the tumour
stroma. CCK-B/gastrin receptor expression in PMN cells
has caught little attention but Iwata et al. recently
identified CCK-B/gastrin receptors in the lamina propria
of rat stomach [26]. The findings of our group and others
confirm that CCK-B/gastrin receptor expression in RNA
preparations from tumour tissue does not necessarily
reflect properties which are unique to cancer cells.

Q 2001 Blackwell Science Ltd, European Journal of Clinical Investigation, 31, 812±820
818 F. Schmitz et al.
Figure 5 Clonal expansion of Colo 320 cells expressing wild-
type or 287V ! F CCK-B/gastrin receptors. Colo 320 cells were
transiently transfected with the cDNAs encoding the wild-type
and mutant CCK-B/gastrin receptor or treated as transfected cells
but not exposed to foreign DNA (mock). Cells were seeded at very
low density to follow clonal expansion of individual cells. Cell
numbers of 15 individual clones were counted at 24, 48 and 72 h
after transfection. Data are means ^ SD of three independent
experiments. Clonal expansion was significantly accelerated by
53´1 ^ 15´9% (*P , 0´05) in Colo 320 cells expressing mutant
287
V ! F CCK-B/gastrin receptors when compared to nontrans-
fected (mock) or CCK-B/gastrin receptor wild-type-transfected
Colo 320 cells.
However, tumours in general have been more and more
characterised as complex tissues with interactions of
primary cancer cells, fibroblasts, immune cells and
endothelial cells [27]. The significant difference in CCK-
B/gastrin receptor expression in colon cancer specimens
when compared to normal colonic tissue, however,
indicates that increased expression of CCK-B/gastrin
receptors may be linked to malignant transformation of
the colonic epithelium. Recent data even suggest that the
mRNA expression levels for gastrin and CCK-B/gastrin
receptors undergo changes during transition from polyps
to invasive adenocarcinomas. Watson and co-workers
Q 2001 Blackwell Science Ltd, European Journal of Clinical Investigation, 31, 812±820
observed in 10 patients with adenocarcinoma of the
colon and 18 subjects with colonic adenomas by quanti-
tative PCR analysis that gastrin mRNA content reaches
peak levels with increasing malignant progression while
CCK-B/gastrin receptor levels where highest at the
adenoma stage [28].
One putative mechanism by which gastrin expression is
regulated in colon cancer cells has been established
recently [12]. Oncogenic ras has been shown to account
for high gastrin levels in colorectal cancers but it remains
unknown to what extent and by which mechanism CCK-
B/gastrin receptor expression is regulated in the colonic
tumorigenesis.
Spontaneous mutations of gastrin receptors have been
described earlier. A CCK-B/gastrin receptor cDNA
encoding a truncated isoform lacking the N-terminal
extracellular domain has been isolated from human
stomach [21]. The mutant receptor, termed DCCK-BR,
was recently found to be co-expressed with the wild-type
receptor in eight out of 13 gastrointestinal tumour cell
lines, including eight cell lines of colorectal origin [19].
From a pharmacological point of view, however, it appears
unlikely that the truncated isoform confers gastrin
sensitivity and growth promoting activity since deletion
of the N-terminal domain results in a . 30-fold decreased
affinity for gastrin [20,21]. Hellmich et al. cloned another
cDNA from human colon cancer encoding a CCK-B/
gastrin receptor splice variant [22]. This splice variant
retains intron IV during RNA processing, resulting in an
in-frame insertion of 69 amino acids within the third
intracellular loop. This CCK-B/gastrin receptor splice
variant exhibits both low- and high-affinity binding sites
for amidated gastrin with IC50 values of 315 nM and
0´2 nM, respectively. Affinity for glycine-extended gastrin
is increased four-fold, with half-maximal displacement of
125
I-labelled gastrin occurring at 318 nM glycine-
extended gastrin when compared to wild-type CCK-B/
gastrin receptors (IC50 ˆ 1´3 mM). The CCK-B/gastrin
receptor mutant 287V ! F identified in the present study
Figure 6 Phosphorylation of ERK-1 and
ERK-2 by mutant 287V ! F and wild type
CCK-B/gastrin receptors. Phosphorylation
status of p44 or p42 MAP kinase was
assessed following stimulation with amidated
gastrin as described in Methods. Extracellular
regulated protein kinases are not
phosphorylated upon stimulation with G17
amide in Colo 320 cells expressing the
287
V ! F mutant CCK-B/gastrin receptor.

retains all the pharmacological characteristics of a wild-
type CCK-B/gastrin receptor, i.e. high affinity for ami-
dated gastrin, CCK-8S and the benzodiazepine-derived
antagonist L-365,260. Mutation 287V ! F does not result
in an increased affinity for glycine-extended gastrin which
has been proposed to mediate mitogenic effects through
receptors other than classical CCK-B/gastrin receptors in
colorectal cancer [11]. The glycine-extended gastrin
receptor first described in AR4-2J cells which are derived
from rat pancreas does not exhibit high affinity for
amidated gastrin or classical CCK-B/gastrin receptor
antagonists [29]. Therefore, binding data do not support
that the single point mutation within the putative third
intracellular receptor segment is the predominant receptor
for nonamidated forms of gastrin. However, we have
shown that (i) CCK-B/gastrin receptor mutations do occur
in colorectal cancer and that (ii) mutations in receptor
domains which are critical for effector pathway activation
and not ligand selectivity may display increased mitogenic
properties. In addition, extracellular regulated mitogen-
activated kinase activity is not required for clonal expan-
sion of Colo 320 cells expressing the 287V ! F CCK-B/
gastrin receptor mutant. Interestingly, this finding is
consistent with glycine-extended gastrin-signalling in
pancreatic AR4-2J cells. In AR4-2J cells, glycine-extended
gastrin and amidated gastrin induce different programmes
of early gene activation [30]. While glycine-extended
gastrin induces acitivity of c-Jun amino terminal kinase,
gastrin exhibits activation of the MAP kinase signalling
pathway. Mutation 287V ! F results in MAP kinase
effector pathway inactivation while clonal expansion from
single cells is increased. Future studies will address which
signal transduction pathways are involved in the mediation
of mitogenic properties of this novel cDNA.
Acknowledgements
The study was supported in part by a research grant from
the Deutsche Froschungsgemeinschaft to Frank Schmitz
(DFG Schm 1073/4±1).
References
1 Brinton LA, Gridley G, Hrubec Z, Hoover R, Fraumeni JF Jr.
Cancer risk following pernicious anaemia. Br J Cancer
1989;59:810±13.
2 Thorburn CM, Friedman GD, Dickinson CJ, Vogelman JH,
Orentreich N, Parsonnet J. Gastrin and colorectal cancer.
Gastroenterology 1998;115:275±80.
3 Potter JD. Colorectal cancer. molecules and populations. J
Natl Cancer Inst 1999;91:916±32.
4 Bos JL, Fearon ER, Hamilton SR, Verlaan-de Vries M, van
Boom JH, van der Eb AJ et al. Prevalence of ras gene
mutations in human colorectal cancers. Nature 1987;327:293.
5 Fearon ER, Hamilton SR, Vogelstein B. Clonal analysis of
human colorectal tumors. Science 1987;238:193.
Q 2001 Blackwell Science Ltd, European Journal of Clinical Investigation, 31, 812±820
CCK-B/gastrin receptors and colorectal cancer 819
6 Groden J, Thliveris A, Samowitz W, Carlson M, Gelbert L,
Albertsen H et al. Identification and characterization of the
familial adenomatous polyposis coli gene. Cell 1991;66:589±
600.
7 Fearon RR, Vogelstein B. A genetic model for colorectal
tumorigenesis. Cell 1990;61:759±67.
8 Watson SA, Durrant LG, Morris DL. Growth-promoting
action of gastrin on human colonic and gastric tumour cells
cultured in vitro. Br J Surg 1988;75:342±5.
9 Singh P, Owlia A, Varro A, Dai B, Rajaraman S, Wood T.
Gastrin gene expression is required for the proliferation and
tumorigenicity of human colon cancer cells. Cancer Res
1996;56:4111±15.
10 Darro F, Camby I, Kruczynski A, Pasteels JL, Martinez J, Kiss R.
Characterisation of the influence of anti-gastrin, anti-epidermal
growth factor, anti-oestradiol, and anti-luteinising hormone
releasing hormone antibodies on the proliferation of 27 cell lines
from the gastrointestinal tract. Gut 1995;36:220±30.
11 Baldwin GS, Shulkes A. Gastrin, gastrin receptors and
colorectal carcinoma. Gut 1998;42:581±4.
12 Nakata H, Wang SL, Chung C, Westwick JK, Tillotson LG.
Oncogenic ras induces gastrin gene expression in colon
cancer. Gastroenterology 1998;115:1144±53.
13 Orbuch M, Venzon DJ, Lubensky IA, Weber HC, Gibril F,
Jensen RT. Prolonged hypergastrinemia does not increase the
frequency of colonic neoplasia in patients with Zollinger±
Ellison syndrome. Dig Dis Sci 1996;41:604±13.
14 Hsing AW, Hansoson LE, McLaughlin JK, Nyren O, Blot WJ,
Ekbom A et al. Pernicious anemia and subsequent cancer.
Cancer 1993;71:745±50.
15 Wang TC, Dockray GJ. Lessons form genetically engineered
animal models. I. Physiological studies with gastrin in
transgenic mice. Am J Physiol 1999;277:G6±G11.
16 Koh TJ, Dockray GJ, Varro A, Cahill RJ, Dangler CA, Fox JG
et al. Gastrin deficiency results in altered gastric differentiation
and decreased colonic proliferation in mice. Gastroenterology
1997;113:1015±25.
17 Shenker A. G protein-coupled receptor structure and
function: the impact of disease-causing mutations. Baillieres
Clin Endocrinol Metab 1995;9:427±51.
18 Schaffer K, McBride EW, Beinborn M, Kopin AS.
Interspecies polymorphisms confer constitutive activity to the
Mastomys cholecystokinin-B/gastrin receptor. J Biol Chem
1998;273:28779±84.
19 McWilliams DF, Watson SA, Crosbee DM, Michaeli D, Seth
R. Coexpression of gastrin and gastrin receptors (CCK-B and
DCCK-B) in gastrointestinal tumour cell lines. Gut
1998;42:795±8.
20 Schmitz F, Pratt DS, Wu MJ, Kolakowski LF Jr, Beinborn M,
Kopin AS. Identification of cholecystokinin-B/gastrin receptor
domains that confer high gastrin affinity: utilization of a novel
Xenopus laevis cholecystokinin receptor. Mol Pharmacol
1996;50:436±41.
21 Miyake A. A truncated isoform of human CCK-B/gastrin
receptor generated by alternative usage of a novel exon.
Biochem Biophys Res Comm 1995;208:230±7.
22 Hellmich MR, Rui XL, Hellmich HL, Fleming RY, Evers BM,
Townsend CM Jr. Human colorectal cancers express a
constitutively active cholecystokinin-B/gastrin receptor that
stimulates growth. J Biol Chem 2000;275:32122±8.
23 Singh P, Xu Z, Dai B, Rajaraman S, Rubin N, Dhruva B.
Incomplete processing of progastrin expressed by human
colon cancer cells: role of noncarboxyamidated gastrins. Am J
Physiol 1994;266:G459±68.
820 F. Schmitz et al.
24 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 2000;278:G390±9.
25 Watson SA, Michaeli D, Morris TM, Clarke P, Varro A,
Griffin N et al. Antibodies raised by gastrimmune inhibit the
spontaneous metastasis of a human colorectal tumour,
AP5LV. Eur J Cancer 2000;35:1286±91.
26 Iwata N, Murayama T, Matsumori Y, Ito M, Nagata A,
Taniguchi T et al. Autocrine loop through cholecystokinin-B/
gastrin receptors involved in growth of human leukemia cells.
Blood 1996;88:2683±9.
Q 2001 Blackwell Science Ltd, European Journal of Clinical Investigation, 31, 812±820
27 Hanahan D, Weinberg RA. The hallmarks of cancer. Cell
2000;100:57±70.
28 Watson SA, Michaeli D, Smith A, McWilliams DF.
Quantification of gastrin and CCKB/gastrin receptor
expression during the adenoma carcinoma sequence in the
colon. Gastroenterology 2000;118:A1607.
29 Seva C, Dickinson CJ, Yamada T. Growth-promoting effects
of glycine-extended progastrin. Science 1994;265:410±12.
30 Todisco A, Takeuchi Y, Seva C, Dickinson CJ, Yamada T.
Gastrin and glycine-extended progastrin processing
intermediates induce different programs of early gene
activation. J Biol Chem 1995;270:28337±41.