chk1 atr

Targeted Inhibition of ATR or CHEK1 ReversesRadioresistance in Oral Squamous Cell CarcinomaCells with Distal Chromosome Arm 11q Loss

Madhav Sankunny,1,2 Rahul A. Parikh,2,3 Dale W. Lewis,1 William E. Gooding,2,4 William S. Saunders,2,5

and Susanne M. Gollin1,2,6,7*

1Departmentof HumanGenetics,Universityof Pittsburgh Graduate School of Public Health,Pittsburgh,PA2Universityof Pittsburgh Cancer Institute,Pittsburgh,PA3Departmentof Medicine,Division of Hematology-Oncology,Universityof Pittsburgh School of Medicine,Pittsburgh,PA4Departmentof Biostatistics,Universityof Pittsburgh Cancer Institute,Pittsburgh,PA5Departmentof Biological Sciences,Universityof Pittsburgh,Pittsburgh,PA6Departmentof Otolaryngology,Universityof Pittsburgh School of Medicine,Pittsburgh,PA7Departmentof Pathology,Universityof Pittsburgh School of Medicine,Pittsburgh,PA

Oral squamous cell carcinoma (OSCC), a subset of head and neck squamous cell carcinoma (HNSCC), is the eighth most

common cancer in the U.S.. Amplification of chromosomal band 11q13 and its association with poor prognosis has been well

established in OSCC. The first step in the breakage-fusion-bridge (BFB) cycle leading to 11q13 amplification involves breakage

and loss of distal 11q. Distal 11q loss marked by copy number loss of the ATM gene is observed in 25% of all Cancer

Genome Atlas (TCGA) tumors, including 48% of HNSCC. We showed previously that copy number loss of distal 11q is

associated with decreased sensitivity (increased resistance) to ionizing radiation (IR) in OSCC cell lines. We hypothesized

that this radioresistance phenotype associated with ATM copy number loss results from upregulation of the compensatory

ATR-CHEK1 pathway, and that knocking down the ATR-CHEK1 pathway increases the sensitivity to IR of OSCC cells with

distal 11q loss. Clonogenic survival assays confirmed the association between reduced sensitivity to IR in OSCC cell lines

and distal 11q loss. Gene and protein expression studies revealed upregulation of the ATR-CHEK1 pathway and flow cytome-

try showed G2M checkpoint arrest after IR treatment of cell lines with distal 11q loss. Targeted knockdown of the ATR-

CHEK1 pathway using CHEK1 or ATR siRNA or a CHEK1 small molecule inhibitor (SMI, PF-00477736) resulted in increased

sensitivity of the tumor cells to IR. Our results suggest that distal 11q loss is a useful biomarker in OSCC for radioresistance

that can be reversed by ATR-CHEK1 pathway inhibition. VC 2013 Wiley Periodicals, Inc.

INTRODUCTION

Cancer is a critical public health problemaccounting for one in four deaths in the U.S. andmore worldwide despite advances in diagnosis andtreatment (Heron et al., 2009; Siegel et al., 2013).Oral (oropharyngeal) squamous cell carcinoma(OSCC) is the eighth most common cancer in theU.S., accounting for an estimated 41,380 new cases(2.5%) in 2013 and 7,890 deaths (1.4%) (Siegelet al., 2013). Worldwide, HNSCC (worldwide sta-tistics include lip, oral cavity, larynx, and pharynx)was diagnosed in 550,319 new patients (4.4%) in2008 and resulted in 305,096 deaths (4.0%) (Ferlayet al., 2010; Bray et al., 2013). Over the past fourdecades, the 5-year relative survival rates haveimproved substantially for oropharyngeal SCC inboth Caucasians (from 54 to 67%) and in AfricanAmericans (from 36 to 45%), although disturbingdisparities remain (Desantis et al., 2013).

Carcinomas are frequently characterized bychromosomal instability, resulting in loss of tumorsuppressor genes and gain or amplification of onco-genes (Ha and Califano, 2002). One of the mostfrequent chromosomal abnormalities in OSCC and

Supported by: NIH/NIDCR/NCI, Grant numbers: R01DE014729,P30CA047904, and P50CA097190; University of Pittsburgh CancerInstitute; Pfizer, Inc., Grant number: WS594969; Jennie K. ScaifeOvarian Cancer Center of Excellence at the University of Pitts-burgh, Pittsburgh Foundation Learmonth Fund, Joan Gaines Can-cer Research Fund at the University of Pittsburgh, includingdonations from Mr. Bill Benter and the Teresa and H. John HeinzIII Charitable Fund.

*Correspondence to: Susanne M. Gollin, Department of HumanGenetics, University of Pittsburgh, Graduate School of PublicHealth, 130 DeSoto Street, A300 Crabtree Hall, Pittsburgh, PA15261, USA. E-mail: [email protected]

Received 9 October 2013; Accepted 15 October 2013

DOI 10.1002/gcc.22125

Published online 25 November 2013 inWiley Online Library (wileyonlinelibrary.com).

VVC 2013 Wiley Periodicals, Inc.

GENES, CHROMOSOMES & CANCER 53:129143 (2014)

other carcinomas is amplification of chromosomalband 11q13, which includes the cyclin D1 gene(CCND1) (Schraml et al., 1999; Gollin, 2001;Huang et al., 2002; Albertson, 2006; Jin et al.,2006; Gibcus et al., 2007). 11q13 amplificationresults from breakage-fusion-bridge (BFB) cycles(Reshmi et al., 2007) and/or chromosome breakageand rearrangement resulting from palindromic seg-mental duplications flanking 11q13 (Gibcus et al.,2007). The first step in the BFB cycle is a chromo-some break distal to the amplified region, possiblyat the FRA11F chromosomal fragile site or as aresult of rearrangement involving segmental dupli-cations resulting in loss of some or all of the distalsegment of chromosome 11q (Shuster et al., 2000;Reshmi et al., 2007).Loss of distal 11q and amplification of chromo-

somal band 11q13 are associated with poor progno-sis in OSCC (Michalides et al., 1995; Akervallet al., 1997; Jin et al., 1998; Jin et al., 2006).Although numerous investigators identified copynumber loss of distal 11q, from 11q14!11qter,centered primarily on 11q22!q23 in a variety ofprimary tumors and cell lines (George et al., 2007;Parikh et al., 2007; Ambatipudi et al., 2011; Swartset al., 2011; Edelmann et al., 2012), we showedthat distal 11q (ATM gene) loss is associated withreduced sensitivity (resistance) to ionizing radia-tion (IR) in OSCC cell lines (Parikh et al., 2007;Henson et al., 2009). In silico copy number analy-sis of the ATM gene shows that distal 11q loss ispresent in a focal peak of deletion containing 60 to80 genes in 25% of The Cancer Genome Atlas(TCGA) more than 7,200 tumors, including 47 to54% of melanomas, HNSCC, esophageal SCC,breast and cervical carcinomas, 30 to 36% of lungSCC, ovarian, prostate, and bladder carcinomas,and 18 to 28% of lung, stomach, colorectal, hepato-cellular, and rectal carcinomas (Beroukhim et al.,2010; Mermel et al., 2011). Thus, based on theAmerican Cancer Society statistics (Siegel et al.,2013) and the TCGA frequencies, at least 330,000of the 1,660,290 new cancer cases expected in theU.S. in 2013 may have distal 11q loss.Distal 11q contains a block of critical DNA dam-

age response (DDR) genes, including ATM(11q22.3), MRE11A (11q21), H2AFX (11q23.3), andCHEK1 (11q24.2). The cornerstone of the DDR toIR is the ATM gene, which is mutated in the rare,pleiotropic autosomal recessive disorder, ataxiatelangiectasia (AT) (Harnden, 1994; Savitsky et al.,1995; Baskaran et al., 1997; Lavin and Shiloh, 1997;Pandita et al., 1999). ATM encodes a 370 kDa pro-tein that is a member of the family of lipid/protein

kinases related to phosphatidylinositol 3-kinase(PI3K), known as the PI3K-related kinases (PIKK)(Keith and Schreiber, 1995; Shiloh, 2003; Abraham,2004). In IR-induced double strand breaks (DSB),the MRE11A-RAD50-NBS1 (MRN) complex playsthe role of a sensor (Stracker et al., 2004). Theprimary transducer of the DSB signal is ATM(Shiloh, 2003). In response to DNA DSB inducedby IR, ATM is rapidly phosphorylated at the serine1981 (ser1981) residue, which facilitates the phos-phorylation of various other proteins involved inthe regulation and repair of DNA damage (Bakken-ist and Kastan, 2003). These effectors of ATMinclude ABL1, BRCA1, TP53, and CHEK2 (Bas-karan et al., 1997; Khanna and Jackson, 2001). Animportant substrate of ATM is the histone H2AX, amember of the histone H2A subfamily that maps todistal 11q (Fernandez-Capetillo et al., 2004). ATMactivates H2AX at the DSB site, converting it intothe phosphorylated form, gH2AX, which thenanchors DNA damage response proteins to the sitesof damage (Stucki et al., 2005). The signaling cas-cade activated by ATM culminates in cell cyclearrest, apoptosis, and DNA repair. Ataxia Telan-giectasia and Rad3-related (ATR) is another PIKKfamily protein involved primarily in signaling thepresence of stalled replication forks and mainte-nance of genomic integrity during S phase, alongwith its partners, ATR interacting protein (ATRIP)and replication protein A (RPA) (Cortez et al.,2001; Zou and Elledge, 2003; Byun et al., 2005).Although ATR plays a primary role in respondingto ultraviolet light (UV) and chemotherapy-inducedgenomic insults, it also appears to play a role inresponding to IR-induced DNA damage (Adamset al., 2006; Myers and Cortez, 2006). The presenceof ATR in nuclear foci after IR treatment suggestsrecruitment of this kinase to sites of DNA damage;ATM is known to regulate the loading of ATR tosites of DNA DSB (Zou and Elledge, 2003; Cua-drado et al., 2006). Unlike the rapid phosphoryla-tion of ATM post-treatment, ATR recruitment andactivation is comparatively delayed, but alsorequires a functional MRN complex (Adams et al.,2006). Even though the ATM and ATR pathwaysare thought to respond primarily to different typesof DNA damage, the two pathways appear to beintimately intertwined.Based on our observation that distal 11q loss

leads to radioresistance in OSCC cell lines, thisstudy tests the hypotheses that (1) distal 11q loss,including loss of the ATM gene is associated withan upregulated ATR-CHEK1 pathway after IR,(2) upregulation of the ATR-CHEK1 pathway

130 SANKUNNY ET AL.

Genes, Chromosomes & Cancer DOI 10.1002/gcc

results in G2-M checkpoint arrest after IR treat-ment and radioresistance, and (3) knockdown ofthe ATR-CHEK1 pathway by RNA interferenceor a targeted small molecule inhibitor resensitizescancer cells with distal 11q loss to IR.

MATERIALS AND METHODS

Cell Culture

The UPCI:SCC OSCC cell lines developed byus (White et al., 2007) were cultured in M10medium comprised of Minimal Essential Medium(MEM) supplemented with 1% nonessential aminoacids (NEAA), 1% L-glutamine, 0.05 mg/ml genta-micin, and 10% fetal bovine serum (FBS) (all fromGIBCO Invitrogen, Grand Island, NY). Normalhuman oral keratinocytes grown in our laboratory(NHOK) and TERT-transfected human keratino-cyte (OKF6/TERT-1 cells, a gift of James Rhein-wald, Brigham and Womens Hospital, HarvardInstitutes of Medicine (Dickson et al., 2000)) celllines were used as controls. NHOK were isolatedfrom normal oral epithelial tissue obtained and dei-dentified by the University of Pittsburgh Head &Neck SPORE Tissue Bank from consenting non-cancer patients under our unified Tissue BankInstitutional Review Board approval. NHOK werecultured in serum-free keratinocyte growthmedium-2 (KGM-2) medium (Lonza, Walkersville,MD) supplemented with bovine pituitary extract(BPE), human epidermal growth factor (hEGF),insulin (bovine), hydrocortisone, GA-1000 (Genta-micin, Amphotericin B), epinephrine and transfer-rin supplied in the KGM-2 BulletKitTM (Lonza) asper the manufacturers instructions. OKF6/TERT-1 cells were cultured in Keratinocyte-SFM supple-mented with 25 lg/ml bovine pituitary extract, 0.2ng/ml EGF, 0.3 mM CaCl2, and penicillin-streptomycin (GIBCO Invitrogen).

Fluorescence In Situ Hybridization (FISH)

Molecular cytogenetic analysis was carried out todetermine the relative copy numbers of ATM,MRE11A, H2AFX, and CHEK1 in OSCC cell linescompared with the chromosome 11 centromere(CEP11/D11Z1, Abbott Molecular Inc., Des Plaines,IL). Molecular cytogenetic analyses by FISH werecarried out and analyzed in the University of Pitts-burgh Cell Culture and Cytogenetics Facility asdescribed previously (Parikh et al., 2007). TheFISH probes were prepared from the followingBAC clones purchased from Childrens Hospital ofOakland Research Institute (CHORI, Oakland, CA;

http://www.chori.org/bacpac): RP11-241D13 (ATM),RP11-685N10 (MRE11A), RP11-892K21 (H2AFX),and RP11-712D22 (CHEK1). BAC DNA waslabeled with Spectrum OrangeTM using a nick trans-lation kit from Abbott Molecular Inc. The percent-age of cells with relative loss, gain, or amplificationwas determined after counting signals in at least 200nuclei per slide. Copy number gain or loss of geneswas determined by the relative ratio of the BACprobe to centromere 11. A ratio of

survival was assessed with proportional hazardsregression by adjusting for these covariates asneeded to determine if 11q loss could be consid-ered independently associated with cancermortality.

Quantitative Real-Time PCR (QRT-PCR)

QRT-PCR was carried out to assess the relativeATR and CHEK1 expression in untreated cell linesand those treated with IR or transfected withCHEK1 or ATR siRNA. RNA extraction for real-time PCR was performed using TRIzol reagent(Gibco Invitrogen) according to the manufac-turers instructions. The extracted RNA was puri-fied using the RNeasy Mini kit (QIAGEN,Germantown, MD) and resuspended in 100 llRNase-free water. The RNA samples were puri-fied of contaminating DNA using a DNAfreeDNase kit (Ambion, Austin, TX) according to themanufacturers instructions. RNA concentrationswere assessed using a SmartSpec 3000 (Bio-RadLaboratories, Hercules, CA) and normalized to 40ng/ll. Reverse transcription was carried out asdescribed earlier (Huang et al., 2002).QRT-PCR of cDNA obtained after reverse tran-

scription was carried out on a 7300 Real-TimePCR System (Applied Biosystems, Foster City,CA), and analysis was done using the relativequantitation method as in Huang et al. (2002).The final concentrations of the QRT-PCR reac-tion components were as follows: 13 TaqmanGene Expression Master Mix and 13 TaqmanGene Expression Assays (Probe/Primer mix forATR, CHEK1 and 18s RNA) (Applied Biosystems).Thermocycler conditions for QRT-PCR were asfollows: 95C for 10 min, 40 cycles of 95C for 15 sand 60C for 60 s using the (Applied Biosystems).The RNA expression levels were quantified rela-tive to Universal Reference cDNA obtained fromClontech (Mountain View, CA).

Immunoblotting

Immunoblotting was carried out as describedpreviously (Parikh et al., 2007) to evaluate expres-sion of ATR-CHEK1 pathway proteins includingATR, CHEK1 and pCHEK1, and to study thefunction of TP53 in our cell lines by analyzingTP53 and p21 protein expression after Adriamycintreatment (0.4 lg/ml in culture medium for 45 hat 37C). The primary antibodies utilized were:ATR (Affinity Bioreagents, Golden, CO, 1:1,000dilution), CHEK1 (Cell Signaling, Danvers, MA,

1:1,000 dilution), pCHEK1 ser345 (Cell Signaling,1:1,000 dilution), pCHEK1 ser317 (Cell Signaling,1:1,000 dilution), TP53 (Santa Cruz Biotechnol-ogy, Inc., 1:750 dilution), and p21 (Santa Cruz Bio-technology, Inc., 1:750 dilution). Appropriatespecies-specific secondary antibodies (Santa CruzBiotechnology, Inc., 1:5,000 dilution) were appliedand target proteins were visualized using theWestern LightningTM Chemiluminescence Rea-gent Plus kit (PerkinElmer Life Sciences, Boston,MA) according to the manufacturers instructions.To verify equal protein loading in the gels, mem-branes were stripped and re-probed with antibod-ies against b-actin (Sigma Immunochemicals, St.Louis, MO, 1:1,000 dilution) or a-Actinin (SantaCruz Biotechnology, Inc., 1:1,000 dilution).

Densitometric Analysis of Protein Bands

Image J software from NIH was used for analyz-ing densities of protein bands in western blots.Relative optical densities were calculated for theprotein being evaluated and the loading controlprotein for all lanes using one of the lanes as a ref-erence. Adjusted densities were then calculatedfor each sample by normalizing the relative den-sity of the protein of interest to the loading controlfor the same.

Cell Cycle Analysis by Flow Cytometry

For cell cycle analysis, cells were seeded in 35mm or 60 mm dishes and allowed to attach over-night. Following the relevant treatments, mock orIR, floating and adherent cells were collected atthe end of 24 h, washed with phosphate-bufferedsaline (PBS), and fixed with 70% ethanol. Thecells were then treated with 80 mg/ml RNase Aand 50 mg/ml propidium iodide (Invitrogen-Molec-ular Probes, Carlsbad, CA) for 45 min at 37C.The stained cells were analyzed using a CoulterEpics XL Flow Cytometer in the UPCI FlowCytometry Facility.

Assessment of Mitotic Segregation Defects

To assess genotoxic stress and chromosomalinstability, we analyzed mitotic segregationdefects in OSCC cell lines with or without distal11q loss that were grown on coverslips. Cells wereeither untreated or treated with IR and thengrown for 18 or 36 h. At the specified timepoint,the cells on coverslips were fixed with 100% meth-anol, dried, stained with DAPI, and mounted ontoslides using antifade. The slides were coded and

132 SANKUNNY ET AL.


1,000 cells were analyzed from each cell line. Thefrequencies of anaphase and interphase bridgesand micronuclei, markers of chromosomal segrega-tion defects resulting from chromosomal aberra-tions, were recorded.

CHEK1 Knockdown by RNA Interference

RNA interference to CHEK1 and ATR was per-formed using Smartpool duplexes for each of thespecific genes, obtained from Dharmacon (Lafay-ette, CO). Nonspecific (scrambled) controlduplexes (Dharmacon) were used as controls. Theduplexes were reconstituted in DNAfree RNAresuspension buffer provided by Dharmaconaccording to the manufacturers instructions. Fortransfection, the carcinoma cell lines were seededin 60 mm dishes or T25 flasks and transfectedwith siRNA duplexes using Lipofectamine 2000(Invitrogen) according to the manufacturersinstructions. The final working siRNA concentra-tion ranged from 90 to 100 nM. We examined cellstreated without vector (untreated controls), cellstransfected with the nonspecific (scrambled) con-trol siRNA (mock transfected controls), and cellstransfected with smartpool siRNA duplexes foreither CHEK1 (siCHEK1 transfected) or ATR(siATR transfected) for all of our siRNA experi-ments. Peak siRNA transfection was seen at theend of 24 to 96 h as assessed by QRT-PCR andimmunoblotting. At 72 h post-transfection, cellswere seeded for clonogenic survival assay.

CHEK1 Knockdown by Small Molecule Inhibitor

PF-00477736, a potent, specific CHEK1 smallmolecule inhibitor (SMI) was a gift from Pfizer,

Inc. (Groton, CT). PF-00477736 selectively inhib-its enzymatic activity of CHEK1, abrogates DNAdamage-induced cell cycle arrest, and increasesthe cytotoxic effect of DNA damaging agents inTP53-defective cell lines (Blasina et al., 2008).Doseresponse curves for the SMI, assessment ofthe optimal time for addition of the SMI in rela-tion to IR treatment, and the effects of monother-apy with the drug or combined treatment with IRwere determined using clonogenic survival assays.

RESULTS

Copy Number Loss of Genes on Distal 11q in

OSCC Cell Lines is Associated with Decreased

Sensitivity to IR

Dual-color FISH with BAC probes to the ATM,MRE11A, H2AFX, and CHEK1 genes was carriedout in a selected series of our OSCC cell linesbased on their utility in clonogenic survival assaysto determine the relative gene copy number com-pared with the chromosome 11 centromere enu-meration probe (CEP11; Fig. 1). Table 1summarizes these FISH results, which suggestthat copy number loss of genes on distal 11q isobserved in a substantial fraction of OSCC celllines.The sensitivity of OSCC cells to IR was

assessed by clonogenic survival assay. The celllines were divided into two groups: distal 11qloss and no distal 11q loss. The IR doses usedfor our experiments were 2.5 Gy and 5 Gy. Resultsare reported as plots of Surviving Fraction vs.radiation dose on a semilog scale. OSCC cell lineswith distal 11q loss showed 63.8 6 1.9% and 17.86 0.9% survival at 2.5 and 5 Gy of IR,

TABLE 1. Summary of the Interphase FISH Results in Selected OSCC Cell Lines

Percentage of cells with copy number loss ofa

Cell lineMRE11A 11q21

(9415046694227040)ATM 11q22-q23

(108093559108239826)H2AFX 11q23.3

(118964584118966177)CHEK1 11q24.2

(125495031125546150)

UPCI:SCC029Bb 90 92 97 99UPCI:SCC040b 100 99 99 99UPCI:SCC084b 95 93 98 97UPCI:SCC131b 98 92 96 98UPCI:SCC136b 99 92 91 96UPCI:SCC104 1 98 96 1UPCI:SCC081b 8 22 97 4UPCI:SCC066 20 3 13 15UPCI:SCC116 14 15 16 10

Interphase FISH was used to assess copy number of ATM, MRE11A, H2AFX and CHEK1 genes relative to the chromosome 11 centromere.aLower ratio of test gene relative to the copy number of the chromosome 11 centromere.bThese cell lines express CCND1 gain or amplification.

TARGETING ATR-CHEK1 IN OSCC WITH 11Q LOSS 133


respectively, whereas the cell lines without distal11q loss showed 23.9 6 2.4% and 2.1 6 0.1% sur-vival, respectively at the same two IR doses(Fig. 2). These results translated to threefoldhigher survival at 2.5 Gy and eightfold higher sur-vival at 5 Gy IR in the OSCC cell lines with distal11q loss compared with the OSCC cell lines with-out distal 11q loss. Similar to our previous results(Parikh et al., 2007), these results showed clearevidence of radioresistance (or increased survival)in OSCC cell lines with distal 11q loss comparedwith OSCC cell lines without distal 11q loss atboth radiation doses tested. Hence, radioresistanceappears to be associated with distal 11q lossmarked by the ATM gene.

Distal 11q (ATM) Loss is Associated with Shorter

Disease-Specific Survival in Patients with OSCC

To examine the effect on disease-specific sur-vival of distal 11q loss as measured by FISH withthe ATM/CEP11 probe set, we carried out Kaplan-

Meier survival estimates in a retrospective seriesof 42 OSCC patients, 35 from our institution andseven from the University of Michigan. Allpatients had surgical resection of their SCCHN

Figure 1. Representative FISH images (A) showing normal copy number of ATM comparedwith CEP11; (B) and (C) showing loss of ATM copy number compared with CEP11; (D) showingnormal copy number of CHEK1 compared with CEP11; and (E) and (F) showing loss of CHEK1copy number compared with CEP11.

Figure 2. Survival of OSCC cell lines after treatment with IR. Thesurviving fraction of cells at specific IR doses is plotted with error bars(6SEM) on a logarithmic scale. Carcinoma cells in the distal 11q lossgroup (UPCI:SCC029B, 040, and 131) showed three times as muchsurvival as cells in the no distal 11q loss group (UPCI:SCC066, 081,and 116).

134 SANKUNNY ET AL.


tumors between 1992 and 1996. Thirty-ninepatients are deceased, 30 subsequent to head andneck cancer and nine from other causes; threepatients were alive without evidence of disease atlast follow-up.The results show that those 27 OSCC patients

with distal 11q loss in at least 80% of cells have amedian disease-specific survival of 18 monthscompared with 65.4 months for the 15 patientswith distal 11q loss in less than 80% of cells (logrank P 5 0.0501) (Fig. 3). We investigatedwhether this difference was influenced by treat-ment with radiation or chemotherapy. Thirty-onepatients underwent chemo- or radiotherapy eitherin the adjuvant setting or for disease recurrence.Applying Cox proportional hazards regression, wefound no interaction between distal 11q loss andexposure to chemo- or radiotherapy (P 5 0.183).The hazard ratio for distal 11q loss alone was 2.27(95% CI 5 0.975.33). The effect of distal 11qloss among the 31 treated patients was similar orslightly more pronounced (P 5 0.022) with a haz-ard ratio of 3.07 (95% CI 5 1.128.38). We alsoinvestigated whether the association between dis-tal 11q loss and disease-specific survival was dueto the confounding influence of other clinical vari-ables. Age and T stage were modestly associatedwith survival, but neither covariate was correlatedwith distal 11q loss. Nonetheless, we estimatedthe hazard ratio for distal 11q loss alone and condi-tional upon age and T stage. The hazard ratio wasunchanged (2.0 alone vs. 2.04 adjusting for ageand T stage). We conclude that distal 11q loss isassociated with increased risk of cancer mortalityindependently of T and N stage and may havepredictive utility for response to chemotherapy orradiotherapy.

Increased Mitotic Segregation Defects After IR

Treatment of Cells with Distal 11q Loss

To assess mitotic defects in OSCC cell linesafter treatment with IR, we measured the frequen-cies of micronuclei and anaphase and interphasechromosome bridges, which represent the result ofmisrepaired DNA and chromosomes and/or defec-tive chromosomal segregation (Fenech et al.,2011). The frequency of each aberration type withor without IR treatment is shown in Table 2 andFigure 4. Cells with distal 11q loss have a greaterthan two-fold increase in micronuclei and a 10-fold increase in interphase bridges 18 and 36 hpost-irradiation compared with cells without distal11q loss. No significant accumulation of anaphasebridges was observed in either group of cell lines.This is not surprising, since most cells would beexpected to continue to cycle, resolving anaphasebridges into aberrant nuclear chromosomes, micro-nuclei, and/or interphase bridges. This result pro-vides additional support to the proposition thatcells with distal 11q loss have a diminished DNAdamage response to IR-induced DNA damagecompared with cells without distal 11q loss.

The ATR-CHEK1 Pathway is Upregulated in

OSCC Cell Lines with Distal 11q Loss

Copy number loss of genes on distal 11qresulted in decreased expression of MRE11A,ATM, and H2AFX and their proteins (Parikh et al.,2007). This observation combined with the radio-resistance phenotype suggests that a decrease inthe ATM pathway results in increased activity of acompensatory pathway in the cells with distal 11qloss. We evaluated the activity of the relatedATR-CHEK1 pathway by examining the proteinexpression of some members of this pathway, spe-cifically, ATR, CHEK1 and its phosphorylatedform. Immunoblotting for CHEK1 proteinrevealed that in untreated OSCC cells, there wasno strong correlation observed between CHEK1protein expression and distal 11q loss as was seenbetween CHEK1 gene expression and distal 11qloss (Parikh et al., 2007). Four of five cell lineswith distal 11q loss (UPCI:SCC029B, 040, 104,131) showed lower CHEK1 protein expressioncompared with the cell lines without 11q loss(UPCI:SCC066, 081, 116) (Fig. 5A). UPCI:SCC136(distal 11q loss) had high CHEK1 expression.OSCC cell lines with distal 11q loss that weretreated with IR showed an increase in CHEK1expression 6 h after radiation compared with celllines without 11q loss (Fig. 5B). This suggests

Figure 3. Kaplan-Meier plot of disease specific survival. Forty-twopatients were classified by distal 11q loss (680%). The median follow-up for 12 censored patients was 5.9 years (range 2 months18 years).P 5 0.0515 by a log rank test.



upregulation of the ATR-CHEK1 pathway afterirradiation of cells with distal 11q loss.We then examined the phosphorylation status of

CHEK1 using antibodies against the phosphoryla-tion sites ser345 and ser317 to confirm upregulationof the ATR-CHEK1 pathway. Ser317 phosphoryla-tion is required for subsequent ser345 phosphoryla-tion in response to DNA damage (Wang et al.,2012). Ser345 phosphorylation is necessary forCHEK1 activation and a proper checkpointresponse after DNA damage and promotes nuclearretention of CHEK1 (Jiang et al., 2003; Niida et al.,2007). Early phosphorylation of ser345 is thought tooccur in an ATR-dependent manner, whereasphosphorylation of ser317 in response to IR isreported to be dependent on ATM and NBS1(Gatei et al., 2003; Wang et al., 2012). As predicted,cells with distal 11q loss showed an increase inCHEK1 ser345 phosphorylation after IR, whichwas absent in the cells without distal 11q loss (Fig.5C), which expressed increased phosphorylation ofCHEK1 ser317, suggesting that ATM is primarilyresponsible for the IR-induced DDR. There wasalso an observable increase in ser317 phosphoryla-tion in cells with distal 11q loss, since they retainsome ATM expression (Fig. 5C). These resultsconfirm increased activity of the ATR-CHEK1pathway in the cells with distal 11q loss comparedwith those without distal 11q loss.

Distal 11q Loss is Associated with Increased S and

G2M Checkpoint Arrest After IR

To study the cell cycle profiles of OSCC cells inresponse to DNA damage, flow cytometry was car-ried out on cell lines treated with 5 Gy IR (Table 3).We observed loss of the G1 checkpoint and

increased G2M accumulation after treatment withIR in the three (of five) cell lines with distal 11qloss. Both cell lines without 11q loss, UPCI:SCC066and UPCI:SCC116, had an intact G1 checkpointand when compared with the NHOK control cells,these cell lines showed increased accumulation ofcells in the G2M phase. Untreated UPCI:SCC104(distal 11q loss) cells had a higher percentage ofG2M phase cells compared with untreatedUPCI:SCC066 (no 11q loss) cells (Fig. 6). Following5 Gy IR, UPCI:SCC066 showed an accumulation ofcells in both the G1 and G2M phases, whileUPCI:SCC104 cells showed predominant accumula-tion in the G2M phase.

CHEK1 or ATR siRNA Resensitizes OSCC Cells to

IR-Induced DNA Damage

Our previous studies suggested that the ATR-CHEK1 pathway plays a role in the radioresistantphenotype observed in the cell lines with distal11q loss. To confirm the role of the ATR-CHEK1pathway in radioresistance in cells with distal 11qloss, we used siRNA specific to CHEK1 or ATR toknock down its expression in cell lines and assessthe effect of this knockdown on resistance.CHEK1 expression was inhibited to a high level(~90%) after knockdown using siRNA specific toCHEK1, while a scrambled nonspecific siRNA didnot inhibit CHEK1 expression after 72 h, asassessed by QRT-PCR and immunoblotting (datanot shown). Similarly, ATR expression was alsoinhibited to a high level (>80%) after siRNA-based knockdown. The effect of CHEK1 and ATRknockdown on sensitivity to IR was assessed byclonogenic survival assays. The cells in the clono-genic survival assay were divided into three

Figure 4. Chromosomal segregation defects after irradiation of tumor cells with distal 11qloss compared with cells without distal 11q loss.

136 SANKUNNY ET AL.


groupsUntreated, Mock transfected, andsiCHEK1 transfected. The Mock transfectedgroup represents the average survival values forthe two controls, one a control treated with thetransfection reagent, Lipofectamine 2000, and theother control treated with a pool of non-specificsiRNA. To maximize the effect of CHEK1 or ATRknockdown, cells were harvested and plated forclonogenic survival assay 48 to 72 h post-transfection and treated with IR 24 h later.The survival of untransfected OSCC cells with

distal 11q loss (UPCI:SCC040, UPCI:SCC029B andUPCI:SCC131) was 65.6 6 0.23% and 18.8 6 0.1%at 2.5 and 5 Gy of IR, respectively (Fig. 7A). Mock-transfected cells showed a slight, but not statisti-cally significant decrease in survival compared withthe untreated cells. The survival of siCHEK1 trans-fected cells was 30.2 6 6.1% and 6.4 6 2.4% at 2.5

and 5 Gy of IR, respectively. The results show thathalf as many OSCC cells with distal 11q loss sur-vived an IR dose of 2.5 Gy after transfection withCHEK1 siRNA compared with cells without 11qloss. Radiation treatment of 5 Gy resulted in three-fold lowered survival in CHEK1 siRNA-transfectedcells compared with untreated and mock-transfected cells. No significant survival differencewas observed between the three treatment condi-tions in tumor cells without distal 11q loss.Similarly, ATR knockdown resulted in resensitiza-

tion exclusively of tumor cells with distal 11q loss toradiation treatment. ATR inhibition in OSCC cellswith loss of distal 11q (UPCI:SCC029B, 040 and131) showed a four-fold decrease in survival at 2.5Gy and a 10-fold decrease at 5 Gy (Fig. 7B). OSCCcells without loss of distal 11q (UPCI:SCC116) didnot show a significant difference in survival after

Figure 5. (A) CHEK1 protein expression in OSCC cells shows thatcells without distal 11q loss tend to have higher CHEK1 expressioncompared with cells with 11q loss; (B) After IR treatment, cells withdistal 11q loss showed an increase in CHEK1 expression compared

with cells without distal 11q loss; (C) Phosphorylation status ofCHEK1 ser345 and ser317 showing increased ATR-dependent ser345phosphorylation in cells with distal 11q loss, but no difference inser317 phosphorylation between cells with or without distal 11q loss.



ATR inhibition. These results suggest that theupregulation of the ATR/CHEK1 pathway plays adirect role in the radioresistance of tumor cells withdistal 11q loss. Knockdown of CHEK1 or ATRexpression by siRNA transfection showed thatinhibiting the activity of the ATR-CHEK1 pathwaycan partially resensitize the cells with distal 11q lossto radiation therapy.

Targeted Inhibition of CHEK1 Radiosensitizes

OSCC Cells

To advance these studies towards the clinic, weused a potent and specific targeted CHEK1 small

molecule inhibitor (SMI), PF-00477736, to inhibitthe enzymatic activity of CHEK1, and thenassessed the effect of this inhibition on survivalafter radiation treatment. CHEK1 inhibitors arereported to selectively increase the efficacy of IRand other DNA damaging agents in cells withdefective G1 checkpoints (e.g., as a result of defec-tive TP53 signaling) while producing minimaldamage in cells without G1 checkpoint defects(Blasina et al., 2008). Before SMI treatment, wetested our cell lines for TP53 functionality bytreating the cells with the chemotherapeuticagent, Adriamycin and then assessing the proteinexpression of TP53 and its downstream target, p21by immunoblotting. The results of these studiesidentified OSCC cell lines with distal 11q loss anddefective TP53 signaling for the SMIexperiments.PF-00477736 dose response curves generated in

OSCC cell lines in our lab and by other investiga-tors (Blasina et al., 2008) showed that the optimaldose range required to achieve maximum effect isbetween 360 and 540 nM (data not shown). Todetermine the best time to add the SMI, clono-genic survival assays were carried out after addingthe SMI at various timepoints around the radiationtherapy. Addition of the SMI 8 h before irradiationwas determined to have the maximum effect oncell killing (data not shown). OSCC cell lines

TABLE 3. Results of Cell Cycle Analysis in OSCC Cell Lines in Response to Ionizing Radiationa

Distal 11q loss status Cell lines

Untreated 5 Gy IR (24 h)

G1 (%) S (%) G2M (%) G1 (%) S (%) G2M (%)

No loss UPCI:SCC066 70 13 16 50 16 32UPCI:SCC116 62 15 22 49 15 32

Loss UPCI:SCC084 60 10 29 32 12 54UPCI:SCC104 53 16 30 22 18 57UPCI:SCC131 67 16 15 32 11 55

aOSCC tumor cells were treated with 5 Gy of IR and grown for 24 h, after which they were fixed and stained for cell cycle analysis.

TABLE 2. Frequency of Mitotic Segregation Defects in OSCC Cell Lines

Cell lines Nuclear appearance Untreated IR (118 h) IR (136 h)

No Distal 11q loss (UPCI:SCC066, UPCI:SCC116) Normal 95.5 90.6 80.3Anaphase bridges 0.1 0.9 0.1

Micronuclei 4.3 9 18.3Interphase bridges 0.1 0.4 1.2

Distal 11q loss (UPCI:SCC029B, UPCI:SCC040) Normal 89.7 74.5 39.3Anaphase bridges 0 0.5 0.3

Micronuclei 7.9 21.8 49Interphase bridges 2.4 3.2 11.4

Figure 6. Comparison of cell cycle profiles between UPCI:SCC066(without distal 11q loss) and UPCI:SCC104 (with distal 11q loss) inresponse to IR. UPCI:SCC066 and 104 were either mock-treated ortreated with 5 Gy IR and allowed to recover for 24 h. At the end of24 h, flow cytometric analyses show loss of the G1 cell cycle check-point in UPCI:SCC104, with most cells accumulating in G2M, and accu-mulation of UPCI:SCC066 cells in both the G1 and G2M cell cyclephases. [Color figure can be viewed in the online issue, which is avail-able at wileyonlinelibrary.com.]

138 SANKUNNY ET AL.


UPCI:SCC040 and UPCI:SCC131 (both with dis-tal 11q loss and defective TP53 signaling), whentreated with 540 nM SMI in combination with IRresulted in a two-fold reduction in survival at 2.5Gy and more than four-fold reduction in survivalat 5 Gy, compared with cells treated with IR alone(Fig. 7C). UPCI:SCC116 (no distal 11q loss, defec-tive TP53 signaling) when treated with the samedose of SMI combined with IR showed no differ-ence in survival between the untreated and SMI1 IR-treated cells at 2.5 or 5 Gy. The SMI alone

did not have a significant deleterious effect onthese cells, but was lethal at 360 nM inUPCI:SCC066 cells (no distal 11q loss, defectiveTP53 signaling) and hTERT-transfected humanOKF6 cells (data not shown). Our results indicatethat the CHEK1 SMI, PF-00477736 mirrored theeffect of the siRNA-based knockdown of theATR-CHEK1 pathway in OSCC cell lines.

DISCUSSION

Although distal 11q copy number loss, fromband 11q14 to 11qter has been reported in manytypes of tumors, its significance had not beeninvestigated until our previous study (Parikh et al.,2007). Copy number loss appears to be centeredaround 11q22 to 11q23, in the vicinity of the ATMgene. We showed copy number loss of the regionfrom 11q21 to 11q24 in about 25% of OSCC,breast and ovarian primary tumors and a high fre-quency of loss in OSCC cell lines (Parikh et al.,2007). The Broad Institute TCGA and Tumor-scape websites support our observations, showingATM copy number loss in about 20 to 25% of allcancers and frequencies of loss as high as 54% incutaneous melanomas and 48% in invasive breastadenocarcinomas and HNSCC. We showed previ-ously that distal 11q loss is associated with resist-ance to radiation therapy and our current resultsconfirm and extend this finding (Parikh et al.,2007). AT cells, lacking a functional ATM protein,are radiosensitive because the presence of ATMprotein 1 h postirradiation is essential for the sur-vival of cells (Choi et al., 2010). In contrast, tumorcells with distal 11q loss retain some ATM expres-sion that based on our results, is sufficient to pro-vide the cells the ability to repair their DSB andsurvive IR treatment. AT cells that lack ATM pro-tein upregulate the ATR-CHEK1 pathway result-ing in a prolonged G2M arrest (Wang et al., 2003).Similarily we observed upregulation of the ATR-CHEK1 pathway and cell cycle arrest in the S andG2M phases after IR treatment of tumor cells withdistal 11q loss. S/G2M arrest and subsequentrepair prevents the entry into mitosis of cells withDNA damage, which would lead to mitotic catas-trophe. Instead, these cells seem to repair thedamaged DNA, enabling survival and resulting indecreased sensitivity to IR. Knockdown of theATR-CHEK1 pathway using siRNA or the PfizerCHEK1 inhibitor resulted in resensitization of thecells with distal 11q loss to IR, which led to celldeath by mitotic catastrophe (Fig. 8). SinceCHEK1 inhibitors are being developed by several

Figure 7. Survival after CHEK1 or ATR knockdown using siRNA orthe SMI (PF-00477736) in OSCC cell lines. The surviving fraction oftreated or untreated cells at doses of 0 and 2.5 Gy is plotted on a log-arithmic scale with error bars (6SEM). (A) Cells transfected withsiCHEK1 in the distal 11q loss group (UPCI:SCC029B, 040, and 131)showed 50% decreased survival compared with the untreated andmock-transfected cells. Survival in the no distal 11q loss group wasnot altered significantly by siCHEK1 transfection. (B) HNSCC cellstransfected with siATR in the distal 11q loss group (UPCI:SCC029B,040, 131) showed fourfold decrease in survival at 2.5 Gy and 10-folddecrease at 5 Gy compared with the untreated and mock transfectedcells. (C) Combined IR and SMI treatment was twice as effective in kill-ing cells with distal 11q loss (UPCI:SCC040 and UPCI:SCC131) com-pared with IR alone. UPCI:SCC116 (without distal 11q loss) did notshow a significant difference in survival when treated with combined IRand SMI compared with IR alone.



companies, our results suggest that distal 11q loss(with particular emphasis on ATM copy numberloss) may be a useful companion diagnostic bio-marker to select the subgroup of OSCC patientsexpected to respond poorly to radiation therapy,but better to combined radiation therapy andCHEK1 or ATR inhibition.11q13 amplification correlates with an aggres-

sive tumor phenotype and poor clinical outcome(Jares et al., 1994; Akervall et al., 1995; Michalideset al., 1995; Fracchiolla et al., 1997; Michalideset al., 1997; Mineta et al., 2000; Miyamoto et al.,2003; Wreesmann et al., 2004a, 2004b; Clark et al.,2009). In a series of 11 OSCC cell lines examinedby chromosomal comparative genomic hybridiza-tion (cCGH), four of five patients whose tumorshad 11q13 amplification/distal 11q loss died,whereas all six patients whose tumors lacked11q13 amplification/distal 11q loss survived (Mar-tin et al., 2008). Therefore, we examined 42OSCC patients by Kaplan-Meier disease-specificsurvival analysis and found that patients whosetumors have distal 11q loss have a significantlyworse outcome than patients whose tumors do nothave 11q loss. Based on these results, we concludethat distal 11q loss is a frequent finding in carcino-mas and correlates with a worse clinical outcome,most likely due to therapeutic resistance.Copy number loss of ATM, MRE11A, and

H2AFX results in decreased expression of thesegenes, reduced gH2AX focus formation inresponse to radiation-induced DSBs and chromo-somal instability (Parikh et al., 2007). The intrinsic

level of chromosomal instability observed inOSCC cell lines with distal 11q loss is higher thanin normal keratinocytes or OSCC cells without11q loss (Parikh et al., 2007). We observed a simi-lar increase in mitotic segregation defects inOSCC cells with distal 11q loss after treatmentwith IR compared with OSCC cells without 11qloss. Increased chromosomal instability is consist-ent with a diminished DDR to IR-induced DNAdamage in cells with distal 11q loss. The homolo-gous recombination repair (HRR) mechanism isconsidered to be responsible for repairing IR-induced damage in the S and G2M phases of thecell cycle. Further, the ATR-CHEK1 pathwayrepairs IR-induced DNA damage by a mechanismindependent of nonhomologous end joining(NHEJ), but dependent on HRR (Wang et al.,2004, 2005). Although cells with distal 11q lossusually arrest in G2M phase after IR, the observedincreased accumulation of mitotic segregationdefects shown by our studies suggests repair by anerror-prone mechanism that enables the cells toprogress through mitosis without undergoingmitotic catastrophe. A third DNA repair pathwaygaining significant attention is microhomology-mediated end joining (MMEJ) which might beinvolved in the repair of DSBs in our model. Fur-ther studies are warranted to clarify which of thesethree DNA repair pathways is responsible forradioresistance of tumor cells with distal 11q loss.Our results show that distal 11q loss is associ-

ated with decreased expression of MRE11A, ATM,and H2AFX, which are important components ofthe ATM-CHEK2 DDR pathway (Parikh et al.,2007). Upregulation of the ATR-CHEK1 pathwaymay be compensating for the decreased function-ing of the ATM-CHEK2 pathway. Despite copynumber loss of the CHEK1 gene, CHEK1 expres-sion, especially after IR treatment, appears to beincreased in cells with distal 11q loss. CHEK1expression is regulated by E2F transcription fac-tors, with CHEK1 and E2F1 (transcriptional acti-vator) expression showing a strong correlation(Verlinden et al., 2007). It has also been reportedthat E2F1 levels are higher in OSCC cell linesfrom highly invasive tumors (Zhang et al., 2000),and that phosphorylation of E2F1 by ATM andCHEK2 result in protein stabilization (Inoueet al., 2007). These findings suggest that transcrip-tional upregulation of the CHEK1 gene could beresponsible for increased CHEK1 expression intumor cell lines. This would explain the observedincrease in CHEK1 expression in spite of copynumber loss in OSCC cell lines. Increased ATR-

Figure 8. In cells without loss of distal 11q, the ATM/CHEK2 path-way is primarily responsible for the DDR to IR-induced DSB and theresponse involves G1 cell cycle arrest, resulting in either cell death orrepair of the damaged DNA. Therefore, inhibition of the ATR-CHEK1pathway has no effect on these cells. In cells with loss of distal 11q,the diminished ATM response to IR-induced damage results in upregu-lation of the ATRCHEK1 pathway. This leads to enhancement of Sand G2M checkpoint arrest following DNA damage and protects cellswith DNA damage from entry into S phase or mitosis. Inhibition ofthe ATRCHEK1 pathway using a specific siRNA or a targeted SMIresults in cell death by mitotic catastrophe in cells entering mitosiswith unrepaired DNA damage.

140 SANKUNNY ET AL.


dependent phosphorylation of CHEK1 ser345 incells with distal 11q loss after IR treatment dem-onstrates upregulation of ATR at the apex of theATR-CHEK1 pathway. Activated CHEK1 phos-phorylates members of the CDC25 family, result-ing in their inhibition. CDC25 phosphatases areknown to dephosphorylate cyclin-dependentkinases (CDK), thereby activating them, resultingin cell cycle progression (Uto et al., 2004). Inhibi-tion of CDC25C phosphatase results in cell cyclearrest at the G2M phase. Thus, upregulation of theATR-CHEK1 pathway is likely responsible forthe G2M arrest observed in tumor cells with distal11q loss. This arrest would enable the tumor cellswith distal 11q loss to repair their DNA damageand progress through the cell cycle without under-going mitotic catastrophe. Loss of the G1 check-point in OSCC leads to an increased number ofcells with unrepaired DNA damage entering the Sand the G2M phases of the cell cycle. Thus,OSCC cells with enhanced G2M phase cell cyclecheckpoints may be able to avoid TP53-independent cell death by premature chromosomecondensation/mitotic catastrophe (Fragkos andBeard, 2011) and may have a survival advantage asdemonstrated by our radioresistant OSCC cells.Our results demonstrate that distal 11q loss is

associated with a decreased DDR, radioresistance,an upregulated ATR-CHEK1 pathway, and loss ofthe G1 checkpoint accompanied by G2M arrest.Since the ATR-CHEK1 pathway appears to func-tion as a compensatory mechanism for a downre-gulated ATM-CHEK2 pathway, knocking it downwith siRNA or a SMI should reverse the observedphenotype. Our results demonstrate that CHEK1or ATR knockdown using siRNA or a CHEK1SMI results in radiosensitization of cells with dis-tal 11q loss. Increased cell death in OSCC cellswith distal 11q loss is thought to be due to syn-thetic lethality resulting from loss of the G1 check-point and induced loss of the G2M checkpoint byCHEK1 or ATR knockdown or inhibition. Syn-thetic lethality after loss of cell cycle checkpointsby CHEK1 inhibition and TP53/p21 loss has beendescribed recently (Origanti et al., 2013). Syn-thetic lethality-inducing drugs in the form of sev-eral targeted CHEK1 SMIs with promising effectsare in the pipeline or clinical trials (Busby et al.,2000; Graves et al., 2000; Tao et al., 2009; Shaheenet al., 2011; Bennett et al., 2012; Borst et al., 2012;Ferrao et al., 2012; Karp et al., 2012; Schenk et al.,2012; Wu et al., 2012).In conclusion, our results clearly show that cells

with distal 11q loss have a decreased DDR and

decreased sensitivity to treatment with IR. Upreg-ulation of the ATR-CHEK1 pathway was alsoobserved in these cells and knockdown of CHEK1or ATR resulted in radiosensitization. Since inhi-bition of CHEK1 and ATR using SMIs is beingdeveloped as a promising targeted anti-tumor ther-apy, our results have substantial translationalvalue. Distal 11q loss could be used as a biomarkerpredictive of a less favorable prognosis anddecreased response to conventional radiation ther-apy. It could also be developed as a companiondiagnostic to determine response to CHEK1 orATR inhibition in combination with radiation orchemotherapy in the treatment of OSCCs andother tumors.

ACKNOWLEDGMENTS

The authors are grateful to Ms. Jianhua Zhou forassistance with the FISH analysis and other techni-cal support. The authors are also grateful for thegift of PF-00477736 from Pfizer, Inc., helpful dis-cussions from Sreesha Srinivasan, Ph.D., WendyLevin, M.D., Indrawan McAlpine, Ph.D., BonnieDutcher, Ph.D., Matthew Cotter, Ph.D., and PingChen, Ph.D., and assistance from Mr. Donnie W.Owens and Ms. Donna Stocker. Molecular cytoge-netic studies were carried out in the University ofPittsburgh Cell Culture and Cytogenetics Facility.In addition, we thank Dr. Andrew Remes from theUniversity of Pittsburgh Office of EnterpriseDevelopment and Dr. Maria Vanegas of the Uni-versity of Pittsburgh Office of Technology Manage-ment for helpful discussions. Flow cytometry wascarried out in the UPCI Flow Cytometry Facility.

REFERENCES

Abraham RT. 2004. PI 3-kinase related kinases: Big players instress-induced signaling pathways. DNA Repair 3:883887.

Adams KE, Medhurst AL, Dart DA, Lakin ND. 2006. Recruit-ment of ATR to sites of ionising radiation-induced DNA dam-age requires ATM and components of the MRN proteincomplex. Oncogene 25:38943904.

Akervall JA, Jin Y, Wennerberg JP, Zatterstrom UK, Kjellen E,Mertens F, Willen R, Mandahl N, Heim S, Mitelman F. 1995.Chromosomal abnormalities involving 11q13 are associated withpoor prognosis in patients with squamous cell carcinoma of thehead and neck. Cancer 76:853859.

Akervall JA, Michalides RJ, Mineta H, Balm A, Borg A, DictorMR, Jin Y, Loftus B, Mertens F, Wennerberg JP. 1997. Amplifi-cation of cyclin D1 in squamous cell carcinoma of the head andneck and the prognostic value of chromosomal abnormalitiesand cyclin D1 overexpression. Cancer 79:380389.

Albertson DG. 2006. Gene amplification in cancer. Trends Genet22:447455.

Ambatipudi S, Gerstung M, Gowda R, Pai P, Borges AM, SchafferAA, Beerenwinkel N, Mahimkar MB. 2011. Genomic profilingof advanced-stage oral cancers reveals chromosome 11q altera-tions as markers of poor clinical outcome. PLoS One 6:e17250.



Bakkenist CJ, Kastan MB. 2003. DNA damage activates ATMthrough intermolecular autophosphorylation and dimer dissocia-tion. Nature 421:499506.

Baskaran R, Wood LD, Whitaker LL, Canman CE, Morgan SE,Xu Y, Barlow C, Baltimore D, Wynshaw-Boris A, Kastan MB,Wang JY. 1997. Ataxia telangiectasia mutant protein activates c-Abl tyrosine kinase in response to ionizing radiation. Nature387:516519.

Bennett CN, Tomlinson CC, Michalowski AM, Chu IM, Luger D,Mittereder LR, Aprelikova O, Shou J, Piwinica-Worms H,Caplen NJ, Hollingshead MG, Green JE. 2012. Cross-speciesgenomic and functional analyses identify a combination therapyusing a CHK1 inhibitor and a ribonucleotide reductase inhibitorto treat triple-negative breast cancer. Breast Cancer Res 14:R109.

Beroukhim R, Mermel CH, Porter D, Wei G, Raychaudhuri S,Donovan J, Barretina J, Boehm JS, Dobson J, Urashima M, McHenry KT, Pinchback RM, Ligon AH, Cho YJ, Haery L, GreulichH, Reich M, Winckler W, Lawrence MS, Weir BA, Tanaka KE,Chiang DY, Bass AJ, Loo A, Hoffman C, Prensner J, Liefeld T,Gao Q, Yecies D, Signoretti S, Maher E, Kaye FJ, Sasaki H,Tepper JE, Fletcher JA, Tabernero J, Baselga J, Tsao MS,Demichelis F, Rubin MA, Janne PA, Daly MJ, Nucera C, LevineRL, Ebert BL, Gabriel S, Rustgi AK, Antonescu CR, Ladanyi M,Letai A, Garraway LA, Loda M, Beer DG, True LD, Okamoto A,Pomeroy SL, Singer S, Golub TR, Lander ES, Getz G, SellersWR, Meyerson M. 2010. The landscape of somatic copy-numberalteration across human cancers. Nature 463:899905.

Blasina A, Hallin J, Chen E, Arango ME, Kraynov E, Register J,Grant S, Ninkovic S, Chen P, Nichols T, OConnor P, AnderesK. 2008. Breaching the DNA damage checkpoint via PF-00477736, a novel small-molecule inhibitor of checkpoint kinase1. Mol Cancer Ther 7:23942404.

Borst GR, McLaughlin M, Kyula JN, Neijenhuis S, Khan A,Good J, Zaidi S, Powell NG, Meier P, Collins I, Garrett MD,Verheij M, Harrington KJ. 2012. Targeted Radiosensitization bythe Chk1 Inhibitor SAR-020106. Int J Radiat Oncol Biol Phys85:11101118.

Bray F, Ren JS, Masuyer E, Ferlay J. 2013. Global estimates ofcancer prevalence for 27 sites in the adult population in 2008.Int J Cancer 132:11331145.

Busby EC, Leistritz DF, Abraham RT, Karnitz LM, Sarkaria JN.2000. The radiosensitizing agent 7-hydroxystaurosporine (UCN-01) inhibits the DNA damage checkpoint kinase hChk1. CancerRes 60:21082112.

Byun TS, Pacek M, Yee MC, Walter JC, Cimprich KA. 2005.Functional uncoupling of MCM helicase and DNA polymeraseactivities activates the ATR-dependent checkpoint. Genes Dev19:10401052.

Choi S, Gamper AM, White JS, Bakkenist CJ. 2010. Inhibition ofATM kinase activity does not phenocopy ATM protein disrup-tion: Implications for the clinical utility of ATM kinase inhibi-tors. Cell Cycle 9:40524057.

Clark ES, Brown B, Whigham AS, Kochaishvili A, Yarbrough WG,Weaver AM. 2009. Aggressiveness of HNSCC tumors dependson expression levels of cortactin, a gene in the 11q13 amplicon.Oncogene 28:431444.

Cortez D, Guntuku S, Qin J, Elledge SJ. 2001. ATR and ATRIP:Partners in checkpoint signaling. Science 294:17131716.

Cuadrado M, Martinez-Pastor B, Murga M, Toledo LI, Gutierrez-Martinez P, Lopez E, Fernandez-Capetillo O. 2006. ATM regu-lates ATR chromatin loading in response to DNA double-strand breaks. J Exp Med 203:297303.

Desantis C, Naishadham D, Jemal A. 2013. Cancer statistics forAfrican Americans, 2013. CA Cancer J Clin 63:151166.

Dickson MA, Hahn WC, Ino Y, Ronfard V, Wu JY, Weinberg RA,Louis DN, Li FP, Rheinwald JG. 2000. Human keratinocytesthat express hTERT and also bypass a p16(INK4a)-enforcedmechanism that limits life span become immortal yet retainnormal growth and differentiation characteristics. Mol Cell Biol20:14361447.

Edelmann J, Holzmann K, Miller F, Winkler D, Buhler A, ZenzT, Bullinger L, Kuhn MW, Gerhardinger A, Bloehdorn J,Radtke I, Su X, Ma J, Pounds S, Hallek M, Lichter P, KorbelJ, Busch R, Mertens D, Downing JR, Stilgenbauer S, DohnerH. 2012. High-resolution genomic profiling of chronic lympho-cytic leukemia reveals new recurrent genomic alterations. Blood120:47834794.

Fenech M, Kirsch-Volders M, Natarajan AT, Surralles J, CrottJW, Parry J, Norppa H, Eastmond DA, Tucker JD, Thomas P.2011. Molecular mechanisms of micronucleus, nucleoplasmic

bridge and nuclear bud formation in mammalian and humancells. Mutagenesis 26:125132.

Ferlay J, Shin HR, Bray F, Forman D, Mathers C, Parkin DM.2010. GLOBOCAN 2008 v1.2. cancer Incidence and MortalityWorldwide: IARC CancerBase No.10. Cancer IAFRO. Lyon,France: International Agency for Research on Cancer.

Fernandez-Capetillo O, Lee A, Nussenzweig M, Nussenzweig A.2004. H2AX: the histone guardian of the genome. DNA Repair3:959967.

Ferrao PT, Bukczynska EP, Johnstone RW, McArthur GA. 2012.Efficacy of CHK inhibitors as single agents in MYC-driven lym-phoma cells. Oncogene 31:16611672.

Fracchiolla NS, Pruneri G, Pignataro L, Carboni N, Capaccio P,Boletini A, Buffa R, Neri A. 1997. Molecular and immunohisto-chemical analysis of the bcl-1/cyclin D1 gene in laryngeal squa-mous cell carcinomas: correlation of protein expression with lymphnode metastases and advanced clinical stage. Cancer 79:11141121.

Fragkos M, Beard P. 2011. Mitotic catastrophe occurs in theabsence of apoptosis in p53-null cells with a defective G1checkpoint. PLoS One 6:e22946.

Gatei M, Sloper K, Sorensen C, Syljuasen R, Falck J, Hobson K,Savage K, Lukas J, Zhou BB, Bartek J, Khanna KK. 2003.Ataxia-telangiectasia-mutated (ATM) and NBS1-dependentphosphorylation of Chk1 on Ser-317 in response to ionizingradiation. J Biol Chem 278:1480614811.

George RE, Attiyeh EF, Li S, Moreau LA, Neuberg D, Li C, FoxEA, Meyerson M, Diller L, Fortina P, Look AT, Maris JM. 2007.Genome-wide analysis of neuroblastomas using high-density sin-gle nucleotide polymorphism arrays. PLoS One 2:e255.

Gibcus JH, Kok K, Menkema L, Hermsen MA, Mastik M, KluinPM, van der Wal JE, Schuuring E. 2007. High-resolution map-ping identifies a commonly amplified 11q13.3 region containingmultiple genes flanked by segmental duplications. Hum Genet121:187201.

Gollin SM. 2001. Chromosomal alterations in squamous cell carci-nomas of the head and neck: window to the biology of disease.Head Neck 23:238253.

Graves PR, Yu L, Schwarz JK, Gales J, Sausville EA, OConnorPM, Piwnica-Worms H. 2000. The Chk1 protein kinase and theCdc25C regulatory pathways are targets of the anticancer agentUCN-01. J Biol Chem 275:56005605.

Ha PK, Califano JA, 3rd. 2002. The molecular biology of laryngealcancer. Otolaryngol Clin North Am 35:9931012.

Harnden DG. 1994. The nature of ataxia-telangiectasia: problemsand perspectives. Int J Radiat Biol 66:S1319.

Henson BJ, Bhattacharjee S, ODee DM, Feingold E, Gollin SM.2009. Decreased expression of miR-125b and miR-100 in oralcancer cells contributes to malignancy. Genes ChromosomesCancer 48:569582.

Heron M, Hoyert DL, Murphy SL, Xu J, Kochanek KD, Tejada-Vera B. 2009. Deaths: final data for 2006. Natl Vital Stat Rep57:1134.

Huang X, Gollin SM, Raja S, Godfrey TE. 2002. High-resolutionmapping of the 11q13 amplicon and identification of a gene,TAOS1, that is amplified and overexpressed in oral cancer cells.Proc Natl Acad Sci USA 99:1136911374.

Inoue Y, Kitagawa M, Taya Y. 2007. Phosphorylation of pRB atSer612 by Chk1/2 leads to a complex between pRB and E2F-1after DNA damage. EMBO J 26:20832093.

Jares P, Fernandez PL, Campo E, Nadal A, Bosch F, Aiza G,Nayach I, Traserra J, Cardesa A. 1994. PRAD-1/cyclin D1 geneamplification correlates with messenger RNA overexpressionand tumor progression in human laryngeal carcinomas. CancerRes 54:48134817.

Jiang K, Pereira E, Maxfield M, Russell B, Goudelock DM,Sanchez Y. 2003. Regulation of Chk1 includes chromatin associ-ation and 14-3-3 binding following phosphorylation on Ser-345.J Biol Chem 278:2520725217.

Jin C, Jin Y, Gisselsson D, Wennerberg J, Wah TS, Stromback B,Kwong YL, Mertens F. 2006. Molecular cytogenetic characteri-zation of the 11q13 amplicon in head and neck squamous cellcarcinoma. Cytogenet Genome Res 115:99106.

Jin Y, Hoglund M, Jin C, Martins C, Wennerberg J, Akervall J,Mandahl N, Mitelman F, Mertens F. 1998. FISH characteriza-tion of head and neck carcinomas reveals that amplification ofband 11q13 is associated with deletion of distal 11q. GenesChromosomes Cancer 22:312320.

Karp JE, Thomas BM, Greer JM, Sorge C, Gore SD, Pratz KW,Smith BD, Flatten KS, Peterson K, Schneider P, Mackey K,

142 SANKUNNY ET AL.


Freshwater T, Levis MJ, McDevitt MA, Carraway HE,Gladstone DE, Showel MM, Loechner S, Parry DA, HorowitzJA, Isaacs R, Kaufmann SH. 2012. Phase I and pharmacologictrial of cytosine arabinoside with the selective checkpoint 1inhibitor Sch 900776 in refractory acute leukemias. Clin CancerRes 18:67236731.

Keith CT, Schreiber SL. 1995. PIK-related kinases: DNA repair,recombination, and cell cycle checkpoints. Science 270:5051.

Khanna KK, Jackson SP. 2001. DNA double-strand breaks: Signal-ing, repair and the cancer connection. Nat Genet 27:247254.

Lavin MF, Shiloh Y. 1997. The genetic defect in ataxia-telangiec-tasia. Annu Rev Immunol 15:177202.

Martin C, Reshmi S, Ried T, Gottberg W, Wilson J, Reddy J,Khanna P, Johnson J, Myers E, Gollin S. 2008. Chromosomalimbalances in oral squamous cell carcinoma: examination of 31cell lines and review of the literature. Oral Oncol 44:369382.

Mermel CH, Schumacher SE, Hill B, Meyerson ML, BeroukhimR, Getz G. 2011. GISTIC2.0 facilitates sensitive and confidentlocalization of the targets of focal somatic copy-number altera-tion in human cancers. Genome Biol 12:R41.

Michalides R, van Veelen N, Hart A, Loftus B, Wientjens E,Balm A. 1995. Overexpression of cyclin D1 correlates withrecurrence in a group of forty-seven operable squamous cell car-cinomas of the head and neck. Cancer Res 55:975978.

Michalides RJ, van Veelen NM, Kristel PM, Hart AA, Loftus BM,Hilgers FJ, Balm AJ. 1997. Overexpression of cyclin D1 indi-cates a poor prognosis in squamous cell carcinoma of the headand neck. Arch Otolaryngol Head Neck Surg 123:497502.

Mineta H, Miura K, Takebayashi S, Ueda Y, Misawa K, HaradaH, Wennerberg J, Dictor M. 2000. Cyclin D1 overexpressioncorrelates with poor prognosis in patients with tongue squamouscell carcinoma. Oral oncology 36:194198.

Miyamoto R, Uzawa N, Nagaoka S, Hirata Y, Amagasa T. 2003.Prognostic significance of cyclin D1 amplification and overexpres-sion in oral squamous cell carcinomas. Oral oncology 39:610618.

Myers JS, Cortez D. 2006. Rapid activation of ATR by ionizingradiation requires ATM and Mre11. J Biol Chem 281:93469350.

Niida H, Katsuno Y, Banerjee B, Hande MP, Nakanishi M. 2007.Specific role of Chk1 phosphorylations in cell survival andcheckpoint activation. Mol Cell Biol 27:25722581.

Origanti S, Cai SR, Munir AZ, White LS, Piwnica-Worms H.2013. Synthetic lethality of Chk1 inhibition combined with p53and/or p21 loss during a DNA damage response in normal andtumor cells. Oncogene 32:577588.

Pandita TK, Westphal CH, Anger M, Sawant SG, Geard CR,Pandita RK, Scherthan H. 1999. Atm inactivation results inaberrant telomere clustering during meiotic prophase. Mol CellBiol 19:50965105.

Parikh RA, White JS, Huang X, Schoppy DW, Baysal BE,Baskaran R, Bakkenist CJ, Saunders WS, Hsu LC, Romkes M,Gollin SM. 2007. Loss of distal 11q is associated with DNArepair deficiency and reduced sensitivity to ionizing radiation inhead and neck squamous cell carcinoma. Genes ChromosomesCancer 46:761775.

Reshmi SC, Roychoudhury S, Yu Z, Feingold E, Potter D,Saunders WS, Gollin SM. 2007. Inverted duplication pattern inanaphase bridges confirms the breakage-fusion-bridge (BFB)cycle model for 11q13 amplification. Cytogenet Genome Res116:4652.

Savitsky K, Bar-Shira A, Gilad S, Rotman G, Ziv Y, Vanagaite L,Tagle DA, Smith S, Uziel T, Sfez S, Ashkenazi M, Pecker I,Frydman M, Harnik R, Patanjali SR, Simmons A, Clines GA,Sartiel A, Gatti RA, Chessa L, Sanal O, Lavin MF, Jaspers NG,Taylor AM, Arlett CF, Miki T, Weissman SM, Lovett M,Collins FS, Shiloh Y. 1995. A single ataxia telangiectasia genewith a product similar to PI-3 kinase. Science 268:17491753.

Schenk EL, Koh BD, Flatten KS, Peterson KL, Parry D, HessAD, Smith BD, Karp JE, Karnitz LM, Kaufmann SH. 2012.Effects of selective checkpoint kinase 1 inhibition on cytara-bine cytotoxicity in acute myelogenous leukemia cells in vitro.Clin Cancer Res 18:53645373.

Schraml P, Kononen J, Bubendorf L, Moch H, Bissig H, NocitoA, Mihatsch MJ, Kallioniemi OP, Sauter G. 1999. Tissue micro-arrays for gene amplification surveys in many different tumortypes. Clin Cancer Res 5:19661975.

Shaheen M, Allen C, Nickoloff JA, Hromas R. 2011. Syntheticlethality: exploiting the addiction of cancer to DNA repair.Blood 117:60746082.

Shiloh Y. 2003. ATM and related protein kinases: Safeguardinggenome integrity. Nat Rev Cancer 3:155168.

Shuster MI, Han L, Le Beau MM, Davis E, Sawicki M, LeseCM, Park NH, Colicelli J, Gollin SM. 2000. A consistent pat-tern of RIN1 rearrangements in oral squamous cell carcinomacell lines supports a breakage-fusion-bridge cycle model for11q13 amplification. Genes Chromosomes Cancer 28:153163.

Siegel R, Naishadham D, Jemal A. 2013. Cancer statistics, 2013.CA Cancer J Clin 63:1130.

Stracker TH, Theunissen JW, Morales M, Petrini JH. 2004. TheMre11 complex and the metabolism of chromosome breaks:The importance of communicating and holding things together.DNA Repair 3:845854.

Stucki M, Clapperton JA, Mohammad D, Yaffe MB, Smerdon SJ,Jackson SP. 2005. MDC1 directly binds phosphorylated histoneH2AX to regulate cellular responses to DNA double-strandbreaks. Cell 123:12131226.

Swarts DR, Claessen SM, Jonkers YM, van Suylen RJ, DingemansAM, de Herder WW, de Krijger RR, Smit EF, Thunnissen FB,Seldenrijk CA, Vink A, Perren A, Ramaekers FC, Speel EJ. 2011.Deletions of 11q22.3-q25 are associated with atypical lung carci-noids and poor clinical outcome. Am J Pathol 179:11291137.

Tao Y, Leteur C, Yang C, Zhang P, Castedo M, Pierre A,Golsteyn RM, Bourhis J, Kroemer G, Deutsch E. 2009. Radio-sensitization by Chir-124, a selective CHK1 inhibitor: effects ofp53 and cell cycle checkpoints. Cell Cycle 8:11961205.

Uto K, Inoue D, Shimuta K, Nakajo N, Sagata N. 2004. Chk1,but not Chk2, inhibits Cdc25 phosphatases by a novel commonmechanism. EMBO J 23:33863396.

Verlinden L, Vanden Bempt I, Eelen G, Drijkoningen M,Verlinden I, Marchal K, De Wolf-Peeters C, Christiaens MR,Michiels L, Bouillon R, Verstuyf A. 2007. The E2F-regulatedgene Chk1 is highly expressed in triple-negative estrogenreceptor /progesterone receptor /HER-2 breast carcinomas. Can-cer Res 67:65746581.

Wang H, Powell SN, Iliakis G, Wang Y. 2004. ATR affecting cellradiosensitivity is dependent on homologous recombinationrepair but independent of nonhomologous end joining. CancerRes 64:71397143.

Wang H, Hu B, Liu R, Wang Y. 2005. CHK1 affecting cell radio-sensitivity is independent of non-homologous end joining. CellCycle 4:300303.

Wang J, Han X, Zhang Y. 2012. Autoregulatory mechanisms of phos-phorylation of checkpoint kinase 1. Cancer Res 72:37863794.

Wang X, Khadpe J, Hu B, Iliakis G, Wang Y. 2003. An overactivatedATR/CHK1 pathway is responsible for the prolonged G2 accu-mulation in irradiated AT cells. J Biol Chem 278:3086930874.

White JS, Weissfeld JL, Ragin CC, Rossie KM, Martin CL,Shuster M, Ishwad CS, Law JC, Myers EN, Johnson JT, GollinSM. 2007. The influence of clinical and demographic risk fac-tors on the establishment of head and neck squamous cell carci-noma cell lines. Oral Oncol 43:701712.

Wreesmann VB, Shi W, Thaler HT, Poluri A, Kraus DH, PfisterD, Shaha AR, Shah JP, Rao PH, Singh B. 2004a. Identificationof novel prognosticators of outcome in squamous cell carcinomaof the head and neck. J Clin Oncol 22:39653972.

Wreesmann VB, Wang D, Goberdhan A, Prasad M, Ngai I,Schnaser EA, Sacks PG, Singh B. 2004b. Genetic abnormalitiesassociated with nodal metastasis in head and neck cancer. Head& Neck 26:1015.

Wu W, Bi C, Bence AK, Um SL, Yan B, Starling JJ, Marshall MS,Beckmann RP. 2012. Antitumor activity of Chk1 inhibitorLY2606368 as a single agent in SW1990 human pancreas ortho-topic tumor model. Proceedings of the 103rd Annual Meetingof the American Association for Cancer Research, 2012 Mar 31-Apr 4, Chicago, IL. Philadelphia (PA): AACR; Cancer Res2012;72(8 Suppl):Abstract nr 1776.

Zhang SY, Liu SC, Johnson DG, Klein-Szanto AJ. 2000. E2F-1gene transfer enhances invasiveness of human head and neckcarcinoma cell lines. Cancer Res 60:59725976.

Zou L, Elledge SJ. 2003. Sensing DNA damage through ATRIPrecognition of RPA-ssDNA complexes. Science 300:15421548.



l

chk1 atr

Documents

number loss of distal

breakageand loss of

oscc cell linesand distal

atrchek1 pathway inhibition

compensatoryatrchek1

chek1 reversesradioresistance

atm copy number loss

ir treatment of cell