- Email: [email protected]
The small GTPase Rac1 is involved in the maintenance of stemness and malignancies in glioma stem-like cells
FEBS Letters 585 (2011) 2331–2338
journal homepage: www.FEBSLetters.org
The small GTPase Rac1 is involved in the maintenance of stemness and malignancies in glioma stem-like cells Chang-Hwan Yoon a, Kyung-Hwan Hyun a, Rae-Kwon Kim a, Hyejin Lee a, Eun-Jung Lim a, Hee-Yong Chung b, Sungkwan An c, Myung-Jin Park d, Yongjoon Suh a, Min-Jung Kim a,e,⇑, Su-Jae Lee a,⇑ a
Department of Chemistry, Research Institute for Natural Sciences, Hanyang University, Seoul 133-791, Republic of Korea Department of Medicine, Hanyang University, Seoul 133-791, Republic of Korea Department of Microbiological Engineering, Kon-Kuk University, Seoul 143-701, Republic of Korea d Division of Radiation Cancer Biology, Korea Institute of Radiological and Medical Sciences, Seoul 139-706, Republic of Korea e Department of Radiation Biology, Environmental Radiation Research Group, Korea Atomic Energy Research Institute, Daejeon, Republic of Korea b c
a r t i c l e
i n f o
Article history: Received 8 February 2011 Revised 7 May 2011 Accepted 26 May 2011 Available online 23 June 2011 Edited by Quan Chen Keywords: Rac1 Glioma stem-like cells Migration and invasion Radiation resistance
a b s t r a c t A subpopulation of cancer cells with stem cell properties is responsible for tumor formation, maintenance, and malignant progression; however, the molecular mechanisms underlying the maintenance of cancer stem-like cell properties have remained unclear. Here, we show that the Rho family GTPase Rac1 is involved in the glioma stem-like cell (GSLC) maintenance and tumorigenicity in human glioma. The Rac1-Pak signaling was markedly activated in GSLCs. Knockdown of Rac1 caused reduction of expression of GSLC markers, self-renewal-related proteins and neurosphere formation. Moreover, down-regulation of Rac1 suppressed the migration, invasion, and malignant transformation in GSLCs. Furthermore, inhibition of Rac1 enhanced radiation sensitivity of GSLCs. These results indicate that the small GTPase Rac1 is involved in the maintenance of stemness and malignancies in GSLCs. Ó 2011 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved.
1. Introduction An increasing body of evidence suggests that a subpopulation of cancer cells with stem-like cell properties is responsible for tumor formation [1,2] and maintenance and malignant progression [3–5]. These rare tumor cells, termed cancer stem-like cells or cancer-initiating cells, are characterized by their strong tumorigenic properties and ability to self-renewal [6,7]. Therefore, development of novel therapeutic strategies that specifically target cancer stemlike cells may more effectively eradicate malignant tumors than conventional treatments, while reducing the risk of relapse and metastasis. It is critical to understand how the properties of cancer stem-like cells make them particularly difficult to eradicate. However, the molecular features of cancer stem-like cells that orchestrate the enrichment and maintenance of stemness and tumorigenic properties remain to be elucidated. Members of the Rho GTPase family Rac, CDC42, and Rho have been shown to be implicated in uncontrolled proliferation, Abbreviations: EGF, epidermal growth factor; bFGF, basic fibroblast growth factor; siRNA, small interfering RNA; shRNA, small hairpin RNA ⇑ Corresponding authors. Address: Department of Chemistry, Research Institute for Natural Sciences, Hanyang University, Seoul 133-791, Republic of Korea (M.-J. Kim). Fax: +82 2 2299 0762. E-mail address: [email protected] (S.-J. Lee).
survival, loss of epithelial cell properties, and abnormal cell migration [8–11], and for that reason, Rho GTPases have been suggested to control the tumor formation and malignant progression in human cancers [12–14]. Emerging evidence indicates that Rac1-dependent signal transduction is essential for malignant cell transformation [15,16]. Moreover, total expression and activity levels of Rac1 are much higher in several types of human cancers than their normal tissues [13,17,18], and a high level of Rac1 plays a fundamental role in cell survival and maintenance of malignant phenotype of high-grade glioma cells [15,19,20]. In these studies, a high expression level of Rac1 is correlated with rapid proliferation and survival of glioma cells, and inhibition of Rac1 greatly reduces glioma proliferation, and induces apoptotic cell death in patient derived glioma cells [21]. These observations have led to the concept that inappropriate expression and activation of Rac1 is involved in the development and promotion of cancer; however, the molecular mechanisms by which small GTPase Rac1 might contribute to these processes have remained obscure. The aim of this study was to find out the role of the Rho GTPase family Rac1 in the maintenance of malignant phenotypes in human glioma cells. Our results show that activation of Rac1 is critically involved in the enrichment of the glioma stem-like cell population and the maintenance of malignancies. The findings of this study
0014-5793/$36.00 Ó 2011 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.febslet.2011.05.070
2332
C.-H. Yoon et al. / FEBS Letters 585 (2011) 2331–2338
may provide a basis for improving therapeutic interventions against malignant brain cancer. 2. Materials and methods 2.1. Cell, chemical reagents, and antibodies U87MG and U373MG (glioma cancer cells) were obtained from the Korean Cell Line Bank (KCLB) and cultured in Dulbecco’s Modified Eagle Medium (DMEM) containing 10% fetal bovine serum (FBS), 100 U/ml penicillin and 100 mg/ml streptomycin in a humidified incubator at 37 °C with 5% CO2 atmosphere. For the process of isolation and identification of glioma stem cells (U87 spheres and U373 spheres), U87MG and U373MG were collected and inoculated into a serum-free medium which contained DMEM/F12 with high glucose medium, 20 ng/ml of epidermal growth factor (EGF), basic fibroblast growth factor (bFGF) and B27 (1X). The cells were placed in incubator on conditions of 37 °C, 5% CO2 atmosphere. In every 3– 4 days the medium was renewed [22,23]. Polyclonal antibody CD133 and Snail were obtained from Abcam, and Rac/cdc42 assay reagent (PAK1 PBD, agarose), polyclonal antibodies to anti-nestin, anti-Musashi1 were purchased from Millipore. The monoclonal antibody Rac1, CDC42 and Annexin V-FITC were obtained from BD Transduction Laboratories™. Polyclonal antibody Slug and Zeb1 were obtained from Santa Cruz. The polyclonal antibody Vimentin was obtained from Thermo Science. 4,6-Diamidino-2-phenylindole (DAPI), EGF and monoclonal anti-b-actin were obtained from Sigma. bFGF was purchased from R&D Systems. Anti-mouse Alexa Fluor 488, anti-rabbit Alexa Flour 488 and B27 were purchased from Invitrogen. CD133 (directly conjugated with phycoerythrin, clone 293C3) and mIgG2b-PE were purchased from Miltenyi Biotech Ltd. 2.2. Spheres culture U87MG and U373MG cells were resuspended in DMEM-F12 (Invitrogen) containing 20 ng/ml of EGF, bFGF and B27 (1X) as a stem cell–permissive medium. Spheres were collected after 5– 7 days, and protein was extracted for additional experiments or dissociated with Accutase (Innovative Cell Technologies Inc.) for expansion [24,25].
was purchased from Samchully Pharmaceutical Co. Ltd. (Seoul, Korea). A control siRNA specific for green fluorescent protein (50 -CCACTACCTGAGCACCCAG-30 ) was used as the negative control. siRNA duplexes (40 nM) were introduced into cells using a Microporator-mini (Invitrogen) by following the procedure recommended by the manufacturer. After 48 h, the medium was added, and cells were harvested for additional experiments. 2.6. Lentiviral shRNA experiments Small hairpin RNA (shRNA) constructs against human Rac1 was purchased from OriGene Technologies Inc. (Rockville, MD) (shRNA Rac1; Clone ID TRCN0000004870, non-targeting shRNA; Sigma– Aldrich SHC002). 293T cells were seeded (3 106 cells/ml) in 60-mm dish. Lentiviral particles were generated using a three-plasmid system, as described previously [27,28]. To standardize lentiviral transduction assays, viral titers were measured in a benchmark cell line, A549. For growth assays, titers corresponding to multiplicities of infection (MOIs) of 5 and 1 in A549 cells were employed. For Rac1 knockdown, cells were plated on day zero at 1 105 cells/ml in 60-mm dishes. Twenty-four hours after infection, cells were treated with 1 lg/ml puromycin for 1 days to eliminate uninfected cells. Media was replaced and after 2 more days, cells were harvested for Western blot or for additional experiments. 2.7. Rac1 activation assay Cells were lysed lysis buffer [40 mM Tris–HCl (pH 8.0), 120 mM NaCl, 0.1% Nonidet-P40] supplemented with protease inhibitors. For Rac/cdc42 assay, Cellular debris was removed by centrifugation at 13 200 rpm for 15 min at 4 °C. Protein concentrations were determined using Bio-Rad protein assay reagents, and total cell lysate (300 lg protein in 0.3 ml) was immunoprecipitated with 10 lg PAK-1 PBD agarose bead (for RAC and CDC42) at 4 °C for 3 h. After washing three times with lysis buffer, the pull-out GTP-bound RAC/CDC42 protein was eluted by boiling in SDS-sample buffer and evaluated by Western blot analysis. 2.8. Western blot analysis
Sphere-forming glioma cells were first distributed into 96-well plates at a density of 1–2 cells/well. After 12 h, individual wells were visually checked for the presence of a single cell. The clones were grown and tested for the neurospheres formation. Neurospheres with a diameter >20 lm were counted using an inverted microscope at days 1, 4, 7 and 10 [26].
Cell lysates were prepared by extracting proteins with lysis buffer [40 mM Tris–HCl (pH 8.0), 120 mM NaCl, 0.1% Nonidet-P40] supplemented with protease inhibitors. Proteins were separated by sodium dodecyl sulfate–polyacrylamide (SDS) gel electrophoresis and transferred to a nitrocellulose membrane (Amersham Biosciences, Arlington Heights, IL). The membrane was blocked with 5% non-fat dry milk in Tris-buffered saline and incubated with primary antibodies for overnight at 4 °C. Blots were developed with a peroxidase-conjugated secondary antibody and proteins visualized by enhanced chemiluminescence (ECL, Amersham) procedures, using the manufacturer’s protocol.
2.4. Flow cytometric analyses
2.9. Invasion and motility assays
Cells were dissociated and resuspended in PBS containing 0.5% bovine serum albumin. For magnetic labeling, CD133 Micro Beads were used (Miltenyi Biotech). Positive magnetic cell separation (MACS) was done using 25MS MACS columns in series. Cells were stained with CD133/2-PE (Miltenyi Biotech) or isotype control antibody (mIgG2b-PE, Miltenyi Biotech) and analyzed on a BD FACSCalibur flow cytometer.
Cells (2 104 cells/well) were suspended in 0.2 ml of serumfree DMEM or serum-free DMEM-F12 supplemented with B27, FGF and hEGF for invasion and motility assays. For invasion assay, the cells were loaded in the upper well of the Transwell chamber (8-mm pore size; Corning, Corning, NY) that was precoated with 10 mg/ml growth factor reduced (BD Matrigel™ matrix – BD Biosciences) on an upper side of the chamber with the lower well filled with 0.8 ml of serum-free DMEM or serum-free DMEMF12 supplemented with B27, FGF and hEGF. After incubation for 48 h at 37 °C, non-invaded cells on the upper surface of the filter were removed with a cotton swab, and migrated cells on the lower surface of the filter were fixed and stained with a Diff-Quick kit
2.3. Sphere colony counting
2.5. Small interfering RNA (siRNA) transfection siRNA targeting of Rac1 was performed using 21 base pairs (including a 2-deoxynucleotide overhang). siRNA target of Rac1
C.-H. Yoon et al. / FEBS Letters 585 (2011) 2331–2338
(Fisher Scientific) and photographed (magnification 200). Invasiveness was determined by counting cells in five microscopic fields per well, and the extent of invasion was expressed as an average number of cells per microscopic field. Cells were imaged with by phase-contrast microscopy (Leica Microsystems, Bannockburn, IL). For motility studies, we used invasion chambers with control inserts that contained the same type of membrane but without the Matrigel coating (one chamber per well of a 24-well plate). 2 104 cells in 0.2 ml of serum-free DMEM or serum-free DMEM-F12 supplemented with B27, FGF and hEGF were added to the apical side of each insert, and 0.8 ml of serum-free DMEM was added to the basal side of each insert. The inserts were processed as described above for the invasion assay. 2.10. Soft agar colony formation assay To examine anchorage-independent growth, a cell suspension (2 104 cells) were suspended in 0.4% agarose with serum-free DMEM-F12 supplemented with B27, FGF and hEGF and seeded in triplicate on 60-mm dishes pre-coated with 0.8% agar in complete growth medium and incubated at 37 °C, 5% CO2. After 12–30 days, colonies were photographed and counted in four randomly chosen fields and expressed as means of triplicates, representative of two independent experiments. 2.11. Tumor xenografts on nude mice After transfection with shRNA, U373 sphere forming glioma cells (4 104) were subcutaneously inoculated to the right flank of athymic Balb/c female nude mice (5 weeks of age). Tumor size was measured with a caliper (calculated volume = shortest diameter2 longest diameter/2) at 5 day intervals. This study was reviewed and approved by the Institutional Animal Care and Use Committee (IACUC) of Korea Institute of Radiological and Medical Sciences (KIRAMS). 2.12. Quantification of cell death FACS analysis using propidium iodide (PI) staining detects cell death by means of the dye entering the cells along with changes in the target cell membrane and DNA damage. For the cell-death assessment, cells were harvested by trypsinization, washed in PBS, and then incubated in propidium iodide (50 ng/ml) for 5 min at room temperature. Then, cells (10 000 per sample) were analyzed on a BD FACSCalibur flow cytometer, using Cell Quest software (BD Biosciences). 2.13. Immunofluorescent staining Immunocytochemistry of tumor spheres was performed as described [5]. Antibodies used were as follows: human anti-CD133 (1:200), -nestin (1:200) and -musashi1 (1:200) for the detection of neural stem. Stainings were visualized using anti-mouse Alexa Fluor 488 and anti-rabbit Alexa Flour 488 (Molecular Probes). Nuclei were counterstained using 4,6-diamidino-2-phenylindole (DAPI; Sigma). Stained cells were visualized with a fluorescencemicroscope (Olympus IX71). 2.14. Statistical analysis All the data in the study were performed in triplicate. The results are described as mean ± S.D. Statistical analysis was performed by one-way analysis of variance (ANOVA) and comparisons among groups were performed by independent sample t-test. Differences were considered significant at values of ⁄P < 0.001.
2333
3. Results and discussion 3.1. Rac1 is involved in the maintenance of glioma stem-like cells Rho family small GTPase Rac1 has been implicated in the tumor formation and malignant progression in human glioma [1–5]. To examine whether Rac1 is involved in the maintenance of glioma stem-like cells, we first examined the activation status of Rac1 in glioma stem-like cells using Rac1-PAK binding assay. As shown in Fig. 1A, binding activity of Rac1 to PAK was markedly increased in U87 and U373 glioma cells grown under sphereforming conditions compared to matched glioma cells grown as a monolayer. There was no change in Rac1 at the total protein level. To further determine whether the activation of Rac1 is required for the maintenance of the glioma stem-like cell population. To further study the involvement of Rac1 in the maintenance of the glioma stem-like cell population, we down-regulated Rac1 in sphere forming U87 and U373 glioma cells by treatment with siRNA. Treatment with siRNA or shRNA targeted to Rac1 specifically down-regulated Rac1, while the expression levels of b-actin and other GTpase, CDC42 were not altered (Fig. S1A). Importantly, treatment with siRNA targeting of Rac1 resulted in the increase of binding activity of Rac1 to PAK and a marked decrease in the number of spheres (Fig. 1B). Moreover, the CD133+ cell population was clearly suppressed by siRNA-mediated knockdown of Rac1 (Fig. 1C). Treatment with siRNA targeted to Rac1 also attenuated expression of glioma stem-like cell marker proteins, CD133, Nestin and Musashi-1 (Fig. 1D). In addition, a fluorescence microscopic analysis of cells immunostained with antibodies specific for the marker proteins clearly revealed that siRNA targeting of Rac1 effectively suppressed expression level of CD133, Nestin, and Musashi-1 (Fig. 1E). These results specifically implicate Rac1 in the maintenance of glioma stem-like cell population.
3.2. Activation of Rac1 is critical for the proliferation and self-renewal in glioma stem-like cells Enhanced self-renewal ability is one of the key characteristics of cancer stem-like cells. To determine whether Rac1 is critical for the proliferation and self-renewal in glioma stem-like cells, we inhibited Rac1 expression with siRNA and examined changes in self-renewal potential using an in vitro self-renewal assay, and also examined changes in the expression levels of the known self-renewal-related proteins Sox2, Notch1/2, and b-catenin. As shown in Fig. 2A, treatment with siRNA targeted to Rac1 clearly inhibited sphere-forming ability of glioma stem-like cells at single cell levels (self-renewal potential). In contrast, down-regulation of Rac1 in monolayer-cultured glioma cells did not suppress cell growth (Fig. S1B), suggesting that Rac1 is required only for the proliferation of glioma stem-like cells, but not monolayer-cultured cells. Consistent with these results, the expression of self-renewal-related proteins Sox2, Notch2 and b-catenin was markedly higher in sphere forming glioma cells, compared to monolayer-cultured glioma cells (Fig. 2B), again confirming that glioma stem-like cells are highly enriched in sphere forming glioma cells more than monolayer-cultured cells. Importantly, treatment with siRNA targeted to Rac1 led to a decrease of Sox2, Notch2, and b-catenin at the protein level in sphere-forming glioma stem-like cells, but did not affect the level of Notch1 (Fig. 2B). Since treatment with siRNA specifically targeted to Rac1 (Fig. S1A) and siRNA-mediated down-regulation of Rac1 was sustained at least for 10 days (Fig. S1C), Rac1 would be directly involved in down-regulation of self-renewal-related proteins, thereby suppressing sphere
2334
C.-H. Yoon et al. / FEBS Letters 585 (2011) 2331–2338
Fig. 1. (A) U87 and U373 monolayers or spheres lysates were immunoprecipitated with 10 lg PAK-1 PBD agarose bead. Total proteins were subjected to immunoblot analysis with anti-Rac1 and b-actin antibodies. b-Actin was used as the loading control. (B) U87 and U373 spheres were transfected with si-RNA of Rac1 or control. (Left) After 48 h, total cell lysates was immunoprecipitated with 10 lg PAK-1 PBD agarose bead. And proteins were subjected to immunoblot analysis with anti-Rac1 and b-actin antibodies. bActin was used as the loading control. (Right) Graph shows the number of spheres per visual field of microscopy. Error bars represent mean ± S.D. of five different fields. An asterisk () indicates P < 0.001 with one-way ANOVA comparison of the control groups to the corresponding Rac1 inhibition groups. (C) U87 and U373 spheres were transfected with either si-RNA of Rac1 or control. After 48 h, cells were analyzed for the expression of glioma-stem cell marker CD133 by flow cytometry. Error bars represent mean ± S.D. of triplicate samples. An asterisk () indicates P < 0.001 with one-way ANOVA comparison of the control groups to the corresponding Rac1 inhibition groups. (D) After 48 h, total cell lysates were subjected to immunoblot analysis with anti-CD133, -Nestin, or -Musashi1 antibodies. b-Actin was used as the loading control. (E) U87 and U373 spheres were transfected with either si-RNA of Rac1 or control. After 48 h, cells were stained with FITC-conjugated with anti-CD133, -Nestin, or -Musashi1 antibodies and imaged by fluorescence microscopy (Olympus IX71). Photomicrographs were taken at magnification 400.
C.-H. Yoon et al. / FEBS Letters 585 (2011) 2331–2338
2335
Fig. 2. (A) U87 and U373 sphere were transfected with either control-siRNA or si-Rac1. Cells were then distributed into 96-well plates at a density of 1 cell/well. The ability of a single clone to form new spheres was tested by microscopy after 10 days. (Left) Images represent changes in spheres size in the presence of control-siRNA or si-Rac1 (magnification 200). (Right) Number of clones forming spheres with a diameter >20 lm was counted using inverted in the presence of control-siRNA or si-Rac1. Error bars represent mean ± S.D. of triplicate samples. An asterisk () indicate respectively P < 0.001 with one-way ANOVA comparison of the control groups to the corresponding Rac1 inhibition groups. (B) U87 and U373 sphere were transfected with either control-siRNA or si-Rac1. After 48 h, total cell lysates were subjected to immunoblot analysis with anti-Notch1/2, -Sox2, -b-catenin, and Rac1 antibodies. b-Actin was used as the loading control.
formation. Our findings are supported by previous studies that down-regulation of Notch, Sox2 and b-catenin led to a decrease of sphere formation [33–35]. To confirm the role of Rac1 in the stemness of glioma stem-like cell population, we also treated patient-derived glioma stem-like cells (X01GB) with siRNA targeting of Rac1, and examined changes in the CD133+ cell population, sphere formation, and self-renewal potential. As shown in Fig. S2A, siRNA-mediated Rac1 knockdown reduced the CD133+ cell population and decreased the number of spheres. The self-renewal potential in patient-derived glioma stem-like cells was also inhibited by down-regulation of Rac1 (Fig. S2B). Taken together, these results suggest that Rac1 plays a critical role in the maintenance of stemness in glioma stem-like cells. 3.3. Rac1 is involved in the maintenance of the malignant properties of glioma stem-like cells We next examined whether Rac1 is required for maintenance of the malignant properties of sphere forming glioma cells. To this end, we treated sphere-forming glioma cells with shRNA and examined anchorage-independent colony formation in soft agar. Contrary to normal cells that do not proliferate without anchorage, tumorigenic cancer cells can grow in a condition without anchorage such as semisolid medium. Thus, colony forming assay in soft agar is widely adapted to measure tumorigenic capacity in vitro, as a good predictor of tumor forming ability in vivo. As shown in Fig. 3A, shRNA-mediated knockdown of Rac1 effectively attenuated
colony formation of sphere forming glioma cells in soft agar. To further test the involvement of Rac1 in the maintenance of malignant potential in sphere forming cells, we inhibited Rac1 expression with siRNA and analyzed changes in cellular invasion and migration. Notably, invasive behavior and cell migration potential of glioma stem-like cells were markedly suppressed by down-regulation of Rac1 (Fig. 3B and C). These results indicate that Rac1 is involved in the maintenance of malignant potential in cancer stem-like cell population. A mesenchymal phenotype is the hallmark of tumor aggressiveness in human malignant glioma. To further test whether Rac1 is critical for maintenance of mesenchymal phenotype in glioma stem-like cells, we knocked down Rac1 with siRNA and analyzed changes in protein expression levels of mesenchymal cell markers N-cadherin and Vimentin, and the transcription repressors Zeb1, Snail, and Slug. As expected, expression levels of mesenchymal cell markers and the transcription repressors were much higher in sphere forming glioma stem-like cells than matched glioma cells grown as a monolayer (data not shown). Importantly, knockdown of Rac1 with siRNA clearly suppressed N-cadherin and Vimentin in glioma stem-like cells (Fig. 3D). Moreover, treatment with siRNA targeted to Rac1 effectively inhibited expression of Zeb1 and Slug proteins, though the protein level of Snail was not altered. These results indicate that Rac1 plays a critical role for the maintenance of mesenchymal phenotype in glioma stem-like cells. Since down-regulation of Rac1 attenuated colony formation in soft agar and suppressed cell migration and invasion in vitro, we next examined whether Rac1 is required for tumor formation in
2336
C.-H. Yoon et al. / FEBS Letters 585 (2011) 2331–2338
Fig. 3. (A) U87 and U373 sphere were infected with either scramble-shRNA or sh-Rac1 virus, and after 24 h with 1 lg/ml puromycin, infected cells (2 104 cells/ml) were seeded in soft agar plate. The colonies were calculated relative to 1000 cells during 4 weeks of plating. (Left) The colony was determined by counting cells in randomly selected four microscopic fields per plate. Photomicrographs were taken at magnification 200. (Right) Graph shows the number of spheres per visual field of microscopy in the presence of scramble-shRNA or sh-Rac1 virus. Error bars represent mean ± S.D. of four different fields. An asterisk () indicates P < 0.001 with one-way ANOVA comparison of the control groups to the corresponding Rac1 inhibition groups. (B) U87 and U373 monolayers or spheres were transfected with either control-siRNA or si-Rac1, and then the cells (2 104 cells/well) were seeded in each Matrigel chamber and cultured in serum-free DMEM-F12 supplemented with B27, FGF and hEGF for 48 h. (Left) Invaded cells were fixed and stained with a Diff-Quick kit and photographed (magnification 100). (Right) Invasiveness was determined by counting cells in randomly selected five microscopic fields per well. Bars represent mean ± S.D. of triplicate samples. An asterisk () indicates P < 0.001 with one-way ANOVA comparison of the control groups to the corresponding Rac1 inhibition groups. (C) U87 and U373 monolayers or spheres were transfected with either control-siRNA or si-Rac1, and then the cells (2 104 cells/well) were seeded in each chamber (without the Matrigel) and cultured in serum-free DMEM-F12 supplemented with B27, FGF and hEGF for 48 h. (Left) Migrated cells were fixed and stained with a Diff-Quick kit and photographed (magnification 100). (Right) Cell migration was determined by counting cells in randomly selected five microscopic fields per well. Error bars represent mean ± S.D. of triplicate samples. An asterisk () indicates P < 0.001 with one-way ANOVA comparison of the control groups to the corresponding Rac1 inhibition groups. (D) U87 and U373 spheres were transfected with either control-siRNA or si-Rac1, and after 48 h, total cell lysates were subjected to immunoblot analysis with anti-N-cadherin, -Vimentin, -Zeb1, -Snail, or -Slug antibodies. b-Actin was used as the loading control.
C.-H. Yoon et al. / FEBS Letters 585 (2011) 2331–2338
2337
Fig. 4. U373 sphere forming glioma cells were transfected with either control scrambled shRNA or sh-Rac1. Transfected cells were then subcutaneously inoculated to the right flank of athymic Balb/c female nude mice (5 weeks of age, n = 4 per group). Tumor size was measured with a caliper (calculated volume = shortest diameter2 longest diameter/2) at 5 day intervals. Representative photos shows differential volume of tumor at 35 days. Error bars represent mean ± S.D. ⁄P < 0.001.
vivo. To this end, we treated sphere-derived U373 cells with siRNA targeted to Rac1 and then transplanted to athymic Balb/c female nude mice subcutaneously. Notably, tumor formation was dramatically attenuated in mice transplanted with U373 sphere-derived glioma cells treated with shRNA Rac1, compared to scrambled shRNA (Fig. 4). Taken together, these results have implications that Rac1 is a critical regulator in vitro and in vivo for maintenance of the malignant properties in glioma stem-like cells. 3.4. Rac1 is important for the maintenance of resistance to anticancer treatments in glioma stem-like cells Cancer stem-like cells have been implicated in the resistance to chemotherapy and radiotherapy [36,37]. As expected, sphereforming glioma cells showed greater resistance to both ionizing radiation and anticancer drug, etoposide than monolayer cells (data not shown), confirming that cancer stem-like cells are enriched in sphere-cultured glioma cells. To examine whether activation of Rac1 is involved in the maintenance of resistance to anticancer treatments in glioma stem-like cells, we inhibited Rac1 expression with siRNA, and then examined changes in sensitivity of glioma stem-like cells to ionizing radiation and etoposide in sphere forming glioma stem-like cells. Though single treatment with ionizing radiation or etoposide or siRNA targeted to Rac1 did not cause significant cell death, pretreatment with siRNA targeted to Rac1 sensitized sphere forming glioma cells to ionizing radiation and etoposide, inducing significant cell death (Fig. 5). Concurrently with cell death, we also observed the increase of Annexin V positive cell population by ionizing radiation or etoposide treatment only in pretreated sphere forming glioma cells with siRNA targeted to Rac1 (Fig. S3), indicating that cell death is due to apoptosis. Taken together, these results suggest that these results indicate that Rac1 is important for the maintenance of resistance to anticancer treatments in glioma stem-like cells. The Rho family small GTPase Rac1 has been implicated in cell survival and malignant transformation in human cancers including glioma [15,19,20,29–31]. Similar observations have led to the concept that inappropriate activation of Rac1 is involved in the development and promotion of glioma [8,9,12,32]. In this study, we further found that the Rho family GTPase Rac1 is involved in the stemness and tumorigenicity in glioma stem-like cells. The Rac1 signaling was markedly activated in sphere forming glioma stem-like cells. Importantly, inhibition of Rac1 signaling with siRNA reduced expression of glioma stem-like cell markers and self-renewal-related proteins as well as sphere formation. Moreover, siRNA-targeting of Rac1 effectively suppressed the migration, invasion, and malignant transformation in sphere forming glioma stem-like cells.
Fig. 5. U87 and U373 spheres were transfected with either control-siRNA or siRac1. Cells were then treated with ionizing radiation (10 Gy) and etoposide (10 lM). After 48 h, cell death was assessed by flow cytometry as described in Section 2. Error bars represent mean ± S.D. of triplicate samples.
Furthermore, siRNA-mediated knockdown of Rac1 effectively enhanced radiation sensitivity of glioma stem-like cells. These results indicate that the small GTPase Rac1 is involved in the maintenance of stemness and malignancies in the glioma stem-like cells. These findings may provide pivotal insights in the context of therapeutic interventions to target glioma stem-like cell population. Acknowledgements This work was supported by the Korea Research Foundation (KRF) and Ministry of Education, Science and Technology (MEST), Korean government, through its National Nuclear Technology Program (2008-2003935) and Medical Research Center Program (2010-0029504). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.febslet.2011.05.070. References [1] Schatton, T. and Frank, M.H. (2008) Cancer stem cells and human malignant melanoma. Pigment Cell Melanoma Res. 21, 39–55. [2] Singh, S.K., Clarke, I.D., Hide, T. and Dirks, P.B. (2004) Cancer stem cells in nervous system tumors. Oncogene 23, 7267–7273. [3] Bao, S. et al. (2006) Stem cell-like glioma cells promote tumor angiogenesis through vascular endothelial growth factor. Cancer Res. 66, 7843–7848. [4] Zhou, J. et al. (2007) Activation of the PTEN/mTOR/STAT3 pathway in breast cancer stem-like cells is required for viability and maintenance. Proc. Natl. Acad. Sci. USA 104, 16158–16163. [5] Inagaki, A., Soeda, A., Oka, N., Kitajima, H., Nakagawa, J., Motohashi, T., Kunisada, T. and Iwama, T. (2007) Long-term maintenance of brain tumor stem cell properties under at non-adherent and adherent culture conditions. Biochem. Biophys. Res. Commun. 361, 586–592. [6] He, S., Nakada, D. and Morrison, S.J. (2009) Mechanisms of stem cell selfrenewal. Annu. Rev. Cell Dev. Biol. 25, 377–406.
2338
C.-H. Yoon et al. / FEBS Letters 585 (2011) 2331–2338
[7] Sullivan, J.P., Minna, J.D. and Shay, J.W. (2010) Evidence for self-renewing lung cancer stem cells and their implications in tumor initiation, progression, and targeted therapy. Cancer Metastasis Rev. 29, 61–72. [8] Murai, T., Miyazaki, Y., Nishinakamura, H., Sugahara, K.N., Miyauchi, T., Sako, Y., Yanagida, T. and Miyasaka, M. (2004) Engagement of CD44 promotes Rac activation and CD44 cleavage during tumor cell migration. J. Biol. Chem. 279, 4541–4550. [9] Parri, M. and Chiarugi, P. (2010) Rac and Rho GTPases in cancer cell motility control. Cell Commun. Signal. 8, 23. [10] Castillo-Lluva, S., Tatham, M.H., Jones, R.C., Jaffray, E.G., Edmondson, R.D., Hay, R.T. and Malliri, A. (2010) SUMOylation of the GTPase Rac1 is required for optimal cell migration. Nat. Cell Biol. 12 (11), 1078–1085. [11] Karlsson, R., Pedersen, E.D., Wang, Z. and Brakebusch, C. (2009) Rho GTPase function in tumorigenesis. Biochim. Biophys. Acta 1796, 91–98. [12] Espina, C. et al. (2008) A critical role for Rac1 in tumor progression of human colorectal adenocarcinoma cells. Am. J. Pathol. 172, 156–166. [13] Hwang, S.L., Hong, Y.R., Sy, W.D., Lieu, A.S., Lin, C.L., Lee, K.S. and Howng, S.L. (2004) Rac1 gene mutations in human brain tumours. Eur. J. Surg. Oncol. 30, 68–72. [14] Gomez del Pulgar, T., Benitah, S.A., Valeron, P.F., Espina, C. and Lacal, J.C. (2005) Rho GTPase expression in tumourigenesis: evidence for a significant link. BioEssays 27, 602–613. [15] Nakada, M., Drake, K.L., Nakada, S., Niska, J.A. and Berens, M.E. (2006) EphrinB3 ligand promotes glioma invasion through activation of Rac1. Cancer Res. 66, 8492–8500. [16] Johnson, E., Seachrist, D.D., DeLeon-Rodriguez, C.M., Lozada, K.L., Miedler, J., Abdul-Karim, F.W. and Keri, R.A. (2010) HER2/ErbB2-induced breast cancer cell migration and invasion require p120 catenin activation of Rac1 and Cdc42. J. Biol. Chem. 285, 29491–29501. [17] Schnelzer, A. et al. (2000) Rac1 in human breast cancer: overexpression, mutation analysis, and characterization of a new isoform, Rac1b. Oncogene 19, 3013–3020. [18] Wang, Z. et al. (2010) Rac1 is crucial for Ras-dependent skin tumor formation by controlling Pak1-Mek-Erk hyperactivation and hyperproliferation in vivo. Oncogene 29, 3362–3373. [19] Chuang, Y.Y., Tran, N.L., Rusk, N., Nakada, M., Berens, M.E. and Symons, M. (2004) Role of synaptojanin 2 in glioma cell migration and invasion. Cancer Res. 64, 8271–8275. [20] Chan, A.Y., Coniglio, S.J., Chuang, Y.Y., Michaelson, D., Knaus, U.G., Philips, M.R. and Symons, M. (2005) Roles of the Rac1 and Rac3 GTPases in human tumor cell invasion. Oncogene 24, 7821–7829. [21] Senger, D.L. et al. (2002) Suppression of Rac activity induces apoptosis of human glioma cells but not normal human astrocytes. Cancer Res. 62, 2131– 2140.
[22] Jin, F. et al. (2008) Comparison between cells and cancer stem-like cells isolated from glioblastoma and astrocytoma on expression of anti-apoptotic and multidrug resistance-associated protein genes. Neuroscience 154, 541– 550. [23] Jin, F. et al. (2008) Paradoxical expression of anti-apoptotic and MRP genes on cancer stem-like cell isolated from TJ905 glioblastoma multiforme cell line. Cancer Invest. 26, 338–343. [24] Singh, S.K., Clarke, I.D., Terasaki, M., Bonn, V.E., Hawkins, C., Squire, J. and Dirks, P.B. (2003) Identification of a cancer stem cell in human brain tumors. Cancer Res. 63, 5821–5828. [25] Suslov, O.N., Kukekov, V.G., Ignatova, T.N. and Steindler, D.A. (2002) Neural stem cell heterogeneity demonstrated by molecular phenotyping of clonal neurospheres. Proc. Natl. Acad. Sci. USA 99, 14506–14511. [26] Lu, F. and Wong, C.S. (2005) A clonogenic survival assay of neural stem cells in rat spinal cord after exposure to ionizing radiation. Radiat. Res. 163, 63–71. [27] Singh, A., Greninger, P., Rhodes, D., Koopman, L., Violette, S., Bardeesy, N. and Settleman, J. (2009) A gene expression signature associated with ‘‘K-Ras addiction’’ reveals regulators of EMT and tumor cell survival. Cancer Cell 15, 489–500. [28] Moffat, J. et al. (2006) A lentiviral RNAi library for human and mouse genes applied to an arrayed viral high-content screen. Cell 124, 1283–1298. [29] Kamai, T., Yamanishi, T., Shirataki, H., Takagi, K., Asami, H., Ito, Y. and Yoshida, K. (2004) Overexpression of RhoA, Rac1, and Cdc42 GTPases is associated with progression in testicular cancer. Clin Cancer Res 10, 4799–4805. [30] Vega, F.M. and Ridley, A.J. (2008) Rho GTPases in cancer cell biology. FEBS Lett. 582, 2093–2101. [31] Sahai, E. and Marshall, C.J. (2002) RHO-GTPases and cancer. Nat. Rev. Cancer 2, 133–142. [32] Baugher, P.J., Krishnamoorthy, L., Price, J.E. and Dharmawardhane, S.F. (2005) Rac1 and Rac3 isoform activation is involved in the invasive and metastatic phenotype of human breast cancer cells. Breast Cancer Res. 7, R965–R974. [33] Grudzien, P. et al. (2010) Inhibition of Notch signaling reduces the stem-like population of breast cancer cells and prevents mammosphere formation. Anticancer Res. 30, 3853–3867. [34] Malanchi, I. et al. (2008) Cutaneous cancer stem cell maintenance is dependent on beta-catenin signalling. Nature 452, 650–653. [35] Xiang, R. et al. (2011) Downregulation of transcription factor SOX2 in cancer stem cells suppresses growth and metastasis of lung cancer. Br. J. Cancer 104, 1410–1417. [36] Zhang, M., Atkinson, R.L. and Rosen, J.M. (2010) Selective targeting of radiation-resistant tumor-initiating cells. Proc. Natl. Acad. Sci. USA 107, 3522–3527. [37] Frosina, G. (2009) DNA repair and resistance of gliomas to chemotherapy and radiotherapy. Mol. Cancer Res. 7, 989–999.
journal homepage: www.FEBSLetters.org
The small GTPase Rac1 is involved in the maintenance of stemness and malignancies in glioma stem-like cells Chang-Hwan Yoon a, Kyung-Hwan Hyun a, Rae-Kwon Kim a, Hyejin Lee a, Eun-Jung Lim a, Hee-Yong Chung b, Sungkwan An c, Myung-Jin Park d, Yongjoon Suh a, Min-Jung Kim a,e,⇑, Su-Jae Lee a,⇑ a
Department of Chemistry, Research Institute for Natural Sciences, Hanyang University, Seoul 133-791, Republic of Korea Department of Medicine, Hanyang University, Seoul 133-791, Republic of Korea Department of Microbiological Engineering, Kon-Kuk University, Seoul 143-701, Republic of Korea d Division of Radiation Cancer Biology, Korea Institute of Radiological and Medical Sciences, Seoul 139-706, Republic of Korea e Department of Radiation Biology, Environmental Radiation Research Group, Korea Atomic Energy Research Institute, Daejeon, Republic of Korea b c
a r t i c l e
i n f o
Article history: Received 8 February 2011 Revised 7 May 2011 Accepted 26 May 2011 Available online 23 June 2011 Edited by Quan Chen Keywords: Rac1 Glioma stem-like cells Migration and invasion Radiation resistance
a b s t r a c t A subpopulation of cancer cells with stem cell properties is responsible for tumor formation, maintenance, and malignant progression; however, the molecular mechanisms underlying the maintenance of cancer stem-like cell properties have remained unclear. Here, we show that the Rho family GTPase Rac1 is involved in the glioma stem-like cell (GSLC) maintenance and tumorigenicity in human glioma. The Rac1-Pak signaling was markedly activated in GSLCs. Knockdown of Rac1 caused reduction of expression of GSLC markers, self-renewal-related proteins and neurosphere formation. Moreover, down-regulation of Rac1 suppressed the migration, invasion, and malignant transformation in GSLCs. Furthermore, inhibition of Rac1 enhanced radiation sensitivity of GSLCs. These results indicate that the small GTPase Rac1 is involved in the maintenance of stemness and malignancies in GSLCs. Ó 2011 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved.
1. Introduction An increasing body of evidence suggests that a subpopulation of cancer cells with stem-like cell properties is responsible for tumor formation [1,2] and maintenance and malignant progression [3–5]. These rare tumor cells, termed cancer stem-like cells or cancer-initiating cells, are characterized by their strong tumorigenic properties and ability to self-renewal [6,7]. Therefore, development of novel therapeutic strategies that specifically target cancer stemlike cells may more effectively eradicate malignant tumors than conventional treatments, while reducing the risk of relapse and metastasis. It is critical to understand how the properties of cancer stem-like cells make them particularly difficult to eradicate. However, the molecular features of cancer stem-like cells that orchestrate the enrichment and maintenance of stemness and tumorigenic properties remain to be elucidated. Members of the Rho GTPase family Rac, CDC42, and Rho have been shown to be implicated in uncontrolled proliferation, Abbreviations: EGF, epidermal growth factor; bFGF, basic fibroblast growth factor; siRNA, small interfering RNA; shRNA, small hairpin RNA ⇑ Corresponding authors. Address: Department of Chemistry, Research Institute for Natural Sciences, Hanyang University, Seoul 133-791, Republic of Korea (M.-J. Kim). Fax: +82 2 2299 0762. E-mail address: [email protected] (S.-J. Lee).
survival, loss of epithelial cell properties, and abnormal cell migration [8–11], and for that reason, Rho GTPases have been suggested to control the tumor formation and malignant progression in human cancers [12–14]. Emerging evidence indicates that Rac1-dependent signal transduction is essential for malignant cell transformation [15,16]. Moreover, total expression and activity levels of Rac1 are much higher in several types of human cancers than their normal tissues [13,17,18], and a high level of Rac1 plays a fundamental role in cell survival and maintenance of malignant phenotype of high-grade glioma cells [15,19,20]. In these studies, a high expression level of Rac1 is correlated with rapid proliferation and survival of glioma cells, and inhibition of Rac1 greatly reduces glioma proliferation, and induces apoptotic cell death in patient derived glioma cells [21]. These observations have led to the concept that inappropriate expression and activation of Rac1 is involved in the development and promotion of cancer; however, the molecular mechanisms by which small GTPase Rac1 might contribute to these processes have remained obscure. The aim of this study was to find out the role of the Rho GTPase family Rac1 in the maintenance of malignant phenotypes in human glioma cells. Our results show that activation of Rac1 is critically involved in the enrichment of the glioma stem-like cell population and the maintenance of malignancies. The findings of this study
0014-5793/$36.00 Ó 2011 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.febslet.2011.05.070
2332
C.-H. Yoon et al. / FEBS Letters 585 (2011) 2331–2338
may provide a basis for improving therapeutic interventions against malignant brain cancer. 2. Materials and methods 2.1. Cell, chemical reagents, and antibodies U87MG and U373MG (glioma cancer cells) were obtained from the Korean Cell Line Bank (KCLB) and cultured in Dulbecco’s Modified Eagle Medium (DMEM) containing 10% fetal bovine serum (FBS), 100 U/ml penicillin and 100 mg/ml streptomycin in a humidified incubator at 37 °C with 5% CO2 atmosphere. For the process of isolation and identification of glioma stem cells (U87 spheres and U373 spheres), U87MG and U373MG were collected and inoculated into a serum-free medium which contained DMEM/F12 with high glucose medium, 20 ng/ml of epidermal growth factor (EGF), basic fibroblast growth factor (bFGF) and B27 (1X). The cells were placed in incubator on conditions of 37 °C, 5% CO2 atmosphere. In every 3– 4 days the medium was renewed [22,23]. Polyclonal antibody CD133 and Snail were obtained from Abcam, and Rac/cdc42 assay reagent (PAK1 PBD, agarose), polyclonal antibodies to anti-nestin, anti-Musashi1 were purchased from Millipore. The monoclonal antibody Rac1, CDC42 and Annexin V-FITC were obtained from BD Transduction Laboratories™. Polyclonal antibody Slug and Zeb1 were obtained from Santa Cruz. The polyclonal antibody Vimentin was obtained from Thermo Science. 4,6-Diamidino-2-phenylindole (DAPI), EGF and monoclonal anti-b-actin were obtained from Sigma. bFGF was purchased from R&D Systems. Anti-mouse Alexa Fluor 488, anti-rabbit Alexa Flour 488 and B27 were purchased from Invitrogen. CD133 (directly conjugated with phycoerythrin, clone 293C3) and mIgG2b-PE were purchased from Miltenyi Biotech Ltd. 2.2. Spheres culture U87MG and U373MG cells were resuspended in DMEM-F12 (Invitrogen) containing 20 ng/ml of EGF, bFGF and B27 (1X) as a stem cell–permissive medium. Spheres were collected after 5– 7 days, and protein was extracted for additional experiments or dissociated with Accutase (Innovative Cell Technologies Inc.) for expansion [24,25].
was purchased from Samchully Pharmaceutical Co. Ltd. (Seoul, Korea). A control siRNA specific for green fluorescent protein (50 -CCACTACCTGAGCACCCAG-30 ) was used as the negative control. siRNA duplexes (40 nM) were introduced into cells using a Microporator-mini (Invitrogen) by following the procedure recommended by the manufacturer. After 48 h, the medium was added, and cells were harvested for additional experiments. 2.6. Lentiviral shRNA experiments Small hairpin RNA (shRNA) constructs against human Rac1 was purchased from OriGene Technologies Inc. (Rockville, MD) (shRNA Rac1; Clone ID TRCN0000004870, non-targeting shRNA; Sigma– Aldrich SHC002). 293T cells were seeded (3 106 cells/ml) in 60-mm dish. Lentiviral particles were generated using a three-plasmid system, as described previously [27,28]. To standardize lentiviral transduction assays, viral titers were measured in a benchmark cell line, A549. For growth assays, titers corresponding to multiplicities of infection (MOIs) of 5 and 1 in A549 cells were employed. For Rac1 knockdown, cells were plated on day zero at 1 105 cells/ml in 60-mm dishes. Twenty-four hours after infection, cells were treated with 1 lg/ml puromycin for 1 days to eliminate uninfected cells. Media was replaced and after 2 more days, cells were harvested for Western blot or for additional experiments. 2.7. Rac1 activation assay Cells were lysed lysis buffer [40 mM Tris–HCl (pH 8.0), 120 mM NaCl, 0.1% Nonidet-P40] supplemented with protease inhibitors. For Rac/cdc42 assay, Cellular debris was removed by centrifugation at 13 200 rpm for 15 min at 4 °C. Protein concentrations were determined using Bio-Rad protein assay reagents, and total cell lysate (300 lg protein in 0.3 ml) was immunoprecipitated with 10 lg PAK-1 PBD agarose bead (for RAC and CDC42) at 4 °C for 3 h. After washing three times with lysis buffer, the pull-out GTP-bound RAC/CDC42 protein was eluted by boiling in SDS-sample buffer and evaluated by Western blot analysis. 2.8. Western blot analysis
Sphere-forming glioma cells were first distributed into 96-well plates at a density of 1–2 cells/well. After 12 h, individual wells were visually checked for the presence of a single cell. The clones were grown and tested for the neurospheres formation. Neurospheres with a diameter >20 lm were counted using an inverted microscope at days 1, 4, 7 and 10 [26].
Cell lysates were prepared by extracting proteins with lysis buffer [40 mM Tris–HCl (pH 8.0), 120 mM NaCl, 0.1% Nonidet-P40] supplemented with protease inhibitors. Proteins were separated by sodium dodecyl sulfate–polyacrylamide (SDS) gel electrophoresis and transferred to a nitrocellulose membrane (Amersham Biosciences, Arlington Heights, IL). The membrane was blocked with 5% non-fat dry milk in Tris-buffered saline and incubated with primary antibodies for overnight at 4 °C. Blots were developed with a peroxidase-conjugated secondary antibody and proteins visualized by enhanced chemiluminescence (ECL, Amersham) procedures, using the manufacturer’s protocol.
2.4. Flow cytometric analyses
2.9. Invasion and motility assays
Cells were dissociated and resuspended in PBS containing 0.5% bovine serum albumin. For magnetic labeling, CD133 Micro Beads were used (Miltenyi Biotech). Positive magnetic cell separation (MACS) was done using 25MS MACS columns in series. Cells were stained with CD133/2-PE (Miltenyi Biotech) or isotype control antibody (mIgG2b-PE, Miltenyi Biotech) and analyzed on a BD FACSCalibur flow cytometer.
Cells (2 104 cells/well) were suspended in 0.2 ml of serumfree DMEM or serum-free DMEM-F12 supplemented with B27, FGF and hEGF for invasion and motility assays. For invasion assay, the cells were loaded in the upper well of the Transwell chamber (8-mm pore size; Corning, Corning, NY) that was precoated with 10 mg/ml growth factor reduced (BD Matrigel™ matrix – BD Biosciences) on an upper side of the chamber with the lower well filled with 0.8 ml of serum-free DMEM or serum-free DMEMF12 supplemented with B27, FGF and hEGF. After incubation for 48 h at 37 °C, non-invaded cells on the upper surface of the filter were removed with a cotton swab, and migrated cells on the lower surface of the filter were fixed and stained with a Diff-Quick kit
2.3. Sphere colony counting
2.5. Small interfering RNA (siRNA) transfection siRNA targeting of Rac1 was performed using 21 base pairs (including a 2-deoxynucleotide overhang). siRNA target of Rac1
C.-H. Yoon et al. / FEBS Letters 585 (2011) 2331–2338
(Fisher Scientific) and photographed (magnification 200). Invasiveness was determined by counting cells in five microscopic fields per well, and the extent of invasion was expressed as an average number of cells per microscopic field. Cells were imaged with by phase-contrast microscopy (Leica Microsystems, Bannockburn, IL). For motility studies, we used invasion chambers with control inserts that contained the same type of membrane but without the Matrigel coating (one chamber per well of a 24-well plate). 2 104 cells in 0.2 ml of serum-free DMEM or serum-free DMEM-F12 supplemented with B27, FGF and hEGF were added to the apical side of each insert, and 0.8 ml of serum-free DMEM was added to the basal side of each insert. The inserts were processed as described above for the invasion assay. 2.10. Soft agar colony formation assay To examine anchorage-independent growth, a cell suspension (2 104 cells) were suspended in 0.4% agarose with serum-free DMEM-F12 supplemented with B27, FGF and hEGF and seeded in triplicate on 60-mm dishes pre-coated with 0.8% agar in complete growth medium and incubated at 37 °C, 5% CO2. After 12–30 days, colonies were photographed and counted in four randomly chosen fields and expressed as means of triplicates, representative of two independent experiments. 2.11. Tumor xenografts on nude mice After transfection with shRNA, U373 sphere forming glioma cells (4 104) were subcutaneously inoculated to the right flank of athymic Balb/c female nude mice (5 weeks of age). Tumor size was measured with a caliper (calculated volume = shortest diameter2 longest diameter/2) at 5 day intervals. This study was reviewed and approved by the Institutional Animal Care and Use Committee (IACUC) of Korea Institute of Radiological and Medical Sciences (KIRAMS). 2.12. Quantification of cell death FACS analysis using propidium iodide (PI) staining detects cell death by means of the dye entering the cells along with changes in the target cell membrane and DNA damage. For the cell-death assessment, cells were harvested by trypsinization, washed in PBS, and then incubated in propidium iodide (50 ng/ml) for 5 min at room temperature. Then, cells (10 000 per sample) were analyzed on a BD FACSCalibur flow cytometer, using Cell Quest software (BD Biosciences). 2.13. Immunofluorescent staining Immunocytochemistry of tumor spheres was performed as described [5]. Antibodies used were as follows: human anti-CD133 (1:200), -nestin (1:200) and -musashi1 (1:200) for the detection of neural stem. Stainings were visualized using anti-mouse Alexa Fluor 488 and anti-rabbit Alexa Flour 488 (Molecular Probes). Nuclei were counterstained using 4,6-diamidino-2-phenylindole (DAPI; Sigma). Stained cells were visualized with a fluorescencemicroscope (Olympus IX71). 2.14. Statistical analysis All the data in the study were performed in triplicate. The results are described as mean ± S.D. Statistical analysis was performed by one-way analysis of variance (ANOVA) and comparisons among groups were performed by independent sample t-test. Differences were considered significant at values of ⁄P < 0.001.
2333
3. Results and discussion 3.1. Rac1 is involved in the maintenance of glioma stem-like cells Rho family small GTPase Rac1 has been implicated in the tumor formation and malignant progression in human glioma [1–5]. To examine whether Rac1 is involved in the maintenance of glioma stem-like cells, we first examined the activation status of Rac1 in glioma stem-like cells using Rac1-PAK binding assay. As shown in Fig. 1A, binding activity of Rac1 to PAK was markedly increased in U87 and U373 glioma cells grown under sphereforming conditions compared to matched glioma cells grown as a monolayer. There was no change in Rac1 at the total protein level. To further determine whether the activation of Rac1 is required for the maintenance of the glioma stem-like cell population. To further study the involvement of Rac1 in the maintenance of the glioma stem-like cell population, we down-regulated Rac1 in sphere forming U87 and U373 glioma cells by treatment with siRNA. Treatment with siRNA or shRNA targeted to Rac1 specifically down-regulated Rac1, while the expression levels of b-actin and other GTpase, CDC42 were not altered (Fig. S1A). Importantly, treatment with siRNA targeting of Rac1 resulted in the increase of binding activity of Rac1 to PAK and a marked decrease in the number of spheres (Fig. 1B). Moreover, the CD133+ cell population was clearly suppressed by siRNA-mediated knockdown of Rac1 (Fig. 1C). Treatment with siRNA targeted to Rac1 also attenuated expression of glioma stem-like cell marker proteins, CD133, Nestin and Musashi-1 (Fig. 1D). In addition, a fluorescence microscopic analysis of cells immunostained with antibodies specific for the marker proteins clearly revealed that siRNA targeting of Rac1 effectively suppressed expression level of CD133, Nestin, and Musashi-1 (Fig. 1E). These results specifically implicate Rac1 in the maintenance of glioma stem-like cell population.
3.2. Activation of Rac1 is critical for the proliferation and self-renewal in glioma stem-like cells Enhanced self-renewal ability is one of the key characteristics of cancer stem-like cells. To determine whether Rac1 is critical for the proliferation and self-renewal in glioma stem-like cells, we inhibited Rac1 expression with siRNA and examined changes in self-renewal potential using an in vitro self-renewal assay, and also examined changes in the expression levels of the known self-renewal-related proteins Sox2, Notch1/2, and b-catenin. As shown in Fig. 2A, treatment with siRNA targeted to Rac1 clearly inhibited sphere-forming ability of glioma stem-like cells at single cell levels (self-renewal potential). In contrast, down-regulation of Rac1 in monolayer-cultured glioma cells did not suppress cell growth (Fig. S1B), suggesting that Rac1 is required only for the proliferation of glioma stem-like cells, but not monolayer-cultured cells. Consistent with these results, the expression of self-renewal-related proteins Sox2, Notch2 and b-catenin was markedly higher in sphere forming glioma cells, compared to monolayer-cultured glioma cells (Fig. 2B), again confirming that glioma stem-like cells are highly enriched in sphere forming glioma cells more than monolayer-cultured cells. Importantly, treatment with siRNA targeted to Rac1 led to a decrease of Sox2, Notch2, and b-catenin at the protein level in sphere-forming glioma stem-like cells, but did not affect the level of Notch1 (Fig. 2B). Since treatment with siRNA specifically targeted to Rac1 (Fig. S1A) and siRNA-mediated down-regulation of Rac1 was sustained at least for 10 days (Fig. S1C), Rac1 would be directly involved in down-regulation of self-renewal-related proteins, thereby suppressing sphere
2334
C.-H. Yoon et al. / FEBS Letters 585 (2011) 2331–2338
Fig. 1. (A) U87 and U373 monolayers or spheres lysates were immunoprecipitated with 10 lg PAK-1 PBD agarose bead. Total proteins were subjected to immunoblot analysis with anti-Rac1 and b-actin antibodies. b-Actin was used as the loading control. (B) U87 and U373 spheres were transfected with si-RNA of Rac1 or control. (Left) After 48 h, total cell lysates was immunoprecipitated with 10 lg PAK-1 PBD agarose bead. And proteins were subjected to immunoblot analysis with anti-Rac1 and b-actin antibodies. bActin was used as the loading control. (Right) Graph shows the number of spheres per visual field of microscopy. Error bars represent mean ± S.D. of five different fields. An asterisk () indicates P < 0.001 with one-way ANOVA comparison of the control groups to the corresponding Rac1 inhibition groups. (C) U87 and U373 spheres were transfected with either si-RNA of Rac1 or control. After 48 h, cells were analyzed for the expression of glioma-stem cell marker CD133 by flow cytometry. Error bars represent mean ± S.D. of triplicate samples. An asterisk () indicates P < 0.001 with one-way ANOVA comparison of the control groups to the corresponding Rac1 inhibition groups. (D) After 48 h, total cell lysates were subjected to immunoblot analysis with anti-CD133, -Nestin, or -Musashi1 antibodies. b-Actin was used as the loading control. (E) U87 and U373 spheres were transfected with either si-RNA of Rac1 or control. After 48 h, cells were stained with FITC-conjugated with anti-CD133, -Nestin, or -Musashi1 antibodies and imaged by fluorescence microscopy (Olympus IX71). Photomicrographs were taken at magnification 400.
C.-H. Yoon et al. / FEBS Letters 585 (2011) 2331–2338
2335
Fig. 2. (A) U87 and U373 sphere were transfected with either control-siRNA or si-Rac1. Cells were then distributed into 96-well plates at a density of 1 cell/well. The ability of a single clone to form new spheres was tested by microscopy after 10 days. (Left) Images represent changes in spheres size in the presence of control-siRNA or si-Rac1 (magnification 200). (Right) Number of clones forming spheres with a diameter >20 lm was counted using inverted in the presence of control-siRNA or si-Rac1. Error bars represent mean ± S.D. of triplicate samples. An asterisk () indicate respectively P < 0.001 with one-way ANOVA comparison of the control groups to the corresponding Rac1 inhibition groups. (B) U87 and U373 sphere were transfected with either control-siRNA or si-Rac1. After 48 h, total cell lysates were subjected to immunoblot analysis with anti-Notch1/2, -Sox2, -b-catenin, and Rac1 antibodies. b-Actin was used as the loading control.
formation. Our findings are supported by previous studies that down-regulation of Notch, Sox2 and b-catenin led to a decrease of sphere formation [33–35]. To confirm the role of Rac1 in the stemness of glioma stem-like cell population, we also treated patient-derived glioma stem-like cells (X01GB) with siRNA targeting of Rac1, and examined changes in the CD133+ cell population, sphere formation, and self-renewal potential. As shown in Fig. S2A, siRNA-mediated Rac1 knockdown reduced the CD133+ cell population and decreased the number of spheres. The self-renewal potential in patient-derived glioma stem-like cells was also inhibited by down-regulation of Rac1 (Fig. S2B). Taken together, these results suggest that Rac1 plays a critical role in the maintenance of stemness in glioma stem-like cells. 3.3. Rac1 is involved in the maintenance of the malignant properties of glioma stem-like cells We next examined whether Rac1 is required for maintenance of the malignant properties of sphere forming glioma cells. To this end, we treated sphere-forming glioma cells with shRNA and examined anchorage-independent colony formation in soft agar. Contrary to normal cells that do not proliferate without anchorage, tumorigenic cancer cells can grow in a condition without anchorage such as semisolid medium. Thus, colony forming assay in soft agar is widely adapted to measure tumorigenic capacity in vitro, as a good predictor of tumor forming ability in vivo. As shown in Fig. 3A, shRNA-mediated knockdown of Rac1 effectively attenuated
colony formation of sphere forming glioma cells in soft agar. To further test the involvement of Rac1 in the maintenance of malignant potential in sphere forming cells, we inhibited Rac1 expression with siRNA and analyzed changes in cellular invasion and migration. Notably, invasive behavior and cell migration potential of glioma stem-like cells were markedly suppressed by down-regulation of Rac1 (Fig. 3B and C). These results indicate that Rac1 is involved in the maintenance of malignant potential in cancer stem-like cell population. A mesenchymal phenotype is the hallmark of tumor aggressiveness in human malignant glioma. To further test whether Rac1 is critical for maintenance of mesenchymal phenotype in glioma stem-like cells, we knocked down Rac1 with siRNA and analyzed changes in protein expression levels of mesenchymal cell markers N-cadherin and Vimentin, and the transcription repressors Zeb1, Snail, and Slug. As expected, expression levels of mesenchymal cell markers and the transcription repressors were much higher in sphere forming glioma stem-like cells than matched glioma cells grown as a monolayer (data not shown). Importantly, knockdown of Rac1 with siRNA clearly suppressed N-cadherin and Vimentin in glioma stem-like cells (Fig. 3D). Moreover, treatment with siRNA targeted to Rac1 effectively inhibited expression of Zeb1 and Slug proteins, though the protein level of Snail was not altered. These results indicate that Rac1 plays a critical role for the maintenance of mesenchymal phenotype in glioma stem-like cells. Since down-regulation of Rac1 attenuated colony formation in soft agar and suppressed cell migration and invasion in vitro, we next examined whether Rac1 is required for tumor formation in
2336
C.-H. Yoon et al. / FEBS Letters 585 (2011) 2331–2338
Fig. 3. (A) U87 and U373 sphere were infected with either scramble-shRNA or sh-Rac1 virus, and after 24 h with 1 lg/ml puromycin, infected cells (2 104 cells/ml) were seeded in soft agar plate. The colonies were calculated relative to 1000 cells during 4 weeks of plating. (Left) The colony was determined by counting cells in randomly selected four microscopic fields per plate. Photomicrographs were taken at magnification 200. (Right) Graph shows the number of spheres per visual field of microscopy in the presence of scramble-shRNA or sh-Rac1 virus. Error bars represent mean ± S.D. of four different fields. An asterisk () indicates P < 0.001 with one-way ANOVA comparison of the control groups to the corresponding Rac1 inhibition groups. (B) U87 and U373 monolayers or spheres were transfected with either control-siRNA or si-Rac1, and then the cells (2 104 cells/well) were seeded in each Matrigel chamber and cultured in serum-free DMEM-F12 supplemented with B27, FGF and hEGF for 48 h. (Left) Invaded cells were fixed and stained with a Diff-Quick kit and photographed (magnification 100). (Right) Invasiveness was determined by counting cells in randomly selected five microscopic fields per well. Bars represent mean ± S.D. of triplicate samples. An asterisk () indicates P < 0.001 with one-way ANOVA comparison of the control groups to the corresponding Rac1 inhibition groups. (C) U87 and U373 monolayers or spheres were transfected with either control-siRNA or si-Rac1, and then the cells (2 104 cells/well) were seeded in each chamber (without the Matrigel) and cultured in serum-free DMEM-F12 supplemented with B27, FGF and hEGF for 48 h. (Left) Migrated cells were fixed and stained with a Diff-Quick kit and photographed (magnification 100). (Right) Cell migration was determined by counting cells in randomly selected five microscopic fields per well. Error bars represent mean ± S.D. of triplicate samples. An asterisk () indicates P < 0.001 with one-way ANOVA comparison of the control groups to the corresponding Rac1 inhibition groups. (D) U87 and U373 spheres were transfected with either control-siRNA or si-Rac1, and after 48 h, total cell lysates were subjected to immunoblot analysis with anti-N-cadherin, -Vimentin, -Zeb1, -Snail, or -Slug antibodies. b-Actin was used as the loading control.
C.-H. Yoon et al. / FEBS Letters 585 (2011) 2331–2338
2337
Fig. 4. U373 sphere forming glioma cells were transfected with either control scrambled shRNA or sh-Rac1. Transfected cells were then subcutaneously inoculated to the right flank of athymic Balb/c female nude mice (5 weeks of age, n = 4 per group). Tumor size was measured with a caliper (calculated volume = shortest diameter2 longest diameter/2) at 5 day intervals. Representative photos shows differential volume of tumor at 35 days. Error bars represent mean ± S.D. ⁄P < 0.001.
vivo. To this end, we treated sphere-derived U373 cells with siRNA targeted to Rac1 and then transplanted to athymic Balb/c female nude mice subcutaneously. Notably, tumor formation was dramatically attenuated in mice transplanted with U373 sphere-derived glioma cells treated with shRNA Rac1, compared to scrambled shRNA (Fig. 4). Taken together, these results have implications that Rac1 is a critical regulator in vitro and in vivo for maintenance of the malignant properties in glioma stem-like cells. 3.4. Rac1 is important for the maintenance of resistance to anticancer treatments in glioma stem-like cells Cancer stem-like cells have been implicated in the resistance to chemotherapy and radiotherapy [36,37]. As expected, sphereforming glioma cells showed greater resistance to both ionizing radiation and anticancer drug, etoposide than monolayer cells (data not shown), confirming that cancer stem-like cells are enriched in sphere-cultured glioma cells. To examine whether activation of Rac1 is involved in the maintenance of resistance to anticancer treatments in glioma stem-like cells, we inhibited Rac1 expression with siRNA, and then examined changes in sensitivity of glioma stem-like cells to ionizing radiation and etoposide in sphere forming glioma stem-like cells. Though single treatment with ionizing radiation or etoposide or siRNA targeted to Rac1 did not cause significant cell death, pretreatment with siRNA targeted to Rac1 sensitized sphere forming glioma cells to ionizing radiation and etoposide, inducing significant cell death (Fig. 5). Concurrently with cell death, we also observed the increase of Annexin V positive cell population by ionizing radiation or etoposide treatment only in pretreated sphere forming glioma cells with siRNA targeted to Rac1 (Fig. S3), indicating that cell death is due to apoptosis. Taken together, these results suggest that these results indicate that Rac1 is important for the maintenance of resistance to anticancer treatments in glioma stem-like cells. The Rho family small GTPase Rac1 has been implicated in cell survival and malignant transformation in human cancers including glioma [15,19,20,29–31]. Similar observations have led to the concept that inappropriate activation of Rac1 is involved in the development and promotion of glioma [8,9,12,32]. In this study, we further found that the Rho family GTPase Rac1 is involved in the stemness and tumorigenicity in glioma stem-like cells. The Rac1 signaling was markedly activated in sphere forming glioma stem-like cells. Importantly, inhibition of Rac1 signaling with siRNA reduced expression of glioma stem-like cell markers and self-renewal-related proteins as well as sphere formation. Moreover, siRNA-targeting of Rac1 effectively suppressed the migration, invasion, and malignant transformation in sphere forming glioma stem-like cells.
Fig. 5. U87 and U373 spheres were transfected with either control-siRNA or siRac1. Cells were then treated with ionizing radiation (10 Gy) and etoposide (10 lM). After 48 h, cell death was assessed by flow cytometry as described in Section 2. Error bars represent mean ± S.D. of triplicate samples.
Furthermore, siRNA-mediated knockdown of Rac1 effectively enhanced radiation sensitivity of glioma stem-like cells. These results indicate that the small GTPase Rac1 is involved in the maintenance of stemness and malignancies in the glioma stem-like cells. These findings may provide pivotal insights in the context of therapeutic interventions to target glioma stem-like cell population. Acknowledgements This work was supported by the Korea Research Foundation (KRF) and Ministry of Education, Science and Technology (MEST), Korean government, through its National Nuclear Technology Program (2008-2003935) and Medical Research Center Program (2010-0029504). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.febslet.2011.05.070. References [1] Schatton, T. and Frank, M.H. (2008) Cancer stem cells and human malignant melanoma. Pigment Cell Melanoma Res. 21, 39–55. [2] Singh, S.K., Clarke, I.D., Hide, T. and Dirks, P.B. (2004) Cancer stem cells in nervous system tumors. Oncogene 23, 7267–7273. [3] Bao, S. et al. (2006) Stem cell-like glioma cells promote tumor angiogenesis through vascular endothelial growth factor. Cancer Res. 66, 7843–7848. [4] Zhou, J. et al. (2007) Activation of the PTEN/mTOR/STAT3 pathway in breast cancer stem-like cells is required for viability and maintenance. Proc. Natl. Acad. Sci. USA 104, 16158–16163. [5] Inagaki, A., Soeda, A., Oka, N., Kitajima, H., Nakagawa, J., Motohashi, T., Kunisada, T. and Iwama, T. (2007) Long-term maintenance of brain tumor stem cell properties under at non-adherent and adherent culture conditions. Biochem. Biophys. Res. Commun. 361, 586–592. [6] He, S., Nakada, D. and Morrison, S.J. (2009) Mechanisms of stem cell selfrenewal. Annu. Rev. Cell Dev. Biol. 25, 377–406.
2338
C.-H. Yoon et al. / FEBS Letters 585 (2011) 2331–2338
[7] Sullivan, J.P., Minna, J.D. and Shay, J.W. (2010) Evidence for self-renewing lung cancer stem cells and their implications in tumor initiation, progression, and targeted therapy. Cancer Metastasis Rev. 29, 61–72. [8] Murai, T., Miyazaki, Y., Nishinakamura, H., Sugahara, K.N., Miyauchi, T., Sako, Y., Yanagida, T. and Miyasaka, M. (2004) Engagement of CD44 promotes Rac activation and CD44 cleavage during tumor cell migration. J. Biol. Chem. 279, 4541–4550. [9] Parri, M. and Chiarugi, P. (2010) Rac and Rho GTPases in cancer cell motility control. Cell Commun. Signal. 8, 23. [10] Castillo-Lluva, S., Tatham, M.H., Jones, R.C., Jaffray, E.G., Edmondson, R.D., Hay, R.T. and Malliri, A. (2010) SUMOylation of the GTPase Rac1 is required for optimal cell migration. Nat. Cell Biol. 12 (11), 1078–1085. [11] Karlsson, R., Pedersen, E.D., Wang, Z. and Brakebusch, C. (2009) Rho GTPase function in tumorigenesis. Biochim. Biophys. Acta 1796, 91–98. [12] Espina, C. et al. (2008) A critical role for Rac1 in tumor progression of human colorectal adenocarcinoma cells. Am. J. Pathol. 172, 156–166. [13] Hwang, S.L., Hong, Y.R., Sy, W.D., Lieu, A.S., Lin, C.L., Lee, K.S. and Howng, S.L. (2004) Rac1 gene mutations in human brain tumours. Eur. J. Surg. Oncol. 30, 68–72. [14] Gomez del Pulgar, T., Benitah, S.A., Valeron, P.F., Espina, C. and Lacal, J.C. (2005) Rho GTPase expression in tumourigenesis: evidence for a significant link. BioEssays 27, 602–613. [15] Nakada, M., Drake, K.L., Nakada, S., Niska, J.A. and Berens, M.E. (2006) EphrinB3 ligand promotes glioma invasion through activation of Rac1. Cancer Res. 66, 8492–8500. [16] Johnson, E., Seachrist, D.D., DeLeon-Rodriguez, C.M., Lozada, K.L., Miedler, J., Abdul-Karim, F.W. and Keri, R.A. (2010) HER2/ErbB2-induced breast cancer cell migration and invasion require p120 catenin activation of Rac1 and Cdc42. J. Biol. Chem. 285, 29491–29501. [17] Schnelzer, A. et al. (2000) Rac1 in human breast cancer: overexpression, mutation analysis, and characterization of a new isoform, Rac1b. Oncogene 19, 3013–3020. [18] Wang, Z. et al. (2010) Rac1 is crucial for Ras-dependent skin tumor formation by controlling Pak1-Mek-Erk hyperactivation and hyperproliferation in vivo. Oncogene 29, 3362–3373. [19] Chuang, Y.Y., Tran, N.L., Rusk, N., Nakada, M., Berens, M.E. and Symons, M. (2004) Role of synaptojanin 2 in glioma cell migration and invasion. Cancer Res. 64, 8271–8275. [20] Chan, A.Y., Coniglio, S.J., Chuang, Y.Y., Michaelson, D., Knaus, U.G., Philips, M.R. and Symons, M. (2005) Roles of the Rac1 and Rac3 GTPases in human tumor cell invasion. Oncogene 24, 7821–7829. [21] Senger, D.L. et al. (2002) Suppression of Rac activity induces apoptosis of human glioma cells but not normal human astrocytes. Cancer Res. 62, 2131– 2140.
[22] Jin, F. et al. (2008) Comparison between cells and cancer stem-like cells isolated from glioblastoma and astrocytoma on expression of anti-apoptotic and multidrug resistance-associated protein genes. Neuroscience 154, 541– 550. [23] Jin, F. et al. (2008) Paradoxical expression of anti-apoptotic and MRP genes on cancer stem-like cell isolated from TJ905 glioblastoma multiforme cell line. Cancer Invest. 26, 338–343. [24] Singh, S.K., Clarke, I.D., Terasaki, M., Bonn, V.E., Hawkins, C., Squire, J. and Dirks, P.B. (2003) Identification of a cancer stem cell in human brain tumors. Cancer Res. 63, 5821–5828. [25] Suslov, O.N., Kukekov, V.G., Ignatova, T.N. and Steindler, D.A. (2002) Neural stem cell heterogeneity demonstrated by molecular phenotyping of clonal neurospheres. Proc. Natl. Acad. Sci. USA 99, 14506–14511. [26] Lu, F. and Wong, C.S. (2005) A clonogenic survival assay of neural stem cells in rat spinal cord after exposure to ionizing radiation. Radiat. Res. 163, 63–71. [27] Singh, A., Greninger, P., Rhodes, D., Koopman, L., Violette, S., Bardeesy, N. and Settleman, J. (2009) A gene expression signature associated with ‘‘K-Ras addiction’’ reveals regulators of EMT and tumor cell survival. Cancer Cell 15, 489–500. [28] Moffat, J. et al. (2006) A lentiviral RNAi library for human and mouse genes applied to an arrayed viral high-content screen. Cell 124, 1283–1298. [29] Kamai, T., Yamanishi, T., Shirataki, H., Takagi, K., Asami, H., Ito, Y. and Yoshida, K. (2004) Overexpression of RhoA, Rac1, and Cdc42 GTPases is associated with progression in testicular cancer. Clin Cancer Res 10, 4799–4805. [30] Vega, F.M. and Ridley, A.J. (2008) Rho GTPases in cancer cell biology. FEBS Lett. 582, 2093–2101. [31] Sahai, E. and Marshall, C.J. (2002) RHO-GTPases and cancer. Nat. Rev. Cancer 2, 133–142. [32] Baugher, P.J., Krishnamoorthy, L., Price, J.E. and Dharmawardhane, S.F. (2005) Rac1 and Rac3 isoform activation is involved in the invasive and metastatic phenotype of human breast cancer cells. Breast Cancer Res. 7, R965–R974. [33] Grudzien, P. et al. (2010) Inhibition of Notch signaling reduces the stem-like population of breast cancer cells and prevents mammosphere formation. Anticancer Res. 30, 3853–3867. [34] Malanchi, I. et al. (2008) Cutaneous cancer stem cell maintenance is dependent on beta-catenin signalling. Nature 452, 650–653. [35] Xiang, R. et al. (2011) Downregulation of transcription factor SOX2 in cancer stem cells suppresses growth and metastasis of lung cancer. Br. J. Cancer 104, 1410–1417. [36] Zhang, M., Atkinson, R.L. and Rosen, J.M. (2010) Selective targeting of radiation-resistant tumor-initiating cells. Proc. Natl. Acad. Sci. USA 107, 3522–3527. [37] Frosina, G. (2009) DNA repair and resistance of gliomas to chemotherapy and radiotherapy. Mol. Cancer Res. 7, 989–999.