Glioblastoma is a severe type of primary brain tumor, and its highly invasive character is considered to be a major therapeutic obstacle. Several recent studies have reported that ionizing radiation (IR) enhances the invasion of tumor cells, but the mechanisms for this effect are not well understood. In this study, we investigated the possible signaling mechanisms involved in IR-induced invasion of glioma cells. IR increased the matrix metalloproteinase (MMP)-2 promoter activity, mRNA transcription, and protein secretion along with the invasiveness of glioma cells lacking functional PTEN (U87, U251, U373, and C6) but not those harboring wild-type (WT)-PTEN (LN18 and LN428). IR activated phosphatidylinositol 3-kinase (PI3K), Akt, and mammalian target of rapamycin, and blockade of these kinases by specific inhibitors (LY294002, Akt inhibitor IV, and rapamycin, respectively) and transfection of dominant-negative (DN) mutants (DN-p85 and DN-Akt) or WT-PTEN suppressed the IR-induced MMP-2 secretion in U251 and U373 cells. In addition, inhibitors of epidermal growth factor receptor (EGFR; AG490 and AG1478), Src (PP2), and p38 (SB203580), EGFR neutralizing antibody, and transfection of DN-Src and DN-p38 significantly blocked IR-induced Akt phosphorylation and MMP-2 secretion. IR-induced activation of EGFR was suppressed by PP2, whereas LY294002 and SB203580 did not affect the activations of p38 and PI3K, respectively. Finally, these kinase inhibitors significantly reduced the IR-induced invasiveness of these cells on Matrigel. Taken together, our findings suggest that IR induces Src-dependent EGFR activation, which triggers the p38/Akt and PI3K/Akt signaling pathways, leading to increased MMP-2 expression and heightened invasiveness of PTEN mutant glioma cells. (Cancer Res 2006; 66(17): 8511-9)
- ionizing radiation
- phosphatidylinositol 3-kinase
Malignant glioma, the most common primary central nervous system tumor, is one of the most aggressive and lethal neoplasms. One of the major limitations in the treatment of glioma is the high prevalence of intrinsic or acquired resistance against radiation ( 1), which is likely associated with intrinsic cellular radioresistance, rapid cellular proliferation, and high invasiveness ( 2, 3). Among these biological factors, local invasive growth is a key feature of malignant brain tumors, showing a high incidence of recurrence even following drastic surgical resection ( 4, 5). Thus, it is important to understand the mechanisms of tumor invasiveness within brain, because this may lead to new strategies for curative treatment of glioma.
Tumor cell invasion involves the proteolytic degradation of extracellular matrix (ECM) components by tumor cell-secreted proteases, including serine proteases, plasminogen activators, and matrix metalloproteinases (MMPs; refs. 6, 7). Elevated levels of MMPs have been found in many tumors and are believed to play an important role in cellular invasion and metastasis ( 6– 8). Previous studies have shown that glioma cells secrete MMPs and their endogenous inhibitors, tissue inhibitors of metalloproteinases (TIMPs), and that high MMP expression is correlated with high aggressiveness in brain tumors ( 7, 9– 11). In particular, increased expression and activity of MMP-2 (72-kDa gelatinase A) and MMP-9 (92-kDa gelatinase B) have been correlated with an increased grade of glioma malignancy ( 10, 12– 16). Expression of MMP-2 is high in glioma cells in vitro and seems to be correlated with the invasive character of these cells ( 12– 14), whereas MMP-9 expression is induced by various stimuli, such as inflammatory cytokines and growth factors ( 15, 16). Thus, MMP-2 and MMP-9 are intriguing candidates for therapeutic manipulation in invasive gliomas.
Glioma invasiveness may be mediated by some of the genetic events commonly found in these tumors. In gliomas, tumor suppressor genes, such as PTEN and p53, are frequently mutated or deleted, whereas oncogenes, such as epidermal growth factor receptor (EGFR) and platelet-derived growth factor receptor (PDGFR), are commonly overexpressed. These genetic alterations can lead to activation of invasiveness-related signal transduction pathways, including focal adhesion kinase (FAK), Ras/mitogen activated protein kinase (MAPK), and phosphatidylinositol 3-kinase (PI3K)/Akt pathways ( 17, 18). The latter pathway is regulated by PTEN (also called MMAC1), a tumor suppressor gene located on human chromosome 10q23.3 ( 19), which is known to regulate cell growth, apoptosis ( 20), and interactions with the ECM, and has been shown to inhibit cell migration, spreading, and focal adhesion ( 21). PTEN dephosphorylates two lipid signal transduction molecules (phosphatidylinositol 3,4,5-triphosphate and phosphatidylinositol 3,4-bisphosphate) involved in the PI3K/Akt pathway ( 22) and exhibits protein phosphatase with dual specificity in vitro against its potential cellular target, FAK ( 23). PTEN has been shown to suppress MMP-2 expression and invasion of glioma cells in a phosphatase-dependent manner ( 24, 25) and has recently been reported to suppress mutant EGFR type III–mediated invasion of glioma cells by blocking FAK phosphorylation ( 26). Furthermore, we recently showed that hyaluronic acid induces expression of MMP-9 and osteopontin and consequent invasion and/or migration of glioma cells and that these events required the protein and lipid phosphatase activities of PTEN, respectively ( 25, 27). Collectively, these earlier findings suggest that the PI3K/Akt pathway might be a good target for new strategies aimed at controlling glioma invasion and migration.
Various stimuli, such as ionizing radiation (IR), growth factor/receptor interactions, and ECM/integrin interactions, have been shown to increase the invasiveness of glioma cells ( 28– 31). Exposure of cancer cells to IR stimulates the expression of MMP-2, MMP-9, Bcl-2, and integrin family members and activates multiple signaling pathways involved in controlling cell survival and repopulation ( 28– 34). However, the precise regulatory mechanisms responsible for IR-induced MMP expression and cancer cell invasion are not well understood.
In the present study, we investigated the some possible mechanisms for the effect of IR on MMP-2 secretion and invasion of glioma cells and found that IR up-regulates glioma cell invasion by inducing MMP-2 and that the Src/EGFR-mediated p38/Akt and PI3K/Akt pathways are critically involved in these events. Our findings suggest that these signaling pathways could be potential targets for improving the efficacy of radiotherapy for malignant gliomas.
Materials and Methods
Reagents and cell culture. Antibodies against EGFR, phosphorylated EGFR, PTEN, Src, and hemagglutinin (HA) were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Antibodies against MMP-2, MMP-9, TIMP-1, and TIMP-2 were from Calbiochem (La Jolla, CA). Antibodies against Akt, p38, and mammalian target of rapamycin (mTOR), and phosphorylated form of Src, Akt, mTOR, and p38 were from Cell Signaling Biotechnology (Beverly, MA). Antibody against p85 subunit of PI3K was purchased from Upstate Biotechnology (Lake Placid, NY). Specific inhibitors of EGFR (AG490 and AG1478), ErbB2 (AG825), PDGFR (SU5402), PI3K (LY294002), Src (PP2), mTOR (rapamycin), MAPK family [PD98059, SB203580 and SP600125, extracellular signal-regulated kinase 1/2 (ERK1/2), p38 and c-Jun NH2-terminal kinase (JNK), respectively], Akt (Akt inhibitor IV), MMP-2 (OA-Hy), and a broad-spectrum kinase inhibitor (genistein) were purchased from Calbiochem. Human glioma cell lines [U87, U251, and U373 (obtained from American Type Culture Collection) and LN18 and LN428 (generously gifts from Dr. Funari, Ludwig Institute for Cancer Research, La Jolla, CA)] and rat glioma cell line C6 were maintained in DMEM supplemented with 10% heat-inactivated fetal bovine serum, 100 units/mL penicillin, and 100 μg/mL streptomycin at 37°C in a humidified atmosphere containing 5% CO2.
Irradiation. Cells were plated in 30-mm dishes and incubated at 37°C under humidified 5% CO2 and 95% air in culture medium until 70% to 80% confluent. Cells were then exposed to γ-rays from a 137Cs γ-ray source [Atomic Energy of Canada, Korea Institute of Radiological and Medical Sciences (KIRAMS), Seoul, Korea] at a dose rate of 3.81 Gy/min.
Transfection and transduction. Wild-type (WT)-PTEN (obtained from Dr. Sun, Yale University, New Haven, CT), mutant PTEN (C124S, protein and lipid phosphatase activity-deficient mutant; generously gift from Dr. Whang, University of North Carolina School of Medicine, North Carolina, NC), dominant-negative (DN) mutant of p85 [DN-p85 (regulatory domain of PI3K)], DN-Src, DN-Akt (generous gift from Dr. Su-Jae Lee, KIRAMS, Seoul, Korea), and WT-MKK6 (generous gift from Dr. Aree Moon, Duksung Women's University, Seoul, Korea) plasmids were transiently transfected into U251 and U373 glioma cells using Effectene transfection reagent (Qiagen, Valencia, CA) by following the supplier's instructions. For retrovirus transduction, U251 and U373 cells were infected with control retrovirus and retrovirus harboring DN-38 and WT-p38 (generous gift from Dr. Su-Jae Lee) for 90 minutes at 37°C. In adenoviral transduction, cells were infected with 50 plaque-forming units per cell of control adenoviral LacZ and adenoviral WT-EGFR (generous gift from Dr. Seung-Hoon Lee, National Cancer Center, Goyang, Gyeonggi, Korea) for 90 minutes at 37°C. After 48 hours, transfected and transduced cells were used for further experiments. Expression of target genes was confirmed by Western blot analysis using the antibodies against target proteins, phosphorylated downstream effector molecules, and/or HA.
Reverse transcription-PCR (RT-PCR). Total RNA was isolated from glioma cells irradiated together with or without actinomycin D (a transcription blocker) or cyclohexamide (a translation blocker) using the RNeasy kit (Qiagen), and complimentary first-strand DNA was generated using the reverse transcription system kit (Bioneer, Daejeon, Korea) according to the manufacturer's protocol. The PCR conditions involved an initial denaturation step at 94°C for 5 minutes followed by 30 cycles at 94°C for 30 seconds, 55°C for 30 seconds, and 72°C for 30 seconds. Oligonucleotide primer sequences used were as follows: MMP-2 sense 5′-CAGGCTCTTCTCCTTTCACAAC-3′ and antisense 5′-AAGCCACGGCTTGGTTTTCCTC-3′ and β-actin sense 5′-CATGGGTCAGAAGGATTCCTAT-3′ and antisense 5′-GCGCTCGGTGAGGATCTTCAT-3′. PCR was done using the PCR Master kit (Bioneer) according to the manufacturer's protocol.
Luciferase reporter assays. pGL3 luciferase reporter vector, containing MMP-2 promoter regions, was provided by Dr. Sang-Oh Yoon (Korea Advanced Institute of Science and Technology, Taejon, Korea). pRL-cytomegalovirus (CMV) vector (Promega, Madison, WI) was used to evaluate transfection efficiency. Cells were seeded in 12-well plates and cultured at 37°C until they reach 50% to 60% confluency. Plasmids were then transiently cotransfected into glioma cells by using Effectene (Qiagen). After 24 hours, cells were irradiated at 5 Gy, and cell lysates were prepared at 6, 12, and 24 hours of incubation using Dual-Luciferase Assay kit (Promega). The luminescence was measured by using MicroLumatPlus LB96V microplate luminometer (EG&G Berthold, Wellesley, MA) according to the manufacturer's protocol.
Gelatin zymography. Glioma cells in subconfluent culture (∼70-80% cell density of confluent culture) were washed and refreshed with serum-free DMEM and then irradiated and incubated for 18 hours. In some experiments, cells were preincubated for 1 hour with various kinase inhibitors before irradiation. The enzymatic activity of electrophoretically separated gelatinolytic enzymes in the conditioned medium of glioma cells were determined by gelatin zymography as described previously ( 25). Zones of gelatinolytic activity were detected as clear bands against a blue background. Densitometric analysis was done using Scion Image NIH Image program.
Western blot analysis. Glioma cells were irradiated at various dose and lysed in lysis buffer as described previously ( 25). After a brief sonication, the lysates were clarified by centrifugation at 12,000 × g for 15 minutes at 4°C, and protein content was measured by the Bradford method. An aliquot (50 μg protein/lane) of the total protein was separated by 10% or 12% SDS-PAGE and blotted to nitrocellulose membrane (0.2 μm; Amersham, Arlington Heights, IL), and Western blot was done as described previously ( 25). Immunoblots were visualized by enhanced chemiluminescence (Amersham) according to the manufacturer's protocol.
Kinase assays. Cells were lysed in lysis buffer. Cell lysates were clarified by centrifugation, and equal amounts of the lysate proteins (500 μg) were immunoprecipitated with an antibody against the p85 subunit of PI3K or p38 and conjugated to protein A/G plus-agarose (Santa Cruz Biotechnology). In PI3K assay, the immune complexes were washed twice with PBS (pH 7.4) containing 1% NP40 and 1 mmol/L Na3VO4, twice with 100 mmol/L Tris-HCl (pH 7.5) containing 500 mmol/L LiCl and 1 mmol/L Na3VO4, and twice with 50 mmol/L Tris-HCl (pH 7.2) containing 150 mmol/L NaCl. The kinase reactions were started by adding 5 mg/mL l-α-phosphatidylinositol (Sigma, St. Louis, MO) in 20 mmol/L HEPES (pH 7.4), 5 mmol/L MnCl2, 10 μmol/L ATP, 5 μCi γ-[32P]ATP, and 2.5 mmol/L EGTA. After 20 minutes of incubation at room temperature, the reactions were quenched by adding 1 mol/L HCl. The phospholipids were extracted using a 1:1 mixture of chloroform and methanol and separated by TLC. Spots were visualized by autoradiography. In p38 kinase assay, the beads were washed thrice with a solution [20 mmol/L Tris-HCl (pH 7.5), 150 mmol/L NaCl, 1 mmol/L EGTA, 1 mmol/L EDTA, and 0.5% NP40] and once with a kinase assay buffer [50 mmol/L Tris-HCl (pH 7.5), 137 mmol/L NaCl, 1 mmol/L MgCl2, 1 mmol/L Na3VO4, 2.5 mmol/L β-glycerophosphate, 2 mmol/L EDTA, and 5 μmol/L ATP] and then subjected to kinase assay. p38 activities were measured in a reaction mixture consisting of a kinase assay buffer, 1 μg activating transcription factor-2 (ATF-2), and 5 μCi γ-[32P]ATP for 20 minutes at 30°C. The reaction was terminated by addition of SDS sample buffer, and the samples were subjected to 10% SDS-PAGE. Phosphorylated ATF-2 was visualized by autoradiography.
In vitro invasion assays. Invasion assays were done using modified Boyden chambers with polycarbonate Nucleopore membrane (Corning, Corning, NY) as described previously ( 25). Invasiveness was determined by counting cells in five microscopic fields per well, and the extent of invasion was expressed as an average number of cells per microscopic field.
IR up-regulates MMP-2 secretion and invasion in functional PTEN-deficient glioma cells. To investigate the role of IR in the invasiveness of glioma cells, we first studied the effect of IR on the secretion of MMPs and TIMPs in various glioma cell lines. Because IR did not trigger any cytotoxic effect on glioma cells after 24 hours of irradiation with up to 10 Gy in serum-free condition, we used <10 Gy and 24 hours of incubation time throughout this study. Following IR (5 Gy) of different human and rat glioma cell lines, gelatin zymography of conditioned medium and Western blot analysis of cell lysates showed significantly induced secretion and expression of MMP-2, but not MMP-9, in some of the cell lines ( Fig. 1A ). Interestingly, IR-induced secretion of MMP-2 was detected in functional PTEN-deficient glioma cell lines (U87, U251, U373, and C6), whereas it was not detectable in glioma cell lines harboring WT-PTEN (LN18 and LN428; Fig. 1A, top). Expression of TIMP-2, but not TIMP-1, was detected in all cell lines but dominantly expressed in LN18 and LN428 cells and IR did not affect the expression of TIMPs in these cells ( Fig. 1A, bottom). We then did Matrigel invasion assays to test the effect of IR on glioma cells invasiveness. As shown in Fig. 1B, IR increased the invasiveness of glioma cells lacking functional PTEN (U251 and U373) but not that of cells harboring WT-PTEN (LN18 and LN428). The IR-induced MMP-2 secretion, expression ( Fig. 1C, top), and in vitro invasion ( Fig. 1C, bottom) of glioma cells (U251 and U373) were dose dependent. In addition, IR-induced invasion was suppressed by the addition of synthetic MMP-2 inhibitor, OA-Hy ( Fig. 1D). Collectively, these results show that IR might induce invasion in glioma cells harboring mutant PTEN through up-regulation of MMP-2 secretion, and we used U251 and U373 glioma cells for further experiments.
IR-induced MMP-2 secretion in glioma cells is mediated at the transcriptional level. Next, we did RT-PCR analysis to determine whether the IR-induced increase of MMP-2 secretion was controlled at the level of gene expressions. As shown in Fig. 2A , IR (5 Gy) up-regulated transcription of MMP-2 mRNA in U251 and U373 glioma cells, and pretreatment with actinomycin D, but not cyclohexamide, blocked both the basal MMP-2 mRNA level and IR-induced MMP-2 up-regulation. Consistent with the results of our RT-PCR analysis, a luciferase assay revealed that IR significantly induced MMP-2 promoter activity in U251 cells ( Fig. 2B), indicating that IR up-regulates expression of the ECM-degrading protease, MMP-2, at the transcriptional level.
PTEN suppresses IR-induced MMP-2 secretion in U251 and U373 glioma cells. Because IR stimulated MMP-2 expression and secretion in glioma cells lacking functional PTEN, we next examined the role of PTEN in this response. Gelatin zymography showed that transfection of WT-PTEN significantly reduced basal secretion of MMP-2 as well as IR-induced MMP-2 up-regulation in U251 and U373 cell lines, whereas transfection of the PTEN mutant plasmid did not affect MMP-2 secretion ( Fig. 3A ). In addition, Western blot analysis showed that IR stimulated the phosphorylation of Akt in both cells, whereas both basal and IR-stimulated levels of Akt phosphorylation were decreased in cells expressing WT-PTEN but not in cells expressing mutant PTEN ( Fig. 3A). Because these data strongly suggested that the PI3K/Akt pathway is critical for IR-induced MMP-2 secretion in these cells, we tested the effect of specific inhibitors of these kinases on the IR-induced secretion of MMP-2. As shown in Fig. 3B, treatment of U251 and U373 cells with a specific inhibitor of PI3K (LY294002) or Akt (Akt inhibitor IV) significantly and dose-dependently blocked IR-stimulated secretion of MMP-2 and phosphorylation of Akt. To further confirm the involvement of the PI3K/Akt pathway, we transfected plasmids encoding HA-tagged DN forms of p85 (DN-p85), the regulatory domain of PI3K, and Akt (DN-Akt) into U251 and U373 cells, respectively. As shown in Fig. 3C, transfection of DN-p85 and DN-Akt suppressed IR-induced MMP-2 secretion and Akt phosphorylation in these cells. We also investigated the involvement of mTOR, a downstream effector molecule of Akt, in IR-induced MMP-2 secretion using rapamycin, a specific inhibitor of mTOR. Figure 3D shows that IR significantly induced mTOR phosphorylation, and pretreatment of U251 and U373 glioma cells with rapamycin suppressed these events in a dose-dependent manner. Taken together, these data strongly suggest that the PI3K/Akt/mTOR signaling pathway is required for IR-induced MMP-2 secretion and that PTEN modulates these events in U251 and U373 human glioma cells.
Src-dependent EGFR phosphorylation is involved in IR-induced MMP-2 secretion and Akt phosphorylation in U251 and U373 glioma cells. Receptor tyrosine kinases (RTK), such as EGFR, are often activated in response to irradiation in various carcinoma cells and are typically overexpressed in glioma cells ( 34, 35). IR also can activate non-RTKs, such as Src, which can signal downstream MAPK pathways in EGFR activation-dependent or activation-independent fashions ( 33, 35, 36). To investigate the involvement of RTKs in the IR-induced secretion of MMP-2, we pretreated U251 and U373 cells with various RTK inhibitors before IR and found that pretreatment of the glioma cells with specific inhibitors of EGFR (AG490 and AG1478) and a broad-spectrum RTK inhibitor (genistein), but not with specific inhibitor of ErbB2 (AG825) or PDGFR (SU5402), significantly inhibited IR-induced MMP-2 secretion and EGFR phosphorylation ( Fig. 4A ). To confirm the involvement of EGFR in this event, we transduced adenovirus harboring WT-EGFR and found that, regardless of IR, overexpression of WT-EGFR significantly enhanced the secretion of MMP-2 and phosphorylation of EGFR and Akt in both cell lines ( Fig. 4B). On the contrary, however, pretreatment of these cells with a neutralizing antibody against EGFR suppressed IR-induced MMP-2 secretion and phosphorylation of EGFR and Akt ( Fig. 4B), indicating that IR-induced EGFR activation is required for the up-regulation of MMP-2 through the Akt signaling pathway. We next examined the possible involvement of Src kinase in these events. Figure 4C shows that IR induced phosphorylation of Src kinase, whereas transfection of DN-Src or pretreatment with a specific inhibitor of Src (PP2) reduced IR-induced MMP-2 secretion and Src and Akt phosphorylation. In addition, PP2 completely blocked EGFR phosphorylation induced by IR, whereas AG1478 did not affect IR-induced phosphorylation of Src ( Fig. 4D), indicating that Src might be an upstream kinase of the EGFR/Akt signaling pathway in irradiated U251 and U373 glioma cells. Taken together, these data strongly suggest that IR-induced MMP-2 secretion is accomplished by Src/EGFR-mediated Akt signal transduction.
p38 is also involved in IR-induced Akt phosphorylation and MMP-2 secretion in U251 and U373 glioma cells. To further understand the mechanism of IR-mediated induction of MMP-2 secretion, we next used specific MAPK inhibitors to examine the roles of three MAPKs. Gelatin zymography showed that SB203580, a specific inhibitor of p38, but not PD98059 and SP600125, specific inhibitors of ERK1/2 and JNK, respectively, significantly abolished both basal and IR-mediated MMP-2 secretion in U251 and U373 human glioma cells ( Fig. 5A ). To confirm the involvement of p38 in IR-induced MMP-2 expression, we transiently infected both cell lines with retrovirus-mediated WT and DN-p38 expression constructs. As expected, infection with DN-p38 markedly reduced the basal and IR-induced secretion of MMP-2 in both cell lines ( Fig. 5B). In contrast, introduction of WT-p38 significantly enhanced the basal secretion of MMP-2, and irradiation further increased this MMP-2 secretion ( Fig. 5B). Interestingly, introduction of DN-p38 significantly suppressed both basal and IR-induced Akt phosphorylation levels, whereas WT-p38 up-regulated Akt phosphorylation ( Fig. 5B). To further confirm this effect, we transiently transfected the cells with a plasmid encoding WT-MKK6, an upstream kinase of p38, and found that expression of WT-MKK6 enhanced MMP-2 secretion, increased the activation of ATF-2 (a downstream target molecule of p38), and increased the basal and IR-induced phosphorylation levels of Akt ( Fig. 5C). Furthermore, pretreatment of cells with AG1478 and PP2, but not with LY294002, significantly suppressed IR-induced p38 kinase activity ( Fig. 5D, top). Finally, a PI3K assay revealed that treatment of cells with AG1478 and PP2, but not SB203580, significantly suppressed basal and IR-induced PI3K activity ( Fig. 5D, bottom). Collectively, these findings indicate that Src/EGFR-mediated MKK6/p38 activation is required for IR-induced Akt activation and subsequent MMP-2 secretion in U251 and U373 glioma cells independent of PI3K-mediated activation of Akt.
Specific inhibitors of Src, EGFR, p38, PI3K, and Akt suppress IR-induced invasion of U251 and U373 glioma cells. Because our results revealed that the Src, EGFR, p38, PI3K, and Akt kinases are critical for IR-induced MMP-2 up-regulation in U251 and U373 cells, we next examined whether the inhibition of these pathways could suppress the IR-induced invasion by these cells. As expected, treatment of U251 and U373 cells with PP2, AG1478, SB203580, LY294002, and Akt inhibitor IV, but not PD98059 or SP600125, ameliorated the IR-induced invasiveness of U251 and U373 glioma cells ( Fig. 6A ). These data strongly indicate that these signaling pathways are important for IR-induced MMP-2 up-regulation and increased glioma cell invasiveness. Therefore, it is highly possible that Src/EGFR-mediated p38/Akt and PI3K/Akt signaling pathways play a key role in IR-induced MMP-2 expression and subsequent invasiveness in U251 and U373 glioma cells ( Fig. 6B).
Despite aggressive treatment strategies, the prognosis of glioblastoma remains poor partly due to its high invasiveness into adjacent brain tissues. Here, we examined the regulatory mechanism of IR in MMP-2 secretion and invasion of glioma cells. We observed that IR enhanced glioma cell invasiveness and that this effect was at least partly mediated by increased MMP-2 expression in PTEN mutated glioma cell lines. Furthermore, IR was found to activate the PI3K/Akt and p38/Akt signaling pathways in U251 and U373 cells, with Src-mediated EGFR phosphorylation acting upstream of these pathways. Collectively, these findings seem to identify candidate signaling pathways that may be used as potential targets for controlling IR-induced glioma invasiveness.
The IR-induced invasiveness of PTEN mutant glioma cells required up-regulation of MMP-2. Elevated levels of MMP-2 have been reported in malignant gliomas ( 6– 12); most glioma cells constitutively express MMP-2, and its expression is correlated with the invasive potential of these tumor cells ( 11– 14, 24, 37, 38). Similar to other members of the metalloproteinase family, the synthesis and function of MMP-2 can be regulated by various factors, not all of which have been fully elucidated. Transforming growth factor-β ( 39), laminin, an ECM component ( 40), interleukin-8 ( 41), and insulin like growth factor-1 ( 42) have all been shown to alter the expression or function of MMP-2. In addition, IR potently induces MMP-2 in various normal cells ( 43– 45) and was also found to induce invasion and metastasis in several types of cancer cells in vitro and in vivo, including gliomas ( 28– 31).
Here, we found that IR potently induced MMP-2 expression and invasion in various glioma cells. Interestingly, IR-induced MMP-2 expression and invasion were restricted to cells lacking functional PTEN, and transfection of WT-PTEN, but not mutant PTEN, in these cells resulted in the suppression of these events. These findings strongly indicate the involvement of PI3K/Akt signaling pathway in IR-induced MMP-2 expression and invasion of glioma cells. Indeed, IR-activated PI3K/Akt pathway, treatment of specific inhibitors, or transfection of DN mutant forms of these kinases significantly reduced these events. In some ways, these findings are contradictory to the report of Wild-Bode et al., who reported that sublethal doses of irradiation up-regulated expression of MMP-2 and MT1-MMP, down-regulated expression of TIMP-2, and increased invasiveness in glioma cells (U87, LN229, and LN18) regardless of PTEN and p53 status ( 28). However, our finding is consistent with that of Cordes et al., who reported that IR did not affect invasion of LN-229 and LN-18 cells harboring WT-PTEN ( 32). These authors also found that MMP-2 and MT1-MMP were up-regulated and TIMP-2 was down-regulated in irradiated A172 and U138 cells having no PTEN protein; however, IR suppressed or did not affect invasion of these cells ( 32). In the present study, we found that a large amount of TIMPs expression was detected in glioma cell lines harboring WT-PTEN but not in cells with mutant PTEN ( Fig. 1B), and IR did not affect the secretion of TIMPs. Despite the large amount of MMP-2 expression, the basal invasive activity of WT-PTEN harboring cells was comparatively weaker than the cells with mutant PTEN, which might possibly be due to the concomitant expression of endogenous MMP inhibitors, TIMP-1 and TIMP-2. On the contrary, gelatin zymography showed high MMP-2 activity in LN18 and LN428 cells, which might be due to the dissociation of TIMP-2 by SDS on the gel during the process of the zymography as described in Materials and Methods. In addition, our previous study showed that introduction of WT-PTEN enhances TIMP-1 and TIMP-2 expressions in glioma cells lacking functional PTEN ( 25). Therefore, the present and earlier studies show some differences. These are most likely due to different experimental conditions and/or different cellular states; nevertheless, it is clear that IR enhances the invasion of glioma cells and MMP-2 might be a critical factor in this event and that PI3K/Akt signaling pathway is involved in this event, partly at least.
Another critical factor in glioma invasion seems to be EGFR, which is overexpressed in up to 50% of malignant glioma cells ( 35). When bound by its ligands (e.g., epidermal growth factor and transforming growth factor-α), EGFR is activated and triggers downstream signaling cascades. In addition, EGFR may be activated by irradiation in various cancer cells, including glioma ( 33– 35), and this effect is believed to be a major cause of radioresistance in glioma cells. Our data showed that IR potently induced EGFR phosphorylation and that blockade of EGFR activation by specific inhibitors and neutralizing antibodies effectively suppressed IR-induced MMP-2 expression and Akt phosphorylation. UV irradiation-mediated EGFR phosphorylation has recently been reported to depend on activation of Src, a non-RTK ( 36), suggesting that activation of Src may also mediate IR-induced EGFR activation. Consistent with this hypothesis, our data showed clearly that IR induced Src phosphorylation in glioma cells and that inhibition of Src phosphorylation by a specific inhibitor (PP2) or DN-Src suppressed IR-induced EGFR phosphorylation, PI3K activation, Akt phosphorylation, and MMP-2 expression. These results indicate that IR-induced MMP-2 expression is most likely accomplished at least partially through Src/EGFR-mediated PI3K/Akt signaling.
Because IR seems to preferentially transmit signals via EGFR to MAPK family members (i.e., ERK1/2, p38, and JNK; ref. 33), we examined the involvement of MAPKs in IR-induced MMP-2 expression and invasion. Of the specific MAPK inhibitors tested, only the p38 inhibitor inhibited IR-induced MMP-2 expression in U373 and U251 glioma cells. Furthermore, we found that IR activated p38 and introduction of DN-p38 blocked IR-induced MMP-2 expression, whereas introduction of WT-MKK6 (an upstream kinase of p38) or WT-p38 enhanced both basal and IR-induced MMP-2 expression. IR-induced activation of p38 was significantly blocked by Src and EGFR inhibitors and DN-Src, indicating that p38 acts as an another downstream target molecule of IR-induced Src/EGFR activation. Interestingly, blockade of p38 effectively suppressed IR-induced Akt phosphorylation but did not affect PI3K activity. In addition, PI3K activity did not affect IR-induced p38 kinase activity. Several studies have shown that p38 regulates Akt phosphorylation in a PI3K-dependent or PI3K-independent fashion under various experimental situations ( 46, 47). Therefore, p38 and PI3K seem to be another important kinases that signal to Akt pathway in glioma cells invasion, partly at least.
Although IR-induced MMP-2 expression and invasion were effectively controlled by the above-mentioned kinases, these signaling molecules were also found to be involved in basal MMP-2 expression and invasion of glioma cells. Furthermore, inhibition of these kinases did not completely reduce the IR-induced MMP-2 expression and invasion of glioma cells. Therefore, it is highly likely that these signaling molecules are not the only ones responsible for IR-induced MMP-2 expression and invasion of glioma cells. Further studies are needed to uncover other critical signaling molecules involved in these events.
In summary, our results indicate that IR induces Src-dependent EGFR activation, which triggers the p38/Akt and PI3K/Akt signaling pathways, leading to increased MMP-2 expression and heightened invasiveness of mutant PTEN glioma cells. Although future work will be required to elucidate whether this signaling occurs in vivo, these findings provide possible new targets for controlling the invasiveness of IR-treated glioma cells, perhaps facilitating a strategy for preventing radioresistance in this deadly cancer.
Grant support: National Nuclear R&D Program of Ministry of Science and Technology (Seoul, Korea).
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Note: C-M. Park and M-J. Park contributed equally to this study.
- Received December 5, 2005.
- Revision received June 19, 2006.
- Accepted June 27, 2006.
- ©2006 American Association for Cancer Research.