Liver metastasis is one of the critical prognostic factors of colorectal cancer and 25% of patients with colorectal cancer have synchronous or metachronous liver metastases. An understanding of the mechanisms causing such liver metastases is therefore essential for ensuring that they are appropriately treated and followed-up.
The product of the c-met proto-oncogene encodes transmembrane tyrosine kinase (1 ) and is the receptor for hepatocyte growth factor (HGF), which regulates both cell motility and cell growth (2 ). c-met is expressed in various normal tissues (3 ) and overexpressed in tumors, including thyroid (4 ), stomach (5 ), pancreatic (6 ,7 ) and colon cancer (8 ,9 ). HGF is detected around the tumors, including liver metastases (10 ). These facts suggest that c-met and HGF play an important role in invasion and metastasis.
Overexpression of c-met in primary colorectal cancer and liver metastases has been reported using Northern blot (9 ). However, it is difficult to quantify mRNA in small clinical samples using Northern blot. We have therefore quantified mRNA expression of c-met in colorectal cancer and liver metastatic tissues by quantitative reverse transcriptase polymerase chain reaction (RT-PCR) using fluorescence-labeled primers and autosequencer and examined the relationship between c-met expression and liver metastasis.
Tumor, adjacent normal mucosa (10 cm away from the tumor), liver metastatic tumor and normal liver tissue from patients with colorectal cancer at the National Cancer Center Hospital were obtained immediately after surgery and frozen and stored in liquid nitrogen. To avoid liver tissue contamination, the metastatic tissues were carefully taken from the periphery of the metastases adjacent to the central necrosis.
Total RNA was extracted from the frozen tissues by the procedure described by Chomcznski and Sacchi (11 ). Randomly primed cDNA was synthesized from 1 µg of total RNA by reverse transcriptase (GIBCO BRL, Gaithersburg, MD, USA) in a total volume of 20 µl and followed by PCR amplification. c-met and [beta]-actin cDNA were amplified using the following primers: c-met (12 ), 5'TGCGAAGTGAAGGGTCTCC3', 5'GGTCACTTCACCTACCGAAA3' (reverse primer); [beta]-actin (13 ), 5'CTGTCTGGCGGCACCACCAT3' (forward primer), 5'GCAACTAAGTCA- TAGTCCGC3' (reverse primer). Forward primers were synthesized and labeled with Cy5 amidyte reagent, a fluorescent dye (Pharmacia, Uppsala, Sweden), using Oligo 1000 DNA synthesizer (Beckman, Fullenton, CA, USA). The reaction mixture contained 5 µl of cDNA as a template, 25 pmol of forward and reverse primers of c-met and [beta]-actin, 20 nmol each dNTP, 10 mM Tris/HCl pH 8.3, 50 mM potassium chloride, 1.5 mM magnesium chloride, 0.001% (w/v) gelatin and 2.5 units of Taq DNA polymerase (Perkin Elmer Cetus, Norwalk, CT, USA) to a total volume of 100 µl. The PCR amplification was performed for 30 cycles in the following conditions: denatured at 94°C for 30 s, annealed at 54°C for 30 s, extended at 72°C for 30 s.
The PCR products containing c-met or [beta]-actin were subcloned into the pUC18 plasmid vector (Pharmacia) and JM109 competent cells (Toyobo, Osaka, Japan) were transformed by this plasmid according to the manufacturer's instructions. The transformed cells were cultured in LB agar (Wako Pure Chemical Co., Osaka, Japan). Plasmids were purified using a Quiagen plasmid kit (Quiagen, Chatsworth, CA, USA) and were used as a control. It was estimated that 1 µg of the plasmids contained cDNA derived from 6.8×1011 copies of c-met or [beta]-actin mRNA.
Two microliters of the PCR products were mixed with 5 µl of the loading buffer containing 90% deionized formamide, 20 mM EDTA and 0.05% bromophenol blue; 1.5 µl of the aliquot was electrophoresed on 6% polyacrylamide gel containing 6M urea at 35W and 55°C for 150 min using ALFred DNA sequencer (Pharmacia). Data were analyzed using the software package Fragment ManagerTM (Pharmacia) equipped with short gel plate (14 ).
Estimated mRNA expression was statistically compared by unpaired t-test.
Serially diluted c-met and [beta]-actin control plasmids were analyzed and 105-108 copies of c-met mRNA and 106-109 copies of [beta]-actin mRNA could be quantified by this PCR (Fig. 1 ). Two peaks of [beta]-actin control plasmids were beyond the scale in this condition. However, these peaks were also usable for quantification because plots of the area under the peaks were on a linear curve.
c-met and [beta]-actin mRNA were examined in tissue from 27 cases of primary colorectal cancer and ten cases of liver metastases (Fig. 2 ). After estimating the copy number of c-met and [beta]-actin mRNA using the control plasmids as a standard, the copy number of c-met mRNA was standardized by the following formula: estimated copy number of c-met mRNA * 108/estimated copy number of [beta]-actin mRNA. This gives numbers of c-met mRNA to 108 copies of [beta]-actin mRNA. The patients' backgrounds and standardized c-met expression are summarized in Table 1 . Mean copy numbers of c-met mRNA in cancer tissues and normal mucosa were 105.5 and 104.5 respectively (Fig. 3 ). The copy number of c-met mRNA in cancer was significantly higher than that of normal mucosa (P < 0.0001). The expression of c-met mRNA in cancer tissue could be compared with that of corresponding normal tissue in 22 of the 27 samples. In 20 of the 22 (91%), c-met was overexpressed in the cancer tissue when compared with its expression in corresponding normal tissue. No correlation was found between c-met expression and Dukes' stage.
Table 1. Clinicopathological backgrounds and c-met expression
Our results show that c-met is expressed more strongly in primary colorectal cancer than in normal mucosa and that c-met expression in liver metastases is higher than that in the corresponding primary cancer. A melanoma cell line that produces liver-specific metastases in a murine metastasis model has a higher level of c-met than other melanoma cell lines and c-met expression correlates with their liver colonization potential (15 ). Increased c-met expression in transformed cells enhances their invasiveness and metastatic potential (16 ). These facts suggest that the overexpression of c-met in primary cancer and metastatic tissue gives colorectal cancer cells an advantage to colonize the liver.
c-met is overexpressed in thyroid (4 ), stomach (5 ), pancreatic (6 ,7 ) and colon cancers (8 ,9 ). Di Renzo et al. showed c-met overexpression in >50% of colorectal tumors including adenomas and a lack of correlation between the c-met overexpresion and tumor stage (9 ). Our results indicate that c-met is overexpressed in >90% of cancers and that there is no relationship between c-met expression and tumor stage.
The increased expression of c-met is explained by a transcriptional level change and gene amplification of c-met. The transcription of c-met is upregulated by cytokines such as interleukin (IL)-1, IL-6, tumor necrosis factor-[alpha] (17 ) and HGF (18 ). Since these cytokines are induced around the tumors, they are thought to increase c-met expression in the tumor. Loss of wild-type p53 also enhances the opportunity for inappropriate c-met expression (19 ). Point mutation of p53 and loss of heterozygosity (LOH) of chromosome 17p are frequently detected in colorectal cancer and the rates of p53 mutation and LOH of 17p are higher in liver metastatic tumors than primary tumors (20 ). Gene amplification of c-met has been detected in gastric and colorectal cancer (9 ,21 ). In colorectal cancer, gene amplification has been detected in 10% of primary cancers and in eight of nine (88%) liver metastases (9 ).
The ligand of c-met, HGF, increases the motility of several normal cell types and cancer cells including colon cancer (22 ). Immunohistochemical analysis has shown that HGF is undetectable in normal liver but is present in liver around tumor metastases (10 ). The serum HGF concentration is increased in certain clinical conditions such as liver failure, cirrhosis (23 ), transcatheter arterial embolization, partial hepatectomy (24 ) and abdominal surgery (25 ). These facts show that HGF production results from cancer cell invasion into the liver, inflammation or surgery and they suggest that HGF activates overexpressed c-met in colorectal cancer and stimulates invasion and metastasis. Transformed NIH 3T3 cells that secrete HGF and express c-met acquire tumorigenecity in nude mice and develop metastatic activity (16 ). This suggests that an interaction between HGF and c-met plays an important role in tumorigenesis. In this case, HGF and c-met work as an autocrine system. However, HGF is not detected in normal liver and the liver metastases of colorectal cancer. As HGF is detected around liver metastases, it serves a paracrine-like role in the development of colorectal cancer liver metastases. HGF is also a potent cell growth stimulator of hepatocytes (26 ), pancreatic cells (27 ) and endothelial cells (28 ,29 ). It therefore acts as an angiogenic factor and induces the development of blood vessels around the metastatic tumor. This angiogenic function has also been thought to stimulate the development of liver metastases. On the other hand, HGF inhibits the cell growth of various tumor cell lines including hepatoma (30 ,31 ) and colon cancer cells (22 ). This indicates that c-met overexpression is undesirable for cancer growth. However, the growth inhibition rate of HGF varies among colon cancer cells and the inhibition is not marked (22 ). These differences in effect of HGF on colon cancer cells may correlate with the growth rate of colorectal cancer liver metastases.
In conclusion, c-met is overexpressed in primary colorectal cancer and liver metastases and c-met mRNA is higher in liver metastases than in primary cancer. These facts suggest that c-met plays a role in the development of colorectal cancer liver metastases.
This work was supported in part by a Grant-in-Aid for the Second Term Comprehensive 10-year Strategy for Cancer Control from the Ministry of Health and Welfare of Japan and by a Grant-in-Aid from Foundation for Promotion of Cancer Research. We thank N. Fukayama for technical assistance.
Japanese Journal of Clinical Oncology
Pages
Introduction
Materials And Methods
Tissues
RT-PCR
Control Plasmids
Electrophoresis
Statistical Analysis
Results
Quantification of c-met and [beta]-actin Control Plasmids
c-met Expression in Colorectal Cancer and Liver Metastasis Tissue
Discussion
Acknowledgments
References
Patient
No.Tumor*
locationHistological
gradeDukes'
stageStandardized copy number
of c-met m-RNA[dagger]
Primary
tumorNormal
mucosaLiver
metsNormal
liver
1
S
Moderate
B
4.8
4.1
2
R
Well
B
5.8
5.7
3
S
Moderate
C
5.9
4.2
4
R
Well
A
6.9
4.2
5
S
Moderate
C
2.9
3.5
6
R
Moderate
B
5.4
3.2
7
A
Well
B
5.7
4.3
8
R
Well
B
6.9
5.8
9
C
Well
B
6.6
5.1
10
A
Poor
C
6.7
4.3
11
S
Moderate
B
6.0
4.3
12
R
Moderate
A
4.5
4.3
13
R
Moderate
C
4.7
5.8
14
S
Moderate
C
6.8
5.3
15
S
Well
B
4.7
3.1
16
R
Well
B
5.1
-[Dagger]
17
S
Well
C
5.8
-
18
R
Moderate
C
4.6
-
19
A
Well
B
5.8
5.5
20
A
Moderate
C
4.6
5.5
21
D
Moderate
D
5.9
4.6
22
A
Well
C
5.7
4.7
23
S
Moderate
D
6.1
-
6.2
-
24
S
Well
D
5.7
4.9
6.4
6.6
25
R
Well
D
7.1
5.5
5.9
6.7
26
R
Moderate
D
3.7
-
5.0
-
27
S
Well
D
4.5
3.7
-
6.3
28§
R
Moderate
A
5.4
6.8
29
T
Moderate
D
6.5
6.7
30
D
Mucinous
C
6.8
5.8
31
R
Moderate
C
5.3
6.6
32
S
Moderate
C
6.1
4.4
33
C
Well
C
7.0
7.0
References
This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: www-admin{at}oup.co.uk
Last modification: 19 May 1998
Copyright© Japanese Journal of Clinical Oncology, 1997.
This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: www-admin{at}oup.co.uk
Last modification: 19 May 1998
Copyright© Japanese Journal of Clinical Oncology, 1997.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  H. Takeuchi, A. Bilchik, S. Saha, R. Turner, D. Wiese, M. Tanaka, C. Kuo, H.-J. Wang, and D. S. B. Hoon c-MET Expression Level in Primary Colon Cancer: A Predictor of Tumor Invasion and Lymph Node Metastases Clin. Cancer Res., April 1, 2003; 9(4): 1480 - 1488. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||