Granulocyte-macrophage colony-stimulating factor upregulates matrix metalloproteinase-2 (MMP-2) and membrane type-1 MMP (MT1-MMP) in human head and neck cancer cells

Cancer Letters 156 (2000) 83±91

www.elsevier.com/locate/canlet

Granulocyte-macrophage colony-stimulating factor upregulates matrix metalloproteinase-2 (MMP-2) and membrane type-1 MMP (MT1-MMP) in human head and neck cancer cells Toshiki Tomita a, Masato Fujii a,*, Yutaka Tokumaru a, Yorihisa Imanishi a, Minoru Kanke a, Taku Yamashita a, Ryuichiro Ishiguro a, Jin Kanzaki a, Kaori Kameyama b, Yoshihide Otani c a

Department of Otolaryngology, Keio University, School of Medicine, 35 Shinanomachi, Shinjuku-ku, Tokyo 160-8582, Japan b Department of Pathology, Keio University, School of Medicine, Tokyo, Japan c Department of Surgery, Keio University, School of Medicine, Tokyo, Japan Received 19 January 2000; received in revised form 4 April 2000; accepted 4 April 2000

Abstract Matrix metalloproteinase-2 (MMP-2) and membrane type 1-MMP (MT1-MMP) play an important role in the invasion and metastasis of head and neck squamous cell carcinoma (HNSCC), but the mechanism of their regulation is not clearly understood. Recently, granulocyte-macrophage colony-stimulating factor (GM-CSF) has been shown to be associated with cancer invasion and metastasis. We hypothesized that GM-CSF may upregulate MMP-2 and/or MT1-MMP expression in HNSCC cells, and may thereby in¯uence their ability to invade and metastasize. We studied the effects of GM-CSF on the production of MMP-2 and MT1-MMP in HNSCC cell lines SAS and HSC-2. Gelatin zymography of conditioned media derived from HNSCC cells revealed a major band of 68 kDa, which was characterized as proMMP-2. GM-CSF stimulated the production of proMMP-2 in both cell lines in a dose-dependent manner. Treatment with 50 ng/ml GM-CSF for 24 h increased the proMMP2 activity 3.4-fold in SAS cells and 2.3-fold in HSC-2 cells compared with untreated controls. Northern blot analyses demonstrated that GM-CSF led to elevated mRNA levels of MMP-2 and MT1-MMP in both cell lines. The results identify GM-CSF as a regulator of MMP-2 and MT1-MMP expression in certain types of HNSCC, and suggest that GM-CSF may contribute to the invasiveness of HNSCC through the regulation of MMP-2 and MT1-MMP expression. q 2000 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Head and neck squamous cell carcinoma; Granulocyte-macrophage colony-stimulating factor; Matrix metalloproteinase-2; Membrane type 1 matrix metalloproteinase

1. Introduction In the progression of cancer metastasis, various enzymes such as matrix metalloproteinases (MMPs), * Corresponding author. Tel.: 181-3-3353-1211 ext. (62897); fax: 181-3-5379-0335. E-mail address: [email protected] (M. Fujii).

serine proteinase and cystine proteinase are necessary to degrade the extracellular matrix (ECM) and basement membrane. Especially, MMP-2 activity correlates well with the invasiveness and metastatic potential of many tumors in vivo and in vitro [1,2]. Membrane type 1-MMP (MT1-MMP) has been cloned and characterized as a speci®c activator of proMMP-2 [3]. Moreover, recent studies have

0304-3835/00/$ - see front matter q 2000 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0304-383 5(00)00446-8

84

T. Tomita et al. / Cancer Letters 156 (2000) 83±91

shown that MT1-MMP itself can degrade the extracellular matrix [4,5]. The expression of MT1-MMP is also implicated in the invasion and metastasis of many tumors in vivo [6±8]. Several observations indicate that paracrine or autocrine growth factors, as well as cancer cell±host tissue interactions, can modulate the expression of genes involved in cancer invasion and metastasis [9]. Granulocyte-macrophage colony-stimulating factor (GM-CSF), a glycoprotein growth factor, in¯uences myelopoiesis by stimulating the differentiation of stem cells to produce granulocytes and monocytes [10]. It has recently been reported to correlate with cancer invasion and metastasis [11±18]. However, the mechanism is under investigation. GM-CSF has been demonstrated to be present in a homogenate of human head and neck squamous cell carcinoma (HNSCC) tissue [18,19]. We hypothesized that GM-CSF secreted in a paracrine or autocrine fashion may regulate MMP-2 and/or MT1-MMP production to modulate the invasion and metastasis of cancer cells. In the present study, we examined the effects of GM-CSF on the expression of proMMP-2 and MT1-MMP in HNSCC cell lines. 2. Materials and methods 2.1. Cell culture and culture conditions The human oral squamous cell carcinoma cell lines, SAS and HSC-2, were provided by the Japanese Collection of Research Bioresources (JCRB), Tokyo, Japan. SAS cells were established from a primary tongue carcinoma [20] and HSC-2 cells from a metastatic lymph node in ¯oor of the mouth carcinoma [21]. SAS and HSC-2 were maintained in Dulbecco's modi®ed Eagle's medium (DMEM) with 45% Ham's F-12 medium and Eagle's minimal essential medium (MEM), respectively, containing 10% fetal bovine serum, 100 IU/ml penicillin, 100 mg/ml streptomycin, and 250 ng/ml amphotericin B. Cells were kept at 378C, in a humidi®ed atmosphere of 95% air and 5% CO2, with medium replacement every 3 days. 2.2. Biological reagents Recombinant human GM-CSF (rhGM-CSF) was purchased from the Anapure Bioscienti®c Co. Ltd.

(Beijing, China). The speci®c activity was 1 £ 107 units/mg. Monoclonal mouse antihuman GM-CSF receptor a, S-50, was purchased from Santa Cruz Biotechnology (Santa Cruz, CA). The GM-CSF receptor consists of an 80-kDa a subunit and a 120kDa b subunit [22]. The b subunit is common in other cytokine receptors, such as interleukin-3 and interleukin-5, but the a subunit is unique to the GM-CSF receptor. We therefore chose S-50 which reacts with the 80-kDa GM-CSF receptor a chain. DMEM/F-12, MEM, and antibiotic were purchased from Gibco BRL. 2.3. Immunocytochemistry Subcon¯uent HNSCC cells on a glass slide were ®xed with 1% formalin for 10 min, treated with 0.3% H2O2 for 15 min, and incubated in normal blocking serum in PBS for 20 min at room temperature. They were incubated overnight at 48C in 1% BSA± PBS and antihuman GM-CSF receptor a, S-50. After washing with PBS, they were incubated with biotinconjugated secondary antibody for 30 min at room temperature. After further washing, they were treated with streptavidin±biotin peroxidase complex for 30 min at room temperature. Diaminobenzidine tetrahydrochloride was used as the chromogen, and the cells were lightly counterstained with hematoxylin. For negative controls, cells were similarly processed without the primary antibodies. 2.4. MTT assay Because GM-CSF has been reported to be a growth factor for several tumor cell lines [23,24], we tested the effects of GM-CSF on the growth of HNSCC cell lines. Cell proliferation was assessed by a tetrazoliumdye-based assay (MTT assay). For the MTT assay, cells were seeded and cultured on a 96-well multititer plate for 24 h in regular culture medium, and then washed and re-fed with serum-free medium in the absence or presence of GM-CSF in varying concentrations (0, 5, 10, 50 ng/ml). After 24 h, 10 ml/well of 5 mg/ml of MTT (3-(4,5 dimethylthiazol-2-yl)-2,5 diphenyl-tetrazolium bromide; Sigma Chemical Co., St. Louis, MO) diluted by phosphate-buffered saline (pH 7.4) was added. The media were aspirated and 100 ml of acid-isopropanol was applied after 4 h of additional incubation. The plates were then placed on

T. Tomita et al. / Cancer Letters 156 (2000) 83±91

85

a shaker for 2 min and read at 570 nm. Each assay was performed in quadruplicate. Differences between values were analyzed by the Fisher's protected least signi®cant difference test. Doses of GM-CSF chosen were based on previous studies [12,13,15,16].

Mannheim, Germany) according to the manufacturer's instructions. Transcribed products were identi®ed by following gel electrophoresis and quanti®ed by direct colorimetric detection using a labeling detection kit (Boehringer Mannheim) [26].

2.5. Gelatin zymography

2.7. Northern blot analysis

A total of 5 £ 104 cells/well were seeded on a 24well culture plate. After incubation for 24 h in regular culture media, the cells were washed and re-fed with serum free medium in the absence or presence of GMCSF in varying concentrations. After 24 h, culture supernatants were collected, cleared of debris by centrifugation at 3000 rev./min for 10 min and stored at 2208C. Zymography was performed according to the modi®ed method of Kawashima et al. [25]. Brie¯y, samples of culture media were mixed with SDS sample buffer in the absence of any reducing agent, incubated at 378C for 30 min, and electrophoresed on 10% SDS±polyacrylamide gels copolymerized with 0.2% gelatin at 48C. After electrophoresis, gels were rinsed twice in 2.5% Triton X-100 and incubated at 378C for 17 h in 0.15 M NaCl, 10 mM CaCl2 and 50 mM Tris±HCl buffer (pH 7.5). Gels were stained with 0.1% Coomassie blue R250, and destained in 5% isopropanol and 8% acetic acid in H2O. Gelatinolytic enzymes were detected as transparent bands on the blue background of Coomassie blue-stained gel. The intensity of each gelatinolytic band was semi-quanti®ed by NIH Image version 1.61 software.

After incubation for 24 h in regular culture medium, cultures were washed and re-fed with serum-free medium with or without 50 ng/ml GM-CSF. After 24 h, total RNA was extracted from cells using ISOGEN-LS RNA isolation reagent as recommended by the supplier (Nippon Gene, Tokyo). Total RNA (10 mg/lane) was subjected to 1% agarose gel electrophoresis with formaldehyde and transferred to a positively charged nylon membrane (Boehringer Mannheim) by capillary blotting with 10 £ SSC overnight. The blotted RNAs were UV-crosslinked (Stratalinker II, Stratagene) to the membrane. The blots were hybridized with DIG-labeled antisense RNA probe. Finally, immunodetection was carried out using DIG Luminescent detection kit (Boehringer Mannheim) following the manufacturer's instructions. The hybridization signals were visualized after exposure on Kodak BioMax MS ®lm for 15 min at room temperature. For quantitative analysis, the intensity of the hybridization signals of MMP-2 and MT1-MMP were semi-quanti®ed and normalized to the corresponding 28S rRNA levels as an internal control visualized by methylene blue staining [27], by densitometric scanning with the aid of NIH Image version 1.61 software.

2.6. Construction of RNA probes Complementary RNA probes of MMP-2 and MT1MMP labeled with digoxigenin (DIG) were generated by an in vitro transcription method for non-RI hybridization. Each complementary DNA (cDNA) of MT1MMP and MMP-2 was kindly provided by Dr H. Sato, Department of Virology and Molecular Oncology, Cancer Research Institute, Kanazawa University. They were inserted into pGEM-4Z and pGEM-3Zf, respectively. The plasmids were linearized by appropriate restriction enzymes to prepare the templates. In vitro transcription and labeling with DIG-UTP were performed simultaneously by appropriate RNA polymerases for the single stranded antisense probes, using the DIG RNA labeling kit (Boehringer

3. Results 3.1. Immunocytochemistry of GM-CSF receptor in HNSCC cell lines Prior to the experiments, the expression of GMCSF receptor in the cell lines was con®rmed by immunocytochemistry. Both cells were immunoreactive with antihuman GM-CSF receptor antibodies, with intense immunostaining in the cytoplasm (Fig. 1a,c). No immunoreactivity was observed when processed in a similar manner without the primary antibodies (Fig. 1b,d). Both cancer cell lines expressed the a subunit of GM-CSF receptor.

86

T. Tomita et al. / Cancer Letters 156 (2000) 83±91

Fig. 1. Microscopic view of SAS cells (a,b) and HSC-2 cells (c,d). Cells were cultured, ®xed, and incubated with antihuman GM-CSF receptor antibodies (a,c) or without antibodies (b,d). Cells were counterstained with hematoxylin. GM-CSF receptor immunoreactivity is observed as diffuse cytoplasmic staining (a,c). Both cancer cell lines express GM-CSF receptors.

Fig. 2. Effects of GM-CSF concentration on gelatinase activity in SAS and HSC-2 cells. After incubation with GM-CSF, conditioned media were collected and analyzed by gelatin zymography and compared with untreated controls (lane 1). The intensity of each gelatinolytic band was scanned and semi-quanti®ed by NIH Image version 1.61 software. Relative gelatinolytic activities corresponding to proMMP-2 (68 kDa type IV collagenase) from duplicate determination are shown.

T. Tomita et al. / Cancer Letters 156 (2000) 83±91

3.2. Effects of GM-CSF on the growth of HNSCC cell lines MTT assay was performed to study the effects of GM-CSF on the proliferation of HNSCC cell lines. GM-CSF showed no signi®cant in vitro proliferative effect on either SAS or HSC-2 cells (data not shown). 3.3. Effects of GM-CSF on MMP secretion and activity in HNSCC cells To study the effects of GM-CSF on MMP-2 secretion, conditioned media of the tumor cells cultured in the presence or absence of GM-CSF for 24 h were subjected to zymography. SAS cells demonstrated gelatinolytic activities of around 68 kDa, which were considered to correspond to a latent form of MMP-2 (proMMP-2). On the other hand, HSC-2 cells demonstrated gelatinolytic activities of around 92 and 68 kDa, which were considered to correspond to latent forms of matrix metalloproteinase-9 (proMMP-9) and proMMP-2, respectively. GM-CSF increased the production of proMMP-2 in both cell

87

lines in a dose-dependent manner. Treatment with GM-CSF at 50 ng/ml increased the 68-kDa activity 3.4-fold in SAS cells and 2.3-fold in HSC-2 cells compared with untreated controls (Fig. 2). However, the gelatinolytic band of proMMP-9 was not obviously induced by the GM-CSF treatment. 3.4. MMP mRNA levels in HNSCC cells treated with GM-CSF Northern blot hybridization was performed to examine whether the augmented secretion of proMMP-2 induced by GM-CSF was accompanied by increased levels of MMP-2 mRNA. SAS cells and HSC-2 cells treated with 50 ng/ml GM-CSF for 24 h demonstrated a 1.5- and 3.4-times higher level of MMP-2 speci®c mRNA transcripts in the untreated controls, respectively (Fig. 3), suggesting that the levels of secreted proMMP-2 are primarily regulated by those of the MMP-2 mRNA level. Subsequently, Northern blot hybridization with MT1-MMP RNA probe was performed. SAS cells and HSC-2 cells treated with 50 ng/ml GM-CSF for 24 h demonstrated a

Fig. 3. Effects of GM-CSF on MMP-2 mRNA levels in SAS and HSC-2 cells. Total RNAs (10 mg) from SAS cells and HSC-2 cells, cultured for 24 h with or without 50 ng/ml GM-CSF, were separated by agarose/formaldehyde gel electrophoresis, transferred to nylon membranes (Boehringer Mannheim), and hybridized with DIG-labeled antisense RNA probes of MMP-2. The intensity of each hybridization signal was determined by densitometry scanning standardized against 28S rRNA and compared with values representing the signal intensity of mRNA controls. GM-CSF (50 ng/ml) increased the intensity of the bands of MMP-2 mRNA 1.5 times in SAS cells and 3.4 times in HSC-2 cells, respectively. Representative of two experiments.

88

T. Tomita et al. / Cancer Letters 156 (2000) 83±91

2.2- and 2.0-times higher level of MT1-MMP speci®c mRNA transcripts in the untreated controls, respectively (Fig. 4).

4. Discussion GM-CSF is produced by T cells, macrophages, mast cells, endothelial cells, and ®broblasts. Because it does not appear in the circulation at detectable levels, it is thought to behave according to a paracrine model where the substance is produced and acts locally [10]. Recent reports have revealed that GMCSF can act on nonhematopoietic cells such as those of small cell lung cancer [28], osteosarcoma, breast cancer [23] and colon cancer [24]. Miyagawa et al. showed that the receptors for GM-CSF were identi®ed in nine of 35 (26%) human nonhematopoietic tumor cell lines including non-small cell lung cancer, stomach cancer, colon cancer, and osteosarcoma cells [29]. Recently, a number of studies have shown that the presence of GM-CSF correlates with cancer invasion and metastasis [11±17]. Young et al. showed that patients with HNSCC whose primary

cancers produced high GM-CSF levels had a high incidence of recurrence or metastasis [18]. However, in these studies, the precise mechanism by which GMCSF promotes cancer invasion and metastasis is not explained. MMP-2 and MT1-MMP have been shown to play a central role in the process of invasion and metastasis of HNSCC. An immunohistochemical study of MMP2 in 46 patients with HNSCC found a high level of expression in 77% of patients with lymph node metastases, but only in 25% of patients without lymph node metastases [30]. An in situ hybridization study of hypopharyngeal squamous cell carcinoma demonstrated that the expression of MMP-2 mRNA correlated well with the outcome of treatment [31]. Overexpression of MT1-MMP mRNA is a general feature of HNSCC [32], and we have previously shown that the expression of MT1-MMP correlates well with cervical lymph node metastasis in HNSCC [33]. On the other hand, little is known about the regulation of MMP-2 and MT1-MMP in human malignancies. MMP-2 is regulated by relatively few polypeptide factors [34]; however, expression of this collagenase is

Fig. 4. Effects of GM-CSF on MT1-MMP mRNA levels in SAS and HSC-2 cells. Total RNAs (10 mg) from SAS cells and HSC-2 cells, cultured for 24 h with or without 50 ng/ml GM-CSF, were separated by agarose/formaldehyde gel electrophoresis, transferred to nylon membranes (Boehringer Mannheim), hybridized with DIG-labeled antisense RNA probes of MT1-MMP. GM-CSF (50 ng/ml) increased the intensity of the bands of MT1-MMP mRNA 2.2 times in SAS cells and 2.0 times in HSC-2 cells, respectively. Representative of two experiments.

T. Tomita et al. / Cancer Letters 156 (2000) 83±91

induced by transforming growth factor-b (TGF-b) in tumorigenic human cell lines [35] and in prostate cancer cells [36], interferon-a (IFN-a) and IFN-g in melanoma cells [37], interleukin-8 (IL-8) in melanoma cells [38], and hepatocyte growth factor/scatter factor (HGF/SF) in glioma cells [39]. Expression of MT1MMP is induced by HGF/SF in glioma cells [39] and by TNF-a or TGF-b in ®brosarcoma cells [40]. Several studies have shown the localization of MMP-2 and MT1-MMP in HNSCC. MMP-2 mRNA was detected in both cancer cells and stromal cells [31], while proMMP-2 protein was localized to a cancer nest [30]. MT1-MMP mRNA was detected in both cancer cells and stromal cells [41,42], while MT1-MMP protein was localized to cancer cells [32]. We hypothesized that GM-CSF, which is secreted in a paracrine or autocrine fashion, may regulate MMP-2 and/or MT1-MMP production by cancer cells, and may thereby in¯uence their ability to invade and metastasize. We ®rst proved the presence of GM-CSF receptor expression in SAS and HSC-2 cells. Although some studies have shown that GM-CSF enhances the in vitro growth of tumor cell lines [23,24], its proliferative effect on tumor cells does not seem to be observed in most cancer cell lines [15,29]. In fact, our present data demonstrated that GM-CSF has no proliferative effect on HNSCC cells. Our data have shown that MMP-2 enzyme activity and the amount of proMMP-2 in HNSCC cells are enhanced in the presence of GM-CSF. Similar results were reported in lung cancer cell lines [12,15] and in a murine colon cancer cell line [43], but the authors only investigated MMP-2 enzyme activity. As shown in Fig. 3, GM-CSF upregulates the steadystate expression of MMP-2 speci®c mRNA levels in HNSCC cells. Our present report ®rst shows that GMCSF enhanced MMP-2 expression through the pretranslational activation. Since overproduction of proMMP-2 is indispensable for the development of the invasive phenotype [2], GM-CSF may contribute to the invasive activity of HNSCC cells through the upregulation of MMP-2. Moreover, we also demonstrated for the ®rst time that GM-CSF upregulates the MT1-MMP expression in HNSCC cells. Although MT1-MMP is an activator of proMMP-2 [3], gelatin zymography indicated no activation of proMMP-2 in cells treated with GM-CSF

89

in our experiment. Overexpression of MT1-MMP in some cell lines resulted in proMMP-2 activation, but no clear association was found between proMMP-2 activation and MT1-MMP expression levels, suggesting the contribution of some other factors [40]. It has recently been reported that the tissue inhibitors of metalloproteinase (TIMP)-2 are necessary for the interaction of MT1-MMP and proMMP-2, and the latter exists on the cell surface only as a complex with TIMP-2 [44]. Therefore the lack of TIMP-2 is a possible cause of there being no activation of proMMP-2. On the other hand, because MT1-MMP can directly degrade the extracellular matrix by itself [4,5], GM-CSF-mediated MT1-MMP upregulation in HNSCC cells may induce their invasive activity. We conclude that in HNSCC cells bearing GMCSF receptor, MMP-2 and MT1-MMP production are upregulated by GM-CSF. Our results suggest that GM-CSF may contribute to the invasiveness of HNSCC cells bearing GM-CSF receptor through the upregulation of MMP-2 and MT1-MMP expression. References [1] L.A. Liotta, P.S. Steeg, W.G. Stetler-Stevenson, Cancer metastasis and angiogenesis: an imbalance of positive and negative regulation, Cell 64 (1991) 327±336. [2] W.G. Stetler-Stevenson, S. Aznavoorian, L.A. Liotta, Tumor cell interactions with the extracellular matrix during invasion and metastasis, Annu. Rev. Cell. Biol. 9 (1993) 541±573. [3] H. Sato, T. Takino, Y. Okada, J. Cao, A. Shinagawa, E. Yamamoto, M. Seiki, A matrix metalloproteinase expressed on the surface of invasive tumour cells, Nature 370 (1994) 61±65. [4] M.P. d'Ortho, H. Stanton, M. Butler, S.J. Atkinson, G. Murphy, R.M. Hembry, MT1-MPP on the cell surface causes focal degradation of gelatin ®lms, FEBS Lett. 421 (1998) 159± 164. [5] M.P. d'Ortho, H. Will, S. Atkinson, G. Butler, A. Messent, J. Gavrilovic, B. Smith, R. Timpl, L. Zardi, G. Murphy, Membrane-type matrix metalloproteinases 1 and 2 exhibit broad-spectrum proteolytic capacities comparable to many matrix metalloproteinases, Eur. J. Biochem. 250 (1997) 751± 757. [6] H. Nakamura, H. Ueno, K. Yamashita, T. Shimada, E. Yamamoto, M. Noguchi, N. Fujimoto, H. Sato, M. Seiki, Y. Okada, Enhanced production and activation of progelatinase A mediated by membrane-type 1 matrix metalloproteinase in human papillary thyroid carcinomas, Cancer Res. 59 (1999) 467±473. [7] H. Ueno, H. Nakamura, M. Inoue, K. Imai, M. Noguchi, H. Sato, M. Seiki, Y. Okada, Expression and tissue localization of membrane-types 1, 2, and 3 matrix metalloproteinases in

90

[8]

[9] [10] [11]

[12]

[13]

[14]

[15]

[16]

[17]

[18]

[19] [20]

T. Tomita et al. / Cancer Letters 156 (2000) 83±91 human invasive breast carcinomas, Cancer Res. 57 (1997) 2055±2060. M. Tokuraku, H. Sato, S. Murakami, Y. Okada, Y. Watanabe, M. Seiki, Activation of the precursor of gelatinase A/72 kDa type IV collagenase/MMP-2 in lung carcinomas correlates with the expression of membrane-type matrix metalloproteinase (MT-MMP) and with lymph node metastasis, Int. J. Cancer 64 (1995) 355±359. P. Mignatti, D.B. Rifkin, Biology and biochemistry of proteinases in tumor invasion, Physiol. Rev. 73 (1993) 161±195. J.C. Gasson, Molecular physiology of granulocyte-macrophage colony-stimulating factor, Blood 77 (1991) 1131±1145. E.C. Kohn, G.H. Hollister, J.D. DiPersio, S. Wahl, L.A. Liotta, E. Schiffmann, Granulocyte-macrophage colony-stimulating factor induces human melanoma-cell migration, Int. J. Cancer 53 (1993) 968±972. X.H. Pei, Y. Nakanishi, K. Takayama, F. Bai, N. Hara, Granulocyte, granulocyte-macrophage, and macrophage colonystimulating factors can stimulate the invasive capacity of human lung cancer cells, Br. J. Cancer 79 (1999) 40±46. X.H. Pei, Y. Nakanishi, K. Takayama, J. Yatsunami, F. Bai, M. Kawasaki, K. Wakamatsu, N. Tsuruta, K. Mizuno, N. Hara, Granulocyte-colony stimulating factor promotes invasion by human lung cancer cell lines in vitro, Clin. Exp. Metastasis 14 (1996) 351±357. K. Takeda, K. Hatakeyama, Y. Tsuchiya, H. Rikiishi, K. Kumagai, A correlation between GM-CSF gene expression and metastases in murine tumors, Int. J. Cancer 473 (1991) 413±420. N. Tsuruta, J. Yatsunami, K. Takayama, Y. Nakanishi, Y. Ichinose, N. Hara, Granulocyte-macrophage-colony stimulating factor stimulates tumor invasiveness in squamous cell lung carcinoma, Cancer 82 (1998) 2173±2183. M.R. Young, Y. Lozano, A. Djordjevic, S. Devata, J. Matthews, M.E. Young, M.A. Wright, Granulocyte-macrophage colony-stimulating factor stimulates the metastatic properties of Lewis lung carcinoma cells through a protein kinase A signal-transduction pathway, Int. J. Cancer 53 (1993) 667±671. M.R. Young, J. Halpin, R. Hussain, Y. Lozano, A. Djordjevic, S. Devata, J.P. Matthews, M.A. Wright, Inhibition of tumor production of granulocyte-macrophage colony-stimulating factor by 1alpha,25-dihydroxyvitamin D3 reduces tumor motility and metastasis, Invasion Metastasis 13 (1993) 169± 177. M.R. Young, M.A. Wright, Y. Lozano, M.M. Prechel, J. Bene®eld, J.P. Leonetti, S.L. Collins, G.J. Petruzzelli, Increased recurrence and metastasis in patients whose primary head and neck squamous cell carcinomas secreted granulocyte-macrophage colony-stimulating factor and contained CD34 1 natural suppressor cells, Int. J. Cancer 74 (1997) 69±74. E.A. Mann, J.D. Spiro, L.L. Chen, D.L. Kreutzer, Cytokine expression by head and neck squamous cell carcinomas, Am. J. Surg. 164 (1992) 567±573. K. Okumura, A. Konishi, M. Tanaka, M. Kanazawa, K. Kogawa, Y. Niitsu, Establishment of high- and low-invasion clones derived for a human tongue squamous-cell carcinoma

[21]

[22]

[23]

[24]

[25]

[26]

[27] [28]

[29]

[30]

[31]

[32]

[33]

cell line SAS, J. Cancer Res. Clin. Oncol. 122 (1996) 243± 248. N. Kamata, K. Chida, K. Rikimaru, M. Horikoshi, S. Enomoto, T. Kuroki, Growth-inhibitory effects of epidermal growth factor and overexpression of its receptors on human squamous cell carcinomas in culture, Cancer Res. 46 (1986) 1648±1653. K. Hayashida, T. Kitamura, D.M. Gorman, K. Arai, T. Yokota, A. Miyajima, Molecular cloning of a second subunit of the receptor for human granulocyte-macrophage colonystimulating factor (GM-CSF): reconstitution of a high-af®nity GM-CSF receptor, Proc. Natl. Acad. Sci. USA 87 (1990) 9655±9659. S. Dedhar, L. Gaboury, P. Galloway, C. Eaves, Human granulocyte-macrophage colony-stimulating factor is a growth factor active on a variety of cell types of nonhemopoietic origin, Proc. Natl. Acad. Sci. USA 85 (1988) 9253±9257. W.E. Berdel, S. Danhauser-Riedl, G. Steinhauser, E.F. Winton, Various human hematopoietic growth factors (interleukin-3, GM-CSF, G-CSF) stimulate clonal growth of nonhematopoietic tumor cells, Blood 73 (1989) 80±83. A. Kawashima, I. Nakanishi, H. Tsuchiya, A. Roessner, K. Obata, Y. Okada, Expression of matrix metalloproteinase 9 (92-kDa gelatinase/type IV collagenase) induced by tumour necrosis factor alpha correlates with metastatic ability in a human osteosarcoma cell line, Virchows Arch. 424 (1994) 547±552. H.J. Holtke, C. Kessler, Non-radioactive labeling of RNA transcripts in vitro with the hapten digoxigenin (DIG); hybridization and ELISA-based detection, Nucleic Acids Res. 18 (1990) 5843±5851. M. Wilkinson, J. Doskow, S. Lindsey, RNA blots: staining procedures and optimization of conditions, Nucleic Acids Res. 19 (1991) 679. M.R. Ruff, W.L. Farrar, C.B. Pert, Interferon gamma and granulocyte/macrophage colony-stimulating factor inhibit growth and induce antigens characteristic of myeloid differentiation in small-cell lung cancer cell lines, Proc. Natl. Acad. Sci. USA 83 (1986) 6613±6617. K. Miyagawa, S. Chiba, K. Shibuya, Y.F. Piao, S. Matsuki, J. Yokota, M. Terada, K. Miyazono, F. Takaku, Frequent expression of receptors for granulocyte-macrophage colony-stimulating factor on human nonhematopoietic tumor cell lines, J. Cell Physiol. 143 (1990) 483±487. J. Kusukawa, Y. Sasaguri, I. Shima, T. Kameyama, M. Morimatsu, Expression of matrix metalloproteinase-2 related to lymph node metastasis of oral squamous cell carcinoma. A clinicopathologic study, Am. J. Clin. Pathol. 99 (1993) 18±23. Y. Miyajima, R. Nakano, M. Morimatsu, Analysis of expression of matrix metalloproteinases-2 and -9 in hypopharyngeal squamous cell carcinoma by in situ hybridization, Ann. Otol. Rhinol. Laryngol. 104 (1995) 678±684. T. Yoshizaki, H. Sato, Y. Maruyama, S. Murono, M. Furukawa, C.S. Park, M. Seiki, Increased expression of membrane type 1-matrix metalloproteinase in head and neck carcinoma, Cancer 79 (1997) 139±144. Y. Imanishi, M. Fujii, Y. Tokumaru, M. Kanke, T. Tomita, K.

T. Tomita et al. / Cancer Letters 156 (2000) 83±91

[34] [35]

[36]

[37]

[38]

[39]

Kameyama, H. Sato, Analysis of expression of MT1-MMP and MMP-2 in human head and neck carcinoma, Head Neck 20 (1998) 457. U. Benbow, C.E. Brinckerhoff, The AP-1 site and MMP gene regulation: what is all the fuss about?, Matrix Biol. 15 (1997) 519±526. P.D. Brown, A.T. Levy, I.M. Margulies, L.A. Liotta, W.G. Stetler-Stevenson, Independent expression and cellular processing of Mr 72,000 type IV collagenase and interstitial collagenase in human tumorigenic cell lines, Cancer Res. 50 (1990) 6184±6191. I. Sehgal, T.C. Thompson, Novel regulation of type IV collagenase (matrix metalloproteinase-9 and -2) activities by transforming growth factor-beta1 in human prostate cancer cell lines, Mol. Biol. Cell 10 (1999) 407±416. E.S. Hujanen, A. Vaisanen, A. Zheng, K. Tryggvason, T. Turpeenniemi-Hujanen, Modulation of M(r) 72,000 and M(r) 92,000 type-IV collagenase (gelatinase A and B) gene expression by interferons alpha and gamma in human melanoma, Int. J. Cancer 58 (1994) 582±586. M. Luca, S. Huang, J.E. Gershenwald, R.K. Singh, R. Reich, M. Bar-Eli, Expression of interleukin-8 by human melanoma cells up-regulates MMP-2 activity and increases tumor growth and metastasis, Am. J. Pathol. 151 (1997) 1105±1113. R. Hamasuna, H. Kataoka, T. Moriyama, H. Itoh, M. Seiki, M. Koono, Regulation of matrix metalloproteinase-2 (MMP-2) by hepatocyte growth factor/scatter factor (HGF/SF) in human glioma cells: HGF/SF enhances MMP-2 expression

[40]

[41]

[42]

[43]

[44]

91

and activation accompanying up-regulation of membrane type-1 MMP, Int. J. Cancer 82 (1999) 274±281. J. Lohi, K. Lehti, J. Westermarck, V.M. Kahari, J. Keski-Oja, Regulation of membrane-type matrix metalloproteinase-1 expression by growth factors and phorbol 12-myristate 13acetate, Eur. J. Biochem. 239 (1996) 239±247. A. Okada, J.P. Bellocq, N. Rouyer, M.P. Chenard, M.C. Rio, P. Chambon, P. Basset, Membrane-type matrix metalloproteinase (MT-MMP) gene is expressed in stromal cells of human colon, breast, and head and neck carcinomas, Proc. Natl. Acad. Sci. USA 92 (1995) 2730±2734. M. Sutinen, T. Kainulainen, T. Hurskainen, E. Vesterlund, J.P. Alexander, C.M. Overall, T. Sorsa, T. Salo, Expression of matrix metalloproteinases (MMP-1 and -2) and their inhibitors (TIMP-1, -2 and -3) in oral lichen planus, dysplasia, squamous cell carcinoma and lymph node metastasis, Br. J. Cancer 77 (1998) 2239±2245. S. Shimizu, Y. Nishikawa, K. Kuroda, S. Takagi, K. Kozaki, S. Hyuga, S. Saga, M. Matsuyama, Involvement of transforming growth factor beta1 in autocrine enhancement of gelatinase B secretion by murine metastatic colon carcinoma cells, Cancer Res. 56 (1996) 3366±3370. A.Y. Strongin, I. Collier, G. Bannikov, B.L. Marmer, G.A. Grant, G.I. Goldberg, Mechanism of cell surface activation of 72-kDa type IV collagenase. Isolation of the activated form of the membrane metalloprotease, J. Biol. Chem. 270 (1995) 5331±5338.