Downregulation of Spry2 by miR-21 triggers malignancy in human gliomas

Abstract

Gliomas are associated with high mortality because of their exceedingly invasive character. As these tumors acquire their invasiveness from low-grade tumors, it is very important to understand the detailed molecular mechanisms of invasion onset. Recent evidences suggest the significant role of microRNAs in tumor invasion. Thus, we hypothesized that deregulation of microRNAs may be important for the malignant progression of gliomas. We found that the aberrant expression of miR-21 is responsible for glioma invasion by disrupting the negative feedback circuit of Ras/MAPK signaling, which is mediated by Spry2. Upregulation of miR-21 was triggered by tumor microenvironmental factors such as hyaluronan and growth factors in glioma cells lacking functional phosphatase and tensin homolog (PTEN), but not harboring wild-type PTEN. Consistently with these in vitro results, Spry2 protein levels were significantly decreased in 79.7% of invasive WHO grade II–IV human glioma tissues, but not in non-invasive grade I and normal tissues. The Spry2 protein levels were not correlated with their mRNA levels, but inversely correlated with miR-21 levels. Taken together, these results suggest that the post-transcriptional regulation of Spry2 by miR-21 has an essential role on the malignant progression of human gliomas. Thus, Spry2 may be a novel therapeutic target for treating gliomas.

Introduction

The distinct ability of gliomas to infiltrate into the brain extracellular matrix makes it impossible to treat these tumors using surgery and radiation therapy. Thus, it is essential to identify key regulators of glioma invasion for the treatment of this incurable disease. Hyaluronan (HA), a principal glycosaminoglycan in the extracellular matrix of the brain, is a critical factor for glioma invasion. It facilitates cell adhesion, motility and proliferation through interactions with receptors, such as CD44 and RHAMM (Merzak et al., 1994; Toole 2004; Park et al., 2008). In spite of recent advances in our understanding of the proximal signal transduction pathways activated by HA, the molecular mechanism of HA-associated motility and invasion is not well understood. Previously, we reported that HA-induced metalloproteinase-9 (MMP-9) secretion and invasion of glioma cells lacking functional PTEN are mediated by ERK1/2 and NF-κB signaling (Park et al., 2002; Kim et al., 2005, 2008). However, it remains unclear how these signalings are amplified in a tumor-specific manner.

microRNAs (miRNAs) are small non-coding RNAs that regulate gene expression by binding to target mRNAs to inhibit their translation or cause their degradation. miRNAs can also affect cell growth, differentiation, apoptosis and so on (Kim and Nam, 2006). Recently, several miRNAs have been identified to be frequently deregulated in the progression of various cancers, and have emerged as potential targets for anti-cancer therapies (Croce, 2009; Negrini et al., 2009). Moreover, many studies have demonstrated the contribution of miRNAs to tumor migration and invasion (Nicoloso et al., 2009). The remarkable phenotypic differences between normal and cancer cells may be because of the differential regulation of intracellular signaling. As miRNAs may function in the fine-tuning of signaling by regulating the translation of signaling regulators, the differential expression of miRNAs may result in such distinct phenotypes. Thus, we hypothesized that the expression of certain miRNAs induced by HA differs between normal and cancer cells, which determines invasion by dissimilarly fine-tuning the corresponding signaling.

In this report, we have shown that HA induces miR-21 expression in glioma cells, but not in normal astrocytes, which increases the amplitude and duration of Ras/MAPK signaling through suppressing its negative feedback regulator, Spry2 expression. The increased signaling results in enhanced glioma invasion by increasing MMP-9 secretion.

Results

HA-induced miR-21 facilitates glioma invasion

As in our previous reports, HA treatment induced MMP-9 secretion (Figures 1a and c) and invasion (Figure 1c) of U87MG glioma cell lines (P<0.01). In contrast, human normal astrocytes did not display MMP-9 secretion (Figure 1a) or invasion (data not shown) after HA treatment. To identify miRNAs that may be involved in HA-induced glioma invasion, we performed miRNA microarray analysis using HA-treated glioma cell lines (U87MG and U373MG). Among those miRNAs that were significantly deregulated in both cell lines by HA treatment (Supplementary Figure 1), we were interested in miR-21, which has been reported to be an oncogene that is overexpressed in various tumors, especially in glioblastomas (Chan et al., 2005; Conti et al., 2009; Krichevsky and Gabriely, 2009). Increased expression of miR-21 has been implicated in carcinogenesis, including the inhibition of apoptosis, promotion of cell proliferation and stimulation of tumor growth by targeting PTEN, TPM1, PDCD4 and so on (Krichevsky and Gabriely, 2009). Recently, it was reported that glioma invasion is induced by increased miR-21 expression, but the mechanism by which this occurs remains unclear (Gabriely et al., 2008). Thus, we selected miR-21 as a candidate miRNA that may have an effect on HA-induced signaling. Indeed, HA treatment upregulated miR-21 in U87MG cells (P<0.01), but not in normal astrocytes (Figure 1b).

Figure 1
figure1

HA-induced glioma invasion by the upregulation of miR-21. (a) Gelatin zymography analysis of astrocytes, HT1080 and U87 cells after treatment with HA (b) Real-time PCR analysis of miR-21 expression in U87MG cells and astrocytes after treatment with HA for 24 h. miRNA levels were evaluated as the ratio between the values obtained from non-treated and HA-treated cells. Data represent the mean values of three independent experiments performed in duplicate. *P<0.01 vs control by the unpaired t-test. (c) Effect of anti-miR-21 on HA-induced glioma invasion. Matrigel invasion assay of control oligo- and anti-miR-21-transfected cells in the presence (HA) or absence (Control) of HA. Gelatin zymography and quantitative PCR were performed to measure the protein and mRNA levels of MMP-9. Data represent the mean values of three independent experiments performed in duplicate. *P<0.01 vs control by the unpaired t-test. **P<0.01 vs HA alone by the unpaired t-test. (d) Effect of miR-21 overexpression on invasion. Matrigel invasion and gelatin zymography were performed for different cell lines transfected with empty (Control) and miR-21 (miR-21) vectors. Data represent the mean values of three independent experiments performed in duplicate. *P<0.01 vs control by the unpaired t-test. (e) Downregulation of HA-induced miR-21 expression by PTEN. PTEN-transfected U87MG and (f) si-PTEN-transfected LN229 were examined for miR-21 expression by real-time PCR. The activity of PTEN and its expression were measured by western blot analysis. Data represent the mean values of three independent experiments performed in duplicate. *P<0.01 vs control by the unpaired t-test. **P<0.05 vs HA alone by the unpaired t-test.

To examine the involvement of miR-21 in HA-induced glioma invasion, we transfected U87MG cells with anti-miR-21, which inhibits the function of miR-21. Anti-miR-21 suppressed HA-induced MMP-9 secretion and invasion (P<0.01) (Figure 1c), suggesting that miR-21 is necessary for glioma invasion. Next, we overexpressed miR-21 in glioma cell lines (U373MG U87MG and LN428) and normal astrocytes. miR-21 overexpression increased MMP-9 secretion and invasion (P<0.01) of glioma cell lines, but not normal astrocytes (Figure 1d), indicating that miR-21 is sufficient for glioma invasion.

Previously, we demonstrated that PTEN regulates HA-induced MMP-9 secretion and invasion in glioma cells (Park et al., 2002). Glioma cells lacking functional PTEN (U87MG and U373MG) could secrete MMP-9 after HA treatment, but ones harboring wild-type (wt) PTEN (LN18 and LN428) could not secrete MMP-9 (Park et al., 2002). Thus, we tested whether PTEN is related to HA-induced miR-21 expression, which facilitates MMP-9 secretion and invasion. Indeed, reintroduction of wt-PTEN into U87MG cells suppressed HA-induced miR-21 expression (P<0.01) (Figure 1e), and knockdown of PTEN by si-PTEN in LN229 cells potentiated it (P<0.01) (Figure 1f), implying that PTEN negatively regulates miR-21 expression. The regulation may be mediated by Akt, as PTEN is a negative regulator of Akt. It is well-known that HA activates Akt through its CD44 receptor (Toole, 2004). Indeed, HA activated Akt in U87MG cells, but Akt inhibitor did not suppress HA-induced miR-21 expression (Supplementary Figure 2), suggesting that the PTEN-mediated regulation of miR-21 is independent of Akt activation.

Taken together, these results suggest that miR-21 has an essential role in HA-induced invasion by inducing MMP-9 expression in glioma cells lacking functional PTEN.

miR-21 facilitates glioma invasion through Ras/MAPK signaling

Previously, we demonstrated that ERK1/2 and NF-κB signalings have important roles in HA-induced invasion of glioma cells lacking functional PTEN (Park et al., 2002; Kim et al., 2005, 2008). To examine the role of miR-21 in these pathways, we performed immunoblot analyses using relevant antibodies. In HA-treated U87MG cells, anti-miR-21 dramatically reduced the amplitude and duration of ERK1/2 phosphorylation (see 10 and 30 min) (Figure 2a), and miR-21 overexpression potentiated ERK1/2 phosphorylation (see 30 and 50 min) (Figure 2b). However, anti-miR-21 did not affect the nuclear translocation of NF-κB subunit p65 (Figures 2c). Moreover, a PKC activator, PMA-induced ERK1/2 was not affected by anti-miR-21 either (Figure 2d). From these results, we speculated that the target of miR-21 might be Ras, an upstream regulator of Raf activation. Thus, we further analyzed the activated Ras status using pull-down assays with GST–Raf–RBD conjugated beads. In the Ras binding assay, anti-miR-21 inhibited HA-induced Ras–Raf binding (Figure 2e), indicating that the target of miR-21 might be a negative regulator of Ras activation. Taken together, these results suggest that miR-21 regulates glioma invasion through Ras/MAPK signaling.

Figure 2
figure2

Potentiation of Ras/MAPK signaling by miR-21. (a) Inhibition of ERK1/2 activation by anti-miR-21. Western blot analysis of phosphorylated ERK1/2 (pERK) was performed using U87MG cells transfected with anti-miR-21 after treatment with HA. (b) Increased duration of active ERK1/2 in U87MG cells overexpressing miR-21. Western blot analysis of phosphorylated ERK1/2 (pERK) in U87MG cells transfected with miR-21 after treatment with HA. (c) Effect of anti-miR-21 on the nuclear translocation of the NF-κB p65 subunit after treatment with HA. C, cytoplasm, N, nucleus. (d) Effect of miR-21 on PMA-induced ERK1/2 activation. (e) Involvement of miR-21 in HA-induced Ras activation. Activated Ras was pulled down by GST–Raf (RBD) after treatment with HA and measured by western blot analysis.

miR-21 targets Spry2 for glioma invasion

In order to identify the exact target of miR-21 responsible for glioma invasion, we uncovered more than 200 potential targets using a miRNA target prediction algorithm, TargetScan (Friedman et al., 2009). Among the targets, we chose Spry2 containing a miR-21 seed match in its 3′ untranslated region (UTR), as it is known to negatively regulate Ras signaling (Cabrita and Christofori, 2008). Moreover, it has been reported that miR-21 targets Spry2 in cardiocytes and colon cancer SW480 cells (Sayed et al., 2008). We constructed a reporter plasmid (wt-UTR), driven by SV40 basal promoter, harboring the 778-nucleotide wt-3′UTR of Spry2 at the 3′ position of luciferase reporter gene. To determine sequence specificity, we constructed the second reporter plasmid (mt-UTR) in which the conserved targeting sequence of miR-21, AUAAGCUA (235–241 of Spry2 3′UTR), was completely mutated (Figure 3a). Transient transfection of miR-21 with the wt-UTR into 293T cells led to the significant decrease of luciferase activity compared with control, but the luciferase activity of mt-UTR-transfected cells was largely unaffected by miR-21 (Figure 3a). Moreover, anti-miR-21 increased Spry2 protein level, but did not affect its mRNA level in U87MG cells (Figure 3b). Taken together, these results suggest that miR-21 downregulates Spry2 in a sequence-specific manner at post-transcriptional level in glioma cells.

Figure 3
figure3

Downregulation of Spry2 by miR-21 is critical for glioma invasion through the disruption of negative feedback inhibition. (a) miR-21 inhibits translation of Spry2 by targeting its 3′UTR. Luciferase activity of 293T cells transfected with pGL3 and pGL3-UTRs vectors, and miR-21. mRNA and protein expression of Spry2 was measured after 24 h of transfection by quantitative PCR and western blot analysis. Data are mean values of three independent experiments performed in duplicate. *P<0.05 vs control by unpaired t-test. (b) Post-transcriptional regulation of Spry2 by miR-21. Spry2 mRNA and protein expression were measured after 24 h of transfection by quantitative PCR and western blot analysis. (c) Role of Spry2 targeting by miR-21 in HA-induced glioma invasion. Matrigel invasion assay of control oligo-, anti-miR-21-, and si-Spry2-transfected U87MG cells in the presence (HA) or absence (Control) of HA. Data represent the mean values of three independent experiments performed in duplicate. *P<0.01 vs control by the unpaired t-test. **P<0.01 vs HA alone by the unpaired t-test. ***P<0.01 vs HA and anti-miR-21 by the unpaired t-test. (d) miR-21 triggers the disruption of negative feedback inhibition by targeting Spry2. Western blot analysis of phosphorylated ERK1/2 (pERK) and Spry2 was carried out in U87MG cells transfected with anti-miR-21 after treatment with HA. (e) Analysis of miR-21 expression by real-time PCR in U87MG cells after treatment with HA for the indicated time. miRNA levels were evaluated as the ratio between the values obtained from non-treated and HA-treated cells. (f) Overexpression of Spry2 inhibits ERK1/2 activation. Western blot analysis of phosphorylated ERK1/2 (pERK) and Spry2 was performed in U87MG cells overexpressing Spry2 or mCherry by a lenti-viral expression system. (g) Inhibition of MMP-9 expression by Spry2 overexpression. Gelatin zymography analysis of Spry2- or mCherry-overexpressing U87MG cells after treatment with HA.

Although miR-21 directly targets Spry2, it was unknown whether the downregulation of Spry2 by miR-21 has a critical role in glioma invasion. Anti-miR-21-transfected U87MG cells displayed decreased HA-induced invasion and ERK1/2 activation, which was partially rescued by the transfection of si-Spry2 (Figure 3c and Supplementary Figure 3). This suggests that the effect of miR-21 on HA-induced glioma invasion is mediated through Spry2.

As mentioned above, Spry2 is known to be a negative feedback regulator of Ras signaling, and, in normal tissues, Ras signaling is tightly regulated by this negative feedback. Thus, prolonged Ras signaling due to the lack of negative feedback regulation may cause deregulated cell invasion. As such, we examined the role of miR-21 in the negative feedback loop. In anti-miR-21-transfected U87MG cells, Spry2 expression was more strongly induced by HA than in control oligo-transfected U87MG cells (Figure 3d). Moreover, HA-induced ERK1/2 phosphorylation at 30 min was decreased by anti-miR-21 (Figure 3d), and real-time PCR analysis revealed the increase of miR-21 during the same period (Figure 3e). These results suggest that miR-21 blocks the negative feedback by inhibiting Spry2 expression, resulting in extended Ras/MAPK signaling and deregulated cell invasion. Consistently, Spry2-overexpressed U87MG cells showed the decrease of ERK1/2 phosphorylation (Figure 3f) and MMP-9 secretion (Figure 3g) compared with mCherry-overexpressed U87MG cells.

miR-21 integrates various growth factor stimuli

Like HA, growth factors such as PDGF, EGF and bFGF increased miR-21 (Figure 4a) and invasion (Figure 4b) of U87MG cells. The increased invasion was inhibited by anti-miR-21 (Figure 4b), suggesting that the growth factor-induced invasion is regulated by miR-21 as well. Moreover, Spry2 overexpression also inhibited growth factor-induced invasion (Figure 4c). Thus, these results suggest that the miR-21/Spry2/Ras/MAPK pathway regulates glioma invasion by integrating various cancer-microenvironmental stimuli, such as HA and growth factors, strongly proposing this pathway as an attractive therapeutic target to block glioma invasion.

Figure 4
figure4

miR-21 mediates growth factor-induced glioma invasion. (a) Upregulation of mature miR-21 after treatment with growth factors. Northern blot analysis of miR-21 in U87MG cells after treatment with growth factors as indicated. (b) Inhibition of invasion by anti-miR-21 after treatment with growth factors. Matrigel invasion assay of control oligo- and anti-miR-21-transfected U87MG cells in the presence of growth factors as indicated. (c) Inhibition of invasion by Spry2 overexpression after treatment with growth factors. Matrigel invasion assay of pCDNA 6.2 vector- (Control) and Spry2-transfected U87MG cells in the presence of growth factors as indicated. Data represent the mean values of three independent experiments performed in duplicate *P<0.01 vs control by the unpaired t-test. **P<0.01 vs control oligo by the unpaired t-test.

Inverse correlation between miR-21 and Spry2 in human glioma tissues

We next determined the expression pattern of Spry2 in human glioma tissues. Among tissues listed in Table 1, there were two normal-tumor pairs. Spry2 was significantly decreased in tumor tissues compared with their normal counterparts from the same patients (Figure 5a) and in different grades of gliomas (Figure 5b). In the tumor tissues, PTEN also diminished, and miR-21 increased (Figure 5a), suggesting that, like in U87MG cells, reduced PTEN augmented miR-21 to suppress Spry2, resulting in enhanced glioma invasion. Strikingly, Spry2 expression was markedly decreased in 79.7% of 69 malignant glioma tissues (WHO grade II–IV) (H-score 1, 55/69), in the cytoplasm of the tumor cells, but not in normal astrocytes (Table 1). Downregulation of Spry2 was similarly observed in different WHO grades of tumor (grade II: 71.4% (10/14); grade III: 88.4% (15/17); and grade IV: 78.9% (30/38)), but not in grade I: 28.6% (4/14) (Table 1). Grade I tumors are recognized as benign tumors. Thus, these results imply that Spry2 is the critical factor that determines the malignancy of gliomas.

Table 1 Summary of Spry2 staining in human gliomas
Figure 5
figure5

Downregulation of Spry2 by miR-21 in human glioma tissues. (a) Real-time PCR evaluation of miR-21 in normal (N) and tumor (T) tissues of patients #20361312 and #31452681. The miR-21 levels were normalized over the normal sample of patient #20361312. Western blot analysis of Spry2, PTEN and GAPDH from the same tissues. Data represent the mean values of three independent experiments performed in duplicate. *, P<0.01 vs normal by the unpaired t-test. (b) The immunohistochemical expression of Spry2 was absent in most of the grade II–IV tumor cells compared with grade I and normal brain tissue (Normal) ( × 400). (c) Inverse correlation between miR-21 and Spry2 in tumor tissues. Spry2 mRNA and miR-21 expression were evaluated by real-time PCR, and the results were normalized to a normal tissue (N1). Protein abundance was determined by western blotting densitometry normalized to GAPDH protein expression and then compared with N1. A correlation coefficient of −0.49869 with P=0.004 indicates an inverse relationship between miR-21 and Spry2 protein. A correlation coefficient of 0.15233 with P=0.43 indicates no relationship between Spry2 mRNA and protein levels.

Furthermore, we investigated the correlation of miR-21 and Spry2 expression in human glioma tissues listed in Table 1, whose frozen sections were available. Real-time PCR and immunoblot analyses of normal and tumor tissues revealed a significant inverse correlation between miR-21 and Spry2 protein levels (P=0.004), whereas did not show any significant correlation between the protein and mRNA levels of Spry2 (P=0.43) (Figure 5c). These results imply that Spry2 is decreased in most of human glioma tissues through post-transcriptional regulation by miR-21, thus promoting tumor invasion. Further, Spry2 may contribute to blocking cancer progression by suppressing the development of glioma invasiveness (that is, by functioning as a tumor suppressor).

Discussion

Gliomas are graded on the WHO consensus-derived scale of I to IV, according to their degree of malignancy (Furnari et al., 2007). Grade I tumors are benign and can be cured by surgical resection. Grade II tumors display low-grade malignancy, but eventually most of them transform into high-grade tumors during long clinical courses. Grade III tumors are anaplastic and show increased proliferation, and grade IV tumors are highly malignant and resistant to chemo/radiotherapy. Although grade II tumors show low malignancy, diffusive infiltration makes them impossible to cure by surgery. Thus, it is very important to understand the detailed molecular mechanism of invasion onset. In the present study, we demonstrated that HA-induced miR-21 potentiates glioma invasion by targeting Spry2, a well-known negative regulator of growth factor signaling (Figure 6). The downregulation of Spry2 increases the amplitude and duration of Ras/MAPK signaling, which dramatically increases MMP-9 expression and invasion. Consistently, Spry2 protein levels were significantly decreased in 79.7% of invasive WHO grade II–IV human glioma tissues, but not in non-invasive WHO grade I tumors and normal tissues. Notably, the mRNA levels of Spry2 remained unchanged, but their protein levels were significantly decreased in human patients. These results imply that the post-transcriptional regulation of Spry2 by miR-21 has an important role in the malignant progression of human gliomas.

Figure 6
figure6

Schematic overview of the proposed molecular mechanism of miR-21-regulated glioma invasion. miR-21 is upregulated by cancer microenvironments in PTEN-malfunctional tumor cells and disrupts Spry2-mediated negative feedback inhibition of Ras/MAPK signaling.

Aberrant miR-21 expression has been shown in various tumors, including gliomas, colon, breast, prostate, pancreas and stomach tumors (Krichevsky and Gabriely, 2009). Increased miR-21 expression in such tumors is associated with cell proliferation, migration, invasion and metastasis (Krichevsky and Gabriely, 2009). These results imply that miR-21 is a key regulatory molecule in cancer development that functions by targeting different signaling molecules. In addition, human miR-21 is known to be localized on chromosome 17q23.2. Although amplification of this region has been observed in several cancers, many other cancers that exhibit high miR-21 expression, including gliomas, did not show any genetic amplification (Roversi et al., 2006). This observation implicates the aberrant expression of miR-21 in gliomas is due to deregulation of its biogenesis, but not by its genetic alterations. Furthermore, we observed that PTEN, a tumor suppressor frequently mutated in gliomas, suppressed HA-induced miR-21 expression. These results suggest that the aberrant expression of miR-21 might be closely related to PTEN-regulated cellular signaling.

Recent studies have reported that Spry2 is a tumor suppressor because it is downregulated in hepatocellular carcinoma, non-small cell lung cancer and breast cancer (Lo et al., 2006). Knockdown of Spry2 expression also accelerates Ras-induced lung cancer development in mice (Shaw et al., 2007). In this study, we showed that HA-induced miR-21 enhanced glioma invasion by targeting Spry2. Thus, our results suggest that the downregulation of Spry2 by miR-21 is the key event that triggers the malignancy of gliomas, and implicates Spry2 as a novel tumor suppressor in malignant gliomas. Moreover, we demonstrated that the downregulation of Spry2 by miR-21 in glioma cells enhances the strength and duration of Ras/MAPK signaling, proposing that HA-induced miR-21 expression triggers glioma invasion by disrupting the negative feedback regulator of Ras/MAPK signaling, Spry2. Similarly, miR-21 augmented MAPK signaling through inhibition of Spry1 in cardiac fibroblasts (Thum et al., 2008). Furthermore, the fact that miR-21 affects not only HA but also growth factor signaling implies that miR-21 merges extracellular stimuli to trigger glioma invasion, and this regulation represents a viable therapeutic target to block glioma invasion.

Cancer is a complex genetic disease caused by the mutation and epigenetic alteration of genes. Thus, novel target identification for therapeutic purposes has been mainly focused on the search for DNA modifications, although few genetic variants are known to reproducibly influence human cancer (Hirschhorn, 2009). Moreover, numerous studies have shown that heterogeneity of DNA alterations exists even among the same class of cancer, thus hampering drug development. Indeed, the frequencies of individual genetic alterations identified in glioblastomas by The Cancer Genome Atlas are less than 55% (for example, EGFR, 45%; PTEN, 36%; ARF, 49%; P16/INK4A, 52% and CDKN2B, 47%) (TCGARN, 2008). Recently, microarray technologies have accelerated novel target identification, but this identification is restricted to DNA or RNA samples. Furthermore, discrepancies between mRNA and protein levels often exist because of post-transcriptional regulation. To solve this problem, proteomics offers the chance to examine whole proteomes directly, though this technique is not yet suitable for massive screening. In the present study, we present a typical example, Spry2. As the The Cancer Genome Atlas did not check the Spry2 protein level, it did not recognize this gene, whose protein levels were altered in as many as 79.7% of our human glioma tissues. A search of the Rembrandt database also showed that Spry2 mRNA levels are little changed or even increased in gliomas (data not shown) (National Cancer Institute, 2005). Moreover, in the glioblastoma study of The Cancer Genome Atlas, Ras/MAPK signaling was not carefully examined because the frequency of Ras mutation was as low as 2%, and other components of the signaling pathway were not altered either. However, in the present study, we discovered that Ras/MAPK signaling has a critical role in the malignancy of gliomas through its negative regulator, Spry2. Thus, our findings suggest that the transcriptome analysis is not sufficient, and post-transcriptional regulation by miRNA will provide new insights for the future study of human cancer.

Taken together, we have identified a novel tumor suppressor in gliomas, Spry2, which is post-transcriptionally regulated by miR-21. Increased miR-21 and decreased Spry2 shown in most of our human glioma tissues emphasize the therapeutic relevance of miR-21 and Spry2 for the future treatment of gliomas.

Materials and methods

Reagents and cell culture

HA was purchased from Sigma Chemical Co. (St Louis, MO, USA) and reconstituted in Dulbecco's modified Eagle's medium (Life Technologies, Inc., Grand Island, NY, USA). Rabbit polyclonal anti-Akt, anti-ERK1/2, anti-MEK, and goat polyclonal anti-actin antibodies were purchased from Cell Signaling Biotechnology (Beverly, MA, USA). Rabbit polyclonal anti-NF-κB p65 was purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Rabbit polyclonal anti-Spry2 was purchased from Upstate Biotechnology (Lake Placid, NY, USA). Matrigel was purchased from Becton Dickinson (Bedford, MA, USA), and gelatin was purchased from Sigma Chemical Co. Human glioma cell lines (U373MG, U87MG, LN18 and LN428) were obtained from the American Type Culture Collection (Manassas, VA, USA). Glioma cells were cultured in Dulbecco's modified Eagle's medium supplemented with 10% heat-inactivated fetal bovine serum, 100 units/ml penicillin and 100 μg/ml streptomycin.

Luciferase assay

293T cells were seeded into six-well plates and transfected with luciferase reporters using the Effectine transfection reagent (Qiagen, Valencia, CA, USA). After transfection, the cells were incubated for two days and harvested for luciferase activity using the Dual-Luciferase Reporter Assay System (Promega, Madison, WI, USA), according to the manufacturer's protocol.

Transfection

The siRNA against Spry2 was purchased from Santa Cruz Biotechnology, and cell transfections with siRNA molecules (50 nM) were carried out using an Amaxa Nucleofector device (Lonza, Walkersville, MD, USA), according to the manufacturer's instructions. To express miR-21, miR-21 was transfected with the Amaxa Nucleofector device, according to the manufacturer's instructions. To inhibit the function of miR-21, anti-miR-21 and the negative control, purchased from Dharmacon (Chicago, IL, USA), were introduced into U87MG cells by Amaxa.

Invasion assays

Glioma invasion assays were performed using modified Boyden chambers with polycarbonate nucleopore membranes (Corning, Corning, NY, USA), as described previously (Park et al., 2002; Kim et al., 2005, 2008). Invasiveness was determined by counting cells in five microscopic fields per well, and the extent of invasion was expressed as the average number of cells per microscopic field.

Gelatin zymography

Glioma cells (about 70–80% confluent) were washed and replenished with serum-free Dulbecco's modified Eagle's medium. Cells were transfected with anti-control or anti-miR21, incubated for 36 h and stimulated with 200 μg/ml HA for 18 h. The volume of conditioned media was normalized according to the cell number. The enzymatic activity of electrophoretically separated gelatinolytic enzymes in the conditioned media of glioma cells was determined by gelatin zymography, as described previously (Park et al., 2002; Kim et al., 2005, 2008). Zones of gelatinolytic activity were detected as clear bands against a blue background.

Nuclear and cytoplasmic extracts and western blot analyses

Cells were either treated with HA (200 μg/ml) for a period of 0–6 h or with 17-AAG for 45 min, followed by HA (200 μg/ml) for 3 h at 37 °C. The cytoplasmic and nuclear extracts were prepared using a CelLytic NuCLEAR Extraction Kit from Sigma, according to the manufacturer's protocol. The nuclear and cytoplasmic extracts (30 μg) were resolved by SDS-polyacrylamide gel electrophoresis and then electrotransferred to nitrocellulose membranes (Schleicher and Schuell, Dassel, Germany). Immunoblots were visualized using the ECL Plus Western blotting detection system (Amersham, Arlington Heights, IL, USA), according to the manufacturer's protocol.

Quantitative PCR

Total RNA was extracted using TRIZOL reagent (Invitrogen, Carlsbad, CA, USA). First strand cDNA was generated using Superscript III (Invitrogen), according to the manufacturer's protocol. PCR amplification was carried out with the following primers: MMP-9, 5′-IndexTermAGTTTGGTGTCGCGGAGCAC-3′ and 5′-IndexTermTACATGAGCGCTTCCGGAAC-3′. PCR reactions were performed using the PCR Master kit (Roche, Mannheim, Germany), according to the manufacturer's protocol.

Real-time PCR

Real-time PCR assays were carried out to detect the expression level of mature miR-21 using the mirVana qRT-PCR miRNA Detection Kit (Ambion, Austin, TX, USA), with SYBR Green I as the fluorescent dye (enabling real-time detection of PCR products), according to the manufacturer's protocol. The cycling conditions were 95 °C for 10 min, followed by 40 cycles of 95 °C for 15 s and 60 °C for 60 s in 7500 real-time PCR system (Applied Biosystems, Carlsbad, CA, USA). The results of the real-time PCR data are presented as Ct values, where Ct is defined as the threshold cycle of PCR at which the amplified product was first detected. To compare the different RNA samples in an experiment, we used the comparative Ct method and compared the RNA expression in samples with that of the control in each experiment. PCR was performed in duplicate for each sample. The results are expressed as mean±s.d. for the relative expression levels compared with the control, and minimum values of three independent experiments were obtained.

Immunohistochemistry

Archived surgically resected glioblastoma specimens were retrieved from the National Cancer Center. Glioblastomas were selected from patients with no previous radiation or chemotherapy treatment. Glioblastoma tissues were fixed in 10% buffered formalin, routinely processed and paraffin embedded. Immunohistochemical studies were conducted on 6 μm sections. Sections were deparaffinized and subjected to heat-induced epitope retrieval by steaming for 15 min. Slides were then incubated with antibodies against Spry2 (1:100, Upstate technology) at 4 °C overnight. Antibodies were detected using the avidin–biotin–peroxidase complex method using 3,3′-diaminobenzidine as the chromogen. Standard positive controls were used throughout. Normal sera served as the negative control. Sections were counterstained with hematoxylin. We scored the overall staining intensities of malignant cells with respect to those observed in the corresponding normal astrocytes. The staining intensity was graded on a scale from 0 to 2 (0: negative; 1: weak positive; 2: strong positive), and the proportion of positive tumor cells was measured for each specimen (0: <10%; 1: 10–50%; 2: >50%). The staining intensity was multiplied by the proportion of positive tumor cells to obtain a semi-quantitative H-score.

Statistical analysis of the data

All values are reported as mean±s.d. Differences were assessed by the two-tailed Student t-test using Excel software; P<0.05 was considered as statistically significant.

References

  1. Cabrita MA, Christofori G . (2008). Sprouty proteins, masterminds of receptor tyrosine kinase signaling. Angiogenesis 11: 53–62.

  2. Chan JA, Krichevsky AM, Kosik KS . (2005). MicroRNA-21 is an antiapoptotic factor in human glioblastoma cells. Cancer Res 65: 6029–6033.

  3. Conti A, Aguennouz M, La Torre D, Tomasello C, Cardali S, Angileri FF et al. (2009). miR-21 and 221 upregulation and miR-181b downregulation in human grade II-IV astrocytic tumors. J Neurooncol 93: 325–332.

  4. Croce CM . (2009). Causes and consequences of microRNA dysregulation in cancer. Nat Rev Genet 10: 704–714.

  5. Friedman RC, Farh KK, Burge CB, Bartel DP . (2009). Most mammalian mRNAs are conserved targets of microRNAs. Genome Res 19: 92–105.

  6. Furnari FB, Fenton T, Bachoo RM, Mukasa A, Stommel JM, Stegh A et al. (2007). Malignant astrocytic glioma: genetics, biology, and paths to treatment. Genes Dev 21: 2683–2710.

  7. Gabriely G, Wurdinger T, Kesari S, Esau CC, Burchard J, Linsley PS et al. (2008). MicroRNA 21 promotes glioma invasion by targeting matrix metalloproteinase regulators. Mol Cell Biol 28: 5369–5380.

  8. Hirschhorn JN . (2009). Genomewide association studies--illuminating biologic pathways. N Engl J Med 360: 1699–1701.

  9. Kim MS, Kwak HJ, Lee JW, Kim HJ, Park MJ, Park JB et al. (2008). 17-Allylamino-17-demethoxygeldanamycin down-regulates hyaluronic acid-induced glioma invasion by blocking matrix metalloproteinase-9 secretion. Mol Cancer Res 6: 1657–1665.

  10. Kim MS, Park MJ, Kim SJ, Lee CH, Yoo H, Shin SH et al. (2005). Emodin suppresses hyaluronic acid-induced MMP-9 secretion and invasion of glioma cells. Int J Oncol 27: 839–846.

  11. Kim VN, Nam JW . (2006). Genomics of microRNA. Trends Genet 22: 165–173.

  12. Krichevsky AM, Gabriely G . (2009). miR-21: a small multi-faceted RNA. J Cell Mol Med 13: 39–53.

  13. Lo TL, Fong CW, Yusoff P, McKie AB, Chua MS, Leung HY et al. (2006). Sprouty and cancer: the first terms report. Cancer Lett 242: 141–150.

  14. Merzak A, Koocheckpour S, Pilkington GJ . (1994). CD44 mediates human glioma cell adhesion and invasion in vitro. Cancer Res 54: 3988–3992.

  15. National Cancer Institute (2005). REMBRANDT home page. http://rembrandt.nci.nih.gov. Accessed 29 September 2009.

  16. Negrini M, Nicoloso MS, Calin GA . (2009). MicroRNAs and cancer--new paradigms in molecular oncology. Curr Opin Cell Biol 21: 470–479.

  17. Nicoloso MS, Spizzo R, Shimizu M, Rossi S, Calin GA . (2009). MicroRNAs--the micro steering wheel of tumour metastases. Nat Rev Cancer 9: 293–302.

  18. Park JB, Kwak HJ, Lee SH . (2008). Role of hyaluronan in glioma invasion. Cell Adh Migr 2: 202–207.

  19. Park MJ, Kim MS, Park IC, Kang HS, Yoo H, Park SH et al. (2002). PTEN suppresses hyaluronic acid-induced matrix metalloproteinase-9 expression in U87MG glioblastoma cells through focal adhesion kinase dephosphorylation. Cancer Res 62: 6318–6322.

  20. Roversi G, Pfundt R, Moroni RF, Magnani I, Van Reijmersdal S, Pollo B et al. (2006). Identification of novel genomic markers related to progression to glioblastoma through genomic profiling of 25 primary glioma cell lines. Oncogene 25: 1571–1583.

  21. Sayed D, Rane S, Lypowy J, He M, Chen IY, Vashistha H et al. (2008). MicroRNA-21 targets Sprouty2 and promotes cellular outgrowths. Mol Biol Cell 19: 3272–3282.

  22. Shaw AT, Meissner A, Dowdle JA, Crowley D, Magendantz M, Ouyang C et al. (2007). Sprouty-2 regulates oncogenic K-ras in lung development and tumorigenesis. Genes Dev 21: 694–707.

  23. TCGARN (2008). Comprehensive genomic characterization defines human glioblastoma genes and core pathways. Nature 455: 1061–1068.

  24. Thum T, Gross C, Fiedler J, Fischer T, Kissler S, Bussen M et al. (2008). MicroRNA-21 contributes to myocardial disease by stimulating MAP kinase signalling in fibroblasts. Nature 456: 980–984.

  25. Toole BP . (2004). Hyaluronan: from extracellular glue to pericellular cue. Nat Rev Cancer 4: 528–539.

Download references

Acknowledgements

This work was supported by the National Cancer Center Grants (1010171, 1010172, and 1110120) and the National Research Foundation Grant (2010-0016811). JHK was supported by the National Cancer Center Grant (1110110) and the National Research Foundation Grant (2010-0016704).

Author information

Correspondence to J B Park or S-H Lee.

Ethics declarations

Competing interests

The authors declare no conflict of interest.

Additional information

Supplementary Information accompanies the paper on the Oncogene website

Supplementary information

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Kwak, H., Kim, Y., Chun, K. et al. Downregulation of Spry2 by miR-21 triggers malignancy in human gliomas. Oncogene 30, 2433–2442 (2011). https://doi.org/10.1038/onc.2010.620

Download citation

Keywords

  • glioma
  • invasion
  • hyaluronan
  • miR-21
  • spry2
  • PTEN

Further reading