The microtubule binding drug EM011 inhibits the growth of paediatric low grade gliomas

Cancer Letters 335 (2013) 109–118

Contents lists available at SciVerse ScienceDirect

Cancer Letters journal homepage: www.elsevier.com/locate/canlet

The microtubule binding drug EM011 inhibits the growth of paediatric low grade gliomas Norbert F. Ajeawung a, Harish C. Joshi b, Deepak Kamnasaran a,c,⇑ a

Pediatric Research Unit, Centre de Recherche du CHUL, Québec, QC, Canada G1V 4G2 Department of Cell Biology, Emory University School of Medicine, Atlanta, GA 30322, United States c Department of Pediatrics, Laval University, Québec, QC, Canada G1V 4G2 b

a r t i c l e

i n f o

Article history: Received 27 November 2012 Received in revised form 1 February 2013 Accepted 2 February 2013

Keywords: Viability Paediatric astrocytomas Low grade glioma Drug discovery Gene expression

a b s t r a c t Low grade gliomas are a heterogeneous group of tumours representing the most common form of neoplasms in the central nervous system among children. Although gross total resection remains the principal treatment, it is often impractical especially for the resection of tumours within eloquent regions of the brain. Instead Radiotherapy is utilised in such cases, but because of its associated toxicities, it is refrained from use among younger children. These limitations coupled with hypersensitivity and toxicities associated with some commonly used chemotherapeutic agents, have ignited the need to search for safer and more effective treatments for paediatric low grade gliomas. In this study, we investigated the EM011 drug on the growth of two pilocytic and one diffuse paediatric astrocytoma cell lines, using an assortment of cancer assays. We discovered that treatments of low grade gliomas with EM011 abrogated cell viability by inducing a decrease in cell proliferation and an arrest in the S and G2M cell cycle phases, followed by a converse increase in apoptosis in a dose and time dependent manner. The cell migratory and invasion indices, as well as anchorage independent growth in soft agarose, were significantly attenuated. These findings were mechanistically associated with a transient release of AIF, a disruption of microtubule architecture, and a decline in the expression of key genes which drive cancer progression including EGFR, mTORC1, JUN and multiple MMPs. In fact, the activity of MMP2 was also perturbed by EM011. These findings, in conjunction with the insignificant adverse side effects established from other studies, make EM011 an appealing chemotherapeutic agent for the treatment of paediatric low grade gliomas. Ó 2013 Elsevier Ireland Ltd. All rights reserved.

1. Introduction A significant portion of the cytoskeleton is composed of an elaborate array of assembled alpha and beta tubulins, giving rise to a filamentous structure referred to as microtubules. Microtubules display periodic cycles of polymerisation and depolymerisation, resulting in a dynamic unstable state necessary for functions such as intracellular transport, cell motility, maintenance of cell shape, and formation of the mitotic spindle. Therefore, alterations in this dynamic state are expected to have profound consequences on the cell. A phenomenon which could be achieved through the interference of tubulin polymerisation cycles, or via the dysregulation of microtubule interacting proteins, or even by aberrancies in the post-translational modifications of tubulins [1]. In fact, this notion of manipulating the dynamic nature of microtubules as a strategy to attenuate cancer progression, has led to the development and ⇑ Corresponding author at: Department of Pediatrics, Laval University, Local RC9800, 2705 Boulevard Laurier, Québec, QC, Canada. Tel.: +1 418 656 4141 (O); fax: +1 418 654 2753. E-mail address: [email protected] (D. Kamnasaran). 0304-3835/$ - see front matter Ó 2013 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.canlet.2013.02.004

testing of new microtubule targeting compounds; some of which are currently being used for the treatment of a variety of cancers [2]. Even though many of these chemotherapeutic agents were validated for their potent anti-neoplastic properties, a wide spectrum of dismal side effects often becomes a limiting factor for their clinical use [3]; hence necessitating the need to investigate other potential chemotherapeutic agents. Indeed, an interesting class of microtubule binding agents that can potently change the dynamics of microtubules leading to growth arrest and apoptosis, are those belonging to the noscapiniod compounds; a sub-family of opioid drugs [4,5]. EM011 is a potent non-toxic brominated derivative of noscapine that possesses a wide spectrum of anti-cancer properties in various in vitro and in vivo cancer models [5]. Despite its safe pharmacological profile, EM011 has never been tested on tumour cells from paediatric gliomas. As a result, we have investigated the anti-neoplastic properties of EM011 on paediatric low grade gliomas, the most common central nervous system tumours among children; and which are histologically and molecularly distinctive in comparison to high grade gliomas [6]. Although a sizable fraction of paediatric low grade gliomas can be surgically ressected, gross total resection is sometimes

110

N.F. Ajeawung et al. / Cancer Letters 335 (2013) 109–118

impractical to accomplish due to the location of the tumour within the central nervous system. Moreover, tumour re-growth after the resection of grade I tumours is common and warrants the need for additional follow-up surgeries [7] or other adjuvant therapies. If left untreated, low grade gliomas can potentially progress into high grade malignancies [8,9], making any form of current therapeutic intervention less effective. Furthermore, paediatric low grade tumours considered to be benign, could sometimes be located in eloquent regions of the brain, thereby decreasing the possibility for gross resection, but yet, imposing cognitive and adaptive deficiencies that threaten patients’ overall well being [10]. Because the incidence of some paediatric glioma cases are within the ages of 0–3 years, Radiotherapy is often delayed or avoided due to increased risk factors like toxicities and tumour recurrence [11,12]. As a result, Chemotherapy is often utilised to improve prognosis but most often with currently used drugs, their toxicity could impact both the short term and long term overall well being of the child; hence warranting the need to investigate the effectiveness of other suitable Chemotherapeutic agents. In fact, chemical agents in cough suppressants like noscapine and its highly related family member EM011, exhibit potent anti-neoplastic properties, while still having no significant adverse side effects even when administered at high doses of 300 mg/kg per body mass in mice [13]; hence making it clinically useful for the treatment of cancers such as paediatric low grade gliomas. In this report, we examined the mechanistic effect of EM011 on the growth and transformation properties of a panel of paediatric low grade glioma cell lines, and further investigated its influence on different cancer progression pathways. Paediatric low grade glioma cell lines treated with EM011 succumb to growth retardation due to changes in the cell cycle and apoptosis, a decrease in migratory/invasive properties and anchorage independent growth. These events are associated with the perturbed expression of a wide variety of key cancer progression genes, and inhibition of microtubule functions in the cytoplasm.

2.4. Methylene blue assay Growth inhibition of cells treated with varying drug concentrations was investigated following staining with 0.1% filtered methylene blue in 20% ethanol. The cells were incubated at room temperature for 5 min, washed gently with PBS followed by colorimetric absorbance measurements taken at 650 nm. 2.5. BrdU ELISA proliferation assay Cell proliferation was investigated using the BrdU ELISA kit (Roche) as described by manufacturer. Briefly, drugs at varying concentrations were administered to each well of a 96 well plate containing 2  103 cells. Prior to analysis, cells were labelled with BrdU for 18 h, fixed onto the surface of the plate, and then incubated with an Anti-BrdU antibody which binds to the BrdU incorporated into the DNA. After PBS washes, the substrate solution was added and colorimetric values were measured at 490 nm with an ELISA plate reader. 2.6. Cell cycle analysis 2  105 Cells were grown in each well of 6-well plates and exposed to varying drug doses daily. Prior to analysis, cells were stained with propidium iodide and analysed by flow cytometry as previously described [17] using the BD FACS canto II software. 2.7. Apoptotic assay 2  105 Cells were grown in each well of 6-well plates at 37 °C, and followed by the administration of varying drug doses. Cell were then trypsinized, washed with PBS, labelled with Annexin V-FITC and Propidium iodide, and finally analysed by flow cytometry. Cells stained with Annexin and those with Propidium were quantified using the BD FACS Canto II software. 2.8. Mitochondrial membrane permeability tests Changes in the mitochondrial membrane potential were investigated using Dioc6(3) as described previously [18]. Prior to drug treatment, 1.0  105 cells were grown in each well of 24-well plates at 37 °C, then treated with varying drug doses for specified periods. Cells were then harvested, washed with PBS, incubated in Dioc6(3) for 15 min, and then analysed by flow cytometry using BD FACS Canto II program. 2.9. Migration

2. Materials and method 2.1. Cell lines and drugs Grade I (R286 and Res186) and grade II (Res259) paediatric low grade glioma cell lines, which were previously characterised [14,15]; were cultured in DMEM/ Ham’s F12 supplemented with 10% FBS and 1X antibiotic–antimycotic. The human neuroprogenitor cell line (ReNcell VM, Millipore) was cultured in ReNcell neural stem cell maintenance medium (Millipore) containing 20 ng/ml epidermal growth factor and basic fibroblast growth factor. A non-neoplastichuman astrocyte cell line was obtained as gift from Dr. Hoke (John Hopkins University) and cultured in DMEM supplemented with 10% FBS, 1X Non-essential amino acid and antibiotic– antimycotic. All cell lines were maintained in a humidified 37 °C incubator with 5% CO2. The synthesis of EM011 was described earlier by us [16], and is currently commercially unavailable. BCNU was obtained from Sigma Aldrich.

2.2. MTS cytotoxic assay

2.0  105 Cells in each well of 24-well plates were incubated at 37 °C. Cell confluency was verified microscopically followed by a transverse scratch at the centre of the plate with a pipette tip. Fresh cell culture medium was replaced, and the cells exposed to varying drug treatments. Images before, during and at the study endpoint, were captured with a digital camera, and the migration index for wound healing was measured. 2.10. Invasion Standard Matrigel Boyden chamber assays with 8 lM pore sizes of insert were undertaken. Cells in serum free media containing drugs at varying doses were in the upper chamber. DMEM/Ham’s F12 containing 10% FBS was added into the lower chamber. After 24 h following administration of EM011, the cells in the upper chamber were carefully swabbed using a cotton pad pre-soaked with PBS. The cells at the bottom of the matrix were fixed in methanol, stained with crystal violet and quantified with a Compound microscope. The mean number of cells taken from five randomly selected fields was quantified.

The cell titre 96 well aqueous one cell proliferation assay kit-MTS (Promega) was used to evaluate cell cytotoxicity as described by manufacturer. Briefly, 2000 cells were plated in each well of 96 well plates, then incubated overnight, and followed by the addition of varying concentrations of drugs. Prior to analysis, the MTS reagent was added and the cells incubated at 37 °C for 2 h followed by colorimetric measurements taken at 490 nm.

2.11. Anchorage independent growth

2.3. Cell death/viability assessment

2.12. Immunofluorescent cytochemistry

The trypan blue exclusion technique was used to quantify living cells from dead cells following treatments. Briefly, cells were treated for the desired periods, trypsinized prior to analysis and stained with 0.4% trypan blue. The cells were loaded in a haemocytometer and counted microscopically. The percentage viability was calculated from the average count for live cells (those that exclude trypan blue) and dead cells (those that allow uptake of typan blue) as previously described [17].

Cells were grown on poly D-lysine/laminin coated slides and then treated with varying doses of EM011 for 24 h. Cells were then fixed in 20 °C ice cold methanol for 5 min, permeabilized with 0.05% Triton X 100, and then blocked with 1% Roche western blotting blocking reagent for 30 min. Fixed cells were hybridized with a 1:200 dilution of the alpha tubulin primary antibody (DM1A) (Santa Cruz) for 1 h at room temperature, washed in PBS, hybridized with protein G-FITC (Abcam) for

Soft agarose assays over an 18 day period in 24 well plates were performed as previously described [17]. Cells were then stained with 0.005% crystal violet for 2 h at room temperature, followed by quantification of the number of colonies with a compound microscope. The mean number of colonies was calculated from five randomly selected fields.

N.F. Ajeawung et al. / Cancer Letters 335 (2013) 109–118 another 1 h, then finally incubated in PBS containing 1 lg/ml propidium iodide. The microtubule morphologies were documented with a Nikon eclipse 80i fluorescence microscope equipped with a digital camera. 2.13. Western blotting Cells were treated with 20 lM EM011 at different intervals and subjected to cell fractionation using the Mitochondrial Isolation Kit for cultured cells (Thermo Scientific). The protein concentrations were determined using the BCA method and 40 lg of cytosolic and mitochondrial protein fractions were electrophoresed on an 8% SDS–PAGE gel. The proteins were transferred to a PVDF membrane using a Semi dry transfer system (Biorad), blocked with 1% western blotting blocking reagent (Roche) and incubated overnight at 4 °C with a 1:200 dilution of the E1 AIF antibody (Santa Cruz Biotech). Finally, the blot was incubated at room temperature with 1:3000 dilution of Protein G horseradish peroxidase conjugate (Biorad). The blots were exposed to the ECL reagent (Amersham) and imaged with the ImageQuant LAS 4000 imaging system (GE Healthcare). 2.14. Gelatin zymography Protein extraction was undertaken with the MPER mammalian protein extraction kit (Thermos Fisher Scientific) as specified by manufacturer. Protein lysates were electrophoresed on a 10% SDS–polyacrylamide gel containing 0.1% gelatin. After electrophoresis, the gel was washed in 2.5% Triton X 100 for 20 min and incubated for 72 h in a zymography developing buffer containing 0.05 M Tris–HCl PH 7.6, 0.02 M NaCl, 0.005 M CaCl2 and 0.02% Brij 35. The gel was washed in distilled water, stained with the Page-blue reagent (Fermentas) and images of white bands corresponding to metalloproteinase gelatinolytic activity on a blue background, were captured with Digital camera. 2.15. Real time reverse transcription (RT) PCR Total RNA was extracted with the GenElute mammalian RNA extraction kit (Sigma) as described by manufacturer (Sigma). cDNA was synthesised using 1 lg of total RNA using the Superscript II Reverse Transcriptase kit (Invitrogen) and oligo DT, as described by the manufacturer, then diluted by 5-fold in sterile distilled water. Real time PCR with the Ssofast evagreen supermix (Biorad) was undertaken with our home-made Cancer Pathway finder PCR array, comprising of primer pairs from 89 genes. With this array, PCR primers for each gene were designed using Primer blast (www.ncbi.nlm.nih.gov/tools/primer-blast). 2.16. Statistical analysis All experiments were done in triplicate, with Bar graphs representing values that are the mean ± SEM. Data analyses were performed using the Graph pad prism software. Multiple comparisons of datasets from treated and untreated samples were performed using the ANOVA and Bonferroni post hoc tests.

3. Results 3.1. EM011 inhibits cell viability and proliferation Since EM011 has been shown to inhibit the viability of various cancer cell lines [5,13], we questioned whether EM011 could similarly influence the growth of paediatric low grade gliomas. To accomplish this, we utilised Grade I (R286 and Res186) and Grade II (Res259) paediatric low grade glioma cell lines, which were previously characterised [14,15]. Viability assays, first undertaken with the Trypan blue assay, demonstrated a significant dose dependent decrease in the number of viable low grade glioma cells compared to the untreated control and/or vehicle treated control, with varying doses of EM011 over a 5 day period (Fig. 1A). The calculated IC50 values were 21 lM for R286 and 41 lM for Res259 and R286; demonstrating that cell line R286 shows the highest sensitivity to EM011. These findings were confirmed using the Methylene blue and MTS viability assays (Fig. 1B and C), where a profound decrease in cell viability compared to the controls, were observed as early as 24 h post-treatment. Since EM011 alters the viability of paediatric low grade glioma cell lines, we next determined whether this was a mechanistic consequence of a change in cell proliferation. Enhanced proliferation, a fundamental feature of malignant cells, is often assessed using Bromodeoxyuridine (BrdU), a synthetic analog of thimidine. The

111

Res259 and Res186 cell lines were previously reported with doubling times of 24 and 48 h respectively [14]. BrDU labelling assays subsequently indicated a significant dose dependent decrease in the proliferation index of low grade glioma cells with EM011 treatments, compared to the untreated control and/or vehicle treated control (Fig. 2A). The most profound effect was observed with EM011 treatments administered to the Res259 cell line. Finally, we examined the effect of EM011 on the viability of two non-transformed or normal cells of the nervous system, namely, human astrocytes and neuroprogenitors. EM011 modestly retarded the viability of both human non-transformed astrocyte cell line (NHA) (Supplemental Fig. 1A) and neuroprogenitor cell line (RenCell VM) (Supplemental Fig. 1B) by day 5, compared to the untreated control and/or vehicle treated control. Hence, the minimal cytotoxic properties of EM011 on normal cells makes it an attractive drug that warrants further investigations as a potent antineoplastic agent to treat low grade gliomas. 3.2. EM011 induces S and G2M growth arrest To further unravel how EM011 mechanistically influences the proliferation of low grade glioma cell lines, we undertook flow cytometric studies. Subsequent flow cytometric analyses of propidium iodide stained cells revealed differential and significant accumulation of cells within the S and G2M phases of the cell cycle following EM011 treatments, in a dose and cell line dependent manner. For instance, treatments of the pilocytic astrocytoma cell line R286 with EM011 induced a significant difference in cells accumulating within the S-phase, but not the G2M phase of the cell cycle over a 5 day period compared to the untreated cells (Fig. 2B). However, the other pilocytic astrocytoma cell line (Res186) showed a different response to EM011 with no significant difference observed on the first day between treated and untreated Res186 cells. However, by the third and fifth days, Res186 cells began to respond significantly in a dose dependent manner with cells accumulating in the S and G2M phases compared to the untreated cells (Fig. 2B). On the contrary, the diffuse astrocytoma cell line Res 259 portrayed a cell cycle profile that was not significantly different from the untreated cells following administration of low doses of EM011. However, administration of 40 lM EM011 was capable of significantly increasing the proportion of cells within the S and G2M phase, when compared to the untreated cells (Fig. 2B). These findings were consistent with those from the Immunofluorescent cytochemistry studies of microtubules labelled with the alpha tubulin antibody. In this manner, a disruption of the microtubule network and the presence of numerous spherically shaped mitotic figures were seen among cells treated with EM011 (Supplemental Fig. 2). This was in contrary to the appearance of a smooth microtubular architecture seen among untreated or vehicle treated cells (Supplemental Fig. 2). In addition, a gradual decrease in G0/G1 populations was observed among all three cell lines subjected to EM011 treatments. This was accompanied instead by the progressive accumulations of cells within the sub-G1 phase (Fig. 3A); hence suggesting EM011was capable of concurrently inducing apoptosis. 3.3. EM011 induces apoptosis Cancer cells continuously thrive because of their ability to evade apoptosis. Induction of apoptosis is therefore a paramount property of effective anti-tumour agents. Since our precursory flow cytometric analyses initially showed evidence of treating low grade glioma cell lines with EM011 could induce the progressive accumulation of cells within the sub-G1 phases of the cell cycle, we next mechanistically validated the consequences of this drug on apoptosis. First by using annexin V-FITC/propidium iodide

112

N.F. Ajeawung et al. / Cancer Letters 335 (2013) 109–118

Fig. 1. Effect of EM011 on cell viability. EM011 induces a significant dose and time dependent decrease in cell viability as determined by the (A) Trypan blue exclusion test, (B) MTS cytotoxicity assay and (C) Methylene blue live cell staining methods. P-values < 0.05 are denoted with an asterisk (⁄) from triplicate experiments. NT = untreated.

staining experiments, we questioned whether EM011 could mediate early as well as late apoptotic events. Flow cytometric and fluorescent microscopic analyses subsequently revealed dose and time dependent increases in annexin V stained cells following EM011 treatments. In this manner, EM011 induced the translocation of phosphatidyl serine from the inner to outer membrane; a common feature of early apoptotic cells. Furthermore, similar increases in the proportion of propidium iodide stained cells (necrotic cells) were observed at different time points, as the dose of EM011 increased (Fig. 3B). Likewise, with 40 lM of BCNU, EM011 could significantly increase the proportion of apoptotic cells compared to the untreated cells in a dose and time dependent manner. Secondly, we investigated whether EM011 mediated apoptosis was mechanistically associated with the loss of mitochondrial membrane potential using DiOC6 (3,30 -dihexyloxacarbocyanine iodide); a dye which at a low concentration could target the mitochondria with altered membrane permeability. Each paediatric glioma cell line had approximately 10% increases in cells with altered mitochondrial membrane following 48 h of treatment of a low 10 lM dose of EM011. Increasing the dose of EM011 to 20 lM or 40 lM further resulted in the loss of mitochondrial membrane potential (Dwm) in 27–49% of low grade glioma cell lines within a 48 h (Fig. 3C and D); suggesting an involvement of the mitochondria in EM011 mediated apoptosis.

3.4. EM011 mediated apoptosis is associated with transient release of AIF from the mitochondria The apoptosis inducing factor (AIF), a crucial mediator of cell death, plays a relevant role in Noscapine mediated apoptosis in glioma cells [19]. Given that a distortion in the mitochondrial

membrane potential is evident from an exposure of glioma cell lines to EM011, we investigated whether this phenomenon could be associated with the release of the mitochondrial resident protein AIF. Subsequent western blot analyses of the fractionated mitochondria and cytosolic proteins revealed the absence of AIF in the cytosolic fractions of untreated cells and the transient accumulation of AIF in cells treated with EM011 during the 3–48 h interval (Supplemental Fig. 3A). Further studies, using immunofluorescent cytochemistry, demonstrated evidence of a nuclear translocation of the released AIF with an associated nuclear degradation (Supplemental Fig. 3B). Nuclear fragmentation is often accompanied by DNA loss and this could partially explain the decline in DNA content from BrdU labelled cells following EM011 treatments (Fig. 2A).

3.5. EM011 blocks migration and invasion Cancer metastasis is among the major cause of recurrence in patients undergoing glioma therapy. Here, we questioned whether EM011 could mechanistically perturb the migratory and invasive properties of paediatric low grade gliomas. Using the wound healing assay, we observed a significant retardation in the migratory potential of EM011 treated cells in a dose and time dependent manner. The decrease in migratory abilities was obvious with the administration of only 10 lM of EM011, but with further declines noted at 20 lM when approximately 36–48% of the panel of low grade glioma cell lines respectively, showed significant decreases in the migration index, compared to the controls. Moreover, the administration of 40 lM EM011 almost abrogated the migration by causing 59–83% decreases in migration among all cell lines (Fig. 4A and C). To further complement these findings, we utilised the Boyden chamber matrigel assay to assess whether EM011

N.F. Ajeawung et al. / Cancer Letters 335 (2013) 109–118

113

Fig. 2. Effect of EM011 on cell proliferation and cell cycle regulation. Over a 5-day period, EM011 induces a dose and time dependent significant decrease in (A) cell proliferation, as judged by BrdU ELISA experiments, and with (B) cells arresting in the S and G2 M phases of the cell cycle. P-values < 0.05 are denoted with an asterisk (⁄) from triplicate experiments. NT = untreated.

could also perturb the invasion of low grade glioma cells through a three dimensional matrix. Consistent with the results from the wound healing assay, EM011 treatments resulted in significant dose dependent decreases in the invasion index of all three cell lines, compared to the untreated cells (Fig. 4B). 3.6. EM011 inhibits anchorage independent growth Clonogenic survival experiments were conducted for an 18 day duration to assess whether EM011 could abrogate the transformation of low grade gliomas. A visible difference in the size and number of colonies in soft agarose was observed following microscopic analyses of treated and untreated cells in a dose dependent manner. Cells treated with EM011 produced fewer colonies with the average size being smaller than those of untreated cells. For example, in Res259, EM011 dose dependently reduced the number of anchorage independent growth colonies in soft agarose by 55– 70% respectively (Fig. 4D). Furthermore, our tests showed that EM011 can efficiently cross the blood brain barrier (data not shown). However, we were unable to undertake correspondingly in vivo assays, since none of these low grade cell lines could grow into xenograft tumours in immuno-deficient mice. 3.7. EM011 dysregulated the expression of key cancer progression genes and disrupted microtubule formation To further mechanistically investigate how EM011 could mediate a decrease in cell viability, proliferation and anchorage independent growth, and with the converse induction of apoptosis

and cell cycle arrest, we examined its ability to alter the expression of a panel of 89 genes involved in cancer progression. As shown in Fig. 5, the expression of a wide number of genes was affected by EM011 including those involved in migration, growth and cytoskeletal organisation. Matrixmetalloproteinase 2 (MMP2) was the most affected and its expression declined by 86 and 245 fold at 3 h and 24 h respectively, following EM011 administration.

3.8. EM011 inhibits matrix metalloproteinase activity Since EM011 declined MMP2 expression, we further examined whether this drug could perturb the gelatinolytic activity of MMPs using zymography. With the exception of MMP2, no obvious gelatinase activities were observed for other MMP9, including MMP9. This could be explained by the fact that MMP2 expression is increased by approximately 260 fold in Res259 cells compared to MMP9. Despite the profound inhibitory effect of EM011 on MMP2 expression, only a slight reduction in MMP2 gelatinolytic activity in conditioned medium was observed between the treated and untreated samples (Supplemental Fig. 4). It is unclear whether the reduced MMP2 gelatinolytic activity of EM011 treated cells, is a consequence of its diminished MMP2 expression. However, it is likely that this finding should be interpreted with caution considering that, EM011 also induces apoptosis; a process involving the release of bioactive molecules in cell culture medium, leading to an overestimation of MMP2 activity.

114

N.F. Ajeawung et al. / Cancer Letters 335 (2013) 109–118

Fig. 3. Effect of EM011 on apoptosis. (A) EM011 induces dose dependent increase in the sub G1 cell fraction as well as in (B) the proportion of early apoptotic (AV+/PI) and late apoptotic (PI+/PI+) cells. (C) Increasing uptake of DiOC6(3) by mitochondrial of EM011 treated cells depicting loss of mitochondrial membrane potential. (D) Representative images from flow cytometric analysis of Dioc6(3) stained cells. ⁄ = p < 0.05; NT = untreated; AV+ = annexin V positive and PI+ = Propidium iodide positive.

4. Discussion Exploitation of the dynamic nature of microtubules remains a paramount strategy for curbing cancer growth and as a result, microtubule interfering agents continue to be at the forefront of

cancer therapy. EM011 is a promising tubulin binding anti-tumour agent that is synthetically derived from the non-toxic naturally prevalent opiate alkaloid known as noscapine. Previously, noscapine was shown with in vivo and/or in vitro studies to abrogate the growth of various malignancies including tenoposide resistant

N.F. Ajeawung et al. / Cancer Letters 335 (2013) 109–118

115

Fig. 4. Effect of EM011 on cell migration, invasion and anchorage independent growth. EM011 induces a significant dose dependent decreases in the (A) migration index as judged from Would healing assays, and the (B) invasion index as judged from Boyden chamber matrigel assays, among paediatric glioma cell lines. (C) Representative images examples from the wound healing assay. (D) Treatments with EM011 also induce a significant dose dependent inhibition of anchorage independent growth in soft agarose over a 21-day period. ⁄ represents P-values < 0.05 and ⁄⁄⁄ represents P-values < 0.0001; NT = untreated.

lymphoblastoid T cells, T lymphoid tumours, melanoma and breast cancer cells [13,20]. Consistent with these studies, treatment of pilocytic and diffuse low grade glioma cell lines derived from children with EM011, mechanistically resulted in profound decreases in cell growth and migration and the converse induction of cell cycle arrest and apoptosis. These effects were further mechanistically associated with dysregulation of the cell architectural and prosurvival genes in low grade gliomas, including the decreased expression of two major cytoskeletal components, beta actin and beta III tubulin; a crucial component of the microtubule network. One of the principal functions of the cytoskeleton, which is often exploited or reprogrammed during cancer metastasis, is cell motility. A major hallmark in the progression of gliomas is their inherent ability to degrade the extra cellular matrix and hence infiltrate the brain parenchyma. Crossing the basement membrane (invasion), is a multistep process requiring the formation of a protrusion (invadopodia) which perforates the extra cellular matrix, thereby allowing the invadopodia to extend towards the stroma compartment [21]. The actin cytoskeleton plays a pivotal role in invadopodia formation through interaction with actin binding proteins such

as Arp2/3, contactin, and confilin [22]. Once formed, critical mediators of extra cellular matrix degradation such as MMPs and other proteins necessary for invadopodia maturation, could be manoeuvred by microtubules or recruited to the invadopodia by actin regulatory protein contactin as well as other proteins such as the exocyst complex and SNARE [23–27]. The importance of microtubules in this process could be deduced from studies with the microtubule interfering compound, Paclitaxel, that blocks cancer cell migration and invasion [28]. The actin-MMP regulatory invasion loop could further be demonstrated by the fact that inhibition of MMPs could disrupt contactin or actin accumulation at the ventral surfaces of invasive cells [25,26,29]. The decline in Beta actin and Beta III tubulin expression in EM011 treated cells coincided with a significant decrease in the migratory and invasive potential of low grade glioma cell lines treated with EM011. In fact, EM011 treated cells had reduced migration capacity as observed with the scratch assay. Likewise, using the Boyden chamber assay, we showed that a significant portion of EM011 treated cells was unable to traverse the matrigel barrier; suggesting a deficiency in enzymes necessary for the

116

N.F. Ajeawung et al. / Cancer Letters 335 (2013) 109–118

Fig. 5. Effect of EM011 on the expression of key cancer progression genes, using a Real-time PCR based cancer pathway finder array. (A) Scatter plot analyses of untreated verses treated cells with EM011 for 3 h and (B) 24 h. Shown are genes with greater than two fold increase (Red circles) or decrease (Green circles) in gene expression (Pvalues < 0.05), following EM011 administration. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

degradation of the matrix proteins. The observed impairment in the invasive capacity of EM011 cells could be explained by the drastic decline in the expression of a wide number of genes involved in the migration of cancerous cells including MTORC1, ACTB, CTSL1, MMP1, MMP2, MMP3 and MMP9, following EM011 treatment. The MMP family of proteases have emerged as critical mediators of cancer progression due to their ability to degrade the basement membrane and extra cellular membrane; thereby facilitating the migration of cancerous cells. MMPs also target a wide number of substrates including angiogenic molecules, apoptotic and cell adhesion substrates, growth factors and receptors and hence, collectively modulate activities related to cell death, inflammation, survival, proliferation and angiogenesis [30]. The degradative effect of MMPs on the extra cellular matrix modulates cell proliferation and migration via the release of bioactive molecules (matrikines) that augment growth through interaction with growth receptor signalling [31,32]. Our Real time PCR array expression analyses of 89 genes depicted MMP2 as the most affected gene following EM011 treatment. Indeed, MMP2 showed an 86-fold decreased expression within the first 3 hours of EM011 treatment and by the 24th hour, the decrease in MMP2 expression succumbed to 245 fold and was further accompanied by more than 50% decreased expression of other MMPs including MMP1, 3 and 9. Such findings were mechanistically confirmed by zymography which also revealed a possible effect of EM011 on MMP2 gelatinase activity. Interestingly, MMP2 and MMP9 levels were reported to be higher in gliomas than in normal brain tissues, and their expression correlates with the increasing gradation and invasiveness of gliomas [33–37]. Blockade of MMP9 expression through anti-sense could attenuate invasion and the in-vivo growth of glioblastoma [34]. In addition, studies in ovarian and colorectal cancer xenograft models using Batimastat, an inhibitor of MMP1, MMP2, MMP3, MMP7 and

MMP9, led to a significant reduction in tumour size and survival in treated mice compared to controls [38,39]. Likewise, administration of BAY-129556 to colon cancer xenograft model, resulted in substantial decrease in primary tumour size as well a significant declines in both liver and pancreatic metastases [40]. Furthermore, in clinic trials, Marismastat, an MMP inhibitor, was reported to be as effective as Gemcitabine in the treatment of human pancreatic carcinoma [41]. Moreover, administration of Marismastat to patients with advanced gastric cancer led to increases in time-todisease progression and survival [42]. Similar increases in survival time of glioblastoma patients were observed when Temozolomide, the currently approved chemotherapeutic agent for glioblastomas, was combined with Marismastat [43]. Other studies have shown that MMPs can also activate growth factors [44,45] which bind to receptors like EGFR; thereby triggering downstream signalling events leading to enhanced cell growth and migration. Likewise, EGFR can induce MMP expression and hence control cancer cell growth and migration [46] to create an invasion–survival feedback loop. EGFR is over expressed in about 45% of patients with gliomas [47]. Interestingly, EGFR and its downstream signalling target mTOR, had more than 2-fold decline in expression at 24 h following EM011 administration in our panel of paediatric low grade glioma cell lines. The importance of decreasing the levels of EGFR and mTOR in cancer, following administration of a non-toxic compound, cannot be overemphasized. EGFR and mTOR are among the most desirable targets of glioma therapy and their relevance is reflected by the vast amount of EGFR/mTOR inhibitors approved for the treatment of various malignancies as well as those currently in clinical trials [47]. For example, EM011 mediated down-regulation of mTORC1, a component of the mTOR complex, could negatively affect S6K1 activation. This might profoundly affect cell migration since S6K1 regulates phosphorylation of focal adhesion molecules such as

N.F. Ajeawung et al. / Cancer Letters 335 (2013) 109–118

focal adhesion kinase, paxillin and p130cas as well as actin cytoskeletal organisation [48,49]. Cell migration can also be regulated by another member of the mTOR complex (mTORC2) which modulates actin cytoskeleton and cell motility through interaction with P-Rex-1 leading to the activation of Rac [50]. Rac proteins belong to the family of Rho GTPases; important mediators of actin reorganisation, cell migration and metastasis. Interestingly, beta actin (ACTB), a major component of the actin cytoskeleton, which was included in our PCR array among other reference genes, had profound decreased expression in EM011 treated cells; hence supporting EM011 as having a role in the inhibition of cell growth and migration. It is also possible that the decrease in migration observed in EM011 treated cells could be mechanistically attributed to other factors, like the influence of EM011 on cell growth signalling, survival and apoptotic pathways, and including its negative regulation of the expression of cathepsin L1 (CTSL1), an important mediator of glioma progression [51]. For example in gliomas, the expression and activity of cathepsin L1 elevates with increasing tumour gradation and pathological invasiveness [52]. In fact, anti-sense mediated inhibition of cathepsin L perturbs invasiveness and immensely augments Staurosporine mediated apoptosis of human brain tumour cells [53]. Similar studies by Sivaparvathi et al. [54], using antibodies to attenuate cathepsin L, resulted in a profound decrease in the invasiveness of glioblastoma cells in vitro. Likewise, CTSL1 could also mediate a pro-apoptotic role through cleavage of Bid, which in turn activates the intrinsic apoptotic pathway via induction of the outer mitochondrial membrane permeabilisation [51] leading to the release of cell death mediators such as Cyc c, Endo G and AIF. Indeed in this study, flow cytometric analyses of DiOC6 stained low grade glioma cells showed an increase in mitochondrial membrane permeability within 48 h following EM011 administration. Consistent with the alteration in mitochondrial membrane potential, is the finding of a significant augmentation in the BAX/BCL2 ratio by 48 h, coupled with the transient release and nuclear translocation of AIF following administration of EM011. Similar observations of Noscapine mediated releases of AIF in glioma cells and EM011 mediated increases in the BAX/ BCL2 ratio were noted in non-small cell lung cancer and was consistent with Caspase 3 activation and PARP cleavage [16]. However, in our panel of paediatric low grade glioma cell lines, an increase in Caspase 3 expression was not observed, even though its inhibitor, XIAP, had significant reduced expression by EM011. In summary, administration of EM011 to paediatric low grade glioma cell lines could significantly induce growth arrest and apoptosis; thereby succumbing to decreases in cell viability, proliferation, anchorage independent growth and cell migration/invasion. These effects were mechanistically associated with alteration in the expression profile of multiple genes involved in cancer progression, and the disruption of microtubule dynamics. Grant support This work is supported by a Doctorate scholarship to N.F.A. from FRSQ, and grants to D.K. from the CHUQ foundation, Fondation des étoiles, Canadian Foundation for Innovation, Laval University Faculty of Medicine Foundation, Natural Sciences and Engineering Research Council, and Fonds de la recherche en santé du Québec, and to H.C.J. from the National Institutes of Health. D.K. is a FRSQ scholar. Acknowledgement We thank Dr. Kam Kamnasaran for comments on this manuscript; Drs. Michael Bobola (University of Washington, USA) and

117

Chris Jones (Institute for Cancer Research, UK) for providing all cell lines used in this study; and Dr. Georgi Maximov for technical assistance. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.canlet.2013. 02.004. References [1] M.A. Jordan, L. Wilson, Microtubules as a target for anticancer drugs, Nat. Rev. Cancer 4 (2004) 253–265. [2] C. Dumontet, M.A. Jordan, Microtubule-binding agents: a dynamic field of cancer therapeutics, Nat Rev Drug Discov. 9 (2010) 790–803. Review. Erratum in: Nat. Rev. Drug. Discov. 9(11) (2010) 897. [3] J. Zhou, P. Giannakakou, Targeting microtubules for cancer chemotherapy, Curr. Med. Chem. Anticancer Agents 5 (2005) 65–71. [4] H.C. Joshi, A. Salil, U. Bughani, P.K.Naik, Noscapinoids – A New Class of Anticancer Drugs Demand Biothechnological InterventionÓ CAB International Medicinal Plant Biothechnology, 2010, pp. 303–320 (Chapter 19). [5] R. Aneja, T. Miyagi, P. Karna, T. Ezell, D. Shukla, M. Vij Gupta, C. Yates, S.R. Chinni, H. Zhau, L.W. Chung, H.C. Joshi, A novel microtubule-modulating agent induces mitochondriallydriven caspase-dependent apoptosis via mitotic checkpoint activation in human prostate cancer cells, Eur. J. Cancer. 46 (2010) 1668–1678. [6] M. Karajannis, J.C. Allen, E.W. Newcomb, Treatment of pediatric brain tumors, J. Cell Physiol. 217 (2008) 584–589. [7] D.C. Bowers, T.P. Krause, L.J. Aronson, A. Barzi, P.C. Burger, B.S. Carson, et al., Second surgery for recurrent pilocytic astrocytoma in children, Pediatr. Neurosurg. 34 (2001) 229–234. [8] A. Broniscer, S.J. Baker, A.N. West, M.M. Fraser, E. Proko, M. Kocak, et al., Clinical and molecular characteristics of malignant transformation of lowgrade glioma in children, J. Clin. Oncol. 25 (2007) 682–689. [9] A. Gajjar, R. Bhargava, J.J. Jenkins, R. Heideman, R.A. Sanford, J.W. Langston, et al., Low-grade astrocytoma with neuraxis dissemination at diagnosis, J. Neurosurg. 83 (1995) 67–71. [10] D.W. Beebe, M.D. Ris, F.D. Armstrong, J. Fontanesi, R. Mulhern, E. Holmes, et al., Cognitive and adaptive outcome in low-grade pediatric cerebellar astrocytomas: evidence of diminished cognitive and adaptive functioning in National Collaborative Research Studies (CCG 9891/POG 9130), J. Clin. Oncol. 23 (2005) 5198–5204. [11] S. Sharif, R. Ferner, J.M. Birch, J.E. Gillespie, H.R. Gattamaneni, M.E. Baser, et al., Second primary tumors in neurofibromatosis 1 patients treated for optic glioma: substantial risks after radiotherapy, J. Clin. Oncol. 24 (2006) 2570– 2575. [12] B. Donahue, Short- and long-term complications of radiation therapy for pediatric brain tumors, Pediatr. Neurosurg. 18 (1992) 207–217. [13] R. Aneja, J. Zhou, S.N. Vangapandu, B. Zhou, R. Chandra, H.C. Joshi, Drugresistant T-lymphoid tumors undergo apoptosis selectively in response to an antimicrotubule agent, EM011, Blood 107 (2006) 2486–2492. [14] D.A. Bax, S.E. Little, N. Gaspar, L. Perryman, L. Marshall, M. Viana-Pereira, T.A. Jones, R.D. Williams, A. Grigoriadis, G. Vassal, P. Workman, D. Sheer, R.M. Reis, A.D. Pearson, D. Hargrave, C. Jones, Molecular and phenotypic characterisation of paediatric glioma cell lines as models for preclinical drug development, PLoS One 4 (2009) e5209. [15] E.T. Valera, M.A. de Freitas Cortez, R.G. de Paula Queiroz, F.M. de Oliveira, M.S. Brassesco, F.M. de Oliveira, M.S. Brassesco, N. Jabado, D. Faury, M.S. Bobola, H.R. Machado, C.A. Scrideli, T.G. Tone, Pediatric glioblastoma cell line shows different patterns of expression of transmembrane ABC transporters after in vitro exposure to vinblastine, Childs Nerv. Syst. 25 (2009) 39–45. [16] J. Zhou, K. Gupta, S. Aggarwal, R. Aneja, R. Chandra, D. Panda, H.C. Joshi, Brominated derivatives of noscapine are potent microtubule-interfering agents that perturb mitosis and inhibit cell proliferation, Mol. Pharmacol. 63 (2003) 799–807. [17] J. Mukherjee, D. Kamnasaran, A. Balasubramaniam, I. Radovanovic, G. Zadeh, T.R. Kiehl, A. Guha, Human schwannomas express activated platelet-derived growth factor receptors and c-kit and are growth inhibited by Gleevec (Imatinib Mesylate), Cancer Res. 69 (12) (2009) 5099–5107. [18] U. Ozgen, S. Savasan, S. Buck, Y. Ravindranath, Comparison of DiOC6 uptake and annexin V labeling for quantification of apoptosis in leukemia cells and non-malignant T lymphocytes from children, Cytometry 42 (2000) 74–78. [19] E.W. Newcomb, Y. Lukyanov, I. Smirnova, T. Schnee, D. Zagzag, Noscapine induces apoptosis in human glioma cells by an apoptosis-inducing factordependent pathway, Anticancer Drugs 19 (2008) 553–563. [20] P. Karna, S.M. Sharp, C. Yates, S. Prakash, R. Aneja, EM011 activates a survivingdependent apoptotic program in human non-small cell lung cancer cells, Mol. Cancer. 8 (2009) 93. [21] M. Schoumacher, R.D. Goldman, D. Louvard, D.M. Vignjevic, Actin, microtubules, and vimentin intermediate filaments cooperate for elongation of invadopodia, J. Cell Biol. 189 (2010) 541–556.

118

N.F. Ajeawung et al. / Cancer Letters 335 (2013) 109–118

[22] M. Oser, H. Yamaguchi, C.C. Mader, J.J. Bravo-Cordero, M. Arias, X. Chen, V. Desmarais, J. van Rheenen, A.J. Koleske, J. Condeelis, Cortactin regulates cofilin and N-WASp activities to control the stages of invadopodium assembly and maturation, J. Cell Biol. 186 (2009) 571–587. [23] S. Linder, The matrix corroded: podosomes and invadopodia in extracellular matrix degradation, Trends Cell Biol. 17 (3) (2007) 107–117. [24] R. Poincloux, F. Lizarraga, P. Chavrier, Matrix invasion by tumour cells: a focus on MT1-MMP trafficking to invadopodia, J. Cell Sci. 122 (2009) 3015–3024. [25] M. Sakurai-Yageta, C. Recchi, G. Le Dez, J.B. Sibarita, L. Daviet, J. Camonis, C. D’Souza-Schorey, P. Chavrier, The interaction of IQGAP1 with the exocyst complex is required for tumor cell invasion downstream of Cdc42 and RhoA, J. Cell Biol. 181 (2008) 985–998. [26] A. Steffen, G. Le Dez, R. Poincloux, C. Recchi, P. Nassoy, K. Rottner, T. Galli, P. Chavrier, MT1-MMP-dependent invasion is regulated by TIVAMP/VAMP7, Curr. Biol. 18 (2008) 926–931. [27] E.S. Clark, A.M. Weaver, A new role for cortactin in invadopodia: regulation of protease secretion, Eur. J. Cell Biol. 87 (2008) 581–590. [28] A. Westerlund, E. Hujanen, M. Hoyhtya, U. Puistola, T. Turpeenniemi-Hujanen, Ovarian cancer cell invasion is inhibited by paclitaxel, Clin. Exp. Metastasis 15 (1997) 318–328. [29] I. Ayala, M. Baldassarre, G. Giacchetti, G. Caldieri, S. Tete, A. Luini, R. Buccione, Multiple regulatory inputs converge on cortactin to control invadopodia biogenesis and extracellular matrix degradation, J. Cell Sci. 121 (2008) 369– 378. [30] H. Hua, M. Li, T. Luo, Y. Yin, Y. Jiang, Matrix metalloproteinases in tumorigenesis: an evolving paradigm, Cell Mol. Life Sci. 68 (23) (2011) 3853–3868. [31] T. Sadowski, S. Dietrich, F. Koschinsky, et al., Matrix metalloproteinase 19 processes the laminin 5 gamma 2 chain and induces epithelial cell migration, Cell Mol. Life Sci. 62 (2005) 870–880. [32] N. Koshikawa, T. Minegishi, A. Sharabi, V. Quaranta, M. Seiki, Membrane-type matrix metalloproteinase-1 (MT1-MMP) is a processing enzyme for human laminin gamma 2 chain, J. Biol. Chem. 280 (2005) 88–93. [33] M. Nakada, Y. Okada, J. Yamashita, The role of matrix metalloproteinases in glioma invasion, Front. Biosci. 8 (2003) e261–e269. [34] P.A. Forsyth, H. Wong, T.D. Laing, N.B. Rewcastle, D.G. Morris, H. Muzik, K.J. Leco, R.N. Johnston, P.M. Brasher, G. Sutherland, D.R. Edwards, Gelatinase-A (MMP-2), gelatinase-B (MMP-9) and membrane type matrix metalloproteinase-1 (MT1-MMP) are involved in different aspects of the pathophysiology of malignant gliomas, Br. J. Cancer. 79 (1999) 1828–1835. [35] T. Nakagawa, T. Kubota, M. Kabuto, N. Fujimoto, Y. Okada, Secretion of matrix metalloproteinase-2 (72 kD gelatinase = type IV collagenase1=4 gelatinase A) by malignant human glioma cell lines: implications for the growth and cellular invasion of the extracellular matrix, J. Neurooncol. 28 (1996) 13–24. [36] A. Nakano, E. Tani, K. Miyazaki, Y. Yamamoto, J. Furuyama, Matrix metalloproteinases and tissue inhibitors of metalloproteinases in human gliomas, J. Neurosurg. 83 (1995) 298–307. [37] S. Kondraganti, S. Mohanam, S.K. Chintala, Y. Kin, S.L. Jasti, C. Nirmala, S.S. Lakka, Y. Adachi, A.P. Kyritsis, F. Ali-Osman, R. Sawaya, G.N. Fuller, J.S. Rao, Selective suppression of matrix metalloproteinase-9 in human glioblastoma cells by antisense gene transfer impairs glioblastoma cell invasion, Cancer Res. 60 (2000) 6851–6855. [38] B. Davies, P.D. Brown, N. East, et al., A synthetic matrix metalloproteinase inhibitor decreases tumor burden and prolongs survival of mice bearing human ovarian carcinoma xenografts, Cancer Res. 53 (1993) 2087–2091.

[39] X. Wang, X. Fu, P.D. Brown, et al., Matrix metalloproteinase inhibitor BB-94 (batimistat) inhibits human colon tumor growth and spread in a patient-like orthotopic model in nude mice, Cancer Res. 54 (1994) 4726–4728. [40] C. Flynn, C. Bull, D. Eberwein et al., Anti-metastatic activity of BAY 12–9566 in a human colon carcinoma HCT 116 orthotopicmodel [abstract no. 2057], in: Proceedings of the 89thAnnual Meeting of the American Association for Cancer Research; 1998 March 28–April 1; New Orleans (LA), p. 301. [41] S.R. Bramhall, M.T. Hallissey, J. Whiting, J. Scholefield, G. Tierney, R.C. Stuart, R.E. Hawkins, P. Mcculloch, T. Maughan, P.D. Brown, M. Baillet, J.W. Fielding, Marimastat as maintenance therapy for patients with advanced gastric cancer: a randomised trial, Br. J. Cancer. 86 (2002) 1864–1870. [42] S.R. Bramhall, A. Rosemurgy, P.D. Brown, C. Bowry, J.A. Buckels, Marimastat as first-line therapy for patients with unresectable pancreatic cancer: a randomized trial, J. Clin. Oncol. 19 (2001) 3447–3455. [43] M.D. Groves, V.K. Puduvalli, K.R. Hess, K.A. Jaeckle, P. Peterson, W.K. Yung, V.A. Levin, Phase II trial of temozolomide plus the matrix metalloproteinase inhibitor, marimastat, in recurrent and progressive glioblastoma multiforme, J. Clin. Oncol. 20 (2002) 1383–1388. [44] D.W. Perng, K.T. Chang, K.C. Su, Y.C. Wu, C.S. Chen, W.H. Hsu, C.M. Tsai, Y.C. Lee, Matrix metalloprotease-9 induces transforming growth factor-b(1) production in airway epithelium via activation of epidermal growth factor receptors, Life Sci. 89 (2011) 204–212. [45] L.J. Xiao, P. Lin, F. Lin, X. Liu, W. Qin, H.F. Zou, L. Guo, W. Liu, S.J. Wang, X.G. Yu, ADAM17 targets MMP-2 and MMP-9 via EGFR-MEK-ERK pathway activation to promote prostate cancer cell invasion, Int. J. Oncol. 40 (2011) 1714–1724. [46] M. Anand, T.E. Van Meter, H.L. Fillmore, Epidermal growth factor induces matrix metalloproteinase-1 (MMP-1) expression and invasion in glioma cell lines via the MAPK pathway, J. Neurooncol. 104 (2011) 679–687. [47] M.W. Ronellenfitsch, J.P. Steinbach, W. Wick, Epidermal growth factor receptor and mammalian target of rapamycin as therapeutic targets in malignant glioma: current clinical status and perspectives, Target Oncol. 5 (3) (2010) 183–191. [48] L. Liu, L. Chen, J. Chung, S. Huang, Rapamycin inhibits F-actin reorganization and phosphorylation of focal adhesion proteins, Oncogene 27 (2008) 4998– 5010. [49] L. Liu, F. Li, J.A. Cardelli, K.A. Martin, J. Blenis, S. Huang, Rapamycin inhibits cell motility by suppression of mTOR-mediated S6K1 and 4E-BP1 pathways, Oncogene 25 (2006) 7029–7040. [50] I. Hernández-Negrete, J. Carretero-Ortega, H. Rosenfeldt, R. Hernández-García, J.V. Calderón-Salinas, G. Reyes-Cruz, J.S. Gutkind, J. Vázquez-Prado, P-Rex1 links mammalian target of rapamycin signaling to Rac activation and cell migration, J. Biol. Chem. 282 (32) (2007) 23708–23715. [51] V. Gocheva, W. Zeng, D. Ke, D. Klimstra, T. Reinheckel, C. Peters, D. Hanahan, et al., Distinct roles for cysteine cathepsin genes in multistage tumorigenesis, Genes Dev. 20 (2006) 543–556. [52] U. Magdolen, J. Krol, S. Sato, M.M. Mueller, S. Sperl, A. Krueger, M. Schmitt, V. Magdolen, Natural inhibitors of tumor associated proteases, Radiol. Oncol. 36 (2002) 131–143. [53] N. Levicar, R.A. Dewey, E. Daley, T.E. Bates, D. Davies, J. Kos, G.J. Pilkington, T.T. Lah, Selective suppression of cathepsin L by antisense cDNA impairs human brain tumor cell invasion in vitro and promotes apoptosis, Cancer Gene Ther. 10 (2003) 141–151. [54] M. Sivaparvathi, M. Yamamoto, G.L. Nicolson, Z.L. Gokaslan, G.N. Fuller, L.A. Liotta, R. Sawaya, J.S. Rao, Expression and immunohistochemical localization of cathepsin L during the progression of human gliomas, Clin. Exp. Metastasis 14 (1996) 27–34.