yu238259 is a novel inhibitor of homology- dependent dna …€¦ · 01/10/2015 · published...

DNA Damage and Repair

YU238259 Is a Novel Inhibitor of Homology-Dependent DNA Repair That Exhibits SyntheticLethality and Radiosensitization in Repair-Deficient TumorsGregory C. Stachelek1,2, Elizabeth Peterson-Roth1, Yanfeng Liu1, Rafael J. Fernandez III1,Luke R.G. Pike1, Jack M. Qian1, Laura Abriola3, Denton Hoyer3,William Hungerford3,Janie Merkel3, and Peter M. Glazer1,2

Abstract

Radiotherapy and DNA-damaging chemotherapy are frequent-ly utilized in the treatment of solid tumors. Innate or acquiredresistance to these therapies remains a major clinical challenge inoncology. The development of small molecules that sensitizecancers to established therapies represents an attractive approachto extending survival and quality of life in patients. Here, wedemonstrate that YU238259, a member of a novel class of DNAdouble-strand break repair inhibitors, exhibits potent syntheticlethality in the setting of DNA damage response and DNA repairdefects. YU238259 specifically inhibits homology-dependentDNA repair, but not non-homologous end-joining, in cell-basedGFP reporter assays. Treatment with YU238259 is not onlysynergistic with ionizing radiation, etoposide, and PARP inhibi-tion, but this synergism is heightened by BRCA2 deficiency.

Further, growth of BRCA2-deficient human tumor xenografts innudemice is significantly delayed byYU238259 treatment even inthe absence of concomitant DNA-damaging therapy. The cyto-toxicity of these small molecules in repair-deficient cells resultsfrom an accumulation of unresolved DNA double-strand breaks.These findings suggest that YU238259 or related small moleculesmay have clinical benefit to patients with advanced BRCA2-negative tumors, either as a monotherapy or as an adjuvant toradiotherapy and certain chemotherapies.

Implications: We have identified a novel series of compoundsthat demonstrate synthetic lethality in DNA repair–deficient celland animal models and have strong potential for clinical trans-lation. Mol Cancer Res; 13(10); 1–9. �2015 AACR.

IntroductionIn cancer therapy, gene expression profiling and mutational

analyses are increasingly being used to inform the optimal choiceof treatment and overall prognosis. Personalized cancer therapiesare rapidly becoming the standard of care in clinical practice (1).Thus, novel drugs targeting molecular pathways known to bedysregulated in cancer may have utility as first-line treatments orwhen treatment has failed. Frequently, the inability to properlycoordinate repair of damaged DNA underlies tumorigenesis anddisease progression in malignancies (2–6). Human cancer syn-dromes have been linked tomutations in nearly every DNA repairpathway. These pathways often exhibit redundancy, and inhibi-tion of additional repair factors may induce synthetic lethality inrepair-deficient tumors while sparing healthy tissue (7–9). DNA

double-strand break (DSB) repair pathways are particularly vul-nerable, as shown by recent observations that inactivation of theRAD52 recombinase is lethal in the context of BRCA1, BRCA2, orPALB2 loss (10).

Synthetic lethality is not limited to defects in a single DNArepair pathway and may also occur across complementary repairpathways. This is perhaps best characterized in the relationshipbetween the homology-dependent repair (HDR) and base exci-sion repair (BER) pathways (11). Incompletely resolved lesionsthat are substrates of BER (e.g., alkylated and damaged bases)generate DNA single-strand breaks (SSB) that persist into S-phase,leading to stalling and collapse of DNA replication forks andinduction of DSBs that must be resolved by HDR. This is exem-plified by the enhanced toxicity of PARP inhibitors in BRCA1- andBRCA2-deficient cells (12). Clinical trials of PARP inhibitors areongoing (13–16). Notably, olaparib has shown potential in thetreatment of BRCA-mutant ovarianmalignancies, which are oftenmetastatic at time of diagnosis, limiting therapeutic options.

Genetic instability is a hallmark of cancer and can quicklyconfer resistance to even targeted monotherapies or drug com-binations. Inhibition of DNA repair pathways can decreaseresistance by altering the response of tumor cells to DNA-damaging chemotherapy and ionizing radiation (IR). We havepreviously shown that inhibition of the BER pathway in aBRCA2-mutant background sensitizes cells to the DNA-meth-ylating agent temozolomide (17). Conversely, even in anHDR-defective background, intact DNA mismatch repair is

1Department of Therapeutic Radiology,Yale School of Medicine, NewHaven, Connecticut. 2Department of Genetics, Yale School of Medi-cine, New Haven, Connecticut. 3Yale Center for Molecular Discovery,West Haven, Connecticut.

Gregory C. Stachelek and Peter M. Glazer both wrote the article.

Corresponding Author: Peter M. Glazer, Yale School of Medicine, 333 Cedar St,P.O. Box 208040, New Haven, CT 06520. Phone: 203-737-2788; Fax: 203-737-1467; E-mail: [email protected]

doi: 10.1158/1541-7786.MCR-15-0036

�2015 American Association for Cancer Research.

MolecularCancerResearch

www.aacrjournals.org OF1

Research. on January 22, 2021. © 2015 American Association for Cancermcr.aacrjournals.org Downloaded from

Published OnlineFirst June 26, 2015; DOI: 10.1158/1541-7786.MCR-15-0036

http://mcr.aacrjournals.org/

necessary for certain DNA lesions to induce cytotoxicity, ren-dering mismatch repair–defective cells highly resistant to alkyl-ation and crosslinking (18, 19). Efforts to identify effectiveradiosensitizing agents have also met with recent success andcontinue to represent an area of high translational potential(20). Given that improving genomic techniques will soonenable profiling of entire gene pathways in tumors, it is likelythat selection of cancer therapies will increasingly be guided byelucidation of the functional status of DNA repair pathways.

We hypothesized that novel small molecules that exploit com-mon DNA repair deficiencies in tumors and synergize withvalidated chemotherapeutic agents may provide the basis forrapid clinical translation. This led us to develop a syntheticlethality screen for small molecules that were preferentially toxicto cells lacking the well-studied repair proteins FANCD2 andBRCA2, but which exhibited low absolute toxicity in isogenic,repair-proficient paired cells. Using the commercially availableChemBridge DIVERSet, we have identified several structurallyrelated compounds that fulfill these criteria and, to our knowl-edge, have not been reported in the literature as antineoplasticagents. Each of the compounds possesses amodified sulfonamidebackbone that appears essential to the cellular activity of this class.Iterative rounds of chemical synthesis and structure–activity rela-tionship analysis of screening hits have produced a lead com-pound, YU238259, with increased potency and solubility.

Our work demonstrates that YU238259 is an inhibitor of theHDR pathway, yielding potent synthetic lethality in the context ofadditional repair defects and allowing it to synergize effectivelywith agents that induce DSBs in cells. Further, we show that thesesulfonamide compounds do not intercalate into DNA and do notinhibit PARP activity. YU238259 and a related compound thatwas an early hit in our screen, YU128440, also demonstratesynthetic lethality in a mouse model using BRCA2-deficienthuman tumor xenografts. These characteristics make YU238259and related molecules promising candidates for translationaldevelopment.

Materials and MethodsChemicals

Etoposide, doxorubicin, and caffeine were purchased fromSigma. Olaparib was purchased from Selleck. Rabbit-derivedantibodies against PTEN (#9552), ATM (#2873), NBS1(#3002), ATRIP (#2737), b-tubulin (#2146), and mouse-derivedantibody against b-actin (#3700) were purchased from CellSignaling Technology. Pooled siRNAs against ATRIP and NBS1,as well as nontargeted control siRNA, were purchased fromDharmacon.

Cell linesDLD-1 and DLD-1 BRCA2-KO human colorectal adenocarci-

noma cell lines (HorizonDiscovery) were cultured inMcCoy's 5Amedium with 10% FBS. U251 glioma cells with inducible PTEN(21) were cultured in DMEM with 10% FBS, 0.5 mg/mL G418,and 10 mg/mL blasticidin. Doxycycline (1 mg/mL) was used forinduction and PTEN expression was confirmed by Western blot.PD20 and PD20þFANCD2 cells were from the OHSU FA CellRepository and were grown in MEM-alpha with 15% FBS. Allother cell lines were cultured in DMEM with 10% FBS. PEO1/4cells (22) were a gift of Dr. Toshiyasu Tanaguchi. HCC1937þvec-tor andHCC1937þBRCA1 (23) cells were a gift ofDr. Zhong Yun.

U2OS DR-GFP and U2OS EJ5-GFP cell lines have been describedpreviously (24, 25). GM05849 ATM-deficient fibroblasts werefrom the ATCC and were stably complemented with wild-typeATM or vector (P.M. Glazer, unpublished data). U2OS cell linewas from the ATCC. All paired lines were tested for repair proteinexpression by Western blot.

Screening librariesChemBridgeDIVERSet and other small-molecule libraries were

formatted as 10 mmol/L DMSO stocks in 384-well plates. Fordose response curves of individual screening compounds, 100mmol/L stocks were prepared and serially diluted by 2-fold inDMSO.

Synthetic lethality screening assayCells were seeded in 20 mL media onto white clear-bottom

384-well plates (Corning) using a MultiDrop Combi ReagentDispenser (Thermo Scientific). Primary screening used PD20 �FANCD2 cells plated at 750 cells/well. Follow-up screening anddose response curves used PEO1 and PEO4 cells plated at 750 and1500 cells/well, respectively. After seeding, the assay plates werecentrifuged at 500 rpm for 10 seconds and incubated overnight.Compound (20 nL) was then transferred from the source plate tothe assay plates using an Aquarius (Tecan) with a 384-well pintool (V&P Scientific). Final concentration of screening com-pounds was 10 mmol/L in 0.1% DMSO. Tamoxifen (final con-centration 30 mmol/L for PD20 cells and 60 mmol/L for PEO1/4cells) was added to positive control wells. Assay plates werecentrifuged at 500 rpm for 10 seconds and then incubated for72 hours. CellTiter-Glo (Promega) was used to measure cellviability in the assaywells according to themanufacturer's instruc-tions. 20 mL/well of CellTiter-Glo was added to each assay plateusing a MultiDrop Combi. Plates were shaken on a ThermomixerR (Eppendorf) at 1,100 rpm for 1 minute and incubated in thedark for 10 minutes at room temperature. Luminescence wasmeasured using an Envision plate reader (PerkinElmer) with0.3 second sampling time per well. Because the positive controlyielded loss of signal, Z' scores were calculated by 1 – (3scþ þ3sc–)/ j mc– – mcþ j , where m and s represent the mean and SD,respectively, of the positive (cþ) and negative (c–) control data.

PARP enzymatic assayPARP activity was assessed using a Universal PARP Chemilu-

minescent Assay (Trevigen) according to the manufacturer'sinstructions. 3-aminobenzamide served as a positive control forPARP inhibition.

Circular dichroism spectroscopyYU128440, doxorubicin, caffeine, or DMSO vehicle were

added at 1:1,000 to a 1 mg/mL solution of 8 kb plasmid DNA inTE buffer to yield drug concentrations of 10 mmol/L, and incu-bated at room temperature for 5minutes. Sampleswere read in 10mm path length quartz micro-cuvettes (Hellma), and CD absor-bance was averaged over four measurements from 230 nm to 330nm using a Chirascan spectrometer (Applied Photophysics).

DSB repair reporter assaysU2OS reporter cell lines (DR-GFP or EJ5-GFP) were pretreated

in triplicate with varying concentrations of YU238259 for 24hours, after which 4 mg of SCE-I plasmid was transfected into

Stachelek et al.

Mol Cancer Res; 13(10) October 2015 Molecular Cancer ResearchOF2




1 � 106 cells/replicate using an Amaxa Nucleofector (Lonza).Transfected cells were reseeded on 6-well plates and cultured withYU238259 for an additional 72 hours. The percentage of GFP-positive cells was quantified by flow cytometry. Data analysis wasperformed using FlowJo software (Tree Star Inc.). Error barsrepresent the SD.

ImmunofluorescenceU2OS cells were pretreated with 25 mmol/L YU238259 or

DMSO vehicle for 1 hour, and cells were then irradiated with10 Gy IR. Cells were fixed at 8 hours after IR, stained withantibodies and Hoechst dye, and imaged. Foci were quantifiedusing the InCell Analyzer algorithm developed by YCMD. Cellswere scored as foci-positive if they contained �15 foci (BRCA1,pDNA-PK) or �20 foci (53BP1).

Neutral comet assayDLD-1 and DLD-1 BRCA2-KO cells were treated with

YU238259 for 4 hours and then irradiated with 5 Gy IR. Cellswere then harvested either immediately or 24 hours after irradi-ation, and a neutral comet assay was performed per manufac-turer's instructions (Trevigen). Comet tail moments were ana-

lyzedusingCometScore software (TriTek) for at least 50 randomlychosen cells per condition. Error bars represent the SD.

Cell cycle analysisDLD-1/BRCA2-KO cells were treated with YU238259 for 6

hours and then harvested by trypsinization. Cells were fixed in70% ethanol, treated with 100 mg/mL RNase, and stained withpropidium iodide. DNA content was analyzed by flow cytometryfor 10,000 cells per condition.

Clonogenic survival assaysCell lineswere plated in triplicate at 500 to1,000 cells/well in 6-

well plates. Cells were treated with YU238259 for 48 hours. Forsynergism assays, cells were pretreated with YU238259 for 24hours, then irradiated or treated with etoposide or olaparib for anadditional 24 hours in the presence of YU238259. For siRNAexperiments, U2OS cells were transfected with ATRIP or NBS1pooled siRNA or nontargeting control per manufacturer instruc-tions, incubated for 72 hours to ensuremaximal knockdown, andthen treated with YU238259 for 48 hours. Cells were cultured for1 to 2 weeks until well-defined colonies had formed, replacingculture media every 2 to 3 days. Cells were briefly permeabilized

A

B

C

FANCD2+/+ FANCD2-/- BRCA2+/+ BRCA2-/-

YU126919

Screening library15,040 cmps

Synthe�c lethalitywith FANCD2 loss

Selec�on of leadcompounds

Valida�on with 12-ptdose responses

Synthe�c lethalitywith BRCA2 loss

8.0Ind.16.21Ind.

YU127190 19.291.1Ind.Ind.

YU128355 4.5313.3 --

YU128440 6.5Ind.4.689.59

YU175534 -- 5.5Ind.

YU238259 -- 8.7Ind.

D

YU238259

YU175534

YU128440YU128355

YU126919 YU127190

LD50 (mmol/L) values of screening compounds

BRCA2+/+ BRCA2-/-

YU126919 69%<20%

YU127190 88%61%

YU128440 67%<20%

YU175534 100%46%

YU238259 98%27%

Maximal percent cytotoxicityof screening compounds

* Ind. = indeterminate; did notachieve 50% cytotoxicity over0-100 mmol/L dose range

OO

OO

O

OO

O

NN

NN

H

NHN

H

HS

OON NH

S O

CICI

O

O

O

O O

OOOO O

BrH

NH

NH

S

SS

S

S

S

CI

CI

N

Figure 1.High-throughput screening reveals a novel class of compounds with synthetic lethality in the context of HDR defects. A, methodology of synthetic lethality screen.B, LD50 values and maximal percent effect values for given sulfonamide compounds were calculated in cells proficient or deficient in either BRCA2 orFANCD2. C, structures of sulfonamide compounds identified in screen of ChemBridge DIVERSet small-molecule library or through structural homology screening.D, structure of lead agent, YU238259.

YU238259: A Novel Inhibitor of Homology-Dependent DNA Repair

www.aacrjournals.org Mol Cancer Res; 13(10) October 2015 OF3




with 0.9% saline solution and stained with crystal violet in 80%methanol. Colonies of�50 cells were then counted visually. Datapoints represent the average of the three values, and error barsrepresent the SD.

Mouse tumor xenografts069(nu)/070(nu/þ) athymic nude mice were purchased

from Harlan. At 4 to 5 weeks of age, mice were injectedsubcutaneously with 3 � 106 DLD-1 or DLD-1 BRCA2-KOcells suspended in 100 mL PBS. Tumor take rate was >80%.When tumors reached 100 mm3 geometric mean volume, themice were injected with 3 mg/kg YU238259 or its 3:1 DMSO:PBS vehicle, or 5 mg/kg YU128440 or its 1:19 DMSO:PBSvehicle (IP, 100 mL total in each case). Treatment was repeated3�/week (Monday/Wednesday/Friday) for a total of 12 dosesof YU238259 and 4 doses of YU128440. Tumor growth wasassessed by external caliper. Mice were euthanized when indi-vidual tumor volumes exceeded 1,000 mm3.

ResultsHigh-throughput screening reveals a class of small moleculesthat are synthetically lethal to HDR-deficient cells

To identify small molecules with synthetic lethality in thecontext of HDR defects, our screen made use of paired cell linesdeficient and proficient in FANCD2 (PD20) and BRCA2(PEO1/4) activity (Fig. 1A). Following optimization, Z' valueswere >0.5, indicating excellent discrimination between positiveand negative controls. Screening compounds that demonstrat-ed increased cytotoxicity in both FANCD2- and BRCA2-defi-cient cells relative to repair-proficient counterparts, as well asachieving a minimum cytotoxic effect, were considered "hits."Initial screening of 15,040 molecules in the ChemBridgeDIVERSet library identified several candidates, four of whichpossessed a similar sulfonamide backbone (Fig. 1C). A fifthcompound, YU175534, was identified through subsequentscreening of structurally homologous molecules across otherlibraries. Multiple rounds of chemical synthesis and structure–activity relationship analysis yielded a derivative compound,YU238259 (Fig. 1D). Relative to the initial screening hits,YU175534 and YU238259 possessed equal or lower LD50sand increased maximal cytotoxicity in BRCA2-deficient cells(Fig. 1B). Given its favorable solubility profile and lowertoxicity in repair-proficient cells, YU238259 was ultimatelyselected as the lead compound for additional study of thisclass of drugs.

Sulfonamide compounds do not inhibit PARP activity orintercalate into DNA

The toxicity of these novel sulfonamide compounds in cellswith HDR defects and their minimal toxicity in repair-proficientcells are strikingly similar to the observed effects of PARP inhi-bition. An in vitro assay of PARP enzymatic activity, however,demonstrated no decrease in PARP activity after treatment witheither YU238259 or an initial screening hit, YU128440 (Fig. 2A).Further, assessment of helical stability of duplex DNA by circulardichroism spectroscopy showed no evidence of intercalation byYU128440 as compared with doxorubicin, a known intercalatingagent (Fig. 2B). Thus, the cytotoxicity of sulfonamide compoundsin repair-deficient cells is unrelated to either of these establishedmechanisms.

YU238259 inhibits homology-dependent DSB repairTo investigate the possibility that inhibition of DSB

repair pathways may underlie the activity of sulfonamides, weutilized the U2OS cell lines DR-GFP and EJ5-GFP, whichcontain integrated GFP genes engineered to report the repairof DNA DSBs via the HDR and NHEJ pathways, respectively(24, 25). These reporters contain a restriction enzyme sequencethat can be cleaved by introduction of an I-SceI expressionplasmid. In each case, repair of the I-SceI–induced DSB via therespective pathway results in expression of full-length, activeGFP (Fig. 3A). YU238259 was added to U2OS DR-GFP orU2OS EJ5-GFP cells 24 hours prior to DSB induction, andtreatment was maintained until cell harvest for FACS analysis.YU238259-treated cells exhibited a significant, dose-dependentdecrease in HDR efficiency, as quantified by percentage of cellsexpressing GFP, but no reduction in NHEJ frequency (Fig. 3B).The slight upward trend in NHEJ activity seen with increasingdoses of YU238259 may suggest a compensatory shift to NHEJin the context of early HDR inhibition. Similarly, analysis of IR-induced foci in U2OS cells pretreated with YU238259 showeddecreased formation of BRCA1 foci, but not of the NHEJ factors53BP1 or pDNA-PK, relative to DMSO vehicle (Fig. 3C). Thepercentage of cells with 15 or more BRCA1 foci per cell wasdecreased by 30% after YU238259 treatment, and this differ-ence was significant at the P < 0.05 level. There was nosignificant difference in the levels of 53BP1 or pDNA-PKfoci-positive cells following treatment. This again demonstratesinhibition of HDR activity by YU238259 with little to no effecton the NHEJ pathway.

-2

-1

0

1

2

330310290270250230

CD a

bsor

banc

e (m

deg)

Wavelength (nm)

DMSO

Doxorubicin

YU128440

Caffeine

0

500

1,000

1,500

2,000

Lum

ines

cenc

e (r

el. u

nits

)

A

YU1284403-AB YU238259B

No substr

ate

No inhibito

r

1,000 mmol/L

500 mmol/L

250 mmol/L

125 mmol/L

100 mmol/L

50 mmol/L

25 mmol/L

12.5 mmol/L

10 mmol/L

5 mmol/L

2.5 mmol/L

1.25 mmol/L

Figure 2.Sulfonamide compounds do not inhibit PARP activity or intercalate into DNA.A, PARP activity in the presence of YU128440, YU238259, or vehicle wasassessed by in vitro enzymatic assay. 3-aminobenzamide, a known PARPinhibitor, was a positive control. B, plasmid DNA was incubated withYU128440, caffeine (a negative control), or a known intercalating agent,doxorubicin. Disruption of the duplex DNA helical structure was analyzed bycircular dichroism spectroscopy.

Stachelek et al.





Using a neutral comet assay, a measure of the number of DSBsin individual cells, we observed that YU238259-treated DLD-1cells, either BRCA2-proficient or BRCA-deficient, had significantlyhigher levels of DSBs 24 hours after 5 Gy IR compared withDMSO-treated controls (Fig. 3D), consistent with impaired res-olution of DSBs owing to HDR inhibition. In DLD-1 cells, nosignificant increase inDSBswas seenwith YU238259 treatment inthe absence of IR, though there was a significant increase in DSBsin YU238259-treated DLD-1 BRCA2-KO cells after 24 hours. Cellcycle analysis of YU238259-treatedDLD-1 andDLD-1BRCA2-KOcells demonstrated a small increase in G1 cell population andsmall decrease in S-phase cell population following treatment,which was heightened in the BRCA2-deficient cells (Fig. 3E),

consistent with early DSB accumulation in the context of BRCA2loss. Taken together, these data suggest that synthetic lethality ofYU238259 in HDR-deficient cells results from accumulation ofunresolvedDSBs following additional inhibition of residualHDRpathway activity.

YU238259 exhibits synthetic lethality with loss of frequentlymutated tumor suppressors

Although initial screening was carried out for syntheticlethality in the context of BRCA2 or FANCD2 loss, the potentinhibition of HDR by YU238259 suggested that syntheticlethality might be observed for other DNA repair- or DNA-damage response-defective cells. Indeed, in paired, isogenic

I-Scel

GFP-donorInac�ve GFP

I-Scel cleavage and HR

GFP+

GFP- donor

I-Scel I-Scelpuro

I-Scel cleavageand NHEJ

GFP GFP+

HDR NHEJ

012345678

Perc

ent G

FP+ c

ells

5.02.51.250.6250

YU238259 (μmol/L)

0

20

40

60

Perc

ent f

oci+ c

ells

DMSO YU238259

pDNA-PK53BP1BRCA1

P < 0.05

DA

B

E

C

0

1

2

3

4

5

6

7

8

Com

et ta

il m

omen

t (no

rmal

ized)

** P < 0.001

* P < 0.01

5 Gy IRNo IR 5 Gy IRNo IR

DLD-1 BRCA2-KO

G1 S G2–M

DMSO

+0

h

YU23

8259

+0

h

DMSO

+24

h

YU23

8259

+24

h

DMSO

+0

h

YU23

8259

+0

h

DMSO

+24

h

YU23

8259

+24

hDM

SO +

0 h

YU23

8259

+0

h

DMSO

+24

h

YU23

8259

+24

h

DMSO

+0

h

YU23

8259

+0

h

DMSO

+24

h

YU23

8259

+24

h

0

10

20

30

40

50

60

Perc

ent c

ell p

opul

a�on

52.5052.50

DLD-1 BRCA2-KO

**

**

*

Figure 3.YU238259 specifically inhibits HDR activity. A, schematic of DR-GFP and EJ5-GFP cell-based reporter assays for HDR and NHEJ, respectively. B, effect of YU238259on the frequency of HDR and NHEJ, as assessed by FACS analysis of GFPþ cells. C, relative percentage of cells positive for DNA repair foci at 8 hours afterirradiation with 10 Gy IR, following pretreatment with 25 mmol/L YU238259 or vehicle. D, neutral comet assay of DLD-1 or DLD-1 BRCA2-KO cells pretreatedwith YU238259 and irradiated with 5 Gy IR. DSB levels, as assessed by comet tail moment, were measured at 0 and 24 hours after irradiation. E, cell cycle analysisof DLD-1 and DLD-1 BRCA2-KO cells treated with YU238259 for 6 hours and stained with propidium iodide.






cell lines deficient or proficient in BRCA2 (DLD-1), ATM(GM05849), and PTEN (U251), we observed that YU238259was significantly more toxic to DNA repair or DNA-damageresponse-deficient cells than their repair-proficient counter-parts (Fig. 4A–C). We can thus conclude that the cytotoxicityof YU238259 is not cell line-specific and that synthetic lethal-ity exists for a variety of cells with DNA repair or damageresponse pathway mutations. We also investigated whethersiRNA-mediated knockdown of the repair factors ATR, ATRIP,and NBS1 would sensitize A549 cells to YU238259 treatment.The overall toxicity of ATR loss precluded any determination ofsynthetic lethality in the case of ATR knockdown; however,siRNA against ATRIP and NBS1 rendered cells sensitive toYU238259 (Fig. 4D). Differences between the repair-deficientand proficient lines were analyzed statistically via the Wilcoxonsigned-rank test and were significant at the P < 0.01 level.

YU238259 sensitizes tumor cells to radiotherapy andDSB-inducing chemotherapy

It has been well established that inhibition of DNA repairpathways can potentiate the cytotoxicity of radiotherapy andchemotherapy (26–28). We hypothesized that HDR inhibitionby YU238259 would enable it to synergize with IR and drugs thatgenerate DSBs. We found that the pretreatment of DLD-1 andDLD-1 BRCA2-KO cells with YU238259 enhanced the effective-ness of IR or etoposide (Fig. 5A, B,D, and E). InDLD-1BRCA2-KO

cells, YU238259 plus IR and YU238259 plus etoposide weremathematically proven to be synergistic at all dose combinations,using the Chou–Talalay mutually nonexclusive model (29).Interestingly, the combination of YU238259 with the PARPinhibitor, olaparib, was also synergistic in DLD-1 BRCA2-KOcells (Fig. 5C and F). This suggests that YU238259 is also ableto potentiate the toxicity of endogenous lesions that subsequentlyprogress into DSBs in the setting of BER inactivation throughPARP inhibition.

Growth of BRCA2-deficient tumor xenografts in nude mice isinhibited by YU238259 treatment

DLD-1 cells have previously been shown to effectively formtumor xenografts in nudemice (30). To assess the in vivo syntheticlethality of our sulfonamide compounds, tumor xenografts weregenerated by subcutaneous injection of 3� 106 DLD-1 or DLD-1BRCA2-KO cells into the flanks of nude mice. The mice thenreceived repeated doses of 3 mg/kg YU238259 or 5 mg/kgYU128440 by intraperitoneal injection. BRCA2-proficient DLD-1 xenografts were unresponsive to treatment, but DLD-1 BRCA2-KO xenografts showed a statistically significant delay in tumorgrowth relative to vehicle for both drugs (Fig. 6A and B). Further,there was a significant overall survival benefit (as measured bydays for individual tumors to reach 4� their volume at time oftreatment) inDLD-1BRCA2-KOmice treatedwithYU128440andincreased survival, though not to a statistically significant degree,

A

10

100

107.552.50

Perc

ent s

urvi

val

DLD-1DLD-1 BRCA2-KO

B

1

10

100

20151050

GM05849 + ATM

GM05849

C

10

100

107.552.50

Perc

ent s

urvi

val

U251+PTENU251

YU238259 (mmol/L)

YU238259 (mmol/L) YU238259 (mmol/L)

D

10

100

107.552.50

CTRLAn�-ATRIPAn�-NBS1

YU238259 (mmol/L)

ATRIP

NBS1Tubulin

PTENATM

Ac�n

GM

GM

+ATM

U25

1

U25

1+PT

EN

moc

k

NT

siRN

AAT

RIPi

NBS

1i

Figure 4.YU238259 demonstrates syntheticlethality with loss of HDR proteins.Paired, isogenic human cell linesdeficient and proficient in (A) BRCA2,(B) ATM, and (C) PTEN were treatedwith increasing doses of YU238259 for48 hours and toxicity assessed byclonogenic survival assay. D, U2OScells were transfected with siRNAsagainst ATRIP, NBS1, or nontargetingcontrol, or mock-transfected.Seventy-two hours later, cells weretreated with YU238259 as above andsurvival assessed by clonogenic assay.Differences between repair-deficientand -proficient lines were analyzedstatistically via the Wilcoxon signed-rank test and were significant at theP < 0.01 level.

Stachelek et al.





in mice treated with YU238259 (Fig. 6C and D). These datarecapitulate the synthetic lethality of YU238259 treatment inBRCA2-deficient cultured cells. Importantly, the bioavailabilityof YU238259 and other sulfonamides has not yet been opti-mized for in vivo treatment. Preliminary pharmacokinetic inves-tigation suggests that there is rapid hepatic metabolism (notshown) that potentially could be slowed through further chem-ical modification of YU238259. Thus, there is potential forimproved in vivo efficacy and the possibility to achieve largereffects at lower doses.

DiscussionExploiting deficiencies in DNA repair found in cancer cells

offers a promising approach to enhance cancer therapy. Agentsthat exploit synthetic lethality typically have increased thera-peutic ratios and the potential for synergism with existingtreatment modalities. Here, we have described the identifica-tion and characterization of a novel class of DNA DSB repair-inhibiting compounds with antitumor activity. Our lead com-pound of this class, YU238259, exhibits synthetic lethality incellular and mouse models of human BRCA2-deficient tumors.YU238259 also demonstrates synergism with IR and etopo-side, which remain in frequent clinical use. Immunofluores-cence studies of IR-induced foci formation by HDR and NHEJfactors show a decrease in BRCA1 foci formation in YU238259-treated cells, but no effect on 53BP1 or pDNA-PK foci. Thissupports data from reporter cell lines indicating thatYU238259 specifically inhibits HDR and suggests a moreproximal pathway inhibition. Ongoing proteomic analysis will

likely be necessary to elucidate the cellular target(s) of thisclass of compounds and more fully define the mechanism ofaction at the molecular level. Importantly, YU238259 does notinhibit PARP activity and so offers a novel mechanism toachieve synthetic lethality with BRCA2 deficiency. Further, weshow that the divergent mechanisms of action and low toxi-cities of PARP inhibition and YU238259 treatment enablethem to synergize effectively. The recent FDA approval ofolaparib for the treatment of advanced ovarian cancer high-lights the potential utility of this therapeutic approach. Addi-tional studies will be useful in determining the degree to whichcotreatment can radiosensitize tumor cells and the spectrum ofDNA repair pathway mutations that confer vulnerability tothese agents.

The observation of synthetic lethality with inactivation ofBRCA2, ATM, and PTEN underscores the potential clinical utilityof YU238259 across a variety of common tumors. Of note,BRCA1/2 mutations account for 5% to 13% of ovarian cancersin developed countries and are associated with higher histologicgrade and poorer prognosis. Approximately 75% of ovariancancers are metastatic at the time of diagnosis (31). For stage IIIand IV ovarian carcinomas, therapeutic options remain limitedand 5-year survival rates are low (32). Our data show increasedcytotoxicity of YU238259 in the patient-derived BRCA2-deficientovarian carcinoma line PEO1 relative to its treatment-resistant,BRCA2–wild-type equivalent, PEO4. Given the need for moreeffective chemotherapy and the striking lethality of ovarian can-cer, novel synthetic lethal agents such as YU238259 that synergizewith olaparib and other emerging therapies have significantclinical potential in this setting.

125 nmol/L

0 nmol/L

250 nmol/L

0 nmol/L

250 nmol/L

500 nmol/L

0 Gy

2.5 Gy

5 Gy

1

10

100

107.552.50

YU238259 (mmol/L)

1

10

100

107.552.50

YU238259 (mmol/L)

1

10

100

107.552.50

Perc

ent s

urvi

val

YU238259 (mmol/L)

1

10

100

107.552.501

10

100

107.552.501

10

100

107.552.50

Perc

ent s

urvi

val

ECA OLAPARIBETOPOSIDEIR

FDB

DLD-1 DLD-1 DLD-1

BRCA2-KO BRCA2-KO BRCA2-KO

Figure 5.YU238259 exhibits synergism with radiotherapy and DNA-damaging chemotherapy that is potentiated by BRCA2 loss. Clonogenic assay of DLD-1 and DLD-1BRCA2-KO cells pretreated with YU238259 for 24 hours and then cotreated with (A and B) IR, (C and D) etoposide, or (E and F) olaparib for an additional24 hours in the presence of YU238259. Analysis using the Chou–Talalay model yielded combination index (CI) values < 0.9 for all dose combinations in the DLD-1BRCA2-KO cell line, mathematically proving synergism. Higher doses of YU238259 were also synergistic with IR in the DLD-1 cell line.






Small-molecule inhibitors ofmany proximal HDR/NHEJ path-way enzymes, including the MRE11/RAD50/NBS1 (MRN) com-plex, ATM, ATR, and DNA-PKcs, have been described (33–36).Our compounds bear little structural similarity to any knownDSBrepair inhibitors. Given the findings, it remains highly plausiblethat the enzymatic target of these compounds is one for whichsmall-molecule inhibitors do not currently exist. We expect thatadditional characterization anddevelopment of this novel class ofrepair inhibitors will be a valuable avenue of inquiry with hightranslational potential.

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

Authors' ContributionsConception and design: G.C. Stachelek, E. Peterson-Roth, D. Hoyer,W. Hungerford, P.M. GlazerDevelopment of methodology: G.C. Stachelek, E. Peterson-Roth, D. Hoyer,W. Hungerford, P.M. GlazerAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): E. Peterson-Roth, Y. Liu, R.J. Fernandez III, L.R.G.Pike, J.M. Qian, L. Abriola, D. Hoyer, W. Hungerford, J. Merkel

Analysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): G.C. Stachelek, E. Peterson-Roth, Y. Liu, R.J. Fernan-dez III, L. Abriola, W. Hungerford, J. Merkel, P.M. GlazerWriting, review, and/or revision of the manuscript: G.C. Stachelek, E. Peter-son-Roth, J. Merkel, P.M. GlazerAdministrative, technical, or material support (i.e., reporting or organizingdata, constructingdatabases):R.J. Fernandez III, J.M.Qian, L. Abriola,D.HoyerStudy supervision: P.M. GlazerOther (compound design and synthesis): D. Hoyer

AcknowledgmentsThe authors thank R. Bindra, J. Sweasy, and R. Jensen for helpful insights.

Grant SupportThis work was supported by a pilot project grant from the Yale Center for

Molecular Discovery (to P.M. Glazer) and by USPHS grant R01 CA168733 (toP.M. Glazer).

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

Received January 20, 2015; revised May 16, 2015; accepted June 10, 2015;published OnlineFirst June 26, 2015.

References1. GinsburgGS. Realizing the opportunities of genomics in health care. JAMA

2013;309:1463–4.2. Leibeling D, Laspe P, Emmert S. Nucleotide excision repair and cancer.

J Mol Histol 2006;37:225–38.3. Klein TJ, Glazer PM. The tumormicroenvironment and DNA repair. Semin

Radiat Oncol 2010;20:282–7.4. Martin SA, Lord CJ, Ashworth A. Therapeutic targeting of the DNA mis-

match repair pathway. Clin Cancer Res 2010;16:5107–13.5. Wallace SS, Murphy DL, Sweasy JB. Base excision repair and cancer. Cancer

letters 2012;327:73–89.6. Marechal A, Zou L.DNAdamage sensing by the ATMandATRkinases. Cold

Spring Harb Perspect Biol 2013;5:pii: a012716.

7. Lin ZP, Ratner ES, Whicker ME, Lee Y, Sartorelli AC. Triapine disruptsCtIP-mediated homologous recombination repair and sensitizes ovar-ian cancer cells to PARP and topoisomerase inhibitors. Mol Cancer Res2014;12:381–93.

8. Origanti S, Cai SR, Munir AZ, White LS, Piwnica-Worms H. Syntheticlethality of Chk1 inhibition combined with p53 and/or p21 loss during aDNA damage response in normal and tumor cells. Oncogene 2013;32:577–88.

9. Konstantinopoulos PA, Wilson AJ, Saskowski J, Wass E, Khabele D. Sub-eroylanilide hydroxamic acid (SAHA) enhances olaparib activity by target-ing homologous recombination DNA repair in ovarian cancer. GynecolOncol 2014;133:599–606.

Days a�er injec�on

Cum

ula�

ve su

rviv

al ra

te

20 30 40 50

0.0

0.5

1.0

1 6 11 16 21 26 31 36 41 46

100

1,000

Tum

or v

olum

e (m

m3 )

Days a�er injec�on Days a�er injec�on

1 6 11 16 21 26 31 36 41 46 51 56

10

100

1,000

DLD-1DLD-1 + CMPDLD-1 BRCA2-KODLD-1 BRCA2-KO + CMP

BA

C

Days a�er injec�on

30 40 50

0.0

0.5

1.0

D

**

nsns

* P < 0.05

* P < 0.05 * P < 0.05

* nsnsns

Figure 6.YU238259 inhibits growth of BRCA2-deficient tumor xenografts in nude mice.Growth curves of subcutaneous DLD-1 orDLD-1 BRCA2-KO tumor xenografts inathymic mice treated with (A) four dosesof 5 mg/kg YU128440 or (B) 12 doses of 3mg/kgYu238259. Survival rate ofmice, asmeasured by number of days required fortumors to reach 4� their volume at timeof treatment, for (C) YU128440 and (D)YU238259.


Stachelek et al.




10. Lok BH, Carley AC, Tchang B, Powell SN. RAD52 inactivation is synthet-ically lethal with deficiencies in BRCA1 and PALB2 in addition to BRCA2through RAD51-mediated homologous recombination. Oncogene2013;32:3552–8.

11. Illuzzi JL, Wilson DM 3rd. Base excision repair: contribution to tumori-genesis and target in anticancer treatment paradigms. Curr Med Chem2012;19:3922–36.

12. Farmer H, McCabe N, Lord CJ, Tutt AN, Johnson DA, Richardson TB, et al.Targeting the DNA repair defect in BRCA mutant cells as a therapeuticstrategy. Nature 2005;434:917–21.

13. Lee J-M, Hays JL, Annunziata CM, Noonan AM, Minasian L, Zujewski JA,et al. Phase I/Ib study of olaparib and carboplatin in BRCA1 or BRCA2mutation-associated breast or ovarian cancer with biomarker analyses.J Natl Cancer Inst 2014;106.

14. Sandhu SK, Schelman WR, Wilding G, Moreno V, Baird RD, Miranda S,et al. The poly(ADP-ribose) polymerase inhibitor niraparib (MK4827) inBRCAmutation carriers and patients with sporadic cancer: a phase 1 dose-escalation trial. Lancet Oncol 2013;14:882–92.

15. Kummar S, Ji J, Morgan R, Lenz HJ, Puhalla SL, Belani CP, et al. A phase Istudy of veliparib in combination with metronomic cyclophosphamidein adults with refractory solid tumors and lymphomas. Clin Cancer Res2012;18:1726–34.

16. Plummer R, Lorigan P, StevenN, Scott L,MiddletonMR,Wilson RH, et al. Aphase II study of the potent PARP inhibitor, Rucaparib (PF-01367338,AG014699), with temozolomide in patients with metastatic melanomademonstrating evidence of chemopotentiation. Cancer Chemother Phar-macol 2013;71:1191–9.

17. Stachelek GC, Dalal S, Donigan KA, Campisi Hegan D, Sweasy JB, GlazerPM. Potentiation of temozolomide cytotoxicity by inhibition of DNApolymerase beta is accentuated by BRCA2 mutation. Cancer Res 2010;70:409–17.

18. Aebi S, Kurdi-Haidar B, Gordon R, Cenni B, Zheng H, Fink D, et al. Loss ofDNA mismatch repair in acquired resistance to cisplatin. Cancer Res1996;56:3087–90.

19. Friedman HS, Johnson SP, Dong Q, Schold SC, Rasheed BK, Bigner SH,et al. Methylator resistance mediated by mismatch repair deficiency ina glioblastoma multiforme xenograft. Cancer Res 1997;57:2933–6.

20. Hehlgans S, Storch K, Lange I, Cordes N. The novel HDAC inhibitorNDACI054 sensitizes human cancer cells to radiotherapy. RadiotherOncol2013;109:126–32.

21. Levitt RJ, Georgescu MM, Pollak M. PTEN-induction in U251 glioma cellsdecreases the expression of insulin-like growth factor binding protein-2.Biochem Biophys Res Commun 2005;336:1056–61.

22. Sakai W, Swisher EM, Jacquemont C, Chandramohan KV, Couch FJ,Langdon SP, et al. Functional restoration of BRCA2 protein by secondary

BRCA2 mutations in BRCA2-mutated ovarian carcinoma. Cancer Res2009;69:6381–6.

23. Scully R, Ganesan S, Vlasakova K, Chen J, Socolovsky M, Livingston DM.Genetic analysis of BRCA1 function in a defined tumor cell line. Mol Cell1999;4:1093–9

24. Gunn A, Stark JM. I-SceI-based assays to examine distinct repair outcomesof mammalian chromosomal double strand breaks. Methods Mol Biol2012;920:379–91.

25. Pierce AJ, Johnson RD, Thompson LH, Jasin M. XRCC3 promotes homol-ogy-directed repair of DNA damage in mammalian cells. Genes Dev1999;13:2633–8.

26. Chernikova SB, Wells RL, Elkind MM.Wortmannin sensitizes mammaliancells to radiation by inhibiting the DNA-dependent protein kinase-medi-ated rejoining of double-strand breaks. Radiat Res 1999;151:159–66.

27. Morgan MA, Parsels LA, Zhao L, Parsels JD, Davis MA, Hassan MC, et al.Mechanism of radiosensitization by the Chk1/2 inhibitor AZD7762involves abrogation of the G2 checkpoint and inhibition of homologousrecombinational DNA repair. Cancer Res 2010;70:4972–81.

28. Taverna P, Hwang HS, Schupp JE, Radivoyevitch T, Session NN, Reddy G,et al. Inhibition of base excision repair potentiates iododeoxyuridine-induced cytotoxicity and radiosensitization. Cancer Res 2003;63:838–46.

29. ChouTC, Talalay P.Generalized equations for the analysis of inhibitions ofMichaelis-Menten and higher-order kinetic systems with two or moremutually exclusive and nonexclusive inhibitors. Eur J Biochem 1981;115:207–16.

30. LinC, YuY, ZhaoHG, YangA, YanH,Cui Y.Combinationof quercetinwithradiotherapy enhances tumor radiosensitivity in vitro and in vivo. Radio-ther Oncol 2012;104:395–400.

31. Lengyel E. Ovarian cancer development and metastasis. Am J Pathol2010;177:1053–64.

32. Matsuda A, Katanoda K. Five-year relative survival rate of ovarian cancer inthe USA, Europe, and Japan. Jpn J Clin Oncol 2014;44:196.

33. Hickson I, Zhao Y, Richardson CJ, Green SJ, Martin NM, Orr AI, et al.Identification and characterization of a novel and specific inhibitor of theataxia-telangiectasia mutated kinase ATM. Cancer Res 2004;64:9152–9.

34. Leahy JJ, Golding BT, Griffin RJ, Hardcastle IR, Richardson C, Rigoreau L,et al. Identification of a highly potent and selective DNA-dependentprotein kinase (DNA-PK) inhibitor (NU7441) by screening of chrome-none libraries. Bioorg Med Chem Lett 2004;14:6083–7.

35. Dupr�eA, Boyer-Chatenet L, Sattler RM,ModiAP, Lee JH,NicoletteML, et al.A forward chemical genetic screen reveals an inhibitor of theMre11-Rad50-Nbs1 complex. Nat Chem Biol 2008;4:119–25.

36. Reaper PM,GriffithsMR, Long JM,Charrier JD,Maccormick S,CharltonPA,et al. Selective killing of ATM- or p53-deficient cancer cells throughinhibition of ATR. Nat Chem Biol 2011;7:428–30.






Published OnlineFirst June 26, 2015.Mol Cancer Res Gregory C. Stachelek, Elizabeth Peterson-Roth, Yanfeng Liu, et al. Radiosensitization in Repair-Deficient TumorsRepair That Exhibits Synthetic Lethality and YU238259 Is a Novel Inhibitor of Homology-Dependent DNA

Updated version

10.1158/1541-7786.MCR-15-0036doi:

Access the most recent version of this article at:

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

Subscriptions

Reprints and

[email protected] at

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

Permissions

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

.http://mcr.aacrjournals.org/content/early/2015/10/01/1541-7786.MCR-15-0036To request permission to re-use all or part of this article, use this link



http://mcr.aacrjournals.org/lookup/doi/10.1158/1541-7786.MCR-15-0036

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

mailto:[email protected]

http://mcr.aacrjournals.org/content/early/2015/10/01/1541-7786.MCR-15-0036


yu238259 is a novel inhibitor of homology- dependent dna …€¦ · 01/10/2015 · published...

Documents