| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Tumor Biology |
Academic Unit of Cancer Studies, Queen’s Medical Centre, University of Nottingham, Nottingham NG7 2UH, United Kingdom [S. A. W., P. A. C., T. M. M.], and Academic Medical Unit, Royal Free Hospital, London, NW3 2PF United Kingdom [M. E. C.]
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Recent studies have shown that by using an antiserum directed against the NH2-terminal extracellular loop of the receptor, denoted GRP1, immunoreactive bands can be detected on both plasma membranes and at intracellular membrane locations of gastrointestinal tumor cell lines by Western blotting (10) . The plasma membrane-associated bands were detected at Mr 74,000, the known molecular weight of the CCKB/gastrin receptor (11) . The intracellular forms were of lower molecular weight and may have represented deglycosylated or degraded forms of the receptor (10) .
A recent study has used the CCKB/gastrin receptor as a trafficking system for the delivery of cytotoxic drugs (12) . By the use of heptagastrin linked to ellipticine via a succinoyl-substituted pentapeptide, a 1000-fold increase in cytotoxicity was found with the receptor-linked drug on CCKB/gastrin receptor-positive NIH 3T3 cells compared with receptor-negative cells. In addition, the drug receptor conjugate inhibited the growth of CCKB/gastrin receptor-positive cells in vivo with a delay in growth and a reduction in size (12) .
Additional studies have followed the intracellular trafficking of the CCKB/gastrin receptor linked to the GRP1 antiserum described above (13) . In these studies, using the rat pancreatic tumor AR42J, antibody-receptor complexes were localized to the nucleus, as confirmed by both light and electron microscopy, as well as Western blotting on isolated nuclear preparations (13) .
The aim of the present study was to evaluate the therapeutic effect of passive immunization using the GRP1 antiserum in a human colorectal liver metastasis model. There is currently little to offer patients with advanced colorectal cancer. Blockading the CCKB/gastrin receptor would be an amenable therapeutic approach because serum gastrin levels have previously been found to be elevated in such patients, raising the possibility that up-regulation of CCKB/gastrin receptor expression may occur (14) .
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The cell line was maintained at 37°C under humidified conditions in 5% CO2 and routinely passaged in RPMI 1640 tissue culture medium (Life Technologies, Inc., Irvine, United Kingdom) containing 10% heat-inactivated FCS (Sigma, Poole, United Kingdom).
Growth of C170HM2 Liver-invasive Tumors in Vivo.
C170HM2 cells were harvested by using 0.025% EDTA and resuspended at
1 x 106
cells/ml-1 in sterile
saline (0.9%; pH 7.3). A 1-ml volume was then injected into the
peritoneal cavity of male nude mice [6 weeks of age; bred within the
Cancer Studies Unit (University of Nottingham, Nottingham, United
Kingdom)]. This resulted in established liver tumors after
approximately 10 days (15)
. Mice were killed at day
40–50, before the clinical condition of the mice was compromised.
United Kingdom Coordinating Committee for Cancer Research guidelines
were adhered to throughout all animal experimentation.
GRP1 Antiserum.
GRP1 antiserum was directed against the
NH2-terminal extracellular epitope of the human
CCKB/gastrin receptor. The peptide was based on the known amino acid
sequence of the human CCKB/gastrin receptor and comprised the epitope
KLNRSVQGTGPGPGASL, which was linked to diphtheria toxoid
(10)
. Rabbits were immunized with the peptide conjugated
to diphtheria toxoid to derive a polyclonal antiserum
(10)
.
Biodistribution of Radiolabeled GRP1 Antiserum.
Polyclonal rabbit anti-GRP1 antiserum and normal rabbit serum were
prepared at a protein concentration of 1
mg/ml-1. These sera
were radiolabeled with Na125I (Sigma,
Gillingham, United Kingdom) by the Iodo-Gen method. Briefly, Iodo-Gen
(Pierce, Chester, United Kingdom) was dissolved in methylene chloride
at a concentration of 400
µg/ml-1 and dispensed
into tubes in 0.3-ml volumes. The contents of the tubes were then
evaporated to dryness under a stream of nitrogen and stored desiccated
at 4°C. A 0.5-ml volume of the antiserum preparation was placed in
the tubes along with 15 MBq of Na125I
(Amersham, Bucks, United Kingdom). The tube contents were mixed and
incubated at room temperature for 10 min. A G25 Sephadex column
(Pharmacia, Uppsala, Sweden) was prepared by washing with 10 ml of
0.9% saline. The radioiodine-antibody mixtures were added and eluted
by adding 0.5-ml fractions of 0.9% saline. The associated
radioactivity of each fraction was measured, and peak fractions were
pooled. The antibody presence was confirmed by measuring the absorbance
at 280 nm.
125I-labeled rabbit anti-GRP1 antiserum and normal rabbit serum were administered to nude mice bearing established C170HM2 liver tumors by tail vein injection of 200 µl of serum containing 2 x 106 cpm radioactivity. Mice were treated with a single injection, and the biodistribution of radiolabel into the liver and liver tumors was compared.
Mice were killed at 6 h and at 1, 3, and 5 days, and the number of cpm was monitored in liver tumor, normal liver tissue, and blood. Results were calculated as cpm/gram-1 tissue and expressed as the ratio of liver tumor:normal liver.
Therapeutic Effect of GRP1 Antiserum.
Nude mice were injected with increasing concentrations of C170HM2 cells
(8 x 105 to 1.5 x 106 cells/mouse; initial i.p. injection)
to establish different liver tumor loads. The number of mice initiated
per group is shown in Table 1
.
|
The clinical condition of the mice was carefully assessed during the study by measuring animal weights and by clinical observation. Mice were palpated toward the end of the study to monitor liver tumor bulk. At sacrifice, the liver tumors were dissected free of normal liver tissue and weighed. Half of each tumor was then frozen for Western blotting analysis, and the other half was formalin fixed and paraffin embedded for histological analysis.
Western Blotting.
Extranuclear membrane extracts were prepared from C170HM2 xenograft
tissue, and Western blotting was performed as described previously
(10)
. Immunoreactive bands were localized using GRP1
antiserum (1:50 dilution of a 2 mg/ml-1
protein stock). Specificity was confirmed by preabsorption (24 h,
4°C) of the GRP1 antiserum with 20 mg/ml-1
GRP1 protein and by the use of normal rabbit serum as a negative
control. Detection of immunoreactive bands was as described previously
(10)
.
Image Analysis of Necrosis in C170HM2 Liver Tumors.
Two experienced observers who were blind to the treatment groups
counted the level of necrosis by using Leica Qwin image
processing and an analysis system run on a Leica Q5001W personal
computer. A program was developed enabling necrotic areas to be
expressed as a percentage of total tumor area using H&E-stained
sections.
The interobserver variation and intraobserver variation was <10.0%.
Apoptosis.
The presence of apoptotic cells in the xenografts was detected by using
the Apoptag Plus in situ apoptosis detection kit (Oncor,
Gaithersburg, MD). This method detects apoptotic cells by direct
immunoperoxidase detection of digoxigenin-labeled genomic DNA (DNA
fragmentation results in free 3' OH groups at the end of DNA fragments
that are tailed with digoxigenin-labeled nucleotides) in 4-µm
sections prepared from paraffin-embedded tissue. Rat mammary gland
epithelium was used as a positive control.
Statistics.
All statistics were performed using the Mintab package for a personal
computer.
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. The results are calculated as cpm/gram of tissue and expressed as a
liver tumor:normal liver ratio for both rabbit GRP1 antiserum
and normal rabbit serum. Each point is representative of two to six
animals with the SDs shown. At the earliest time points (6 h), there
was lower radioactivity in the liver tumor when compared with the
normal liver after treatment with both GRP1 antiserum and normal rabbit
serum. After 24 h, there was a 1.5-fold increase in the level of
GRP1 antiserum localized within the liver tumor in comparison to the
normal liver tissue. Localization of normal rabbit serum in the liver
tumor remained less than that present in the normal liver (Fig. 1
;
significance values shown on figure legends). Enhanced liver tumor
localization of GRP1 antiserum was maintained for 5 days.
|
(initial inoculum, 8.0 x 105
cells/ml-1), B
(initial inoculum, 1.0 x 106
cells/ml-1), and
C (initial inoculum, 1.5 x 106
cells/ml-1). All lesions
were macroscopically visible because the tumors invade through the
external liver capsule; the smallest lesion measured was 4
mm2 (weight, 0.05 gram).
|
. The distribution was
significantly different when comparing the control group with the test
group (P < 0.005, 
2
test).
With a cell inoculum of 1.0 x 106
cells/mouse, there were non-tumor-bearing animals in both groups, and
the number of non-tumor-bearing animals was again in the GRP1
antiserum-treated group (54% in the GRP1 antiserum-treated group
compared with 12.5% in the normal rabbit serum-treated group, Fig. 2B
). In the control group, 75% of mice had >1.0
liver tumor, whereas in the test group, none of the animals had >1.0
liver tumor, and the distribution was significantly different between
the two groups as determined by a 2 test
(P < 0.0001). When the cell inoculum was
increased to 1.5 x 106
cells/mouse, all of the control mice had liver tumors, resulting in a
100% take rate, as compared with a 72% take rate in the test group
(Fig. 2C)
. In the control group, 70% of mice had >1.0
tumor/liver, and 9% of mice in the GRP1 antiserum-treated group had
>1.0 tumor/liver. This distribution was significantly different in the
control versus test group (P < 0.005, 2 test).
Tumor weight at therapy termination was measured in normal rabbit serum
control- and GRP1 antiserum-treated groups for all tumor cell inocula.
The results are shown in Fig. 3
. With increasing cell inocula, the final tumor weight within the liver
rose accordingly in the control groups from a mean weight of 0.37 gram
to 1.89 grams. In all experiments, treatment with GRP1 antiserum
resulted in a significant reduction in tumor burden: (a) a
73% reduction was seen with a cell inocula of 8.0 x 105 cells/mouse (P = 0.0155, Mann-Whitney nonparametric U test); (b) a
64% reduction was seen with a cell inoculum of 1.0 x 106 cells/mouse (P = 0.003, Mann-Whitney nonparametric U test); and
(c) a 60% reduction was seen with a cell inoculum of
1.5 x 106 cells/mouse
(P = 0.0068, Mann-Whitney nonparametric
U test).
|
shows animal weights over time for mice treated with a cell inoculum of
1.5 x 106 cells/mouse. The
treatment had no effect on the normal weight gain or clinical condition
of the mice.
|
, shows representative areas from the
normal rabbit serum- and GRP1 antiserum-treated liver tumors,
respectively.
|
c) as a positive control. No apoptosis was evident in the tumors treated
with normal rabbit serum (Fig. 6a)
, whereas apoptotic cells
were evident in viable leading edge tumor tissue in the GRP1
antiserum-treated tumors (Fig. 6b)
.
|
. Fig. 7a
shows the immunoreactive bands of C170HM2 cells
grown in vitro (10)
. Fig. 7b
details
the GRP1 immunoreactive bands from C170HM2 xenografts treated with
either normal rabbit serum or GRP1 antiserum. In normal rabbit
serum-treated xenografts, there is an immunoreactive band at
Mr 74,000 and a fainter band at around
Mr 45,000 that is known to be an
intracellular form of a receptor previously exposed on the plasma
membrane (10)
.
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
C170HM2 cells express the CCKB/gastrin receptor at both the mRNA level (17) and the protein level by Western blotting (10) . They also express gastrin mRNA and secrete gastrin-like peptides (17 , 10) . Thus, C170HM2 is a potentially gastrin-sensitive colonic tumor. Antiserum directed against the CCKB/gastrin receptor had a strong therapeutic effect on liver-invasive growth at the level of tumor take and on tumor proliferation. Thus, such an antiserum may blockade the effect of serum and tumor-associated gastrin peptides.
GRP1 antibody-receptor complexes have previously been shown by electron microscopy to be internalized within the cell and to localize to the nucleus (13) , which has been confirmed by confocal microscopy (21) . In the present study, GRP1 antiserum has been shown to localize within C170HM2 tumors to a greater extent than in normal liver tissue. This would indicate tumor uptake in a specific manner and may represent a second mechanism by which GRP1 antiserum may be mediating a tumor-inhibitory effect. Thus, by internalizing within the cell and following transport to the nucleus, the antibody may have nonspecific inhibitory effects. Indeed, apoptosis was shown in the present study in the viable tissue within GRP1 antiserum-treated tumors. This indicates that in addition to the cytostatic effects mediated by the removal of a potent growth factor, the cellular machinery may have been induced to undergo programmed cell death in response to nuclear disturbance caused by the presence of an antibody. The level of cell death could potentially be further increased by linking the antibody to a cytotoxic agent, as has been described previously (12) .
Necrosis was modestly increased in the GRP1 antiserum-treated tumors compared with tumors from control serum-treated mice. This may reflect the fact that all tumors in the GRP1 antiserum-treated groups were smaller than the corresponding tumors in the normal rabbit serum-treated groups. The latter develop necrosis in response to poor vascularization over a certain size, whereas the small tumors in the test group potentially lack this nonspecific onset of necrosis.
Western blotting performed on xenograft tissue revealed that in the normal rabbit serum control-treated xenografts, the predominant receptor expressed was the Mr 74,000 isoform, which is the reported molecular weight of the classical CCKB/gastrin receptor (11) . The lower molecular weight receptor (Mr 45,000) was also evident. This is an intracellular form of the CCKB/gastrin receptor expressed previously on the cell membrane (10) . In xenografts treated with GRP1 antiserum, the Mr 45,000 form of the receptor appeared to be expressed at a greater level. This would indicate that plasma membrane-associated receptors were internalizing to a greater extent, which may have been due to antibody binding and antibody-receptor complex internalization. Therefore, it would appear that in addition to direct competition for the binding of plasma gastrin, down-regulation of external receptors might also have occurred. The intermediate molecular weight isoform detected by Western blotting has not been observed previously. The significance of this observation is not known, but it may be artifactual because the receptor antibody complex may mask degradation sites responsible for complete cleavage to the Mr 45,000 form.
In conclusion, antiserum directed against the NH2-terminal portion of the human CCKB/gastrin receptor has localized within and caused growth inhibition and apoptosis of a human liver-invasive colorectal tumor in vivo. Because there is little to offer patients with advanced disease in terms of therapeutic agents, treatment with antibodies raised against the GRP1 epitope would be a potential therapeutic modality either alone or linked to cytotoxic/radiolabeled agents in a targeted approach.
|
FOOTNOTES |
|---|
1 Supported by Grant R8GAWA004 (Aphton
Corporation). 
2 To whom requests for reprints should be
addressed, at Academic Unit of Cancer Studies, D Floor, West Block,
Queen’s Medical Centre, University of Nottingham, Nottingham NG7 2UH,
United Kingdom. Phone: 44-0-115-9709248; Fax: 44-0-115-9709902; E-mail: sue.watson{at}nottingham.ac.uk
3 The abbreviations used are: CCK,
cholecystokinin; GRP, gastrin receptor protein.
4 D. McWilliams, personal communication.
Received 12/28/99. Accepted 8/11/00.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
CCK-B in gastrointestinal tumour cell lines. Gut, 42: 795-798, 1998.This article has been cited by other articles:
|
|
 |
|
  M. Dufresne, C. Seva, and D. Fourmy Cholecystokinin and gastrin receptors. Physiol Rev, July 1, 2006; 86(3): 805 - 847. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y.-S. Guo, J.-Z. Cheng, G.-F. Jin, J. S. Gutkind, M. R. Hellmich, and C. M. Townsend Jr. Gastrin Stimulates Cyclooxygenase-2 Expression in Intestinal Epithelial Cells through Multiple Signaling Pathways. EVIDENCE FOR INVOLVEMENT OF ERK5 KINASE AND TRANSACTIVATION OF THE EPIDERMAL GROWTH FACTOR RECEPTOR J. Biol. Chem., December 6, 2002; 277(50): 48755 - 48763. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Stubbs, K. Khan, S. A. Watson, K. Savage, A. P. Dhillon, and M. E. Caplin Endocytosis of Anti-CCK-B/Gastrin Receptor Antibody and Effect on Hepatoma Cell Lines J. Histochem. Cytochem., September 1, 2002; 50(9): 1213 - 1217. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
, rabbit anti-GRP-1 antiserum. The rabbit anti-GRP1 antiserum
and normal rabbit serum were radiolabelled with
Na125I by the Iodo-Gen method. Radiolabeled sera
were then administered to nude mice bearing established C170HM2 liver
tumors by tail vein injection delivering 2 x 106 cpm. Mice were treated with a single injection, and the
biodistribution of radiolabel into the liver and liver tumor was
compared.
) normal rabbit serum-treated and
(hatched bars) rabbit GRP-1 antiserum-treated mice
injected with the following cell innocula: A,
8.0 x 105 cells/mouse; B,
1.0 x 106 cells/mouse; and
C, 1.5 x 106 cells/mouse.
Nude mice were injected with increasing C170HM2 cell numbers to
establish different tumor levels within the liver. Antisera (rabbit
GRP1 antiserum or normal rabbit serum) were administered daily (0.5 ml
volume) by tail vein injection from day 0. At sacrifice, the liver
tumors were dissected free of normal liver tissue and counted.
) rabbit GRP-1 antiserum after injection of nude
mice with the C170HM2 cell line is shown. The clinical condition of the
mice was carefully assessed during the study by measurement of animal
weights on a weekly basis.
