- Email: [email protected]
Expression of 32-kDa Laminin-Binding Protein mRNA in Colon Cancer Tissues
JOURNAL OF SURGICAL RESEARCH ARTICLE NO.
61, 120–126 (1996)
0091
Expression of 32-kDa Laminin-Binding Protein mRNA in Colon Cancer Tissues DONG PO PEI, M.D., YOUNG HAN, PH.D., DEEPAK NARAYAN, M.D., DANIEL HERZ, M.D., AND THANJAVUR S. RAVIKUMAR, M.D.1 Surgical Oncology, Department of Surgery, The Cancer Institute of New Jersey, CABM Room 223, 679 Hoes Lane, Piscataway, New Jersey 08854 Submitted for publication August 22, 1994
Colorectal cancer initiation and progression are associated with stepwise genetic alterations. We and others have shown that a gene encoding for a 32-kDa putative laminin-binding protein (LBP-32) is overexpressed during colorectal cancer progression by Northern blots analysis. Northern blots cannot indicate the heterogeneity of expression from cell to cell and the distribution pattern of gene expression within a given tumor. In order to overcome these problems, we examined the LBP-32 mRNA expression in colorectal carcinomas by in situ hybridization. LBP-32 mRNA expression in 30 cases of primary and metastatic colorectal cancers and their respective adjacent normal tissues were detected by in situ hybridization using 35S-UTP radiolabeled antisense riboprobes. The results showed that LBP-32 mRNA was expressed at a low level in the normal colonic mucosa adjacent to the tumor compared with colon cancer tissues. Its expression in poorly differentiated colorectal cancer was much higher than that in well- and moderately differentiated colorectal cancer. More importantly, the LBP32 mRNA was expressed more highly in the invasive lesions of the cancer and liver metastases compared with the cancer lesions in situ. Our results imply that in situ hybridization is a powerful tool in evaluating the changes in gene expression in the cancer cells and LBP-32 mRNA expression is related to progression, invasion, and metastasis of colorectal cancer. q 1996 Academic Press, Inc.
INTRODUCTION
Colorectal carcinoma is the second leading cause of cancer death in the United States. About 152,000 new cases are expected in 1993 and more than 57,000 of them will die of this disease [1]. Invasion and metastasis are the greatest obstacles to the successful treatment and the major reasons for cancer patient mortal1 To whom correspondence should be addressed at The Cancer Institute of New Jersey, CABM Room 223, 679 Hoes Lane, Piscataway, New Jersey 08854. Fax: (908) 2355475.
0022-4804/96 $18.00 Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
AID
JSR 4574
/
m4784$$$28
ity. Therefore, the identification of new genetic and biochemical markers that would predict the invasive and metastatic potential of a cancer at an early stage will be important from the diagnostic and therapeutic standpoint. The process of tumor invasion and metastasis is charactered by a series of interactions between tumor cells and basement membranes (BM). Tumor cells have to penetrate this natural tissue barrier during the invasion and metastasis [2]. Laminin (Mr Å 800,000) is a prominent component of BM and promotes many biological activities including cell adhesion, cell spreading, morphogenesis, differentiation, proliferation, mitogenesis, neutritic outgrowth, and cell migration [3, 4]. Many of these activities are mediated through the cell surface laminin-binding protein (LBP) or laminin receptors (LR) [5, 6]. Human LR complementary DNA (cDNA) clones were originally identified from the cDNA expression library of human endothelial cells by means of screening with a monoclonal antibody against the 67-kDa LR [7]. Subsequently, a full length human cDNA clone for human LBP was identified by the differential screening of a cDNA library from a poorly differentiated human colon carcinoma cell line [8]. Its sequence was homologous to that of the 67-kDa LR cDNA. The coding capacity of the full length cDNA clone, however, is for nascent protein with a calculated molecular mass of 32 kDa (LBP-32) [8]. It was suggested that the LBP-32 and 67-kDa LR are antigenically related and a precursor product relationship exists between them [9–12]. The function of LBP-32 is not yet defined. Attachment of laminin to tumor cells enhances several properties of the malignant phenotype and has been associated with increased tumorigenic and metastatic potential of tumor cells in vivo and in vitro [13–18]. LBP-32 and its mRNA were overexpressed in colorectal carcinoma and its expression has significant correlation with Dukes’ classification [8, 12, 15, 19–20]. It was also found that antisense RNA of LBP-32 inhibited tumor cell attachment and invasion in vitro [21]. These findings suggest that LBP-32 has a role in the colorectal cancer progression, invasion, and metastasis.
120
01-29-96 09:16:13
srga
AP: Surg Res
PEI ET AL.: LAMININ-BINDING PROTEIN mRNA IN COLON CANCER
121
dried, and resuspended in diethyl pyrocarbonate (DEPC)-treated distilled water containing RNasin (Boehringer Mannheim, Indianapolis, IN) and 10 mM DTT. Incorporation rate was over 70%. The probes were used on the day of synthesis.
In Situ Hybridization GEM-4
FIG. 1. The schematic diagram of the plasmid P used for in vitro transcription of antisense and sense probes of LBP-32 mRNA.
In situ hybridization was performed as described by Zeller et al. [25]. Pretreatment of slides included the dewaxing, rehydration, acid
In previous studies, an overexpression of LBP-32 mRNA in human colorectal cancer compared with adjacent normal colonic mucosae was demonstrated using Northern blots analysis. Northern blots analysis is relatively crude and does not provide insight into the changes of gene expression in a subpopulation within a given tumor. The technique of in situ hybridization allows us to identify and localize the gene expression at the single cell level [22, 23]. In this study, we used in situ hybridization to assess the heterogeneity of LBP-32 mRNA expression in colorectal cancer and correlate the expression of LBP-32 mRNA with tumor progression and metastasis. MATERIALS AND METHODS
Cell Line Clone A, a poorly differentiated human colon carcinoma cell line, was kindly provided by Dr. Lan Bo Chen, Dana-Farber Cancer Institute, Boston, Massachusetts. Clone A cells were incubated in 45% Dulbecco’s modification of Eagle’s essential medium (DMEM)/ 45% RPMI 1640 (RPMI)/10% fetal bovine serum (FBS) (GIBCO/ BRL, Grand Island, NY). Cell cultures were incubated on slides at 377C in humidified atmosphere air and 5% CO2 . Seventy percent confluent cells were fixed by 4% paraformaldehyde for in situ hybridization.
Tissues Thirty surgical specimen pairs of colorectal carcinoma tissues and adjacent normal colorectal tissues were collected from the operating room through the tissue retrieval services of Yale Comprehensive Cancer Center and the Cancer Institute of New Jersey. Among these, there were 24 cases of primary colorectal carcinomas including 14 cases of well- and moderately differentiated colorectal carcinomas and 10 cases of poorly differentiated colorectal carcinomas, and six cases of metastatic colorectal carcinomas in the liver. The diagnoses in all cases were confirmed by routine histiopathology. All tissues were fixed, dehydrated, and embedded in paraffin. Then sections of 5-mm thickness were mounted on aminoalkylsilane-treated glass slides [24] and the slides were stored at 47C.
Probe Preparation A 334-bp HindIII-Sac I fragment of human LBP-32 cDNA (J-9) [8] was inserted into the HindIII-SacI sites of PGEM-4 vector (Promega, Biological Research Products, Madison, WI) with flanking SP6 and T7 promotors, respectively. The resulting plasmid was linearized with either HindIII or EcoRI downstream of the coding sequence to permit generation of antisense or sense riboprobes by SP6 or T7 RNA polymerase, respectively (Fig. 1). The transcription of riboprobes was carried out as described [25] in vitro in the presence of 35S-a-UTP (DuPont, NEN Research Product, Boston, MA) and SP6 and T7 RNA polymerase (Boehringer Mannheim, Indianapolis, IN). The DNA template was removed by RNase-free DNase I (Sigma, Chemical Co., St. Louise, MO) digestion. The labeled transcription products were precipitated,
AID
JSR 4574
/
m4784$$$29
01-29-96 09:16:13
FIG. 2. The expression of LBP-32 mRNA in the human colon carcinoma cell line Clone A (A). The specific grains disappear on the slides hybridized with sense probe (B) and on the slides treated with RNase (C). The distribution and number of nonspecific grains (B, C) are quite different from the specific grains (A). (Hematoxylin and eosin counterstaining, 1400.)
srga
AP: Surg Res
122
JOURNAL OF SURGICAL RESEARCH: VOL. 61, NO. 1, FEBRUARY 15, 1996
TABLE 1 The Expression of LBP-32 mRNA in Normal and Tumor Tissues Cases Primary colorectal carcinoma Well- or moderately differentiated Poorly differentiated Liver metastases of colon carcinoma
24 14 10 6
Normal tissues adjacent to tumor 4.50 3.60 5.75 2.63
{ { { {
2.66 2.49 2.48 1.49
Tumor tissues 15.22 12.47 19.07 14.26
{ { { {
4.54 3.54 2.48 4.28
P1 õ 0.0001 P2 õ 0.0001 P3 õ 0.0001
Note. Mean grains/cell { SD. P1: Primary colon cancer vs normal colon adjacent to the tumor. P2: Poorly differentiated tumor vs well- or moderately differentiated tumor. P3: Liver metastasis vs normal liver adjacent to the tumor.
denaturation in NaCl and heat denaturation in 21 standard saline citrate (SSC) at 707C, followed by digestion with 5 mg/ml of pronase E (Sigma, Chemical Co., St. Louis, MO) at 377C. After digestion was stopped by dipping the slides in glycine, the slides were postfixed in 4% paraformaldehyde. The slides were equilibrated with 10 mM DTT and blocked by blocking solution at 457C (blocking solution: 10 mM DTT, 0.185% iodioacetamide and 0.125% N-ethylmaleimide in 1 1 PBS). The slides were then equilibrated with 0.1 M triethanolamine (TEA) (pH 8.0) and acetylated in 0.25 and 0.5% acetic anhydrate, 5 min each, at room temperature. Finally the slides were washed twice in 2 1 SSC, dehydrated, and air dried. Before hybridization, the probes were then mixed with the hybridization mix containing 50% formamide, 0.3 M NaCl, 1 mM EDTA (pH 8.0), 10 mM Tris-Cl (pH 8.0), 1 1 Denhardt’s solution, 500 mg/ ml yeast tRNA, 500 mg/ml poly A (Boehringer Mannheim, Indianapolis, IN), 50 mM DTT and 10% polyethylene glycol. The concentration of labeled probes was 1.5 1 105 cpm/ml. Twenty microliters of probes were applied evenly to each section. Hybridization was performed in a moist chamber containing moist chamber solution {50% formamide, 0.3 M NaCl, 1 mM EDTA (pH 8.0) and 10 mM Tris-Cl (pH 7.4)} for 4 hr at 457C with gentle agitation.
Posthybridization Wash After Hybridization, the slides were washed twice in wash solution A (50% formamide, 2 1 SSC and 20 mM b-mercaptoethanol), 15 min each, at 557C, twice in wash solution B (50% formamide, 2 1 SSC, 20 mM b-mercaptoethanol and 0.25% Triton-100), 15 min each, at 557C, and twice in wash solution C (2 1 SSC and 20 mM b-mercaptoethanol), 2 min each, at room temperature. The slides were then
treated with RNase A (20 mg/ml) and RNase T1 (1 U/ml) (Boehringer Mannheim, Indianapolis, IN) for 30 min at 377C. After RNase treatment, the slides were washed twice in wash solution C, 30 min each, at 457C, twice in wash solution A, 30 min each, 457C, and twice in 2 1 SSC, 5 min each, at room temperature. After washing, the slides were dehydrated in graded ethanol containing 0.3 M ammonium acetate and air dried.
Autoradiography Autoradiography was performed by dipping the slides into Kodak NTB-2 emulsion diluted 1:1 with distilled water at 457C. After 5 days exposure at 47C, the slides were developed, fixed, and counterstained with hematoxylin and eosin, dehydrated, cleared, and mounted as for immunohistochemistry [25].
Control Experiments The seperate set of sections were hybridizated with sense probes for specificity control. Similarly concurrent assay control sections were processed by treating with RNase before hybridization with antisense probe.
Quantitive Analysis of the Results The expression of LBP mRNA was analyzed by examining the silver grains on the sections in a bright field under light microscopy with high power (1400) and counting the number of silver grains of 50 cells in at least 10 different fields for every case. For each case, the number of positive grains per cell was determined by subtracting
FIG. 3. The expression of LBP-32 mRNA in the normal colon control, primary colon carcinoma, normal colon adjacent to the tumor, liver metastases, and normal liver adjacent to tumor. Each column represents the mean grains per cell.
AID
JSR 4574
/
m4784$$$29
01-29-96 09:16:13
srga
AP: Surg Res
PEI ET AL.: LAMININ-BINDING PROTEIN mRNA IN COLON CANCER
123
FIG. 4. LBP-32 mRNA expression in poorly differentiated colon cancer (A). moderately differentiated colon cancer (B) the specific grains on (A) and (B) disappear in the section treated with RNase before hybridization with antisense riboprobe (C) and the section hybridized with sense riboprobe (D). (Hematoxylin and eosin counterstaining, 1400.)
the average grains per cell calculated in the control sections hybridized with sense probes from the average grains per cell obtained in the sections hybridized with antisense probes (positive grains/cell Å average grains/cell in antisense section-average grains/cell in sense section).
Statistical Analysis The statistical significant differences between colorectal tumor and adjacent normal colon, and colonic tumor in situ and invasive tumors were determined by paired Student’s t test, and the differences between groups, i.e., poorly differentiated vs well- and moderately differentiated colorectal cancers, primary colorectal cancers vs metastatic colorectal cancers, and tumors in situ vs liver metastatic lesions were calculated by unpaired Student’s t test.
RESULTS
Using in situ hybridization, we found that the expression of LBP-32 mRNA is higher in the Clone A cells and the grains (positive) produced by the antisense probe tend to be arranged in groups or clumps around or on the cells (Fig. 2A). The nonspecific grains were, on the other hand, scattered randomly across the sections hybridized with sense probe (Fig. 2B) as well as the
AID
JSR 4574
/
m4784$$$29
01-29-96 09:16:13
RNase-treated sections hybridized with the antisense probe (Fig. 2C). The results of LBP-32 mRNA expression in human colon cancer tissues are described as mean grains/cell { SD and summarized in Table 1. The mean grains in primary colorectal carcinomas and metastatic colorectal carcinomas in the liver were much more than that in the matched adjacent normal colonic tissues and normal liver (t1 Å 15.474, P1 õ 0.001, and t2 Å 8.903, P2 õ 0.001) (Fig. 3). The expression of LBP-32 mRNA was related to the differentiated of tumor cells. The grains were more abundantly expressed in poorly differentiated colorectal cancers than in well- and moderately differentiated colorectal cancers (t Å 5.060, P õ 0.001) (Fig. 4A and B). The most significant finding in this study was that there was a relationship between expression of LBP-32 mRNA and tumor invasion and metastasis. Six cases of primary colorectal carcinomas showed both lesions in situ and invasive lesions in one section. Data comparing the expression of LBP-32 mRNA in the invasive components vs the tumors in situ are summarized in Table 2. The mean grains was greater in the invasive components compared to the lesions in situ (t Å 6.883, P
srga
AP: Surg Res
124
JOURNAL OF SURGICAL RESEARCH: VOL. 61, NO. 1, FEBRUARY 15, 1996
FIG. 5. The LBP-32 mRNA expression in the tumor lesion in situ (A) and invasive component of the tumor (B). The picture (A) and (B) were taken from the same section of colon cancer specimen. The upper part of (A) also shows that the LBP-32 mRNA expression in the normal colon adjacent to the tumor. (Hematoxylin and eosin counterstaining, 1400.)
õ 0.001) (Fig. 5). Comparison of liver metastatic lesions (Table 1) with tumor lesions in situ (Table 2) showed that the expression of LBP-32 mRNA was significantly higher in the liver metastases than in the tumor lesions in situ (t Å 2.834, P õ 0.005). DISCUSSION
In situ hybridization provides distinct advantages over Northern blots analysis of gene expression in the
AID
JSR 4574
/
m4784$$$30
01-29-96 09:16:13
tissue specimen. Although the Northern blots analysis has been routinely used to quantify mRNA level from total cellular extracts of homogenized tissue specimen, this technique is only useful to assess the average expression of mRNA in a particular sample, but does not provide information about heterogeneity from cell to cell. In situ hybridization is valuable for demonstrating the specific expression of a particular mRNA in tissues of mixed cell populations. Since many specimens contain both normal and tumor tissues or both carcinoma
srga
AP: Surg Res
PEI ET AL.: LAMININ-BINDING PROTEIN mRNA IN COLON CANCER
TABLE 2 The Expression of LBP-32 mRNA in the Tumors in Situ and Invasive Tumor Components in Individual Patient Samples Cases
Tumor tissue lesion in situ
Tumor tissue invasive lesion
1 2 3 4 5 6 Mean
8.80 6.16 4.56 4.96 8.34 10.46 7.21 { 2.35
14.08 10.88 7.76 10.78 11.32 18.18 12.17 { 3.57
P õ 0.001
Note. Mean grains/cell { SD.
in situ and invasive carcinoma. Northern blots analysis will not identify which components express a high level of a particular mRNA. In contrast, in situ hybridization can compare the mRNA expression levels in different components within the same tissue section. To validate the specificity of the LBP-32 mRNA antisense probe, we used Clone A cell line, known to overexpress LBP-32 mRNA [20], as a positive control. A sense probe and RNase treatment were used as additional controls. Our results demonstrated that the antisense probe is specific to the LBP-32 mRNA. In this study we found that the mRNA was much higher in the colon cancer cells than in the normal colon epithelial cells adjacent to the tumor. Significantly, we found that the LBP-32 mRNA in poorly differentiated colon carcinoma were much more than that in moderately to well-differentiated colon carcinoma. We also demonstrated that the expression of LBP-32 mRNA was higher in the invasive carcinoma lesions and in the liver metastatic lesions than that in the tumors in situ. The expression of LBP-32 was demonstrated to be higher in the human colon cancer tissue by monoclonal antibody to LBP-32 and this overexpression is associated with increased expression of LBP-32 mRNA as well as with the invasive phenotype of colon carcinoma [12]. Our observations extend previous findings and indicate a mechanistic causal relationship between tumor invasion, metastasis, and LBP-32 mRNA expression. Previous studies have indicated that poorly differentiated colon carcinoma cell lines adhered to a surface coated with laminin or reconstituted BM extracts to a significantly greater extent than moderately and welldifferentiated cell lines [26]. It was also shown that invasive cells had the highest level of specific lamininbinding activity [27]. Some investigators have shown that LBP-32 mRNA was higher in fetal liver and developing kidney than that the corresponding normal adult organs [28, 29]. Northern blots analysis demonstrated that LBP-32 mRNA was overexpressed in primary colon carcinoma compared with adjacent normal colonic epithelium, and poorly differentiated carcinoma expressed a higher level of LBP-32 mRNA than do well-
AID
JSR 4574
/
m4784$$$30
01-29-96 09:16:13
125
differentiated carcinomas. The expression of LBP-32 mRNA in the primary colon cancer had a significant correlation with Dukes’ classification and cancer progression, and liver metastases of colon cancer also expressed a high level of LBP mRNA [8, 12, 19–20]. Furthermore, antisera to a bacterial fusion protein of the LBP was proved to inhibit the adhesion of hepatocyte to laminin by 30% [10] and antisense RNA of LBP-32 was demonstrated to inhibit tumor cell attachment and invasiveness in vitro by in vitro cell invasion assays [21]. These findings together with our results in this study suggest that expression of LBP-32 mRNA is related to tumor cell differentiation, attachment, invasion, and metastasis. It is not clear how the increased expression of laminin-binding protein is associated with the malignant phenotype of the cancer cells in the differentiation cascade and how the LBP-32 mRNA plays a role in tumor cell invasion and metastasis. It was reported that ability of cell to bind the laminin may be controlled by the amount of laminin receptor mRNA of the cells available for translation [7]. This may imply that augmentation of the number of laminin receptor molecules on tumor cell surface may faciliate the interaction and stable attachment of cancer cells to laminin. The binding of tumor cells to laminin may actually trigger a positive feedback response in the cancer cells leading to increasing production of LBP or LR mRNA [30, 31]. Furthermore, the interaction between laminin and tumor cell surface may promote the secretion of active basement membrane degrading enzymes, thus facilitating the invasion of the tumor cell through the basement membrane barrier [32, 33]. In conclusion, there is heterogeneity of LBP-32 mRNA expression in the cancer cells. We postulate that those cancer cells in primary colon carcinoma with a higher level of LBP-32 mRNA expression may have a higher metastatic potential. Patients with colon carcinoma who express a high level of LBP-32 mRNA may have a poor prognosis. The mechanism by which the expression of LBP-32 is increased in metastatic colon cancer cells has not been elucidated. Retrospective and prospective studies analyzing LBP-32 mRNA expression with follow up of patients could provide clues to the prognostic indicator potential of LBP-32 mRNA overexpression. Furthermore, it would help us to understand the mechanism of tumor cell invasion and metastasis, and develop the therapeutic strategies for the treatment of colon carcinoma using this gene product as a target. ACKNOWLEDGMENT The authors thank Dr. Peter Amenta for his assistance in photography.
REFERENCES 1. Bering, C. C., Squires, T. S., Tong, T. Cancer statistics, 1993. CA Cancer J. Clin. 43: 7–26, 1993.
srga
AP: Surg Res
126
JOURNAL OF SURGICAL RESEARCH: VOL. 61, NO. 1, FEBRUARY 15, 1996
2. Aznavoorian, S., Murphy, A. N., Stetler-Stevenson, W. G., Liotta, L. A. Molecular aspects of tumor cell invasion and metastasis. Cancer 71: 1368–1383, 1993. 3. Kleinman, H. K., Cannon, F. B., Laurie, G. W., Hasell, J. R., Aumailley, M., Terranova, V. P., Martin, G. R., et al. Biological activities of laminin. Cell Biochem. 27: 317–325, 1985. 4. Beck, K., Hunter, I., Engel, J. Structure and function of laminin: Anatomy of a multidomain glycoprotein. FASEB J. 4: 148–160, 1990. 5. Rao, C. N., Basky, S. H., Terranova, V. P., Liotta, L. A. Isolation of a tumor cell laminin receptor. Biochem. Biophys. Res. Commun. 111: 804–808, 1983. 6. Runyan, R. B., Versalovic, J., Shur, B. D. Functionally distinct laminin receptors mediate cell adhesion and spreading. The requirement for surface galactosyl-transferase in cell spreading. J. Cell Biol. 107: 1863–1871, 1988. 7. Wewer, U. M., Liotta, L. A., Jaye, M., Ricca, G. A., Drohan, W. N., Claysmith, A. P., Rao, C. N., et al. Altered levels of laminin receptor mRNA in various human carcinoma cells that have different abilities to bind laminin. Proc. Natl. Acad. Sci. USA 83: 7137–7141, 1986. 8. Yow, H., Wong, J. M., Chen, H. S., Lee, C., Steele, G. D., Jr., Chen, L. B. Increased mRNA expression of a laminin-binding protein in human colon carcinomas: Complete sequence of a full length cDNA encoding the protein. Proc. Natl. Acad. Sci. USA 85: 6394–6398, 1988. 9. Rao, C. N., Castronovo, V., Schmittk, M. C., Wewer, U. M., Claysmith, A. P., Liotta, L. A., Sobel, M. E. Evidence for a precursor of the high-affinity metastasis-associated murine laminin receptor. Biochemistry 28: 7476–7486, 1989. 10. Clement, B., Segui-Real, B., Savagner, P., Kleiman, H. K., Yamada, Y. Hepatocytic attachment to laminin is mediated through multiple receptors. J. Cell Biol. 110: 185–192, 1990. 11. Castronovo, V., Claysmith, A. P., Barker, K. T., Cioce, V., Krutzsch, H. C., Sobel, M. E. Biosynthesis of the 67kD high affinity laminin receptor. Biochem. Biophys. Res. Commun. 177: 177–183, 1991. 12. Cioce, V., Castronovo, V., Shmookler, B. M., Garbisa, S., Grigioni, W. F., Liotta, L. A., Sobel, M. E. Increased expression of the laminin receptor in human colon cancer. J. Natl. Cancer Inst. 83: 29–36, 1991. 13. Basky, S. H., Rao, C. N., Williams, J. E., Liotta, L. A. Laminin molecular domains which alter metastasis in a murine model. J. Clin. Invest. 74: 843–848, 1984. 14. Malinoff, H. L., McCoy, J. P., Varani, J., Wicha, M. S. Metastatic potential of murine fibrosarcoma cells is influenced by cell surface laminin. Int. J. Cancer 33: 651–655, 1984. 15. Hand, P. H., Thor, A., Schlom, J., Rao, C. N., Liotta, L. A. Expression of laminin receptor in normal and carcinomatous tissues as defined by a monoclonal antibody. Cancer Res. 45: 2713–2719, 1985. 16. Iwamoto, Y., Robey, F. A., Graf, J., Sasaki, M., Kleinman, H. K., Yamada, Y., Martin, G. R. YIGSR, a synthetic laminin pentapeptide inhibits experimental metastasis. Science 238: 1132–1134, 1987. 17. Castronovo, V., Colin, C., Claysmith, A. P., Chen, P. H. S., Lifrange, E., Lambotte, R., Krutzsch, H., et al. Immunodetection of the metastasis-associated laminin receptor in human breast cancer cells obtained by fine needle aspiration biopsy. Am. J. Pathol. 137: 1373–1381, 1990. 18. Kanemoto, K., Reich, R., Royce, L., Greatorex, D., Adler, S. H.,
AID
JSR 4574
/
m4784$$$31
01-29-96 09:16:13
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
Shiraishi, N., Martin, G. R., et al. Identification of an amino acid sequence from the laminin A Chain that stimulates metastasis and colleganase IV production. Proc. Natl. Acad. Sci. USA 87: 2279–2283, 1990. Stallmach, A., Schuppan, D., Dax, J., Hanski, C., Riecken, E. O. Identification of laminin-binding protein in cell membranes of a human colon adenocarcinoma cell line. Gut 31: 70– 76, 1990. Mafune, K. I., Ravikumar, T. S., Wong, J. M., Yow, H., Chen, L. B., Steele, G. D. Expression of a Mr 32,000 laminin-binding protein messenger RNA in human colon carcinoma correlates with disease progression. Cancer Res. 50: 3888–3891, 1990. Mafune, K. I., Ravikumar, T. S. Antisense RNA of 32 kDa laminin-binding protein inhibits attachment and invasion of a human colon carcinoma cell lines. J. Surg. Res. 52: 340–346, 1992. Stoler, M. H. In situ hybridization: A research Technique as routine diagnostic test? Arch. Pathol. Lab. Med. 117: 478–481, 1993. Campo, E., Monteagudo, C., Castronovo, V., Claysmith, A. P., Fernandez, P. L., Sobel, M. E. Detection of laminin receptor mRNA in human cancer cell lines and colorectal tissues by in situ hybridization. Am. J. Pathol. 141: 1073–1083, 1992. Rentrop, M., Knapp, B., Winter, H., Schweizer, J. Aminoalkylsilane-treated glass slides as support for in situ hybridization of keratin cDNAs to frozen tissue sections under varying fixation and pretreatment conditions. Histochem. J. 18: 271–276, 1986. Zeller, R., Royers, M., Watkins, S. In situ hybridization and immunohistochemistry. In F. M. Ausubel, and R. Brent, et al. (Eds.), Current Protocol in Molecular Biology. New York: Green Publishing Associate Inc., 1991. Pp. 14.0.3–14.6.13. Daneker, G. W., Piazza, A. J., Steele, G. D., Jr., Mercurio, A. M. Relationship between extracellular matrix interactions and degree of differentiation in human colon carcinoma cell lines. Cancer Res. 49: 681–686, 1989. Aznavoorian, S., Liotta, L. A., and Kupchik, H. Z. Characteristics of invasive and noninvasive human colorectal adenocarcinoma cells. J. Natl. Cancer. Inst. 82: 1485–1493, 1990. Rescan, P. Y., Clement, B., Yamada, Y., Segui-Real, B., Baffet, G., Guguen-Guillouzo, C., Guillouzo, A. Differential expression of laminin chains and receptor (LBP-32) in fetal and neoplastic hepatocytes compared to normal adult hepatocytes in vivo and in culture. Am. J. Pathol. 137: 701–709, 1990. Laurie, G. W., Horikoshi, S., Killen, P. D., Segui-Real, B., Yamada, Y. In situ hybridization reveals temporal and spatial changes in cellular expression of mRNA for a laminin receptor, laminin, and basement membrane (Type IV) collagen in the developing kidney. J. Cell Biol. 109: 1351–1362, 1989. Castronovo, V., Taraboletti, G., Liotta, L. A., Sobel, M. E. Modulation of laminin receptor expression by estrogen and progestins in human breast cancer cell lines. J. Natl. Cancer Inst. 81: 781– 786, 1989. Castronovo, V., Sobel, M. E. Laminin and fibronectin increase the steady state level of the 67-kDa high affihity metastasisassociated laminin receptors mRNA in human cancer cells. Biochem. Biophys. Res. Commun. 168: 1110–1117, 1990. Turpeeniemi-Hujanen, T., Thorgeirsson, U. P., Rao, C. N., Liotta, L. A. Laminin increases the release of type IV collagenase from malignant cells. J. Biol. Chem. 261: 1883–1889, 1986. Castronovo, V., Taraboletti, G., Sobel, M. E. Laminin receptor complementary DNA-deduced synthetic peptide inhibits cancer cell attachment to endothelium. Cancer Res. 51: 5672–5678, 1991.
srga
AP: Surg Res
61, 120–126 (1996)
0091
Expression of 32-kDa Laminin-Binding Protein mRNA in Colon Cancer Tissues DONG PO PEI, M.D., YOUNG HAN, PH.D., DEEPAK NARAYAN, M.D., DANIEL HERZ, M.D., AND THANJAVUR S. RAVIKUMAR, M.D.1 Surgical Oncology, Department of Surgery, The Cancer Institute of New Jersey, CABM Room 223, 679 Hoes Lane, Piscataway, New Jersey 08854 Submitted for publication August 22, 1994
Colorectal cancer initiation and progression are associated with stepwise genetic alterations. We and others have shown that a gene encoding for a 32-kDa putative laminin-binding protein (LBP-32) is overexpressed during colorectal cancer progression by Northern blots analysis. Northern blots cannot indicate the heterogeneity of expression from cell to cell and the distribution pattern of gene expression within a given tumor. In order to overcome these problems, we examined the LBP-32 mRNA expression in colorectal carcinomas by in situ hybridization. LBP-32 mRNA expression in 30 cases of primary and metastatic colorectal cancers and their respective adjacent normal tissues were detected by in situ hybridization using 35S-UTP radiolabeled antisense riboprobes. The results showed that LBP-32 mRNA was expressed at a low level in the normal colonic mucosa adjacent to the tumor compared with colon cancer tissues. Its expression in poorly differentiated colorectal cancer was much higher than that in well- and moderately differentiated colorectal cancer. More importantly, the LBP32 mRNA was expressed more highly in the invasive lesions of the cancer and liver metastases compared with the cancer lesions in situ. Our results imply that in situ hybridization is a powerful tool in evaluating the changes in gene expression in the cancer cells and LBP-32 mRNA expression is related to progression, invasion, and metastasis of colorectal cancer. q 1996 Academic Press, Inc.
INTRODUCTION
Colorectal carcinoma is the second leading cause of cancer death in the United States. About 152,000 new cases are expected in 1993 and more than 57,000 of them will die of this disease [1]. Invasion and metastasis are the greatest obstacles to the successful treatment and the major reasons for cancer patient mortal1 To whom correspondence should be addressed at The Cancer Institute of New Jersey, CABM Room 223, 679 Hoes Lane, Piscataway, New Jersey 08854. Fax: (908) 2355475.
0022-4804/96 $18.00 Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
AID
JSR 4574
/
m4784$$$28
ity. Therefore, the identification of new genetic and biochemical markers that would predict the invasive and metastatic potential of a cancer at an early stage will be important from the diagnostic and therapeutic standpoint. The process of tumor invasion and metastasis is charactered by a series of interactions between tumor cells and basement membranes (BM). Tumor cells have to penetrate this natural tissue barrier during the invasion and metastasis [2]. Laminin (Mr Å 800,000) is a prominent component of BM and promotes many biological activities including cell adhesion, cell spreading, morphogenesis, differentiation, proliferation, mitogenesis, neutritic outgrowth, and cell migration [3, 4]. Many of these activities are mediated through the cell surface laminin-binding protein (LBP) or laminin receptors (LR) [5, 6]. Human LR complementary DNA (cDNA) clones were originally identified from the cDNA expression library of human endothelial cells by means of screening with a monoclonal antibody against the 67-kDa LR [7]. Subsequently, a full length human cDNA clone for human LBP was identified by the differential screening of a cDNA library from a poorly differentiated human colon carcinoma cell line [8]. Its sequence was homologous to that of the 67-kDa LR cDNA. The coding capacity of the full length cDNA clone, however, is for nascent protein with a calculated molecular mass of 32 kDa (LBP-32) [8]. It was suggested that the LBP-32 and 67-kDa LR are antigenically related and a precursor product relationship exists between them [9–12]. The function of LBP-32 is not yet defined. Attachment of laminin to tumor cells enhances several properties of the malignant phenotype and has been associated with increased tumorigenic and metastatic potential of tumor cells in vivo and in vitro [13–18]. LBP-32 and its mRNA were overexpressed in colorectal carcinoma and its expression has significant correlation with Dukes’ classification [8, 12, 15, 19–20]. It was also found that antisense RNA of LBP-32 inhibited tumor cell attachment and invasion in vitro [21]. These findings suggest that LBP-32 has a role in the colorectal cancer progression, invasion, and metastasis.
120
01-29-96 09:16:13
srga
AP: Surg Res
PEI ET AL.: LAMININ-BINDING PROTEIN mRNA IN COLON CANCER
121
dried, and resuspended in diethyl pyrocarbonate (DEPC)-treated distilled water containing RNasin (Boehringer Mannheim, Indianapolis, IN) and 10 mM DTT. Incorporation rate was over 70%. The probes were used on the day of synthesis.
In Situ Hybridization GEM-4
FIG. 1. The schematic diagram of the plasmid P used for in vitro transcription of antisense and sense probes of LBP-32 mRNA.
In situ hybridization was performed as described by Zeller et al. [25]. Pretreatment of slides included the dewaxing, rehydration, acid
In previous studies, an overexpression of LBP-32 mRNA in human colorectal cancer compared with adjacent normal colonic mucosae was demonstrated using Northern blots analysis. Northern blots analysis is relatively crude and does not provide insight into the changes of gene expression in a subpopulation within a given tumor. The technique of in situ hybridization allows us to identify and localize the gene expression at the single cell level [22, 23]. In this study, we used in situ hybridization to assess the heterogeneity of LBP-32 mRNA expression in colorectal cancer and correlate the expression of LBP-32 mRNA with tumor progression and metastasis. MATERIALS AND METHODS
Cell Line Clone A, a poorly differentiated human colon carcinoma cell line, was kindly provided by Dr. Lan Bo Chen, Dana-Farber Cancer Institute, Boston, Massachusetts. Clone A cells were incubated in 45% Dulbecco’s modification of Eagle’s essential medium (DMEM)/ 45% RPMI 1640 (RPMI)/10% fetal bovine serum (FBS) (GIBCO/ BRL, Grand Island, NY). Cell cultures were incubated on slides at 377C in humidified atmosphere air and 5% CO2 . Seventy percent confluent cells were fixed by 4% paraformaldehyde for in situ hybridization.
Tissues Thirty surgical specimen pairs of colorectal carcinoma tissues and adjacent normal colorectal tissues were collected from the operating room through the tissue retrieval services of Yale Comprehensive Cancer Center and the Cancer Institute of New Jersey. Among these, there were 24 cases of primary colorectal carcinomas including 14 cases of well- and moderately differentiated colorectal carcinomas and 10 cases of poorly differentiated colorectal carcinomas, and six cases of metastatic colorectal carcinomas in the liver. The diagnoses in all cases were confirmed by routine histiopathology. All tissues were fixed, dehydrated, and embedded in paraffin. Then sections of 5-mm thickness were mounted on aminoalkylsilane-treated glass slides [24] and the slides were stored at 47C.
Probe Preparation A 334-bp HindIII-Sac I fragment of human LBP-32 cDNA (J-9) [8] was inserted into the HindIII-SacI sites of PGEM-4 vector (Promega, Biological Research Products, Madison, WI) with flanking SP6 and T7 promotors, respectively. The resulting plasmid was linearized with either HindIII or EcoRI downstream of the coding sequence to permit generation of antisense or sense riboprobes by SP6 or T7 RNA polymerase, respectively (Fig. 1). The transcription of riboprobes was carried out as described [25] in vitro in the presence of 35S-a-UTP (DuPont, NEN Research Product, Boston, MA) and SP6 and T7 RNA polymerase (Boehringer Mannheim, Indianapolis, IN). The DNA template was removed by RNase-free DNase I (Sigma, Chemical Co., St. Louise, MO) digestion. The labeled transcription products were precipitated,
AID
JSR 4574
/
m4784$$$29
01-29-96 09:16:13
FIG. 2. The expression of LBP-32 mRNA in the human colon carcinoma cell line Clone A (A). The specific grains disappear on the slides hybridized with sense probe (B) and on the slides treated with RNase (C). The distribution and number of nonspecific grains (B, C) are quite different from the specific grains (A). (Hematoxylin and eosin counterstaining, 1400.)
srga
AP: Surg Res
122
JOURNAL OF SURGICAL RESEARCH: VOL. 61, NO. 1, FEBRUARY 15, 1996
TABLE 1 The Expression of LBP-32 mRNA in Normal and Tumor Tissues Cases Primary colorectal carcinoma Well- or moderately differentiated Poorly differentiated Liver metastases of colon carcinoma
24 14 10 6
Normal tissues adjacent to tumor 4.50 3.60 5.75 2.63
{ { { {
2.66 2.49 2.48 1.49
Tumor tissues 15.22 12.47 19.07 14.26
{ { { {
4.54 3.54 2.48 4.28
P1 õ 0.0001 P2 õ 0.0001 P3 õ 0.0001
Note. Mean grains/cell { SD. P1: Primary colon cancer vs normal colon adjacent to the tumor. P2: Poorly differentiated tumor vs well- or moderately differentiated tumor. P3: Liver metastasis vs normal liver adjacent to the tumor.
denaturation in NaCl and heat denaturation in 21 standard saline citrate (SSC) at 707C, followed by digestion with 5 mg/ml of pronase E (Sigma, Chemical Co., St. Louis, MO) at 377C. After digestion was stopped by dipping the slides in glycine, the slides were postfixed in 4% paraformaldehyde. The slides were equilibrated with 10 mM DTT and blocked by blocking solution at 457C (blocking solution: 10 mM DTT, 0.185% iodioacetamide and 0.125% N-ethylmaleimide in 1 1 PBS). The slides were then equilibrated with 0.1 M triethanolamine (TEA) (pH 8.0) and acetylated in 0.25 and 0.5% acetic anhydrate, 5 min each, at room temperature. Finally the slides were washed twice in 2 1 SSC, dehydrated, and air dried. Before hybridization, the probes were then mixed with the hybridization mix containing 50% formamide, 0.3 M NaCl, 1 mM EDTA (pH 8.0), 10 mM Tris-Cl (pH 8.0), 1 1 Denhardt’s solution, 500 mg/ ml yeast tRNA, 500 mg/ml poly A (Boehringer Mannheim, Indianapolis, IN), 50 mM DTT and 10% polyethylene glycol. The concentration of labeled probes was 1.5 1 105 cpm/ml. Twenty microliters of probes were applied evenly to each section. Hybridization was performed in a moist chamber containing moist chamber solution {50% formamide, 0.3 M NaCl, 1 mM EDTA (pH 8.0) and 10 mM Tris-Cl (pH 7.4)} for 4 hr at 457C with gentle agitation.
Posthybridization Wash After Hybridization, the slides were washed twice in wash solution A (50% formamide, 2 1 SSC and 20 mM b-mercaptoethanol), 15 min each, at 557C, twice in wash solution B (50% formamide, 2 1 SSC, 20 mM b-mercaptoethanol and 0.25% Triton-100), 15 min each, at 557C, and twice in wash solution C (2 1 SSC and 20 mM b-mercaptoethanol), 2 min each, at room temperature. The slides were then
treated with RNase A (20 mg/ml) and RNase T1 (1 U/ml) (Boehringer Mannheim, Indianapolis, IN) for 30 min at 377C. After RNase treatment, the slides were washed twice in wash solution C, 30 min each, at 457C, twice in wash solution A, 30 min each, 457C, and twice in 2 1 SSC, 5 min each, at room temperature. After washing, the slides were dehydrated in graded ethanol containing 0.3 M ammonium acetate and air dried.
Autoradiography Autoradiography was performed by dipping the slides into Kodak NTB-2 emulsion diluted 1:1 with distilled water at 457C. After 5 days exposure at 47C, the slides were developed, fixed, and counterstained with hematoxylin and eosin, dehydrated, cleared, and mounted as for immunohistochemistry [25].
Control Experiments The seperate set of sections were hybridizated with sense probes for specificity control. Similarly concurrent assay control sections were processed by treating with RNase before hybridization with antisense probe.
Quantitive Analysis of the Results The expression of LBP mRNA was analyzed by examining the silver grains on the sections in a bright field under light microscopy with high power (1400) and counting the number of silver grains of 50 cells in at least 10 different fields for every case. For each case, the number of positive grains per cell was determined by subtracting
FIG. 3. The expression of LBP-32 mRNA in the normal colon control, primary colon carcinoma, normal colon adjacent to the tumor, liver metastases, and normal liver adjacent to tumor. Each column represents the mean grains per cell.
AID
JSR 4574
/
m4784$$$29
01-29-96 09:16:13
srga
AP: Surg Res
PEI ET AL.: LAMININ-BINDING PROTEIN mRNA IN COLON CANCER
123
FIG. 4. LBP-32 mRNA expression in poorly differentiated colon cancer (A). moderately differentiated colon cancer (B) the specific grains on (A) and (B) disappear in the section treated with RNase before hybridization with antisense riboprobe (C) and the section hybridized with sense riboprobe (D). (Hematoxylin and eosin counterstaining, 1400.)
the average grains per cell calculated in the control sections hybridized with sense probes from the average grains per cell obtained in the sections hybridized with antisense probes (positive grains/cell Å average grains/cell in antisense section-average grains/cell in sense section).
Statistical Analysis The statistical significant differences between colorectal tumor and adjacent normal colon, and colonic tumor in situ and invasive tumors were determined by paired Student’s t test, and the differences between groups, i.e., poorly differentiated vs well- and moderately differentiated colorectal cancers, primary colorectal cancers vs metastatic colorectal cancers, and tumors in situ vs liver metastatic lesions were calculated by unpaired Student’s t test.
RESULTS
Using in situ hybridization, we found that the expression of LBP-32 mRNA is higher in the Clone A cells and the grains (positive) produced by the antisense probe tend to be arranged in groups or clumps around or on the cells (Fig. 2A). The nonspecific grains were, on the other hand, scattered randomly across the sections hybridized with sense probe (Fig. 2B) as well as the
AID
JSR 4574
/
m4784$$$29
01-29-96 09:16:13
RNase-treated sections hybridized with the antisense probe (Fig. 2C). The results of LBP-32 mRNA expression in human colon cancer tissues are described as mean grains/cell { SD and summarized in Table 1. The mean grains in primary colorectal carcinomas and metastatic colorectal carcinomas in the liver were much more than that in the matched adjacent normal colonic tissues and normal liver (t1 Å 15.474, P1 õ 0.001, and t2 Å 8.903, P2 õ 0.001) (Fig. 3). The expression of LBP-32 mRNA was related to the differentiated of tumor cells. The grains were more abundantly expressed in poorly differentiated colorectal cancers than in well- and moderately differentiated colorectal cancers (t Å 5.060, P õ 0.001) (Fig. 4A and B). The most significant finding in this study was that there was a relationship between expression of LBP-32 mRNA and tumor invasion and metastasis. Six cases of primary colorectal carcinomas showed both lesions in situ and invasive lesions in one section. Data comparing the expression of LBP-32 mRNA in the invasive components vs the tumors in situ are summarized in Table 2. The mean grains was greater in the invasive components compared to the lesions in situ (t Å 6.883, P
srga
AP: Surg Res
124
JOURNAL OF SURGICAL RESEARCH: VOL. 61, NO. 1, FEBRUARY 15, 1996
FIG. 5. The LBP-32 mRNA expression in the tumor lesion in situ (A) and invasive component of the tumor (B). The picture (A) and (B) were taken from the same section of colon cancer specimen. The upper part of (A) also shows that the LBP-32 mRNA expression in the normal colon adjacent to the tumor. (Hematoxylin and eosin counterstaining, 1400.)
õ 0.001) (Fig. 5). Comparison of liver metastatic lesions (Table 1) with tumor lesions in situ (Table 2) showed that the expression of LBP-32 mRNA was significantly higher in the liver metastases than in the tumor lesions in situ (t Å 2.834, P õ 0.005). DISCUSSION
In situ hybridization provides distinct advantages over Northern blots analysis of gene expression in the
AID
JSR 4574
/
m4784$$$30
01-29-96 09:16:13
tissue specimen. Although the Northern blots analysis has been routinely used to quantify mRNA level from total cellular extracts of homogenized tissue specimen, this technique is only useful to assess the average expression of mRNA in a particular sample, but does not provide information about heterogeneity from cell to cell. In situ hybridization is valuable for demonstrating the specific expression of a particular mRNA in tissues of mixed cell populations. Since many specimens contain both normal and tumor tissues or both carcinoma
srga
AP: Surg Res
PEI ET AL.: LAMININ-BINDING PROTEIN mRNA IN COLON CANCER
TABLE 2 The Expression of LBP-32 mRNA in the Tumors in Situ and Invasive Tumor Components in Individual Patient Samples Cases
Tumor tissue lesion in situ
Tumor tissue invasive lesion
1 2 3 4 5 6 Mean
8.80 6.16 4.56 4.96 8.34 10.46 7.21 { 2.35
14.08 10.88 7.76 10.78 11.32 18.18 12.17 { 3.57
P õ 0.001
Note. Mean grains/cell { SD.
in situ and invasive carcinoma. Northern blots analysis will not identify which components express a high level of a particular mRNA. In contrast, in situ hybridization can compare the mRNA expression levels in different components within the same tissue section. To validate the specificity of the LBP-32 mRNA antisense probe, we used Clone A cell line, known to overexpress LBP-32 mRNA [20], as a positive control. A sense probe and RNase treatment were used as additional controls. Our results demonstrated that the antisense probe is specific to the LBP-32 mRNA. In this study we found that the mRNA was much higher in the colon cancer cells than in the normal colon epithelial cells adjacent to the tumor. Significantly, we found that the LBP-32 mRNA in poorly differentiated colon carcinoma were much more than that in moderately to well-differentiated colon carcinoma. We also demonstrated that the expression of LBP-32 mRNA was higher in the invasive carcinoma lesions and in the liver metastatic lesions than that in the tumors in situ. The expression of LBP-32 was demonstrated to be higher in the human colon cancer tissue by monoclonal antibody to LBP-32 and this overexpression is associated with increased expression of LBP-32 mRNA as well as with the invasive phenotype of colon carcinoma [12]. Our observations extend previous findings and indicate a mechanistic causal relationship between tumor invasion, metastasis, and LBP-32 mRNA expression. Previous studies have indicated that poorly differentiated colon carcinoma cell lines adhered to a surface coated with laminin or reconstituted BM extracts to a significantly greater extent than moderately and welldifferentiated cell lines [26]. It was also shown that invasive cells had the highest level of specific lamininbinding activity [27]. Some investigators have shown that LBP-32 mRNA was higher in fetal liver and developing kidney than that the corresponding normal adult organs [28, 29]. Northern blots analysis demonstrated that LBP-32 mRNA was overexpressed in primary colon carcinoma compared with adjacent normal colonic epithelium, and poorly differentiated carcinoma expressed a higher level of LBP-32 mRNA than do well-
AID
JSR 4574
/
m4784$$$30
01-29-96 09:16:13
125
differentiated carcinomas. The expression of LBP-32 mRNA in the primary colon cancer had a significant correlation with Dukes’ classification and cancer progression, and liver metastases of colon cancer also expressed a high level of LBP mRNA [8, 12, 19–20]. Furthermore, antisera to a bacterial fusion protein of the LBP was proved to inhibit the adhesion of hepatocyte to laminin by 30% [10] and antisense RNA of LBP-32 was demonstrated to inhibit tumor cell attachment and invasiveness in vitro by in vitro cell invasion assays [21]. These findings together with our results in this study suggest that expression of LBP-32 mRNA is related to tumor cell differentiation, attachment, invasion, and metastasis. It is not clear how the increased expression of laminin-binding protein is associated with the malignant phenotype of the cancer cells in the differentiation cascade and how the LBP-32 mRNA plays a role in tumor cell invasion and metastasis. It was reported that ability of cell to bind the laminin may be controlled by the amount of laminin receptor mRNA of the cells available for translation [7]. This may imply that augmentation of the number of laminin receptor molecules on tumor cell surface may faciliate the interaction and stable attachment of cancer cells to laminin. The binding of tumor cells to laminin may actually trigger a positive feedback response in the cancer cells leading to increasing production of LBP or LR mRNA [30, 31]. Furthermore, the interaction between laminin and tumor cell surface may promote the secretion of active basement membrane degrading enzymes, thus facilitating the invasion of the tumor cell through the basement membrane barrier [32, 33]. In conclusion, there is heterogeneity of LBP-32 mRNA expression in the cancer cells. We postulate that those cancer cells in primary colon carcinoma with a higher level of LBP-32 mRNA expression may have a higher metastatic potential. Patients with colon carcinoma who express a high level of LBP-32 mRNA may have a poor prognosis. The mechanism by which the expression of LBP-32 is increased in metastatic colon cancer cells has not been elucidated. Retrospective and prospective studies analyzing LBP-32 mRNA expression with follow up of patients could provide clues to the prognostic indicator potential of LBP-32 mRNA overexpression. Furthermore, it would help us to understand the mechanism of tumor cell invasion and metastasis, and develop the therapeutic strategies for the treatment of colon carcinoma using this gene product as a target. ACKNOWLEDGMENT The authors thank Dr. Peter Amenta for his assistance in photography.
REFERENCES 1. Bering, C. C., Squires, T. S., Tong, T. Cancer statistics, 1993. CA Cancer J. Clin. 43: 7–26, 1993.
srga
AP: Surg Res
126
JOURNAL OF SURGICAL RESEARCH: VOL. 61, NO. 1, FEBRUARY 15, 1996
2. Aznavoorian, S., Murphy, A. N., Stetler-Stevenson, W. G., Liotta, L. A. Molecular aspects of tumor cell invasion and metastasis. Cancer 71: 1368–1383, 1993. 3. Kleinman, H. K., Cannon, F. B., Laurie, G. W., Hasell, J. R., Aumailley, M., Terranova, V. P., Martin, G. R., et al. Biological activities of laminin. Cell Biochem. 27: 317–325, 1985. 4. Beck, K., Hunter, I., Engel, J. Structure and function of laminin: Anatomy of a multidomain glycoprotein. FASEB J. 4: 148–160, 1990. 5. Rao, C. N., Basky, S. H., Terranova, V. P., Liotta, L. A. Isolation of a tumor cell laminin receptor. Biochem. Biophys. Res. Commun. 111: 804–808, 1983. 6. Runyan, R. B., Versalovic, J., Shur, B. D. Functionally distinct laminin receptors mediate cell adhesion and spreading. The requirement for surface galactosyl-transferase in cell spreading. J. Cell Biol. 107: 1863–1871, 1988. 7. Wewer, U. M., Liotta, L. A., Jaye, M., Ricca, G. A., Drohan, W. N., Claysmith, A. P., Rao, C. N., et al. Altered levels of laminin receptor mRNA in various human carcinoma cells that have different abilities to bind laminin. Proc. Natl. Acad. Sci. USA 83: 7137–7141, 1986. 8. Yow, H., Wong, J. M., Chen, H. S., Lee, C., Steele, G. D., Jr., Chen, L. B. Increased mRNA expression of a laminin-binding protein in human colon carcinomas: Complete sequence of a full length cDNA encoding the protein. Proc. Natl. Acad. Sci. USA 85: 6394–6398, 1988. 9. Rao, C. N., Castronovo, V., Schmittk, M. C., Wewer, U. M., Claysmith, A. P., Liotta, L. A., Sobel, M. E. Evidence for a precursor of the high-affinity metastasis-associated murine laminin receptor. Biochemistry 28: 7476–7486, 1989. 10. Clement, B., Segui-Real, B., Savagner, P., Kleiman, H. K., Yamada, Y. Hepatocytic attachment to laminin is mediated through multiple receptors. J. Cell Biol. 110: 185–192, 1990. 11. Castronovo, V., Claysmith, A. P., Barker, K. T., Cioce, V., Krutzsch, H. C., Sobel, M. E. Biosynthesis of the 67kD high affinity laminin receptor. Biochem. Biophys. Res. Commun. 177: 177–183, 1991. 12. Cioce, V., Castronovo, V., Shmookler, B. M., Garbisa, S., Grigioni, W. F., Liotta, L. A., Sobel, M. E. Increased expression of the laminin receptor in human colon cancer. J. Natl. Cancer Inst. 83: 29–36, 1991. 13. Basky, S. H., Rao, C. N., Williams, J. E., Liotta, L. A. Laminin molecular domains which alter metastasis in a murine model. J. Clin. Invest. 74: 843–848, 1984. 14. Malinoff, H. L., McCoy, J. P., Varani, J., Wicha, M. S. Metastatic potential of murine fibrosarcoma cells is influenced by cell surface laminin. Int. J. Cancer 33: 651–655, 1984. 15. Hand, P. H., Thor, A., Schlom, J., Rao, C. N., Liotta, L. A. Expression of laminin receptor in normal and carcinomatous tissues as defined by a monoclonal antibody. Cancer Res. 45: 2713–2719, 1985. 16. Iwamoto, Y., Robey, F. A., Graf, J., Sasaki, M., Kleinman, H. K., Yamada, Y., Martin, G. R. YIGSR, a synthetic laminin pentapeptide inhibits experimental metastasis. Science 238: 1132–1134, 1987. 17. Castronovo, V., Colin, C., Claysmith, A. P., Chen, P. H. S., Lifrange, E., Lambotte, R., Krutzsch, H., et al. Immunodetection of the metastasis-associated laminin receptor in human breast cancer cells obtained by fine needle aspiration biopsy. Am. J. Pathol. 137: 1373–1381, 1990. 18. Kanemoto, K., Reich, R., Royce, L., Greatorex, D., Adler, S. H.,
AID
JSR 4574
/
m4784$$$31
01-29-96 09:16:13
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
Shiraishi, N., Martin, G. R., et al. Identification of an amino acid sequence from the laminin A Chain that stimulates metastasis and colleganase IV production. Proc. Natl. Acad. Sci. USA 87: 2279–2283, 1990. Stallmach, A., Schuppan, D., Dax, J., Hanski, C., Riecken, E. O. Identification of laminin-binding protein in cell membranes of a human colon adenocarcinoma cell line. Gut 31: 70– 76, 1990. Mafune, K. I., Ravikumar, T. S., Wong, J. M., Yow, H., Chen, L. B., Steele, G. D. Expression of a Mr 32,000 laminin-binding protein messenger RNA in human colon carcinoma correlates with disease progression. Cancer Res. 50: 3888–3891, 1990. Mafune, K. I., Ravikumar, T. S. Antisense RNA of 32 kDa laminin-binding protein inhibits attachment and invasion of a human colon carcinoma cell lines. J. Surg. Res. 52: 340–346, 1992. Stoler, M. H. In situ hybridization: A research Technique as routine diagnostic test? Arch. Pathol. Lab. Med. 117: 478–481, 1993. Campo, E., Monteagudo, C., Castronovo, V., Claysmith, A. P., Fernandez, P. L., Sobel, M. E. Detection of laminin receptor mRNA in human cancer cell lines and colorectal tissues by in situ hybridization. Am. J. Pathol. 141: 1073–1083, 1992. Rentrop, M., Knapp, B., Winter, H., Schweizer, J. Aminoalkylsilane-treated glass slides as support for in situ hybridization of keratin cDNAs to frozen tissue sections under varying fixation and pretreatment conditions. Histochem. J. 18: 271–276, 1986. Zeller, R., Royers, M., Watkins, S. In situ hybridization and immunohistochemistry. In F. M. Ausubel, and R. Brent, et al. (Eds.), Current Protocol in Molecular Biology. New York: Green Publishing Associate Inc., 1991. Pp. 14.0.3–14.6.13. Daneker, G. W., Piazza, A. J., Steele, G. D., Jr., Mercurio, A. M. Relationship between extracellular matrix interactions and degree of differentiation in human colon carcinoma cell lines. Cancer Res. 49: 681–686, 1989. Aznavoorian, S., Liotta, L. A., and Kupchik, H. Z. Characteristics of invasive and noninvasive human colorectal adenocarcinoma cells. J. Natl. Cancer. Inst. 82: 1485–1493, 1990. Rescan, P. Y., Clement, B., Yamada, Y., Segui-Real, B., Baffet, G., Guguen-Guillouzo, C., Guillouzo, A. Differential expression of laminin chains and receptor (LBP-32) in fetal and neoplastic hepatocytes compared to normal adult hepatocytes in vivo and in culture. Am. J. Pathol. 137: 701–709, 1990. Laurie, G. W., Horikoshi, S., Killen, P. D., Segui-Real, B., Yamada, Y. In situ hybridization reveals temporal and spatial changes in cellular expression of mRNA for a laminin receptor, laminin, and basement membrane (Type IV) collagen in the developing kidney. J. Cell Biol. 109: 1351–1362, 1989. Castronovo, V., Taraboletti, G., Liotta, L. A., Sobel, M. E. Modulation of laminin receptor expression by estrogen and progestins in human breast cancer cell lines. J. Natl. Cancer Inst. 81: 781– 786, 1989. Castronovo, V., Sobel, M. E. Laminin and fibronectin increase the steady state level of the 67-kDa high affihity metastasisassociated laminin receptors mRNA in human cancer cells. Biochem. Biophys. Res. Commun. 168: 1110–1117, 1990. Turpeeniemi-Hujanen, T., Thorgeirsson, U. P., Rao, C. N., Liotta, L. A. Laminin increases the release of type IV collagenase from malignant cells. J. Biol. Chem. 261: 1883–1889, 1986. Castronovo, V., Taraboletti, G., Sobel, M. E. Laminin receptor complementary DNA-deduced synthetic peptide inhibits cancer cell attachment to endothelium. Cancer Res. 51: 5672–5678, 1991.
srga
AP: Surg Res