Downregulation of MMP-9 in ERK-mutated stable transfectants inhibits glioma invasion in vitro

Abstract

We previously showed that enhanced expression of MMP-9, an endopeptidase that digests basement-membrane type IV collagen, is related to tumor progression in vitro and in vivo; antisense-MMP-9 stably transfected clones were less invasive than untransfected parental cells and did not form tumors in nude mice. In this study, we examined the role of ERK-1 in the regulation of MMP-9 production and the invasive behavior of the human glioblastoma cell line SNB19, in which ERK1 is constitutively activated. SNB19 cells were stably transfected with mt-ERK, a vector encoding ERK-1 cDNA in which the conserved lysine at codon 71 was changed to arginine, thus impairing the catalytic efficiency of this enzyme. Gelatin zymography showed reduced levels of MMP-9 in the mt-ERK-transfected cell lines relative to those in vector-transfected and parental control cells. Reductions in MMP-9 protein mRNA levels were also detected in the mt-ERK-transfected cells by Western and Northern blotting. The mt-ERK-transfected cells were much less invasive than parental or vector control cells in a Matrigel invasion assay and in a spheroid coculture assay. Thus an ERK-dependent signaling pathway seems to regulate MMP-9 mediated glioma invasion in SNB19 cells; interfering with this pathway could be developed into a therapeutic approach, which aims at a reduction of cancer cell invasion.

Introduction

To invade and spread through surrounding normal tissue, tumor cells must degrade multiple elements of the ECM, including fibronectin, laminin, and type IV collagen (Matrisian, 1992). Of the several families of ECM-degrading enzymes, the most extensive are the matrix metalloproteinases (MMPs). Studies from our laboratory have shown that expression Mr.92 000 type IV collagenases (MMP-9) by glial tumor cells increases with the degree of malignancy of those tumor cells (Rao et al., 1996). Many growth factors (e.g., epidermal growth factor (EGF), hepatocyte growth factor (HGF)), cytokines (e.g., tumor necrosis factor alpha (TNF-α)), tumor promoters (e.g., tetradecanoyl phorbol acetate (TPA)), and oncogenes (e.g., RAS, Src) induce MMP-9 protein expression through the activation of its gene promoter by signal transduction pathways (Gum et al., 1997).

Several studies have identified signal transduction pathways involved in the regulation of MMP-9 expression in keratinocytes (Mccawley et al., 1999; Zeigler et al., 1999) and tumor cells (Gum et al., 1996, 1997; Simon et al., 1998). A major mechanism through which signals from extracellular stimuli are transmitted to the nucleus involves activation of kinases related to the mitogen-activated protein kinase (MAPK) superfamily (for review see Robinson and Cobb, 1997). To date, the involvement of at least three subgroups of MAPK family members have been identified in a wide range of cellular responses to extracellular signals. The enzymes in the first subgroup, named extracellular signal-regulated kinases (ERKs) are activated through sequential phosphorylation of the upstream kinases Raf and MEK. Activation of Raf, in turn, is achieved through its interaction with membrane bound farnesylated Ras. The classical Raf/MEK/ERK mitogenic cascade is strongly activated upon stimulation of cells with growth factors, serum, and phorbol esters like phorbol 12-myristate 13-acetate (PMA). For the other two MAPK subgroups, c-Jun N-terminal kinase/stress-activated protein kinase (JNK/SAPK) and p38MAPK, homologous signal transduction pathways have been described. These latter two subgroups of the MAPK family are only weakly activated by mitogens, but are highly stimulated on exposure to inflammatory cytokines such as TNF-α and IL-1 and a wide variety of environmental stress inducers.

The MMP-9 gene covers 13 exons spanning 7.7 kb and is transcribed into a 2.5 kb mRNA (Sato and Seiki, 1993). The 5′ flanking sequence, which includes 670 nucleotides, contains putative binding sites for AP-1, NF-κB, Sp1, and Ap-2 (Sato and Seiki, 1993). The Ap-1 transcription complex appears to play an essential role in stimulating transcriptional activation of MMP-9 (Sato et al., 1993; Gum et al., 1996). In its active form, the AP-1 complex may comprise homodimers of c-jun or heterodimers between c-Fos, c-jun and ATF2 (Karin et al., 1997; Smeal et al., 1989; Whisler et al., 1997). c-jun proteins are activated by N-terminal phosphorylation of specific serine residues (ser 63/73) that appear to be exclusively activated by jun-N-terminal kinase (JNKs), also known as stress-activated protein kinases (SAPKs). It has therefore been suggested that the JNK pathway is necessary in mediating MMP-9 production (Gum et al., 1997). However, c-Fos activation is not as strictly regulated. In this regard, the transcription factor Elk-1, when phosphorylated induces c-Fos transcription which can be mediated either by activation of JNK or extracellular-regulated protein kinase (ERK) depending on cell type and cellular stimuli (Cavigelli et al., 1995). Stimulation of the MMP-9 gene by tumor necrosis factor-alpha (TNF-α) is mediated partly through the NF-κB and Sp1 motifs located 600 and 558 nucleotides upstream of the transcriptional start site (Sato and Seiki, 1993). Gum et al. (1996) have shown that the mutation of previously undescribed AP-1 and PEA3 motifs located at −553 and −540, respectively, severely impairs the ability of ras to induce the MMP-9 gene in an ovarian cancer cell line. Thus, the cis elements of the promoter and the transacting factors regulating MMP-9 production seem to differ depending on the cell type and the stimulus.

Using transient transfection assays, we have earlier shown that MMP-9 promoter activity is regulated by both the ERK and JNK dependent signaling pathways (Lakka et al., 2000). It is likely that the ERK and JNK pathways are activated and work in concert to stimulate MMP-9 promoter activity in SNB19 glioblastoma cells. In this study we examined the effects of interfering with the signaling through the MAPK pathway on the expression of MMP-9 production and glioma invasion. We stably transfected SNB19 cells with a mutated ERK cDNA (mt ERK) to impair the expression of MMP-9. Our results suggest that ERK-dependent signaling molecules regulate MMP-9 expression in SNB19 cells and that interfering with this pathway reduces both MMP-9 synthesis and invasiveness in gliomas.

Results

Cell proliferation and ERK expression in parental cells, vector-transfected controls, and mt-ERK-transfected SNB19 cells

To determine any change in the growth pattern of SNB19 cell stably transfected with mt ERK, we performed a growth curve on these cells along with parental SNB19 cells and vector controls (Figure 1). The growth pattern of the mt ERK cells was similar to that of the parental and vector controls. Considering that ERK1 is constitutively activated in SNB19 cells, we first examined the activation status of the ERK subgroup of MAPK in the SNB19 glioblastoma cell line, a line that normally produces large amounts of MMP-9, by using Western blotting. Both parental SNB19 cells and cells that had been transfected with a construct mt-ERK encoding a kinase-inactive ERK1 (Frost et al., 1994) were treated with phorbol myristate acetate (PMA) (50 ng/ml), after which the cell extracts were subjected to 10% SDS–PAGE, transferred onto nitrocellulose membranes, and probed with an anti-ERK antibody that reacts with both human and murine ERK1 and ERK2. The results indicate that the parental SNB19 cells contained constitutively activated ERK1 and ERK2, the amounts of which were higher in the presence of PMA. The mt-ERK-transfected cells, in contrast, produced much lower levels of ERK in the presence of PMA (Figure 2).

Figure 1
figure1

Proliferation assay. Briefly 1×104 SNB19 parental, Vector controls (PCEP4) and mt ERK stable transfectant cells were seeded in 100 mm dishes and cell number was counted at the indicated time periods in triplicates. Mean (±s.d.) values from three separate experiments are shown

Figure 2
figure2

Western blot analysis of ERK protein expression. SNB19 cells were plated at 1×105 cells/well in six-well plates. The cells were treated with 200 nm of PMA wherever indicated. Total cell lysate was collected, and equal amounts of protein (20 μg) from each treatment group were run on 10% SDS-polyacrylamide gels. Separated proteins were transferred onto nitrocellulose membranes and probed with Phospho-ERK monoclonal antibody in an enhanced chemiluminescence protocol

MMP-9 enzymatic activity and mRNA and protein levels in parental cells and stable transfectants by gelatin zymography, Northern and Western blotting

As expected, parental cells and cells transfected with empty vector (PCEP4) expressed MMP-9 in response to PMA, whereas in the absence of PMA these cells expressed only MMP-2 (Figure 3A). In contrast, the stable mt-ERK clones expressed very little or no MMP-9 activity and showed no significant change in MMP-2 activity in the presence of PMA. Quantitatively, MMP-9 enzymatic activity was 10–14 times higher in the parental cells and vector controls than in the mt-ERK clones. After gelatin zymography confirmed that MMP-9 activity was reduced in the mt-ERK clones, we assessed MMP-9 protein levels in parental cells, PECP4 vector controls, and mt-ERK clones by Western blotting with antibodies specific to MMP-9. Use of an anti-MMP-9 antibody revealed that MMP-9 protein was present in the parental and vector-control cells treated with PMA, but MMP-9 protein was not detected in the mt-ERK transfectants regardless of PMA treatment (Figure 3B). We further characterized mRNA expression in parental cells, vector controls, and mt-ERK transfected clones. The parental cells and the vector controls expressed MMP-9 mRNA in the presence of PMA (Figure 3C). In contrast, we could not detect MMP-9, with or without PMA stimulation, in the mt-ERK transfected cells. Probing the stripped membranes with GAPDH verified that similar amounts of mRNA had been loaded onto the Northern blots. Quantification by densitometry showed that the amount of MMP-9 mRNA was 10–12 times (P<0.001) higher in parental and vector clones compared to the mt-ERK clones.

Figure 3
figure3

Determination of MMP-9 enzyme activity, protein and mRNA expression in SNB19, vector, and ERKmt construct transfected cells. (A) Gelatin zymograms of the conditioned medium from parental SNB19, empty vector (PCEP4) and ERK mt clones grown in the presence (*) and absence of PMA. Conditioned medium containing equal amounts of protein (20 μg) was mixed with Laemmli sample buffer and run on 10% SDS-polyacrylamide gels containing 0.1% gelatin. (B) Western blot analysis of MMP-9 protein expression. SNB19 cells were plated at 1×105 cells/well in six-well plates in the presence or absence of PMA. Conditioned medium was collected, and equal amounts of protein (20 μg) from each treatment group were run on 10% SDS-polyacrylamide gels. Separated proteins were transferred onto nitrocellulose membranes and probed with an anti-MMP-9 monoclonal antibody in an enhanced chemiluminescence protocol. (C) Northern blot analysis of MMP-9 expression by SNB19 and transfected cells. Total RNA was isolated and 10 μg was electrophoresed on 1.2% agarose gels, blotted onto a nylon membrane, and hybridized with a 32P-labeled MMP-9 cDNA probe. After removal of the radiolabeled probe, the membrane was rehybridized with a GAPDH cDNA probe to check the relative amounts of mRNA that had been loaded onto the gels

Electrophoretic mobility shift assay for Ap-1 and NF-κB

Transcription of the MMP-9 gene yields a 2.3 kb mRNA (Wilhelm et al., 1989) and is regulated by a 670 bp upstream sequence (Sato and Seiki, 1993), which includes motifs corresponding to transcription factors Ap-1, NF-κB, PEA3 and SP-1 binding sites. The synthesis and activity of transcription factors in the Ap-1 (Bogoyevitch et al., 1995; Agarwal et al., 1995) and NF-κB (Li and Sedivy, 1993) can be regulated by the extracellular signal-regulated kinase. Considering the importance of the Ap-1 sites at −533 (Gum et al., 1996) and −79 (Sato et al., 1993) and the NF-κB site at −600 (Sato and Seiki, 1993) in the regulation of MMP-9 promoter activity, we performed electrophoretic mobility shift assays with nuclear extracts of the parental, vector and mt-ERK stable clones to identify whether the extracts bound to any of these sites. Figure 4 shows a specific AP-1 complex that was present after incubation of the radioactive oligonucleotide with the nuclear extract in parental and vector clones, and this complex was slightly increased in the presence of PMA. In contrast, this complex was undetectable in mt-ERK clones even in the presence of PMA. A specific NF-κB complex was present in parental and vector clones and this complex was increased in the presence of PMA; but this specific NF-κB complex was undetectable in mt-ERK clones irrespective of PMA treatment (Figure 5). The binding of the transcription factors to the oligonucleotide was specific since a 100-fold excess of cold-oligonucleotide competed for the binding proteins to the radiolabeled oligonucleotide and this band was absent in the oligonucleotide not incubated with nuclear extract (Figures 4 and 5).

Figure 4
figure4

Binding of SNB19 nuclear extracts (NE) with MMP-9 promoter-specific Ap1 oligonucleotides. NE extracts were prepared as described in Materials and methods from SNB19 and transfected cell with and without treatment with PMA. Double-stranded oligonucleotides (50 fmol) were incubated with nuclear extracts (4 μg) or a 100-fold molar excess of unlabeled cold probes

Figure 5
figure5

Binding of SNB19 nuclear extracts (NE) with NF-κB oligonucleotides. NE extracts were prepared as described in Materials and methods from SNB19, ERK-mt and transfected cells with and without treatment with PMA. Double-stranded oligonucleotides (50 fmol) were incubated with nuclear extracts (4 μg) or a 100-fold molar excess of unlabeled cold probes

Invasiveness of mt-ERK-transfected SNB19 cells

Since these findings suggest that MMP-9 is regulated by an ERK-dependent mechanism, we next investigated whether interfering with this regulatory pathway would influence the invasive capacity of this glioma cell line. When cells were plated at a density of 0.5×106 cells/ml in the upper chamber, staining of mt-ERK cells that invaded through the Matrigel was much less compared to parental and vector clones (Figure 6A). Quantitative analysis of the number of cells by 3–(4,5-dimethylthiazol-2-yl)–2,5-diphenyltetrazolium bromide assay (Mohanam et al., 1993) showed that 50% parental cells and 53% vector cells invaded to the lower side of the membrane at 48 h (Figure 6B). In contrast only 13% of mt-ERK cells invaded, with a significant reduction (P<0.001) compared to parental and vector controls.

Figure 6
figure6

Matrigel invasion assay. SNB19 cells and transfected clones were allowed to invade through transwell inserts (8-μm pores) coated with 0.7 mg/ml Matrigel for 24 h. Cells in the lower chamber were then stained (A) and quantified (B) as described in the text. Each value is the mean±s.d. of five different experiments. P<0.001

In co-culture assay, spheroids consisting of parental SNB19 cells, vector-transfected cells (PCEP4), or mt-ERK-transfected cells were stained with DiI (which fluoresces red) and cultured with fetal rat brain cell aggregates, which were stained with DiO (which fluoresces green). The parental- and vector-cell spheroids merged with the rat brain aggregates within 24 h, and by 48 and 72 h the tumor spheroids had progressively invaded the rat brain aggregates, thereby reducing the brain aggregate volume. In contrast, the mt-ERK-transfected spheroids failed to invade rat brain aggregates even at 72 h, although the two cell types did become attached to each other after 24 h (Figure 7A). Quantitative analysis of the fetal brain aggregate volume remaining after coculture indicated only 10–15% remaining brain volume after coculture with parental cells, vector controls, compared with 95–98% in mt-ERK clones (Figure 7B). The invasiveness of fetal rat brain aggregates was significantly decreased in mt-ERK clones compared with parental and vector clones (P<0.001).

Figure 7
figure7

(A) Confocal laser scanning images of SNB19 spheroids and rat brain aggregates cocultures scanned at a depth of 100 μm at 24, 48 and 72 h. Visible are the rat brain aggregates (green fluorescence) and the parental or vector-transfected or ERK mt transfected tumor cell spheroids (red fluorescence). (B) Quantitation of remaining fetal rat brain aggregates by tumor spheroids as described in Materials and methods. Data shown are the mean±s.d. from four separate experiments from each group (P<0.001)

Discussion

Although both positive and negative control of collagenase activity at the appropriate site is essential for normal tissue development, abnormal expression of collagenases may contribute to disease processes such as tumor growth and metastasis. Several reports have shown that glioblastomas produce significantly higher levels of MMP-9 than do lower-grade tumors and normal brain tissue (Rao et al., 1993, 1996; Forsyth et al., 1998; Nakada et al., 1999). We recently demonstrated that stably transfected antisense MMP-9 glioblastoma cells produce less MMP-9 protein and mRNA and they are less invasive both in vitro and in vivo models (Kondraganti et al., 2000). In the current study, we sought to determine the involvement of ERK in the regulation of MMP-9 production in glioma cells invasiveness. We utilized a dominant negative mutant ERK to characterize the role of endogenous ERK in mediating glioma invasion as well as in mediating endogenous AP-1 activity, which has been shown to be essential for MMP-9 gene expression. In the present study, mt-ERK clones showed decreased levels of MMP-9, enzymatic activity, protein and mRNA levels compared to parental and vector clones. Further, mt-ERK cells were much less invasive in Matrigel and spheroid assays than parental and vector cells even in the presence of PMA.

The results of the present study showed that AP-1 and NF-κB binding transcription factors were significantly reduced in ERK1-mutated clones even in the presence of PMA. The Ap-1 complex was shown to be essential in the production of several MMPs including MMP-9 (Sato and Seiki, 1993). Gum et al. (1996) showed that mutation of previously undescribed AP-1 and PEA3 motifs located at −553 and −540 respectively, severely impaired the ability of ras to induce MMP-9 in an ovarian cancer cell line. Genersch et al. (2000) suggested that AP-1 site located at −79, which was described as being important for v-src- and phorbol ester-specific induction of MMP-9 expression in tumor cells (Gum et al., 1996; Sato and Seiki, 1993) was unlikely to play a role in endothelial cells. However, c-jun activation is ERK independent and activated by a distinct MAPK signaling module, SEK/JNK (Derijard, 1994). Therefore, it seems reasonable that co-ordinate activation of ERK and JNK pathways are necessary to generate an increase in MMP-9 production. One possible mechanism by which sustained MAPK activation could result in MMP-9 induction is through regulation of an essential transcription factor such as c-fos. Expression of this immediate early gene depends on MAPK activation; further phosphorylation of c-fos by MAPK enhances MMP-9 expression (Chen et al., 1994). Thus, in one possible signaling scenario, initial activation may serve to enhance c-fos transcriptional activity and Ap-1-dependent expression of the MMP-9 gene (Gum et al., 1997). In conjunction with our findings that JNK activation is not affected by blocking ERK activity (data not shown) it is possible that these pathways are activated independently and co-coordinately in SNB19 glioma cell line. Zeigler et al. (1999) have shown coordinate and prolonged activation of the endogenous JNK as well as ERK pathways are necessary to generate an increase in MMP-9 production in growth factor (HGF or EGF) stimulated keratinocytes.

A functional NF-κB site occurs in the proximal stimulatory region of the MMP-9 promoter (Sato and Seiki, 1993; Gum et al., 1996; Chintala et al., 1998; Lakka et al., 2000; Kondraganti et al., 2000) and deletion of this site reduces up-regulation of reporter gene constructs in response to phorbol ester and TNF-α. Bond et al. (2001) have shown an absolute requirement of NF-κB activity in addition to the essential role played by the AP-1 transcription factor, for MMP-9 production in rabbit and vascular smooth muscle cells in response to a wide variety of stimuli including cytokines, growth factors and phorbol esters. They have also noticed similar observation in rabbit dermal fibroblasts (Bond et al., 1998). Huang et al. (2001) have demonstrated that blocking of NF-κB activity in human prostate cancer cells is associated with suppression of angiogenesis, invasion, and metastasis. NF-κB signaling blockade significantly inhibited in vitro and in vivo expression of three major pro-angiogenic molecules, VEGF, IL-8 and MMP-9, and hence decreased neoplastic angiogenesis. They have further shown that inhibition of NF-κB activity in PC-3M cells also resulted in the down-regulation of MMP-9 mRNA and collagenase activity, resulting in decreased invasion through Matrigel (Huang et al., 2001).

In our earlier work (Lakka et al., 2000) we have shown that transient transfection of SNB19 cells with mt ERK or mt JNK repressed MMP-9 promoter activity suggesting that interfering with either pathway could result in inhibiting MMP-9 expression. The inability of the dn transfected cells to activate both these pathways to a certain threshold level might account for their inability to produce MMP-9. Regulation of MMP-9 activation by various stimuli in different cellular settings may involve different signal transduction pathways. Inhibition of ERK by MEK specific inhibitors blocked MMP-9 expression in breast cancer cells (Yao et al., 2001) and decreased MMP-9 production and attenuated the in vivo invasiveness in head and neck squamous carcinoma cells (Simon et al., 1998). MMP-9 expression was completely abolished by blocking p38 activation in breast and ovarian cancer cell lines (Yao et al., 2001; Simon et al., 2001). The Ras-dependent expression of MMP-9 in OVCAR-3 cells, studied by the effect of transient expression of activated Ha-ras on MMP-9 expression, was found to be mediated through a MEK-1-independent signaling pathway (Gum et al., 1996). In contrast, disruption of ERK- and JNK-dependent signaling was shown to decrease endogenous constitutive MMP-9 expression in æM-SCC-1 cells (Gum et al., 1997). In keratinocytes, JNK and p38 MAPK cascades were described as being insufficient for growth factor-induced MMP-9 expression. Instead, prolonged activation of ERK signaling was thought to be necessary to induce MMP-9 expression by growth factors (Mccawley et al., 1999; Zeigler et al., 1999). In endothelial cells, ERK1/2 has been reported to represent the convergence point for TNF-α as well as PMA-induced expression of MMP-9 (Genersch et al., 2000). In the present study we show that interfering with the ERK signaling pathway inhibits MMP-9 expression and glioma invasion in vitro. Since there are more than one signaling pathway molecules involved in MMP-9 expression, there could be a cross-talk between signaling pathways so that blocking one of the signaling pathways may affect the activation/inhibition of another pathway. It is also possible that signals from different signaling pathway molecules could converge on a common target whose activation turns on MMP-9 expression. We are in the process of investigating these possibilities.

In the present study we demonstrate that PMA activates the classical mitogenic MEK/ERK since stimulation with PMA caused phosphorylation of ERK1/2 in SNB19 cells. This phosphorylation was significantly decreased in SNB19 cells transfected with mt ERK. On the other hand phorbol ester, which is a potent stimulus of MMP-9 expression is via c-Raf-1 (Sozeri et al., 1992), an ERK1 but not a JNK (Cavigelli et al., 1995) activator. In the present study we could not detect MMP-9 protein and mRNA expression in mt ERK stable transfected SNB19 cells even with PMA stimulation. Thus it is more likely that its inductive effects on synthesis of this collagenases is propagated along an ERK-dependent signaling cascade. With respect to transcriptional control of MMP-9 gene, we could show that the already extensively studied AP-1 sites at −79 and −553 and NF-κB at −600 are involved in PMA stimulated MMP-9 expression in SNB19 glioblastoma cells. This finding is supported by other studies by Sato and Seiki (1993) wherein they show that induction of MMP-9 promoter activity brought about by PMA treatment required the AP-1, NF-κB and Sp-1 motifs. More apparently our results support the conclusion that blocking or interrupting the ERK1 signaling cascade may play an important role in affecting cellular responses such as migration, MMP-9 expression and invasive capacity that require de novo MMP-9 protein and mRNA expression.

Materials and methods

Cell lines and DNA construct

SNB19 cell lines were purchased from the American Type Culture Collection (Manassas, VA, USA) and maintained in Dulbecco's modified Eagle's medium (DMEM) F12 containing 10% fetal calf serum, 100 μg/ml streptomycin, and 100 units/ml penicillin in a humidified atmosphere containing 5% CO2 at 37°C. The ERK1 construct encodes the ERK1 cDNA, in which the conserved lysine at codon 71 was changed to arginine, thus impairing the catalytic efficiency of the enzyme (Frost et al, 1996).

Transient transfections

Glioblastoma cells were cotransfected at 70% confluence with the PCEP4 vector containing the mutated cDNA for ERK (mt ERK) in serum-free medium; control cells were transfected with the empty vector (PCEP4). The medium was replaced with serum-containing medium after 5 h and allowed to incubate for 48 h. The growth of cells in medium containing hygromycin allowed selection of drug-resistant colonies that had stably integrated the hygromycin-resistant gene-containing vector.

Proliferation assay

Cell proliferation was assessed by seeding 1×104 of SNB19, empty vector (PCEP4) and mt ERK transfected cells in 100 mm tissue culture dishes. Cell number was counted at the indicated times in triplicates.

Gelatin zymography

Protein (20 μg) from culture supernatants was denatured in the absence of a reducing agent and subjected to electrophoresis on a 7.5% polyacrylamide gel containing 0.1% (w/v) gelatin. The gel was incubated at room temperature for 2 h in the presence of 2.5% Triton X-100 and subsequently at 37°C overnight in a buffer containing 10 mM CaCl2, 0.15 M NaCl, and 50 mM Tris (pH 7.5). The gel was stained for protein with 0.25% Coomassie blue and photographed in a light box. Proteolysis was detected as a white zone in a dark field.

Western blot analysis

For immunoblotting studies, cell extracts from the parental cells and stable transfectants were generated in a phosphate-buffered saline solution containing 1% NP40, 0.5% sodium deoxycholate, 0.1% SDS, and 1 μM PMSF (RIPA buffer), denatured in the presence of a reducing agent and subjected to electrophoresis on a 10.5% SDS–PAGE gel. The resolved proteins were transferred to a nitrocellulose membrane. The membrane was then blocked with a solution containing 3.0% bovine serum albumin and incubated with antibodies for ERKs (#93, Santa Cruz Biotechnology) and MMP-9 (Oncogene Science, Cambridge, MA, USA) and an anti-rabbit horseradish-peroxidase conjugate. Reactive proteins were visualized by electrochemiluminescence (Amersham, Arlington Heights, IL, USA) according to the manufacturer's recommendations.

Nuclear extraction procedure

Nuclear extracts were prepared as described elsewhere (Schreiber et al., 1989; Chaturvedi et al., 1994). Briefly, 2×106 cells were treated with 300–400 μl of lysis buffer (50 mM KCl, 0.5% Nonidet P-40, 25 mM HEPES (N-2 hydroxyethylpiperazine-N′-2 ethanesulfonic acid, pH 7.8), 25 mM phenylmethylsulfonyl fluoride, 10 μg/ml of leupeptin, 20 μg/ml of aprotinin, and 100 μM dithiothreitol) on ice for 4 min. After 1 min of centrifugation at 14 000 r.p.m., the supernatant was removed and saved as a cytoplasmic extract. The nuclei were washed once with the same volume of buffer without Nonidet P40 and then 50–100 μl of extraction buffer (500 mM KCl and 10% glycerol with the same concentrations of HEPES, phenylmethylsulfonyl fluoride, leupeptin, aprotinin, and dithiothreitol as in the lysis buffer) was added depending on pellet size and mixed by pipetting several times. After centrifugation at 14 000 r.p.m. for 5 min, the supernatant was harvested as a nuclear protein extract and stored at −70°C. Protein concentration was quantified with a protein assay reagent kit (Bio-Rad, CA, USA).

Electrophoretic mobility shift assay

Electrophoretic mobility shift assay (EMSA) was performed by incubating 4 μg of nuclear extract (prepared as described above) for 15 min at 37°C with 8 fmol of 32P-end-labeled double-stranded oligonucleotide containing AP-1 and NF-κB binding site. The DNA–protein complex formed was separated from free oligonucleotide on 7% native polyacrylamide gels, and the gel was then dried. The film was developed after overnight exposure at −70°C. For supershift assays, the nuclear extracts were incubated with AP-1 for 30 min at 37°C before the complex was analysed by EMSA.

Matrigel invasion assay

Invasiveness of SNB19 cells was determined in vitro by measuring the distance that the cells invaded through Matrigel-coated (Collaborative Research, Inc., Boston, MA, USA) transwell inserts (Costar, Cambridge, MA, USA) according to a previously described procedure (Mohanam et al., 1993). Briefly, transwell inserts with an 8-μm pore size were coated with Matrigel in cold serum-free DMEM at a final concentration of 0.78 mg/ml. Cells were trypsinized and washed three times with serum-free medium, after which 1×106 cells were added in triplicate wells and incubated for 48 h. Cells that passed through the filters into the lower wells were quantified as described before (Mohanam et al., 1993) and expressed as a percentage at the sum of the cells in the upper and lower wells. Cells on the lower side of the membrane were fixed, stained with Hema-3 and photographed.

Spheroid invasion assays

Invasiveness of glioma spheroids was also measured in a three-dimensional model described previously (Go et al., 1997). In this model, glioma spheroids were created, stained with the fluorescent dye DiI, and co-cultured with fetal rat brain aggregates, which had been stained with DiO. At different intervals, serial 1-μm-thick sections were obtained from the surface through the center of the cocultures with a confocal laser-scanning microscope. DiI fluorescence was detected by using an argon laser at 488 nm with a band-pass filter at 520–560 nm, and DiO fluorescence with a helium/neon laser at 543 nm with a long-pass filter at 590 nm. The volumes of fetal brain aggregate or tumor spheroid that remained at 24, 48 and 72 h of cocultures were quantified with the formula as described previously (Kondraganti et al., 2000). Briefly, the areas of red or green fluorescence, indicating the amounts of spheroid or brain aggregate remaining, are calculated in each confocal plane by using NIH Image, a software package from the National Institutes of Health, and those areas are multiplied by the distance between each sample section. The total volume of the spheroid or remaining brain aggregate is then calculated by adding all of the individual volumes down to the center of the spheroid or brain aggregate. The calculated volume is then doubled to approximate the complete brain aggregate volume.

References

  1. Agarwal S, Corbley MJ, Roberts TM . 1995 Oncogene 11: 427–438

  2. Bogoyevitch MA, Ketterman AJ, Sugden PH . 1995 J. Biol. Chem. 270: 29710–29717

  3. Bond M, Chase AJ, Baker AH, Newby AC . 2001 CardVas. Res. 50: 556–565

  4. Bond M, Fabunmi RP, Baker AH, Newby AC . 1998 FEBS Letts. 435: 29–34

  5. Cavigelli M, Dolfi F, Claret F, Karin M . 1995 EMBO J. 14: 5957–5964

  6. Chaturvedi MM, LaPushin R, Aggarwal BB . 1994 J. Biol. Chem. 269: 14575–14583

  7. Chen Q, Kinch MS, Lin TH, Burridge K, Juliano RL . 1994 J. Biol. Chem. 6: 26602–26605

  8. Chintala SK, Sawaya R, Aggarwal BB, Majumder S, Giri DK, Kyritsis AP, Gokaslan ZL, Rao JS . 1998 J. Biol. Chem. 273: 13545–13551

  9. Derijard B, Hibi M, Wu IM, Barrett T, Su B, Deng T, Karim M, Davis RJ . 1994 Cell 76: 1025–1037

  10. Forsyth PA, Dickinson-Laing T, Gibson A, Rewcastle NB, Sutherland G, Johnston R . 1998 J. Neuro-Oncol. 36: 21–29

  11. Frost JA, Geppert TD, Cobb MH, Feramisco JR . 1994 Proc. Natl. Acad. Sci. USA 91: 3844–3848

  12. Genersch E, Hayess K, Nuenfeld Y, Haller H . 2000 J. Cell Sci. 113: 4319–4330

  13. Go Y, Chintala SK, Mohanam S, Gokaslan ZL, Bjerkvig R, Oka K, Nicolson GL, Sawaya R, Rao JS . 1997 Clin. Exp. Metastasis 15: 440–446

  14. Gum R, Lengyel E, Juarez J, Chen JH, Sato H, Seiki M, Boyd D . 1996 J. Biol. Chem. 271: 10672–10682

  15. Gum R, Wang H, Lengyel E, Juarez J, Boyd D . 1997 Oncogene 14: 1481–1493

  16. Huang S, Pettaway CA, Uehara H, Bucana CD, Fidler IJ . 2001 Oncogene 20: 4188–4197

  17. Karin M, Liu ZG, Zandi E . 1997 Curr. Opin. Cell Biol. 9: 24–246

  18. Kondraganti S, Mohanam S, Chintala SK, Kin Y, Jasti SL, Chandrasekar N, Lakka SS, Adachi Y, Kyritsis AP, Ali-Osman F, Sawaya R, Fuller GN, Rao JS . 2000 Cancer Res. 60: 6851–6855

  19. Lakka SS, Jasti SL, Kyritsis AP, Yung WKA, Ali-Osman F, Rao JS . 2000 Clin. Exp. Metastasis 18: 245–252

  20. Li S, Sedivy JM . 1993 Proc. Natl. Acad. Sci. USA 90: 9247–9251

  21. Matrisian LM . 1992 Bioessays 14: 455–463

  22. McCawley LJ, Li S, Wattenberg EV, Hudson LG . 1999 J. Biol. Chem. 274: 4337–4353

  23. Mohanam S, Sawaya R, McCutcheon I, Ali-Osman F, Boyd D, Rao JS . 1993 Cancer Res. 53: 4143–4147

  24. Nakada M, Nakamura H, Ikeda E, Fujimoto N, Yamashita J, Sato H, Seiki M, Okada Y . 1999 Am. J. Pathol. 154: 417–428

  25. Rao JS, Steck PA, Mohanam S, Stetler-Stevenson WG, Liotta LA, Sawaya R . 1993 Cancer Res. 53: 2208–2211

  26. Rao JS, Yamamoto M, Mohanam S, Gokaslan ZL, Stetler-Stevenson WG, Rao VH, Fuller GN, Liotta LA, Nicholson GL, Sawaya R . 1996 Clin. Exp. Metastasis 14: 12–18

  27. Robinson MJ, Cobb MH . 1997 Curr. Opin. Cell Biol. 9: 180–186

  28. Sato H, Kita M, Seiki M . 1993 J. Biol. Chem. 268: 23460–23468

  29. Sato H, Seiki M . 1993 Oncogene 8: 395–405

  30. Schreiber E, Matthias P, Muller MM, Schaffner W . 1989 Nucleic Acids Res. 17: 6419

  31. Simon C, Goepfert H, Boyd D . 1998 Cancer Res. 58: 1135–1139

  32. Simon C, Simon M, Vucelic G, Hicks MJ, Plinkert PK, Koitschev A, Zenner P . 2001 Exp. Cell. Res. 271: 344–355

  33. Smeal T, Angel P, Meek J, Karin M . 1989 Genes Dev. 3: 2091–2100

  34. Sozeri O, Vollmer K, Liyanage M, Frith D, Kour G, Mark III GE, Stabel S . 1992 Oncogene 11: 2259–2262

  35. Whisler RL, Chen M, Beiquing L, Carle KW . 1997 Cell Immunol. 175: 41–50

  36. Wilhelm SM, Collier IE, Marmer BL, Eisen AZ, Grant G, Goldberg G . 1989 J. Biol. Chem. 264: 17213–17221

  37. Yao J, Xiong S, Klos K, Nguyen N, Grijalva R, Li P, Yu D . 2001 Oncogene 20: 8066–8074

  38. Zeigler ME, Chi Y, Schmidt T, Varani J . 1999 J. Cell Physiol. 18: 271–284

Download references

Acknowledgements

Supported by National Cancer Institute Grants CA-76350, CA-75557 and CA-85216 (JS Rao).

Author information

Correspondence to Jasti S Rao.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Lakka, S., Jasti, S., Gondi, C. et al. Downregulation of MMP-9 in ERK-mutated stable transfectants inhibits glioma invasion in vitro. Oncogene 21, 5601–5608 (2002) doi:10.1038/sj.onc.1205646

Download citation

Keywords

  • extracellular signal regulated kinase
  • invasion
  • gliomas
  • stable transfectants
  • MMP-9

Further reading