METHOD Open AccessA genome scale overexpression screen to revealdrug activity in human cellsAnthony Arnoldo1,2,3, Saranya Kittanakom1,2,3, Lawrence E Heisler1,2,3,4, Anthony B Mak1,3, Andrey I Shukalyuk5,Dax Torti4, Jason Moffat1,3, Guri Giaever1,3,6,7 and Corey Nislow1,2,3,4,7*AbstractTarget identification is a critical step in the lengthy and expensive process of drug development. Here, we describea genome-wide screening platform that uses systematic overexpression of pooled human ORFs to understanddrug mode-of-action and resistance mechanisms. We first calibrated our screen with the well-characterized drugmethotrexate. We then identified new genes involved in the bioactivity of diverse drugs including antineoplasticagents and biologically active molecules. Finally, we focused on the transcription factor RHOXF2 whose overexpressionconferred resistance to DNA damaging agents. This approach represents an orthogonal method for functionalscreening and, to our knowledge, has never been reported before.BackgroundBiological systems tend to remain phenotypically stable inthe face of environmental challenges and genetic changes[1]. As such, genetic perturbation has become an efficienttechnique to dissect cellular functions. In this study, weused the concept of modulating gene dosage in humancells to gain insight into drug mode of action. Understand-ing the primary mechanism of action as well as the poten-tial polypharmacological effects of a drug can provideinsight into how to reduce detrimental side effects, un-cover new applications for novel indication and explain re-sistance mechanisms [2-4].Drug resistance in malignant tissues can be categorizedinto three main mechanisms: (i) drug distribution/metabol-ism (pharmacokinetics), (ii) heterogeneity of cancer cellsand (iii) tumor micro-environment [5,6]. Among the othercellular mechanisms, gene overexpression (for example, byamplification of the drug target) can titrate a drug’s effect.This is exemplified by the classic case of methotrexate re-sistance through the amplification of the gene DHFR inneoplastic tissue from an individual with disseminatedsmall-cell lung cancer that relapsed during methotrexatechemotherapy [7]. Other forms of overexpression resistanceinclude the up-regulation of 1) efflux pumps (for example,ATP-binding cassette transporters), 2) survival mechanisms(for example, anti-apoptotic proteins), 3) DNA damagerepair, 4) pathways for drug inactivation and 5) the overex-pression of target isotypes. Additionally, proteins down-stream of the inhibited target can be modulated in such amanner as to bypass the toxic effect of a drug.To dissect some of these mechanisms by which drugsact within the cell, we postulated that, when overexpressed,genes conferring resistance to a lethal chemical treatmentcan illuminate the drug mode of action. Overexpressinggenes to confer resistance to an otherwise toxic compoundis a well-established concept. Early studies showed that astreptomycin resistance gene cassette could function in abacterial tetracycline resistant plasmid [8] and it wasshown in other studies that cloning of the mouse dihydro-folate reductase into a bacterial plasmid provided resist-ance to trimethoprim in Escherichia coli [9]. A large-scaleapproach using the overexpression of a yeast genomicDNA library to identify genes conferring resistance to spe-cific drugs has validated the concept of gene dosage for thediscovery of drug targets in eukaryotes [10].A logical extension of this concept is to employ newlyavailable biological tools that enable the systematic perturb-ation of all protein function in human cultured cells usingRNA interference (RNAi), CRISPR (clustered regularlyinterspaced short palindromic repeats), cDNA libraries,transposons or small molecule inhibitors in combination* Correspondence: corey.nislow@ubc.ca1Department of Molecular Genetics, University of Toronto, Toronto, M5S 3E1,Canada2Banting and Best Department of Medical Research, University of Toronto,Toronto, M5S 3E1, CanadaFull list of author information is available at the end of the article© 2014 Arnoldo et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the CreativeCommons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, andreproduction in any medium, provided the original work is properly credited. The Creative Commons Public DomainDedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article,unless otherwise stated.Arnoldo et al. Genome Medicine 2014, 6:32http://genomemedicine.com/content/6/4/32with gene overexpression. Although testing all importanttherapeutic drugs against all proteins from the approxi-mately 20,000 genes of the human genome in all differenti-ated cell types is not currently feasible, we report the initialdevelopment of a gain-of-function approach to identifydrug mode of action by the overexpression of 12,200 hu-man ORFs (hORFs) [11,12] in the human cell line HEK293.In summary, we have developed a new experimentalpipeline to identify genes whose up-regulation suppressesthe toxic effect of chemicals in human cells and have suc-cessfully applied this strategy to seven pharmacologicalcompounds.MethodsCell culture and plasmidsHuman embryonic kidney HEK293 (ATCC), HEK293T(ATCC), non-small cell lung carcinoma NCI-H1299 (ATCC),chondrosarcoma SW1353 (gift from Johanne Martel-Pelletier, University of Montreal, Canada) and pancreaticadenocarcinoma HPAC (ATCC) cells were maintainedin DMEM (Wisent Inc. Montreal, Quebec, Canada) sup-plemented with 10% fetal bovine serum (Gibco, Carlsbad,California, US) at 37°C and 5% CO2. The rtTA express-ing HEK293_M2 (ATCC), human breast adenocarcin-oma MCF7_M2 (ATCC) and adenocarcinomic humanalveolar basal epithelial A549_M2 (ATCC) cells weremaintained in DMEM (Wisent Inc.) supplemented with10% fetal bovine serum (Gibco) at 37°C and 5% CO2.Blast phase chronic myelogenous leukemia K562 cells(gift from Reinhart Reithmeier, University of Toronto,Canada) were maintained in RPMI1640 (Sigma, St. Louis,Missouri, US) supplemented with 10% fetal bovineserum (Gibco) at 37°C and 5% CO2. Multiple myelomaU266B1 cells (gift from Aaron Schimmer, PrincessMargaret Cancer Centre, Canada) were maintained inIscove's modified Dulbecco's medium (IMDM; Gibco) sup-plemented with 10% fetal bovine serum (Gibco) at 37°Cand 5% CO2. Thyroid gland medullary carcinoma TT cells(gift from Gilbert Cote, MD Anderson Cancer Center,USA) were maintained in Ham’s F-12 K (Gibco) supple-mented with 10% fetal bovine serum (Gibco) at 37°C and5% CO2.The lentiviral destination vector pLD-T-IRES-Venus-WPRE-STOP was cloned using the following procedure:the Gateway cassette from pLS-Dest-EcF (gift of Dr TonyPawson, University of Toronto, Canada) was subcloned intopLJM17 (J Moffat, unpublished) using the restriction en-zymes MluI and XbaI, resulting in the pLD-puro-TRE (T)vector. IRES-Venus-WPRE were PCR amplified from thepSLIK-Venus [13] using the following primers: XcmI_IR-ES_F (5′-GCGCCTTTTCCAAGGCAGCCCTGGAATTCCGCCCCTCTCCCTCC) and NsiI_WPRE_R (5′-AAACAATGCATGTCGACGCGGGGAGGCGGCCCAAAGGGAGATCC). The IRES-Venus-WPRE amplicon was clonedinto the pLD-puro-T vector using the restriction en-zymes XcmI and NsiI to replace the human phospho-glycerate kinase promoter and puromycin resistancegene. A stop codon was introduced directly downstreamof the Gateway cassette by digesting pLD-T-IRES-Venus-WPRE with the restriction enzyme XbaI and bluntedusing DNA polymerase I, large (Klenow) fragment (NewEngland Biolabs Inc., Beverly, Massachusetts, US).The piggyBac PB-TGcMV-Neo plasmid used for the hitconfirmation was derived from the PB-TET (AddGene,Cambridge, Massachusetts, US). The 'IRES-beta-Geo'fragment from PB-TET was replaced by the 'promoterPGK-neomycin resistance gene' fragment by homologousrecombination in Saccharomyces cerevisiae. Briefly, PB-TET was digested with RsrII and ApaI to remove thebeta-geo gene. The fragment coding for the promoterPGK and neomycin resistance gene, the URA3 cassetteand the yeast 2 μ-origin of replication were PCR amplifiedand cloned into the digested PB-TET plasmid by homolo-gous recombination in yeast. Yeast was then transformedwith the p414-Cre plasmid in order to excise the URA3cassette and the 2 μ-origin of replication after expressionof the Cre recombinase.All short hairpin RNAs (shRNAs) against the RHOXF2and green fluorescent protein (GFP) mRNAs were de-rived from the RNAi Consortium (TRC) lentiviral librar-ies and obtained in the pLKO.1 vector. Five shRNAsagainst RHOXF2 were tested and were designed againstthe following regions: 5′-CGGGATGAGAGATGATTACTT (RHOXF2_shRNA_1); 5′-GACGAGAAAGAACTACAGGAT (RHOXF2_shRNA_2); 5′-AGAAGCATGAATGTGACTGAA (RHOXF2_shRNA_3); 5′-GAGGGCATTAATGGCAAGAAA (RHOXF2_shRNA_4); 5′-GTCGCTTACTGAAGAGGTCAA (RHOXF2_shRNA_5).Library preparation, virus production and lentiviralinfectionThe 12,212 hORF collection (representing 10,214 distinctgenes) represents version 3.1 of an ongoing effort to createa complete human set of protein-encoding genes [11,12].Initially based on the Mammalian Gene Collection cDNAcollection, that work transfers full-length ORFs (excludingthe 5′ and 3′ mRNA untranslated regions) into a Gatewaysystem. Version 3.1 was obtained from Open Biosystems(Thermo Scientific, Waltham, Massachusetts, US) and re-groups non-fully sequenced verified and non-clonal ORFs(limitations corrected in the 8.1 version [14]). The col-lection was divided into 34 minipools of 376 hORFs;for each minipool, the hORFs were cloned en massefrom the pDONR223 into the lentiviral expression vec-tor pLD-T-IRES-Venus-WPRE-STOP by Gateway LRreaction. After electroporation, transformants were se-lected on Luria Broth (LB) plus ampicillin and the lenti-viral vector was extracted.Arnoldo et al. Genome Medicine 2014, 6:32 Page 2 of 16http://genomemedicine.com/content/6/4/32Lentivirus was produced by normalizing the amount ofDNA for the 34 hORF minipools and by co-transfectionwith the packaging plasmid psPAX2 and the envelopeplasmid pMD2.G into the packaging HEK293T cells usingFuGENE (Roche Mississauga, Ontario, Canada).HEK293_M2 cells were then infected at a multiplicity ofinfection of 0.3. Doxycycline (2 μg/ml) was added to in-duce the expression of the Venus fluorescent protein andVenus-positive cells were sorted using a BD FACS AriaCell Sorter (East Ruherford, New Jersey, US). Because theexpression of the gene coding for Venus is linked to theexpression of the hORFs, selection of the fluorescent cellsallowed the selection for non-silent, stably integratedlentivirus as well as functional inducible hORFs.Genomic DNA extraction, library preparation and dataanalysis for the hORFeome representation inHEK293_M2 cellsWe thawed 6.3 × 106 HEK293_M2 cells harboring thevirally integrated human ORFeome version 3.1 collectionand these were cultured for 1 day to recover. GenomicDNA was recovered from 16 × 106 cells and 6 μg of gen-omic DNA (gDNA were used to amplify the collection(three PCR reactions with 2 μg of gDNA). FollowingPCR, the amplified ORFs were used directly to preparea Nextera sequencing library (Illumina Nextera DNASample Preparation Kit, San Diego, California, US) andsequenced on an Illumina MiSeq according to the man-ufacturer’s protocol. Fifty-nucleotide paired-end readswere aligned against a reference database consisting ofthe ORF sequences in the hORFeome 3.1 collectionusing bwa (version 0.6.1). Coverage at each base positionin each reference sequence was determined using thebedtools program genomeCoverageBed (v2.14.2). A cus-tom Perl script was used to summarize the coveragelevels to determine the mean coverage for each referencesequence and the proportion of each reference sequencewith different levels of coverage. Raw data have been de-posited in the ArrayExpress repository under accessionnumber E-MTAB-2498.hORFeome drug target screenAliquots of 500,000 HEK293_M2 cells were seeded in aT175 flask and grown for 15 hours before the induction ofhORF expression by addition of doxycycline. After 18 hoursof induction, cells formed micro-colonies that were cul-tured in the presence of drug or DMSO (as control) for 2to 3 weeks until distinct colonies were detectable.After each screen, surviving cells were harvested andgDNA was extracted as follows. Cells were lysed usingSNET buffer (10 mM Tris pH 8.0, 0.1 M EDTA, 0.5%SDS, 0.1 mg/ml Proteinase K, 25 μg/ml RNAse A). gDNAwas isolated using one volume of phenol/chloroform/isoa-myl alcohol (25:24:1) pH 8.0, precipitated by adding twovolumes of 95% ethanol (-20°C), washed with ice cold 70%ethanol and resuspended in Tris-HCl pH 8.0.During the screen selection experiments, only a minorityof hORFs protected HEK293_M2s against the cytotoxiceffect of the drug. Compared to the initial cell population,HEK293_M2 cells harboring the most resistant genesshould be over-represented. In order to further amplifythe set of resistant genes from the resulting reduced popu-lation, hORFs were PCR amplified from only 80 ng of gen-omic DNA. The sequences of the forward and reverseprimers specific for the inserted hORFs were 5′-CGGTACCCGGGGATCCTCTAGTCAGCTGAC and 5′-CCATTTGTCTCGAGGTCGAGAATTCTAGCTAGAATC, re-spectively. Each reaction was carried out in a 50 μl volumecontaining 25 μl of 2× Phusion Flash High-Fidelity MasterMix (Finnzymes, Espoo, Finland), 200 nM of each primerand 80 ng of genomic DNA. The PCR profile was: 1 mi-nute at 98°C for one cycle; 10 s at 98°C, 20 s at 65°C, 4 mi-nutes at 72°C for 35 cycles; 10 minutes at 72°C for onecycle. PCR products were purified using QIAquick PCRPurification Kit (QIAGEN Venlo, Limburg, Netherlands),150 ng of purified PCR product was biotinylated usingthe BioPrime DNA Labeling Kit (Invitrogen, Carlsbad,California, US) and unincorporated biotin-14-dCTP wasremoved by passing the samples through Sephadex G-50columns (GE Healthcare Fairfield, Connecticut, US).Sample (150 ng) was added to the Affymetrix GeneChipHuman Gene 1.0 ST array and hybridized at 45°C for17 hours with a rotation of 60 rpm. Chips were washed andstained with SAPE (2× MES staining buffer, 20 mg/mL bo-vine serum albumin (BSA) and 1 mg/ml streptavidin-phycoerythrin), washed on an Affymetrix fluidics stationand scanned.Representation of the complete hORFeome collectionwas assessed using the same experimental conditionswith some modifications. gDNA was extracted from atleast 16 × 106 cells (approximately 1,300 cells per hORF),3 PCR reactions were performed using 2 μg of gDNAeach (approximately 75 genomes per hORFs), two bio-tinylation reactions were performed using 500 ng ofcombined PCR products and 3.5 μg of sample was addedto an Affymetrix GeneChip Human Gene 1.0 ST array.Microarray raw data have been deposited in theArrayExpress repository [15] under accession number E-MTAB-2493.Selection and individual validation of over-representedhORFsThe hORFs conferring resistance to the drug were iden-tified in a step-wise procedure. Screen results were firstvisualized by plotting the log2 of the signal intensity foreach hORF retrieved via PCR from cells cultured in thepresence of drug (on the x-axis) and by plotting the log2ratio of the signal intensity for each hORF of the cellsArnoldo et al. Genome Medicine 2014, 6:32 Page 3 of 16http://genomemedicine.com/content/6/4/32cultured in the presence of drug divided by the signal in-tensity for each hORF of the cells grown in the presenceof DMSO (on the y-axis). Genes with a log2(drug/DMSO)>3 and a log2(drug) >6 enrichment were then listed andranked. Primary hits selected for validation were thosepreviously enriched genes in common among the drugscreen replicates (at least two independent experiments).For validation, 10,000 HEK293-M2 cells were seeded in300 μl of DMEM plus 10% iFBS plus penicillin/strepto-mycin in a 48-well plate pretreated with poly-L-lysine(Sigma, P4832). After 12 hours of incubation, gene expres-sion was induced by the addition of doxycycline (2 μg/mlfinal concentration). After 18 hours of induction, drugwas added and cell growth was assessed 3 days after drugaddition by sulforhodamine B (SRB) assay as previouslydescribed [16].ImmunostainingThe HEK293_M2 + RHOXF2 stable cell line (PB-TGcMV-Neo) were seeded in an eight-well chamber slide (BDFalcon, East Ruherford, New Jersey, US.) and cultivatedovernight in the presence or absence of doxycyclin (2 μg/mlfinal). Cells were treated with 50 nM of mitomycin C for2 days before fixation with 2% paraformaldehyde andpermeabilization with 0.3% Triton X-100. Cells wereblocked for 30 minutes with blocking/dilution buffer(10% goat serum, 0.5% NP-40, 0.5% saponin, 1×phosphate-buffered saline (PBS)) and incubated with theanti-phospho-histone H2A.X (serine 139, clone JBW301;Millipore, Billerica, Massachusetts, US) overnight at 4°C.Cells were then incubated with an Alexa 546 anti-mouseantibody for 1 hour at room temperature, incubated withDAPI (0.8 μg/ml; Sigma) for 10 minutes and mountedusing ProLong Gold (Invitrogen). Images were capturedusing a 40× dry objective. Nuclei and γ-H2A.X foci werequantified using CellProfiler [17].RNA extraction and library preparation for gene expressionprofiling of the RHOXF2 overexpressing cell lineThe HEK293_M2 + RHOXF2 stable cell line (PB-TGcMV-Neo) was seeded in 10-cm dishes, cultivated 24 hours inthe presence or absence of doxycycline (2 μg/ml final) be-fore being treated with 40 nM of mitomycin C or DMSO(control) during a 2-day period. Total RNA was extracted(RNeasy Mini Kit, QIAGEN), mRNA-focused librarieswere generated from 1 μg of total RNA (Illumina TruSeqRNA Sample Preparation Kit V.2) and sequenced using anIllumina HiSeq2000 according to the Illumina protocol.Raw data have been deposited in the ArrayExpress reposi-tory under accession number E-MTAB-2497.RNA-sequencing data analysisSingle-end reads, 51 nucleotides in length, were gener-ated on the Illumina HiSeq. Sequence data were alignedto the UCSC hg19 reference genome using TopHat(v2.0.0), provided with a RefSeq GTF file and instructedto align only across known junctions. Differential ex-pression, as fragments per kilobase of exon per millionfragments mapped (FPKM) differences, were generatedby analysis with cuffdiff (v1.1.0) using the indicated com-parisons. Based on their differential expression, potentialactivated and repressed isoforms were identified and re-lated enriched biological processes were assessed usingGOrilla [18].RHOXF2 quantitative RT-PCRTotal RNA from 20 different normal human tissues(each pool from 3 donors) were obtained from the First-Choice Human Total RNA Survey Panel (Ambion, Austin,Texas, US). Total RNA (5 μg) from each tissue was usedfor oligo(dT)12-18-primed reverse transcription using theSuperScript II reverse transcriptase (Invitrogen). Quantita-tive PCR was performed using the LightCycler 480 SYBRGreen I Master (Roche) and the LightCycler 480 Systemwith the following primers:QhRHOXF1_1F (5′-ACCGTGTTCTACTGCCTGAGTGTA) and QhRHOXF1_1R (5′-TTCATGCCGTTCTCGTGGTTCACA) on RHOXF1 exon 1; QhRHOXF1_2F (5′-TGGAGGAGCTGGAAAGTGTT) and QhRHOXF1_2.2R(5′-GGCCCTTTTATTCTTAAACC) spanning RHOXF1exon 1 and exon 3; QhRHOXF2_1F (5′-CCGGACCAGTGTAGCCAGTA) and QhRHOXF2_1R (5′-TCTTTTTCTTCTCCGCCTTG) spanning RHOXF2 exon 1 and exon 2;QhRHOXF2_2F (5′-ATGGTGCTGTCGCTTACTGA) andQhRHOXF2_2R (5′-TCGAGGTCTCCTTCCCATAG) onRHOXF2 exon 2; QhRHOXF2_4F (5′-CAGCGGGATGAGAGATGATT) and QhRHOXF2_4R (5′-TTGGGGAATGTGAAAGAAGG) on RHOXF2 exon 3.Housekeeping/reference genes were: QhCYCG_F (5′-CTTGTCAATGGCCAACAGAGG) and QhCYCG_R (5′-GCCCATCTAAATGAGGAGTTGGT) on CYCG (cyclophilinG); QhGUSB_F (5′-ACGCAGAAAATATGTGGTTGGA)and QhGUSB_R (5′-GCACTCTCGTCGGTGACTGTT)on GUSB (beta-glucuronidase); QhActbF1 (5′-GAAGTCCCTTGCCATCCTAAAAG) and QhActbR1 (5′-AGGACTGGGCCATTCTCCTTA) on ACTB (beta-actin); QhEe-f1a1F2 (5′-CTGAACCATCCAGGCCAAAT) and QhEef1a1R2 (5′-AGCCGTGTGGCAATCCA) on EEF1A1 (eukary-otic translation elongation factor 1 alpha 1); QhGAPDH_F1 (5′-CATGAGAAGTATGACAACAGCCT) andQhGAPDH_R1 (5′-AGTCCTTCCACGATACCAAAGT)on GAPDH (glyceraldehyde-3-phosphate dehydrogenase).RHOXF2 immunohistochemistryFormalin-fixed, paraffin-embedded sections of humannormal testis were deparaffinized for 12 hours at 58°Cand rehydrated. Heat-induced epitope retrieval was insodium citrate buffer pH 6.0 using a pressure cookerArnoldo et al. Genome Medicine 2014, 6:32 Page 4 of 16http://genomemedicine.com/content/6/4/32[19]. All antibodies were diluted in Tris-buffered Saline(TBS) with 1% BSA. Sections were blocked in 10% goatserum (Gibco) for 2 hours at room temperature. Theanti-RHOXF2 primary antibody (Sigma, HPA003314)was used at a 1:150 dilution and incubated overnight at 4°Cin a humid chamber. Endogenous peroxidase activity wassuppressed by incubating the section in 0.3% H2O2 for15 minutes. Horse radish peroxidase (HRP)-conjugated sec-ondary antibody (anti-rabbit IgG-HRP; GE Healthcare) wasused at a 1:250 dilution and incubated for 1 hour at roomtemperature. 3,3′-Diaminobenzidine (Abcam) was used aschromogen. Sections were counterstained with Mayer’shematoxylin (Sigma) before dehydration and mounting.Western blottingFor the detection of endogenous RHOXF2, total cell ly-sates were prepared from 200,000 cells. Briefly, sampleswere lysed in 2× SB loading buffer (60 mM Tris-HClpH6.8, 3% SDS, 10% glycerol, 0.05% bromophenol blue,10% beta-mercaptoethanol) and subjected to SDS-PAGE(12%). The RHOXF2-specific mouse polyclonal antibody(Abcam, ab67811) and the tubulin specific rat antibody(Abcam, ab6160) were used at a dilution of 1:800 and1:40,000 respectively. Immunological complexes were vi-sualized by enhanced chemiluminescence using HRP-conjugated anti mouse or anti rat IgG.Drug dose-response curvesK562 cells (100,000) were cultured in 600 μl of media ina 48-well plate for 24 hours before the addition of di-luted drug. Two days after drug treatment, the percent-age of drug inhibition was assessed by counting K562using a Coulter particle count and cell analyser andcompared to cells grown in the presence of DMSO only.For the adherent HPAC, SW1353, and NCI-H1299 celllines, 5,000 cells were seeded in 300 μl of media in 96-well plate and incubated for 24 hours prior to drugaddition. Following two days of drug treatment, the per-centage of drug inhibition was estimated by SRB assay[16] and compared to cells grown in DMSO as control.ResultsDevelopment of the assay: DHFR and methotrexateWe developed an original phenotypic assay in which themammalian cell line HEK293_M2 was used to identifyhORFs capable of rescuing small molecule toxicity whenoverexpressed (Figure 1; Additional file 1). As a proof ofconcept, we rescued methotrexate toxicity by overex-pression of its target, the gene encoding the enzymedihydrofolate reductase (DHFR).We first cloned the human DHFR gene into the Gateway-compatible lentiviral vector pLD-T-IRES-Venus-WPRE-STOP where gene expression is under the control of theTet-On promotor (Additional file 2). After transductionof the rtTA-expressing cell line HEK293_M2 by lentivirus,we detected a high level of protein in the presence ofdoxycycline by western blotting (Figure 2a; Additional file3). No protein was detected in the absence of doxycycline,suggesting robust control of transgene expression. Cellswere next exposed to increasing doses of methotrexate inthe absence or presence of different doses of doxycycline.Increasing levels of DHFR overexpression in HEK293_M2cells rescued methotrexate toxicity in a dose-dependentmanner, demonstrating that the lentiviral construct wasfunctional (Figure 2b; Additional file 3).We implemented our approach on a larger scale bycloning, en masse, a pool of 376 random hORFs (includingDHFR) into the lentiviral vector. Corresponding lentiviruswere used to generate a population of HEK293_M2 cellswith the stably integrated hORFs. Gene expression wasinduced as before by addition of doxycycline and cellswere exposed to a lethal concentration of methotrexate.Ten days after treatment, surviving cells were harvested,genomic DNA was extracted, and hORFs were PCR amp-lified and hybridized on an Affymetrix GeneChip HumanGene 1.0 ST array to determine which hORFs conferredresistance to methotrexate. DHFR was found as the pre-dominant gene in this pool of 376 hORFs whose overex-pression provided resistance to methotrexate, validatingour strategy at the scale of several hundred pooled clones(Additional file 4).The same approach was next applied to the completecollection of 12,212 hORFs. Here, we divided the collec-tion into 34 minipools of 376 hORFs each, transferredthem en masse into the pLD-T-IRES-Venus-WPRE-STOPlentivirus vector and then pooled 34 minipools together toproduce sufficient quantities of lentivirus for the infectionof HEK293_M2 cells at a multiplicity of infection of 0.3(Figure 1). Because induction of the Venus marker isdependent on hORF expression, addition of doxycyclinefor 24 hours generated sufficient fluorescence to sortthose cells with stable lentivirus integration (Additionalfile 5). The presence and relative abundance of the virallyintegrated hORFs were assessed using next-generation se-quencing. Based on the identification of at least 95% ofthe sequence with a minimum of 10× coverage, 87% ofthe mappable hORFs were considered as present in theHEK293_M2 cells. Further analysis of the sequencing re-sults demonstrated a size-dependent relative abundanceof the hORFs, with smaller clones being more representedin the population (Additional file 6). An aliquot of 500,000HEK293_M2 cells (an average of 41 cells per hORF) wasseeded in a T175 flask, and grown for 15 hours beforethe induction of hORF expression by addition of doxy-cycline. After 18 hours of doxycycline induction, cellsformed micro-colonies, which were cultured in the pres-ence of methotrexate for 2 weeks. hORFs derived fromthe surviving cells were identified by PCR amplificationArnoldo et al. Genome Medicine 2014, 6:32 Page 5 of 16http://genomemedicine.com/content/6/4/32and microarray hybridization. As we found for the mini-pool experiment, DHFR was the top gene conferring re-sistance to methotrexate when overexpressed (Figure 2c).These results encouraged us to test the same strategy fordrugs with diverse modes of action.Screens with additional drugsWe tested six additional diverse drugs, including theantifolate aminopterin, the lipase inhibitor orlistat,the Hsp90 inhibitor 17-(dimethylaminoethylamino)-17-demethoxygeldanamycin (a geldanamycin analog) andLentiviruspLD-T-IRES-Venus-WPRE-STOPCell SortinghORFxSpectpDONR233-hORFxE.coli stocks12,212 ORFsMOI= 0.3HEK293_M2 (rtTA)Drug selectionTetOndrugHitgDNA PCRmicroarraygeneSignal intensityIRES VenushORFxAmpNGSabdoxFigure 1 A gain-of-function cell-based assay to characterize the mode of action of small chemical compounds. (a) A collection of 12,212hORFs was cloned en masse into the lentiviral expression vector pLD-IRES-Venus-WPRE-STOP. The resulting constructs were used to producelentivirus and infect the rtTA-expressing cell line HEK293_M2 at a multiplicity of infection (MOI) of 0.3. Sorting of the Venus-positive cells resulted in theisolation of human cells with the functional integrated lentiviral constructs. (b) After seeding, doxycycline (dox) was added to the cells in order to in-duce the expression of the hORFs. After selection in the presence of a lethal dose of drug, surviving cells were harvested, gDNA extracted and the na-ture of the hORFs providing resistance to the chemical identified after hybridization on microarray. NGS, next-generation sequencing.c-tubulinDHFR DHFRLog2(methotrexate/DMSO)Log2(methotrexate signal intensity)12,212 hORFs[methotrexate] nMPercentage survival0 0.004 0.016 0.063 2dox μg/ml02040608010015 60 240 9600 ug/ml dox0.0039 ug/ml dox0.0156 ug/ml dox0.0625 ug/ml dox2 ug/ml doxabFigure 2 DHFR was identified as the top candidate during a large-scale screen against methotrexate. (a) Immunoblotting showing theconditional expression of DHFR in HEK293_M2 cells. Addition of 4 ng/ml, 16 ng/ml, 63 ng/ml and 2,000 ng/ml of doxycycline (dox) inducedDHFR protein expression in the stable cell line HEK293_M2. Alpha tubulin was used as a loading control. For clarity, protein expression level at0.0032 ng/ml of doxycycline has been omitted. (b) Single DHFR overexpression rescued methotrexate toxicity in HEK293_M2 cells. The genecoding for DHFR was cloned into the lentiviral vector and used to create stable HEK293_M2 cells able to conditionally express DHFR. Dose-response curves of the cells grown in the presence of an increasing concentration of doxycycline were calculated after 2 days of exposure to 15,30, 60, 125, 250, 500 and 1,000 nM of methotrexate and compared with those for cells cultured in the presence of DMSO as control. Cell survivalwas quantified using a SRB assay. Errors bars represent the standard deviation for triplicate assays. (c) Identification of genes conferring resistanceto methotrexate when overexpressed. HEK293_M2 cells harboring the 12,212 hORF collection were grown in the presence of a lethal dose ofmethotrexate. The nature of the hORFs conferring resistance to the drug was identified by plotting the log2 of the signal intensity for each hORFin the cells cultured in the presence of methotrexate on the x-axis; and by plotting the log2 ratio of the signal intensity for each hORF of the cellscultured in the presence of the drug divided by the signal intensity for each hORF of the cells grown in the presence of DMSO on the y-axis.Arnoldo et al. Genome Medicine 2014, 6:32 Page 6 of 16http://genomemedicine.com/content/6/4/32three DNA damaging agents (bleomycin, cisplatin andmitomycin C) based on the diversity of their mode ofaction (Table 1).Each drug was screened at least twice and the top hits(those hORFs most enriched in the final sample ofcells; see Methods) were selected for validation. Of 123candidates, we successfully cloned 120 hORFs intothe modified Gateway-compatible piggyBac vector PB-TGcMV-Neo (Additional file 7) and generated 120 inde-pendent HEK293_M2 stable cell lines. Protection againstthe drug toxicity was assessed two days after culturingcells in the presence of four concentrations of each drugwith or without doxycycline induction. Relative cell dens-ities were compared with cells containing an empty vectorcontrol construct and were quantified by SRB assay. Atotal of 17 unique hits were confirmed to rescue their re-spective drug toxicity (Table 2). The number of validatedhits varied considerably according to the compound, withbleomycin and 17-DMAG having low validation rates (7%and 3%, respectively) and, in contrast, cisplatin and mito-mycin C having high validation levels (27% and 42%, re-spectively) (Table 2). Several scenarios, both technical andbiological, may explain this modest confirmation rate.First, PCR amplification of the hORFs from gDNA andprobe cross-hybridization on microarrays can be expectedto generate a certain percentage of false positives, despiteeach screen having been performed in duplicate. Second,since this version of the hORFeome library is not fullysequence verified, we speculate that some genes containmutations that impair proper expression. We attempted tominimize the impact of mutations by restriction verifyingfour independent colonies from the original pDONR233bacterial stock, cloning positive clones as a pool for sub-sequent cloning into the PB-TGcMV-Neo vector andtransfection into HEK293_M2 cells. Finally, the biologicalmechanisms by which these agents work will influence thenumber of hits obtained from each screen, and thereforeaffect the number of hits that can be confirmed. For thisproof-of-principle study we chose to be conservative,selecting more hits for validation at the cost of a highfalse-positive rate.Screens for the antifolates methotrexate and aminopterinFor the antifolate screens, the single confirmed ORF wasDHFR. This result highlights the fact that when a singlehORF confers a strong growth advantage, it can domin-ate the other clones in the population and potentiallymask the detection of other potential protective genes.One way to address this and potentially improve the dy-namic range for this drug class would be to repeat thesescreens with a pool of hORFs lacking DHFR.Among the potential expected hits present in our col-lection was the enzyme gamma-glutamyl hydrolase(GGH), which catalyzes the removal of polyglutamatesfrom methotrexate. Therefore, overexpression of GGHwould be expected to decrease the intracellular presenceTable 1 Characteristics of screened compoundsDrug name Screen concentration Category* Clinical use Known target(s) ReferenceMethotrexate 60 nM Antimetabolites/ immunosuppressants Antineoplastic DHFR [20]AntirheumaticDermatologic agentAminopterin 20 nM Antimetabolites (discontinued) Dermatologic agent (phase I) DHFR, FPGS(?) [21]Antineoplastic agent (phase II)Orlistat 20 μM Antiobesity preparations, excludingdiet productsAnti-obesity agent LPL [22]PNLIP [23]FASNBleomycin 1 μM Cytotoxic antibiotics and relatedsubstancesAntineoplastic agent DNA [24]LIG1LIG3Cisplatin 2 μM Other antineoplastic agents Antineoplastic agent DNA [25]Mitomycin C 80 nM Cytotoxic antibiotics and relatedsubstancesAntineoplastic agent DNA [26]17_DMAG 70 nM Experimental Hsp90 inhibitor Antineoplastic agent (phases I and II) Hsp90 [27]HSP90AA1HSP90AA2HSP90B1HSP90AB1Arnoldo et al. Genome Medicine 2014, 6:32 Page 7 of 16http://genomemedicine.com/content/6/4/32of methotrexate-polyglutamate, thereby increasing theefflux of methotrexate out of the cell and decreasingcytotoxicity. However, GGH overexpression was demon-strated to be insufficient to produce resistance to thedrug in a human fibrosarcoma (HT-1080) and a humanbreast carcinoma (MCF-7) cell line [28].DHFRL1 (dihydrofolate reductase like-1) consistentlyappeared as a top hit for screens with both methotrexateand aminopterin but failed to confer resistance when in-dividually expressed in HEK293_M2 cells. Based on thestrong similarity between DHFR and DHFRL1 (a total of14 non-synonymous changes), we hypothesized that theobserved unspecific signal was probably due to cross-hybridization. Interestingly, a similar problem occurredwhen the same strategy using the same hORFeome li-brary was tested in the model organism S. cerevisiae.Both DHFR and DHFRL1 were enriched in the screensagainst methotrexate (and their expression confirmed atthe protein level) but only DHFR conferred methotrex-ate resistance (data not shown).Orlistat screensOrlistat is a lipase inhibitor that reportedly targets LPL(lipoprotein lipase), PNLIP (pancreatic lipase) and FASN(fatty acid synthase). Although LPL is absent from our col-lection, PNLIP and FASN are present. However, genes forthe latter two did not confer overexpression resistance.Our screen identified the gene coding for monoglycer-ide lipase (MGLL) as a resistance hit. MGLL hydrolyzesintracellular triglyceride stores in adipocytes and othercells to fatty acids and glycerol. The enzyme might alsobe involved in the hydrolysis of monoglycerides [29]. Invitro experiments suggest that orlistat can inhibit theMGLL-like activity in rat-cerebellar membranes and ratcerebellar homogenate [30], rendering the human MGLLa potential direct target for orlistat.Another validated hit, PON3, is also involved in lipidmetabolism. The gene encodes paraoxonase 3, whichis secreted into the bloodstream and associates withhigh-density lipoprotein. The protein can also rapidlyhydrolyze lactones and inhibit the oxidation of low-density lipoprotein.Bleomycin screensAlthough the exact mechanism of action of bleomycin isunknown, available evidence indicates that its main modeof action is via the inhibition of DNA synthesis, with add-itional evidence for its inhibition of RNA and, to a lesserextent, protein synthesis [31].We show that overexpression of IMMT (inner mem-brane protein, mitochondrial) partially rescues bleomycintoxicity. Bleomycin treatment induces damage of both nu-clear and mitochondrial DNA [32]. A recent study demon-strated that the mitochondrial localization of PARP-1Table 2 Validated hitsDrug Gene name Accession number Description Number of validated/testedcandidate genesMethotrexate DHFR BC000192 Dihydrofolate reductase 1/3Aminopterin DHFR BC000192 Dihydrofolate reductase 1/9Bleomycin IMMT BC002412 Inner membrane protein, mitochondrial (mitofilin) 2/30STX3 BC007405 Syntaxin 3Cisplatin MAP2K1IP1 BC026245 Mitogen-activated protein kinase kinase 1 interacting protein 1 4/15MMD BC026324 Monocyte to macrophage differentiation-associatedPTPN2 BC016727 Protein tyrosine phosphatase, non-receptor type 2RHOXF2 BC021719 Rhox homeobox family, member 2Mitomycin C C20orf54 BC009750 Chromosome 20 open reading frame 54 8/19CCDC45 BC009518 Coiled-coil domain containing 45ELF5 BC029743 E74-like factor 5 (ets domain transcription factor)PTPN2 BC008244 Protein tyrosine phosphatase, non-receptor type 2RHOXF2 BC021719 Rhox homeobox family, member 2TMEM150 BC050466 Transmembrane protein 150USPL1 BC038103 Ubiquitin specific peptidase like 1ZFP64 BC012759 Zinc finger protein 64 homolog (mouse)17-DMAG EIF4B BC073139 Eukaryotic translation initiation factor 4B 1/29Orlistat MGLL BC006230 Monoglyceride lipase 2/18PON3 BC070374 Paraoxonase 3Arnoldo et al. Genome Medicine 2014, 6:32 Page 8 of 16http://genomemedicine.com/content/6/4/32requires interaction with IMMT and that it is involved inthe maintenance of mitochondrial DNA integrity [33]. Theauthors also suggest that IMMT overexpression results inan increase of PARP-1 in the extracellular compartment.We therefore speculate that IMMT overexpression recruitssufficient PARP-1 into the mitochondrion to maintainmitochondrial DNA integrity and promote cell survival.Cisplatin and mitomycin C screensRHOXF2 and PTPN2 were found to suppress the tox-icity of cisplatin and mitomycin C. Together with theantifolates, these DNA damaging agents were the onlyscreens yielding identical hits. Also, as for methotrexateand aminopterin, both these alkylating agents have thesame mechanism of action.Another validated hit for mitomycin C, ZFP64 (zincfinger protein 64 homolog), is noteworthy because it hasbeen shown to be potentially phosphorylated by ATMand ATR, the major signal transducing kinases of theDNA damage response in response to ionizing radiation[34]. This observation is consistent with a potential rolefor ZFP64 in the repair of DNA damage generated byboth alkylating agents and ionizing radiation.RHOXF2 overexpression confers resistance to several DNAdamaging agentsAmong the top candidates from the screens and confirma-tions, we focused on RHOXF2, a relatively uncharacterizedmember of the homeobox gene family, because this gene,when overexpressed, provided resistance to both cisplatinand mitomycin C. RHOXF2 clearly reduced drug-inducedtoxicities when overexpressed in the HEK293_M2 cell linetreated with cisplatin (1 to 2.5 μM) or with mitomycin C(20 to 80 nM). Rescue efficiency ranged from 45 to 60%and 30 to 42%, respectively (Figure 3a). This observationwas confirmed using an independent cell biological assayin which overexpression of this transcription factor signifi-cantly reduced the formation of γ-H2A.X foci in cellstreated with 50 nM of mitomycin C for 2 days (Figure 3b).We next asked if RHOXF2′s protective effect was re-stricted to cisplatin and mitomycin C or if this gene, whenoverexpressed, could also suppress the toxicity of DNAdamaging agents with different mechanisms of action. Ac-cordingly, we assessed the cytotoxicity of antimetabolites(methotrexate, 5-fluorouracil), a radiomimetic (bleomycin),replication inhibitors (camptothecin, doxorubicin, etopo-side), mitaplatin, 4-nitroquinoline-1-oxide and several ex-perimental platinum anticancer agents (PT-ACRAMTU(EN), PT-ACRAMTU(PN), PT-ATUCA, PT-AMIDIN) onHEK293-M2 cells expressing or not the transcription factor(Figure 3c; Additional file 8). With the exception of thetopoisomerase I inhibitor camptothecin, all drugs whosetoxicity was suppressed by RHOXF2 overexpression dir-ectly bind to DNA to exert their effects.These observations motivated us to ask (i) if and howRHOXF2, an uncharacterized transcription factor, mod-ulates gene expression in HEK293_M2 cells and (ii) ifthe nature of the regulated genes can illuminate the pro-tective response to DNA damaging agents. We culturedHEK293_M2 cells with or without overexpression ofRHOXF2 in the presence of 40 nM of mitomycin Cor DMSO (control) and profiled their transcriptionalresponse by RNA-seq. The sequencing data confirmedthe RHOXF2 transcript was absent in the uninducedHEK293_M2 samples and verified its strong induction inthe presence of doxycycline regardless of the drug treat-ment (Additional file 9). Neither RHOXF1 nor RHOXF2Bexpression was affected by RHOXF2 overexpression(Additional file 9). Interestingly, and despite their limitednumber, Gene Ontology analysis of the genes whoseexpression was induced during RHOXF2 overexpres-sion showed that the most significantly enriched path-ways are related to DNA damage processes (P < 7 × 10-5and P < 3 × 10-4), stress response (P < 3 × 10-4), apop-tosis (P < 4 × 10-4) and cell cycle (P < 7 × 10-4) (Figure 3d;Additional file 9). As anticipated, in the absence ofRHOXF2 and the presence of mitomycin C, we found astrong enrichment for genes related to the response tostress (P < 3 × 10-5) and regulation of protein metabolicprocess (P < 9.5 × 10-5). Finally, in the presence of RHOXF2,genes involved in the regulation of signal transduction(P < 4 × 10-4) were induced during drug treatment. There-fore, we suggest that when treated with DNA damagingagents, the protective effect conferred by overexpressionof RHOXF2 is due, in part, to its activation of several keygenes involved in the response to DNA damage.Finally, we asked if the RHOXF2-dependent protectiveeffect observed in HEK293_M2 cells was restricted toone cell type. We then tested if RHOXF2 overexpressioncould confer resistance to the rtTA-expressing humanbreast adenocarcinoma cell line MCF7_M2 and theadenocarcinomic human alveolar basal epithelial cellline A549_M2. As quantified by both a two-fold shiftin IC50 (44.91 μM to 88.17 μM) and by the 21% decreasein toxicity at 3 μM, RHOXF2 overexpression protectedMCF7_M2 cells against cisplatin-mediated toxicity(Additional file 10a). In the absence of RHOXF2, addi-tion of doxycycline resulted in a modest increase in IC50(41.06 μM to 58.59 μM; empty vector). The effect ofthe vector alone, while detectable, is less than that ob-served in the presence of RHOXF2. In contrast, althoughRHOXF2 overexpression in A549_M2 cells providedsome apparent resistance (shift in IC50 from 448 nM to1,099 nM), the measured effect was, in fact, largely dueto doxycycline (shift in IC50 from 718 nM to 1,454 nMwith empty vector) (Additional file 10b). In conclusion,our experiments suggest that the protective effect ofRHOXF2 is not confined to a single genetic backgroundArnoldo et al. Genome Medicine 2014, 6:32 Page 9 of 16http://genomemedicine.com/content/6/4/32and expression of the transcription factor is sufficient toprotect both HEK293 and MCF7 cells against the DNAdamaging agent cisplatin. As with all large-scale screens,the biological significance of this observation should befurther confirmed, for example, by interrogating a largerpanel of cell lines.Based on its viability and transcriptional phenotypes inHEK293_M2 cells, we further investigated the potentialrole for RHOXF2 in neoplastic cells.In normal tissue, RHOXF2 expression is confined to testisbut RHOXF2 is expressed in several cancer cell linesIn a small-scale study performed by northern blot analysison normal tissues, expression of the transcription factorRHOXF2 was initially reported to be restricted to testis[35]. We performed a large-scale systematic survey interro-gating a more representative panel of 20 different normalhuman tissues (3 donors per tissue type) by quantitativeRT-PCR (Figure 4a). We first confirmed RHOXF2′s testis--2.5-2-1.5-1-0.500.520 40 60 80 mitomycin C (nM)Difference change in absorption/no drug controlrescue (%)0 36.95 42.20 35.69 29.95Empty vector, no doxEmpty vector, with doxRHOXF2 vector, no doxRHOXF2 vector, with dox0************ab50nM mitomycin C-dox +dox+dox50nM mitomycin C DMSOc* Anatomical Therapeutic Chemical Classification SystemDrug Type* Mode of action RHOXF2 overexpression rescueCisplatin Other antineoplastic agents DNA cross linking, alkylation RescueMitomycinC Cytotoxic antibiotics and related substances DNA cross linking, alkylating like RescueMitaplatin None Bind to nuclear DNA and mitochondrial pyruvate dehydrogenase RescuePT-ACRAMTU(EN) None Bind DNA RescuePT-ACRAMTU(PN) None Bind DNA RescuePT-ATUCA None Bind DNA RescueCamptothecin None Topoisomerase I inhibitor RescueMethotrexate Antimetabolites, Immunosuppressants Inhibit the metabolism of folic acid No rescue5-fluorouracil Antimetabolites Pyrimidine analog, irreversible inhibition of thymidylate synthase No rescueBleomycin Cytotoxic antibiotics and related substances Intercalation or interaction with DNA minor groove No rescueDoxorubicin Cytotoxic antibiotics and related substances Intercalating DNA, inhibition of topoisomerase II progression No rescueEtoposide Plant alkaloids and other natural products Topoisomerase II inhibitor No rescueHydroxyurea Other antineoplastic agents Ribonucleotidereductase inhibition No rescue4 -nitroquinoline -1-oxide None DNA adduct forming agent No rescuePT-AMIDIN None Bind DNA No rescue-log(P-value)0 1 2 3 4 5regulation of signal transductionregulation of execution phase of apoptosiscovalent chromatin modificationhistone acetylationregulation of cell communicationmitomycin C vs no drug (with dox) 0 1 2 3 4 5signal transduction in response to DNA damageresponse to hyperoxiaDNA damage checkpointstress-induced premature senescenceinflammatory cell apoptotic processregulation of interphase of mitotic cell cycleintracellular receptor mediated signaling pathwayd dox vs no dox (no drug)-log(P- value)-log(P-value)mitomycin C vs no drug (no dox) 0 1 2 3 4 5response to stressregulation of protein metabolic processpositive regulation of innate immune responseheart developmentresponse to DNA damage stimulusintrinsic apoptotic signaling pathwaytoll-like receptor signaling pathwaynegative regulation of cell cyclepositive regulation of histone H3-K4 methylationregulation of protein kinase activitypositive regulation of gene expression-2-1.5-1-0.500.51 1.5 2 2.5 cisplatin (μM)Difference change in absorption/no drug control0 rescue (%)61.76 53.53 51.25 45.84*0*******Figure 3 RHOXF2 overexpression in HEK293_M2 cells conferred resistance to a wide variety of DNA damaging agents. (a) In HEK293_M2cells, RHOXF2 overexpression rescued cisplatin and mitomycin C toxicity. Stable RHOXF2-expressing cells were cultured in the presence of anincreasing concentration of cisplatin or mitomycin C and their growth was compared to stable cells with the empty vector PB-TGcMV-Neo.The effect of RHOXF2 expression on cell viability was measured two days after drug exposure and compared to cells cultured in the absenceof drug as a 100% viability control. P < 0.05, TukeyHSD test: *significant difference between RHOXF2 vector with and without doxycycline(dox); **significant difference between RHOXF2 vector with doxycycline and empty vector with doxycycline. Error bars represent standard errorof the mean (n = 4). (b) Reduction in the number of γ-H2A.X foci by RHOXF2 overexpression in HEK293_M2 cells. Cells expressing or notRHOXF2 were treated with 50 nM of mitomycin C (approximately IC10) for 2 days and stained for the presence of γ-H2A.X foci. FollowingRHOXF2 overexpression, the number of γ-H2A.X foci per nucleus dropped from 20.71 ± 1.49 to 15.73 ± 0.92 (24% rescue). Nuclear DNA wasstained with DAPI. Quantification of γ-H2A.X foci in >250 cells from triplicate experiments. Scale bar: 10 μm. (c) In HEK293_M2 cells, RHOXF2overexpression conferred resistance to various DNA damaging agents. RHOXF2 drug suppression capacity was tested using DNA damagingagents with various modes of action as previously described for cisplatin and mitomycin C. (d) Gene Ontology analysis of upregulated genesin HEK293_M2 cells. Cells expressing or not RHOXF2 were treated with 40 nM of mitomycin C for 2 days. Based on their P-values, the mostenriched biological processes are shown.Arnoldo et al. Genome Medicine 2014, 6:32 Page 10 of 16http://genomemedicine.com/content/6/4/32specificity and then investigated which specific testis celltypes expressed RHOXF2 protein by immunohistochemis-try on normal testis samples. We detected a strong signalfor RHOXF2 in spermatogonia and primary spermatocytesbut not in spermatids (Figure 4b). Because RHOXF2 geneexpression was detected in early stage spermatogenic cellsand not in later stages, we hypothesized that RHOXF2might be involved in the self-renewal or maintenance ofthe undifferentiated state of the testis germ cells.Computational analysis predicts RHOXF2 to be a tes-ticular cancer candidate gene [36] and, in fact, the tran-script was detected in several types of human testicular1020304050607080901000.25 1 4 1601020304050607080901000.125 1 8 64012345678RHOXF2_1F+1RRHOXF2_2F+2RRHOXF2_4F+4Rab cd[cisplatin] (μM)Percentage survival*************[mitomycin C] (μM) ***********Percentage survival *e f740000780000820000860000740000780000820000860000RHOXF2 gene expression level / housekeeping genesRHOXF2_shRNAGFP_shRNARHOXF2_shRNAGFP_shRNAtubulinRHOXF2tubulinRHOXF20.016 0.0630Final cell numberFinal cell numberFigure 4 RHOXF2 is expressed in various cancer cell lines and modulates K562 resistance to DNA damaging agents. (a) In normalhuman tissues, RHOXF2 was exclusively detected in testis. Total RNA from 20 distinct tissues (mix of 3 different donors) was reverse transcribedinto first strand cDNA and used as template for quantitative PCR. Three distinct pairs of primers covering exons 1, 2 and 4 were used. RHOXF2expression level was calculated relative to the signal obtained for the amplification of housekeeping genes (CYCG, GUSB, ACTB, EEF1A1, GAPDH).(b) Representative example of immunohistochemical staining for RHOXF2 in normal testis. Section has been counterstained with Mayer’shematoxylin. (c) RHOXF2 was detected in various tumor cell lines by western blotting. (d) RHOXF2 shRNA validation in the human chronicmyelogenous leukemia cell line K562. WT, wild type. (e,f) Partial depletion of RHOXF2 increased K562 sensitivity to cisplatin and mitomycin C.Final numbers of viable cells were calculated in the absence of drug (inset). Student’s t-test; *P < 0.05; **P < 0.01; means ± standard error of themean for n = 4 experiments.Arnoldo et al. Genome Medicine 2014, 6:32 Page 11 of 16http://genomemedicine.com/content/6/4/32cancers [37]. Furthermore, in the present study, we showthat RHOXF2 overexpression conferred resistance to cis-platin (Figure 3a) and that RHOXF2 is found in severalother cancer cell lines, the non-small cell lung carcin-oma NCI-H1299, the blast phase chronic myelogenousleukemia K562, the multiple myeloma U266B1, the thy-roid gland medullary carcinoma TT (CRL-1803) and thechondrosarcoma SW1353 (Figure 4c; Additional file 11).In the chronic myelogenous leukemia K562 cell line,RHOXF2 depletion modulates the response toantineoplastic agentsOur study showed that RHOXF2 overexpression confersresistance to cisplatin and mitomycin C in HEK293_M2cells, in which the protein is not endogenously expressed(Figure 4c; Additional file 9). We first generalized thisobservation by positing that overexpression or ectopic ex-pression of certain genes greatly reduces some detrimentaldrug effects on the growth of particular cancer cell lines.We then reasoned that if this is truly drug-gene specific,the opposite mechanism should experimentally verifyit. More specifically, depletion of the endogenouslyexpressed protein should confer sensitivity to the cell lineduring drug treatment. The most direct translation of thisgene-drug specificity could then be exploited to under-stand drug resistance in cancer cell lines. Therefore, weasked if sensitivity to DNA damaging agents was observedwhen RHOXF2 is knocked down in cancer cells wherethe protein is normally expressed. We generated severalstable cell lines to deplete RHOXF2 (NCI-H1299, K562,SW1353, and the human pancreatic HPAC as negativecontrol) with five shRNAs that target the mRNA. TwoshRNAs against GFP were used as a negative control. TwoshRNAs showed a strong knock-down of the protein in allthree RHOXF2-positive cell lines (Figure 4d; Additionalfile 12e,h).K562 showed a significant increase in sensitivity forboth cisplatin (IC50: 3.231 μM to 2.052 μM) and mitomy-cin C (IC50: 377.8 nM to 250.4 nM) when the transcrip-tion factor was depleted, demonstrating the potentialrole of RHOXF2 in response to DNA damaging agentsand particularly in response to DNA crosslinking agents(Figure 4e,f). Independent K562 transduction resulted inidentical sensitivity to the drugs, showing the phenotype isindependent of the lentiviral integration site. We ruled outthe possibility that the calculated drug sensitivity simplyreflected a growth difference among cell lines by demon-strating that the RHOXF2_shRNA and GFP_shRNA con-trol cell lines divided at the same rate (Figure 4e,f, insets).Finally, we assessed the robustness of our knock-downs inthe presence of the drugs by confirming that the modula-tion of RHOXF2 protein level is maintained even whenthe K562 cells were exposed to cisplatin or mitomycin Cat IC70 (Additional file 13a,b).As expected, no change was measured for the RHOXF2-negative HPAC cells (Additional file 12a,b). Interestingly,no increase in sensitivity to cisplatin and mitomycin C wasobserved for NCI-H1299 and SW1353 knock-down celllines (Additional file 12c,d,f,g).The modest increase in sensitivity of K562 cells to DNAdamaging agents can be explained by different factors.First, transcription factors are generally present in limitedquantities in cells [38-40]. Here, although clearly depletedin the presence of the shRNAs, RHOXF2 is still presentat a relatively high level in K562 cells (Figure 4d). There-fore, we speculate that despite the efficient knock-down,the limited shift in IC50 is in part due to the presence ofenough protein to sustain a certain level of resistance dur-ing drug treatment. In order to test our first hypothesis,a complete knock-out of the transcription factor wouldbe necessary (for example, using CRISPR technology).Secondly, a recent study in mouse embryonic stem cellsreports that knock-down of a few transcription factorsis associated with substantial transcriptome change [41].These data not only indicate the robustness of the pluripo-tency gene network but also suggest that perhaps RHOXF2knock-down hardly perturbs the transcription factor net-work in the K562 cellular context.DiscussionWe have reduced to practice an elegant method to simul-taneously evaluate thousands of genes for their capacity toprovide a survival advantage in the presence of a toxicdose of drug using cells expressing the entire hORF collec-tion and a microarray or sequencing-based readout.The availability of hORF collections represents greatprogress compared to traditional approaches (randommutagenesis, cDNA libraries, and so on) because it per-mits the systematic interrogation of a gene's biologicalfunction in a targeted manner. Such gain-of-function ap-proaches would be bolstered when combined with thereciprocal loss-of-function approach [42]. To date, sev-eral studies have already explored the availability of suchcollections. Most of the studies tested the hORFs func-tion in an array (well-based) format [43-45] that simpli-fies the readout and analysis of the results. Here, wetested the function of thousands of ORFs simultaneouslyin a competitive manner. Our pool approach greatly re-duced the tedious and expensive upstream steps of clon-ing, transduction and testing of the genes. The primarychallenges to the pool approach are (i) potential loss ofORFs during cloning and infection (leading to represen-tation variability), (ii) the potential masking effect ofoverrepresented clones in a population during selection(for example, DHFR in the case of methotrexate) and(iii) the necessity for dedicated analysis of the final results(hits prioritization). In addition to future enhancements ofthe human ORFeome collection, other promising gain-of-Arnoldo et al. Genome Medicine 2014, 6:32 Page 12 of 16http://genomemedicine.com/content/6/4/32function approaches to assess drug mode of action arethe development of functional variomics technology[46] and the exploration of human isoform space in hu-man cell lines.In developing the assay, we first determined the opti-mal period of time between the beginning of the geneexpression and the addition of the chemical compound.To address that question, we quantified the number ofVenus-positive cells 12 hours, 24 hours and 48 hoursafter induction of expression of the hORF collection. Be-cause the human gene and the Venus marker are co-transcribed, quantifying the level of Venus expression byflow cytometry serves as a proxy for the level of humangene expression. Because the number of Venus-positivecells peaks at 24 hours after doxycycline addition(Additional file 5), we postulated that 18 hours post-induction would be sufficient time after which to adddrug to test the human gene for any protective effects.We then addressed the sensitivity of the technique byimproving the ratio between the number of seeded cells,the growth area and the time of drug addition. Seeding500,000 cells in a T175 flask 33 hours before drugaddition provided enough time and density for isolatedcells to form separate micro-colonies that should over-express only one or a few genes. This 'clonogenic assayapproach' was more sensitive to lower drug concentra-tions and provided a better selective environment forthe surviving cells, presumably by minimizing signalingfrom a large population of dying cells that would arise athigher plating densities. Similarly, adding fresh mediumand drug every 48 hours for the duration of the screenimproved the selection process by washing away thedead and dying adherent cells that could have interferedwith the subsequent screen readout.During the course of the screen, the cell populationfollowed a stereotypical pattern of growth, starting witha dramatic decrease in cell number and disappearance ofthe majority of cells, followed by a stabilization of thepopulation and concluding with the development ofmedium to large size colonies originating from the initialsurviving cells. Daily visual inspection of the cell popula-tion was crucial to determine when to stop the drug se-lection and harvest the cells, typically after 20 to 30doubling times.Several refinements may enhance this assay. First, im-proving the coverage and expression quality of the hu-man genes using a more complete collection of fullysequenced ORFs as such collections become available[14]. Moreover, as illustrated in Additional file 6, relativeabundance of the virally integrated hORFs is size-dependent. Therefore, to correct for that variation, prep-aration of the normalized amount of plasmid DNA forlentivirus production could be done as follows: i) reducethe pool size (less than 376 hORFs per pool as in ourstudy); ii) hORFs should be grouped by gene length. Onecan also match the particular drug under interrogationto disease-appropriate cell types and environmentalconditions. Whenever possible, experiments should beperformed with an initial high number of cells to guar-antee the full representation of the collection duringdrug target screening. Although DNA damaging resist-ance was observed in the RHOXF2-overexpressingHEK293_M2 cells, it would be informative to furtherstudy if that observation is dependent on RHOXF2 ex-pression status. Finally, next-generation sequencing willfacilitate the throughput of hit detection and reducethe number of false positives arising from microarraycross-hybridization.Our study revealed several genes whose functions re-late to the action of the drugs tested. RHOXF2 is of aparticular interest because of (i) its strong rescue pheno-type in HEK293_M2 cells, (ii) its capacity to suppressboth cisplatin and mitomycin C toxicity, and (iii) thedearth of functional information on this transcriptionfactor. Homeobox genes encode transcription factorsthat play a central role during embryogenesis. RHOXF2belongs to the Rhox family of genes, which are expressednot only during embryogenesis but also in adult repro-ductive tissue [47]. In normal tissue, RHOXF2 expres-sion is likely restricted to testis but its function remainslargely unknown [35,48]. Here, we demonstrated that (i)the presence of RHOXF2 was sufficient to lower bothcisplatin and mitomycin C toxicity in HEK293_M2 cells,and (ii) the transcription factor might exert its effect viathe activation of genes involved in response to DNAdamage. Furthermore, RHOXF2 protein expression wasdetected in several cancer cell lines. In one of these lines(chronic myelogenous leukemia/K562) knocking downthe protein increased the toxicity for both DNA dam-aging agents. Interestingly, although no growth disad-vantage was observed in K562 cells partially depletedfor RHOXF2 (Figure 4e,f ), its depletion via shRNA in-creased the sensitivity of this leukemia cell line to severalpharmacological compounds. We therefore speculatethat although not a driver of tumorigenesis, RHOXF2could confer on the neoplastic cells a selective advantageduring drug treatment.ConclusionsWe demonstrate the usefulness of our approach in orderto understand drug mechanisms of action by combininggenetic perturbation with drug treatment. Here we reca-pitulated examples of drug resistance and uncoveredseveral unanticipated drug-target interactions by virtueof the suppression of chemical toxicity observed upongene overexpression. We also provide a step-by-stepprotocol for performing such screens and for analyzingthe data (Additional file 1). Moreover, our innovativeArnoldo et al. Genome Medicine 2014, 6:32 Page 13 of 16http://genomemedicine.com/content/6/4/32overexpression approach could also be applied to identifygenes whose up-regulation could be toxic or enhance cellproliferation in different cell types. For example, the assaycould be useful to uncover tumor-specific activity in thecontext of genome engineering as well as for the discoveryof cell-type-specific oncogenes.Additional filesAdditional file 1: Step-by-step protocol for a genome-widemammalian overexpression drug screen.Additional file 2: Plasmid map of the lentiviral vector pLD-T-IRES-Venus-WPRE-stop.Additional file 3: Original immunoblotting showing the conditionalexpression of DHFR in HEK293_M2 cells. Addition of 4 ng/ml, 16 ng/ml,32 ng/ml, 63 ng/ml and 2,000 ng/ml of doxycycline induced DHFR proteinexpression in the stable cell line HEK293_M2. Expression level was comparedto alpha tubulin as a loading control. The asterisk marks a human proteinthat cross-reacts with the anti-DHFR antibody.Additional file 4: DHFR was identified as the predominant gene inthis pool of 376 hORFs whose overexpression provided resistanceto methotrexate. HEK293_M2 cells harboring a minipool collection of376 hORFs (including DHFR) were grown in the presence of a lethal doseof methotrexate. The nature of the hORFs conferring resistance to thedrug was identified by plotting the log2 of the signal intensity for eachhORF in the cells cultured in the presence of methotrexate on the x-axis;and by plotting the log2 ratio of the signal intensity for each hORF of thecells cultured in presence of the drug divided by the signal intensity foreach hORF of the cells grown in presence of DMSO on the y-axis.Additional file 5: Detection of Venus-expressing HEK293_M2 cellsafter 12, 24 and 48 hours of induction. (a,b) HEK293_M2 cells stablytransduced with the hORFeome collection (b) or not (a) were cultured inthe presence of doxycycline (2 μg/ml) for 12, 24 and 48 hours. After geneinduction, cell number and intensity of the Venus fluorescence weremeasured by flow cytometry. The percentage of Venus positive cells wascompared to the total number of living cells measured.Additional file 6: Human ORFeome relative abundance inHEK293_M2s at T0.Additional file 7: Plasmid map of the transposon-based vectorPB-TGcMV-Neo.Additional file 8: In HEK293_M2 cells, RHOXF2 overexpressionconferred resistance to various DNA damaging agents. Stable RHOXF2cells was cultured in the presence of an increasing concentration of DNAdamaging agents and their growth was compared to the stable cells withthe empty vector PB-TGcMV-Neo. The effect of RHOXF2 expression oncell viability was measured two days after drug exposure and compared tocells cultured in the absence of drug as a 100% viability control. TukeyHSDtest, P < 0.05; *significant difference between RHOXF2 vector with andwithout doxycycline; **significant difference between RHOXF2 vector withdoxycycline and empty vector with doxycycline. Error bars representstandard error of the mean (n = 4).Additional file 9: RHOXF2 raw FPKM values and Gene Ontologyenrichment.Additional file 10: RHOXF2 overexpression rescued cisplatin toxicityin HEK293_M2 and MCF7_M2 but not A549_M2 cells. Stable RHOXF2cells were cultured in the presence of an increasing concentration ofcisplatin and their growth was compared to the stable cells with the emptyvector PB-TGcMV-Neo. The effect of RHOXF2 expression on cell viabilitywas measured three days after drug exposure and compared to cellcultured in the absence of drug as a 100% viability control. Lines show thenonlinear fit of a variable-slope dose-response model for (a) MCF7_M2 cells,(b) A549_M2 cells and (c) HEK293_M2 cells as positive control; (d) resultingIC50s. Data analysis was performed using the 'drc' package in R. Experimentswere done in triplicate.Additional file 11: RHOXF2 was detected in various tumor cell linesby western blotting.Additional file 12: RHOXF2 knocked-down in HPAC, NCI-H1299 andSW1353 cells does not modulate cisplatin and mitomycin Csensitivities. (a-g) Dose-response curves for HPAC (a,b), NCI-H1299 (c,d)and SW1353 (f,g) were established after 2 days of exposure to cisplatin(a,c,f) or mitomycin C (b,d,g). Percentage of growth inhibition wascalculated by comparing the number of cells treated with drug to thenumber of cells cultured in media with DMSO as control. Error barsrepresent standard error of the mean (n = 4). RHOXF2 depletion inNCI-H1299 (e) and SW1353 cells (h). Total cell extracts for each cell linewere used to detect the presence of RHOXF2 by western blotting.Expression level was compared to the alpha tubulin as loading control.Additional file 13: RHOXF2 knocked down in K562 cells. (a,b) Celllines in which RHOXF2 and GFP were knocked down were grown in thepresence of 35 μM cisplatin (a) or 7 μM of mitomycin C (b) for 2 days.Total cell extracts for each cell line were used to detect the presence ofRHOXF2 by western blotting. Expression level was compared to alphatubulin as loading control. (c) Total cell extracts from a RHOXF2-expressingcell line and a cell line with GFP knocked down but overexpressing RHOXF2were used to detect the presence of RHOXF2 by western blotting.AbbreviationsBSA: bovine serum albumin; DHFR: dihydrofolate reductase;DMEM: Dulbecco's modified Eagle's medium; gDNA: genomic DNA;GGH: gamma-glutamyl hydrolase; hORF: human ORF; HRP: horse radishperoxidase; MGLL: monoglyceride lipase; ORF: open reading frame;PBS: phosphate-buffered saline; PCR: polymerase chain reaction; RNAi: RNAinterference; shRNA: short hairpin RNA; SRB: sulforhodamine B.Competing interestsThe authors declare that they have no competing interests.Authors’ contributionsAA, SK, GG and CN contributed to the conception and design of this project.AA and CN developed the methodology. AA and SK performed experiments.ABM constructed the pLD-T-IRES-Venus-WPRE-STOP vector. AIS performedquantitative PCR experiments. DT performed the Illumina RNA sequencing.AA and LEH analyzed and interpreted the data. AA, SK, LEH, AIS, JM, GG andCN wrote the manuscript. CN supervised the study. All authors read andapproved the final manuscript.AcknowledgementsCN and GG were supported by a grant from the Canadian Cancer Society(#20830) and the NHGRI.Author details1Department of Molecular Genetics, University of Toronto, Toronto, M5S 3E1,Canada. 2Banting and Best Department of Medical Research, University ofToronto, Toronto, M5S 3E1, Canada. 3Terrence Donnelly Centre for Cellularand Biomolecular Research, University of Toronto, 160 College Street,Toronto, Ontario M5S 3E1, Canada. 4Donnelly Sequencing Center, Universityof Toronto, 160 College Street, Toronto, Ontario M5S 3E1, Canada. 5Instituteof Biomaterials and Biomedical Engineering, University of Toronto, 170College Street, Toronto M5S 3E3, Canada. 6Department of PharmaceuticalSciences, University of Toronto, 144 College Street, Toronto, Ontario M5S3M2, Canada. 7Department of Pharmaceutical Sciences, University of BritishColumbia, 6619-2405 Wesbrook Mall, Vancouver, BC V6T 1Z3, Canada.Received: 3 September 2013 Accepted: 22 April 2014Published: 29 April 2014References1. Jarosz DF, Taipale M, Lindquist S: Protein homeostasis and the phenotypicmanifestation of genetic diversity: principles and mechanisms. Annu RevGenet 2010, 44:189–216.2. Nobeli I, Favia AD, Thornton JM: Protein promiscuity and its implicationsfor biotechnology. Nat Biotechnol 2009, 27:157–167.3. Hopkins AL: Network pharmacology: the next paradigm in drugdiscovery. Nat Chem Biol 2008, 4:682–690.Arnoldo et al. Genome Medicine 2014, 6:32 Page 14 of 16http://genomemedicine.com/content/6/4/324. Strebhardt K, Ullrich A: Paul Ehrlich’s magic bullet concept: 100 years ofprogress. Nat Rev Cancer 2008, 8:473–480.5. Agarwal R, Kaye SB: Ovarian cancer: strategies for overcoming resistanceto chemotherapy. Nat Rev Cancer 2003, 3:502–516.6. Seruga B, Ocana A, Tannock IF: Drug resistance in metastatic castration-resistantprostate cancer. Nat Rev Clin Oncol 2011, 8:12–23.7. Curt GA, Carney DN, Cowan KH, Jolivet J, Bailey BD, Drake JC, Chien SongKS, Minna JD, Chabner BA: Unstable methotrexate resistance in humansmall-cell carcinoma associated with double minute chromosomes.N Engl J Med 1983, 308:199–202.8. Cohen SN, Chang AC, Boyer HW, Helling RB: Construction of biologicallyfunctional bacterial plasmids in vitro. Proc Natl Acad Sci U S A 1973,70:3240–3244.9. Chang AC, Nunberg JH, Kaufman RJ, Erlich HA, Schimke RT, Cohen SN:Phenotypic expression in E. coli of a DNA sequence coding for mousedihydrofolate reductase. Nature 1978, 275:617–624.10. Rine J, Hansen W, Hardeman E, Davis RW: Targeted selection of recombinantclones through gene dosage effects. Proc Natl Acad Sci U S A 1983,80:6750–6754.11. Rual JF, Hirozane-Kishikawa T, Hao T, Bertin N, Li S, Dricot A, Li N, RosenbergJ, Lamesch P, Vidalain PO, Clingingsmith TR, Hartley JL, Esposito D, Cheo D,Moore T, Simmons B, Sequerra R, Bosak S, Doucette-Stamm L, Le Peuch C,Vandenhaute J, Cusick ME, Albala JS, Hill DE, Vidal M: Human ORFeomeversion 1.1: a platform for reverse proteomics. Genome Res 2004,14:2128–2135.12. Lamesch P, Li N, Milstein S, Fan C, Hao T, Szabo G, Hu Z, Venkatesan K,Bethel G, Martin P, Rogers J, Lawlor S, McLaren S, Dricot A, Borick H, CusickME, Vandenhaute J, Dunham I, Hill DE, Vidal M: hORFeome v3.1: a resourceof human open reading frames representing over 10,000 human genes.Genomics 2007, 89:307–315.13. Shin KJ, Wall EA, Zavzavadjian JR, Santat LA, Liu J, Hwang JI, Rebres R,Roach T, Seaman W, Simon MI, Fraser ID: A single lentiviral vectorplatform for microRNA-based conditional RNA interference andcoordinated transgene expression. Proc Natl Acad Sci U S A 2006,103:13759–13764.14. Yang X, Boehm JS, Salehi-Ashtiani K, Hao T, Shen Y, Lubonja R, Thomas SR,Alkan O, Bhimdi T, Green TM, Johannessen CM, Silver SJ, Nguyen C, MurrayRR, Hieronymus H, Balcha D, Fan C, Lin C, Ghamsari L, Vidal M, Hahn WC,Hill DE, Root DE: A public genome-scale lentiviral expression library ofhuman ORFs. Nat Methods 2011, 8:659–661.15. ArrayExpress. [http://www.ebi.ac.uk/arrayexpress/]16. Vichai V, Kirtikara K: Sulforhodamine B colorimetric assay for cytotoxicityscreening. Nat Protoc 2006, 1:1112–1116.17. Carpenter AE, Jones TR, Lamprecht MR, Clarke C, Kang IH, Friman O, GuertinDA, Chang JH, Lindquist RA, Moffat J, Golland P, Sabatini DM: Cell Profiler:image analysis software for identifying and quantifying cell phenotypes.Genome Biol 2006, 7:R100.18. Eden E, Navon R, Steinfeld I, Lipson D, Yakhini Z: GOrilla: a tool fordiscovery and visualization of enriched GO terms in ranked gene lists.BMC Bioinformatics 2009, 10:48.19. Abcam Immunohistochemistry Procedure. [http://docs.abcam.com/pdf/protocols/Short_IHC-P_protocol.pdf]20. Vander Heiden MG: Targeting cancer metabolism: a therapeutic windowopens. Nat Rev Drug Discov 2011, 10:671–684.21. Farber S, Diamond LK, Mercer RD, Sylvester RF, Wolff JA: Temporaryremissions in acute leukemia in children produced by folic acidantagonist, 4-aminopteroyl-glutamic acid (aminopterin). N Engl J Med1948, 238:787–793.22. Shi Y, Burn P: Lipid metabolic enzymes: emerging drug targets for thetreatment of obesity. Nat Rev Drug Discov 2004, 3:695–710.23. Hogan S, Fleury A, Hadvary P, Lengsfeld H, Meier MK, Triscari J, Sullivan AC:Studies on the antiobesity activity of tetrahydrolipstatin, a potent andselective inhibitor of pancreatic lipase. Int J Obes 1987, 11:35–42.24. Rose JL, Reeves KC, Likhotvorik RI, Hoyt DG: Base excision repair proteinsare required for integrin-mediated suppression of bleomycin-inducedDNA breakage in murine lung endothelial cells. J Pharmacol Exp Ther2007, 321:318–326.25. Kelland LR: Emerging drugs for ovarian cancer. Expert Opin Emerg Drugs2005, 10:413–424.26. Iyer VN, Szybalski W: A molecular mechanism of mitomycin action: linking ofcomplementary DNA strands. Proc Natl Acad Sci U S A 1963, 50:355–362.27. Egorin MJ, Lagattuta TF, Hamburger DR, Covey JM, White KD, Musser SM,Eiseman JL: Pharmacokinetics, tissue distribution, and metabolismof 17-(dimethylaminoethylamino)-17-demethoxygeldanamycin(NSC 707545) in CD2F1 mice and Fischer 344 rats. Cancer ChemotherPharmacol 2002, 49:7–19.28. Cole PD, Kamen BA, Gorlick R, Banerjee D, Smith AK, Magill E, Bertino JR:Effects of overexpression of gamma-Glutamyl hydrolase on methotrexatemetabolism and resistance. Cancer Res 2001, 61:4599–4604.29. Karlsson M, Reue K, Xia YR, Lusis AJ, Langin D, Tornqvist H, Holm C: Exon-intronorganization and chromosomal localization of the mouse monoglyceridelipase gene. Gene 2001, 272:11–18.30. Saario SM, Laitinen JT: Monoglyceride lipase as an enzyme hydrolyzing2-arachidonoylglycerol. Chem Biodivers 2007, 4:1903–1913.31. Chen J, Stubbe J: Bleomycins: towards better therapeutics. Nat Rev Cancer2005, 5:102–112.32. Shen CC, Wertelecki W, Driggers WJ, LeDoux SP, Wilson GL: Repair ofmitochondrial DNA damage induced by bleomycin in human cells.Mutat Res 1995, 337:19–23.33. Rossi MN, Carbone M, Mostocotto C, Mancone C, Tripodi M, Maione R, Amati P:Mitochondrial localization of PARP-1 requires interaction with mitofilin andis involved in the maintenance of mitochondrial DNA integrity. J Biol Chem2009, 284:31616–31624.34. Matsuoka S, Ballif BA, Smogorzewska A, McDonald ER 3rd, Hurov KE, Luo J,Bakalarski CE, Zhao Z, Solimini N, Lerenthal Y, Shiloh Y, Gygi SP, Elledge SJ:ATM and ATR substrate analysis reveals extensive protein networksresponsive to DNA damage. Science 2007, 316:1160–1166.35. Wayne CM, MacLean JA, Cornwall G, Wilkinson MF: Two novel human X-linkedhomeobox genes, hPEPP1 and hPEPP2, selectively expressed in the testis.Gene 2002, 301:1–11.36. Hofmann O, Caballero OL, Stevenson BJ, Chen YT, Cohen T, Chua R,Maher CA, Panji S, Schaefer U, Kruger A, Lehvaslaiho M, Carninci P,Hayashizaki Y, Jongeneel CV, Simpson AJ, Old LJ, Hide W: Genome-wideanalysis of cancer/testis gene expression. Proc Natl Acad Sci U S A 2008,105:20422–20427.37. Korkola JE, Houldsworth J, Chadalavada RS, Olshen AB, Dobrzynski D, Reuter VE,Bosl GJ, Chaganti RS: Down-regulation of stem cell genes, including those ina 200-kb gene cluster at 12p13.31, is associated with in vivo differentiationof human male germ cell tumors. Cancer Res 2006, 66:820–827.38. Vaquerizas JM, Kummerfeld SK, Teichmann SA, Luscombe NM: A census ofhuman transcription factors: function, expression and evolution. Nat RevGenet 2009, 10:252–263.39. Ishihama Y, Schmidt T, Rappsilber J, Mann M, Hartl FU, Kerner MJ, FrishmanD: Protein abundance profiling of the Escherichia coli cytosol. BMCGenomics 2008, 9:102.40. Yang VW: Eukaryotic transcription factors: identification, characterizationand functions. J Nutr 1998, 128:2045–2051.41. Nishiyama A, Sharov AA, Piao Y, Amano M, Amano T, Hoang HG, Binder BY,Tapnio R, Bassey U, Malinou JN, Correa-Cerro LS, Yu H, Xin L, Meyers E,Zalzman M, Nakatake Y, Stagg C, Sharova L, Qian Y, Dudekula D, Sheer S,Cadet JS, Hirata T, Yang HT, Goldberg I, Evans MK, Longo DL, Schlessinger D,Ko MS: Systematic repression of transcription factors reveals limitedpatterns of gene expression changes in ES cells. Sci Rep 2013, 3:1390.42. Hoon S, Smith AM, Wallace IM, Suresh S, Miranda M, Fung E, Fung E,Proctor M, Shokat KM, Zhang C, Davis RW, Giaever G, St Onge RP, Nislow C:An integrated platform of genomic assays reveals small-moleculebioactivities. Nat Chem Biol 2008, 4:498–506.43. Skalamera D, Dahmer M, Purdon AS, Wilson BM, Ranall MV, Blumenthal A,Gabrielli B, Gonda TJ: Generation of a genome scale lentiviral vectorlibrary for EF1alpha promoter-driven expression of human ORFs andidentification of human genes affecting viral titer. PLoS One 2012,7:e51733.44. Skalamera D, Ranall MV, Wilson BM, Leo P, Purdon AS, Hyde C, Nourbakhsh E,Grimmond SM, Barry SC, Gabrielli B, Gonda TJ: A high-throughput platformfor lentiviral overexpression screening of the human ORFeome. PLoS One2011, 6:e20057.45. Reece-Hoyes JS, Barutcu AR, McCord RP, Jeong JS, Jiang L, MacWilliams A,Yang X, Salehi-Ashtiani K, Hill DE, Blackshaw S, Zhu H, Dekker J, Walhout AJ:Yeast one-hybrid assays for gene-centered human gene regulatorynetwork mapping. Nat Methods 2011, 8:1050–1052.46. Huang Z, Chen K, Zhang J, Li Y, Wang H, Cui D, Tang J, Liu Y, Shi X, Li W,Liu D, Chen R, Sucgang RS, Pan X: A functional variomics tool forArnoldo et al. Genome Medicine 2014, 6:32 Page 15 of 16http://genomemedicine.com/content/6/4/32discovering drug-resistance genes and drug targets. Cell Rep 2013,3:577–585.47. Maclean JA 2nd, Chen MA, Wayne CM, Bruce SR, Rao M, Meistrich ML,Macleod C, Wilkinson MF: Rhox: a new homeobox gene cluster. Cell 2005,120:369–382.48. MacLean JA 2nd, Wilkinson MF: The Rhox genes. Reproduction 2010,140:195–213.doi:10.1186/gm549Cite this article as: Arnoldo et al.: A genome scale overexpressionscreen to reveal drug activity in human cells. Genome Medicine 2014 6:32.Submit your next manuscript to BioMed Centraland take full advantage of: • Convenient online submission• Thorough peer review• No space constraints or color figure charges• Immediate publication on acceptance• Inclusion in PubMed, CAS, Scopus and Google Scholar• Research which is freely available for redistributionSubmit your manuscript at www.biomedcentral.com/submitArnoldo et al. Genome Medicine 2014, 6:32 Page 16 of 16http://genomemedicine.com/content/6/4/32