Cancer Cell Biology
Tenascin-C upregulates matrix metalloproteinase-9 in breast cancer cells: Direct and synergistic effects with transforming growth factor β1
Article first published online: 21 FEB 2003
Copyright © 2003 Wiley-Liss, Inc.
International Journal of Cancer
Volume 105, Issue 1, pages 53–60, 20 May 2003
How to Cite
Kalembeyi, I., Inada, H., Nishiura, R., Imanaka-Yoshida, K., Sakakura, T. and Yoshida, T. (2003), Tenascin-C upregulates matrix metalloproteinase-9 in breast cancer cells: Direct and synergistic effects with transforming growth factor β1. Int. J. Cancer, 105: 53–60. doi: 10.1002/ijc.11037
- Issue published online: 27 MAR 2003
- Article first published online: 21 FEB 2003
- Manuscript Accepted: 23 DEC 2002
- Manuscript Revised: 18 NOV 2002
- Manuscript Received: 28 JUN 2002
- Ministry of Education, Culture, Sports, Science and Technology of Japan
- matrix metalloproteinase;
- breast cancer
Tenascin-C (TN-C) and matrix metalloproteinases (MMPs) are highly expressed in cancer tissues and probably promote cell migration during cancer progression. TN-C and MMPs are often co-localized in areas of active tissue remodeling in pathologic conditions, suggesting reciprocal regulation. To investigate whether TN-C regulates MMPs expression in cancer cells, we first exposed mammary cancer cells derived from TN-C-deficient mice to TN-C and examined MMPs expression. TN-C was then compared with fibronectin (FN), laminin (LN), basic fibroblast growth factor (b-FGF) and transforming growth factor-beta1 (TGF-β1). Results of endpoint RT-PCR, quantitative real-time RT-PCR and gelatin zymography demonstrated that TN-C, strongly and dose dependently, upregulates MMP-9 expression in murine mammary cancer cells. TN-C weakly induced MMP-2, MMP-3 and MMP-13. FN and LN induced MMP-9 to lesser extents compared with TN-C. b-FGF had no effect on MMP-9 expression. TGF-β1 induced MMP-9 expression in a dose-dependent manner, and this induction was significantly enhanced by addition of TN-C. TN-C and TGF-β1 also upregulated MMP-9 expression in the human breast cancer cell line MDA-MB-231. Neutralization with specific anti-TGF-β1 antibody showed decreased expression of MMP-9, indicating that TGF-β controls the baseline MMP-9 expression by a direct autocrine mechanism. Under neutralization of TGF-β, addition of TN-C still upregulated MMP-9. Conversely, neutralization of endogenous TN-C (in a TN-C-positive mammary cancer cell line) downregulated TGF-β-induced MMP-9 expression. Thus, TN-C induces MMP-9 expression directly and by collaboration with TGF-β. These findings reveal a novel role of TN-C in breast cancer progression. © 2003 Wiley-Liss, Inc.
Tumor invasion and subsequent development of metastasis are complex processes that involve multiple and coordinated steps: dissolution of basement membranes, modulation of tumor cell adhesion, migration and growth at both the primary and the secondary sites. Tumor cells that produce and/or induce factors that can assist in these steps have an advantage for successful invasion and development of metastasis. Tenascin-C (TN-C), a large 6-armed glycoprotein of the extracellular matrix (ECM), is suspected of playing a significant role in cancer invasion. In a variety of malignant tumor tissues, TN-C is highly expressed by tumor cells themselves and stromal cells such as fibroblasts and endothelial cells.1, 2, 3, 4 Previous histopathologic studies using antibodies and nucleotide probes have demonstrated that higher TN-C expression is associated with tumor progression and an unfavorable outcome of patients.4, 5, 6, 7 In breast cancer, cases with dense TN-C deposition also show a high risk of distant metastasis, local and distant recurrence and adverse patient outcome.5, 8 The progressive nature of tumors producing TN-C may be attributed to the properties of TN-C that promote cell migration and proliferation, as evaluated from in vitro studies.9, 10
Matrix metalloproteinases (MMPs), a class of zinc-dependent enzymes, have been implicated in a proteolytic cascade for ECM degradation. They are collectively capable of degrading all the constituents of the ECM. Like TN-C, MMPs are actively produced by both tumor and stromal cells in cancer tissues including breast cancers, suggesting an involvement in tumor invasion and metastasis.11, 12, 13, 14 MMPs are secreted as proenzymes and require activation for their proteolytic activity by the MMP cascade and serum-derived serine proteinases.15 Genes of most MMPs are inducible. Growth factors and oncogenic transformation are among the effectors.16, 17, 18, 19 Transforming growth factor-β (TGF-β) may be an effector or a suppressor depending on the target cell types.20, 21, 22 However, in advanced cancers, increased expression of MMPs is considered to be regulated by the cancer tissue microenvironment in which the TGF-β system plays an important role.23
Thus, both TN-C and MMPs are upregulated in breast cancer tissues. It is tempting to consider that TN-C could induce MMPs expression and activity. Supporting this hypothesis, upregulation of MMPs was observed in synovial fibroblasts plated on a mixed substrate of fibronectin (FN) and TN-C but not on FN alone.24 In addition, during development and remodeling of mammary tissues, TN-C deposition coincides with the expression of MMPs.25, 26 To explore the regulatory pathways of MMPs in breast cancer tissues, we examined the effects of TN-C on the induction of MMPs in murine and human mammary cancer cells. In our study, we found that TN-C induces expression of some MMPs, especially MMP-9, in mammary cancer cells and that TN-C could enhance effects of TGF-β on MMP-9 induction, indicating a new function of TN-C in cancer progression.
MATERIAL AND METHODS
Animals, cells and cell culture
TN-C-deficient mice originally generated by Saga et al.27 were backcrossed with GRS/A mice for more than 12 generations and were congenic. Mammary cancer cell lines were established from a murine mammary tumor (MMT) spontaneously developed in the TN-C-deficient homozygote, then cloned and named 5E, 7E and 1C. Other mammary cancer cell lines named 4B8 and 6E7 were cloned from MMT of a wild-type GRS/A mouse.28 Human breast cancer cell line MDA-MB-231 was obtained from the American Type Culture Collection. These cell lines were maintained in Iscove's modified Dulbecco's medium (IMDM; Sigma, St. Louis, MO) supplemented with 10% fetal bovine serum (FBS; Gibco-BRL, Gaithersburg, MD) at 37°C under a 5% CO2 atmosphere. For induction experiments, cells were plated in Falcon 6-well plates (Becton Dickinson, Franklin Lakes, NJ) at a concentration of 5 × 105 cells/well in 2 ml of 0.1% FBS IMDM and incubated for 48 hr. The medium was then replaced with 2 ml of medium containing the substance under investigation for MMP-9 induction: Basic fibroblast growth factor (b-FGF; 50 ng/ml; Wako Life Science, Osaka, Japan), TGF-β1 (5 ng/ml; Boehringer Mannheim, Mannheim, Germany), FN (20 μg/ml, Gibco-BRL, Gaithersburg, MD), Laminin (LN; 20 μg/ml; Gibco-BRL) and TN-C (10 μg/ml). The incubation was continued for another 24 hr. TN-C was isolated and purified from conditioned medium of the human glioma cell line U-251MG.9
For cultures on ECM substrates, the wells were coated as previously reported.24 Freshly trypsinized cells were washed 3 times in a 0.1% FBS IMDM medium and plated for 24 hr in wells coated with FN (30 μg/ml), TN-C (10 μg/ml) or FN and TN-C substrate mixture.
Conditioned media for gelatin zymography were collected, spun at 2,000g for 5 min at 4°C to remove cellular debris and used immediately or stored at −80°C. Total RNA was isolated from the cells as described below.
The secreted metalloproteinases were detected by zymography as previously reported.29 Briefly, 20 μl of conditioned medium was mixed with an equal volume of sample buffer (100 mM Tris-HCl, pH 6.8, 20% glycerol, 0.2% bromophenol blue). Fifteen microliters of this mixture was then applied to a 10% SDS-polyacrylamide gel containing 1 mg/ml of gelatin. After electrophoresis the gels were washed twice in 100 mM NaCl, 2.5% Triton X-100, 50 mM Tris-HCl buffer (pH 7.5) and incubated for 24 hr at 37°C in 5 mM CaCl2, 50 mM Tris-HCl buffer (pH 8.0). The gels were then stained in 0.1% Coomassie Brilliant Blue R250 in 40% methanol and 10% acetic acid for 1 hr and destained in the same solution without the dye. Gelatinolytic enzymes were detected as clear bands against the blue background of the gels. Gels were scanned in a digital scanner, and densitometric measurement was performed with NIH Image (version 1.56) software.
Analysis of MMP mRNAs using reverse transcription polymerase chain reaction
Total RNAs were extracted with Isogen (Wako Life Science, Osaka, Japan), a premixed RNA isolation reagent based on the acid guanidium thiocyanate-phenol-chloroform extraction method, following the supplied protocol. Reverse transcription (RT) polymerase chain reaction (PCR) (total RNA 1 μg/reaction) was performed using First Strand cDNA Synthesis Kit and oligo-dT primers (Boehringer Mannheim). Aliquots of cDNA were subjected to PCR using the primers shown in Table I for 27 cycles. An initial denaturation step was employed at 95°C for 10 min, followed by repeat cycles of 94°C for 30 sec (denaturation), 55°C for 30 sec (annealing) and 72°C for 1 min (elongation). The samples were then incubated at 72°C for 7 min and stored at 4°C. The amplified products were analyzed by electrophoresis in 2% agarose gels, stained with ethidium bromide (0.1 μg/ml) and detected by ultraviolet light. The images were captured on a digital camera and stored on a disk using Documentation System (Amersham Pharmacia Biotech, Buckingamshire, UK). The bands were quantitated by NIH Image (version 1.56) software.
|mRNA||Primer sequence||Product size (bp)||Position on cDNA|
Quantitative real-time RT-PCR
Quantitative real-time RT-PCR was performed using the ABI Prism 7700 Sequence Detection System (Applied Biosystems/Perkin-Elmer, Foster City, CA). The TaqMan PCR Universal Master Mix reagent, MicroAmp optical tubes and MicroAmp caps were also from Applied Biosystems. The PCR primers and the TaqMan fluorogenic probes were designed using the Primer Express software program (Applied Biosystems) and ordered from Takara Biomedicals (Shiga, Japan). Their sequences are shown in Table II. The sequences of primers and probes for human MMP-9 and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) have been previously reported.30 The standard curve used was based on serial dilutions of one cDNA sample from RT products ranging from 10−1 to 10−4. Amplification mixes (50 μl) contained the sample cDNA (1 μl, around 10 ng), 200 nM each primer (1 μl), 40 nM probe (0.2 μl), TaqMan PCR Universal Master Mix reagent (25 μl) and sterile double distilled water (21.8 μl). The thermal cycling conditions comprised 2 min at 50°C, 10 min at 95°C and 40 cycles at 95°C for 15 sec and 60°C for 1 min. GAPDH was used as the housekeeping gene.
|mRNA||Primer and probe sequences||Product size (bp)||Position on cDNA|
The levels of TGF-β1 in conditioned media of TN-C-null cells (treated or not with 10 μg/ml of TN-C) were assayed using a kit obtained from R&D Systems (Duoset®, Minneapolis, MN), according to the supplied protocol.
Neutralization of TGF-β and TN-C bioactivities
For TGF-β neutralization, TN-C-null cells were treated with 50 μg/ml of monoclonal anti-TGF-β antibody (clone 1D11, R&D Systems, Minneapolis, MN) or 1 μg/ml of recombinant human latent associated peptides (LAP) (R&D Systems). The antibody and LAP are capable of neutralizing TGF-β1, β2 and β3. In another set of experiments TN-C (1 μg/ml) was added to the medium at the time of TGF-β neutralization. For TN-C neutralization, wild-type TN-C-positive cells 6E7 were treated with 10 μg/ml of a rabbit polyclonal, affinity purified, human anti-TN-C antibody31 or a control normal rabbit IgG, in the presence or absence of TGF-β1 (5 ng/ml).
The results are expressed as the mean ± the standard deviation (SD) of at least 3 experiments and comparisons are based on the unpaired t-test (Student's t-test). A p-value < 0.05 was considered significant.
TN-C induces expression of MMPs, especially MMP-9, in murine mammary cancer cell lines derived from TN-C-deficient mice
Initially we examined the expression of MMP-2, MMP-3, MMP-9, MMP-11, MMP-13 and MMP-14 by RT-PCR in TN-C-null cancer cells grown under serum starvation conditions with or without the addition of 10 μg/ml of TN-C. As shown in Figure 1, the addition of TN-C resulted in significant upregulation of MMP-9 and apparent upregulation of MMP-2, MMP-3 and MMP-13 mRNAs. MMP-11 and MMP-14 were almost not affected by the addition of TN-C. Among MMPs expressed by mammary cancer cells, it was previously reported that MMP-9 is upregulated coincidentally with malignant transformation and tumor progression.32, 33 With this in mind, we decided to further test MMP-9 upregulation in different TN-C-null cancer cell lines using quantitative real-time PCR. As shown in Figure 2a, 2 cell lines, 5E and 1C, exhibited increased expression of MMP-9 mRNA by TNC, whereas 7E did show constitutive overexpression of MMP-9. To confirm this finding at the protein level, gelatin zymography was performed on the culture medium of all 3 cell lines. A prominent gelatinolytic activity consistent with the latent form of murine MMP-9 (105 kDa) was observed in all cell lines after treatment with 10 μg/ml TN-C (Fig. 2b). Cell line 7E secreted a noticeable amount of MMP-9 even without TN-C treatment. The 7E cells might have undergone a phenotypical change possibly to compensate for the TN-C-null condition. This cell line was withdrawn from subsequent experiments in our study.
ECM proteins induce MMP-9 in murine mammary cancer cells
We compared the effect of TN-C on the induction of MMP-9 to those of other ECM glycoproteins, FN and LN. The addition of soluble FN and LN increased MMP-9 expression in the cancer cells. Whereas TN-C increased MMP-9 mRNA more than 5-fold, FN showed a 4-fold increase and LN showed a 3-fold increase. These differences were statistically significant (Fig. 3a). Induction of MMP-9 mRNA was also translated by a proportional increase in the gelatinolytic activities in the conditioned medium (data not shown). Next, although we examined the synergistic effects in a combination of soluble TN-C and FN or LN, no synergy was observed. It is well known that cells in culture react differently to ECM proteins, depending on whether the proteins are soluble or coated to a surface. When cells were cultured on substrates of TN-C, FN or a mixture of FN and TN-C, the highest expression of MMP-9 was observed on TN-C alone. A mixture of FN and TN had a higher expression than FN alone (Fig. 3b and c) but lower than TN-C alone.
TN-C enhances the effect of TGF-β1 on MMP-9 induction in mammary cancer cells
TN-C was compared with b-FGF and TGF-β1. b-FGF was previously reported to induce MMP-9 in cancer cells.34, 35 However, in mammary cancer cells used in our study, it had no effect on the expression of MMP-9 mRNA. In contrast, 5 ng/ml of TGF-β1 induced a 4-fold increase, not significantly different from TN-C (Fig. 4a). The gelatinolytic activity in the conditioned medium was also increased with no apparent difference between TGF-β1 and TN-C (Fig. 4β). We then studied the effect of TN-C and TGF-β1 concentrations on MMP-9 upregulation and found that both TN-C and TGF-β1 increased MMP-9 expression in a dose-dependent manner (Fig. 5a). In addition, using a human breast cancer cell line, MDA-MB-231, we confirmed the effect of TN-C on MMP-9 induction in human cells. As shown in Figure 5b, treatment of human cells with TN-C and TGF-β1 increased the MMP-9 mRNA expression 7-fold and 5-fold, respectively. Because TGF-β1 is a known inducer of ECM proteins including TN-C,36 we tested TN-C induction by TGF-β1 using real-time PCR in wild-type murine mammary cancer cells (4B8 and 6E7). Treatment of the wild-type cells with TGF-β1 increased the expression of TN-C mRNA and MMP-9 mRNA up to 7-fold and 53-fold, respectively (data not shown). Having found that TN-C induces MMP-9 and that both TN-C and MMP-9 are induced by TGF-β1, we decided to compare the TN-C and TGF-β1 effects on other MMPs expression in TN-C-null and wild-type cancer cells. We found that MMP-9 was the only MMP induced by both TN-C and TGF-β1 in all the cell lines examined (Table III). To further clarify this close association between TN-C, TGF-β1 and MMP-9, we examined whether TN-C could modulate the effects of TGF-β1 on MMP-9 expression. When TN-C-null cells were treated with lower but effective concentrations of TGF-β1 (0.1 ng/ml) and TN-C (1 μg/ml), a synergistic effect on MMP-9 induction by TN-C and TGF-β1 was observed (Fig. 6). Next, the endogenous expression of TGF-β1 was tested by RT-PCR and ELISA. By RT-PCR, all the cell lines showed a constitutive expression of TGF-β1 that was not apparently affected by the presence or absence of TN-C treatment (data not shown). In the ELISA assay after activation by acidic treatment, the conditioned media of TN-C-null cells treated with or without 10 μg/ml of TN-C also showed significant levels of total TGF-β1 (1.27 and 1.67 ng/ml respectively, latent TGF-β1 included). Therefore, using a neutralizing antibody of TGF-β and latent associated polypeptide (LAP), we tried to minimize the endogenous TGF-β1's effect and clarify whether the TN-C effect on MMP-9 expression is direct or indirect by enhancing an autocrine effect of TGF-β1 secreted by the cancer cells. Treatment of TN-C-null cells with the monoclonal antibody significantly decreased the expression of MMP-9 mRNA from the baseline level and after TN-C treatment. The treatment with LAP showed the same tendency, although the difference was not significant. In the condition where TGF-β activity was neutralized, the addition of TN-C still showed a 3-fold increase in MMP-9 mRNA (Fig. 7a). Furthermore, neutralization of endogenously secreted TN-C by wild-type cells with a polyclonal anti-TN-C antibody significantly reduced the TGF-β1-induced upregulation of MMP-9 (Fig. 7b), although the upregulation of TN-C mRNA by TGF-β1 was not affected (data not shown). Thus, MMP-9 expression in breast cancer cells could be directly regulated by TGF-β1 and TN-C, and the TGF-β1 effect on MMP-9 expression is evidently enhanced by the addition of TN-C.
Our present study of MMPs expression in mammary cancer cells has demonstrated that TN-C upregulates MMP-9 expression in mammary cancer cells and enhances TGF-β-induced MMP-9 expression. TN-C, TGF-β and MMP-9 are all known to be upregulated in breast cancer tissues and expressed in both the stromal and the epithelial compartments of tumors.1, 2, 5, 11, 31, 37–40 TGF-β is known to play an important role in tumor progression through selection of malignant cells that escape TGF-β-induced cell growth inhibition and through the upregulation of ECM and protease expression by cancer cells.41 In agreement, our results also show that the mammary cancer cells used in our study constitutively express TGF-β1 mRNA and that the addition of TGF-β1 upregulates the TN-C and MMP-9 mRNAs expression in these cells. Breast cancer invasion is accompanied by the loss of basement membrane23 and the accumulation of TN-C at the invasion front.2, 8 MMP-9 is known, on the other hand, to participate in the proteolysis of basement membrane materials.15 Interestingly, TN-C, which is not a permanent component of the basement membrane, is known to resist MMP-9 proteolytic activity.42, 43 TN-C deposition and MMPs upregulation are coincidentally shown during development and in remodeling tissues.25, 26, 44, 45 By showing that TN-C upregulates MMP-9 in mammary cancer cells, our study confirms and extends the findings of previous reports that placed TN-C upstream in the regulation of some MMPs, including MMP-9, in the synovial fibroblasts and interstitial cells of heart valves.24, 45 By comparing TN-C to FN and LN, we have shown that TN-C is the most determinant ECM factor for MMP-9 upregulation resulting from cell-ECM interactions studied so far. The results of our experiments with soluble and substrate-bound TN-C, FN and LN indicate that MMP-9 upregulation is strongly dependent on TN-C concentration when mammary cancer cells are exposed to mixtures of ECM proteins. Thus, in addition to the previously reported role of TN-C in cancer cell growth and migration,9, 10 the accumulation of TN-C at the focal site of breast cancer invasion probably also contributes to the degradation of the basement membrane by focally upregulating MMP-9 expression.
Although we have observed an induction of MMP-2, MMP-3, MMP-9 and MMP-13 after TN-C treatment of the mammary cancer cells used in our study, our results concentrate on MMP-9 because its upregulation was the most evident. Other studies have also shown particular expression patterns of MMP-9 among the various MMPs described in breast cancers.11, 46, 47 For instance, the MMP-9 gene seems to be inducible in breast cancer cells in culture, whereas the MMP-2 gene seems to be constitutively expressed.32 The MMP-9 distribution in breast cancer tissues is more focal and restricted to some breast cancers, whereas MMP-2 expression tends to be diffuse and ubiquitous.12, 48 Furthermore, several lines of evidence implicate MMP-9 in breast cancer progression.17, 33, 49
Our data showing that TGF-β upregulates MMP-9 in mammary cancer cells are in agreement with a previous report that showed induction in human bone-metastasizing breast cancer cells.22 TGF-β has also been reported to induce MMP-2 in fibroblasts, and MMP-9 in colon and prostatic cancer cells.20, 21, 50, 51 Considering that TGF-β and TN-C are expressed together in the tumor microenvironment and that both induce MMP-9, we investigated whether their effects were synergistic on MMP-9 induction. We showed that TN-C enhances the effects of TGF-β on MMP-9 expression. By neutralizing constitutively expressed TGF-β with a monoclonal antibody, we have shown a specific downregulation of MMP-9 mRNA. Other MMPs were not affected on real-time PCR (data not shown). This strongly suggests that TGF-β plays a primary role in the control of baseline MMP-9 mRNA level, via a direct autocrine mechanism. Our data also showed that TN-C was still capable of inducing MMP-9 mRNA, despite the addition to the culture of neutralizing antibodies against TGF-β. In addition, blocking of endogenously secreted TN-C with a polyclonal antibody resulted in the downregulation of TGF-β-induced MMP-9 expression. These findings indicate that TN-C can directly stimulate MMP-9 synthesis and that signaling from separated receptors for TGF-β and TN-C could enhance MMP-9 induction. Because TGF-β also stimulates TN-C expression, secreted TGF-β and TN-C could synergistically activate MMP-9 expression of cancer cells in an autocrine manner, followed by progression of cancer cell invasion and metastasis.
As other inducers of MMP-9 in mammary cells, epidermal growth factor (EGF) and amphiregulin in SKBR-3 cells, insulin-like growth factor I (IGF-I), bcl-2 overexpression in MCF-7 cells and FN fragments (FN120) in normal mammary cells have been reported.52, 53, 54, 55 Recently, interactions between TN-C and other growth factor signaling pathways have been found. Epidermal growth factor-like repeats of TN-C can bind and directly activate epidermal growth factor receptor (EGFR).56 TN-C was also shown to act through the αvβ3 integrin to promote EGFR phosphorylation in vascular smooth muscle cells.44 Thus, further studies are necessary to explore the functional domains and receptors of TN-C and intracellular signaling for the induction of MMP-9 expression in cancer cells. In addition, the TN-C-null cell lines used in our study can be considered useful for the investigation of TN-C functional domains.
We thank Dr. E.C. Gabazza for help with the ELISA assay.
- 23Synthesis and degradation of basement membranes and extracellular matrix and their regulation by TGF-β in invasive carcinomas [review]. Intern J Oncol 2001;18: 669–81., , .