- Email: [email protected]
Epidermal growth factor receptor gene amplification is correlated with laminin-5 γ2 chain expression in oral squamous cell carcinoma cell lines
Cancer Letters 175 (2002) 197–204 www.elsevier.com/locate/canlet
Epidermal growth factor receptor gene amplification is correlated with laminin-5 g2 chain expression in oral squamous cell carcinoma cell lines Yukiko Ono 1, Yukihiro Nakanishi, Masahiro Gotoh, Michiie Sakamoto, Setsuo Hirohashi* Pathology Division, National Cancer Center Research Institute, 1-1, Tsukiji 5-chome, Chuo-ku, Tokyo 104-0045, Japan Received 25 May 2001; received in revised form 25 June 2001; accepted 5 July 2001
Abstract Both epidermal growth factor receptor (EGFR) gene amplification and laminin (Ln)-5 g2 chain overexpression have been reported to be poor prognostic factors in patients with squamous cell carcinoma (SCC) of the head and neck. Here we report our investigation of the relationship between EGFR gene amplification and Ln-5 g2 chain expression in seven SCC cell lines, since both epidermal growth factor (EGF) signaling and Ln-5 g2 have been reported to be involved in cell motility. The degree of correlation between EGFR gene amplification and Ln-5 g2 chain expression was evaluated by Southern and Western blot analyses. EGFR gene amplification was detected in all SCC cell lines at levels 5–50 times those in DNA from normal liver tissue. EGFR gene amplification increased with Ln-5 g2 chain protein expression in seven cell lines, showing close correlation between EGFR gene amplification and Ln-5 g2 chain protein expression. In order to show the causal relationship, we analyzed the effects of transforming growth factor-a (TGF-a), tyrosine kinase inhibitor of EGFR, and neutralizing antibody against EGFR, on the expression of Ln-5 g2 in these cell lines. In two cell lines in which EGFR gene amplification was low , expression of both protein and mRNA of the Ln-5 g2 chain increased in the presence of TGF-a, and Ln-5 g2 chain expression was inhibited by neutralizing antibody against EGFR. In all cell lines, Ln-5 g2 chain expression was inhibited by tyrosine kinase inhibitor which acts selectively on the EGFR signal transduction pathway under the stimulus of TGF-a. These results suggest that EGFR gene amplification and the EGFR signaling pathway can act as positive regulators on the induction of the Ln-5 g2 chain secreted by tumor cells. q 2002 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Epidermal growth factor receptor; Laminin-5 g2 chain; Gene amplification; Overexpression; Squamous cell carcinoma cell line
1. Introduction
* Corresponding author. Tel.: 181-3-3542-2511 ext. 4101; fax: 181-3-3248-2463. E-mail address: [email protected] (S. Hirohashi). 1 Present address: Division of Reconstructive Surgery for Oral and Maxillofacial Region, Department of Tissue Regeneration and Reconstruction, Course for Oral Life Science, Niigata University Graduate School of Medical and Dental Sciences, Niigata, Japan.
Human epidermal growth factor receptor (EGFR) is a 170-kDa transmembrane protein with intrinsic tyrosine kinase activity that regulates cell growth in response to binding of epidermal growth factor (EGF) or transforming growth factor-a (TGF-a), the major ligands for EGFR [1,2]. Overexpression or gene amplification of EGFR has been associated with a vari-
0304-3835/02/$ - see front matter q 2002 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0304-383 5(01)00682-6
198
Y. Ono et al. / Cancer Letters 175 (2002) 197–204
ety of human malignancies, including cancer of the head and neck [3–6]. An abnormal level of EGFR is correlated closely with tumor size and stage in squamous cell carcinoma of the upper aerodigestive tract, suggesting that quantitative estimation of EGFR expression may be of prognostic value for this type of cancer [5,6]. EGFR ligands are present during all stages of wound repair, suggesting that they also play an important role in cell migration [7,8]. EGF has been shown to stimulate the migration and proliferation of both normal and tumor cells, including normal mammary epithelial cells, fibroblasts, renal, prostate and squamous cell carcinoma cells [9–13]. Laminins are a family of glycoproteins found in the extracellular matrix. Their functions include the development and maintenance of the basement membrane, and the regulation of cell adhesion, migration and differentiation [14–16]. Structurally, the basic laminin molecule is a cross-shaped heterotrimer of polypeptide chains, consisting of one heavy chain (a) and two light chains (b and g) [14,15]. Differences in these chains and their combinations produce a variety of laminin isoforms which are tissue-specific and probably have different functions [16]. So far, 11 separate isoforms have been identified [17]. Laminin-5 is composed of a3, b3, g2 chains, of which the g chain is specific to laminin-5 [18–21]. It is a major adhesion protein, serving to anchor epithelial cells on the basement membrane, and plays an important role in cell migration; for instance, it has been found in migrating keratinocytes within healing skin wounds [21,22]. In addition, specific cleavage of the laminin-5 g2 chain (Ln-5 g2) by matrix metalloprotease-2 has been reported to be critical to cell migration during tumor invasion and tissue remodeling [23]. Ln-5 g2 has been reported to be present at the leading edge of various human carcinoma tissues [24– 29]. We have previously reported that increased Ln-5 g2 immunoreactivity, which may reflect a high invasive potential of cancer cells, is a factor indicating a poor prognosis for patients with squamous cell carcinoma of the tongue [30]. With regard to the correlation between Ln-5 g2 and EGF, it has been reported that EGF strongly enhances the gene expression of the three subunits of laminin-5 in human cancer cell lines [31]. Our unpublished data provide evidence that EGFR expression is correlated significantly with Ln-5 g2 expression, and Ln-5 g2
overexpression and EGFR overexpression were evident in tumors showing infiltrative growth, suggesting that EGFR might influence the invasive activity of tumor cells through overexpression of Ln-5 g2. Taken together, interaction between Ln-5 g2 and EGFR is suggested; therefore, in this study, we investigated the relationship between EGFR gene amplification and Ln-5 g2 expression in oral SCC cell lines. 2. Materials and methods 2.1. Cell lines and tissue sample Seven oral SCC cell lines were used in this study: HSC-2, HSC-3, HSC-4, HO-1-N-1, HO-1-u-1, Ca922 and SAS. All were obtained from the Human Science Research Resources Bank (Osaka, Japan). A431 is a well characterized vulvar squamous cell carcinoma known to overproduce EGFR mRNA by 20-100 fold, and was used as a positive control of overexpression and gene amplification [32]. Normal liver tissue obtained from the specimen surgically resected for metastatic liver tumor was used as a control for normal diploid DNA. All cells used were cultured in RPMI 1640 supplemented with 10% fetal calf serum (FCS), penicillin, and streptomycin sulfate. All cell cultures were incubated at 378C in a humidified atmosphere of 95% air 1 5% CO2. 2.2. Compounds TGF-a, EGF and the LA1 monoclonal antibody directed against the extracellular domain of the EGFR were purchased from Upstate Biotechnology (Lake Placid, NY). The tyrophostin AG1478 (BIOMOL Res. Labs., Inc., Plymouth Meeting, PA) used in this study is a protein-tyrosine kinase inhibitor with selectivity for the EGFR signal transduction pathway. Normal mouse IgG1 was acquired from Sigma Chemical Co. (St. Louis, MO). Mouse monoclonal antibody 1–97, specific to the Ln-5 g2, was established in our laboratory [30]. 2.3. Cell culture A total of 1 £ 105 cells were plated in six-well tissue culture plates (Iwaki Glass, Tokyo, Japan) in growth medium (RPMI-1640 supplemented with 10% FCS).
Y. Ono et al. / Cancer Letters 175 (2002) 197–204
After incubation (12 h), the medium was removed, and the cells were washed using fresh serum-free medium supplemented with 0.2% bovine serum albumin (BSA). After 18 h, the cells were washed with 0.2% BSA/RPMI-1640 and incubated in 0.2% BSA/RPMI1640 containing 2.5 mM AG1478 dissolved in 100% DMSO, the equivalent concentration of DMSO (0.1%) served as a control, 0.5 mg/ml LA1, 0.5 mg/ml normal mouse IgG1, 10 ng/ml TGF-a, 2% FCS or 10% FCS, respectively. Two hours later, 10 ng/ml TGF-a or no growth factor was added to the medium containing AG1478, DMSO, LA1 and normal mouse IgG1. Cells were cultured in 0.2 % BSA/RPMI-1640 as a control. After 24 h, RNA or protein was extracted. 2.4. Probes The EGFR cDNA probe was a 2.4-kbp ClaI cDNA insert of pE7 which was obtained from the Human Science Research Resources Bank (Osaka, Japan). The b-actin cDNA probe was obtained from Takara Biomedicals (Otsu, Japan). The cyclin D1 probe was provided by Dr Sasaki, Genetics Division, National Cancer Center Research Institute, Tokyo, Japan. cDNA probes for the Ln-5 g2 chain were obtained by reverse transcriptase-polymerase chain reaction (RT-PCR). Nucleotide sequences of primers for the Ln-5 g2 chain were as reported by Mizushima et al. [31]. Nucleotide sequences of RT-PCR products were confirmed with a 310 Genetic Analyzer (PerkinElmer, Foster City, CA). 2.5. Southern blotting Genomic DNA was isolated according to the standard procedure. Briefly, DNA was extracted from each cell line and normal liver tissue by proteinase K digestion, followed by treatment with phenol and chloroform, and ethanol precipitation. Ten micrograms of each DNA sample was digested with EcoRI overnight at 378C. The restriction fragments obtained were separated by electrophoresis in 0.8% agarose gel and transferred onto Nitro Plus membranes by capillary blotting. Lambda phage DNA digested with HindIII was used as a size marker. The membranes were prehybridized for 2 h and hybridized for 18 h with shaking at 428C in 50% formamide, 0.1 M PIPES-NaOH (pH 6.8), 0.65 M NaCl, 5 mM EDTA (pH 8.0), 0.1 g/ml dextran sulfate, and 0.2
199
mg/ml denatured salmon sperm DNA. Randomprimed 32P-labeled probes (Rediprime II kit, Amersham Pharmacia Biotech, Piscataway, NJ) were used in the hybridization fluid. The membranes were washed and exposed to Kodak XAR-5 film using an intensifying screen. To assess variations in loading and transfer of DNA samples, the membranes of the same blot were rehybridized with a cDNA probe for b-actin, a single-copy gene located on the same chromosome as the EGFR gene. The signal intensity of the EGFR, cyclin D1 and b-actin autoradiograms was measured with an image densitometer (model GS700, Bio Rad Labs., Hercules, CA). The relative amplification level of EGFR and cyclin D1 was calculated as the ratio to b-actin after subtracting the background signals. The degree of amplification was expressed as the ratio of EGFR or cyclin D1/b-actin relative to that of the normal control. 2.6. Northern blotting Total RNA was extracted from human oral SCC cell lines and from human normal liver tissue using TRIZOL reagent (GIBCO BRL, Gaithersburg, MD) according to the manufacturer’s protocol. RNA purity and yield were measured by spectrophotometric absorbance at 260 and 280 nm. The quality of the isolated RNA samples was routinely confirmed by examining the integrity of the 28S and 18S ribosomal RNA bands by electrophoresis on agarose-formaldehyde gels. Total RNA (10 mg/lane) was fractionated by electrophoresis on a formamide-containing 1% agarose gel, transferred to Nitro Plus membranes by capillary blotting, and then cross-linked by exposure to the ultraviolet light of a transilluminator. 2.7. Western blotting Cells were extracted with lysis buffer (10 mM HEPES, pH 7.4, 150 mM NaCl, 1 mM EDTA, 1% Triton X-100, 1% NP-40, 1 mg/ml NaN3, 0.5 mM Na3VO4, 1 mM PMSF, 2 mg/ml leupeptin, 2 mg/ml Pepstatin A) on ice for 30 min before centrifugation (12 000 £ g, 15 min). Cell lysates (10 mg protein) diluted in 2 £ Laemmli sample buffer containing 5% 2-mercaptoethanol were heated for 5 min at 958C, then separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) using 7.5% polyacrylamide gels. Separated proteins were
200
Y. Ono et al. / Cancer Letters 175 (2002) 197–204
transferred to polyvinylidene difluoride (PVDF) membranes (Bio Rad Labs). After incubation with mouse monoclonal antibody 1–97 diluted 1:2000 at 48C overnight, the blots were detected using a horseradish peroxidase-conjugated anti-mouse IgG (IBL, Fujioka, Japan) diluted 1:3000 and ECL Western Blotting Detection Reagents (Amersham Pharmacia Biotech), as instructed by the suppliers. The signal intensity of the Ln-5 g2 autoradiograms was measured
with an image densitometer. The degree of Ln-5 g2 expression was expressed as the ratio of the corrected band density to that of HO-1-N-1.
3. Results Southern blot analysis was used to measure the degree of EGFR gene amplification in seven oral
Fig. 1. (a) EGFR gene amplification in seven human oral squamous cell carcinoma cell lines, A431 (vulvar carcinoma cell line) and normal liver tissue. Lane 1, normal liver tissue; lane 2, HSC-2; lane 3, HSC-3; lane 4, HSC-4; lane 5, HO-1-N-1; lane 6, HO-1-u-1; lane 7, Ca9-22; lane 8, SAS; lane 9, A431. Genomic DNA (10 mg) was digested with EcoRI, size fractionated by 0.8% agarose gel electrophoresis, and transferred to a Nitro Plus membrane. The membrane was subsequently hybridized with a EGFR cDNA probe and autoradiographed. Lambda phage DNA digested with HindIII was used as size marker. Band intensities of EGFR were corrected for differences in DNA loading and transfer by reprobing the blots with cDNA probe for b-actin, a single-copy gene located on the same chromosome as the EGFR gene. The autoradiograms were analyzed by densitometry and the EGFR value was corrected by dividing by the b-actin value. The corrected scanning ratio was then plotted for each individual sample. The degree of amplification of the EGFR gene was expressed as the ratio of the corrected band density to that of normal liver tissue. Molecular masses (in kb) are shown on the left side. kb, kilobases. (b) Cyclin D1 gene amplification in seven human oral squamous cell carcinoma cell lines, A431 and normal liver tissue. Lane 1, normal liver tissue; lane 2, HSC-2; lane 3, HSC-3; lane 4, HSC-4; lane 5, HO-1-N-1; lane 6, HO-1-u-1; lane 7, Ca9-22; lane 8, SAS; lane 9, A431. The corrected scanning ratio was plotted for each individual sample. The degree of amplification for the cyclin D1 gene was expressed as the ratio of the corrected band density to that of normal liver tissue. Molecular masses (in kb) are shown on the left side. kb, kilobases. (c) Immunoblot analysis of Ln-5 g2 in seven human oral squamous cell carcinoma cell lines incubated in serum-free medium supplemented with 0.2% BSA. Lane 2, HSC-2; lane 3, HSC-3; lane 4, HSC-4; lane 5, HO1-N-1; lane 6, HO-1-u-1; lane 7, Ca9-22; lane 8, SAS. Whole-cell lysates (10 mg protein) were separated by 7.5% SDS-PAGE under reducing conditions, transferred to PVDF membranes and immunoblotted with 1–97, the mouse monoclonal antibody against Ln-5 g2; 155- and 105-kDa proteins were detected with 1–97. The degree of Ln-5 g2 expression was expressed as the ratio of the corrected band density to that of HO-1-N1. Molecular masses (in kDa) are shown on the left side. kDa, kiloDaltons.
Y. Ono et al. / Cancer Letters 175 (2002) 197–204
SCC cell lines. EGFR gene amplification was detected in all SCC cell lines at levels ranging from 5 to 50 times that in DNA from normal liver tissue (Fig. 1a). EGFR gene amplification increased proportional to LN-5 g2 chain protein expression, showing close correlation between EGFR gene amplification and Ln-5 g2 chain protein expression under no stimulus (Fig. 1a,c). Cyclin D1 gene amplification was detected in all SCC cell lines at levels ranging from two to five times that in DNA from normal liver tissue (Fig. 1b). However, there was no correlation between cyclin D1 gene amplification and LN-5 g2 chain expression. For the elucidation of causal relationship, we analyzed the effect of TGF-a, tyrosine kinase inhibitor of EGFR, and neutralizing antibody against EGFR, on the expression of Ln-5 g2 in the seven cell lines. mRNA expression of the Ln-5 g2 chain (Fig. 2b) in the seven cell lines was almost correlated with protein expression of those cell lines under no stimulus (Fig.
201
2a,b). Under the stimulus of TGF-a, both protein and mRNA of the Ln-5 g2 chain were highly expressed in HO-1-N-1 and HO-1-u-1 (Fig. 2a,b). These two cell lines showed low levels of EGFR gene amplification (Fig. 1a). In HSC-4, expression of Ln-5 g2 chain protein under the stimulus of 2 and 10% FCS was increased to a greater degree than that under the stimulus of TGF-a (Fig. 2a). In HSC-3 and HO-1N-1, expression of Ln-5 g2 chain protein under the stimulus of 10% FCS was more than that under the stimulus of TGF-a (Fig. 2a). In HO-1-N-1 and HO-1-u-1, expression of Ln-5 g2 chain protein was not inhibited by tyrophostin AG1478 under no stimulus (Fig. 3a). Under the stimulus of TGF-a, expression of Ln-5 g2 chain protein was inhibited by neutralizing antibody against EGFR (LA1) in HO-1-N-1 and weakly in HO-1-u-1 (Fig. 3b). Under the stimulus of TGF-a, expression of Ln-5 g2 chain protein was inhibited by tyrosine kinase
Fig. 2. (a) The effect of 2% FCS (lane 2), 10% FCS (lane 3) and 10 ng/ml TGF-a (lane 4) on the expression of Ln-5 g2 in seven human oral squamous cell carcinoma cell lines was determined by Western blot analysis. As control, cells were cultured in serum-free medium supplemented with 0.2% BSA (lane 1). Molecular masses (in kDa) are shown on the left side. (b) Expression of Ln-5 g2 mRNA in seven human oral squamous cell carcinoma cell lines, A431 (vulvar carcinoma cell line) and HFL1 (lung fibroblast cell line) incubated in the absence (2) or presence (1) of 10 ng/ml TGF-a in serum-free medium supplemented with 0.2% BSA. Total RNA (10 mg) was size-fractionated by agarose gel electrophoresis and transferred to a Nitro Plus membrane. The membrane was subsequently hybridized with Ln-5 g2 cDNA probe and autoradiographed. The blot was rehybridized with a b-actin cDNA probe to verify equal RNA sample loading. Molecular masses (in kb) are shown on the left side. g2, laminin-5 g2 chain.
202
Y. Ono et al. / Cancer Letters 175 (2002) 197–204
inhibitor (AG1478) with selectivity for EGFR signal transduction pathway in all seven oral squamous cell carcinoma cell lines (Fig. 3b).
4. Discussion Cell movement is known to be modulated by signals from the extracellular environment. However, the intracellular signal transduction pathways that lead to this biological response have not been fully clarified. Our present investigations have revealed that growth factors might influence the invasive activity of tumor cells by affecting the structure of extracellular matrix components secreted by tumor cells. Our study has clearly shown that EGFR gene amplification correlates well with Ln-5 g2 chain expression, and that Ln-5 g2 chain expression is regulated by the EGFR signal transduction pathway in some cell lines. In the cell lines HO-1-N-1 and HO-1-u-1, which show low EGFR gene amplification and low expression of the Ln-5 g2 chain with no stimulus, Ln-5 g2 chain expression tended to be high under the stimulation of TGF-a. Therefore, in these cell lines, Ln-5 g2 chain expression is considered to be regulated in a ligand-dependent manner. It is also considered that,
in the cell lines that show a high degree of EGFR gene amplification, EGFR is already fully autophosphorylated. On the other hand, in HSC-3, HSC-4 and HO-1N-1, the expression of Ln-5 g2 chain under the stimulus of 2 and 10% FCS was greater than that under the stimulus of TGF-a, suggesting that Ln-5 g2 chain expression in these cell lines is regulated by a signaling pathway other than that of EGFR as well. In HO-1-N-1 and HO-1-u-1, Ln-5 g2 chain expression was inhibited by neutralizing antibody against EGFR (LA1) under the stimulus of TGF-a, and in all cell lines Ln-5 g2 chain expression was inhibited by the tyrosine kinase inhibitor unique to EGFR (AG1478) under the stimulus of TGF-a. These results indicate that Ln-5 g2 chain expression is regulated by the stimulation of TGF-a, and that the tyrosine kinase phosphorylation of EGFR may be involved in the signaling pathway responsible for regulation of Ln-5 g2 chain expression. In HO-1-N-1 and HO-1-u-1, whose Ln-5 g2 chain expression was not inhibited by LA1 in the absence of stimulus, Ln-5 g2 chain expression increased in the presence of TGF-a. These data suggest that these cell lines autosecrete little or no TGF-a. Comparing with other cell lines HO-1-N-1 and HO-1-u-1 showed a tendency to form cell nests in vitro (data not shown). It would be of
Fig. 3. The effects of 2.5 mM AG1478 dissolved in 100% DMSO, the equivalent concentration of DMSO (0.1%) served as a control, 0.5 mg/ml LA1 and 0.5 mg/ml normal mouse IgG1 on the expression of Ln-5 g2 in seven human oral squamous cell carcinoma cell lines, determined by Western blot analysis. (a) AG1478 (lane 1), DMSO (lane 2), LA1 (lane 3), and normal mouse IgG1 (lane 4), and with (lane 5), or without (lane 6) 10 ng/ml TGF-a were added to the serum-free medium containing 0.2% BSA. (b) AG1478 (lane 1), DMSO (lane 2), LA1 (lane 3), and normal mouse IgG1 (lane 4) were added to the serum-free medium containing 0.2% BSA supplemented with 10 ng/ml TGF a. As a control, cells were cultured in serum-free medium containing 0.2% BSA supplemented with (lane 5), or without (lane 6) 10 ng/ml TGF-a. Molecular masses (in kDa) are shown on the left side.
Y. Ono et al. / Cancer Letters 175 (2002) 197–204
great interest to clarify the correlation between Ln-5 g2 chain expression and its response to TGF-a and the biological behavior of those two cell lines in vitro and in vivo. Mizushima et al. reported that, in a gastric adenocarcinoma cell line (STKM-1) and A431, the levels of Ln-5 g2 chain mRNA increased after treatment with EGF [31]. In this study we showed that EGFR gene amplification closely correlated with Ln-5 g2 chain expression, that Ln-5 g2 chain expression was regulated by stimulation with TGF-a in some cell lines, and that tyrosine phosphorylation of EGFR might be involved in the signaling pathway regulating Ln-5 g2 chain expression. However, at present, the whole mechanisms for the high level of Ln-5 g2 chain expression in cancer cells remain to be elucidated. It has recently been reported that metalloprotease inhibitors reduce cell proliferation in direct proportion to their effects on TGF-a release, and also reduce the growth of EGF-responsive tumorigenic cell lines [32,33]. It has also been reported that constitutive Ln-5 g2 cleavage by both membrane type 1 (MT1)matrix metalloprotease (MMP) and MMP2 or MT1MMP plays an important role in tumor cell migration[34]. Therefore it is suggested that Ln-5 g2 chain expression might also be associated with activation of MMP. In the present study we found that Ln-5 g2 chain expression is closely correlated with EGFR gene amplification in seven oral SCC cell lines. Therefore, in addition to increased cell proliferation, augmented cell motility caused by EGFR gene amplification through overexpression of the LN-5 g2 chain could be one of the reasons for the unfavorable clinical outcome of cases with EGFR gene amplification.
Acknowledgements Source of financial support: Y. Ono is a recipient of a Research Resident Fellowship from the Foundation for Promotion of Cancer Research. This research 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, Labour and Welfare, Japan.
203
References [1] S. Cohen, G. Carpenter, L. King Jr., Epidermal growth factor receptor-protein kinase interactions. Co-purification of receptor and epidermal growth factor-enhanced phosphorylation activity, J. Biol. Chem. 255 (10) (1980) 4834–4842. [2] M. Boni-Schnetzler, P.F. Pilch, Mechanism of epidermal growth factor receptor autophosphorylation and high-affinity binding, Proc. Natl. Acad. Sci. USA 84 (22) (1987) 7832– 7836. [3] T.A. Libermann, N. Razon, A.D. Bartal, Y. Yarden, J. Schlessinger, H. Soreq, Expression of epidermal growth factor receptors in human brain tumors, Cancer Res. 44 (2) (1984) 753–760. [4] D.E. Neal, C. Marsh, M.K. Bennett, P.D. Abel, R.R. Hall, J.R.C. Sainsbury, A.L. Harris, Epidermal growth factor receptors in human bladder cancer: comparison of invasive and superficial tumors, Lancet 1 (1985) 366–368. [5] J. Santini, J.L. Formento, M. Francoual, G. Milano, M. Schneider, O. Dassonville, F. Demard, Characterization, quantification, and potential clinical value of the epidermal growth factor receptor in head and neck squamous cell carcinoma, Head Neck 13 (2) (1991) 132–139. [6] Y. Kitagawa, M. Ueda, N. Ando, S. Ozawa, N. Shimizu, M. Kitajima, Further evidence for prognostic significance of epidermal growth factor gene amplification in patients with esophageal squamous cell carcinoma, Clin. Cancer Res. 2 (5) (1996) 909–914. [7] C.P. Kiritsy, A.B. Lynch, S.E. Lynch, Role of growth factors in cutaneous wound healing: a review, Crit. Rev. Oral Biol. Med. 4 (5) (1993) 729–760. [8] S. Blotnick, G.E. Peoles, M.R. Freeman, T.J. Eberlein, M. Klagsbrun, T lymphocytes synthesize and export heparinbinding epidermal growth factor-like growth factor and basic fibroblast growth factor, mitogens for vascular cells and fibroblasts: differential production and release by CD4 1 and CD8 1 T cells, Proc. Natl. Acad. Sci. USA 91 (8) (1994) 2890–2894. [9] M.A. Matthay, J.P. Thiery, F. Lafont, F. Stampfer, B. Boyer, Transient effect of epidermal growth factor on the motility of an immortalized mammary epithelial cell line, J. Cell Sci. 106 (1993) 869–878. [10] H. Xie, M.A. Pallero, K. Gupta, P. Chang, M.F. Ware, W. Witke, D.J. Kwiatkowski, D.A. Lauffenburger, J.E.M. Ullrich, A. Wells, EGF receptor regulation of cell motility: EGF induces disassembly of focal adhesions independently of the motility-associated PLC g signaling pathway, J. Cell Sci. 111 (1998) 615–624. [11] J.T. Price, H.M. Wilson, N.E. Haites, Epidermal growth factor (EGF) increases the in-vitro invasion, motility and adhesion interactions of the primary renal cell carcinoma cell line, A704, Eur. J. Cancer 32 (11) (1996) 1977–1982. [12] T. Turner, P. Chen, L.J. Goodly, A. Wells, EGF receptor signaling enhances in-vivo invasiveness of DU-145 human prostate carcinoma cells, Clin. Exp. Metastasis 14 (4) (1996) 409–418. [13] I. Rabinovitz, A. Toker, M. Mercurio, Protein kinase C-depen-
204
[14] [15]
[16] [17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
Y. Ono et al. / Cancer Letters 175 (2002) 197–204 dent mobilization of the a6b4 integrin from hemidesmosomes and its association with actin-rich cell protrusions drive the chemotactic migration of carcinoma cells, J. Cell Biol. 146 (5) (1999) 1147–1159. R. Timpl, Structure and biological activity of basement membrane proteins, Eur. J. Biochem. 180 (3) (1989) 487–502. K. Beck, I. Hunter, J. Engel, Structure and function of laminin: anatomy of a multidomain glycoprotein, FASEB J. 4 (2) (1990) 148–160. J. Engel, Laminins and other strange proteins, Biochemistry 31 (44) (1992) 10643–10651. J.H. Miner, B.L. Patton, S.I. Lentz, D.J. Gilbert, W.D. Snider, N.A. Jenkins, N.G. Copeland, J.R. Sanes, The laminin alpha chains: expression, developmental transitions, and chromosomal locations of alpha-5, identification of heterotrimeric laminins 8-11, and cloning of a novel alpha 3 isoform, J. Cell Biol. 137 (3) (1997) 685–701. P. Kallunki, K. Sainio, R. Eddy, M. Byers, T. Kallunki, H. Sariola, K. Beck, H. Hirvonen, T.B. Shows, K. Tryggvason, A truncated laminin chain homologous to the B2 chain: structure, spatial expression, and chromosomal assignment, Cell Biol. 119 (3) (1992) 679–693. P. Rousselle, G.P. Lunstrum, D.R. Keene, R.E. Burgeson, Kalinin: an epithelium-specific basement membrane adhesion molecule that is a component of anchoring filaments, J. Cell Biol. 114 (3) (1991) 567–576. D.R. Gerecke, D.W. Wagman, M.F. Champliaud, R.E. Burgeson, The complete primary structure for a novel laminin chain, the laminin B1k chain, J. Biol. Chem. 269 (15) (1994) 11073– 11080. M.C. Ryan, R. Tizard, D.R. Van Devanter, W.G. Carter, Cloning of the LamA3 gene encoding the alpha 3 chain of the adhesive ligand epiligrin. Expression in wound repair, J. Biol. Chem. 269 (36) (1994) 22779–22787. H. Larjava, T. Salo, K. Haapasalmi, R.H. Kramer, J. Heino, Expression of integrins and basement membrane components by wound keratinocytes, J. Clin. Invest. 92 (3) (1993) 1425– 1435. G. Giannelli, J. Falk-Marzillier, O. Schiraldi, W.G. StetlerStevenson, V. Quaranta, Induction of cell migration by matrix metalloprotease-2 cleavage of laminin-5, Science 277 (1997) 225–228 (J. Clin. Invest. 1993: 92(3): 1425-1435). C. Pyke, J. Romer, P. Kallunki, L.R. Lund, E. Ralfkiar, K. Dano, K. Tryggvason, The gamma 2 chain of kalinin/laminin 5 is preferentially expressed in invading malignant cells in human cancers, Am. J. Pathol. 145 (4) (1994) 782–791.
[25] C. Pyke, S. Salo, E. Ralfkiaer, J. Romer, K. Dano, K. Tryggvason, Laminin-5 is a marker of invading cancer cells in some human carcinomas and is co-expressed with the receptor for urokinase plasminogen activator in budding cancer cells in colon adenocarcinomas, Cancer Res. 55 (18) (1995) 4132– 4139. [26] Y. Soini, M. Maatta, S. Salo, K. Tryggvason, H. AutioHarmainen, Expression of the laminin gamma 2 chain in pancreatic adenocarcinoma, J. Pathol. 180 (3) (1996) 290– 294. [27] M. Maatta, Y. Soini, P. Paakko, S. Salo, K. Tryggvason, H. Autio-Harmainen, Expression of the laminin gamma 2 chain in different histological types of lung carcinoma. A study by immunohistochemistry and in situ hybridization, J. Pathol. 188 (4) (1999) 361–368. [28] B. Skyldberg, S. Salo, E. Eriksson, U. Aspenblad, B. Moberger, K. Tryggvason, G. Auer, Laminin-5 as a marker of invasiveness in cervical lesions, J. Natl. Cancer Inst. 91 (1999) 1882–1887. [29] N. Koshikawa, K. Moriyama, H. Takamura, H. Mizushima, Y. Nagashima, S. Yanoma, K. Miyazaki, Overexpression of laminin g2 chain monomer in invading gastric carcinoma cells, Cancer Res. 59 (1999) 5596–5601. [30] Y. Ono, Y. Nakanishi, Y. Ino, T. Niki, T. Yamada, K. Yoshimura, M. Saikawa, T. Nakajima, S. Hirohashi, Clinicopathologic significance of laminin-5 g2 chain expression in squamous cell carcinoma of the tongue: immunohistochemical analysis of 67 lesions, Cancer 85 (11) (1999) 2315–2321. [31] H. Mizushima, Y. Miyagi, Y. Kikkawa, N. Yamanaka, H. Yasumitsu, K. Misugi, K. Miyazaki, Differential expression of Laminin-5/Ladsin subunits in human tissues and cancer cell lines and their induction by tumor promoter and growth factors, J. Biochem. 120 (1996) 1196–1202. [32] D.J. Giard, S.A. Aaronson, G.J. Todaro, P. Arnstein, J.H. Kersey, H. Dosik, W.P. Parks, In-vitro cultivation of human tumors: establishment of cell lines derived from a series of solid tumors, J. Natl. Cancer Inst. 51 (1973) 1417–1423. [33] J. Dong, L.K. Opresko, P.J. Dempsey, D.A. Lanffenburger, R.J. Coffey, H.S. Wiley, Metalloprotease-mediated ligand release regulates autocrine signaling through the epidermal growth factor receptor, Proc. Natl. Acad. Sci. USA 96 (11) (1999) 6235–6240. [34] N. Koshikawa, G. Giannelli, V. Cirulli, K. Miyazaki, V. Quaranta, Role of cell surface metalloprotease MT1-MMP in epithelial cell migration over laminin-5, J. Cell Biol. 148 (3) (2000) 615–624.
Epidermal growth factor receptor gene amplification is correlated with laminin-5 g2 chain expression in oral squamous cell carcinoma cell lines Yukiko Ono 1, Yukihiro Nakanishi, Masahiro Gotoh, Michiie Sakamoto, Setsuo Hirohashi* Pathology Division, National Cancer Center Research Institute, 1-1, Tsukiji 5-chome, Chuo-ku, Tokyo 104-0045, Japan Received 25 May 2001; received in revised form 25 June 2001; accepted 5 July 2001
Abstract Both epidermal growth factor receptor (EGFR) gene amplification and laminin (Ln)-5 g2 chain overexpression have been reported to be poor prognostic factors in patients with squamous cell carcinoma (SCC) of the head and neck. Here we report our investigation of the relationship between EGFR gene amplification and Ln-5 g2 chain expression in seven SCC cell lines, since both epidermal growth factor (EGF) signaling and Ln-5 g2 have been reported to be involved in cell motility. The degree of correlation between EGFR gene amplification and Ln-5 g2 chain expression was evaluated by Southern and Western blot analyses. EGFR gene amplification was detected in all SCC cell lines at levels 5–50 times those in DNA from normal liver tissue. EGFR gene amplification increased with Ln-5 g2 chain protein expression in seven cell lines, showing close correlation between EGFR gene amplification and Ln-5 g2 chain protein expression. In order to show the causal relationship, we analyzed the effects of transforming growth factor-a (TGF-a), tyrosine kinase inhibitor of EGFR, and neutralizing antibody against EGFR, on the expression of Ln-5 g2 in these cell lines. In two cell lines in which EGFR gene amplification was low , expression of both protein and mRNA of the Ln-5 g2 chain increased in the presence of TGF-a, and Ln-5 g2 chain expression was inhibited by neutralizing antibody against EGFR. In all cell lines, Ln-5 g2 chain expression was inhibited by tyrosine kinase inhibitor which acts selectively on the EGFR signal transduction pathway under the stimulus of TGF-a. These results suggest that EGFR gene amplification and the EGFR signaling pathway can act as positive regulators on the induction of the Ln-5 g2 chain secreted by tumor cells. q 2002 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Epidermal growth factor receptor; Laminin-5 g2 chain; Gene amplification; Overexpression; Squamous cell carcinoma cell line
1. Introduction
* Corresponding author. Tel.: 181-3-3542-2511 ext. 4101; fax: 181-3-3248-2463. E-mail address: [email protected] (S. Hirohashi). 1 Present address: Division of Reconstructive Surgery for Oral and Maxillofacial Region, Department of Tissue Regeneration and Reconstruction, Course for Oral Life Science, Niigata University Graduate School of Medical and Dental Sciences, Niigata, Japan.
Human epidermal growth factor receptor (EGFR) is a 170-kDa transmembrane protein with intrinsic tyrosine kinase activity that regulates cell growth in response to binding of epidermal growth factor (EGF) or transforming growth factor-a (TGF-a), the major ligands for EGFR [1,2]. Overexpression or gene amplification of EGFR has been associated with a vari-
0304-3835/02/$ - see front matter q 2002 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0304-383 5(01)00682-6
198
Y. Ono et al. / Cancer Letters 175 (2002) 197–204
ety of human malignancies, including cancer of the head and neck [3–6]. An abnormal level of EGFR is correlated closely with tumor size and stage in squamous cell carcinoma of the upper aerodigestive tract, suggesting that quantitative estimation of EGFR expression may be of prognostic value for this type of cancer [5,6]. EGFR ligands are present during all stages of wound repair, suggesting that they also play an important role in cell migration [7,8]. EGF has been shown to stimulate the migration and proliferation of both normal and tumor cells, including normal mammary epithelial cells, fibroblasts, renal, prostate and squamous cell carcinoma cells [9–13]. Laminins are a family of glycoproteins found in the extracellular matrix. Their functions include the development and maintenance of the basement membrane, and the regulation of cell adhesion, migration and differentiation [14–16]. Structurally, the basic laminin molecule is a cross-shaped heterotrimer of polypeptide chains, consisting of one heavy chain (a) and two light chains (b and g) [14,15]. Differences in these chains and their combinations produce a variety of laminin isoforms which are tissue-specific and probably have different functions [16]. So far, 11 separate isoforms have been identified [17]. Laminin-5 is composed of a3, b3, g2 chains, of which the g chain is specific to laminin-5 [18–21]. It is a major adhesion protein, serving to anchor epithelial cells on the basement membrane, and plays an important role in cell migration; for instance, it has been found in migrating keratinocytes within healing skin wounds [21,22]. In addition, specific cleavage of the laminin-5 g2 chain (Ln-5 g2) by matrix metalloprotease-2 has been reported to be critical to cell migration during tumor invasion and tissue remodeling [23]. Ln-5 g2 has been reported to be present at the leading edge of various human carcinoma tissues [24– 29]. We have previously reported that increased Ln-5 g2 immunoreactivity, which may reflect a high invasive potential of cancer cells, is a factor indicating a poor prognosis for patients with squamous cell carcinoma of the tongue [30]. With regard to the correlation between Ln-5 g2 and EGF, it has been reported that EGF strongly enhances the gene expression of the three subunits of laminin-5 in human cancer cell lines [31]. Our unpublished data provide evidence that EGFR expression is correlated significantly with Ln-5 g2 expression, and Ln-5 g2
overexpression and EGFR overexpression were evident in tumors showing infiltrative growth, suggesting that EGFR might influence the invasive activity of tumor cells through overexpression of Ln-5 g2. Taken together, interaction between Ln-5 g2 and EGFR is suggested; therefore, in this study, we investigated the relationship between EGFR gene amplification and Ln-5 g2 expression in oral SCC cell lines. 2. Materials and methods 2.1. Cell lines and tissue sample Seven oral SCC cell lines were used in this study: HSC-2, HSC-3, HSC-4, HO-1-N-1, HO-1-u-1, Ca922 and SAS. All were obtained from the Human Science Research Resources Bank (Osaka, Japan). A431 is a well characterized vulvar squamous cell carcinoma known to overproduce EGFR mRNA by 20-100 fold, and was used as a positive control of overexpression and gene amplification [32]. Normal liver tissue obtained from the specimen surgically resected for metastatic liver tumor was used as a control for normal diploid DNA. All cells used were cultured in RPMI 1640 supplemented with 10% fetal calf serum (FCS), penicillin, and streptomycin sulfate. All cell cultures were incubated at 378C in a humidified atmosphere of 95% air 1 5% CO2. 2.2. Compounds TGF-a, EGF and the LA1 monoclonal antibody directed against the extracellular domain of the EGFR were purchased from Upstate Biotechnology (Lake Placid, NY). The tyrophostin AG1478 (BIOMOL Res. Labs., Inc., Plymouth Meeting, PA) used in this study is a protein-tyrosine kinase inhibitor with selectivity for the EGFR signal transduction pathway. Normal mouse IgG1 was acquired from Sigma Chemical Co. (St. Louis, MO). Mouse monoclonal antibody 1–97, specific to the Ln-5 g2, was established in our laboratory [30]. 2.3. Cell culture A total of 1 £ 105 cells were plated in six-well tissue culture plates (Iwaki Glass, Tokyo, Japan) in growth medium (RPMI-1640 supplemented with 10% FCS).
Y. Ono et al. / Cancer Letters 175 (2002) 197–204
After incubation (12 h), the medium was removed, and the cells were washed using fresh serum-free medium supplemented with 0.2% bovine serum albumin (BSA). After 18 h, the cells were washed with 0.2% BSA/RPMI-1640 and incubated in 0.2% BSA/RPMI1640 containing 2.5 mM AG1478 dissolved in 100% DMSO, the equivalent concentration of DMSO (0.1%) served as a control, 0.5 mg/ml LA1, 0.5 mg/ml normal mouse IgG1, 10 ng/ml TGF-a, 2% FCS or 10% FCS, respectively. Two hours later, 10 ng/ml TGF-a or no growth factor was added to the medium containing AG1478, DMSO, LA1 and normal mouse IgG1. Cells were cultured in 0.2 % BSA/RPMI-1640 as a control. After 24 h, RNA or protein was extracted. 2.4. Probes The EGFR cDNA probe was a 2.4-kbp ClaI cDNA insert of pE7 which was obtained from the Human Science Research Resources Bank (Osaka, Japan). The b-actin cDNA probe was obtained from Takara Biomedicals (Otsu, Japan). The cyclin D1 probe was provided by Dr Sasaki, Genetics Division, National Cancer Center Research Institute, Tokyo, Japan. cDNA probes for the Ln-5 g2 chain were obtained by reverse transcriptase-polymerase chain reaction (RT-PCR). Nucleotide sequences of primers for the Ln-5 g2 chain were as reported by Mizushima et al. [31]. Nucleotide sequences of RT-PCR products were confirmed with a 310 Genetic Analyzer (PerkinElmer, Foster City, CA). 2.5. Southern blotting Genomic DNA was isolated according to the standard procedure. Briefly, DNA was extracted from each cell line and normal liver tissue by proteinase K digestion, followed by treatment with phenol and chloroform, and ethanol precipitation. Ten micrograms of each DNA sample was digested with EcoRI overnight at 378C. The restriction fragments obtained were separated by electrophoresis in 0.8% agarose gel and transferred onto Nitro Plus membranes by capillary blotting. Lambda phage DNA digested with HindIII was used as a size marker. The membranes were prehybridized for 2 h and hybridized for 18 h with shaking at 428C in 50% formamide, 0.1 M PIPES-NaOH (pH 6.8), 0.65 M NaCl, 5 mM EDTA (pH 8.0), 0.1 g/ml dextran sulfate, and 0.2
199
mg/ml denatured salmon sperm DNA. Randomprimed 32P-labeled probes (Rediprime II kit, Amersham Pharmacia Biotech, Piscataway, NJ) were used in the hybridization fluid. The membranes were washed and exposed to Kodak XAR-5 film using an intensifying screen. To assess variations in loading and transfer of DNA samples, the membranes of the same blot were rehybridized with a cDNA probe for b-actin, a single-copy gene located on the same chromosome as the EGFR gene. The signal intensity of the EGFR, cyclin D1 and b-actin autoradiograms was measured with an image densitometer (model GS700, Bio Rad Labs., Hercules, CA). The relative amplification level of EGFR and cyclin D1 was calculated as the ratio to b-actin after subtracting the background signals. The degree of amplification was expressed as the ratio of EGFR or cyclin D1/b-actin relative to that of the normal control. 2.6. Northern blotting Total RNA was extracted from human oral SCC cell lines and from human normal liver tissue using TRIZOL reagent (GIBCO BRL, Gaithersburg, MD) according to the manufacturer’s protocol. RNA purity and yield were measured by spectrophotometric absorbance at 260 and 280 nm. The quality of the isolated RNA samples was routinely confirmed by examining the integrity of the 28S and 18S ribosomal RNA bands by electrophoresis on agarose-formaldehyde gels. Total RNA (10 mg/lane) was fractionated by electrophoresis on a formamide-containing 1% agarose gel, transferred to Nitro Plus membranes by capillary blotting, and then cross-linked by exposure to the ultraviolet light of a transilluminator. 2.7. Western blotting Cells were extracted with lysis buffer (10 mM HEPES, pH 7.4, 150 mM NaCl, 1 mM EDTA, 1% Triton X-100, 1% NP-40, 1 mg/ml NaN3, 0.5 mM Na3VO4, 1 mM PMSF, 2 mg/ml leupeptin, 2 mg/ml Pepstatin A) on ice for 30 min before centrifugation (12 000 £ g, 15 min). Cell lysates (10 mg protein) diluted in 2 £ Laemmli sample buffer containing 5% 2-mercaptoethanol were heated for 5 min at 958C, then separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) using 7.5% polyacrylamide gels. Separated proteins were
200
Y. Ono et al. / Cancer Letters 175 (2002) 197–204
transferred to polyvinylidene difluoride (PVDF) membranes (Bio Rad Labs). After incubation with mouse monoclonal antibody 1–97 diluted 1:2000 at 48C overnight, the blots were detected using a horseradish peroxidase-conjugated anti-mouse IgG (IBL, Fujioka, Japan) diluted 1:3000 and ECL Western Blotting Detection Reagents (Amersham Pharmacia Biotech), as instructed by the suppliers. The signal intensity of the Ln-5 g2 autoradiograms was measured
with an image densitometer. The degree of Ln-5 g2 expression was expressed as the ratio of the corrected band density to that of HO-1-N-1.
3. Results Southern blot analysis was used to measure the degree of EGFR gene amplification in seven oral
Fig. 1. (a) EGFR gene amplification in seven human oral squamous cell carcinoma cell lines, A431 (vulvar carcinoma cell line) and normal liver tissue. Lane 1, normal liver tissue; lane 2, HSC-2; lane 3, HSC-3; lane 4, HSC-4; lane 5, HO-1-N-1; lane 6, HO-1-u-1; lane 7, Ca9-22; lane 8, SAS; lane 9, A431. Genomic DNA (10 mg) was digested with EcoRI, size fractionated by 0.8% agarose gel electrophoresis, and transferred to a Nitro Plus membrane. The membrane was subsequently hybridized with a EGFR cDNA probe and autoradiographed. Lambda phage DNA digested with HindIII was used as size marker. Band intensities of EGFR were corrected for differences in DNA loading and transfer by reprobing the blots with cDNA probe for b-actin, a single-copy gene located on the same chromosome as the EGFR gene. The autoradiograms were analyzed by densitometry and the EGFR value was corrected by dividing by the b-actin value. The corrected scanning ratio was then plotted for each individual sample. The degree of amplification of the EGFR gene was expressed as the ratio of the corrected band density to that of normal liver tissue. Molecular masses (in kb) are shown on the left side. kb, kilobases. (b) Cyclin D1 gene amplification in seven human oral squamous cell carcinoma cell lines, A431 and normal liver tissue. Lane 1, normal liver tissue; lane 2, HSC-2; lane 3, HSC-3; lane 4, HSC-4; lane 5, HO-1-N-1; lane 6, HO-1-u-1; lane 7, Ca9-22; lane 8, SAS; lane 9, A431. The corrected scanning ratio was plotted for each individual sample. The degree of amplification for the cyclin D1 gene was expressed as the ratio of the corrected band density to that of normal liver tissue. Molecular masses (in kb) are shown on the left side. kb, kilobases. (c) Immunoblot analysis of Ln-5 g2 in seven human oral squamous cell carcinoma cell lines incubated in serum-free medium supplemented with 0.2% BSA. Lane 2, HSC-2; lane 3, HSC-3; lane 4, HSC-4; lane 5, HO1-N-1; lane 6, HO-1-u-1; lane 7, Ca9-22; lane 8, SAS. Whole-cell lysates (10 mg protein) were separated by 7.5% SDS-PAGE under reducing conditions, transferred to PVDF membranes and immunoblotted with 1–97, the mouse monoclonal antibody against Ln-5 g2; 155- and 105-kDa proteins were detected with 1–97. The degree of Ln-5 g2 expression was expressed as the ratio of the corrected band density to that of HO-1-N1. Molecular masses (in kDa) are shown on the left side. kDa, kiloDaltons.
Y. Ono et al. / Cancer Letters 175 (2002) 197–204
SCC cell lines. EGFR gene amplification was detected in all SCC cell lines at levels ranging from 5 to 50 times that in DNA from normal liver tissue (Fig. 1a). EGFR gene amplification increased proportional to LN-5 g2 chain protein expression, showing close correlation between EGFR gene amplification and Ln-5 g2 chain protein expression under no stimulus (Fig. 1a,c). Cyclin D1 gene amplification was detected in all SCC cell lines at levels ranging from two to five times that in DNA from normal liver tissue (Fig. 1b). However, there was no correlation between cyclin D1 gene amplification and LN-5 g2 chain expression. For the elucidation of causal relationship, we analyzed the effect of TGF-a, tyrosine kinase inhibitor of EGFR, and neutralizing antibody against EGFR, on the expression of Ln-5 g2 in the seven cell lines. mRNA expression of the Ln-5 g2 chain (Fig. 2b) in the seven cell lines was almost correlated with protein expression of those cell lines under no stimulus (Fig.
201
2a,b). Under the stimulus of TGF-a, both protein and mRNA of the Ln-5 g2 chain were highly expressed in HO-1-N-1 and HO-1-u-1 (Fig. 2a,b). These two cell lines showed low levels of EGFR gene amplification (Fig. 1a). In HSC-4, expression of Ln-5 g2 chain protein under the stimulus of 2 and 10% FCS was increased to a greater degree than that under the stimulus of TGF-a (Fig. 2a). In HSC-3 and HO-1N-1, expression of Ln-5 g2 chain protein under the stimulus of 10% FCS was more than that under the stimulus of TGF-a (Fig. 2a). In HO-1-N-1 and HO-1-u-1, expression of Ln-5 g2 chain protein was not inhibited by tyrophostin AG1478 under no stimulus (Fig. 3a). Under the stimulus of TGF-a, expression of Ln-5 g2 chain protein was inhibited by neutralizing antibody against EGFR (LA1) in HO-1-N-1 and weakly in HO-1-u-1 (Fig. 3b). Under the stimulus of TGF-a, expression of Ln-5 g2 chain protein was inhibited by tyrosine kinase
Fig. 2. (a) The effect of 2% FCS (lane 2), 10% FCS (lane 3) and 10 ng/ml TGF-a (lane 4) on the expression of Ln-5 g2 in seven human oral squamous cell carcinoma cell lines was determined by Western blot analysis. As control, cells were cultured in serum-free medium supplemented with 0.2% BSA (lane 1). Molecular masses (in kDa) are shown on the left side. (b) Expression of Ln-5 g2 mRNA in seven human oral squamous cell carcinoma cell lines, A431 (vulvar carcinoma cell line) and HFL1 (lung fibroblast cell line) incubated in the absence (2) or presence (1) of 10 ng/ml TGF-a in serum-free medium supplemented with 0.2% BSA. Total RNA (10 mg) was size-fractionated by agarose gel electrophoresis and transferred to a Nitro Plus membrane. The membrane was subsequently hybridized with Ln-5 g2 cDNA probe and autoradiographed. The blot was rehybridized with a b-actin cDNA probe to verify equal RNA sample loading. Molecular masses (in kb) are shown on the left side. g2, laminin-5 g2 chain.
202
Y. Ono et al. / Cancer Letters 175 (2002) 197–204
inhibitor (AG1478) with selectivity for EGFR signal transduction pathway in all seven oral squamous cell carcinoma cell lines (Fig. 3b).
4. Discussion Cell movement is known to be modulated by signals from the extracellular environment. However, the intracellular signal transduction pathways that lead to this biological response have not been fully clarified. Our present investigations have revealed that growth factors might influence the invasive activity of tumor cells by affecting the structure of extracellular matrix components secreted by tumor cells. Our study has clearly shown that EGFR gene amplification correlates well with Ln-5 g2 chain expression, and that Ln-5 g2 chain expression is regulated by the EGFR signal transduction pathway in some cell lines. In the cell lines HO-1-N-1 and HO-1-u-1, which show low EGFR gene amplification and low expression of the Ln-5 g2 chain with no stimulus, Ln-5 g2 chain expression tended to be high under the stimulation of TGF-a. Therefore, in these cell lines, Ln-5 g2 chain expression is considered to be regulated in a ligand-dependent manner. It is also considered that,
in the cell lines that show a high degree of EGFR gene amplification, EGFR is already fully autophosphorylated. On the other hand, in HSC-3, HSC-4 and HO-1N-1, the expression of Ln-5 g2 chain under the stimulus of 2 and 10% FCS was greater than that under the stimulus of TGF-a, suggesting that Ln-5 g2 chain expression in these cell lines is regulated by a signaling pathway other than that of EGFR as well. In HO-1-N-1 and HO-1-u-1, Ln-5 g2 chain expression was inhibited by neutralizing antibody against EGFR (LA1) under the stimulus of TGF-a, and in all cell lines Ln-5 g2 chain expression was inhibited by the tyrosine kinase inhibitor unique to EGFR (AG1478) under the stimulus of TGF-a. These results indicate that Ln-5 g2 chain expression is regulated by the stimulation of TGF-a, and that the tyrosine kinase phosphorylation of EGFR may be involved in the signaling pathway responsible for regulation of Ln-5 g2 chain expression. In HO-1-N-1 and HO-1-u-1, whose Ln-5 g2 chain expression was not inhibited by LA1 in the absence of stimulus, Ln-5 g2 chain expression increased in the presence of TGF-a. These data suggest that these cell lines autosecrete little or no TGF-a. Comparing with other cell lines HO-1-N-1 and HO-1-u-1 showed a tendency to form cell nests in vitro (data not shown). It would be of
Fig. 3. The effects of 2.5 mM AG1478 dissolved in 100% DMSO, the equivalent concentration of DMSO (0.1%) served as a control, 0.5 mg/ml LA1 and 0.5 mg/ml normal mouse IgG1 on the expression of Ln-5 g2 in seven human oral squamous cell carcinoma cell lines, determined by Western blot analysis. (a) AG1478 (lane 1), DMSO (lane 2), LA1 (lane 3), and normal mouse IgG1 (lane 4), and with (lane 5), or without (lane 6) 10 ng/ml TGF-a were added to the serum-free medium containing 0.2% BSA. (b) AG1478 (lane 1), DMSO (lane 2), LA1 (lane 3), and normal mouse IgG1 (lane 4) were added to the serum-free medium containing 0.2% BSA supplemented with 10 ng/ml TGF a. As a control, cells were cultured in serum-free medium containing 0.2% BSA supplemented with (lane 5), or without (lane 6) 10 ng/ml TGF-a. Molecular masses (in kDa) are shown on the left side.
Y. Ono et al. / Cancer Letters 175 (2002) 197–204
great interest to clarify the correlation between Ln-5 g2 chain expression and its response to TGF-a and the biological behavior of those two cell lines in vitro and in vivo. Mizushima et al. reported that, in a gastric adenocarcinoma cell line (STKM-1) and A431, the levels of Ln-5 g2 chain mRNA increased after treatment with EGF [31]. In this study we showed that EGFR gene amplification closely correlated with Ln-5 g2 chain expression, that Ln-5 g2 chain expression was regulated by stimulation with TGF-a in some cell lines, and that tyrosine phosphorylation of EGFR might be involved in the signaling pathway regulating Ln-5 g2 chain expression. However, at present, the whole mechanisms for the high level of Ln-5 g2 chain expression in cancer cells remain to be elucidated. It has recently been reported that metalloprotease inhibitors reduce cell proliferation in direct proportion to their effects on TGF-a release, and also reduce the growth of EGF-responsive tumorigenic cell lines [32,33]. It has also been reported that constitutive Ln-5 g2 cleavage by both membrane type 1 (MT1)matrix metalloprotease (MMP) and MMP2 or MT1MMP plays an important role in tumor cell migration[34]. Therefore it is suggested that Ln-5 g2 chain expression might also be associated with activation of MMP. In the present study we found that Ln-5 g2 chain expression is closely correlated with EGFR gene amplification in seven oral SCC cell lines. Therefore, in addition to increased cell proliferation, augmented cell motility caused by EGFR gene amplification through overexpression of the LN-5 g2 chain could be one of the reasons for the unfavorable clinical outcome of cases with EGFR gene amplification.
Acknowledgements Source of financial support: Y. Ono is a recipient of a Research Resident Fellowship from the Foundation for Promotion of Cancer Research. This research 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, Labour and Welfare, Japan.
203
References [1] S. Cohen, G. Carpenter, L. King Jr., Epidermal growth factor receptor-protein kinase interactions. Co-purification of receptor and epidermal growth factor-enhanced phosphorylation activity, J. Biol. Chem. 255 (10) (1980) 4834–4842. [2] M. Boni-Schnetzler, P.F. Pilch, Mechanism of epidermal growth factor receptor autophosphorylation and high-affinity binding, Proc. Natl. Acad. Sci. USA 84 (22) (1987) 7832– 7836. [3] T.A. Libermann, N. Razon, A.D. Bartal, Y. Yarden, J. Schlessinger, H. Soreq, Expression of epidermal growth factor receptors in human brain tumors, Cancer Res. 44 (2) (1984) 753–760. [4] D.E. Neal, C. Marsh, M.K. Bennett, P.D. Abel, R.R. Hall, J.R.C. Sainsbury, A.L. Harris, Epidermal growth factor receptors in human bladder cancer: comparison of invasive and superficial tumors, Lancet 1 (1985) 366–368. [5] J. Santini, J.L. Formento, M. Francoual, G. Milano, M. Schneider, O. Dassonville, F. Demard, Characterization, quantification, and potential clinical value of the epidermal growth factor receptor in head and neck squamous cell carcinoma, Head Neck 13 (2) (1991) 132–139. [6] Y. Kitagawa, M. Ueda, N. Ando, S. Ozawa, N. Shimizu, M. Kitajima, Further evidence for prognostic significance of epidermal growth factor gene amplification in patients with esophageal squamous cell carcinoma, Clin. Cancer Res. 2 (5) (1996) 909–914. [7] C.P. Kiritsy, A.B. Lynch, S.E. Lynch, Role of growth factors in cutaneous wound healing: a review, Crit. Rev. Oral Biol. Med. 4 (5) (1993) 729–760. [8] S. Blotnick, G.E. Peoles, M.R. Freeman, T.J. Eberlein, M. Klagsbrun, T lymphocytes synthesize and export heparinbinding epidermal growth factor-like growth factor and basic fibroblast growth factor, mitogens for vascular cells and fibroblasts: differential production and release by CD4 1 and CD8 1 T cells, Proc. Natl. Acad. Sci. USA 91 (8) (1994) 2890–2894. [9] M.A. Matthay, J.P. Thiery, F. Lafont, F. Stampfer, B. Boyer, Transient effect of epidermal growth factor on the motility of an immortalized mammary epithelial cell line, J. Cell Sci. 106 (1993) 869–878. [10] H. Xie, M.A. Pallero, K. Gupta, P. Chang, M.F. Ware, W. Witke, D.J. Kwiatkowski, D.A. Lauffenburger, J.E.M. Ullrich, A. Wells, EGF receptor regulation of cell motility: EGF induces disassembly of focal adhesions independently of the motility-associated PLC g signaling pathway, J. Cell Sci. 111 (1998) 615–624. [11] J.T. Price, H.M. Wilson, N.E. Haites, Epidermal growth factor (EGF) increases the in-vitro invasion, motility and adhesion interactions of the primary renal cell carcinoma cell line, A704, Eur. J. Cancer 32 (11) (1996) 1977–1982. [12] T. Turner, P. Chen, L.J. Goodly, A. Wells, EGF receptor signaling enhances in-vivo invasiveness of DU-145 human prostate carcinoma cells, Clin. Exp. Metastasis 14 (4) (1996) 409–418. [13] I. Rabinovitz, A. Toker, M. Mercurio, Protein kinase C-depen-
204
[14] [15]
[16] [17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
Y. Ono et al. / Cancer Letters 175 (2002) 197–204 dent mobilization of the a6b4 integrin from hemidesmosomes and its association with actin-rich cell protrusions drive the chemotactic migration of carcinoma cells, J. Cell Biol. 146 (5) (1999) 1147–1159. R. Timpl, Structure and biological activity of basement membrane proteins, Eur. J. Biochem. 180 (3) (1989) 487–502. K. Beck, I. Hunter, J. Engel, Structure and function of laminin: anatomy of a multidomain glycoprotein, FASEB J. 4 (2) (1990) 148–160. J. Engel, Laminins and other strange proteins, Biochemistry 31 (44) (1992) 10643–10651. J.H. Miner, B.L. Patton, S.I. Lentz, D.J. Gilbert, W.D. Snider, N.A. Jenkins, N.G. Copeland, J.R. Sanes, The laminin alpha chains: expression, developmental transitions, and chromosomal locations of alpha-5, identification of heterotrimeric laminins 8-11, and cloning of a novel alpha 3 isoform, J. Cell Biol. 137 (3) (1997) 685–701. P. Kallunki, K. Sainio, R. Eddy, M. Byers, T. Kallunki, H. Sariola, K. Beck, H. Hirvonen, T.B. Shows, K. Tryggvason, A truncated laminin chain homologous to the B2 chain: structure, spatial expression, and chromosomal assignment, Cell Biol. 119 (3) (1992) 679–693. P. Rousselle, G.P. Lunstrum, D.R. Keene, R.E. Burgeson, Kalinin: an epithelium-specific basement membrane adhesion molecule that is a component of anchoring filaments, J. Cell Biol. 114 (3) (1991) 567–576. D.R. Gerecke, D.W. Wagman, M.F. Champliaud, R.E. Burgeson, The complete primary structure for a novel laminin chain, the laminin B1k chain, J. Biol. Chem. 269 (15) (1994) 11073– 11080. M.C. Ryan, R. Tizard, D.R. Van Devanter, W.G. Carter, Cloning of the LamA3 gene encoding the alpha 3 chain of the adhesive ligand epiligrin. Expression in wound repair, J. Biol. Chem. 269 (36) (1994) 22779–22787. H. Larjava, T. Salo, K. Haapasalmi, R.H. Kramer, J. Heino, Expression of integrins and basement membrane components by wound keratinocytes, J. Clin. Invest. 92 (3) (1993) 1425– 1435. G. Giannelli, J. Falk-Marzillier, O. Schiraldi, W.G. StetlerStevenson, V. Quaranta, Induction of cell migration by matrix metalloprotease-2 cleavage of laminin-5, Science 277 (1997) 225–228 (J. Clin. Invest. 1993: 92(3): 1425-1435). C. Pyke, J. Romer, P. Kallunki, L.R. Lund, E. Ralfkiar, K. Dano, K. Tryggvason, The gamma 2 chain of kalinin/laminin 5 is preferentially expressed in invading malignant cells in human cancers, Am. J. Pathol. 145 (4) (1994) 782–791.
[25] C. Pyke, S. Salo, E. Ralfkiaer, J. Romer, K. Dano, K. Tryggvason, Laminin-5 is a marker of invading cancer cells in some human carcinomas and is co-expressed with the receptor for urokinase plasminogen activator in budding cancer cells in colon adenocarcinomas, Cancer Res. 55 (18) (1995) 4132– 4139. [26] Y. Soini, M. Maatta, S. Salo, K. Tryggvason, H. AutioHarmainen, Expression of the laminin gamma 2 chain in pancreatic adenocarcinoma, J. Pathol. 180 (3) (1996) 290– 294. [27] M. Maatta, Y. Soini, P. Paakko, S. Salo, K. Tryggvason, H. Autio-Harmainen, Expression of the laminin gamma 2 chain in different histological types of lung carcinoma. A study by immunohistochemistry and in situ hybridization, J. Pathol. 188 (4) (1999) 361–368. [28] B. Skyldberg, S. Salo, E. Eriksson, U. Aspenblad, B. Moberger, K. Tryggvason, G. Auer, Laminin-5 as a marker of invasiveness in cervical lesions, J. Natl. Cancer Inst. 91 (1999) 1882–1887. [29] N. Koshikawa, K. Moriyama, H. Takamura, H. Mizushima, Y. Nagashima, S. Yanoma, K. Miyazaki, Overexpression of laminin g2 chain monomer in invading gastric carcinoma cells, Cancer Res. 59 (1999) 5596–5601. [30] Y. Ono, Y. Nakanishi, Y. Ino, T. Niki, T. Yamada, K. Yoshimura, M. Saikawa, T. Nakajima, S. Hirohashi, Clinicopathologic significance of laminin-5 g2 chain expression in squamous cell carcinoma of the tongue: immunohistochemical analysis of 67 lesions, Cancer 85 (11) (1999) 2315–2321. [31] H. Mizushima, Y. Miyagi, Y. Kikkawa, N. Yamanaka, H. Yasumitsu, K. Misugi, K. Miyazaki, Differential expression of Laminin-5/Ladsin subunits in human tissues and cancer cell lines and their induction by tumor promoter and growth factors, J. Biochem. 120 (1996) 1196–1202. [32] D.J. Giard, S.A. Aaronson, G.J. Todaro, P. Arnstein, J.H. Kersey, H. Dosik, W.P. Parks, In-vitro cultivation of human tumors: establishment of cell lines derived from a series of solid tumors, J. Natl. Cancer Inst. 51 (1973) 1417–1423. [33] J. Dong, L.K. Opresko, P.J. Dempsey, D.A. Lanffenburger, R.J. Coffey, H.S. Wiley, Metalloprotease-mediated ligand release regulates autocrine signaling through the epidermal growth factor receptor, Proc. Natl. Acad. Sci. USA 96 (11) (1999) 6235–6240. [34] N. Koshikawa, G. Giannelli, V. Cirulli, K. Miyazaki, V. Quaranta, Role of cell surface metalloprotease MT1-MMP in epithelial cell migration over laminin-5, J. Cell Biol. 148 (3) (2000) 615–624.