Polo-Like Kinase 1 Inhibition Causes Decreased Proliferation by Cell Cycle Arrest, Leading to Cell Death in Glioblastoma

ORIGINAL ARTICLE Polo-like kinase 1 inhibition causes decreased proliferation by cell cycle arrest, leading to cell death in glioblastoma

JA Pezuk1, MS Brassesco2, AG Morales1, JC de Oliveira1, RG de Paula Queiroz2, HR Machado3, CG Carlotti Jr3, L Neder4, CA Scrideli2 and LG Tone2

Glioblastoma (GBM) is one of the most aggressive central nervous system tumors with a patient’s median survival of o1 year. Polo-like kinases (PLKs) are a family of serine/threonine kinases that have key roles in cell cycle control and DNA-damage response. We evaluated PLK1, 2, 3 and 4 gene expression in 8 GBM cell lines and 17 tumor samples, and analyzed the effect of the PLK1 inhibition on SF188 and T98G GBM cell lines and 13 primary cultures. Our data showed PLK1 overexpression and a variable altered expression of PLK2, 3 and 4 genes in GBM tumor samples and cell lines. Treatments with nanomolar concentrations of BI 2536, BI 6727, GW843682X or GSK461364 caused a significant decrease in GBM cells proliferation. Colony formation was also found to be inhibited (Po0.05), whereas apoptosis rate and mitotic index were significantly increased (Po0.05) after PLK1 inhibition in both GBM cell lines. Cell cycle analysis showed an arrest at G2 (Po0.05) and cell invasion was also decreased after PLK1 inhibition. Furthermore, simultaneous combinations of BI 2536 and temozolomide produced synergistic effects for both the cell lines after 48 h of treatment. Our findings suggest that PLK1 might be a promising target for the treatment of GBMs.

Cancer Gene Therapy (2013) 20, 499–506; doi:10.1038/cgt.2013.46; published online 26 July 2013 Keywords: glioblastoma; Polo-like kinases; cell cycle; PLK1 inhibition.

INTRODUCTION Other groups of ATP competitors PLK1 inhibitors are thiophene 11 Glioblastoma (GBM) is one of the most aggressive tumors, benzimidazole derivates, like GSK461364A, and GSK461364A, an 10 representing the main primary brain neoplasia in adults.1 In imidazotriazine; both of them are highly efficient PLK1 inhibitors children GBM is less frequent, representing 0.6–7.9% of all that rapidly forms a reversible complex with PLK1. On this basis, gliomas.2 Standard treatments consist of surgery, radiotherapy the aim of our study was to evaluate the differential expression of and chemotherapy with temozolomide (TMZ).3 Nonetheless, this PLKs in GBM samples and to test the in vitro effects of PLK1 tumor is still one of the hardest to treat, with high recurrence inhibition on GBM cell lines and primary cultures. rates favored by the capacity of tumor cells to invade/infiltrate adjacent tissue, high proliferation, and cellular and genetic heterogeneity.4 Many therapy modalities have been tested in MATERIALS AND METHODS order to improve patient outcome, although their efficacy is still Tumor samples below expectations. Thus, the search for novel therapeutic targets Samples were collected under informed consent from patients submitted is still urgent. to surgical resection at the University Hospital of Ribeiraõ Preto. In all, 17 Polo-like kinases (PLKs) belong to a family of five highly GBM samples (13 adults (76.47%) and 11 males (64.70%)) were used to conserved serine/threonine kinases (PLK1, PLK2, PLK3, PLK4 and evaluate gene expression and 13 tumor samples were used to obtain PLK5) that have key roles in cell cycle progression.5 PLK1 is the primary GBM cultures to be used in PLK1 inhibition tests. Five non- most studied member of this family and is crucial to cell division, neoplastic white matter samples were obtained from epileptic patients 6 undergoing temporal lobectomy. The study was approved by the Ethics from chromosome maturation in G2 to cytokinesis. In normal Committee of the institution (Proc. 7328/2009). tissues PLK1 is found only in proliferating cells. Increased PLK1 gene expression has been described in different neoplasias, a change correlated with prognosis in some cancers.7 Cell culture The inhibition of PLK1 has shown to cause cell cycle arrest The adult human GBM cell lines U251, U138, U87, T98G, U343 and MO59K and to increase apoptosis in several models.8 The inhibition of were purchased from ATCC (Manassas, VA, USA). The LN319 cell line PLK1 has a cytotoxic effect on cells, and because of this several was provided by Dr Frank Furnari, Ludwig Institute for Cancer Research, inhibitors have been developed.9 Among them, the majority of CA, USA, and the pediatric SF188 cell line was provided by Drs Nada PLK1 inhibitors are ATP competitors that act on the ATP-binding Jabado and Damien Faury, McGill University, Canada. The T98G and SF188 cell lines were chosen to be used in functional tests, due to the fact that pocket of the kinase.10 Among them, there is the BI 2536, 11 they have PLK1 overexpression in similar pattern to GBM samples and a dihydropteridinone derivate, and the BI 6727, which is because they are different from each other, representing the heterogeneity 12 the second-in class dihydropteridinone derivate, that have present in GBMs tumors. T98G is resistant to treatment with TMZ and shown to efficiently inhibit specifically the activity of PLK1. SF188 is a pediatric cell line.

1Department of Genetics, Faculty of Medicine of Ribeiraõ Preto, University of Saõ Paulo; 2Department of Pediatrics, Faculty of Medicine of Ribeiraõ Preto, University of Saõ Paulo; 3Department of Surgery and Anatomy, Faculty of Medicine of Ribeiraõ Preto, University of Saõ Paulo and 4Department of Pathology, Faculty of Medicine of Ribeiraõ Preto, University of Saõ Paulo. Correspondence: Professor JA Pezuk, Laborato´rio de Pediatria—Bloco G, Hospital das Clıńicas da Faculdade de Medicina de Ribeiraõ Preto, University of Saõ Paulo, Avenue Bandeirantes, 3900 Bairro Monte Alegre, Ribeiraõ Preto, Sao Paulo CEP 14048-900, Brazil. E-mail: [email protected] Received 17 May 2013; accepted 19 May 2013; published online 26 July 2013 Polo-like kinases in glioblastoma JA Pezuk et al 500 Cells were cultured in HAM-F10 (Life Technologies, Carlsbad, CA, USA) different concentrations of BI 2536 or BI 2536 combined with TMZ supplemented with 10% fetal bovine serum, penicillin (100 U ml À 1) and (250 mmol l À 1) and cultured for an additional 48 h. Briefly, the treated cells À 1 streptomycin (100 mgml )at371C in a humidified 5% CO2 incubator. were trypsinized, centrifuged and mixed with 2 ml of the dye solution, 25 ml Thirteen primary cultures were obtained from fresh GBM samples. propidium iodide (5 mgmlÀ 1), fluorescein diacetate in dimethyl sulfoxide Briefly, after aseptic collection, samples were minced and enzymatically (15 mglÀ 1), Hoechst 33342 in water (2 mgmlÀ 1), and incubated at 37 1C for disaggregated for 1 h in 0.5% collagenase type IV (Life Technologies). Cells 5 min. Samples were then analyzed by fluorescence microscopy with a were then centrifuged; the collagenase was removed and replaced with triple filter. Five hundred nuclei were analyzed per treatment and cells medium. Cultures were then kept at 37 1C in a humidified 5% CO2 were scored and categorized according to differential staining. incubator. Only cells in the third passage were used. Cytogenetic analysis and chromosome preparations Drug and treatments A total of 2 Â 105 cells were seeded in 25-cm2 tissue culture flasks BI 2536, BI 6727, GW843682X and GSK461364 were purchased from Axon containing 5 ml of culture medium. After 24 h, cells were treated with BI Medchem (Groningen, the Netherlands). Stock solutions were prepared 2536 or BI 2536 combined with TMZ at different concentrations and then in dimethyl sulfoxide according to the manufacturer’s instructions and incubated for 24, 48, 72 and 96 h until harvesting. After each period, cells stored at À 80 1C. For experiments, cells were treated with nanomolar were trypsinized, hypotonically treated (0.075M KCl), and fixed with concentrations of BI 2536 (5, 10, 50, 100 and 150 nmol l À 1) for 24, 48, 72 methanol:glacial acetic acid (3:1). Cells were then dropped onto glass 13 and 96 h as described in the literature. Vehicle alone was used as control. slides, allowed to air dry and stained with Giemsa. The mitotic index To confirm the effect of PLK1 inhibition, the other compounds were tested represents the percentage of cells in mitosis at a specific culture condition on both the cell lines in the proliferation assay, using 50, 100 and (1000 cells were scored). 150 nmol l À 1 of BI 6727; 300, 600 and 1200 nmol l À 1 of GW843682X and 75, 150 and 300 nmol l À 1 of GSK461364. TMZ was extracted from the commercial capsule TEMODAL (Schering- Cell cycle analysis. Plough, Rio de Janeiro, Brazil) and diluted in water, according to the For cell cycle analysis, 2 Â 105 cells were seeded on 25-cm2 flasks and treated manufacturer’s instructions, considering 85% of the active principle. with different BI 2536 concentrations for 24, 48, 72 and 96 h. Cells were then Micromolar concentrations were used in all experiments (50, 100 and trypsinized and fixed in 70% ethanol, stained with propidium iodide, and 250 mmol l À 1). analyzed with a Guava Personal Cell Analysis system (Guava Technologies, Each experiment was performed in triplicate and repeated in three sets Hayward, CA, USA) according to the protocol provided by the manufacturer. of tests. Percentages of cells in G0/G1, S, or G2/M phase were determined and processed using the GUAVA Cytosoft software, version 4.2.1. Quantitative real-time RT-PCR Total RNA was isolated using Trizol Reagent (Life Technologies, Carlsbad, Invasion assay CA, USA) and reverse-transcribed using the High Capacity kit (Applied The Matrigel-coated invasion assay (BD Bioscience, Bedford, MA, USA) allows Biosystems, Foster City, CA, USA). Quantitative real-time RT-PCR the evaluation of the capacity of the cell to degrade the extracellular was performed in triplicate in 10 ml reactions using inventoried TaqMan component, once the Matrigel contains laminin, proteolgycans and collagens. (Life Technologies) probes for PLK1, 2, 3 and 4 (Hs00153444_m1, In this study, 5 Â 105 cells were treated with different concentrations of BI Hs00198320_m1, Hs00177725_m1 and Hs00179514_m, respectively, 2536 (10, 50, 100 and 150 nmol l À 1) and transferred to the top of Matrigel- Applied Biosystems) in the ABI Prism 7500 Sequence Detector (Applied coated invasion chambers (24-well insert, 8-mm pore size; Becton Dickinson, Biosystems). Hypoxanthine guanine phosphoribosyl transferase NJ, USA) according to the manufacturer’s protocol. After 24-h incubation, non- (4310890E0) and TATA-binding protein (4310891E) were used as endogen- invading cells were removed from the upper surface of the membrane by ous controls.14 A pool of four white matter samples was used as a calibrator. scrubbing with moistened swabs. The invasive cells attached to the lower Relative quantification was performed by the 2 À DDCt method.15 surface of the membrane insert were fixed in 100% methanol for 10 min and stained with Giemsa (Sigma-Aldrich, Sao Paulo, Brazil). Membranes were then Measurement of cell growth by the XTT cell proliferation assay removed from the insert housing with scalpel blade, placed on a microscope 3 slide, mounted with Entellan (BD Bioscience) and coverslipped. Invading A total of 2.2 Â 10 cells per well were seeded in 96-well plates and allowed cells were photographed under the microscope at Â 100 magnification and to adhere overnight. Next, BI 2536 was added at different concentrations counted with the CytolabView software (Applied Spectral Imaging (ASI), and incubated for 24, 48, 72 and 96 h. After each period, the culture Migdal Ha’Emek, Israel). medium was replaced with medium containing 10 ml of XTT dye (3 mg ml À 1) (XTT II; Roche Molecular Biochemicals, Indianapolis, IN, USA) in each well. The plates were incubated for 2 h at 37 1C and results mea- Statistical analysis sured at 450 nm using an iMark microplate reader (Bio-Rad Laboratories, Statistical analyses were performed using the SigmaStat software 3.5 (Jandel Philadelphia, PA, USA). Cell growth was also monitored at selected intervals Scientific Company, San Rafael, CA, USA). One-way repeated measures by Trypan blue exclusion. For drug combination analysis, the cell lines were analysis of variance followed by the Holm–Sidak pairwise multiple treated in three different schedules. For simultaneous treatment, cells were comparison was used. All tests were carried out for a ¼ 0.05. Effective concomitantly exposed to the different concentrations of BI 2536 and concentrations (IC50) were analyzed using the CalcuSyn software 2.0 (Biosoft, 250 mmol/l of TMZ. For sequential treatments, cells were either pretreated Ferguson, MO, USA). This program provides a measure of the combined for 24 h with the different concentrations of BI 2536 and exposed to drug interactions by the generation of a combination index (CI) value. The CI TMZ, or pretreated with TMZ and exposed to BI 2536 after 24 h. At both is based on the multiple drug-effect equation of Chou and Talalay17 and sequential conditions, cultures were evaluated 48 h after adding the defines the drug interactions as synergistic CI value o1, additive for CI value second drug. ¼ 1 and antagonistic for CI value 41. Calcusyn software was also used to calculate the dose reduction index of drug combinations that estimates the Colony formation assay extent to which the dose of one or more agents in the combination can be reduced to achieve effect levels that are comparable with those achieved Clonogenic assays were performed according to Franken et al.16 Single-cell with single agents. Figures for PLKs gene expression were obtained using suspensions of 200 cells were seeded and treated for 24 h with BI 2536 at À 1 the GraphPad Prism software, version 5.0.0 (GraphPad, La Jolla, CA, USA). 5, 10 and 50 nmol l concentrations combined or not with 50 mM TMZ for SF188 or 100 mmol l À 1 TMZ for T98G. Next, culture medium was replaced with drug-free medium. Cell cultures were then incubated for 7 days and the colonies were fixed and stained with Giemsa. Only colonies with 450 RESULTS cells were counted. Gene expression. All cell lines and tumor samples showed increased levels of PLK1 Detection of cell death expression when compared to the white matter pool (Figure 1). A total of 3 Â 104 cells were seeded on six-well plates and allowed to Upregulation of PLK2 was observed in all cell lines when compared attach. After 24 h, the medium was replaced and cells were treated with to the control. In GBM samples, a high expression of PLK2 was

Cancer Gene Therapy (2013), 499 – 506 & 2013 Nature America, Inc. Polo-like kinases in glioblastoma JA Pezuk et al 501

Figure 1. Relative PLK1-4 gene expression in glioblastoma (GBM) cell lines and GBM tumor samples. HPRT and TBP were used as endogenous genes and a pool of white matter (WM) samples was used as a calibrator; PLK1 expression levels were depicted separately for the cell lines used in functional studies: T98G (16 times higher than WM) and SF188 (11 times higher than WM). observed for only five patients (Figure 1). Seven out of eight cell BI 2536 induces mitotic arrest in GBM cell lines lines showed PLK3 downregulation whereas 50% of tumor BI 2536 treatment induced a strong G2/M arrest in SF188 and samples showed lower levels of this gene (Figure 1). All GBM T98G cells when compared to the controls at all times tested cell lines showed upregulation of PLK4, and all but three tumor (Figure 3). This effect was also demonstrated by the significant samples showed similar expression patterns (Figure 1). increase of mitotic cells up to over 20% in both GBM cell lines after 24 h (Po0.05) (Figure 4a). Cytogenetic analysis also evidenced a distinctive morphological appearance of treated cells with multi- PLK1 inhibition causes a decrease in cell proliferation in vitro ple micronuclei (data not shown). BI 2536 significantly inhibited the growth of both GBM cell lines when compared to control (0.1% dimethyl sulfoxide) at all times BI 2536 increases cell death in GBM cells tested (Po0.05), causing a significant reduction in viability in this cell lines (Figure 2a). The drug presented a maximum effect at BI 2536 treatment mediated a significant increase in the death 72 h, reducing proliferation by 69 and 50% for SF188 and T98G, rates that rise to B30% for SF188 and to B60% for T98G (Po0.05) respectively. The median dose effect (IC ) values for BI 2536 were (Figure 4b). For SF188 cells, such an increase was statistically 50 À 1 determined as 7.75 nmol l À 1 for SF188 and 79.07 nmol l À 1 for significant at all times for the 100 nmol l treatment; however, at T98G after 72 h of treatment. lower concentrations, the increase in dead cells was not always BI 2536 also inhibited cell proliferation (Po0.05) in primary observed, and this was not much different when BI 2536 was cultures after 48 h of treatment with 10, 50 and 100 nmol l À 1 combined with TMZ. For T98G cells, the increase of cell death À 1 when compared to the control (Figure 2b); similar results was significant after treatment with 50 and 100 nmol l at all were obtained for combinations with TMZ. When PLK1 expression times, and this increase was also observed with the combined À 1 levels were considered, IC50 values corresponded to 1.44 mmol l treatments. for GBM primary cultures with low PLK1 expression (o7.3), whereas the IC50 values for patients with higher levels of PLK1 BI 2536 inhibits cell invasion in GBM À 1 expression (47.3) corresponded to 2.57 mmol l after 48 h of The Matrigel assay was used to observe how cells can invade treatment (Figure 2c). through a matrix that represents the neutrophils in the human BI 6727, GW843682X and GSK461364 also inhibited cell prolifera- body. Invasion assays using transwell chambers coated with tion (Po0.05) in both the cell lines (Figure 2e) after 72 h of Matrigel showed significant reductions in invasion in a dose- treatment. The GW843682X compound showed to be more effective dependent manner for SF188 and T98G cells (Figure 4c), with in the inhibition of cell proliferation in SF188 cell lines; meanwhile, a highest reduction of 440% for T98G and 430% for SF188 at the GSK461364 was more efficient for T98G cell lines. À 1 50 nmol l of BI 2536 (Pp0.05).

BI 2536 decreases the clonogenic capacity of GBM cell lines BI 2536 interacts synergistically with TMZ Clonogenic capacity shows the cell renewal potential of a cell after In order to determine the ability of BI 2536 to sensitize GBM cell treatment; this corresponds to a long-term response after lines to the conventionally used TMZ, we measured the effects of treatment, and because of this, lower concentrations are normally combined treatments and tested different schedules with either used. PLK1 inhibition by BI 2536 signiﬁcantly reduced the simultaneous or sequential drug exposure. As shown in Table 1, BI clonogenicity of both cell lines compared to the control 2536 acted synergistically (CIo1) with TMZ in both cell lines in a (Po0.05) (Figure 2d). The clonogenic capacity of the SF188 cell treatment schedule-dependent manner after 48 h. Better response line was reduced by 63, 80 and 93% after treatment with 5, 10 and was observed for simultaneous treatment with very strong 50 nmol l À 1, respectively, and by 82, 83 and 93% when 5, 10 and synergism (CIo0.1) for both the cell lines. Interestingly, very high 50 nmol l À 1 were combined with 50 mmol l À 1 of TMZ. In T98G dose reduction index values (ranging from 127.63 to 142.06) cells, BI 2536 reduced the clonogenic capacity by 35, 79 and 98% were observed for the SF188 cell line. Similarly, sequential BI 2536/ at 5, 10 and 50 nmol l À 1, respectively, and, when combined with TMZ administration produced a synergistic effect at all concentra- 100 mmol l À 1 of TMZ, the inhibition corresponded to 39, 76 and tions of BI 2536 tested for SF188, whereas synergism was 98%, respectively. The IC50 for the SF188 cells was 2.28 nmol l À 1, only observed for T98G cells at 10 and 50 nmol l À 1. Conversely, and for T98G cells, it was 6.1 nmol l À 1. when both cells lines were pretreated with TMZ and exposed to

& 2013 Nature America, Inc. Cancer Gene Therapy (2013), 499 – 506 Polo-like kinases in glioblastoma JA Pezuk et al 502

Figure 2. (a) Proliferation inhibition of SF188 (left) and T98G (right) cell lines after treatment with BI 2536. (b) Proliferation inhibition of primary glioblastoma (GBM) cultures (n ¼ 13) after treatment with BI 2536 and BI 2536 combined with temozolomide (TMZ) after 48 h. (c) Proliferation inhibition of primary GBM cultures after treatment with BI 2536 according to PLK1 gene expression levels. (d) BI 2536 potently abrogated the clonogenic capacity of both GBM cell lines treated with BI 2536 in SF188 (left) and T98G (right) cell lines. (e) Proliferation inhibition of SF188 (left) and T98G (right) cell lines after treatment with BI 6727(1 ¼ 50 nmol l À 1,2¼ 100 nmol l À 1 and 3 ¼ 150 nmol l À 1), GW843682X (1 ¼ 300 nmol l À 1, 2 ¼ 600 nmol l À 1 and 3 ¼ 1200 nmol l À 1) and GSK461364 (1 ¼ 75 nmol l À 1,2¼ 150 nmol l À 1 and 3 ¼ 300 nmol l À 1) after 72 h. *Po0.05.

different concentrations of BI 2536, antagonistic effects were and more proliferative phenotypes. Moreover, PLK1 inhibition by observed (CI41). siRNA decreased cell viability in vitro and in xenograft models.22–23 PLK2 hyperexpression, on the other hand, has been reported in osteosarcomas,24 where its downregulation has been associated DISCUSSION with chemotherapy resistance in epithelial ovarian cancer.25 Both GBM is the most aggressive tumor of the central nervous system.18 genes (PLK1 and PLK2) are involved in cell cycle progression. Even with the advent of TMZ, current treatments show very Compelling evidence shows that clinical response to cytotoxic limited benefits regarding survival and prognosis.19 Different drugs is generally significantly higher for rapidly proliferating approaches are now being used to identify new molecular tumor cells,26–28 such as those with higher PLK1/PKL2 levels. markers for therapeutic purposes, among which, the members Consequently, albeit predicting tumor aggressiveness increased of the PLK family have been identified as potential targets in cell proliferation may result in a better response to chemotherapy, different tumors. Increased PLK1 expression has previously been increasing the patient’s survival time. demonstrated in GBM cell lines and tumor samples when Similar to PLK2, PLK3 shows a peak of expression in G1, although compared to low-grade gliomas,20 reinforcing the hypothesis of it can be found during all cell cycle phases due to its high PLK1 being an important factor in tumor maintenance.21 Recently, stability.29 Our results showed downregulation of PLK3 in most Lee et al.22 demonstrated via microarray profiling of 467 human cases when compared to the white matter pool. Lower levels of GBM samples, a close association between high PLK1 expression PLK3 have been observed in different tumors, suggesting that its

Cancer Gene Therapy (2013), 499 – 506 & 2013 Nature America, Inc. Polo-like kinases in glioblastoma JA Pezuk et al 503

Figure 3. BI 2536 treatments induced G2/M arrest in SF188 (a) and T98G (b) Glioblastoma (GBM) cell lines at all times tested. decrease could contribute to increased genomic instability due treatment; however, this was not always observed when cells were to its participation in the cell response to DNA damage and treated first with BI 2536 and even less frequent when cells were cellular stress.7 pretreated with TMZ, probably due to interaction in the inhibition PLK4 has an important function in mitotic progression and pathways. Sensitization caused by BI 2536 has been previously is also involved in centriole duplication.7 While loss of demonstrated for NVP-AEW541 (inhibitor of insulin-like growth heterozygosity and downregulation of PLK4 have been factor-1 receptor) in biliary tract cancer,39 pointing to the described in hepatocellular carcinoma,30 its hyperexpression has possibility of using BI 2536 in multimodality therapy. Although a been reported for colon carcinoma.31 Our results were also decrease in proliferation was found in GBM primary tumor controversial: all the cell lines studied showed upregulation of this cultures, this was less efficient than for cell lines; this could be gene, though some tumor samples showed low levels of PLK4. explained by the fact that in tumor samples PLK2 gene expression Functional studies of PLK1 inhibition have recently reinforced was down-regulated (see Figure 1), and as a tumor suppressor, the important role of these proteins in tumor progression and it could be responsible for the poor efficiency in decreasing maintenance. PLK1 inhibition by RNAi demonstrated to have proliferation. Besides proliferation, the ability of cancer cells to antitumoral effects, causing mainly mitotic arrest, aneuploidy and form colonies is also a fundamental requirement in the metastasis apoptosis.32 It has been proposed that the response to PLK1 process. Similar to the decrease in clonogenic capacity after inhibition is dependent on TP53;9 SF188 has been reported as treatment observed here, Renner et al.35 obtained 40 and 65% mutated, however, T98G has controversial reports.33 Among the reduction of the clonogenic potential of leukemia cells treated different compounds being developed that inhibit PLK1, BI 2536 with BI 2536; moreover, Morales et al.38 and Oliveira et al.36 have has shown anticancer activities against different tumor cells.34 shown decreased clonogenic formation in osteosarcoma cell lines Herein, we demonstrated that BI 2536, BI 6727, GW843682X and and melanoma, respectively. GSK4613640 efficiently decrease proliferation in GBM cells. Hu Moreover, the in vitro effects of BI 2536 on SF188 and T98G in et al.29 also demonstrated that the inhibition of PLK1 using cell cycle progression showed an accumulation of cells with a siRNA in SF188 cell lines decreases cell growth. Diminished doubled DNA content. This G2/M increase has been previously proliferation after PLK1 inhibition has also been reported in reported after PLK1 inhibition in osterosarcoma cell lines,38 in adenocarcinoma,13 leukemia,35 melanoma,36 cervical adeno- melanoma cells36 and in breast cancer cells.40 Further cytogenetic carcinoma37 and osteosarcoma cells.38 Synergistic effects with analysis demonstrated a higher accumulation (B20%) of dividing TMZ were also observed in both the cell lines at simultaneous cells after 24 h. BI 2536 has been shown to cause mitotic arrest in

& 2013 Nature America, Inc. Cancer Gene Therapy (2013), 499 – 506 Polo-like kinases in glioblastoma JA Pezuk et al 504

Figure 4. (a) BI 2536 treatments induced blockage in mitosis, with higher accumulation of dividing cells after 24 h of treatment in SF188 (left) and T98G (right) cell lines. (b) Increased cell death after treatment with BI 2536 and BI 2536 þ temozolomide (TMZ) in SF188 (left) and T98G (right) glioblastoma (GBM) cells lines at all times tested. (c) Both cell lines presented a signiﬁcant decrease in invasion rate after 24 h treatment with BI 2536. *Po0.05.

Table 1. Median dose effect analysis was employed to characterize the interactions between BI 2536 and TMZ in GBM cell lines in simultaneous and sequential treatments

T98G Concomitant TMZ Pretreatment with BI 2536 Pretreatment with TMZ

BI 2536 AF AF CI DRI AF CI DRI AF CI DRI

10 nmol l À 1 0.22 0.81 0.083 12.867 0.4 0.662 2.146 0.26 1.565 1.157 50 nmol l À 1 0.43 0.88 0.054 21.720 0.46 0.970 2.719 0.33 2.413 1.603 100 nmol l À 1 0.46 0.89 0.056 23.882 0.51 1.112 1.237 0.28 6.505 1.276

SF188 Concomitant TMZ Pretreatment with BI 2536 Pretreatment with TMZ

BI 2536 AF AF CI DRI AF CI DRI AF CI DRI

10 nmol l À 1 0.42 0.903 0.008 127.63 0.53 0.349 3.56 0.28 1.356 1.27 50 nmol l À 1 0.48 0.909 0.007 140.03 0.56 0.521 4.00 0.47 0.910 2.82 100 nmol l À 1 0.53 0.9099 0.007 142.067 0.6 0.604 4.69 0.45 1.68 2.61

Combination index (CI) values o1 correspond to a synergistic interaction Dose reduction index (DRI) reﬂects the fold reduction in the required concentration of tested agents when used in combination to achieve the comparable affected fraction (AF). Synergistic interactions are marked in gray.

Cancer Gene Therapy (2013), 499 – 506 & 2013 Nature America, Inc. Polo-like kinases in glioblastoma JA Pezuk et al 505 37 13 cervix adenocarcinoma and in xenografted mice. However, as 4 Brandes AA, Tosoni A, Franceschi E, Reni M, Gatta G, Vecht C. Glioblastoma in shown by the present results, after longer treatment, the mitotic adults. Crit Rev Oncol Hematol 2008; 67: 139–152. index decreased to normal levels. It has been demonstrated that 5 Strebhardt K. Multifaceted polo-like kinases: drug targets and antitargets for siRNA-mediated PLK1 inhibition also interferes with the correct cancer therapy. Nat Rev Drug Discov 2010; 9: 643–660. assembly of the anaphase-promoting complex, preventing 6 Bruinsma W, Raaijmakers JA, Rh Medema. Switching Polo-like kinase-1 on and off cytokinesis.41 In turn, after repeated division efforts, cells in time and space. Trends Biochem Sci 2012; 37: 534–542. 7 Winkles JA, Alberts GF. Differential regulation of polo-like kinase 1, 2, 3, and 4 proceed toward apoptosis and death. gene expression in mammalian cells and tissues. Oncogene 2005; 24: 260–266. Cytological studies demonstrated that BI 2536-treated cells die 8 Chopra P, Sethi G, Dastidar SG, Ray A. Polo-like kinase inhibitors: an 42 mostly due to mitotic catastrophe, which is markedly enhanced emerging opportunity for cancer therapeutics. Expert Opin Investig Drugs 2010; in cells lacking TP53 function such as SF188 and T98G, and is 19: 27–43. characterized by changes in nuclear morphology43 that precedes 9 Sanhaji M, Louwen F, Zimmer B, Kreis NN, Roth S, Yuan J. Polo-like kinase 1 inhi- cell death.44 Herein, we showed a high increase in cell death in bitors, mitotic stress and the tumor suppressor p53. Cell Cycle 2013; 12: 1340–1351. both the cell lines after treatment with BI 2536, although this 10 Garuti L, Roberti M, Bottegoni G. Polo-like kinases inhibitorsCurr Med Chem 2012; increase was not enhanced by combining with TMZ. Increased cell 19: 3937–3948. death has also been reported in several other tumor cells after 11 Johnson EE, Stewart KD, Woods KW, Giranda VL, Luo Y. Pharmacological and functional comparison of the polo-like kinase family: insight into inhibitor and PLK1 inhibition with BI 2536,40,45 and even in SF188 cell line after 23 substrate specificity. Biochemistry 2007; 46: 9551–9563. inhibition of PLK1 with siRNA. 12 Murugan RN, Park JE, Kim EH, Shin SY, Cheong C, Lee KS et al. Plk1-targeted small Moreover, the ability of cells to cross through coated chambers molecule inhibitors: molecular basis for their potency and specificity. Mol Cells and migrate was also significantly decreased by treatment with BI 2011; 32: 209–220. 46 2536 in our models. Similar results were reported by Zhang et al. 13 Steegmaier M, Hoffmann M, Baum A, Leńa´rt P, Petronczki M, Krssa´k M. BI 2536, a in bladder carcinoma cells treated with the PLK1-inhibitor potent and selective inhibitor of polo-like kinase 1, inhibits tumor growth in vivo. scytonemin, and after silencing by siRNA in colorectal cancer Curr Biol 2007; 17: 316–322. cells.47 Although the underlying mechanisms by which PLK1 14 Valente V, Teixeira SA, Neder L, Okamoto OK, Oba-Shinjo SM, Marie SK et al. inhibition might contribute to the suppression of cell invasion are Selection of suitable housekeeping genes for expression analysis in glioblastoma still unclear, recent data demonstrated that PLK1 affects invasion using quantitative RT-PCR. BMC Mol Biol 2009; 10:17. 15 Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time by phosphorylating vimentin and downregulating cell surface b1 À DDCt 48 quantitative PCR and the 2 method. Methods 2001; 25: 402–408. integrin. 16 Franken NA, Rodermond HM, Stap J, Haveman J, van Bree C. Clonogenic assay of GBM treatment is hampered by its inherent chemoresistance cells in vitro. Nat Protoc 2006; 1: 2315–2319. and its ability to invade/infiltrate surrounding tissues. Although 17 Chou TC, Talalay O. Quantitative analysis of dose-effect relationships: the efforts have been made in order to improve survival, most drugs combined effects of multiple drugs or enzyme inhibitors. Adv Enzyme Regul 1984; developed so far have shown disappointing results in clinical trials. 22: 27–55. In the present study, we showed that PLK1 inhibition not only has 18 Khasraw M, Lassman AB. Advances in the treatment of malignant gliomas. antiproliferative effects limiting invasion of tumor cells but also Curr Oncol Rep 2012; 12: 26–33. increases the cytotoxicity of TMZ. 19 Stupp R, Newlands E. New approaches for temozolomide therapy: use in newly BI 2536 proved to be well-tolerated in intravenous dose diagnosed glioma. Semin Oncol 2001; 28: 19–23. 20 Dietzmann K, Kirches E, von Bossanyi K, Jachau K, Mawrin C. Increased human regimens, efficiently crossed the brain–blood-barrier and inhibited 13 polo-like kinase-1 expression in gliomas. J Neurooncol 2001; 53: 1–11. tumor growth in vivo in several models. Although, the use of BI 21 Wasch R, Hasskarl J, Schnerch D, Lu¨bbert M. BI_2536-targeting the mitotic kinase 2536 in patients has been shown to be safe in phase I clinical Polo-like kinase 1 (Plk1). Recent Results Cancer Res 2010; 184: 215–218. 49–50 trials, controversially, some adverse effects in phase II clinical 22 Lee C, Fotovati A, Triscott J, Chen J, Venugopal C, Singhal A et al. Polo-Like trials have been pointed out.51–52 Even so, our data support Kinase 1 (PLK1) inhibition kills glioblastoma multiforme brain tumour cells in part the likelihood of using PLK1 as a therapeutic target, opening through loss of SOX2 and delays tumour progression in mice. Stem Cells 2012; 30: pharmacologic perspectives for the treatment of GBMs. The future 1064–1075. direction of this research will lead us to perform experiments in 23 Hu K, Lee C, Qiu D, Fotovati A, Davies A, Abu-Ali S et al. Small interfering RNA xenografted mouse models. library screen of human kinases and phosphatases identifies polo-like kinase 1 as a promising new target for the treatment of pediatric rhabdomyosarcomas. Mol Cancer Ther 2009; 8: 3024–3035. 24 Shen T, Li Y, Yang L, Xu X, Liang F, Liang S et al. Upregulation of Polo-like kinase 2 CONFLICT OF INTEREST gene expression by GATA-1 acetylation in human osteosarcoma MG-63 cells. The authors declare no conflict of interest. Int J Biochem Cell Biol 2012; 44: 423–429. 25 Syed N, Coley HM, Sehouli J, Koensgen D, Mustea A, Szlosarek P et al. Polo-like kinase Plk2 is an epigenetic determinant of chemosensitivity and clinical ACKNOWLEDGEMENTS outcomes in ovarian cancer. Cancer Res 2011; 71: 3317–3327. 26 Mito K, Kashima K, Kikuchi H, Daa T, Nakayama I, Yokoyama S. Expression of We are grateful to Dr Elza Tiemi Sakamoto-Hojo and Ana Paula de Lima Montaldi Polo-Like Kinase (PLK1) in non-Hodgkin’s lymphomas. Leuk Lymphoma 2005; 46: (MSc) for facilitating cell cycle analysis, to Kleiton Borges Silva (MSc) for helping with 225–231. the primary tumor cultures expansion, and to Veridiana Kiill Suazo (BSc) for technical 27 Amadori D, Volpi A, Maltoni R, Nanni O, Amaducci L, Amadori A et al. Cell assistance. We also thank the Foundation for Research Support of Saõ Paulo (FAPESP proliferation as a predictor of response to chemotherapy in metastatic breast # 2009/50118-2 and 2009/11053-2) and the Coordination of Improvement of Higher cancer: a prospective study. Breast Cancer Res Treat 1997; 43: 7–14. Education Personnel (CAPES) for financial support. 28 Mitchison T.J. The proliferation rate paradox in antimitotic chemotherapy. Mol Biol Cell 23: 1–6 2012. 29 Archambault V, Glover DM. Polo-like kinases: conservation and divergence in their REFERENCES functions and regulation. Nat Rev Mol Cell Biol 2009; 10: 265–275. 1 Yang S, Liu J, Wang T, Li X, You C. Cerebellar glioblastoma multiforme: a retro- 30 Pellegrino R, Calvisi DF, Ladu S, Ehemann V, Staniscia T, Evert M et al. Oncogenic spective study of 28 patients at a single institution. Int J Neurosci 2013. and tumor suppressive roles of polo-like kinases in human hepatocellular carci- doi:10.3109/00207454.2013.791292. noma. Hepatology 2010; 51: 857–868. 2 Khalatbari MR, Hamidi M, Moharamzad Y. Glioblastoma multiforme with very 31 Rosario CO, Ko MA, Haffani YZ, Gladdy RA, Paderova J, Pollett A et al. Plk4 is rapid growth and long-term survival in children: report of two cases and review of required for cytokinesis and maintenance of chromosomal stability. Proc Natl the literature. Childs Nerv Syst 2011; 27: 1347–1352. Acad Sci USA 2010; 107: 6888–6893. 3 Slotty PJ, Siantidis B, Beez T, Steiger HJ, Sabel M. The impact of improved 32 Macmillan JC, Hudson JW, Bull S, Dennis JW, Swallow CJ. Comparative expression treatment strategies on overall survival in glioblastoma patients. Acta Neurochir of the mitotic regulators SAK and PLK in colorectal cancer. Ann Surg Oncol 2001; 8: (Wien) 2013; 155: 959–963. 729–7402.

& 2013 Nature America, Inc. Cancer Gene Therapy (2013), 499 – 506 Polo-like kinases in glioblastoma JA Pezuk et al 506 33 Soussi T. Handbook OF p53 Mutation in Cell Lines. Version 1.0 07/2007 43 Ianzini F, Mackey MA. Mitotic Catastrophe. In: Gewirtz DA, Holt SE (eds). Cancer http://p53.free.fr/Database/Cancer_cell_lines/cell_lines_1.0.pdf. Drug Discovery and Development Apoptosis, Senescence and Cancer. S Grant 34 Guan R, Tapang P, Leverson JD, Giranda VL, Luo Y. Small interfering RNA- Humana Press: Totowa, NJ, USA, 2007. mediated Polo-like kinase 1 depletion preferentially reduces the survival of 44 Vakifahmetoglu H, Olsson M, Zhivotovsky B. Death through a tragedy: mitotic p53-defective, oncogenic transformed cells and inhibits tumor growth in animals. catastrophe. Cell Death Differ 2008; 15: 1153–1162. Cancer Res 2005; 65: 2698–2704. 45 Yim H, Erikson RL. Polo-like kinase 1 depletion induces DNA damage in early S 35 Renner AG, Dos Santos C, Recher C, Bailly C, Creáncier L, Kruczynski A et al. prior to caspase activation. Mol Cell Biol 2009; 29: 2609–2621. Polo-like kinase 1 is overexpressed in acute myeloid leukemia and its 46 Zhang Z, Zhang G, Kong C. High expression of polo-like kinase 1 is associated with inhibition preferentially targets the proliferation of leukemic cells. Blood 2009; the metastasis and recurrence in urothelial carcinoma of bladder. Urol Oncol 2011. 114: 659–662. doi:10.1016/urolonc.2011.11.028. 36 Oliveira JC, Brassesco MS, Pezuk JA´ , Morales AG, Valera ET, Montaldi AP et al. In 47 Han DP, Cui JT, Lu AG, Chen XH, Feng B, Zong YP et al. Influence of silencing Polo- vitro PLK1 inhibition by BI 2536 decreases proliferation and induces cell-cycle like kinase 1 on migration and invasion of colorectal cancer cells. Zhonghua Wei arrest in melanoma cells. J Drugs Dermatol 2012; 11: 587–592. Chang Wai Ke Za Zhi 2011; 14: 61–64. 37 Pezuk JA, Brassesco MS, Oliveira JC, Morales AG, Montaldi AP, Sakamoto-Hojo ET 48 Rizki A, Mott JD, Bissell MJ. Polo-like kinase 1 is involved in invasion through et al. Antiproliferative in vitro effects of BI 2536-mediated PLK1 inhibition on extracellular matrix 2007Cancer Res 67: 11106–11110. cervical adenocarcinoma cells. Clin Exp Med 2013; 13: 75–80. 49 Ellis PM, Chu QS, Leighl N, Laurie SA, Fritsch H, Gaschler-Markefski B et al. 38 Morales AG, Brassesco MS, Pezuk JA, Morales AG, Montaldi AP, A Phase I open-label dose-escalation study of intravenous BI 2536 together Sakamoto-Hojo ET et al. BI 2536-mediated PLK1 inhibition suppresses HOS and with pemetrexed in previously treated patients with non–small–cell lung cancer. MG-63 osteosarcoma cell line growth and clonogenicity. Anticancer Drugs 2011; Clin Lung Cancer 2013; 14: 19–27. 22: 995–1001. 50 Frost A, Mross K, Steinbild S, Hedbom S, Unger C, Kaiser R et al. Phase i study of 39 Wolf S, Lorenz J, Mossner J, Wiedmann M. Treatment of biliary tract cancer with the Plk1 inhibitor BI 2536 administered intravenously on three consecutive days NVP-AEW541: mechanisms of action and resistance. World J Gastroenterol 2010; in advanced solid tumours. Curr Oncol 2012; 19: e28–e35. 16: 156–166. 51 Schoffski P, Blay JY, De Greve J, Brain E, Machiels JP, Soria JC et al. Multicentric 40 Maire V, Ne´mati F, Richardson M, Vincent-Salomon A, Tesson B, Rigaill G et al. parallel phase II trial of the polo-like kinase 1 inhibitor BI 2536 in patients with Polo-like Kinase 1: a potential therapeutic option in combination with conven- advanced head and neck cancer, breast cancer, ovarian cancer, soft tissue tional chemotherapy for the management of patients with triple-negative breast sarcoma and melanoma. The first protocol of the European Organization cancer. Cancer Res 2012; 73: 813–823. for Research and Treatment of Cancer (EORTC) Network of Core Institutes (NOCI). 41 Liu X, Choy E, Harmon D, Vincent-Salomon A, Tesson B, Rigaill G et al. Inhibition of Eur J Cancer 2010; 46: 2206–2215. polo-like kinase 1 leads to the suppression of osteosarcoma cell growth in vitro 52 Sebastian M, Reck M, Waller CF, Kortsik C, Frickhofen N, Schuler M et al. The efficacy and in vivo. Anticancer Drugs 2011; 22: 444–453. and safety of BI 2536, a novel Plk-1 inhibitor, in patients with stage IIIB/IV non-small 42 Plyte S, Musacchio A. PLK1 inhibitors: setting the mitotic death trap. Curr Biol cell lung cancer who had relapsed after, or failed, chemotherapy: results from an 2007; 17: R280–R283. open-label, randomized phase II clinical trial. JThoracOncol2010; 5: 1060–1067.

Cancer Gene Therapy (2013), 499 – 506 & 2013 Nature America, Inc.