Targeting Protein Kinase CK2 Suppresses Prosurvival ...clincancerres.aacrjournals.org/content/clincanres/19/23/6484.full.pdf · Targeting Protein Kinase CK2 Suppresses Prosurvival

Cancer Therapy: PreclinicalSee related article by Ji and Lu, p. 6335

Targeting Protein Kinase CK2 Suppresses ProsurvivalSignaling Pathways and Growth of Glioblastoma

Ying Zheng1, Braden C. McFarland1, Denis Drygin4, Hao Yu1, Susan L. Bellis1, Hyunsoo Kim2,Markus Bredel3, and Etty N. Benveniste1

AbstractPurpose: Gliomas are the most frequently occurring primary malignancies in the brain, and

glioblastoma is the most aggressive of these tumors. Protein kinase CK2 is composed of two catalytic

subunits (a and/or a‘) and two b regulatory subunits. CK2 suppresses apoptosis, promotes neoangio-

genesis, and enhances activation of the JAK/STAT, NF-kB, PI3K/AKT, Hsp90, Wnt, and Hedgehog

pathways. Aberrant activation of the NF-kB, PI3K/AKT, and JAK/STAT-3 pathways is implicated in

glioblastoma progression. As CK2 is involved in their activation, the expression and function of CK2 in

glioblastoma was evaluated.

Experimental Design and Results: Analysis of 537 glioblastomas from The Cancer Genome Atlas

Project demonstrates the CSNK2A1 gene, encoding CK2a, has gene dosage gains in glioblastoma (33.7%),

and is significantly associated with the classical glioblastoma subtype. Inhibition of CK2 activity by CX-

4945, a selective CK2 inhibitor, or CK2 knockdownby siRNA suppresses activation of the JAK/STAT, NF-kB,and AKT pathways and downstream gene expression in human glioblastoma xenografts. On a functional

level, CX-4945 treatment decreases the adhesion and migration of glioblastoma cells, in part through

inhibition of integrin b1 and a4 expression. In vivo, CX-4945 inhibits activation of STAT-3, NF-kB p65, and

AKT, and promotes survival of mice with intracranial human glioblastoma xenografts.

Conclusions:CK2 inhibitorsmay be considered for treatment of patients with glioblastoma.Clin Cancer

Res; 19(23); 6484–94. �2013 AACR.

IntroductionGlioblastoma is the most deadly primary malignant

brain tumor. Genomic abnormalities in glioblastomahave allowed molecular classification into four subtypes:classical, proneural, neural, and mesenchymal (1).Despite the combination of surgery followed by radio-therapy and chemotherapy, median survival of glioblas-toma patients is 12 to 15 months (2). Aberrant activationof signaling pathways has been implicated in glioblas-toma, particularly receptor tyrosine kinases (RTK) suchas EGFR and PDGFRA (3). However, clinical trials target-ing these individual RTKs have been disappointing,due to concurrent activation of multiple RTKs in glio-blastomas (3). Other pathways including JAK/STAT andNF-kB contribute to glioblastoma progression (4–7). The

JAK/STAT pathway transmits signals from interleukin(IL)-6 family cytokines by activation of STAT-3, whichinduces expression of genes that regulate antiapoptoticbehavior, angiogenesis, and proliferation. Furthermore,STAT-3 is a master regulator of the mesenchymal glio-blastoma phenotype (4). We and others have demon-strated that levels of activated STAT-3 are elevated inglioblastoma tissues (8, 9). JAK1 and JAK2 are alsoactivated in human glioblastoma xenografts (10). TheNF-kB signaling pathway is also constitutively activatedin glioblastoma (11, 12), partly due to deletion of IkBa,an inhibitor of NF-kB (5).

Protein kinase CK2 is composed of two catalytic (aand/or a’) and two b regulatory subunits. CK2 is aconstitutively active serine-threonine kinase that plays afundamental role in maintaining cell survival throughproproliferative, antiapoptotic, and proangiogenic signal-ing (13, 14). Elevated CK2 expression and activity hasbeen demonstrated in blood tumors and solid tumors(14). CK2 is a novel interaction partner of JAK1/2,potentiates JAK and STAT-3 activation (15), and regulatesexpression of IL-6 (16). The NF-kB and PI3K/AKT path-ways are also positively regulated by CK2, promoting cellsurvival and inhibiting apoptosis (13). These character-istics identify CK2 as an attractive therapeutic target. CX-4945 is the first selective orally bioavailable CK2 inhibitorto advance into human clinical trials (17, 18). CX-4945

Authors' Affiliations: Departments of 1Cell, Developmental and Integra-tive Biology, 2Pathology, and 3Radiation Oncology, University of Alabamaat Birmingham, Birmingham, Alabama; and 4Cylene Pharmaceuticals, SanDiego, California

Note: Supplementary data for this article are available at Clinical CancerResearch Online (http://clincancerres.aacrjournals.org/).

Corresponding Author: Etty N. Benveniste, University of Alabama atBirmingham, 1900 University Blvd., Birmingham, AL 35294-0006. Phone:205-934-7667; Fax: 1-205-975-6748; E-mail: [email protected]

doi: 10.1158/1078-0432.CCR-13-0265

�2013 American Association for Cancer Research.

ClinicalCancer

Research

Clin Cancer Res; 19(23) December 1, 20136484

on August 15, 2019. © 2013 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from

Published OnlineFirst September 13, 2013; DOI: 10.1158/1078-0432.CCR-13-0265

http://clincancerres.aacrjournals.org/

inhibits the activity of CK2a/CK2a’, suppresses cell-cycleprogression, angiogenesis, and PI3K/AKT signaling, andexhibits antitumor activity in breast, pancreatic, andprostate xenograft models (17, 19). In the clinical setting,CX-4945 inhibits CK2, leading to durable disease stabi-lization (�16 weeks) in 20% of treated patients whilebeing well tolerated (20).A glioblastoma-specific protein interaction analysis

identified CK2a as an important node connecting cell-cycle–associated proteins (21). Nonspecific CK2 inhibi-tors such as TBB and DMAT exhibit cytotoxic potentialtowards glioblastoma cells in vitro (22), Ellipticine andbenzopyridoindole derivatives, which inhibit CK2 activ-ity, display antitumor activity in a flank model of glio-blastoma (23), and azonaphthalene derivatives, allostericinhibitors of CK2, inhibit growth of U373-MG cells in theflank (24). Downregulation of CK2 by siRNA inducesdeath of the glioblastoma cell line M059K (25). In thisstudy, we analyzed expression of CK2 subunits andassociation with different subtypes of glioblastoma, usingThe Cancer Genome Atlas (TCGA) database. We evalu-ated the effect of inhibiting CK2a/CK2a’ activity orsilencing CK2a, CK2a’, and CK2b expression on theJAK/STAT, NF-kB, and AKT signaling pathways in glio-blastoma, and downstream functional effects such as cell-cycle progression, apoptosis, adhesion, migration, andsenescence. Finally, we tested the antitumor efficacy ofCX-4945 in flank and intracranial human glioblastomaxenograft models in vivo.

Materials and MethodsGene copy number variation and genomic analyses

Five-hundred and thirty seven glioblastoma samplesfrom the TCGA (http://cancergenome.nih.gov/) were usedas a genomic discovery set. Raw Affymetrix Genome-WideHuman SNP Array 6.0 and Agilent Human Genome CGHMicroarray 244A gene dosage data, Affymetrix HumanGenome U133 Plus 2.0 Array and Agilent 244K CustomGene Expression data, and clinical data were retrieved fromthe Open-Access and Controlled-Access Data Tiers Portal(https://tcga-data.nci.nih.gov/tcga/) of TCGA and prepro-cessed for downstream analyses. Gene-level copy numbervariation (CNV) was estimated using the circular binarysegmentation algorithm from the "snapCGH" packageusing an R code, as described (5). Gene dosage segmentswere classified as chromosomal ‘gain’ or ‘loss’ if the absolutevalue of the predicted dosage was more than 0.75 timesthe interquartile range of the difference between observedand predicted values for each region. CNV data processedusing the Genomic Identification of Significant Targets inCancer (GISTIC2) algorithm were retrieved from the BroadInstitute at http://gdac.broadinstitute.org/runs/analyses__2012_03_21/data/GBM/20120321/. The GISTIC2 algo-rithm (26) thresholded estimated CNV values to �2, �1,0, 1, and 2 representing homozygous deletion, heterozy-gous (single copy) deletion, diploid normal copy, low-levelcopy number amplification (i.e., gene dosage gains), orhigh-level copy number amplification, respectively. Genesmapped onto the human genome coordinates using theUniversity of California, Santa Cruz cgData HUGO probe-Map were visualized using the Cancer Genomics Browser(https://genome-cancer.ucsc.edu/).

Cells and reagentsU87-MG and U251-MG lines were authenticated as

described (10). Antibody to CK2b was from EMD Bios-ciences. CX-4945 was provided by Cylene Pharmaceuti-cals. Recombinant IL-6, sIL-6R, EGF, TNF-a, and IL-1bwere from R&D Systems. Antibodies to p-Y-STAT-3, STAT-3, p-Y-STAT-5, STAT-5, NF-kB p65S536, AKTS473, andAKT were from Cell Signaling Technology. Antibodies toCK2a, CK2a’, p-Y-JAK2, and NF-kB p65 were from SantaCruz Biotechnology. Antibodies to Ki-67 and CK2a forimmunohistochemistry and JAK2 were from Millipore,NF-kB p65S529 antibody from Abcam, AKTS129 anti-body from Abnova, and integrin b1 antibody from BDBiosciences. Normal human brain lysate was provided byDr. Harald Sontheimer [University of Alabama at Bir-mingham (UAB), Birmingham, AL].

Glioblastoma xenograft tumorsHuman glioblastoma xenograft tumors X1016, X1046,

and X1066 were maintained in the flank of nude mice withapproval (APN #100908862) of the UAB InstitutionalAnimal Care and Use Committee (10). For subcutaneousimplantation, 200 mL of X1046 tumor slurry was injectedinto flanks of female nude mice and tumor size measuredby calipers and calculated as described (10). Mice were

Translational RelevanceWe demonstrate that protein kinase CK2, indispens-

able for cell survival and a regulator of signaling cascadesinvolved in glioblastoma tumorigenesis, may be a noveltherapeutic target for treatment of glioblastoma. Wedemonstrate for the first time that CSNK2A1, the geneencoding CK2a, shows frequent gene dosage gains inhuman glioblastoma tissues.We further determined thatin classical glioblastoma, these gene dosage gains occurin more than 50% of cases, highlighting patients withclassical glioblastoma as a population that may beresponsive to CK2-modulating therapeutics. UtilizingCX-4945, the first selective CK2 inhibitor to advanceinto human clinical trials, we demonstrate the therapeu-tic effects of inhibiting CK2 activity in flank and intra-cranial models of glioblastoma-patient derived cells invivo. Abnormal STAT-3 activity has been implicated inthe resistance of glioblastoma to temozolomide, and weshow that CK2 expression/activity is critical for STAT-3activation in glioblastomas, highlighting the potential ofCK2 inhibitors being used in combination with stan-dards of care. Given that safety of CK2 inhibition inhumans has been established, targeting this kinase mayprovide a new therapeutic venue for treatment ofglioblastoma.

CK2 Promotes GBM Progression

www.aacrjournals.org Clin Cancer Res; 19(23) December 1, 2013 6485




randomized to vehicle or CX-4945 treatment, which wasadministered intraperitoneally twice a day at 75 mg/kgfrom days 3 to 5, and then orally twice a day from days6 to 40. Intraperitoneal injections were discontinuedbecause of irritation at the injection site. At day 40, micewere euthanized and tumors excised and snap frozen.Percent tumor growth inhibition (TGI) value was calcu-lated as 100 � [1 � (Treated Day 40 � Treated Day 1)/(Vehicle Day 40 � Vehicle Day 1)]. For intracranial implan-tation, 0.5 � 106 X1046 or X1016 cells in 5 mL ofmethylcellulose were injected 2 mm anterior and 1 mmlateral to the bregma at a depth of 2 mm, as described(10). Vehicle or CX-4945 (75 mg/kg) was administeredorally twice a day from day 5 for 4 consecutive weeks.Survival of mice was monitored.

ImmunoblottingSamples were lysed in radioimmunoprecipitation buffer

for 30minutes, andapproximately 20–50mgof total proteinused for immunoblotting (10).

RNA interferenceU251-MG cells were transfected with 100 nmol/L of

CK2a, CK2a’, CK2b, or nontarget siRNA for 48 hours usingDharmacon transfection reagent 1 (15).

Real-time RT-PCR and TaqMan gene expression assaysRNA was extracted using TRIzol (Invitrogen). TaqMan

primers were used, qRT-PCR performed, and data analyzedusing the comparative Ct method (10).

Immunofluorescence and phalloidin stainingX1016 or U251-MG cells were fixed with 2% paraformal-

dehyde, stained with antibodies to CK2a or CK2a’, or withAlexa Fluor 488 phalloidin, and 40, 6–diamidino–2–phe-nylindole (DAPI), and analyzed by immunofluorescencemicroscopy.

Colorimetric cell adhesion assayXenograft X1046 was treated with CX-4945 for 16 hours,

and cells seeded on fibronectin-coated plates (15 mg/mL) inserum-free medium for 1 hour. The cells were then fixed,stained with crystal violet, and absorbancemeasured at 540nm.

Scratch assayX1066 cells were treated with CX-4945 in serum-free

medium for 4 hours. The cells were then scratched usinga p200pipette tip, and imaged at 0 and16hourswith a LeicaMZ16FA stereoscope equipped with a Leica DFC 490 CCDcamera. The unhealed area was quantified using the ImageJ1.41o program.

WST-1 proliferation assayCells were seeded in triplicate at 1.5� 104 cells per well

in the absence or presence of CX-4945. WST-1 reagentwas added and absorbance measured at 450 nm against655 nm (15).

Apoptosis assayCells were treated with CX-4945 for 24 hours, trypsi-

nized, stained with Annexin V and propidium iodide, andexamined by flow cytometry using FlowJo 7.5.5 software(15).

Cell-cycle analysisCells were treated with CX-4945 for 24 hours, fixed with

70%ethanol overnight, stainedwithpropidium iodide, andthe percentage of cells in different cell-cycle stages wasdetermined by FlowJo 7.5.5 software (15).

Densitometric and statistical analysesLevel of significance was determined by Student t test

distribution. Kaplan–Meier survival curves were drawnusing SigmaPlot 11.2 software, and Log-Rank significancetest was performed, and 95% confidence intervals formedi-an survival were computed. P < 0.05 was considered statis-tically significant. Linear regression analyses and graphicalmodel validation were executed using R software. Scatter-plots and locally weighted least squares smooths were usedto confirm the suitability of linear regression analyses, andstatistical significance of these relationships was assessedaccording to the P value for the estimated slope of theregression line. Two-way contingency table analysis,unpaired t test, and Wilcoxon rank-sum test were used asappropriate. ORs in the two-way contingency table analysisand 95% confidence intervals were computed usingWoolf’smethod for variance estimation.

ResultsCSNK2A1 shows frequent gene dosage gains inglioblastoma

CNV analysis in 537 glioblastomas from the TCGA data-base indicates that CSNK2A1, the gene encoding CK2a andmapping to chromosome 20p13kC, shows low-level ampli-fications (i.e., gene dosage gains) in 33.7% of tumors (Fig.1A). In most cases, these gene dosage gains are broadlyspanning and involve whole chromosome 20, suggestingCSNK2A1 as one of potentiallymultiple target genes drivingthese aberrations. Significantly higher CSNK2A1 mRNAlevels were detected in glioblastomas with CSNK2A1 genedosage gains (Fig. 1B). Among 490 glioblastoma sampleswith molecular subtype information, CSNK2A1 gene dos-age gain ismore common (50.7%) in classical glioblastomathan in non-classical glioblastoma (21.3%) (Fig. 1C). CNVanalysis ofCSNK2A2, the gene encodingCK2a’ at 16q21kC,revealed only sporadic gene dosage gains, and analysis ofthe CSNK2B gene, encoding CK2b, revealed a modestpercentage of deletion (7.3%) at 6p21.3kC (not shown).

CK2 is required for JAK/STAT activation inglioblastoma cells

Activation of the JAK/STAT-3 pathway is implicated inglioblastoma progression and propagation of glioblastomastem cells (27–30). We tested whether inhibition of CK2affects STAT activation. Three human glioblastoma xeno-grafts, X1016, X1046, and X1066, which have detectable

Zheng et al.

Clin Cancer Res; 19(23) December 1, 2013 Clinical Cancer Research6486




basal STAT-3 activation (10) and two human glioma lines,U251-MG and U87-MG, were used. Expression of CK2a,CK2a’, and CK2b was detected in the glioblastoma xeno-grafts and cell lines as well as normal human brain lysate(Fig. 2A). Expression of CK2a (Supplementary Fig. S1A andS1B) and CK2a’ (Supplementary Fig. S1B) is detected inboth the cytoplasm and nucleus. U251-MG cells transfectedwithCK2a, CK2b, orCK2a’ siRNAswere stimulatedwith IL-6 and sIL-6R, and examined for phospho-tyrosine STAT-3

levels. The combination of IL-6 and sIL-6R was used topromote optimal STAT-3 activation (31). Downregulationof CK2a, CK2b or CK2a’ expression led to reduced IL-6–induced STAT-3 activation, with CK2a and CK2a’ siRNAhaving the most pronounced effect (Fig. 2B). CK2a siRNAcauses decreased CK2b levels (Fig. 2B, lanes 2 and 6),whereas CK2b siRNA resulted in decreased CK2a’ levels(Fig. 2B, lanes 3 and 7); both phenomena have beenpreviously observed (25, 32, 33). Basal and IL-6–induced

Figure 1. Gene dosage gain of CSNK2A1 in glioblastoma (GBM). A, gene-level CNV analysis across chromosome 20 in 537 glioblastomas. Gene dosagesaremapped according to gene order on chromosome 20.CSNK2A1 shows gene dosage gains in 33.7%of tumors (green line,CSNK2A1 locus on 20p13kC).The RefSeq gene track shows coding gene locations. The proportions diagram shows the frequency of gene dosage gains along chromosome 20.Color intensity is proportional to the deviation from zero. Gene-dosage values indicate the log2 ratio of red (R, Cy5) to green (G, Cy3) intensity of thefluorescence dye (or log2R/G). B, gene dosage gain of CSNK2A1 is associated with significant gain of CSNK2A1mRNA expression in 482 and 428 patientswhose tumors were profiled on two different platforms at the Broad Institute (BI) and University of North Carolina at Chapel Hill (UNC), respectively. Values forgene dosage and gene expression are presented as log2R/G ratios. Locally weighted least squares (LOWESS) smooth fits (in red) confirmed theappropriateness of the linear regression models; P value indicates statistical significance according to estimated slope of the regression line. Thecorresponding box plots show the distribution of CSNK2A1 expression in tumors containing and lacking CSNK2A1 gains; P value calculated via Wilcoxonrank-sum test. C, gene-dosage profiles for CSNK2A1 across 490 glioblastomas along with its relationship to four molecular subtypes of glioblastoma.A corresponding two-way contingency-table analysis reveals a significant association of CSNK2A1 gain with the classical subtype. CI, confidence interval.






STAT-3 activation was inhibited by the selective CK2 inhib-itor CX-4945 in a dose-dependent manner in X1066 (Fig.2C), X1046 (Supplementary Fig. S2A) and U251-MG cells(Supplementary Fig. S2B). IL-6–induced JAK2 activationwas inhibited by CX-4945 in X1066 (Fig. 2D). Total JAK2levels were reduced after CX-4945 treatment (Fig. 2D),suggesting CX-4945 may affect JAK2 stability. OncostatinM (OSM), another IL-6 family member, is elevated inglioblastoma tumors and activates STAT-3 (34). Expressionof OSM-induced STAT-3 target genes was inhibited by CX-4945 (Fig. 2E). A bioinformatics analysis in glioma sug-gested CK2 could be downstream of EGFRvIII (35), themutated oncogenic form of EGFR common in glioblastoma(36). EGF can activate the STAT pathway (37); thus, weexamined CK2 involvement in this system. EGF-inducedSTAT-5 and STAT-3 activation was inhibited by CX-4945(Fig. 2F), as was expression of EGF-induced c-Myc, a STAT-3target gene (Fig. 2G). EGF-induced STAT-3 activation wasinhibited to a greater extent by CX-4945 than STAT-5.

CK2 is required for NF-kB p65 activation inglioblastoma cells

CK2 positively regulates the NF-kB pathway by promot-ing IkB degradation and enhancing DNA binding throughp65 phosphorylation (13), and TNF-a induces phosphor-

ylation of p65 serine 529 by CK2 (38). TNF-a induced p65phosphorylation was partially inhibited by knockdown ofCK2a, CK2b, or CK2a’ (Fig. 3A). TNF-a induced p65phosphorylation (Fig. 3B) and expression of downstreamtargets, IkBa and IL-8 (Fig. 3C), was suppressed byCX-4945in human glioblastoma xenografts, as was IL-1b–inducedp65 phosphorylation (Fig. 3D).

Inhibition of CK2 suppresses constitutive AKTactivation in glioblastoma cells

Mutations or deletions in PI3K and PTEN have definedthe PI3K/AKT pathway as one of the core pathways dysre-gulated in glioblastoma (39). CK2 positively regulates thePI3K pathway by affecting PTEN stability (40) and phos-phorylating serine 129 of AKT, which promotes its catalyticactivity (41). Constitutive phosphorylation of serine 129and 473 was detected in glioblastoma cells, which wasinhibited by CX-4945 (Supplementary Fig. S3).

CK2 inhibition decreases adhesion and migration ofglioblastoma cells

CK2 regulates cell morphology and the cytoskeleton (42,43). When glioblastoma cells were treated with CX-4945,they became retracted and rounded after treatment for 12to 16 hours, suggesting reduced cell adhesion (Fig. 4A).

X1016

A

D

F G

E

B CX1046 X1066 U251 U87 NB

CK2α

CK2α’

CK2β

GAPDH

GAPDH

Non-target siRNA

CK2α siRNA

CX-4945 (μmol/L) − − 2 10

CK2β siRNA

CK2α’ siRNA

IL-6 + sIL-6R

P-Y-STAT3

P-Y-JAK2

IL-6 + sIL-6R −

JAK2

GAPDH

CX-4945 (μmol/L) − 2 10 − 2 10

P-Y-STAT3

IL-6 + sIL-6R − − −

CX-4945 (μmol/L)

CX-4945 (μmol/L)

−−

− 5

0 16.1 29.7

0 88.1 96.0

10

P-Y-STAT5

P-Y-STAT3

EGF

Percentage inhibition


STAT5

STAT3



STAT3

STAT3

CK2α

CK2β

CK2α’

1 2 3 4

1 2 3 4

0 77.2 58.9 87.6

1 2 3 4 5 6 7 8

1 2 3 4 5 6

ACTIN

0 0 71.2 0 38.5 100

4

3

2

1

0EGF −

− − 5 10

mR

NA

fold

induction

mR

NA

fold

induction

25

20

15

10

5

0

c-Myc

SOCS3 IL-6 Mcl-1 Pim-1

UT

OSM

CX-4945 (2 μmol/L) + OSM

CX-4945 (10 μmol/L) + OSM

Figure 2. Inhibition of CK2 suppresses STAT activation. A, twenty micrograms of xenografts X1016, X1046, and X1066, glioblastoma (GBM) lines U251-MGand U87-MG, and normal brain (NB) lysate were immunoblotted with indicated antibodies. B, U251-MG cells were transfected with 100 nmol/L ofnon-target,CK2a,CK2b, orCK2a' siRNAs for 48hours, then stimulatedwith10ng/mLof IL-6 and25ng/mLof sIL-6R for 10minutes.CandD, xenograft X1066was treated with CX-4945 in serum-free medium for 4 hours, and then stimulated as above. Lysates were immunoblotted with indicated antibodies.E, U251-MGcellswere pretreatedwithCX-4945 in serum-freemedium for 4 hours, and then stimulatedwithOSM (5 ng/mL) for 1 hour.mRNAwas analyzedbyqRT-PCR. �, P < 0.05. F, U251-MG cells were pretreated with CX-4945 in serum-free medium for 4 hours, and then stimulated with EGF (50 ng/mL)for 10 minutes. G, U251-MG cells were serum-starved overnight, treated with CX-4945 for 4 hours, and stimulated with EGF for 1 hour. mRNA was analyzedby qRT-PCR. �, P < 0.05.

Zheng et al.





Staining for Phalloidin showed that there were more actinbundles on the membrane, and actin stress fibers appearedto be disorganized and impaired after CX-4945 treatment(Fig. 4B). U251-MG cells transfected with CK2a and/orCK2a’ siRNAs also became retracted and rounded, anddisplayed disorganized actin (Fig. 4C). A small-scale PCRarray was performed to identify genes involved in adhesionthat may be affected by CX-4945. Among the genes iden-tified, integrin a4 (ITGA4) and integrin b1 (ITGB1) weredownregulated by CX-4945 (Fig. 4D), as was ITGB1 proteinexpression (Fig. 4E). U251-MG cells transfected with CK2aor CK2a’ siRNAs have decreased ITGB1 protein expression(Fig. 4F). After glioblastoma, cells were treated with CX-4945 for 12 hours to inhibit ITGB1 expression (Supple-mentary Fig. S4A), cellswerewashed and incubatedwithoutCX-4945 for 12 hours. Expression of ITGB1 recoveredcompared with untreated control (Supplementary Fig.S4B), and cell shape was restored (Supplementary Fig.S4C), suggesting the effect of CX-4945 on glioblastoma cellmorphology and integrin expression is reversible. ITGA4/ITGB1 function as receptors for fibronectin (44) and CX-4945 inhibited glioblastoma cell adhesion to fibronectin(Fig. 4G). Furthermore, CX-4945 inhibited glioblastomacell migration (Fig. 4H and I).

Effects of inhibition of CK2 on glioblastoma cellfunctionTreatment with CX-4945 suppressed xenograft cell

growth (Fig. 5A and B and Supplementary Fig. S5). CK2aor CK2a plus CK2a’ siRNAs also decreased glioblastomacell growth (Fig. 5C). CX-4945 induced apoptosis (Fig. 5D)and increased the percentage of glioblastoma cells in G2–M

phase, and decreased that in G1 and S phases (Fig. 5E). CX-4945 treatment caused senescence in glioblastoma cellsafter 72 hours (Supplementary Fig. S6), and 10 mmol/L ofCX-4945 abrogated glioblastoma cell survival after 5 days(Supplementary Fig. S7). Thus, CK2 inhibition abrogatesmany glioblastoma biologic functions that are critical totumor growth and survival.

CX-4945 inhibits glioblastoma tumor growth in vivoCX-4945was tested in a subcutaneously implanted xeno-

graft model, using Xenograft X1046 (classical subtype), asassessed using the CLANC algorithm (1). CX-4945 treat-ment significantly inhibited tumor growth and exhibited76.4% TGI (Fig. 6A). At day 40, both groups of mice wereeuthanized 3 hours after the last administration of CX-4945or vehicle. Constitutively active STAT-3,NF-kBp65 andAKTwas detected in tumors of the vehicle-treated group, whileactivation of these pathways was barely detectable in 3 of 4mice of the CX-4945–treated group (Fig. 6B). The onetumor from the CX-4945 group that had high levels ofSTAT-3, NF-kB p65 and AKT activation happened to be thelargest tumor, thus having a poor response to CX-4945. CX-4945 treatment did not affect body weight (Fig. 6C) or thenumber of white blood cells, red blood cells, or levels ofhemoglobin (Fig. 6D). CX-4945 was then tested in theintracranial xenograft model. CX-4945 treatment signifi-cantly increased survival of X1046 tumor-bearing mice,with the median survival time after tumor implantationincreasing from 38 (95% confidence interval: 35.6–40.4) to59 days (95% confidence interval: 50.2–67.8; Fig. 6E). Theintracranialmodelwas repeatedutilizing another xenograft,X1016 (classical subtype), in which mice were sacrificed

Non-target siRNA

A B

C D

CK2α siRNA

CK2α

CK2β

CK2α’

GAPDH

CK2β siRNA

CK2α’ siRNA

TNF-α

TNF-αp65-S529

p65 S529Percentage inhibition

Percentage inhibitionp65

p65

p65

ACTIN

CX-4945 (μmol/L)

p65 S529


CX-4945 −−

− −

0 1 10 0 1 10

1 2 3 4 5 6 7 8

0 50.8 44.8 24.7

1 2 3 4 5 6

1 2 3 4 5 6

0 25.5 41.3 0 3.4 20.0

35

30

25

20

15

10

5

0

IκBα IL-8

mR

NA

fo

ld in

du

ctio

n

UTTNF-αCX-4945 + TNF-α

GAPDH

IL-1β (ng/mL) 0 0.5 0.5 1 1

0 54.1 0 33.6 0 24.8

− − −

Figure 3. Inhibition of CK2suppresses NF-kB activation. A,U251-MG cells were transfectedas described in Fig. 2B, thenstimulated with TNF-a (1 ng/mL)for 10 minutes. B, X1066 wastreated with CX-4945 for 4 hours inserum-free medium, and thenstimulated as above.C, X1066wastreated with CX-4945 (10 mmol/L)for 4 hours in serum-free medium,and then stimulatedwith TNF-a for1 hour. mRNA was analyzed byqRT-PCR. �, P < 0.05. D, X1046was pretreated with CX-4945(10 mmol/L) for 4 hours in serum-free medium and then stimulatedwith IL-1b for 10 minutes.






during treatment to evaluate the antitumor effects of CX-4945. Constitutively active STAT-3,NF-kBp65 andAKTwasdetected in intracranial X1016 tumors of the vehicle-treatedgroup, while activation of these pathways was diminishedinmice treatedwithCX-4945 (Fig. 6F). Thepercentage ofKi-67 positive cells in intracranial X1016 tumors was lower inCX-4945–treated mice compared with vehicle-treated mice(Fig. 6G).

DiscussionThe exactmechanism responsible forCK2overexpression

in cancer is not known. We report for the first time thatCSNK2A1, the gene encoding CK2a, harbors frequent gene

dosage gains in glioblastoma, and such gains correlate withincreases in CK2a mRNA. This information suggests thegenetic basis leading to overexpression of CK2amRNA andprotein expression inpatientswith glioblastoma.Moreover,we identify a particular subset of the disease, classicalglioblastoma, that has more than 50% frequency of thisalteration, highlighting patients with classical glioblasto-ma as a population that may be responsive to CK2-modulating therapeutics. Genome-wide CNV analysis inglioblastoma has been limited to evidence that chromo-some 20 harbors frequent gains in gene dosage that maybe driven by several oncogenic targets (45). Our novelfinding that CK2 expression/activity is required for

UTA B

C

D E F

G H I

CX-4945

UT CX-4945

X1016

U87-MG

U251-MG

1.2

1

0.8

0.6

0.4

0.2

0

1.2

1

0.8

0.6

0.4

0.2

0

mR

NA

exp

ressio

n

Ad

he

sio

n o

f ce

lls

(arb

itra

ry v

alu

e)

ITGA4 ITGB1

UT

CX-4945

Nontarget

siRNA CK2α siRNACK2α + CK2α’

siRNACK2α’ siRNA

CX-4945 (μmol/L)

CX-4945 (10 μmol/L)CX-4945 (μmol/L) CX-4945 (2 μmol/L)

0 h 16 h



0 2.1 29.0 0 7.2 63.0 0 3.2 50.4

Time (h) 12 18 24

ITGB1

ACTIN

− 1 10 − 1 10 − 1 10

1 2 3 4 5 6 7 8 9

Nontarget siRNA

CK2α siRNA

CK2α’ siRNA

CK2α’

GAPDH

CK2α

ITGB1

− −

− −

− −

0 79.3 92.6

120

80

40

0

0 2 10

CX-4945 (μmol/L)

“Unheale

d”

are

a

0 1 10

Figure 4. Inhibition of CK2 suppresses adhesion andmigration of glioblastoma (GBM) cells. A, X1016was treatedwithCX-4945 (10 mmol/L) for 16 hours (top), andU87-MG (middle) and U251-MG cells (bottom) were treated for 12 hours. Scale bar is 2 mm. B, X1016 was treated with CX-4945 (10 mmol/L) for 16 hours, andstainedwithAlexaFluor 488phalloidin andDAPI.C,U251-MGcellswere transfectedwith 100nmol/Lof non-target,CK2a, orCK2a' siRNAs, orCK2a (50 nmol/L)plus CK2a' (50 nmol/L) siRNAs for 48 hours, and stained with Alexa Fluor 488 phalloidin and DAPI. The scale bar is 14 mm. D, X1016 cells were treatedwith CX-4945 (10 mmol/L) for 16 hours. mRNA was analyzed by qRT-PCR. �, P < 0.05. E, X1046 was treated with CX-4945 and lysates immunoblotted withindicated antibodies. F, U251-MG cells were transfected with 100 nmol/L of non-target, CK2a, or CK2a' siRNAs for 48 hours, and lysates immunoblottedwith indicated antibodies. G, X1046 was treated with CX-4945 for 16 hours, and seeded onto fibronectin-coated plates for 1 hour. The cells were thenfixed and stained with crystal violet, and absorbance wasmeasured at 540 nm. The normalized value of untreated cells was set to 1. �, P < 0.001. H, X1066 cellswere incubated in the absence or presence of CX-4945 for 4 hours and scratched with a p200 pipette tip, and images were taken at 0 and 16 hours later. Thescale bar is 1 mm. I, images from the assay were quantified. Representative of three experiments (�, P < 0.01; ��, P < 0.001).

Zheng et al.





activation of prosurvival pathways, including the JAK/STAT, NF-kB, and PI3K/AKT pathways in glioblastomas,suggests that the CSNK2A1 gene could be one of theimportant oncogenic drivers for the selection of chromo-some 20 gains during gliomagenesis.Silencing CK2 expression with CK2 siRNA or inhibiting

CK2 activity with CX-4945 abrogates JAK/STAT activationand target gene expression in glioblastomas. While theregulation of JAK2 and STAT-3 by CK2 has been previouslydescribed by us and others (15, 16, 46), our discovery thatCK2 is involved in EGF-induced STAT-3 and STAT-5 acti-vation is novel and requires separate investigation. Abnor-mal STAT-3 activity has been implicated in the resistanceof glioblastoma to temozolomide (47), highlighting thepotential of CK2 inhibitors being used in combinationwith standards of care. CK2 inhibition also resulted in adecrease in total JAK2 levels. HSP90 stabilizes numerousproteins including JAK2, and CK2 is essential for thechaperone function of HSP90 (48). Thus, inhibiting CK2will in turn inhibit HSP90 function, thereby promotingJAK2 degradation, whichmay be another novelmechanismby which STAT activation is inhibited.Silencing CK2 expression or suppressing CK2 activity

also abrogates NF-kB p65 activation and expression ofNF-kB target genes in glioblastomas. The IkBa gene isheterozygously deleted in glioblastoma, leading to NF-kBactivation (5). Analysis of TCGA data indicates that genedosage gain of CSNK2A1 and heterozygous deletion of

NFKBIA show a pattern of mutual exclusion (not shown).Deletion of the NFKBIA gene is associated with the non-classical subtypes of glioblastoma (5) and CSNK2A1 genedosage gains are significantly associated with the classicalsubtype (Fig. 1). Considering that both alterations canlead to constitutive activation of NF-kB, these data suggestthat classical and non-classical tumors select for distinctgenetic aberrations that may have a similar effect inthe pathogenesis of glioblastoma, namely activation ofNF-kB.

Our results indicate that inhibition of CK2 causes cellrounding and actin disorganization, decreases adhesion toextracellular matrix components such as fibronectin, andsuppresses migration of glioblastomas. Expression of integ-rins including ITGB1 and ITGA4 depends on CK2 expres-sion or activity in glioblastoma cells, which is the firstassociation of CK2 and integrin expression. Interestingly,ITGA4 and ITGB1 are expressed at higher levels in humanglioblastoma samples compared with nontumor controls(RembrandtDatabase. Accessed2013 January). In addition,activation of ITGB1 increases the invasiveness of malignantglioma (49). Therefore, inhibitionof integrin expression is anovel mechanism by which CK2 inhibitors cause cell mor-phology changes and decrease cell adhesion andmigration.The molecular basis underlying how CK2 regulates integrinexpression is of interest given that ITGA4 and ITGB1 pro-moters contain binding sites for both STAT and NF-kBtranscription factors.

0.5

0.4

0.3

0.2

0.1

0

A B

D E

C

0.4

0.3

0.2

0.1

0

1.2

1

0.8

0.6

0.4

0.2

024 48 72 96 (h) (h) (h)24 48 72 96 24 48 72 96

0 1 100 1 10

100%

80%

60%

40%

Perc

enta

ge o

f cells

20%

0%

9080706050403020100

Annexin

V+

cells

(%

)

CX-4945 (μmol/L) CX-4945 (μmol/L)

O.D

.450

O.D

.450

O.D

.450

X1046 X1066 U251-MG

24 h

48 h

G2--M

S

G1Sub-G1

UT

CX-4945 (2 μmol/L)

CX-4945 (5 μmol/L)

CX-4945 (10 μmol/L)

CK2α siRNA

NT siRNA

CK2α’ siRNA

CK2α+α’ siRNA

Figure 5. Inhibition of CK2 suppresses glioblastoma (GBM) cell survival. X1046 (A) and X1066 (B) cells were treated with CX-4945, and cell survivalwas measured by the WST-1 assay. C, U251-MG cells were transfected with 100 nmol/L of non-target, CK2a, or CK2a' siRNAs, or CK2a (50 nmol/L) plusCK2a' (50 nmol/L) siRNAs for 48 hours, and survival was determined by the WST-1 assay. Triplicate experiments and error bars show � S.D. �, P < 0.05.D, U251-MG cells were treated with CX-4945 for 24 or 48 hours. Cells were stained with Annexin V and propidium iodide and examined by flow cytometry.Triplicate experiments and error bars show � S.D. �, P < 0.05. E, U251-MG cells were treated with CX-4945 for 24 hours, fixed overnight, stained withpropidium iodide, and digested with RNase. The percentage of cells in the sub-G1, G1, S, and G2–M phase was examined by flow cytometry.






Inhibition of CK2 activity by CX-4945 suppresses in vivoglioblastoma xenograft growth and promotes survival. In aglioblastoma xenograft flank model, administration of CX-4945 inhibits activation of STAT-3, NF-kB p65, and AKT,and suppresses tumor growth. Importantly, treatment withCX-4945 inhibits activation of STAT-3,NF-kBp65, and AKTin intracranial glioblastoma tumors and promotes the sur-vival of mice bearing intracranial human glioblastomatumors. These findings describe the first use of any CK2inhibitor in orthotopic models of glioblastoma using pri-mary human glioblastoma xenografts. CK2 is a remarkablynodal kinase, and its upregulation in glioblastoma has astrong impact on cellular processes indispensable for cancercell survival. Targeting this pleiotropic kinase that influ-ences multiple signaling cascades involved in glioblastomaprogression may prove more effective than strategies thattarget a single pathway.

Disclosure of Potential Conflicts of InterestD.Drygin is employed (other than primary affiliation; e.g., consulting) as

a Vice President, Biology inCylene Pharmaceuticals. Nopotential conflicts ofinterest were disclosed by the other authors.

Authors' ContributionsConception and design: Y. Zheng, E.N. BenvenisteDevelopment of methodology: Y. Zheng, S.L. Bellis, M. BredelAcquisitionofdata (provided animals, acquired andmanagedpatients,provided facilities, etc.): Y. Zheng, B.C. McFarland, H. Yu, M. BredelAnalysis and interpretation of data (e.g., statistical analysis, biosta-tistics, computational analysis): Y. Zheng, H. Yu, H. Kim, M. BredelWriting, review, and/or revision of themanuscript: Y. Zheng, D. Drygin,H. Kim, M. Bredel, E.N. BenvenisteAdministrative, technical, or material support (i.e., reporting or orga-nizing data, constructing databases): Y. Zheng, B.C. McFarland, H. KimStudy supervision: E.N. Benveniste

AcknowledgmentsThe authors thankDr. YanceyGillespie (UAB) for providing glioblastoma

xenografts,Dr. Fang-Tsyr Lin (Baylor) for helpful discussion, and Tanvi MSinha, Bing Wang, and Dr. Jianbo Wang (UAB) for assistance with imaging.

30

25

20

15

10

5

0

0 3 5 7 10 13 17 20 24 27 31 34 38 40

3 5 7 10 13 17 20 24 27 31 34 38 40Time (d)

Time (d)

Weig

ht (g

)

Vehicle

CX-4945

Vehicle

CX-4945

Vehicle CX-4945

Vehicle

CX-4945

900

800

700

600

500

400

300

200

100

0

X1046 flank

A

E F

G

B C

DTum

or

siz

e (

mm

3)

i.p. oral

P-Y-STAT3

STAT3

p65-S529

p65-S536

p65

AKT-S129

AKT-S473

AKT

CK2α

ACTIN

p65-S529

p65-S536

p65

AKT-S129

AKT-S473

AKT

GAPDH

2.5

2

1.5

1

0.5

0

Arb

itra

ry u

nits

WBC NE LY MO EO BA RBC Hb

CX-4945 1(9.3%)

CX-4945 2

(8.2%)

Vehicle 1(11.8%)

Vehicle 2(10.6%)

Vehicle CX-4945

P-Y-STAT3

STAT3

1 2 3 1 2 3

X1046 Intracranial Tumor

Gavage, bid

CX-4945 (75 mg/kg), n = 15Vehicle, n = 14

Log-rank P < 0.001

1.0

0.8

0.6

0.4

0.2

0.00 20 40 60 80

Survival, days after tumor implantation

Pro

port

ion s

urv

ivin

g

n.s.n.s.

Figure 6. CX-4945 inhibits in vivo growth of xenograft glioblastoma tumors. A–D, xenograft X1046 was injected subcutaneously into nude mice. Cages wererandomized and vehicle (n¼ 4) or 75mg/kg of CX-4945 (n¼ 4) was administered twice a day intraperitoneally from days 3 to 5, and by oral gavage twice a dayfrom day 6. Tumor size (A) and body weight (C) were measured on the indicated days. On day 40, all mice were euthanized. Data represent mean � SEM.�, P < 0.05; ��, P < 0.005; ��, P < 0.001. B, tumors were homogenized and lysates immunoblotted with indicated antibodies. D, blood from vehicle andCX-4945–treated mice was obtained by cheek bleeding before euthanasia on day 40, and analyzed with HEMAVET950. The numbers of white blood cells,neutrophils, lymphocytes, monocytes, eosinophils, basophils, red blood cells, and hemoglobin are shown in arbitrary units. n.s., not significant. E,a total of 5� 105 cells/5 mL of xenograft X1046 were injected intracranially. Cages were randomized and vehicle (n¼ 14) or 75mg/kg of CX-4945 (n¼ 15) wasadministered orally twice a day for 28 days starting at day 5. Survival was monitored and mice were euthanized upon moribund. Kaplan–Meier survival withLog-Rank analysis was performed. F, a total of 5 � 105 cells/5 mL of xenograft X1016 were injected intracranially. Cages were randomized, vehicle(n¼ 3) or 75mg/kg of CX-4945 (n¼ 3) was administered orally twice a day starting at day 3, andmice were euthanized between days 13 and 15. Tumors werehomogenized and lysates immunoblotted with indicated antibodies. G, a total of 5 � 105 cells/5 mL of X1016 were injected intracranially. Cages wererandomized and vehicle (n ¼ 2) or 75 mg/kg of CX-4945 (n ¼ 2) was administered orally twice a day starting at day 3, and mice were euthanized at day 18.Brain tissue from vehicle and CX-4945–treated mice was fixed in formalin and embedded in paraffin, and sections were stained with anti-Ki-67 antibody andcounterstained with hematoxylin. Representative images and Ki-67þ percentages (in brackets) are shown. The scale bar is 50 mm.

Zheng et al.





We thank theUABNeuroscienceMolecular Detection Core (P30NS047466)for assistance with immunohistochemistry.

Grant SupportThis work was supported by NIH grant CA158534 (to E.N. Benveniste), a

grant from the Southeastern Brain Tumor Foundation (to E.N. Benveniste),and a grant from the American Brain Tumor Association in honor of PaulFabbri (to B.C. McFarland).

The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely to indicatethis fact.

Received January 28, 2013; revised August 2, 2013; accepted September 5,2013; published OnlineFirst September 13, 2013.

References1. Verhaak RGW, Hoadley KA, Purdom E, Wang V, Qi Y, Wilkerson MD,

et al. Integrated genomic analysis identifies clinically relevant subtypesof glioblastoma characterized by abnormalities in PDGFRA, IDH1,EGFR, and NF1. Cancer Cell 2010;17:98–110.

2. Stupp R, Mason WP, van den Bent MJ, Weller M, Fisher B, TaphoornMJ, et al. Radiotherapy plus concomitant and adjuvant temozolomidefor glioblastoma. N Engl J Med 2005;352:987–96.

3. Dunn GP, Rinne ML, Wykosky J, Genovese G, Quayle SN, Dunn IF,et al. Emerging insights into the molecular and cellular basis ofglioblastoma. Genes Dev 2012;26:756–84.

4. CarroMS, LimWK, AlvarezMJ, Bollo RJ, Zhao X, Snyder EY, et al. Thetranscriptional network for mesenchymal transformation of braintumours. Nature 2010;463:318–25.

5. Bredel M, Scholtens DM, Yadav AK, Alvarez AA, Renfrow JJ, ChandlerJP, et al. NFKBIA deletion in Glioblastomas. N Engl J Med 2011;364:627–37.

6. Doucette TA, Kong LY, Yang Y, Ferguson SD, Yang J, Wei J, et al.Signal transducer and activator of transcription 3 promotes angiogen-esis and drives malignant progression in glioma. Neuro Oncol2012;14:1136–45.

7. Atkinson GP, Nozell SE, Benveniste ET. NF-kappaB and STAT3signaling in glioma: targets for future therapies. Expert Rev Neurother2010;10:575–86.

8. Brantley EC,Nabors LB,GillespieGY,Choi YH, PalmerCA,HarrisonK,et al. Loss of protein inhibitors of activated STAT-3 expression inglioblastoma multiforme tumors: implications for STAT-3 activationand gene expression. Clin Cancer Res 2008;14:4694–704.

9. Lo HW, Cao X, Zhu H, Ali-Osman F. Constitutively activated STAT3frequently coexpresses with epidermal growth factor receptor in high-grade gliomas and targeting STAT3 sensitizes them to Iressa andalkylators. Clin Cancer Res 2008;14:6042–54.

10. McFarland BC,Ma JY, LangfordCP, Gillespie GY, YuH, ZhengY, et al.Therapeutic potential of AZD1480 for the treatment of human glio-blastoma. Mol Cancer Ther 2011;10:2384–93.

11. Raychaudhuri B, Han Y, Lu T, Vogelbaum MA. Aberrant constitutiveactivation of nuclear factor kB in glioblastoma multiforme drivesinvasive phenotype. J Neuro Oncol 2007;85:39–47.

12. Zhao X, Laver T, Hong S, Twitty G, DeVos A, DeVosM, et al. An NF-kBp65-cIAP2 link is necessary formediating resistance to TNF-a inducedcell death in gliomas. J Neuro Oncol 2011;102:367–81.

13. Duncan JS, Litchfield DW. Too much of a good thing: The role ofprotein kinase CK2 in tumorigenesis and prospects for therapeuticinhibition of CK2. Biochim Biophys Acta 2008;1784:33–47.

14. Ruzzene M, Pinna LA. Addiction to protein kinase CK2: A commondenominator of diverse cancer cells?Biochim Biophys Acta 2010;1804:499–504.

15. ZhengY, QinH, Stuart F, Deng L, Litchfield DW, Terfferi A, et al. ACK2-dependent mechanism for activation of the JAK-STAT signaling path-way. BLOOD 2011;118:156–66.

16. Drygin D, Ho CB, Omori M, Bliesath J, Proffitt C, Rice R, et al. Proteinkinase CK2 modulates IL-6 expression in inflammatory breast cancer.Biochem Biophysical Res Commun 2011;415:163–7.

17. Siddiqui-Jain A, Drygin D, Streiner N, Chua P, Pierre F, O'Brien SE,et al. CX-4945, an orally bioavailable selective inhibitor of proteinkinase CK2, inhibits prosurvival and angiogenic signaling and exhibitsantitumor efficacy. Cancer Res 2011;70:10288–98.

18. Pierre F, Chua PC, O'Brien SE, Siddiqui-Jain A, Bourbon P, HaddachM, et al. Discovery and SAR of 5-(3-chlorophenylamino)ben-

zo[c][2,6]naphthyridine-8-carboxylic acid (CX-4945), the first clinicalstage inhibitor of protein kinaseCK2 for the treatment of cancer. JMedChem 2011;54:635–54.

19. Pierre F, Chua PC, O'Brien SE, Siddiqui-Jain A, Bourbon P, HaddachM, et al. Pre-clinical characterization of CX-4945, a potent and selec-tive small molecule inhibitor of CK2 for the treatment of cancer. MolCell Biochem 2011;356:37–43.

20. Marschke RF, Borad MJ, McFarland RW, Alvarez RH, Lim JK,Padgett CS, et al. Findings from the phase I clinical trials of CX-4945, an orally available inhibitor of CK2. ASCO Meeting Abstracts2011;29:3087.

21. Ladha J, Donakonda S, Agrawal S, Thota B, Srividya MR, Sridevi S,et al.Glioblastoma-specificprotein interaction network identifiesPP1Aand CSK21 as connecting molecules between cell cycle-associatedgenes. Cancer Res 2010;70:6437–47.

22. Kaminska B, Ellert-Miklaszewska A, Oberbek A,Wisniewski P, Kaza B,Makowska M, et al. Efficacy and mechanism of anti-tumor action ofnew potential CK2 inhibitors toward glioblastoma cells. Int J Oncol2009;35:1091–100.

23. Prudent R, Moucadel V, Nguyen CH, Barette C, Schmidt F, Florent JC,et al. Antitumor activity of pyridocarbazole and benzopyridoindolederivatives that inhibit protein kinase CK2. Cancer Res 2010;70:9865–74.

24. Moucadel V, Prudent R, Sautel CF, Teillet F, Barette C, Lafanechere L,et al. Antitumoral activity of allosteric inhibitors of protein kinase CK2.Oncotarget 2011;2:997–1010.

25. Olsen BB, Issinger OG, Guerra B. Regulation of DNA-dependentprotein kinase by protein kinase CK2 in human glioblastoma cells.Oncogene 2010;29:6016–26.

26. Mermel CH, Schumacher SE, Hill B, MeyersonML, Beroukhim R, GetzG. GISTIC2.0 facilitates sensitive and confident localization of thetargets of focal somatic copy-number alteration in human cancers.Genome Biol 2011;12:R41.

27. WangH, Lathia JD,WuQ,Wang J, Li Z,Heddleston JM, et al. TargetingInterleukin 6 signaling suppresses glioma stem cell survival and tumorgrowth. Stem Cells 2009;27:2393–404.

28. Guryanova OA, Wu Q, Cheng L, Lathia JD, Huang Z, Yang J, et al.Nonreceptor tyrosine kinase BMX maintains self-renewal and tumor-igenic potential of glioblastoma stem cells by activating STAT3. Can-cer Cell 2011;19:498–511.

29. Cao Y, Lathia JD, Eyler CE, Wu Q, Li Z, Wang H, et al. Erythropoietinreceptor signaling through STAT3 is required for glioma stem cellmaintenance. Genes Cancer 2010;1:50–61.

30. Puram SV, Yeung CM, Jahani-Asl A, Lin C, de la Iglesia N, Konopka G,et al. STAT3-iNOS signaling mediates EGFRvIII-induced glial prolifer-ation and transformation. J Neurosci 2012;32:7806–18.

31. Van Wagoner N, Oh J-W, Repovic P, Benveniste EN. IL-6 productionby astrocytes: autocrine regulation by IL-6 and the soluble IL-6 recep-tor. J Neurosci 1999;19:5236–44.

32. Lou DY, Dominguez I, Toselli P, Landesman-Bollag E, O'Brien C,SeldinDC. Thealpha catalytic subunit of protein kinaseCK2 is requiredfor mouse embryonic development. Mol Cell Biol 2008;28:131–9.

33. Seldin DC, Landesman-Bollag E, Farago M, Currier N, Lou D, Dom-inguez I. CK2 as a positive regulator of Wnt signalling and tumour-igenesis. Mol Cell Biochem 2005;274:63–7.

34. Repovic P, Fears CY, Gladson CL, Benveniste EN. Oncostatin-Minduction of vascular endothelial growth factor expression in astro-glioma cells. Oncogene 2003;22:8117–24.






35. Joughin BA, Naegle KM, Huang PH, Yaffe MB, Lauffenburger DA,White FM. An integrated comparative phosphoproteomic and bio-informatic approach reveals a novel class of MPM-2 motifs upre-gulated in EGFRvIII-expressing glioblastoma cells. Mol Biosyst2009;5:59–67.

36. Gadji M, Crous AM, Fortin D, Krcek J, Torchia M, Mai S, et al. EGFreceptor inhibitors in the treatment of glioblastoma multiform: oldclinical allies and newly emerging therapeutic concepts. Eur J Phar-macol 2009;625:23–30.

37. Quesnelle KM, Boehm AL, Grandis JR. STAT-mediated EGFR signal-ing in cancer. J Cell Biochem 2007;102:311–9.

38. Wang D, Westerheide SD, Hanson JL, Baldwin AS Jr. Tumor necrosisfactor a-induced phosphorylation of RelA/p65 on Ser529 is controlledby casein kinase II. J Biol Chem 2000;275:32592–7.

39. Network, T.C.G.A.R. Comprehensive genomic characterizationdefines human glioblastoma genes and core pathways. Nature2008;455:1061–8.

40. Torres J, Pulido R. The tumor suppressor PTEN is phosphorylated bythe protein kinase CK2 at its C terminus. J Biol Chem 2001;276:993–8.

41. Di Maira G, Salvi M, Arrigoni G, Marin O, Sarno S, Brustolon F, et al.Protein kinase CK2 phosphorylates and upregulates Akt/PKB. CellDeath Differ 2005;12:668–77.

42. Canton DA, Litchfield DW. The shape of things to come: an emergingrole for protein kinase CK2 in the regulation of cell morphology and thecytoskeleton. Cell Signal 2006;18:267–75.

43. KramerovAA,GolubAG,Bdzhola VG,YarmolukSM,AhmedK,BretnerM, et al. Treatment of cultured human astrocytes and vascular endo-thelial cells with protein kinaseCK2 inhibitors induces early changes incell shape and cytoskeleton. Mol Cell Biochem 2011;349:125–37.

44. Garcia-Pardo A, Wayner EA, Carter WG, Ferreira OC Jr. Human Blymphocytes define an alternative mechanism of adhesion to fibro-nectin. The interaction of the alpha 4 beta 1 integrin with theLHGPEILDVPST sequence of the type III connecting segment issufficient to promote cell attachment. J Immunol 1990;144:3361–6.

45. Bredel M, Scholtens DM, Harsh GR, Bredel C, Chandler JP, RenfrowJJ, et al. A network model of a cooperative genetic landscape in braintumors. JAMA 2009;302:261–75.

46. Piazza FA, Ruzzene M, Gurrieri C, Montini B, Bonanni L, Chioetto G,et al. Multiple myeloma cell survival relies on high activity of proteinkinase CK2. Blood 2006;108:1698–707.

47. Kohsaka S, Wang L, Yachi K, Mahabir R, Narita T, Itoh T, et al. STAT3inhibition overcomes temozolomide resistance in glioblastoma bydownregulating MGMT expression. Mol Cancer Ther 2012;11:1289–99.

48. Miyata Y. Protein kinase CK2 in health and disease: CK2: the kinasecontrolling the Hsp90 chaperone machinery. Cell Mol Life Sci2009;66:1840–9.

49. Edwards LA, Woolard K, Son MJ, Li A, Lee J, Ene C, et al. Effect ofbrain- and tumor-derived connective tissue growth factor on gliomainvasion. J Natl Cancer Inst 2011;103:1162–78.

Zheng et al.





2013;19:6484-6494. Published OnlineFirst September 13, 2013.Clin Cancer Res Ying Zheng, Braden C. McFarland, Denis Drygin, et al. Pathways and Growth of GlioblastomaTargeting Protein Kinase CK2 Suppresses Prosurvival Signaling

Updated version

10.1158/1078-0432.CCR-13-0265doi:

Access the most recent version of this article at:

Material

Supplementary

http://clincancerres.aacrjournals.org/content/suppl/2013/12/02/1078-0432.CCR-13-0265.DC2

http://clincancerres.aacrjournals.org/content/suppl/2013/09/13/1078-0432.CCR-13-0265.DC1Access the most recent supplemental material at:

Cited articles

http://clincancerres.aacrjournals.org/content/19/23/6484.full#ref-list-1

This article cites 49 articles, 16 of which you can access for free at:

Citing articles

http://clincancerres.aacrjournals.org/content/19/23/6484.full#related-urls

This article has been cited by 8 HighWire-hosted articles. Access the articles at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected]

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://clincancerres.aacrjournals.org/content/19/23/6484To request permission to re-use all or part of this article, use this link



http://clincancerres.aacrjournals.org/lookup/doi/10.1158/1078-0432.CCR-13-0265

http://clincancerres.aacrjournals.org/content/suppl/2013/09/13/1078-0432.CCR-13-0265.DC1 http://clincancerres.aacrjournals.org/content/suppl/2013/12/02/1078-0432.CCR-13-0265.DC2

http://clincancerres.aacrjournals.org/content/suppl/2013/09/13/1078-0432.CCR-13-0265.DC1 http://clincancerres.aacrjournals.org/content/suppl/2013/12/02/1078-0432.CCR-13-0265.DC2

http://clincancerres.aacrjournals.org/content/19/23/6484.full#ref-list-1

http://clincancerres.aacrjournals.org/content/19/23/6484.full#related-urls

http://clincancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://clincancerres.aacrjournals.org/content/19/23/6484


Documents

Targeting Protein Kinase CK2 Suppresses Prosurvival ...clincancerres.aacrjournals.org/content/clincanres/19/23/6484.full.pdf · Targeting Protein Kinase CK2 Suppresses Prosurvival