dna damage and repair - molecular cancer research · implications: these ﬁndings demonstrate a...

DNA Damage and Repair

Systematic Screen IdentifiesmiRNAsThat Target RAD51 andRAD51D to Enhance Chemosensitivity

Jen-Wei Huang1,2,3,5, Yemin Wang1,2,3, Kiranjit K. Dhillon1,2,3, Philamer Calses1,2,3,5, Emily Villegas2,3,Patrick S. Mitchell2,5, Muneesh Tewari2,3,4, Christopher J. Kemp2,3, and Toshiyasu Taniguchi1,2,3

AbstractHomologous recombinationmediates error-free repair ofDNAdouble-strandbreaks (DSB). RAD51 is an essential

protein for catalyzing homologous recombination and its recruitment toDSBs ismediated bymany factors includingRAD51, its paralogs, and breast/ovarian cancer susceptibility gene products BRCA1/2. Deregulation of these factorsleads to impairedDNA repair, genomic instability, and cellular sensitivity to chemotherapeutics such as cisplatin andPARP inhibitors. microRNAs (miRNA) are short, noncoding RNAs that posttranscriptionally regulate geneexpression; however, the contribution of miRNAs in the regulation of homologous recombination is not wellunderstood. To address this, a library of human miRNA mimics was systematically screened to pinpoint severalmiRNAs that significantly reduce RAD51 foci formation in response to ionizing radiation in human osteosarcomacells. Subsequent study focused on two of the strongest candidates, miR-103 and miR-107, as they are frequentlyderegulated in cancer. Consistent with the inhibition of RAD51 foci formation, miR-103 and miR-107 reducedhomology-directed repair and sensitized cells to various DNA-damaging agents, including cisplatin and a PARPinhibitor. Mechanistic analyses revealed that both miR-103 and miR-107 directly target and regulate RAD51 andRAD51D, which is critical for miR-103/107–mediated chemosensitization. Furthermore, endogenous regulation ofRAD51D by miR-103/107 was observed in several tumor subtypes. Taken together, these data show that miR-103andmiR-107overexpression promotes genomic instability andmaybeused therapeutically to chemosensitize tumors.

Implications: These findings demonstrate a role for miR-103 and -107 in regulating DNA damage repair, there-by identifying new players in the progression of cancer and response to chemotherapy. Mol Cancer Res; 11(12);1564–73. �2013 AACR.

IntroductionHomologous recombination is an error-free pathway of

DNA double-strand break (DSB) repair. DSBs occur in thegenome following exposure to exogenous DNA damagingagents such as ionizing radiation or from endogenous cellularevents such as replication fork stalling and collapse (1).RAD51 is essential for homologous recombination and isrecruited to sites of DNA damage to mediate repair; thisrecruitment can be visualized immunocytochemically as theappearance of distinct nuclear foci (2). RAD51 loading at sitesof DNA damage requires many factors including the breastand ovarian cancer susceptibility proteins (BRCA1, BRCA2/

FANCD1, and PALB2/FANCN) and the RAD51 paralogs(RAD51B, RAD51C/FANCO, RAD51D, XRCC2, andXRCC3; refs. 3, 4). Defects in homologous recombinationlead to genomic instability and mutations in some homolo-gous recombination genes, such as BRCA1, BRCA2, PALB2,RAD51C, and RAD51D, confer cancer susceptibility (5). Onthe other hand, homologous recombination defects rendercancer cells hypersensitive to DNA-damaging chemothera-peutics such as interstrand DNA crosslinking agents andinhibitors of PARP and topoisomerases (6). Thus, identifyingand understanding novel modes of homologous recombina-tion regulation is critical for understanding the pathobiologyof cancer as well as shaping more effective treatment.microRNAs (miRNA) are short noncoding RNAs that

posttranscriptionally downregulate gene expression throughtranslational repression and mRNA destabilization (7).There are approximately 1,000 miRNAs encoded by thehuman genome, which are estimated to collectively regulateapproximately 60% of protein-coding transcripts (8) andthus the majority of biologic processes. Furthermore, aber-rantly expressed miRNAs can initiate or drive the progres-sion of diseases such as cancer (7).Some miRNAs have been implicated in the regulation

of factors required for DNA DSB repair (9, 10). miR-24and the breast cancer–associated miRNAs miR-182 and

Authors' Affiliations: 1Howard Hughes Medical Institute, Chevy Chase,Maryland;Divisionsof 2HumanBiology, 3PublicHealthSciences, and4ClinicalResearch, Fred Hutchinson Cancer Research Center; and 5Molecular andCellular Biology Program, University of Washington, Seattle, Washington

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

Corresponding Author: Toshiyasu Taniguchi, Fred Hutchinson CancerResearch Center, 1100 Fairview Avenue North, C1-015, Seattle, WA98109-1024. Phone: 206-667-7283; Fax: 206-667-5815; E-mail:[email protected]

doi: 10.1158/1541-7786.MCR-13-0292

�2013 American Association for Cancer Research.

MolecularCancer

Research

Mol Cancer Res; 11(12) December 20131564

on May 26, 2020. © 2013 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst October 2, 2013; DOI: 10.1158/1541-7786.MCR-13-0292

http://mcr.aacrjournals.org/

miR-146a/b-5p target BRCA1 (11–13), whereas miR-1245targets BRCA2 (14). miR-24 also targets theDSB-associatedhistone variant, H2AX, as does miR-138 (15, 16). Werecently reported that miR-96 targets RAD51 (17). How-ever, miRNAs that regulate homologous recombinationhave not been systematically identified. In this study, wedeveloped a cell-based screen to identify miRNAs whoseoverexpression altered RAD51 nuclear foci formation fol-lowing ionizing radiation. We present the results of thescreen and the characterization of two of the strongest hits,the paralogous miR-103 and miR-107.

Materials and MethodsCell linesU2OS, HeLa, and H1299 were purchased from the

American Type Culture Collection. The PEO1 C4-2 clonehas been described previously (18).MDA-MB-231was a giftof the Paulovich laboratory (Fred Hutchinson CancerResearch Center, FHCRC, Seattle, WA). These humancancer cell lines were authenticated by short tandem repeatDNA profiling (Bio-Synthesis Inc.) and used for this studywithin 6 months of resuscitation. Mouse embryonic fibro-blasts were a gift of the Clurman laboratory (FHCRC).RAT-1 fibroblasts were a gift of the Fausto/Campbelllaboratory (University of Washington, Seattle, WA).H1299 cells were cultured in RPMI-1640, supplementedwith 10% FBS and 2 mmol/L L-glutamine in a humidified5%CO2-containing atmosphere at 37�C. All other cell lineswere grown in Dulbecco's Modified Eagle Medium supple-mented with 10% FBS and 2 mmol/L L-glutamine.

Plasmids, siRNAs, miRNA mimics/inhibitors, and virusproductionThe 30UTRs (untranslated regions) of human RAD51

(position 1 to 980) and RAD51D (position 2 to 1139) werecloned into the XbaI site of pGL3-control (a gift of the Ferolaboratory, FHCRC). Putative binding sites of miR-103/107 were mutated using the QuikChange Kit (Stratagene).All constructs were verified by Sanger sequencing.To obtain miR-103 and miR-107 expression constructs,

miR-103-2 and miR-107 precursors with flankingsequences (150 bp on each end) were PCR amplified andcloned into pLemiR vector (Open Biosystems). Lentiviruswas produced as previously described (16).The FLAG-RAD51 expression construct was made by

initially modifying pRAD51-Luc (19). First, we cloned theRAD51 cDNA (including the 50UTR but not 30UTR) andinserted anN-terminal FLAG-epitope tag.Next, we ligated aRAD51 promoter–containing fragment (�2,931 to þ5)from pRAD51-Luc and the FLAG-RAD51 cDNA into thepENTR1A no ccDB entry vector (Addgene plasmid 17398;ref. 20). Finally, we recombined FLAG-RAD51 into thepLenti-X1-GFP-Zeo DEST vector (Addgene plasmid19277; ref. 20). The luciferase control was made by ligatingthe RAD51 promoter–containing fragment from pRAD51-Luc and the luciferase open reading frame from pGL3-control into the entry vector and then recombining as above.PEO1 C4-2 and U2OS cells transduced with pLenti-X1-

FLAG-RAD51 or pLenti-X1-Luc lentivirus were selectedwith Zeocin (Life Technologies) at 0.2 mg/mL for 1 and 2weeks, respectively.RAD51D cDNA was cloned into the NcoI and BamHI

sites of pMMP-IRES-puro. Retroviruses were produced aspreviously described (16). PEO1 C4-2 cells transducedwith pMMP-IRES-puro (empty vector) or pMMP-IRES-puro-RAD51D retroviruses were selected with puromycinat 2 mg/mL for 3 days.miRNAmimics (Dharmacon) andmiRNAmimic negative

control (miR-Neg2; Dharmacon), siRNAs targeting BRCA2(21), RAD51 (#1 target sequence, 50-AACTAATCAGGTG-GTAGCTCA-30; #2 and #3 from ref. 22), and RAD51D(target sequence, 50-CTGGGTGGAAATAAGCTTATA-30;Qiagen and Sigma) and negative controls [nontargetingsiRNA (21) or an equimolar pool of nontargeting siRNA,siLuc (17), and siAllstars (Qiagen)] were transfected at 10nmol/L using HiPerFect (Qiagen). Locked nucleic acid–based negative (negative control B) or a combination ofmiR-103 and miR-107 power inhibitors (Exiqon) weretransfected at 15 to 50 nmol/L with HiPerFect (Qiagen).

Immunofluorescence microscopyCellswere grownoncoverslips, irradiated (10Gy), and then

simultaneously fixed and permeabilized as described previ-ously (16). Cells were immunostained for RAD51 (H-92;SantaCruzBiotechnology; 1:200). Imageswere acquiredwitha microscope (TE2000; Nikon) and analyzed usingMetaVue(Universal Imaging) and ImageJ (NIH, Bethesda, MD). Atleast 100 cells per experimental point were scored and cellswith five or more foci were considered foci positive. Eachexperiment was independently replicated at least three times.

miRNA mimic library screeningU2OS cells were transfected (20 nmol/L; HiPerFect;

Qiagen) with the Human miRIDIAN miRNA mimiclibrary (v10.1; Dharmacon) in glass-bottom 96-well plates(#655892; Greiner Bio-One). Two days posttransfection,cells were irradiated, simultaneously fixed and permeabi-lized, and immunostained for RAD51 as described above.Images were acquired with the Cellomics Arrayscan micro-scope (Thermo Scientific) and processed as describedpreviously (16). Using DAPI staining to define nuclei, thenumber of RAD51 foci–positive cells (containing at least 10RAD51 subnuclear foci) was quantified using the automatedcounter. The Z-score was calculated on the basis of theformula Z ¼ ðX � mncÞ=s, where X was the individualsample score, mnc was the mean of negative controls in eachplate and s was the SD of the whole population. Screeningwas independently replicated three times.

Western blot analysisWhole-cell extracts were prepared and resolved by poly-

acrylamide gel electrophoresis as described (16) and trans-ferred onto nitrocellulose membranes. Antibodies againstBRCA2 (Ab-2; Calbiochem), CtIP (ab20163; Abcam),Dicer(ab14601; Abcam), PALB2 [N200; a gift of the Xia labora-tory, University of Medicine and Dentistry of New Jersey

miR-103/miR-107 and Homologous Recombination

www.aacrjournals.org Mol Cancer Res; 11(12) December 2013 1565




(UMDNJ), New Brunswick, NJ], RPA70 (#2267; CellSignaling Biotechnology), from Novus: RAD51C (NB100-177), RAD51D (NB100-166), XRCC3 (NB100-165),and from Santa Cruz Biotechnology: ATR (N-19), BRCA1(D-9), CHK1 (G-4), RAD51 (H-92), XRCC2 (N-20), andactin (Sc-1616-R) were probed with horseradish peroxidase–conjugated anti-mouse, anti-rabbit (GE Biosciences), or anti-goat IgG (sc-2020; Santa Cruz Biotechnology). Chemilumi-nescencewas used for detection andmembraneswere exposedto film or digitally scanned with an ImageQuant LAS 4000(GE Biosciences). Films were digitalized using a standardscanner and images were processed using Photoshop CS(Adobe Systems, Inc.). Protein expression was quantifiedfrom digital images using ImageQuant TL (GE Biosciences).

Real-time quantitative reverse transcription PCRTotal RNAwas extracted using TRizol (Life Technologies)

and reverse-transcribed using the TaqMan miRNA ReverseTranscription Kit or the TaqMan cDNA Reverse Transcrip-tionKit (Life Technologies). TheTaqManMiRNAAssayKitor the Gene Expression Kit (Life Technologies) was used forquantitative PCR (qPCR). The comparativeCt value methodwas used for transcript quantification.miR-103 andmiR-107expression was normalized to RNU24 or RNU44 snoRNA.RAD51 and RAD51D expression was normalized to 18SrRNA or TBP mRNA.

Luciferase assayU2OS cells were seeded in 24-well plates (5 � 104/well)

and cotransfected with miRNA mimics (0.5 nmol/L; HiPer-Fect), pRL-TK (50ng/well;MirusTransIT-LT-1; a gift of theFero laboratory, FHCRC), and pGL3-control vectors con-taining empty, wild-type ormutantRAD51 (or RAD51D) 30-UTR sequences (200 ng/well). Two days posttransfection,cells were lysed and the luciferase activities assayed using theDual-Luciferase Assay Kit (Promega). Relative luciferaseactivity was calculated by normalizing the ratio of Firefly/Renilla luciferase to that of miR-Neg–transfected cells.

Homologous recombination assayU2OS DR-GFP cells (a gift from Drs. Maria Jasin and

Koji Nakanishi, Memorial Sloan-Kettering Cancer Center,New York, NY; ref. 23) were transfected with siRNAs ormiRNA mimics and then transfected 24 hours later withpCBASce (I-SceI expression vector) or empty vector. Twodays after plasmid transfection, cells were harvested andanalyzed using a FACSCalibur Analyzer to determine thepercentages of GFP-positive cells.

Cell-cycle analysisCells were pulse-labeled with BrdU, fixed, and then

stained for DNA content and BrdU incorporation asdescribed previously (17). Flow cytometry was performedto determine the cell-cycle distribution.

Survival assayTransfected or transduced cells were seeded into 12-well

plates at 5� 103 per well (H1299 and HeLa) or 6� 103 per

well (U2OS and PEO1 C4-2) and treated with cisplatin,etoposide, camptothecin, paclitaxel (Sigma), or AZD2281(Axon Medchem) at indicated concentrations. After incu-bation for 5 to 7 days, cell survival was assessed by crystalviolet staining as described previously (16). Clonogenicassays were performed as described previously (17).

Statistical analysisThe Student or paired t tests were used to evaluate the

significance of differences between 2 groups of data (Excel;Microsoft Inc.). All results were expressed as mean � SD orSEM. A P value of less than 0.05 was considered significant.

ResultsmiR-103 and miR-107 inhibit DNA damage–inducedRAD51 foci formationWe devised a screen to systematically identify miRNAs

that regulate homologous recombination by using ionizingradiation–induced RAD51 foci formation as an indicator ofhomologous recombination proficiency (Fig. 1A). We over-expressed a library of 810 human miRNAs in U2OS cells,treated with ionizing radiation (10 Gy), and then fixed andimmunostained cells for RAD51 6 hours after irradiation.The percentage of RAD51 foci–positive cells (containing atleast 10 RAD51 subnuclear foci) was then determined usingautomated microscopy and software-assisted foci quantita-tion. Fifty-two miRNAs were found to significantly reducethe percentage of RAD51 foci–positive cells following irra-diation (Z-score <�2), including miR-96 (17), whereas onemiRNA (miR-372) increased the percentage of RAD51foci–positive cells (Z-score > 2; Fig. 1B and SupplementaryTable S1).miR-103/107, a pair of paralogousmiRNAs that differ at a

single nucleotide and likely regulate overlapping targets dueto seed region identity (Fig. 1C), stood out as the mostpotent inhibitors of RAD51 foci formation (Fig. 1B). miR-103 and/or miR-107 are aberrantly expressed in several solidcancers (24–28) and their overexpression in breast cancercells promotes tumor invasiveness (29). Furthermore, miR-103/107 are upregulated after doxorubicin-induced DNAdamage (30–32). Therefore, we focused on miR-103/107for further investigation.miR-103/107 overexpression by miRNA mimic transfec-

tion (confirmed by qPCR; data not shown) reduced thepercentage of RAD51 foci–positive cells following ionizingradiation in U2OS, HeLa, and an ovarian cancer cell line,PEO1 C4-2 (Fig. 1D; Supplementary Fig. S1A and S1B).RAD51 foci formation in response to cisplatin was similarlyinhibited (Supplementary Fig. S1C). miR-103/107 over-expression by lentiviral transduction of miR-103/107 pre-cursors also reduced the percentage of RAD51 foci–positiveU2OS cells following ionizing radiation (SupplementaryFig. S2A and S2B). These data demonstrate that miR-103/107 overexpression impair DNA damage–inducedRAD51 foci formation in a cell type-independent manner.Homologous recombination and RAD51 loading are

coupled to cell-cycle progression (33). miR-103/107

Huang et al.

Mol Cancer Res; 11(12) December 2013 Molecular Cancer Research1566




overexpression did not significantly alter cell-cycle distribu-tion in either untreated or irradiated cells (SupplementaryFig. S1D–S1F and data not shown). Therefore, the reduc-tion in RAD51 foci formation mediated by miR-103/107overexpression is likely independent of changes in cell-cycleprogression.

miR-103/107 impair homologous recombination andpromote chemosensitivity to DNA damaging agentsNext, we examined the effect ofmiR-103/107 overexpres-

sion on homologous recombination–mediated DSB repairusing U2OS DR-GFP cells (23). The overexpression ofmiR-103/107 significantly reduced homologous recombi-nation efficiency as judged by the percentage of GFP-positive cells following induction of a DSB by I-SceI (Fig.2A), consistentwith the reduction inRAD51 foci formation.Proficient homologous recombination is required for

cellular resistance to agents that induce DNA DSBs in S-phase, such as cisplatin, topoisomerase poisons (e.g., camp-tothecin and etoposide), and PARP inhibitors. As expected,miR-103/107 overexpression sensitized U2OS, HeLa, andPEO1 C4-2 cells to cisplatin and a PARP inhibitor,AZD2281 (Fig. 2B; Supplementary Fig. S3A–S3C). Over-expression of miR-103/107 also sensitized U2OS cells to

camptothecin and etoposide (Fig. 2C). Furthermore, che-mosensitization to these DNA damaging agents wasobserved when miR-103/107 were overexpressed by lenti-viral transduction (Supplementary Figs. S2E and S4C). Incontrast, miR-103/107 overexpression did not sensitize cellsto paclitaxel (Fig. 2D; Supplementary Fig. S3A and S3B), amicrotubule-stabilizing agent, suggesting that chemosensi-tivity is specific to DNA damage.

miR-103/107 regulate RAD51 expressionWe observed a reduction in the overall intensity of

RAD51 immunostaining upon miR-103/107 overexpres-sion (Fig. 1D; Supplementary Fig. S1A and S1B). Consis-tently, miR-103/107 overexpression was found to reduceRAD51protein as well asmRNAexpression in several cancercell lines (Fig. 3A and B; Supplementary Figs. S2C and S2D,S4A and S4B). The downregulation of Dicer (29) served as acontrol for miR-103/107 mimic transfection (Fig. 3A).Using the TargetScan algorithm (34), we noted that humanRAD51 is a predicted target of miR-103/107 and theirputative binding site in the 30UTR of RAD51 is conservedin rats but not mice (Fig. 3C and Supplementary Fig. S5A).Consistent with this observation, miR-103/107 downregu-lated RAD51 protein expression in RAT-1 rat–derived

4

2

0

−2

−4

0 400 800Z-s

co

re

miR-96miR-103-1miR-107miR-103-2

U2OS

A

C

D

BmiRNA mimics

library (810)

48 h20 nmol/L

6 h

lR: 10 Gy

lmmunostaining for RAD51

Cellomics ArrayScan analysis

Foci quantitation

% RAD51 foci- No. of RAD51 RAD51Z-Score Positive cells Foci/cell Intensity/cell

miR-Neg

miR-103-1

miR-103-2

miR-107

siRAD51 #1

−

−2.83

−3.44

−3.16

−5.23

67.05

36.97

40.21 15.05 90.22

41.71 15.09 106.46

1.51 3.04 52.47

13.80 91.79

29.19 189.44

U2OS + lR (10 Gy)

DA

Pl

RA

D5

1

miR-Neg miR-103 miR-107

miR

-10

7

miR

-10

3

miR

-Ne

g

siR

AD

51

D

siB

RC

A2

siL

uc

siL

uc

75

50

25

0

IR (10 Gy):

Pe

rce

nta

ge

of

RA

D5

1 fo

ci-p

ositiv

e

ce

lls (

≥ 5

fo

ci)

Figure 1. miR-103 and miR-107negatively regulate DNA damage–induced RAD51 foci formation.A, screening strategy for theidentification of miRNAs thatregulate ionizing radiation–induced RAD51 foci formation inU2OS cells. B, the mean Z-scoresof eachmiRNA from the screen areranked (n ¼ 3). A table of imageanalysis for miR-103, miR-107,and siRAD51 control is presented.C, mature sequences of miR-103and miR-107 differ at a singlenucleotide (�). Seed regions areunderlined. D, U2OS cellstransfectedwith indicatedmiRNAsor siRNAs (10 nmol/L) wereirradiated (10 Gy) 48 hours later,fixed 6 hours after ionizingradiation, and immunostainedfor RAD51. Nuclei werecounterstained with DAPI. RAD51foci–positive cells contained atleast five nuclear foci (n ¼ 3;�SEM). Representative imagesare shown (scale bar, 20 mm).






fibroblasts but not in mouse embryonic fibroblasts (Sup-plementary Fig. S5B).To determine whether RAD51 is a direct target of

miR-103/107, we generated a luciferase reporter constructcontaining the human RAD51 30UTR. miR-103/107 sig-nificantly reduced the activity of this construct relative tomiR-Neg (Fig. 3D and Supplementary S5C). Point muta-tions introduced in the putative miR-103/107–binding siteof the construct relieved this repression (Fig. 3C and D andSupplementary Fig. S5C), suggesting that miR-103/107directly target RAD51 through a single site in its 30UTR.To determine whether downregulation of RAD51 is

relevant to the miR-103/107–mediated regulation ofhomologous recombination, we tried to complement defectsin RAD51 foci formation and chemoresistance by ectopi-cally expressing FLAG-tagged RAD51 without its 30UTR.FLAG-RAD51 was found to be functional as its expressionrestored ionizing radiation–induced RAD51 foci formationand chemoresistance to cisplatin and AZD2281 in RAD51-depleted cells (Supplementary Fig. S6A and S6B and datanot shown). In bothU2OS and PEO1C4-2 cells transfectedwith miR-103 or miR-107, stable expression of FLAG-RAD51 at least partially complemented ionizing radia-tion–induced RAD51 foci formation as well as chemoresis-

tance to cisplatin and AZD2281 (Supplementary Fig. S7A–S7C). These results demonstrate that RAD51 is one of thecrucial targets of miR-103/107 in the regulation of homol-ogous recombination.

miR-103 and miR-107 directly regulate RAD51DTo identify other homologous recombination–relevant

miR-103/107 targets, we analyzed the expression of severalfactors that regulate RAD51 foci formation (Fig. 3A). Wefound consistent downregulation of RAD51D protein andmRNA expression when miR-103/107 are overexpressed inseveral cell lines (Fig. 3A and B, S2D and S5D). RAD51Dis a predicted target of miR-103/107 by TargetScan (Fig.3C). Using luciferase reporter constructs containing theRAD51D wild-type or mutant 30UTRs, we determinedthat miR-103/107 directly target the 30UTR of RAD51D(Fig. 3D and Supplementary Fig. S5E).The overexpression of ectopic RAD51D partially rescued

ionizing radiation–induced RAD51 foci formation in miR-107–overexpressing PEO1 C4-2 cells (Fig. 4A and B).Chemoresistance to cisplatin and AZD2281 were also par-tially complemented by RAD51D expression (Supplemen-tary Fig. S8). Interestingly, the coexpression of RAD51 andRAD51D largely reversed miR-107–mediated inhibition of

100

75

50

25

0

100

75

50

25

0

miR

-Neg

miR

-103

miR

-107

siC

ontr

ol

siB

RC

A2

Perc

enta

ge H

R r

epair

effic

iency

12.4 %A

B

C D

miR-Neg miR-103

miR-107

2.1 %

1.8 %

GF

P

Red

miR-Neg

miR-103

miR-107

1

0.5

0

1

0.5

00.01 0.1 1 10 0.01 0.1 1 10

AZD2281 (μmol/L)

Surv

ivin

g fra

ction

1

0.5

0

1

0.5

0

1

0.5

0Surv

ivin

g fra

ction

Surv

ivin

g fra

ction

Cisplatin (μmol/L)

0.0001 0.001 0.01 0.1 1 0.1 1 10 100 0.01 0.1 1 10 100

Etoposide (μmol/L) Camptothecin (nmol/L) Paclitaxel (μmol/L)

Figure 2. miR-103 and miR-107impair homologous recombinationand promote chemosensitivity toDNA damaging agents. A, U2OSDR-GFP cells transfected withindicated miRNAs (10 nmol/L)were transfected the next day withan I-SceI expression vector andharvested 48 hours later for flowcytometry. U2OS DR-GFP cellsdepleted of BRCA2 (10 nmol/L),served as a control of impairedhomologous recombination.Representative flow cytometryprofiles with gated GFP-positivepopulations are shown (n ¼ 4;�SEM). Homologousrecombination repair efficiency ispresented as a percentage of themiR-Neg- or siControl–transfectedcontrols. B and D, U2OS cellstransfectedwith indicatedmiRNAs(10 nmol/L) were reseeded 48hours later to assess for cellgrowth in response to cisplatin andAZD2281 (B), etoposide andcamptothecin (C), and paclitaxel(D). Cell growth is expressed as afraction of the untreated control(n ¼ 3; �SEM).

Huang et al.





RAD51 foci formation and chemosensitivity to cisplatin andAZD2281 (Fig. 4C). Our results indicate that RAD51 andRAD51D are the major targets through which miR-107regulate RAD51 foci formation and chemoresistance.

Endogenous miR-103/107 regulate RAD51D expressionTo determine whether endogenous miR-103/107 regu-

late RAD51 and RAD51D expression, we inhibited endog-enous miR-103/107 in H1299 (non–small cell lung carci-noma cell line) and MDA-MB-231 (breast cancer cell line),which abundantly express these miRNAs (29, 35, 36). Anti-sense oligonucleotides (PI-103 and PI-107) effectivelyinhibited miR-103/107 (Fig. 3E and S9A-C). Upon inhi-bition of miR-103/107, we observed a modest upregulationof RAD51D protein and mRNA in these cell lines and noappreciable change in RAD51 expression (Fig. 3E and F andS9D). This suggests that endogenous miR-103/107 have a

significant role in regulating the expression of RAD51D, butnot RAD51, in these cell lines.Next, we looked for inverse relationships between the

expression of miR-103/107 and RAD51 and RAD51D inthe NCI-60 panel of cancer cell lines (37). Using CellMiner(37), we found no significant inverse correlation betweenmiR-103/107 and RAD51, RAD51D or several other targetmRNAs in the NCI-60 panel (Supplementary Table S2). Asthe NCI-60 panel is heterogenous with respect to tissueorigin (37), we additionally analyzed expression datasetsfrom themajority of tumor subtypes in TheCancer GenomeAtlas (http://cancergenome.nih.gov/) to identify tissue-spe-cific correlations between miR-103/107 and RAD51 andRAD51D in clinical tumor samples (see SupplementaryMethods; Supplementary Table S3). Although no signifi-cant inverse correlations between miR-103/107 and RAD51mRNA were found in any of the tumor subtypes analyzed,

HeLaU2OS1

0.75

0.5

0.25

0

1

0.75

0.5

0.25

0

RAD51

RAD51D

miR

-Neg

miR

-107

miR

-Neg

Rela

tive

mR

NA

expre

ssio

n

miR

-107

PEO1C4-2

A

C

D

E F

BHeLa U2OS

miR

-Neg

miR

-103

miR

-107

miR

-Neg

miR

-103

miR

-107

miR

-Neg

miR

-103

miR

-107

ATRBRCA1

BRCA2

PALB2

Dicer

CtIPRPA70

CHK1RAD51

RAD51DActin

U2OS

1.25

1

0.75

0.5

0.25

0

1.25

1

0.75

0.5

0.25

0Re

lative

lu

cife

rase

activity

Re

lative

pro

tein

exp

ressio

n

Control RAD51 RAD51mut

Control RAD51D RAD51Dmut

H1299H1299 + NSC

NSC miR-103 miR-107

miR-Neg

miR-107

1.5

1

0.5

0

Pl-Neg

Pl-103/107

RAD51 RAD51D

Pl-Neg (30 nmol/L):

Pl-103/107 (15 nmol/L ea):

RAD51

RAD51D

Actin

Figure 3. miR-103 and miR-107directly target RAD51 andRAD51D. A, HeLa, U2OS, andPEO1 C4-2 cells transfected withindicated miRNAs (10 nmol/L)were harvested 48 hours later forWestern blotting. B, HeLa andU2OS cells transfected withindicated miRNAs (10 nmol/L)were harvested 48 hours later forreal-time PCR. C, putative miR-103/107 binding sites in the30UTRs of RAD51 and RAD51D.The asterisks denote positions ofsubstitutions made to generatemutant 30UTR constructs. Theseed sequences of miR-103/107are underlined. D, U2OS cellscotransfected with indicatedmiRNAs (0.5 nmol/L) andluciferase constructs (x-axis) wereassayed for luciferase activity 48hours later (n ¼ 8–9; �SEM;Student t test: �, P < 0.05;��,P < 0.005). E, H1299 cells stablyexpressing non-silencing control(NSC), miR-103 or miR-107 wereharvested 2 days after transfectionwith indicated miRNA inhibitors(30 nmol/L) for Western blotting. F,quantitation of target expression inH1299 cells stably expressingNSC and transfected withindicated miRNA inhibitors, as in(E). RAD51 and RAD51D proteinlevelswere normalized to actin andthen expressed relative to PI-Neg–transfected cells (n ¼ 3; �SD).






RAD51D mRNA expression was found to be inverselycorrelated with miR-103/107 levels in lung adenocarcino-mas (Pearson r ¼ �0.2266 and �0.2756 for miR-103 andmiR-107, respectively; P < 0.05) and uterine corpus endo-metrial carcinomas (Pearson r¼�0.1193 and�0.1387 formiR-103 and miR-107, respectively; P < 0.05) and withmiR-103 levels in stomach adenocarcinomas (Pearson r ¼�0.2246, P < 0.05; Supplementary Fig. S10 and Supple-mentary Table S3).

DiscussionWe conducted an unbiased screen of a miRNA library to

systematically identify miRNAs that regulate homologousrecombination. Overexpression of the paralogous miR-103and miR-107, the two strongest hits from our screen,inhibited ionizing radiation–induced RAD51 foci forma-

tion, impaired homologous recombination, and conferredchemosensitivity to DNA damaging agents. Downregula-tion of RAD51 and RAD51D as a consequence of directtargeting was critical to the regulation of homologousrecombination by miR-103/107 (Fig. 4D). To date, severalmiRNAs have been implicated in the regulation of repair ofDNA DSBs (9, 10). The deregulation of some of thesemiRNAs has been suggested to contribute to genomicinstability and chemosensitivity in cancer. It is thereforeplausible that cancer-associated deregulation of miR-103and miR-107 (24–28) potentially contributes to genomicinstability and chemosensitivity of tumors.In this and our previous work, we have characterized

several miRNAs (miR-103, miR-107, and miR-96) thatposttranscriptionally downregulate RAD51 expression.RAD51 has long been suspected of being a tumor suppressorin cancer. It plays an essential role in homologous

cDNA:

A

C

D

B

miR-Neg:miR-107:

Luc

FLAG-RAD51RAD51

RAD51D

ACTIN

+Luc/

empty

+Luc/

RAD51D

+RAD51/

RAD51D

+RAD51/

empty

75

50

25

0miR-Neg:

Perc

enta

ge o

f R

AD

51

foci-positiv

e c

ells

(≥5

foci)

miR-107:

cDNA: Luc/empty

RAD51/empty

RAD51/RAD51D

Luc/RAD51D

Luc/empty +

Luc/empty +

RAD51/RAD51D

RAD51/RAD51D

miR-Neg

+ miR-Neg

+ miR-107

miR-107

1

0.5

0

1

0.5

0

0.05 0.5

Su

rviv

ing

fra

ctio

n

5 0.2 2 20

Cisplatin (μmol/L) AZD2281 (μmol/L)

miR-103/107

RAD51 RAD51D

RAD51 Loading

Homologous recombination

Cellular resistance to DNA-

damaging agents

(e.g., cisplatin, PARP inhibitors)

Figure 4. RAD51 and RAD51Dtargeting is relevant to miR-103/107–mediated regulation ofRAD51 foci formation andchemosensitivity. A, PEO1 C4-2cells stably expressing theindicated cDNAs were transfectedwith miR-Neg or miR-107 (10nmol/L), irradiated (10Gy) 48 hourslater, fixed 6 hours after ionizingradiation, and immunostained withanti-RAD51. Nuclei werecounterstained with DAPI. RAD51foci–positive cells contain at least5 nuclear foci (n ¼ 3–4; �SEM;Student t test: �, P < 0.05;��, P < 0.01; #, not significant).B and C, indicated PEO1 C4-2stable cell lines transfected withmiR-Neg or miR-107 (10 nmol/L)were (B) harvested 48 hours laterfor Western blotting and (C)reseeded 48 hours later to assesscell growth in response to cisplatinand AZD2281. Cell growth isexpressed as a fraction of theuntreated control (n¼ 3–4;�SEM;Student t test: �, P < 0.05;��, P < 0.005; #, not significant).D, our model showing that theregulation of homologousrecombination and cellularresistance to DNA-damagingagents by miR-103/107 ismediated through the targeting ofRAD51 and RAD51D.

Huang et al.





recombination and the deregulation of several RAD51mediators such as BRCA1, BRCA2, and some RAD51paralogs confer cancer susceptibility. RAD51 expression isdownregulated in some cancers (38, 39), but cancer-associated alleles of RAD51 are rare (40). Alternatives toRAD51 dysfunction may therefore exist. One potentialmechanism is the regulation of its expression by cancer-associated miRNAs such as miR-103, miR-107, and miR-96. In addition, several other hits from our screen arealso predicted to target RAD51 (e.g., miR-221/222, miR-124a/506, miR-494, and others) and may collectivelycontribute to the overall regulation of homologous recom-bination (Supplementary Table S1).In addition to RAD51, we found that miR-103/107

can target RAD51D, which was also relevant to miR-103/107–mediated regulation of homologous recombina-tion (Fig. 4A and S8). Although ectopically expressingRAD51 andRAD51D largely restoredmiR-103/107–medi-ated impairment of RAD51 foci formation and chemore-sistance (Fig. 4A and C), we cannot exclude the possibilitythat other targets may be involved. We found that twoRAD51 paralogs, XRCC2 and XRCC3, were also consis-tently downregulated in miR-103/107–overexpressing cells(Supplementary Fig. S5D). Decreased XRCC2 levels arepossibly due to RAD51D downregulation (4), but themechanism of XRCC3 downregulation is unknown.Though a predicted target, our luciferase assays indicatedthat miR-103/107 do not target the XRCC3 30UTR (datanot shown). The role of XRCC3 in RAD51 foci formation isunclear (4, 41), but it has additional roles in the late stages ofhomologous recombination and in checkpoint signaling (3).Therefore, its repression bymiR-103/107may contribute tosome of the phenotypes observed.miR-103/107 also target Dicer and TRBP2, which are

both components in the miRNA biogenesis pathway (Fig.3A; refs. 29, 42). Both of these factors have been implicatedin promoting chemoresistance to cisplatin (43). Further-more, Dicer has been implicated in DNA damage responsesignaling (44) and homologous recombination (45). Inaddition to RAD51 and RAD51D, these and other targetsmay be relevant to miR-103/107–mediated regulation ofhomologous recombination and chemosensitivity.Our data suggest that the physiologic significance of

miRNA-mediated regulation of homologous recombinationmay be cell type- or tissue specific. Our miRNA knockdownstudies demonstrated a role for endogenous miR-103/107 inregulating RAD51D expression in some cell lines (Fig. 3Fand Supplementary Fig. S9D). Consistently, miR-103/107expression is inversely correlated with RAD51D mRNAexpression in several tumor subtypes (Supplementary Fig.S10). Collectively, these findings raise the possibility thatendogenous miR-103/107 target RAD51D to regulateRAD51-mediated homologous recombination and chemo-sensitivity in certain tissues or cell types. We were unable tofind evidence that RAD51 is a target of endogenous miR-103/107 (Fig. 3F; Supplementary Fig. S9D and Supple-mentary Tables S2 and S3). As the overall regulation of geneexpression is complex and miRNAs are generally considered

"fine-tuners" of protein expression, it is perhaps not sur-prising that neither significant derepression of RAD51expression upon miR-103/107 inhibition (Fig. 3F) or sig-nificant inverse correlation between miR-103/107 andRAD51 (and some other targets of miR-103/107) wasobserved (Supplementary Tables S2 and S3). Importantly,our findings do not exclude the possibility that endogenousmiR-103/107 may target RAD51 and/or RAD51D in morespecific cellular contexts.One possibility is that these miRNAs suppress homolo-

gous recombination in cellular states in which DSB repairwould be risky or not economical, such as during cellulardifferentiation (15). Consistent with this, miR-24 and miR-182, which respectively repress H2AX and BRCA1, areupregulated in differentiating leukemia cell lines (13, 15).Intriguingly, miR-103 and miR-107 are upregulated duringretinoic acid–induced differentiation of cultured neuroblas-toma and leukemia cells, respectively (46, 47). With addi-tional targets in cell-cycle regulation such as CDK6 (26, 31),cyclin E, and CDK2 (48), miR-103/107 potentially couplethe regulation of DNA repair and proliferation duringcellular differentiation.Another possibility is that these miRNAs regulate DNA

repair in response to DNA damage thereby modulatingcellular outcome (17). miR-103/107 are upregulated inresponse to doxorubicin treatment through p53-dependenttranscriptional and posttranscriptional mechanisms (30–32). On the basis of our findings, this upregulation couldpotentially inhibit DNA repair and sensitize cells to DNAdamage. Although it may be counterintuitive for a cell tosuppress DNA repair in response to cellular stress, thissuppression may serve to push a damaged cell towardapoptosis (15). Alternatively, restricting or limiting homol-ogous recombination may prevent illegitimate recombina-tion thereby promoting genomic stability.Targeting homologous recombination in cancer has gar-

nered increasing interest because of the resultant hypersen-sitivity to agents that promote DSB formation in S-phase(e.g., cisplatin, PARP inhibitors; ref. 6). miR-103/107 over-expression in combination with cisplatin or other DNAdamaging agents may therefore have therapeutic utility inthe chemosensitization of tumors. Interestingly, the use ofmiR-107 as a therapeutic agent in cancer treatment has beenproposed (49). It is tempting to ask whether combining aDNA damaging agent with nanoparticle-delivered miR-107will chemosensitize tumors (49).In summary, we identified the paralogous miR-103/107

as inhibitors of homologous recombination and chemore-sistance to DNA damaging agents. These miRNAs as well asothers from our screen help provide a more complete pictureof the cellular regulation of DNA repair and may also betherapeutically useful.During the preparation of this article, miR-107 was

identified in a screen for miRNAs that confer PARP inhib-itor sensitivity (50), consistent with our findings. Ourindependent studies both identified RAD51 as a criticaltarget of miR-103/107 in the regulation of homologousrecombination and chemosensitivity. In contrast, our






study also demonstrated that the targeting of RAD51D bymiR-103/107 contributes to their regulation ofDNA repair.Furthermore, we demonstrate that endogenous miR-103/107may have a more significant role in regulating RAD51Dthan RAD51.

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

Authors' ContributionsConception and design: J.-W. Huang, Y. Wang, T. TaniguchiDevelopment of methodology: J.-W. Huang, Y. Wang, E. Villegas, P.S. Mitchell,C.J. Kemp, T. TaniguchiAcquisition of data (provided animals, acquired and managed patients, providedfacilities, etc.): J.-W. Huang, Y. Wang, K.K. Dhillon, P. Calses, T. TaniguchiAnalysis and interpretation of data (e.g., statistical analysis, biostatistics, compu-tational analysis): J.-W. Huang, Y. Wang, K.K. Dhillon, P. Calses, E. Villegas,T. TaniguchiWriting, review, and/or revision of the manuscript: J.-W. Huang, Y. Wang,K.K. Dhillon, P. Calses, P.S. Mitchell, C.J. Kemp, T. TaniguchiAdministrative, technical, or material support (i.e., reporting or organizing data,constructing databases): P. Calses, T. TaniguchiStudy supervision: T. Taniguchi

AcknowledgmentsThe authors thank Drs. Maria Jasin, Koji Nakanishi, Bing Xia, Matthew Fero,

Amanda Paulovich, Bruce Clurman, and Jean Campbell for reagents. The authors alsothank Dr. Celine Jacquemont for critical reading of this article, Dr. BrighamMechamfor technical assistance with Synapse, and Dr. Kavita Garg for helpful discussions. Theauthors thank all members of the Kemp and Taniguchi laboratories for technicalsupport and helpful discussions.

Grant SupportThis work was supported by the Howard Hughes Medical Institute, the NIH/

NHLBI (R21 HL092978; to T. Taniguchi), the NIH/NCI (R01 CA125636; to T.Taniguchi), Fanconi Anemia Research Fund (to T. Taniguchi), and Fred HutchinsonCancer Research Center and Listwin Family Foundation funding (NewDevelopmentfunds; to M. Tewari). J.-W. Huang and P. Calses are supported by PHS NRSA 2T32GM007270 from NIGMS. Y. Wang is a research fellow supported by the CanadianInstitute of Health Research. K.K. Dhillon is supported by the ChromosomeMetabolism and Cancer Training grant (T32 CA 09657) and the Thomsen FamilyPostdoctoral Fellowship.

The costs of publication of this article were defrayed in part by the payment of pagecharges. This article must therefore be herebymarked advertisement in accordance with18 U.S.C. Section 1734 solely to indicate this fact.

ReceivedMay 31, 2013; revised September 10, 2013; accepted September 12, 2013;published OnlineFirst October 2, 2013.

References1. Heyer WD, Ehmsen KT, Liu J. Regulation of homologous recombina-

tion in eukaryotes. Annu Rev Genet 2010;44:113–39.2. Haaf T, Golub EI, Reddy G, Radding CM, Ward DC. Nuclear foci of

mammalian Rad51 recombination protein in somatic cells after DNAdamage and its localization in synaptonemal complexes. Proc NatlAcad Sci U S A 1995;92:2298–302.

3. Suwaki N, Klare K, Tarsounas M. RAD51 paralogs: roles in DNAdamage signalling, recombinational repair, and tumorigenesis. SeminCell Dev Biol 2011;22:898–905.

4. Chun J, Buechelmaier ES, Powell SN. Rad51 paralog complexesBCDX2 and CX3 act at different stages in the BRCA1-BRCA2-depen-dent homologous recombination pathway. Mol Cell Biol 2013;33:387–95.

5. Walsh T, Casadei S, Lee MK, Pennil CC, Nord AS, Thornton AM, et al.Mutations in 12 genes for inherited ovarian, fallopian tube, and peri-toneal carcinoma identified by massively parallel sequencing. ProcNatl Acad Sci U S A 2011;108:18032–7.

6. Evers B, Helleday T, Jonkers J. Targeting homologous recombinationrepair defects in cancer. Trends Pharmacol Sci 2010;31:372–80.

7. Farazi TA, Spitzer JI, Morozov P, Tuschl T. miRNAs in human cancer.J Pathol 2011;223:102–15.

8. FriedmanRC, FarhKK, BurgeCB,Bartel DP.MostmammalianmRNAsare conserved targets of microRNAs. Genome Res 2009;19:92–105.

9. Chowdhury D, Choi YE, Brault ME. Charity begins at home: non-coding RNA functions in DNA repair. Nat Rev Mol Cell Biol 2013;14:181–9.

10. Wang Y, Taniguchi T. MicroRNAs and DNA damage response: impli-cations for cancer therapy. Cell Cycle 2013;12:32–42.

11. Garcia AI, Buisson M, Bertrand P, Rimokh R, Rouleau E, Lopez BS,et al. Downregulation of BRCA1 expression by miR-146a and miR-146b-5p in triple negative sporadic breast cancers. EMBO Mol Med2011;3:279–90.

12. Lal A,Navarro F,MaherCA,Maliszewski LE, YanN,O'Day E, et al.miR-24 Inhibits cell proliferation by targeting E2F2, MYC, and other cell-cycle genes via binding to "seedless" 30UTR microRNA recognitionelements. Mol Cell 2009;35:610–25.

13. Moskwa P, Buffa FM, Pan Y, Panchakshari R, Gottipati P, MuschelRJ, et al. miR-182-mediated downregulation of BRCA1 impactsDNA repair and sensitivity to PARP inhibitors. Mol Cell 2011;41:210–20.

14. Song L, Dai T, Xie Y, Wang C, Lin C, Wu Z, et al. Upregulation of miR-1245 by c-myc targets BRCA2 and impairs DNA repair. J Mol Cell Biol2012;4:108–17.

15. Lal A, Pan Y, Navarro F, Dykxhoorn DM, Moreau L, Meire E, et al. miR-24-mediated downregulation of H2AX suppresses DNA repair in ter-minally differentiated blood cells. Nat Struct Mol Biol 2009;16:492–8.

16. Wang Y, Huang JW, Li M, Cavenee WK, Mitchell PS, Zhou X, et al.MicroRNA-138 modulates DNA damage response by repressing his-tone H2AX expression. Mol Cancer Res 2011;9:1100–11.

17. Wang Y, Huang JW, Calses P, Kemp CJ, Taniguchi T. MiR-96 down-regulates REV1 and RAD51 to promote cellular sensitivity to cisplatinand PARP inhibition. Cancer Res 2012;72:4037–46.

18. Sakai W, Swisher EM, Jacquemont C, Chandramohan KV, Couch FJ,Langdon SP, et al. Functional restoration of BRCA2 protein by sec-ondary BRCA2 mutations in BRCA2-mutated ovarian carcinoma.Cancer Res 2009;69:6381–6.

19. Hine CM, Seluanov A, Gorbunova V. Use of the Rad51 promoter fortargeted anti-cancer therapy. Proc Natl Acad Sci U S A 2008;105:20810–5.

20. CampeauE, Ruhl VE, Rodier F, SmithCL, RahmbergBL, Fuss JO, et al.A versatile viral system for expression and depletion of proteins inmammalian cells. PLoS ONE 2009;4:e6529.

21. Sakai W, Swisher EM, Karlan BY, Agarwal MK, Higgins J, Friedman C,et al. Secondary mutations as a mechanism of cisplatin resistance inBRCA2-mutated cancers. Nature 2008;451:1116–20.

22. Yata K, Lloyd J, Maslen S, Bleuyard JY, Skehel M, Smerdon SJ, et al.Plk1 and CK2 act in concert to regulate Rad51 during DNA doublestrand break repair. Mol Cell 2012;45:371–83.

23. Weinstock DM, Nakanishi K, Helgadottir HR, Jasin M. Assaying dou-ble-strand break repair pathway choice in mammalian cells using atargeted endonuclease or the RAG recombinase. Methods Enzymol2006;409:524–40.

24. Boren T, Xiong Y, Hakam A, Wenham R, Apte S, Wei Z, et al. Micro-RNAs and their target messenger RNAs associated with endometrialcarcinogenesis. Gynecol Oncol 2008;110:206–15.

25. Gottardo F, Liu CG, Ferracin M, Calin GA, Fassan M, Bassi P, et al.MicroRNA profiling in kidney and bladder cancers. Urol Oncol2007;25:387–92.

26. LeeKH, LottermanC,Karikari C,OmuraN, FeldmannG,HabbeN, et al.Epigenetic silencing of microRNA miR-107 regulates cyclin-depen-dent kinase 6 expression in pancreatic cancer. Pancreatology2009;9:293–301.

27. Roldo C, Missiaglia E, Hagan JP, Falconi M, Capelli P, Bersani S, et al.MicroRNA expression abnormalities in pancreatic endocrine and aci-nar tumors are associated with distinctive pathologic features andclinical behavior. J Clin Oncol 2006;24:4677–84.

Huang et al.





28. Takahashi Y, Forrest AR, Maeno E, Hashimoto T, Daub CO, Yasuda J.MiR-107andMiR-185 can induce cell-cycle arrest in humannon–smallcell lung cancer cell lines. PLoS ONE 2009;4:e6677.

29. Martello G, Rosato A, Ferrari F, Manfrin A, Cordenonsi M, Dupont S,et al. A microRNA targeting dicer for metastasis control. Cell2010;141:1195–207.

30. Bohlig L, Friedrich M, Engeland K. p53 activates the PANK1/miRNA-107 gene leading to downregulation of CDK6 and p130 cell-cycleproteins. Nucleic Acids Res 2011;39:440–53.

31. Suzuki HI, Yamagata K, Sugimoto K, Iwamoto T, Kato S, Miyazono K.ModulationofmicroRNAprocessing byp53.Nature 2009;460:529–33.

32. YamakuchiM, LottermanCD, BaoC, Hruban RH, KarimB,Mendell JT,et al. P53-induced microRNA-107 inhibits HIF-1 and tumor angiogen-esis. Proc Natl Acad Sci U S A 2010;107:6334–9.

33. Huertas P, Jackson SP. Human CtIP mediates cell-cycle control ofDNA end resection and double strand break repair. J Biol Chem2009;284:9558–65.

34. Lewis BP, Burge CB, Bartel DP. Conserved seed pairing, often flankedby adenosines, indicates that thousands of human genes are micro-RNA targets. Cell 2005;120:15–20.

35. Chen PS, Su JL, Cha ST, Tarn WY, Wang MY, Hsu HC, et al. miR-107promotes tumor progression by targeting the let-7 microRNA in miceand humans. J Clin Invest 2011;121:3442–55.

36. Selcuklu SD, Yakicier MC, Erson AE. An investigation of microRNAsmapping to breast cancer related genomic gain and loss regions.Cancer Genet Cytogenet 2009;189:15–23.

37. Reinhold WC, Sunshine M, Liu H, Varma S, Kohn KW, Morris J, et al.CellMiner: a web-based suite of genomic and pharmacologic tools toexplore transcript and drug patterns in the NCI-60 cell line set. CancerRes 2012;72:3499–511.

38. Munshi NC, Hideshima T, Carrasco D, Shammas M, Auclair D, DaviesF, et al. Identification of genes modulated in multiple myeloma usinggenetically identical twin samples. Blood 2004;103:1799–806.

39. YoshikawaK, Ogawa T, Baer R, Hemmi H, HondaK, Yamauchi A, et al.Abnormal expression of BRCA1 and BRCA1-interactive DNA-repairproteins in breast carcinomas. Int J Cancer 2000;88:28–36.

40. Kato M, Yano K, Matsuo F, Saito H, Katagiri T, Kurumizaka H, et al.Identification of Rad51 alteration in patients with bilateral breastcancer. J Hum Genet 2000;45:133–7.

41. Somyajit K, Basavaraju S, Scully R, Nagaraju G. ATM- and ATR-mediated phosphorylation of XRCC3 regulates DNA double-strandbreak-induced checkpoint activation and repair. Mol Cell Biol2013;33:1830–44.

42. Shu J, Kren BT, Xia Z, Wong PY, Li L, Hanse EA, et al. GenomewidemicroRNA downregulation as a negative feedback mechanism in theearly phases of liver regeneration. Hepatology 2011;54:609–19.

43. Pouliot LM, Shen DW, Suzuki T, Hall MD, Gottesman MM. Contribu-tions of microRNA dysregulation to cisplatin resistance in adenocar-cinoma cells. Exp Cell Res 2013;319:566–74.

44. Francia S, Michelini F, Saxena A, Tang D, de Hoon M, Anelli V, et al.Site-specific DICER and DROSHA RNA products control the DNA-damage response. Nature 2012;488:231–5.

45. WeiW, Ba Z, GaoM,WuY, Ma Y, Amiard S, et al. A role for small RNAsin DNA double-strand break repair. Cell 2012;149:101–12.

46. Annibali D, Gioia U, Savino M, Laneve P, Caffarelli E, Nasi S. A newmodule in neural differentiation control: two microRNAs upregulatedby retinoic acid, miR-9 and -103, target the differentiation inhibitor ID2.PLoS ONE 2012;7:e40269.

47. Garzon R, Pichiorri F, Palumbo T, Visentini M, Aqeilan R, Cimmino A,et al. MicroRNA gene expression during retinoic acid-induced differ-entiation of human acute promyelocytic leukemia. Oncogene2007;26:4148–57.

48. Liao Y, Lonnerdal B. Global microRNA characterization reveals thatmiR-103 is involved in IGF-1 stimulated mouse intestinal cell prolifer-ation. PLoS ONE 2010;5:e12976.

49. Piao L, Zhang M, Datta J, Xie X, Su T, Li H, et al. Lipid-basednanoparticle delivery of Pre-miR-107 inhibits the tumorigenicity ofhead and neck squamous cell carcinoma. Mol Ther 2012;20:1261–9.

50. Neijenhuis S, Bajrami I, Miller R, Lord CJ, Ashworth A. Identification ofmiRNA modulators to PARP inhibitor response. DNA Repair 2013;12:394–402.






2013;11:1564-1573. Published OnlineFirst October 2, 2013.Mol Cancer Res Jen-Wei Huang, Yemin Wang, Kiranjit K. Dhillon, et al. RAD51D to Enhance ChemosensitivitySystematic Screen Identifies miRNAs That Target RAD51 and

Updated version

10.1158/1541-7786.MCR-13-0292doi:

Access the most recent version of this article at:

Material

Supplementary

http://mcr.aacrjournals.org/content/suppl/2013/10/03/1541-7786.MCR-13-0292.DC1

Access the most recent supplemental material at:

Cited articles

http://mcr.aacrjournals.org/content/11/12/1564.full#ref-list-1

This article cites 50 articles, 14 of which you can access for free at:

Citing articles

http://mcr.aacrjournals.org/content/11/12/1564.full#related-urls

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

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

Subscriptions

Reprints and

[email protected]

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

Permissions

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

.http://mcr.aacrjournals.org/content/11/12/1564To 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-13-0292

http://mcr.aacrjournals.org/content/suppl/2013/10/03/1541-7786.MCR-13-0292.DC1

http://mcr.aacrjournals.org/content/11/12/1564.full#ref-list-1

http://mcr.aacrjournals.org/content/11/12/1564.full#related-urls

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

mailto:[email protected]

http://mcr.aacrjournals.org/content/11/12/1564


dna damage and repair - molecular cancer research · implications: these ﬁndings demonstrate a...

Documents