a role for homologous recombination and abnormal cell-cycle … · therapeutic discovery a role for...

Therapeutic Discovery

A Role for Homologous Recombination and AbnormalCell-Cycle Progression in Radioresistance ofGlioma-Initiating Cells

Yi Chieh Lim1,2, Tara L. Roberts1,2, Bryan W. Day1, Angus Harding3, Sergei Kozlov1, Amanda W. Kijas1,Kathleen S. Ensbey1, David G. Walker4, and Martin F. Lavin1,2

AbstractGlioblastomamultiforme (GBM) is themost common formofbrain tumorwith apoorprognosis and resistance

toradiotherapy.Recentevidencesuggests thatglioma-initiatingcellsplayacentral role inradioresistance through

DNAdamage checkpoint activation and enhanced DNA repair. To investigate this inmore detail, we compared

the DNA damage response in nontumor forming neural progenitor cells (NPC) and glioma-initiating cells

isolated fromGBMpatient specimens.Asobserved forGBMtumors, initial characterization showed that glioma-

initiating cells have long-term self-renewal capacity. They expressmarkers identical toNPCsandhave the ability

to form tumors in an animalmodel. In addition, these cells are radioresistant to varying degrees,which could not

be explained by enhanced nonhomologous end joining (NHEJ). Indeed, NHEJ in glioma-initiating cells was

equivalent, or in some cases reduced, as compared with NPCs. However, there was evidence for more efficient

homologous recombination repair in glioma-initiating cells. We did not observe a prolonged cell cycle nor

enhancedbasal activationof checkpointproteins as reportedpreviously.Rather, cell-cycledefects in theG1–Sand

S-phase checkpoints were observed by determining entry into S-phase and radioresistant DNA synthesis

following irradiation. These data suggest that homologous recombination and cell-cycle checkpoint abnormal-

itiesmaycontribute to the radioresistanceofglioma-initiating cells and thatbothprocessesmaybesuitable targets

for therapy. Mol Cancer Ther; 11(9); 1863–72. �2012 AACR.

IntroductionGliomas are the most common adult brain cancers,

comprising 80% of diagnosed cases (1). Anaplastic astro-cytoma (grade III) and glioblastoma multiforme (GBM;grade IV) tumors are highly lethal (1). Standard therapy iscombined local aggressive cytoreductive surgery followedby radiotherapy. Unfortunately, GBM are particularlyresistant to conventional treatment, leading to recurrenceof brain tumors. A pool of glioma-initiating cells (2), simi-lar to neural progenitor cells (NPC; ref. 3), can surviveexogenousdamage, suchas the lethaldouble strandbreaks(DSB), to repopulate tumor cells following treatment.DSBs are highly detrimental to the structural integrity

of chromosomes. To counteract this damage, normal and

cancer cells initiate the DNA-damage response. DSBs aresensed primarily by theMRN complex; Mre11, Rad50, andNbs1 proteins that rapidly relocates to the site of damageand recruits ATM (ataxia-telangiectasia mutated) kinase,resulting in downstream phosphorylation cascades thatinitiate cell-cycle arrest and DNA repair (4). There are 2major DSB pathways engaged during DNA repair. Non-homologous end joining (NHEJ) is initiated predominantlyduring the G1-phase of the cell cycle and because themajority of cells are in G1-phase, NHEJ is considered themajor pathway for DSB repair. The second repair mecha-nism occurs during late S- and G2-phases. Homologousrecombination is an error-freemethod of repair using sisterchromatids as templates to replace damaged DNA (5). Incontrast,NHEJreliesonligasesandexcisionrepairenzymestoadherebrokenDNAends thatcan introducespontaneousmutations (6).Dependingontheseverityof insult, cellswitha high mutational burden may initiate apoptosis to mini-mize damage to the genome and prevent mutations frombeing maintained through subsequent cell divisions.

Compared with normal cells, cancer cells haveenhanced DNA repair pathways, which confer greatersurvival. Investigations characterizing subpopulations ofaberrant stem cells have shown efficient DSB repairthrough various DNA damage markers, but no specificpathways have been identified to explain the increasedsurvival (7). Recent reports also provided evidence fordifferences in radiosensitivity between subpopulations

Authors' Affiliations: 1Queensland Institute of Medical Research; 2Uni-versity of Queensland Centre for Clinical Research, Royal Brisbane Hos-pital Campus, Herston; 3University of Queensland, Diamantina Institute,Level 4, Princess Alexandra Hospital, Woollongabba; and 4Briz Brain andSpine, Level 10, The Wesley Hospital, Chasely Street, Auchenflower,Queensland, Australia

Note: Supplementary data for this article are available at Molecular CancerTherapeutics Online (http://mct.aacrjournals.org/).

Corresponding Author: Martin F. Lavin, Locked Bag 200 Royal BrisbaneHospital, Herston, Queensland 4029, Australia. Phone: 61-733620341;Fax: 61-733620106; E-mail: [email protected]

doi: 10.1158/1535-7163.MCT-11-1044

�2012 American Association for Cancer Research.

MolecularCancer

Therapeutics

www.aacrjournals.org 1863

on August 24, 2021. © 2012 American Association for Cancer Research. mct.aacrjournals.org Downloaded from

Published OnlineFirst July 6, 2012; DOI: 10.1158/1535-7163.MCT-11-1044

http://mct.aacrjournals.org/

of glioma-initiating cells (8–10), suggesting disparity inDNA repair efficiency. Whether DNA repair is a majormechanism of radioresistance in glioma-initiating cellsremains controversial. To further understand these DNArepair pathways, we compared the DNA damageresponse to ionizing radiation (IR) in nontumor formingNPCs (11, 12) and glioma-initiating cells isolated fromGBM tumors. The significance of this comparison is thepotential for identifying key proteins or pathways uniqueto glioma-initiating cells. In our study, increased survivalof glioma-initiating cells following IR was observed.Although, glioma-initiating cells had a higher proportionof surviving cells, attenuated checkpoint kinase activationsuggested inadequate cell-cycle arrest at G1–S, allowing aportion of G1 cells to enter S-phase. NHEJ activity wasreduced in several glioma-initiating cells lines, but thesedifferences did not account for altered survival. Interest-ingly, glioma-initiating cells seemed to favor the homol-ogous recombination repair pathway. These data indicatethat glioma-initiating cells have an ineffectiveG1–S check-point allowing damaged cells to enter S-phase and usehomologous recombination to repair the genome.

Material and MethodsGlioma-initiating cells extraction and maintenance

Glioma-initiating cells enrichment from tumors from 3GBM patients was carried out as described by Piccirilloand colleagues (13); resulting lines were designated asL1b, L2b, and L3b. Briefly, tumors were dissociated intosingle-cell suspensions before enriching for stem cells/neurospheres in serum-free media supplemented with20 ng/mL EGF and 10 ng/mL basic fibroblast growthfactor. U251 and U87 neurosphere cultures were obtainedfrom Day and colleagues (14) and also underwent theglioma-initiating cells enrichment process. No authentica-tion was carried out by the authors. Neurosphere culturesweremaintained at 37�C in 5%CO2with supplements (12)for 4 to 8 weeks before characterization, as described inSupplementary Methods. The neurosphere media is aselective culture that enriches for progenitors and precur-sors, but notdifferentiated cells.All primary and tumor celllines grown in neurosphere media will contain a mixedpopulation of progenitor and precursor cells. Throughoutthis article, we will refer to the entire population of cellswithin the cultures as glioma-initiating cells.

Nontumorigenic NPCs derived from fetal brain tissueare commercially available (ReNcell; Millipore) and werepreviously characterized by Reynolds and colleagues (11,12, 15). For experiments, neurosphere cultures weregrown for 5 days, before trypsinizing with 0.25% tryp-sin-EDTA (Invitrogen) in PBS into single-cell suspensionsbefore all indicated treatments (12).

Survival assayDissociated neurospheres were prepared as duplicates

before IR. Protocol was previously described (16). Detailsare in Supplementary Methods.

Nonhomologous end joining assayThe NHEJ reporter plasmid pEGFP-N1 (Clontech) was

digestedwithHindIII (NewEnglandBiolabs) andpurifiedby gel extraction kit (Qiagen), and aliquots analyzed bygel electrophoresis to confirm complete digestion. Neuro-sphere lines were transfected using a Gene PulserII(Biorad) at 220 V, 960 mF, infinite resistance. A total of1 � 106 dissociated cells were transfected with 1 mg oflinearized pEGFP-N1, and in parallel with 1 mg circular-ized pEGFP-N1 as control for transfection efficiency. GFPexpression was measured by FACScan (Becton Dickin-son) 48 hours later.

DNA pulse labelingIododeoxyuridine (IdU) analog (100 mm) was added

into neurosphere media. Cells were placed for 30minutesat 37�C before fixation and cytospinning onto Superfrostplus slides (see Supplementary Methods). Slides wereplaced for 1 minute in lysis buffer (50 mmol/L Tris-HClpH8.0, 1 mmol/L EDTA, 0.1% SDS) at room temperature.Slides were rinsed with PBS before incubating in 1 mol/LHCl, for 5minutes on ice, followed by 2mol/LHCl, for 10minutes at room temperature and 10 minutes at 37�C.Immunolabeling is described in SupplementaryMethods.

Homologous recombination assayNeurosphere lines were stably transfected as described

abovewith 1mgof pDR-GFPand stable integrants selectedwith 1 to 8 mg/mL puromycin (Sigma; ref. 17). Analysiswas carried out 48 hours after I-Sce I�/þ transfection aspreviously indicated (18).

Fiber assay for DNA replicationThe DNA fiber assay was conducted as previously

described (19). Total DNA fibers were scored (�300)for each cell line. The percentage of new initiations wasexpressed as fold change [new initiations / (continuingþnew initiations)].

ResultsLong-termgrowthpotential of glioma-initiating cellsand NPCs

The gold standard to determine the presence of normalstem or progenitor cells is to evaluate their self-renewaland lineage differentiation (20, 21). Assays for glioma-initiating cells additionally also require determination oftumor formation in an animal model. We first examinedwhether glioma-initiating cells had stem cell markers andself-renewal capacity comparable with NPCs. The profileof stem cell markers differed between glioma-initiatingcells cultures, but U251 and L1b showed near identicalmarker expression toNPCs (Supplementary Fig. S1A).Allcultures exceptU87displayedCD49f expression (22). Twoother stem cellmarkers, Sox2 andNestin,were detected inall glioma-initiating cells and NPCs with 80% to 95% and78% to 93% positive cells, respectively, by IF (Supplemen-tary Fig. S1A). When glioma-initiating cells and NPCswere platedwith 5% fetal calf serum, both cell types began

Lim et al.

Mol Cancer Ther; 11(9) September 2012 Molecular Cancer Therapeutics1864




to differentiate within 5 to 7 days (Supplementary Fig.S2A). Neural (b-III tubulin) and glial fibrillary acidicprotein [(GFAP) and mitogen-activated protein (MAP)]makers were analyzed 2 weeks later. Glioma-initiatingcells expressed all lineage markers with percentageexpression being 17% to 40% (b-III tubulin), 17% to 31%(MAP), and 18% to 41% (GFAP) across all glioma-initiat-ing cells cultures. Lineage distribution in differentiatedNPCwas 31% (b-III tubulin), 18% (GFAP), and17% (MAP;Supplementary Fig. S2B and S2C). As surface markerexpression is only one method to define stem cell popula-tions, we also evaluated long-term self-renewal usingserial dilution. Single cells were seeded from passage 1,and expansion between passages was measured (Supple-mentaryFig. S1B).All lines expanded constantly as shownby an unchanged slope of a straight line.Finally, to show that glioma-initiating cells could

recapitulate tumors in vivo, we used a Scid/Nod tumor-igenesis model (23). Intracranially injected tumor neu-rospheres (U251) and primary glioma (L2b) produced

glioblastoma in mice (Supplementary Fig. S1C). Surviv-al curves are shown in Supplementary Fig. S1D. Despitelimited CD133 expression, they were capable of formingtumors. This finding is consistent with recent publica-tions showing that both CD133-positive and CD133-negative glioma-initiating cells can induce tumor devel-opment (24). Histologic analysis showed tumors withhighly dense region of marked nuclear atypia andpseudopalisades-like formation in surrounding tissue,all characteristic of GBM (Supplementary Fig. S1C).

Glioma-initiating cells cultures are radioresistantWe next determined whether glioma-initiating cells

cultures had radioresistant properties (25). Following 2GyIR exposure, cell survival was determined by trypan blueassay (19). All glioma-initiating cells cultures had moreviable cells over time compared with NPCs (Fig. 1A; P <0.05). Survival for glioma-initiating cells cultures rangedfrom 40% to 60% 96 hours after IR versus 20% for NPCs.glioma-initiating cells cultures also showed increased

Figure 1. Survival in glioma-initiating cells depends onproliferation. The difference in cellcount at each time point isnormalized back to its own individualunirradiated sample and viability isexpressed as a percentage. A andB, time-dependent (A) and dose-dependent (B) survival ofnontumorigenic and tumorigeniccultures after irradiation. C, glioma-initiating cells total apoptosis(percentage at 24 and 48 hours) after2Gy irradiation. Columns represent 2independent experiments combined.D, proliferation rate of neurospherecultures determined by CFSEstaining. Single cells were untreated(0Gy) or treatedwith indicated dosesof IR and analyzed by flow cytometry96 hours later. Color with therespective numeric symbolrepresents the number ofgenerations each line hasundergone.

DNA Damage Responses in Glioma-Initiating Cells

www.aacrjournals.org Mol Cancer Ther; 11(9) September 2012 1865




survival with doses from 2 to 10 Gy (Fig. 1B). Alternateanalysis using neurosphere size (23, 26) showed NPCsafter 2 Gy had a 0.51-fold reduction in size (Supplemen-tary Fig. S3A). In comparison, glioma-initiating cells had a0.89- to 0.66-fold reduction, suggesting a lesser impact ofIR-induced damage on glioma-initiating cells growth (P <0.01). At 5Gy dose,NPCs reduced to 0.38 the original size,whereas the fold change in glioma-initiating cells sizewas0.76 to 0.51 (P < 0.01).

We next determined whether glioma-initiating cellssurvival was due to accelerated repopulation or apoptosisresistance (27, 28). Radiation-induced DNA damagecaused comparable cell death in glioma-initiating cellsand NPCs at 24 and 48 hours post-IR (Fig. 1C and Sup-plementary Fig. S3B). We next used CFSE labeling tomeasure proliferation. CFSE incorporates into the mem-brane and daughter cells inherit half of the dye aftermitosis. Individual colors illustrate cell divisions (gen-eration numbers; Fig. 1D). All glioma-initiating cellsproliferated at higher rates than NPCs following irra-diation indicating that continuous proliferation ratherthan apoptosis resistance in glioma-initiating cellsaccounted for increased survival. In detail, 96 hourspost-2 Gy IR showed reduced growth with NPC under-going 5 to 6 divisions. U251, L2b, and L3b showed littlechange in proliferation with U251 and L2b capable of 6to 7 divisions and L3b 5 divisions. At 5 Gy dose, U251underwent 7 to 8 divisions, L2b and L3b 4 to 5 divisions,whereas NPC proliferation reduced to 3 to 4 divisions.At high dose (10 Gy), U251, L2b, and L3b cells prolif-erated for 4 to 6 divisions versus 2 to 3 for NPCs.

For long-term proliferation after 2 Gy irradiation, serialpassage showedmarked reduction in NPCs during initialpassages (1–2) before returning to normal growth frompassage 3 (Supplementary Fig. S4A). After 5 Gy reductioningrowthwasobserved inglioma-initiating cells lines, butwith faster recovery. A reduction in growthwas observedin NPCs with recovery from passage 5 (SupplementaryFig. S4B). Conversely, L2b, U251, and U87 showed recov-ery from passage 3, followed by L1b and L3b at passage 4.

Delayed DSB repair in glioma-initiating cells duringearly recovery

The survival, serial passage, and neurosphere formationdata indicated that glioma-initiating cells might haveenhanced DNA repair, allowing rapid growth post-IR. Toexamine the DNA damage repair response, the levels ofproteins central to DSB recognition and response weremeasured by immunoblotting. Interestingly, U87 andL3b had reduced protein expression compared with NPCs(Fig. 2A). InU87, totalATMandMre11 levelswere reduced(29). Similarly, protein levels ofMRN complexmembers inL3b were diminished. In contrast, L1b and L2b displayedincreased MRN protein levels, whereas no significant dif-ference was identified betweenNPCs and U251. Radiationdid not cause any consistent change in levels of DNAdamage response proteins that might account for changesin radioresistance. We next investigated whether differ-

ences in the efficiency of DNA DSB repair might explainradioresistance by following gH2AX foci formation andresolution (Fig. 2B). Because themaximumfoci number at 1hour differed between cultures, the rate of repair wasexpressed as fold change by normalizing to their ownneurosphere lines. In NPCs, efficient repair was identifiedby rapid reduction in gH2AX foci at 6 hours (reduced to17%;Fig. 2C).Conversely,glioma-initiating cells repairwasslow at early time points (<6 hours) ranging from 33% to73% (P< 0.01). Repair ofDNADSBwasnot completeduntil12 to 24 hours in glioma-initiating cells.

NHEJ function does not explain glioma-initiatingcells radioresistance

To investigate repairmechanisms responsible for radio-resistance, we measured NHEJ activity in vivo by recapit-ulating the joining of DNA ends with a linearized GFPexpression vector (Fig. 3A). NHEJ activitywas detected inall neurosphere cultures (Fig. 3B) but was variablebetween glioma-initiating cells cultures. In U251 andU87, NHEJ efficiency was reduced to 23% (P < 0.018) and19% (P < 0.024) compared with NPCs (46%). L2b hadmodestly reducedNHEJ activity to approximately 70% ofNPCs, whereas L3b was comparable with NPCs. L1b hadhigherNHEJ activity (P < 0.034) thanNPCs.As phosphor-ylation of DNA-PKcs correlates with NHEJ activation,DNA-PKcs protein levels and its phosphorylation weremeasured by immunoblotting. Total DNA-PKcs levels didnot change appreciably between glioma-initiating cellsand NPC cultures (Fig. 3C). However, DNA-PKcs phos-phorylation in U251 andU87 cultures was attenuated andconsistent with the reduced levels of NHEJ in these cells.

When treated with DNA-PKcs inhibitor (DNA-PKi) toinhibit NHEJ activity before IR (Fig. 3D), U251 and L2bcontinued to show better survival (Fig. 3D). U251 at 24hours showed no significant difference in viable cellsbetween DNA-PKi treated (76%) and IR only (79%). Evenat 96 hours, viable cells were similar (47% vs. 53%, respec-tively). Partial radiosensitivity occurred in L2b,withDNA-PKi treated (35%) and IR alone (46%) at 96 hours. Incontrast, NPCs showed increased radiosensitivity. At 24hours, DNA-PKi–treated NPCs showed 35% viability ver-sus IR alone (57%). Results at 96 hours were similar: DNA-PKi treated (11%)versus IRalone (21%)showing thatNPCsare more dependent on NHEJ than glioma-initiating cells.

Increased homologous recombination repair inglioma-initiating cells

Homologous recombination is the second major path-way involved in resolution of DSBs (30). Given thereduced levels of NHEJ for many glioma-initiating cells,we quantified homologous recombination using stablytransfected pDR-GFP neurosphere lines containing afunctional but interrupted GFP sequence with a singleDSB site that is inducible only by I-Sce I expression anda downstream complementary nonfunctional GFPsequence that is required for recombination (Fig. 4A;ref. 31). After I-Sce I expression to create a single DSB,

Lim et al.





GFP-positive cells were 0.74% for NPCs but homolo-gous recombination repair was approximately 4-foldhigher in U251 (3.76%, P < 0.0002) and L2b (3.94%, P< 0.0005). L3b had a less efficient homologous recom-bination repair (1.18%, P < 0.009). Homologous recom-bination efficiency was confirmed by measuring Rad51foci accumulation and disassembly (Fig. 4B and Sup-plementary Fig. S5; ref. 32). After IR, maximal numbersof Rad51 positive cells (53%–66%) were achieved by 6hours for L2b and U251, and followed by rapid lossbetween 12 to 24 hours. In contrast, NPCs showedgradual increase with Rad51-positive cells (26%) at 12hours and slow loss of foci (12%) at 36 hours, supportingslower homologous recombination activity. L3b wasintermediate between NPCs, and U251 and L2b (Fig.4B). Brca1, a second marker for homologous recombi-nation was also analyzed (Supplementary Fig. S6A;refs. 33, 34). Results showed that appearance and lossof Brac1 foci was similar to Rad51, again supportingrapid homologous recombination in glioma-initiatingcells (Supplementary Fig. S6B).

Following treatment with Rad51 siRNA (Fig. 4C), bothU251 and L2b showed a drastic decrease in cell viabilitypost-IR (Fig. 4D). At 24 hours, Rad51 siRNA-treated U251showed approximately 21% viable cells versus controlsiRNA treated (76%). By 96 hours, Rad51 siRNA-treatedU251 had very low cell viability (�10%) versus controlsiRNA (55%). L2b behaved similarly; 24 hours, Rad51siRNA L2b (26%) versus control siRNA (71%) and 96hours, Rad51 siRNA L2b (9%) versus control siRNA(39%). These results contrastedwithNPCs, inwhichRad51and control siRNA treated NPCs showed a similar per-centage of viable cells (56% vs. 55%, respectively, at 24hoursand15%vs. 23%at96hours). This clearly showedthegreater importance of homologous recombination for thesurvival of glioma-initiating cells compared with NPCs.

DNA synthesis inhibition after IR is deficient inglioma-initiating cells

Asglioma-initiating cells hadahigherpercentage of cellsundergoing homologous recombination, we hypothesizedglioma-initiating cells had unperturbed entry into S-phase

Figure 2. Ionizing radiation-inducedDNA damage resolves slowly inglioma-initiating cells. A, cells weretreated with and without 2 Gy IR asindicated (þor�) and harvested after1 hour. DNA damage responseprotein levels were determined byimmunoblotting. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH)was used as a loading control. B,cells were irradiated (2 Gy) andharvested at indicated times.Representative images of glioma-initiating cells and NPC with gH2AXfoci (green), Hoechst nucleus stain(blue), and merged. C, for each timepoint, the average foci number wasscored and expressed as foldchange against each individualneurosphere line at maximum foci(1 hour). From statistical analysis,�, P < 0.05, ��, P < 0.01.






postirradiation. To examine this, we used IdU pulse label-ing to measure DNA synthesis (Fig. 5A). Three glioma-initiating cells were selected (U251, L2B, and L3b) becausethey showed different efficiencies of homologous recom-bination. NPCs, U251, L2b, and L3b had 25%, 30%, 29%,and 31% IdU-positive cells, respectively (Fig. 5B). Follow-ing IR, IdU-positive cellswere reduced2-fold forNPCs,butin glioma-initiating cells no reduction occurred (30%–32%IdU positive), indicating that cells continued to enter S-phaseafter IR (P< 0.01). This is supportedbyasynchronouscell-cycle analysis showingunperturbed entry into S-phasein glioma-initiating cells (P < 0.05) after 5Gy, indicating anattenuated S-phase checkpoint (Fig. 5C).

DNA initiations remain relatively unchanged inglioma-initiating cells after irradiation

We next examined capacity of glioma-initiating cells toinitiate DNA synthesis following IR. Cells were pulse-labeled with CldU before irradiation to visualize ongoingDNA synthesis, followed by a pulse with IdU post-IR todetectnew initiations (Fig. 6A). Because initiationsdifferedbetweenneurosphere lines,wenormalized tounirradiatedcells and compared fold change after DNA damage. In

NPCs, new initiations reduced by 66% (Fig. 6B). Interest-ingly, U87, L1b, and L2b presented with relatively unal-tered new initiation post-IR (P < 0.0003). Partial inhibitionof new initiations occurred in U251 (42% reduction, P <0.023) and L3b (32% reduction, P < 0.008). But, all glioma-initiating cells maintained a higher level of new DNAreplication after IR compared with NPCs. Thus, glioma-initiating cells exhibit radioresistant DNA synthesis, indi-cating attenuation of the intra-S-phase checkpoint (35).

To investigate further, we examined ATM activationand phosphorylation of downstream substrates (SMC1,Chk1, and Chk2; ref. 36). ATM activity was attenuated inU87 with remaining glioma-initiating cells similar toNPCs (Fig. 6C). U87 also had reduced levels of SMC1,Chk1, and Chk2 proteins, and radiation-induced phos-phorylation of these proteinswasdefective (Fig. 6C).U251and L1b, which had normal levels of ATM and SMC1,showed reduced radiation-induced phosphorylation ofthese proteins, includingChk2. Activation of p53was alsomeasured. Basal p53 level was comparable with NPCs inglioma-initiating cells, exceptU87 andL1b inwhich levelswere low. p53 phosphorylation was detected in NPCs,U251, L2b, and L3bwith a stronger response in U251 (Fig.

Figure 3. Glioma-initiating cells donot show enhanced NHEJ. A,NHEJ events were quantifiedwithin the gated region (box) after48 hours. Total GFP fluorescenceevents from damaged (linearizedDNA) plasmid-transfected cellswere scored against intact(circular) plasmid transfected cellsfor each line to account fortransfection efficiency. B, graphquantifies the efficiency of NHEJ.C, phosphorylated (PDNA-PK) andtotal DNA-PKcs protein levelsdetermined by immunoblotting 1hour after IR. GAPDH is a loadingcontrol. D, efficiency of NHEJ with(DNA-PKi) and without [dimethylsulfoxide (DMSO)] DNA-PKcs

inhibitor. Neurosphere cultureswere treated with DNA-PKi orDMSO 1 hour before IR treatment.Cell viability was measured by thetrypan blue exclusion assay. Fromstatistical analysis, �, P < 0.05,��, P < 0.01.

Lim et al.





6C). No phosphorylation was detected for U87 or L1b,which might be explained by low p53 protein.Overall, our data suggested that defective checkpoint

activation in glioma-initiating cells allows unperturbed S-phase entry resulting in preferential use of the homolo-gous recombination pathway.

DiscussionThe mechanism for radioresistance in glioma has

remained unclear because previous studies comparedbetween subpopulations of aberrant glioma stem cellswithout using normal neuronal stem cells (8, 9). Humanneural stem cells are an ideal control, given their normalrepair profile and nontumor–forming nature (37). Weused the recently developed immortalized human NPCsas a control in this study to allow for comparison betweenpopulations with "normal" (NPCs) and aberrant (glioma-initiating cells) radiosensitivity. TheNPCswere similar toneural stem cells, as shownby their extensive self-renewalcapacity and multipotency, allowing formation of differ-ent lineages. NPCs sustain self-renewal and proliferation,

without rapidly developing chromosomal abnormalities(11). Given the capacity to compare with NPCs, it seemedrelevant to focus on the overall glioma-initiating cellspopulation, while incorporating NPC as a benchmark todetermine the efficiency of DNA damage responses.

Cancer stem cells (CSC), expressing CD133 have beenimplicated as the radioresistant population responsible fortumor development (2, 7). CD133þ glioma cells showedincreased survived post-IR versus CD133� tumor cells (7).Chk2 kinase inhibition reversed this radioresistance. How-ever, recent data cast doubt on these early findings. Underhypoxic conditions, CD133� CSCs expressed CD133þ (38,39). Also, in vivo studies showed that CD133� CSCs drivetumorigenesis in nude rats (24, 40). In our study, weshowed Sox2 and Nestin expression in all cultures, butnoted differences in CD49f expression between variousglioma-initiating cells lines (41, 22). Despite low CD133expression, all glioma-initiating cells induced tumorformation when injected intracranially into mice, suggest-ing CD133 expression does not correlate withtumorigenic potential. Further studies also showed similar

Figure 4. Increased homologousrecombination repair in glioma-initiating cells. A, stable clones ofNPC, U251, and L2b cultures withintegrated pDR-GFP vector wereeither mock transfected or withinducible I-Sce I�/þ vector.Homologous recombinationefficiency was measured 48 hourslater by flow cytometry to detectGFP-positive cells (above diagonalline). Graph quantifies the efficiencyof homologous recombination. B,cells treated with 5Gy IR andimmunolabeled to detect Rad51 fociat the indicated time points.Representative images show Rad51foci (green), Hoechst stain (blue), andmerged. Graph quantifies thenumber of cells with Rad51 foci. C,cells were irradiated after treatmentwith either Rad51 or negative controlsiRNA and harvested 48 hours later.Levels ofRad51andGAPDH (loadingcontrol) were determined byimmunoblotting. D, cells weretreated with negative control orRad51 siRNA; after 48 hours, cellsreceived IR and viable cellsmeasured by trypan blue exclusionassay. From statistical analysis,�, P < 0.05, ��, P < 0.01.






radioresistant properties between CD133þ/� populationsfollowingDNAdamage (8, 9);measurement of gH2AX focifailed to distinguish significant differences betweenCD133þ and CD133� CSCs (24, 40). However, time pointswere limited to 1 hour (maximum foci) and 24 hours(background levels). These time points would not detectdifferences in the rate of DNA DSB repair, as seen here,where the ratewas slower in glioma-initiating cells, though

theyweremore radioresistant thanNPCs (Fig. 1A and 1B).Supporting data from self-renewal and neurosphere sizecomparison also indicates glioma-initiating cells havegreater recovery than NPCs after IR. Surprisingly, NHEJwas reduced 2-fold in U251 and U87, the more radioresis-tantneurospherecultures.Thiswasborneout in total repairofDNADSBs,asdeterminedby lossof gH2AXfoci. In theseexperiments, the rate of DSB repair was reduced at initial

Figure 6. Glioma-initiating cells continue to initiate replication after ionizing radiation via defective cell-cycle checkpoint. A, diagram illustrating the use of 2different nucleoside analogs to study DNA replication after IR-induced damage. Representative images of replication signals fromDNA fibers; ongoing forks,unidirectional and bidirectional illustrated by continuous and dashed arrows (left and middle) and new initiation as dash arrows only (right). B, newinitiations after DNA damage. To quantify the frequency of firing of origins of replication, the number of red fibers (dashed arrows) was divided by thetotal sum of red and green-red signals (continuous and dashed arrows). Data in graph are normalized to the individual neurosphere line and average newinitiations shown as fold change. C, immunoblot of phosphorylated (indicated by prefix P) and total protein (ATM, SMC1, Chk1, Chk2, and p53) followingmock and 2Gy. From statistical analysis, �, P < 0.05, ��, P < 0.01.

Figure 5. Ionizing radiation does notinhibit DNA synthesis in glioma-initiating cells. Cells wereincubated with 10 mmol/L IdUimmediately before either mock or5 Gy treatment before fixation at 1hour. Columns representpercentage of cells undergoingDNA synthesis. A, IF imagesshowing IdU incorporation foci(red), Hoechst stain (blue), andmerged. B, data quantificationrepresenting the averagepercentage of IdU-labeled nuclei.C, cellswere eithermock or 5Gy IR-treated and cell-cycle stagedetermined 1 hour after exposure.Cells stained with propidium iodidewere quantified by flow cytometry.From statistical analysis,�, P < 0.05, ��, P < 0.01.

Lim et al.





time points and more breaks remained at later times.Combined data indicates that DNA DSB repair, at least byNHEJ, doesnot explain radioresistance inglioma-initiatingcells, as shown by DNA-PKcs inhibition having a lessereffect on glioma-initiating cells than NPCs (Fig. 3D). Incontrast to homologous recombination repair, as deter-mined by pDR-GFP expression, formation and resolutionof Rad51 and Brca1 foci was more efficient in the radio-resistant cultures (U251 and L3b). Glioma-initiating cellstreated with Rad51 siRNA showed increased radiosensi-tivity versus control siRNA-treated cells. Conversely,Rad51 knockdownhad little effect onNPC radiosensitivityconfirming the importance of homologous recombinationparticularly in glioma-initiating cells (Fig. 4D). Thereforeincreased dependence on homologous recombination cou-pledwithdefective cell-cycle checkpoints could explain theincreased glioma-initiating cells radioresistance.Cancer cells are characterized by abnormalities in

cell-cycle control, including prolonged activation of themitotic checkpoint, constitutive activation of the G1–Scheckpoint by p53, and activation by the ATM-53BP1-Chk2 pathway. Constitutive activation can cause selectivepressure in tumors and subsequent inactivation of DNAdamage responses (42, 43). Human embryonal carcinomacells were shown to be defective for the G1–S checkpointafter DNA damage but exhibited both S-phase and G2–Mdelay (44). However, when these cells were differentiatedwith retinoic acid, they failed to arrest in G1 and quicklyexited S-phase arresting in G2–M. In our study, glioma-initiating cells (U251, L2b and L3b) continued to enter S-phase after IR at rates comparable with untreated cells,showing failure of the G1–S checkpoint. Glioma-initiatingcells also exhibited radioresistant DNA synthesis. In anormal cell, DNA damage accumulating from cells pass-ing through S-phase carrying DNA damage wouldincrease cell death. In contrast, tumor cells circumventcell-cycle checkpoints and continue undergoing cell divi-sionswith a "DNAdamage burden"without inducing celldeath, thus resulting in increased survival. Interestingly,this phenomenon has previously been associated withhuman genetic disorders characterized by radiosensitiv-ity and chromosomal instability and is a determinant of adefective intra-S-phase checkpoint (36). It seems likelythat the G2 checkpoint is still intact in glioma-initiatingcells. This contrasts with data for embryonic carcinomas,which displayed prominent S- and G2-phase checkpoints(44). Although enhanced basal activation of Chk1 andChk2 was shown in CD133þ glioma cells (9), we saw noevidence of this in glioma-initiating cells. These results arein keeping with the cell-cycle data because we did notobserve a delay at the G1–S checkpoint, as reported forCD133þglioma cells (9). Bao and colleagues (7) postulatedthat cell-cycle delaymight represent a generalmechanism

for genome protection in glioma-initiating cells. Thispattern of radiation-induced signaling in glioma-initiat-ing cells and its relationship to radioresistance is difficultto explain, as IR-induced activation of ATM kinaseappears normal in all glioma-initiating cells except U87.Failure to observe normal radiation signaling throughSMC1, Chk1, and Chk2 in U87 can be explained byreduced protein levels. However, reduced signalingthrough SMC1 in U251 and L1b is intriguing, givennormal ATM activation and SMC1 protein levels. Phos-phorylation of SMC1 is important for maintaininggenome integrity and enhancing cell survival (45). Thereduced reliance on SMC1 phosphorylation in glioma-initiating cells may be compensated for enhanced homol-ogous recombination and cell-cycle checkpoint changes.

In summary, we showed that glioma-initiating cells aremore radioresistant thanNPC, but the extent of resistanceis variable. On the basis of these results, we suggest thattreatment of patients with drugs (i.e., ATM or ATR inhib-itor) specifically targeting the homologous recombinationpathway or blocking S-phase transition may be noveltherapeutic approaches. This approach may be moreeffective on GBM than surrounding tissues, as homolo-gous recombination is less used in NPCs.

Disclosure of Potential Conflicts of InterestNo potential conflicts of interest were disclosed.

Authors' ContributionsConception and design: Y.C. Lim, T.L. Roberts, B.W. Day, S. Kozlov, D.G.Walker, M.F. LavinDevelopment of methodology: Y.C. Lim, T.L. Roberts, B.W. Day, A.W.Kijas, M.F. LavinAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): Y.C. Lim, B.W. Day, A. Harding, K.S. Ensbey, M.F. LavinAnalysis and interpretation of data (e.g., statistical analysis, biostatis-tics, computational analysis): Y.C. Lim, T.L. Roberts, M.F. LavinWriting, review, and/or revision of the manuscript: Y.C. Lim, T.L.Roberts, S. Kozlov, M.F. LavinAdministrative, technical, or material support (i.e., reporting or orga-nizing data, constructing databases): K.S. EnsbeyStudy supervision: T.L. Roberts, D.G. Walker, M.F. Lavin

AcknowledgmentsThe authors thank Professor Reynolds for the glioma-initiating cells

lines, Dr. Stacey for CFSE dye, Dr. Richard for pDR-GFP construct, PaulaHall andGraceChojnowski for assistancewith flowcytometry, andTraceyLaing for assistance with typing the manuscript.

Grant SupportThe study was supported by grants from the Briz Brain and Spinal

Foundation, and the National Health and Medical Research Council ofAustralia to M.F. Lavin.

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 December 22, 2011; revised May 29, 2012; accepted June 29,2012; published OnlineFirst July 6, 2012.

References1. Schwartzbaum JA, Fisher JL, Aldape KD, Wrensch M. Epidemiology

and molecular pathology of glioma. Nat Clin Prac Neurol 2006;2:494–503.

2. Singh SK, Clarke ID, Terasaki M, Bonn VE, Hawkins C, Squire J, et al.Identification of a cancer stem cell in human brain tumors. Cancer Res2003;63:5821–8.






3. Temple S. The development of neural stem cells. Nature 2001;414:112–7.

4. Dupre A, Boyer-Chatenet L, Gautier J. Two-step activation of ATM byDNA and the Mre11–Rad50–Nbs1 complex. Nat Struct Mol Biol2006;13:451–7.

5. Dronkert MG, Beverloo B, Johnson RD, Hoeijmakers JJ, Jasin M,Kanaar R. Mouse RAD54 affects DNA double-strand breakrepair and sister chromatid exchange. Mol Cell Biol 2000;20:3147–56.

6. Wilson TE, Grawunder U, Lieber MR. Yeast DNA ligase IV mediatesnon-homologous DNA end joining. Nature 1997;388:495–8.

7. BaoS,WuQ,McLendonRE,HaoY,ShiQ,HjelmelandAB, et al. Gliomastem cells promote radioresistance by preferential activation of theDNA damage response. Nature 2006;444:756–60.

8. McCord AM, Jamal M, Williams ES, Camphausen K, Tofilon PJ.CD133þ glioblastoma stem-like cells are radiosensitive with a defec-tive DNA damage response compared with established cell lines. ClinCancer Res 2009;15:5145–53.

9. Ropolo M, Daga A, Griffero F, Foresta M, Casartelli G, Zunino A, et al.Comparative analysis of DNA repair in stem and nonstem glioma cellcultures. Mol Cancer Res 2009;7:383–9.

10. Vescovi AL, Galli R, Reynolds BA. Brain tumour stem cells. Nat RevCancer 2006;6:425–36.

11. Donato R, Miljan EA, Hines SJ, Aouabdi S, Pollock K, Patel S, et al.Differential development of neuronal physiological responsiveness intwo human neural stem cell line. BMC Neurosci 2007;8:36–47.

12. Reynolds BA, Weiss S. Generation of neurons and astrocytes fromisolated cells of the adult mammalian central nervous system. Science1992;255:1707–10.

13. Piccirillo SM, Reynolds BA, Zanetti N, Lamorte G, Binda E, Broggi G,et al. Bone morphogenetic proteins inhibit the tumorigenic potential ofhuman brain tumour-initiating cells. Nature 2006;444:761–5.

14. Day BW, Stringer BW, Spanevello MD, Charmsaz S, Jamieson PR,Ensbey KS, et al. ELK4 neutralization sensitizes glioblastoma toapoptosis through downregulation of the anti-apoptotic proteinMcl-1. Neuro Oncol 2011;13:1202–12.

15. Breier JM, Radio NM, Mundy WR, Shafer TJ. Development of a high-throughput screening assay for chemical effects on proliferation andviability of immortalized human neural progenitor cells. Toxicol Sci2008;105:119–33.

16. Suraweera A, Becherel OJ, Chen P, Rundle N, Woods R, Nakamura J,et al. Senataxin, defective in ataxia oculomotor apraxia type 2, isinvolved in the defense against oxidative DNA damage. J Cell Biol2007;177:969–79.

17. Allen C, Halbrook J, Nickoloff JA. Interactive competition betweenhomologous recombination and non-homologous end joining. MolCancer Res 2003;1:913–20.

18. Zhang J, Ma Z, Treszezamsky A, Powell SN. MDC1 interacts withRad51 and facilitates homologous recombination. Nat Struct Mol Biol2005;12:902–9.

19. Gatei M, Jakob B, Chen P, Kijas AW, Becherel OJ, Gueven N, et al.ATM-dependent phosphorylation of Rad50 regulates DNA repair andcell cycle control. J Biol Chem 2011;286:31542–56.

20. FellingRJ,SnyderMJ,RomankoMJ,RothsteinRP,Ziegler AN,YangZ,et al. Neural stem/progenitor cells participate in the regenerativeresponse to perinatal hypoxia/ischemia. J Neurosci 2006;26:4359–69.

21. Wan F, Zhang S, Xie R, Gao B, Campos B, Herold-Mende C, et al. Theutility and limitations of neurosphere assay, CD133 immunophenotyp-ing and side population assay in glioma stem cell. Brain Pathol2010;20:1750–3639.

22. Lathia JD, Gallagher J, Heddleston JM,Wang J, Eyler CE, MacSwordsJ, et al. Integrin alpha 6 regulates glioblastoma stem cells. Cell StemCell 2010;6:421–32.

23. Galli R, Binda E, Orfanelli U, Gritti BC, Vitis SD, Fiocco R, et al. Isolationand characterization of tumorigenic, stem-like neural precursors fromhuman glioblastoma. Cancer Res 2004;64:7011–21.

24. Wang J, Sakariassen P�, Tsinkalovsky O, Immervoll H, Bøe SO,Svendsen A, et al. CD133 negative glioma cells form tumors in nuderats andgive rise toCD133positive cells. Int JCancer 2007;122:761–8.

25. ShuHG, KimMM, Chen P, Furman F, Julin CM, Israel MA. The intrinsicradioresistance of glioblastoma-derived cell lines is associated with afailure of p53 to induce p21BAX expression. Proc Natl Acad Sci U S A1998;95:14453–8.

26. Deleyrolle LP, ErickssonG,Morrison BJ, Lopez JA, BurrageK, BurrageP, et al. Determination of somatic and cancer stem cell self-renewingsymmetric division rate using sphere assays. PLoS One 2011;6:e15844.

27. Marcu L, Doorn TV, Olver I. Modelling of post-irradiation acceleratedrepopulation in squamous cell carcinomas. Phys Med Biol 2004;49:3767–79.

28. ThamesHD,RuifrokAC,Milas L, HunterN,MasonKA, TerryNHA, et al.Accelerated repopulation during fractionated irradzation of a murineovarian carcinom: Downregulation of apoptotsis as a possible mech-anism. Int J Radiat Oncol Biol Phys 1996;35:951–62.

29. Boosch Md, Bree RT, Lowndes NF. The MRN complex: coordinatingand mediating the response to broken chromosomes. EMBO J2003;4:844–9.

30. Saleh-Gohari N, Helleday T. Conservative homologous recombina-tion preferentially repairs DNA double-strand breaks in the Sphase of the cell cycle in human cells. Nucleic Acids Res 2004;32:3683–8.

31. Allen C, Kurimasa A, Brenneman MA, Chen DJ, Nickoloff JA. DNA-dependent protein kinase suppresses double-strand break-inducedand spontaneous homologous recombination. Proc Natl Acad SciU S A 2002;99:3758–63.

32. TarsounasA,DaviesD,West SC.BRCA2-dependent and independentformation of RAD51 nuclear foci. Oncogene 2003;22:115–1123.

33. Cousineau I, Abaji C, Belmaaza A. BRCA1 regulates RAD51 func-tion in response to DNA damage and suppresses spontaneoussister chromatid replication slippage: Implications for sister chro-matid cohesion, genome stability and carcinogenesis. Cancer Res2005;65:11384–91.

34. ZhouC,HuangP, Liu J. The carboxyl-terminal of BRCA1 is required forsubnuclear assembly of RAD51 after treatment with cisplatin but notionizing radiation in human breast and ovarian cancer cells. BiochemBiophys Res Commun 2005;336:925–60.

35. Falck J, Mailand N, Syljua�sen RG, Bartek J, Lukas J. The ATM–Chk2–

Cdc25A checkpoint pathway guards against radioresistant DNA syn-thesis. Nature 2001;410:842–7.

36. Lavin MF. Ataxia-telangiectasia: from a rare disorder to a paradigm forcell signalling and cancer. Nat Rev Mol Cell Biol 2008;9:759–69.

37. Sidman RL, Li J, Stewart GR, Clarke J, Yang W, Snyder EY, et al.Injection of mouse and human neural stem cells into neonatal Nie-mann–Pick A model mice. Brain Res 2007;1140:195–204.

38. Blazek ER, Foutch JL, Guitta M. Daoy medulloblastoma cells thatexpress CD133 are radioresistant relative to CD133þ and the CD133-sector is enlarged by hypoxia. Int J Radiat Oncol Biol Phys 2007;67:1–5.

39. Griguer CE, Oliva CR, Gobin E, Marcorelles P, Benos DJ, Jr JL, et al.CD133 is a marker of bioenergetic stress in human glioma. PLoS One2008;3:e3655.

40. Beier D, Hau P, ProescholdtM, Lohmeier A,Wischhusen Jr, Oefner PJ,et al. CD133þ and CD133- Glioblastoma-derived cancer stem cellsshowdifferential growth characteristics andmolecular profiles. CancerRes 2007;66:7843–8.

41. Graham V, Khudyakov J, Ellis P, Pevny L. SOX2 functions to maintainneural progenitor identity. Neuron 2003;39:749–65.

42. Bartek J, Lukas C, Lukas J. Checking on DNA damage in S phase. NatRev Mol Cell Biol 2004;5:792–804.

43. Halazonetis TD. Constitutively active DNA damage checkpoint path-ways as the driving force for the high frequency of p53 mutations inhuman cancer. DNA Repair 2004;3:1057–62.

44. Wang X, Lui VH, Poon RP, Lu P, Poon RC. DNA damage mediated Sand G2 checkpoints in human embryonal carcinoma cells. Stem Cells2009;27:568–76.

45. Yazdi PT, Wang Y, Zhao S, Patel N, Lee E, Qin J. SMC1 is a down-stream effector in the ATM/NBS1 branch of the human S-phasecheckpoint. Genes Dev 2002;16:571–82.

Lim et al.





2012;11:1863-1872. Published OnlineFirst July 6, 2012.Mol Cancer Ther Yi Chieh Lim, Tara L. Roberts, Bryan W. Day, et al. Progression in Radioresistance of Glioma-Initiating CellsA Role for Homologous Recombination and Abnormal Cell-Cycle

Updated version

10.1158/1535-7163.MCT-11-1044doi:

Access the most recent version of this article at:

Material

Supplementary

http://mct.aacrjournals.org/content/suppl/2012/07/06/1535-7163.MCT-11-1044.DC1

Access the most recent supplemental material at:

Cited articles

http://mct.aacrjournals.org/content/11/9/1863.full#ref-list-1

This article cites 45 articles, 14 of which you can access for free at:

Citing articles

http://mct.aacrjournals.org/content/11/9/1863.full#related-urls

This article has been cited by 3 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://mct.aacrjournals.org/content/11/9/1863To request permission to re-use all or part of this article, use this link



http://mct.aacrjournals.org/lookup/doi/10.1158/1535-7163.MCT-11-1044

http://mct.aacrjournals.org/content/suppl/2012/07/06/1535-7163.MCT-11-1044.DC1

http://mct.aacrjournals.org/content/11/9/1863.full#ref-list-1

http://mct.aacrjournals.org/content/11/9/1863.full#related-urls

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

mailto:[email protected]

http://mct.aacrjournals.org/content/11/9/1863


a role for homologous recombination and abnormal cell-cycle … · therapeutic discovery a role for...

Documents