Thermal Proteome Profiling Identifies Oxidative-Dependent Inhibition Of

Published OnlineFirst February 4, 2020; DOI: 10.1158/0008-5472.CAN-19-2069

CANCER RESEARCH | TRANSLATIONAL SCIENCE

Thermal Proteome Proﬁling Identiﬁes Oxidative-Dependent Inhibition of the Transcription of Major Oncogenes as a New Therapeutic Mechanism for Select Anticancer Compounds Sylvain Peuget1, Jiawei Zhu1, Gema Sanz1, Madhurendra Singh1, Massimiliano Gaetani2,3, Xinsong Chen4, Yao Shi1, Amir Ata Saei2, Torkild Visnes3,4,5, Mikael S. Lindstrom€ 2,3, Ali Rihani1, Lidia Moyano-Galceran1, Joseph W. Carlson4, Elisabet Hjerpe4, Ulrika Joneborg6, Kaisa Lehti1, Johan Hartman4, Thomas Helleday3,4,7, Roman Zubarev2,3, and Galina Selivanova1

ABSTRACT ◥ Identification of the molecular mechanism of action (MoA) of prerequisite for targeting the mRNA transcription machinery, bioactive compounds is a crucial step for drug development but thus conferring cancer selectivity to these compounds. Further- remains a challenging task despite recent advances in technology. more, DNA repair factors involved in homologous recombina- In this study, we applied multidimensional proteomics, sensitiv- tion were among the most prominently repressed proteins. In ity correlation analysis, and transcriptomics to identify a com- cancer patient samples, RITA, AF, and Onc-1 sensitized to poly mon MoA for the anticancer compounds RITA, aminoflavone (ADP-ribose) polymerase inhibitors both in vitro and ex vivo. (AF), and oncrasin-1 (Onc-1). Global thermal proteome profiling These findings might pave a way for new synthetic lethal com- revealed that the three compounds target mRNA processing bination therapies. and transcription, thereby attacking a cancer vulnerability, transcriptional addiction. This led to the preferential loss of expres- Significance: These findings highlight agents that target tran- sion of oncogenes involved in PDGF, EGFR, VEGF, insulin/IGF/ scriptional addiction in cancer cells and suggest combination MAPKK, FGF, Hedgehog, TGFb, and PI3K signaling pathways. treatments that target RNA processing and DNA repair pathways Increased reactive oxygen species level in cancer cells was a simultaneously as effective cancer therapies.

Introduction principle that the heat-induced unfolding of proteins changes upon binding to ligand. Using quantitative mass spectrometry–based pro- Comprehensive assessment of mechanism of action (MoA) and teomics, TPP measures ligand-induced thermal stabilization or desta- drug–protein interactions in living cells constitutes a major challenge bilization of proteins (2). The changes in melting proteome reflect for rationalizing a drug's therapeutic and adverse effects. Multiple direct drug–protein binding but also the changes in posttranslational strategies for target deconvolution have been developed, but many of modifications and protein–protein interactions. Therefore, the shift them include compound modification/labeling (1). Recently devel- in thermal stability of proteins could be used not only for the prediction oped thermal proteome profiling (TPP) enables such an assessment in of direct drug targets, but also for an unbiased identification of a physiologically relevant cellular environment without compound molecular MoA. modification and at a proteome-wide level. TPP is based on the Cancer cells are subject to multiple stresses, including genotoxic, proteotoxic, replicative, metabolic, and oxidative stresses. Tumor cells 1 Department of Microbiology, Tumor and Cell Biology, Karolinska Institutet, adapt to detrimental stressful conditions by dysregulated gene expres- Stockholm, Sweden. 2Department of Medical Biochemistry and Biophysics, sion. This creates a dependence on continuous active transcription of Karolinska Institutet, Stockholm, Sweden. 3Science for Life Laboratory, Stock- holm, Sweden. 4Department of Oncology-Pathology, Karolinska Institutet, oncogenes (3). Transcriptional addiction emerged as a potential Karolinska University Hospital, Stockholm, Sweden. 5Department of Biotech- “Achilles’ heel” of cancer cells that can be therapeutically exploited. nology and Nanomedicine, SINTEF Industry, Trondheim, Norway. 6Division of However, targeting transcription for selective elimination of cancer Obstetrics and Gynecology, Department of Women's and Children's Health, cells has been considered difficult due to potential toxicity for normal 7 Karolinska Institutet, Karolinska University Hospital, Stockholm, Sweden. Wes- tissues. The development of CDK7 and CDK12/13 inhibitors has ton Park Cancer Centre, Department of Oncology and Metabolism, University of challenged this paradigm: although acting on numerous genes, they Sheffield, Sheffield, United Kingdom. exert a highly specific effect on gene expression (3). These compounds Note: Supplementary data for this article are available at Cancer Research selectively affect cancer cells, leading to a preferential loss of oncogenic Online (http://cancerres.aacrjournals.org/). gene expression, along with DNA damage response genes (4). Attack- Corresponding Authors: Galina Selivanova, Karolinska Institute, Biomedicum ing addiction to a single oncogene is the basis of targeted therapies (5), € C8, Solnavagen 9, Stockholm SE-17165, Sweden. Phone: 46-8-52486302; Fax: but invariably leads to emergence of resistance clones due to cancer 46-8-342651; E-mail: [email protected]; and Sylvain Peuget, Phone: 46-8- ’ 52486273; E-mail: [email protected] cells plasticity. Therefore, targeting addiction to multiple oncogenes, either with a single compound or combinatorial treatments, is an Cancer Res 2020;80:1538–50 attractive anticancer strategy to limit acquired resistance. doi: 10.1158/0008-5472.CAN-19-2069 In this study, we applied multidimensional approaches, including 2020 American Association for Cancer Research. TPP, to identify the MoA and pathways affected by selective anticancer

AACRJournals.org | 1538

TPP Reveals the Mechanism of Selective Anticancer Compounds

compounds RITA, aminoflavone AF, and oncrasin-1 (Onc-1) and ciated drug screens were used to calculate the drug sensitivity Pearson investigated how these new findings might provide opportunities for correlation coefficient between RITA, AF, and Onc-1, as well as the novel combination therapies. correlation between RITA, DNA-damaging agents, and topoisomerase-1 inhibitors. The 50% growth inhibitory concentration (GI50) values of drugs were used for calculations. Materials and Methods Cell lines TPP MCF7, T47D, A375, U2OS, OVCAR-3, 184A1, MCF10A, and BJ- MCF7 cells were treated with 1 mmol/L RITA, 1 mmol/L AF, or hTERT were obtained from ATCC. BJ-hTert HRasV12ER-Tam cells 5 mmol/LOnc-1for2hours(RITAandAF)or6hours(Onc-1)in were a kind gift from Reuven Agami (Netherlands Cancer Institute, biological duplicates. TPP was performed as described previous- Amsterdam, The Netherland). U2OS cMycER-Tam cells were obtained ly (7). Briefly, cells from each biological replicate were split equally from Martin Eilers (University of Wurzburg,€ Wurzburg,€ Germany). into 10 parts; each aliquot was heated at a different temperature CRISPR/Cas9-mediated p53 deletion was performed in MCF7 and point ranging from 37Cto67C. Soluble proteins were then A375 cells using sgRNA-targeting TP53 exon 4. Cell lines were analyzed by quantitative LC/MS-MS. For each protein, the melting Mycoplasma T T ≥ authenticated by ATCC and contamination was tested temperature ( m) was calculated and proteins with |D m| 0.5 C monthly using MycoAlert Mycoplasma Detection Kit (Lonza) accord- and P < 0.05 were considered as hits for subsequent analyses. ing to the manufacturer's instructions. All experiments were per- Sample processing and data analysis are detailed in Supplementary formed within 20 passages from frozen stocks. Further details can be Materials and Methods, and the complete list of TPP hits for each found in Supplementary Materials and Methods and Supplementary compound treatment in MCF7 cells can be found in Supplementary Table S1. Table S3. Gene ontology (GO) statistical overrepresentation analysis was Clinical samples performed using PANTHER (Protein ANalysis THrough Evolu- Experiments on primary cancer cells from patients with breast tionary Relationships 13.1 (http://pantherdb.org; ref. 8) against all cancer and ascites fluid from patients with metastatic high-grade human genes. PANTHER GO-Slim biological processes enriched in serous ovarian cancer (HGSOC) were approved by the Regional TPP hits with Fisher adjusted P < 0.01 were considered as affected Ethical Board of Stockholm (Dnr 2016/957-31, 2017/742-32, and by each treatment. The complete list of GO-Slim biological pro- 2016/1197-31/1) in compliance with the declaration of Helsinki. cesses enriched in TPP hits can be found in Supplementary Samples were taken from patients who signed written informed Table S4. Functional protein interaction network analysis was consent. Detailed information about isolation procedure and treat- performed using STRING 10.5 (http://string-db.org; ref. 9). The ments can be found in Supplementary Materials and Methods. STRING network was built with all proteins affected by at least one treatment. Only interactions with the highest score (>0.900) from Compounds and treatments experiments or curated databases were considered. The three main RITA (NSC652287) and aminoflavone (NSC686288) were obtained clusters of the network were identified by k-means clustering, and from the NCI, while oncrasin-1 was from Santa Cruz Biotechnology. each cluster was analyzed for GO statistical overrepresentation Unless stated otherwise, cells were treated in complete medium with 1 using PANTHER to identify its biological function. mmol/L RITA, 1 mmol/L AF, or 5 mmol/L Onc-1. a-amanitin, ActD, BMH21, camptothecin, cycloheximide, irinotecan, MG132, nordihy- Microarrays droguaiaretic acid, Nutlin-3, and resveratrol were purchased from Expression profiling of MCF7 cells treated with 1 mmol/L RITA, Sigma-Aldrich; cisplatin, doxorubicin, olaparib, and talazoparib from 10 mmol/L Nutlin (8 hours), or with 1 mmol/L AF (8 hours), 5 mmol/L Selleckchem; and THZ531 from ApexBio Technology. For combina- Onc-1 (16 hours) was performed using GeneChip HG-U219 or HT tion experiment with PARP inhibitors, cells were pretreated 2 days HG-U133þ PM 24-Array Plate, respectively. The datasets generated in with olaparib or talazoparib and then treated with the combination for this study are available in Gene Expression Omnibus database (https:// the indicated time. www.ncbi.nlm.nih.gov/geo/) under accession number GSE86880 and GSE128329, and in Supplementary Table S5. Data analysis is detailed Antibodies in Supplementary Materials and Methods. All antibodies used in the study are detailed in Supplementary Materials and Methods. Cell viability assays Cell viability was assessed by resazurin assay. Briefly, cells were qPCR treated for 24–72 hours and incubated for 2 hours with 5 mmol/L Total RNA extraction and cDNA synthesis were performed using resazurin (Sigma-Aldrich) before fluorescence measurement on a Aurum total RNA and iScript cDNA Synthesis Kits (Bio-Rad) accord- VICTOR X5 plate reader (PerkinElmer). For crystal violet staining ing to the manufacturer's instructions. mRNA quantification was experiments, cells were treated for 1–5 days, fixed with ethanol, and performed by qRT-PCR using Sso Advanced Universal SYBRGreen stained with 0.2% crystal violet. Quantification was performed using SuperMix (Bio-Rad). GAPDH and RPL13A were used as housekeeping ImageJ software. genes. Error bars represent SD from mean of at least three independent experiments. The sequences of qPCR primers used are detailed in Reactive oxygen species measurement Supplementary Table S2. Cells were treated as indicated for 8 hours, washed, and incubated for 30 minutes with 10 mmol/L DCFDA in serum-free medium. Correlation sensitivity analysis Then cells were trypsinized, washed twice with PBS, and fluores- NCI-60 analysis tools provided by CellMiner (https://discover.nci. cence was analyzed by a FACSCalibur Flow Cytometer (BD nih.gov/cellminer/; ref. 6) and the NCI-60 cell line panel with asso- Biosciences).

AACRJournals.org Cancer Res; 80(7) April 1, 2020 1539

Peuget et al.

5-Ethynyluridine labeling occurred only upon RITA treatment. These results were confirmed at Cells on glass coverslips were treated for the indicated time, and protein level (Supplementary Fig. S2C). Furthermore, comparison of 1 mmol/L 5-ethynyluridine (EU) was added to the medium 1 hour genome-wide expression upon RITA, Nutlin-3, camptothecin, cisplat- before cell fixation. EU was labeled using the Click-iT RNA Alexa Fluor in, and doxorubicin revealed a very distinct transcriptional profile 594 Imaging Kit (Thermo Fisher Scientific) according to the manu- upon RITA treatment (Supplementary Fig. S2D). Taken together, facturer's instructions. our analysis of proteomics, transcriptomics, and cancer cell sensitivity profiles upon RITA revealed a response different from that induced Homologous recombination reporter assay by DNA-damaging compounds, ROS inducers, or topoisomerase Homologous recombination (HR) reporter assay was performed in inhibitors. U2OS cells harboring stably integrated reporter construct for HR (DR- Notably, sensitivity correlation analysis of NCI-60 database GFP) as described previously (10). To generate a DSB, U2OS DR-GFP revealed a significant correlation between the sensitivity of cancer cells were transfected with I-SceI expression vector pCMV3xnls-I-SceI cells to RITA and the two anticancer compounds AF and Onc-1 (R2 ¼ (1 mg) using Fugene HD (Promega). Twenty-four hours after trans- 0.5938 and R2 ¼ 0.6466, respectively; Fig. 1B and C). fection, cells were treated with RITA, AF, or Onc-1 for 16 hours and AF [5-amino-2-(4-amino-3-fluorophenyl)-6,8-difluoro-7-methylchromen- GFP fluorescence due to HR of I-SceI cleavage site was monitored by 4-one; NSC 686288] displays an antiproliferative and proapoptotic flow cytometry. activity in tumor cells (15). It induces replication-dependent DNA lesions and S-phase–dependent apoptosis (16), although its MoA has Statistical analysis not been fully elucidated. Onc-1 (1-[(4-chlorophenyl)methyl]-1H- Unless otherwise stated, statistical significance was calculated using indole-3-carboxaldehyde) and its derivative NSC-743380, identified two-tailed Student t test. in a screen for selective targeting of mutant K-Ras cancer cells (17), exhibit antitumor activity in several tumors types (18). Onc-1 MoA remains unclear, but involves interference with RNA polymerase II Results (RNA Pol II) transcription and STAT3 inhibition (19, 20). Similarity of phenotypes induced in cancer cells by RITA, AF, Cancer selectivity is crucial for the success of anticancer com- and Onc-1 pounds. Therefore, we compared the effect of these compounds on RITA (reactivation of p53 and induction of tumor cells apoptosis) the growth of normal and cancer cells (Fig. 1D). RITA, AF, and was identified by us in a screen for molecules that selectively kill cells Onc-1 efficiently reduced cell viability in breast cancer cells MCF7 expressing wild-type (wt) p53. RITA is a potent anticancer agent and T47D, whereas they had a limited effect on normal mammary in vitro and in vivo and is highly selective for cancer, but not normal epithelial cells 184A1 and MCF10A and nontransformed BJ-hTert cells (11, 12). Using several pairs of isogenic cancer cells with wtp53/ fibroblasts. knockout (KO) p53, we found that the contribution of p53 to cell death Furthermore, we found that the treatment of MCF7 cells with AF or upon RITA is cell type–dependent (Supplementary Fig. S1A–S1J). Onc-1 phenocopied RITA effects on apoptosis and DNA damage While in melanoma A375 cells, RITA effects were completely p53- response, although the kinetics was slightly different for Onc-1 dependent (Supplementary Fig. S1I and S1J), in MCF7 cells the (Fig. 1E; Supplementary Fig. S3A). The three compounds activated contribution of p53 to the effects of RITA was quite weak, although the JNK pathway (Fig. 1E), in-line with previous studies (19, 21, 22). detectable (Supplementary Fig. S1A–S1H), which is in-line with We found previously that RITA inhibits the antioxidant enzyme recently published data (13). thioredoxin reductase 1 (TrxR1; ref. 23), increasing intracellular ROS To address the p53-independent MoA of RITA, we took advan- level. Similarly, AF and Onc-1 induced a strong oxidative stress in tage of ProTargetMiner database, containing proteome profiles for different cell lines (Fig. 1F; Supplementary Fig. S3B). 56 anticancer drugs in A549 lung cancer cells (14). Because protein Because growth suppression by RITA depends on the down- abundances depend not only on mRNA levels, but also on trans- regulation of several crucial oncogenes (12, 22), we tested whether lation rate and protein stability, proteome response can be highly AF and Onc-1 have similar effects. Indeed, MDM2, PPM1D, MCL1, specific to drug MoA. We compared the proteome signature upon and MDMX were inhibited on mRNA and protein level upon RITA treatment to 20 other anticancer compounds. Unsupervised treatment with all three compounds, along with the induction of clustering revealed the similarity of RITA and antimetabolic drugs proapoptotic PUMA and NOXA (Fig. 2A and B; Supplementary raltitrexed and methotrexate profiles (Fig. 1A). However, RITA Fig. S3C). Interestingly, while p21 mRNA was induced, its protein profile did not cluster with DNA-damaging compounds such as level was downregulated by AF and Onc-1, similar to the effects we bleomycin or doxorubicin, suggesting that its MoA is different from have shown previously for RITA (24). Finally, as RITA, AF, and that of DNA-damaging drugs. Onc-1 all stabilize p53 in MCF7 cells (Fig. 1E), we performed Next, we searched the NCI-60 cancer cell lines pharmacology viability assays in p53 KO MCF7 and showed that apoptosis caused database for compounds eliciting a sensitivity profile similar to RITA. by all three compounds was largely p53-independent in these cells Consistent with our chemical proteomics analysis, RITA sensitivity (Supplementary Fig. S3D). profile did not correlate with those of DNA-damaging drugs (such as Furthermore, we observed a remarkably similar expression profile cisplatin and oxaliplatin), topoisomerase inhibitors (camptothecin, upon RITA, AF, and Onc-1, and identified 3,154 common differen- irinotecan, and topotecan), or reactive oxygen species (ROS)-inducing tially expressed genes (DEG) in MCF7 cells (Fig. 2C and D). Strikingly, compounds (doxorubicin; Supplementary Fig. S2A). We also com- oncogenes were overrepresented among common repressed genes pared transcriptional changes induced by RITA treatment with those (Fig. 2E). Analysis of gene expression data revealed significant down- upon cisplatin, doxorubicin, camptothecin, irinotecan, or Nutlin-3 in regulation of oncogenes representing major oncogenic pathways, MCF7 cells (Supplementary Fig. S2B). While the response of p53- such as PDGF, EGFR, VEGF, insulin/IGF/MAPKK, FGF, Hedgehog, activated genes PUMA (BBC3), NOXA (PMAIP1), and p21 (CDKN1A) TGF-b, and PI3K signaling pathways (Fig. 2F and G; Supplementary was similar, the repression of oncogenes MDM2, PPM1D, and MCL1 Fig. S3E).

1540 Cancer Res; 80(7) April 1, 2020 CANCER RESEARCH

TPP Reveals the Mechanism of Selective Anticancer Compounds

A B RITA S O OH RITA OH S TOP1 inhibitors NH O TOP2 inhinbitors F 2 Tubulin inhibitors O Oncrasin-1 Inducon of DSB F MDM2 inhibitors N Anmetabolites F Cl Aminoﬂavone NH2 O 1 0.5 C 0 R² = 0.5938 R² = 0.6466 3 ) 3 ) −0.5 50 −1 50 2 2 1 1 0 0 RITA 5-FU −1 −1 Nutlin Sensivity to AF (GI Carmofur Paclitaxel Genistein Docetaxel Etoposide idarubicin Epirubicin −2 Ralrexed

Sensivity−2 to Onc-1 (GI Irinotecan Bleomycin Vincrisne Topotecan Oxaliplan Teniposide Floxuridine

Doxorubicin −2 −1 210 3 −1−2 3210 Methotrexate Camptothecin Sensivity to RITA (GI50) Sensivity to RITA (GI50) D 1.2 1 0.8 MCF7 T47D 0.6 184A1 0.4 MCF10A

Relave viability BJ-hTert 0.2 0 −4 −3 −2 −1 0 1 −4 −3 −2 −1 0 1 −4 −3 −2 −1 0 1 Log [RITA] (μmol/L) Log [AF] (μmol/L) Log [Onc-1] (μmol/L) E RITA AF Onc-1 0816 24 08 16240816 24 h F p53 120 DMSO PARP RITA Cleav. PARP 80 AF γH2AX Onc-1 40 p-JNK Cells (count)

JNK 0 100 101 102 103 p-c-JUN DCFDA β-Acn

Figure 1. AF and Onc-1 phenocopy the effects of RITA in cancer cells. A, Unsupervised clustering of proteome proﬁles induced in lung carcinoma A549 cells by a panel of 20

anticancer compounds, including RITA. B, Chemical structures of RITA, AF, and Onc-1. C, Correlation of sensitivity (GI50) between RITA and AF (left) and between RITA and Onc-1 (right) in the NCI-60 cell lines panel. D, Cell viability of cancer cells (red) or normal immortalized cells (gray) upon treatment by RITA, AF, or Onc-1. E, PARP cleavage (indicating apoptosis induction), gH2AX (induction of DNA damage response), and phospho-JNK and phospho-c-JUN protein levels in MCF7 cells treated with RITA, AF, or Onc-1. F, Induction of ROS by RITA, AF, or Onc-1 in MCF7 measured by DCFDA.

TPP identifies RNA processing as the common cell process actions observed upon treatment as a reflection of the key affected affected by RITA, AF, and Onc-1 biological processes. Because RITA, AF, and Onc-1 induced a highly similar cell To minimize the indirect events, we analyzed proteome changes response, we hypothesized that they could have a common molecular at early time points (2 hours for RITA and AF, and 6 hours for Onc- target, and/or target a common pathway. To identify pathways affected 1). We identified 144, 182, and 188 hits, defined as proteins with a fi T ≥ by the three compounds, we performed TPP in MCF7 cells. Identifying signi cant thermal shift |D m| 0.5 C upon treatment with RITA, a direct drug target with TPP is challenging, as many variations in Onc-1, and AF, respectively (Fig. 3A and B), with a prominent protein interactions measured in the cells could be secondary, overlap among them (Fig. 3B; Supplementary Fig. S4A). We indirect events. Instead, the proteome-wide variations in protein identified 90 hits common for two treatments and eight hits interactions provide important clues as to which pathways are common for all three treatments. We performed a GO analysis affected by a drug. Therefore, we considered the changes of inter- using PANTHER to identify biological processes affected (Fig. 3C).

AACRJournals.org Cancer Res; 80(7) April 1, 2020 1541

Peuget et al.

A RITA AF Onc-1 B BBC3 PMAIP1 CDKN2A MDM2 PPM1D MCL1 MDM4 0 8 16 24 0816 24 0816 24 h 5 p21 4 MDM2 *FC) Wip1 2 3 MCL1 2 MDMX 1 β-Acn 0 D C 0.810 DEGs −1 RITA

0.846 3,555 Relave expression level (log −2 AF Onc-1 RITA RITA AF Onc-1 −3 1,647 639 3,154 E 333 16% 1.92E-15 Fracon of oncogenes AF 744 1,136 1.34E-9 4.01E-5 among repressed genes Onc-1 12% Fracon of oncogenes Down Up 8% in the whole-gene set RITA RITA 1,391 4% 2,940 567 0% 376 1,641 RITA AF ONC-1 346 733 1,014 228 976 548 371 AF 343 859 G AF Onc-1 Onc-1 AF Onc-1 RITA F -log (P) 0 132 4 ACVR2A MAP3K5 PDGF signaling pathway IRS2 FC common diﬀerenally expressed genes (3,154) GLI3 2 PIK3R1 GnRH receptor pathway SKI

Log Insulin/IGF pathway-MAPKK/MAPK cascade IRS1 VAV3 FGF signaling pathway RRAS2 ACVR1 p53 pathway EP300 IGF1R Hedgehog signaling pathway UBR5 MAP3K1 Insulin/IGF pathway-AKT signaling cascade ARHGAP5 RASA1 p53 pathway feedback loops 2 SMURF2 CBLB TGF-β signaling pathway SMURF1 FC common FC common diﬀerenally

2 PIK3C2A High Low EGF receptor signaling pathway ELF4 Hypoxia response via HIF acvaon expressed oncogenes (357) PIK3CB

Log JAK2 PI3K pathway FGFR2 RP56KA3 T-cell acvaon P PIK3C2B -log ( ) PIK3CA CCKR signaling map Percentage GNAQ Angiogenesis PDGF pathway VEGF signaling pathway Insulin/IGF pathway High Low Angiogenesis/EGFR/FGF 0204010 30 Hedgehog pathway TGF-β pathway Percentage PI3K/AKT pathway

Figure 2. RITA, AF, and Onc-1 repress the transcription of oncogenes. A, Downregulation of selected oncogenes upon treatment by RITA, AF, or Onc-1 detected by Western blotting. B, mRNA level of p53 target genes (PUMA, NOXA,andp21) and RITA-repressed oncogenes after treatment with RITA, AF, or Onc-1 measured by qPCR. C, Heatmap of common DEGs (P ≤ 0.05) in MCF7 cells upon treatment with RITA, AF, or Onc-1, as assessed by microarray analysis. D, Overlap between DEGs, up-, and downregulated genes upon the three compounds. E, Enrichment of oncogenes among repressed genes upon RITA, AF, and Onc-1 in MCF7 cells. Signiﬁcance was calculated by using Fisher exact test. F, GO analysis of oncogenic pathways downregulated upon treatments with RITA, AF, and Onc-1. Shown are signiﬁcantly

affected pathways (Padj < 0.01). G, Heatmap of common differentially expressed oncogenes (P ≤ 0.05) upon treatment with the three compounds.

Notably, most of the GO biological processes terms enriched in TPP As TPP detects the changes of interactions for a given protein, other hits were related to RNA processing, including splicing and RNA proteins interacting with a TPP hit are likely to be also affected. To have metabolism. The overlap between the GO biological processes terms a more integrated view of the changes in protein complexes involved, was striking, with RNA processing being the common process we generated a protein–protein interaction network using STRING to affected by all three treatments (Fig. 3D). Mitochondrial translation visualize the functional interactions affected by RITA, AF, and Onc-1 was also signiﬁcantly affected upon AF treatment, but not upon (Fig. 3E; Supplementary Fig. S4B). Using unbiased k-means clustering, RITA and Onc-1. we identiﬁed three main clusters of interactions in the network

1542 Cancer Res; 80(7) April 1, 2020 CANCER RESEARCH

TPP Reveals the Mechanism of Selective Anticancer Compounds

A RITA AF Onc-1 4.3 4.3 4.3

3.3 3.3 3.3 P P P 10 10 10 2.3 2.3 2.3 -Log -Log -Log 1.3 1.3 1.3

0.3 0.3 0.3 −10 −8 −6 −4 −2 0246810 −10 −8 −6 −4 −2 0246810 −10 −8 −6 −4 −2 0246810 ΔT (°C) ΔT (°C) ΔT (°C) B m m -Log (P) m Proteins C 0 246 RITA RNA splicing, via transesterificaon reacons 98 mRNA splicing, via spliceosome mRNA processing 19 19 8 tRNA aminoacylaon for protein translaon mRNA processing Onc-1 AF 44 tRNA metabolic process RNA metabolic process 117 111 mRNA splicing, via spliceosome Cellular aminoacid metabolic process Glycogen metabolic process D mRNA 3’-end processing RITA Biological Pteridine-containingcompound metabolicprocess -Log (P) 2 RNA splicing, via transesterificaon reacons processes 1 RITA AF 2 Mitochondrial translaon Onc-1 AF 7 mRNA processing Onc-1 tRNA metabolic process Percentage RNA metabolic process PFAS 01020515 E ATP6V1G1 NUBP1 Percentage ETFA ETFB PDHA1 NUF2 NARFL BCKDHA NUBP2 MRPL53 PDHB ATP6V1B2 PRKCI MRPL46 EIF4H NUMB MRPS5 STRING proteins TANK UBLCP1 MRPS34 ATP6V1E1 STK38 FAM96B CIAO1 PPP1R12A MRPS21 interacon network PDK3 CLNS1A MRPL45 MRPL10 EIF3G RIPK1 GALE PPP2R1B MRPL27 MRPS26 TGS1 PPP2R5C UGDH K-means clustering CASP2 LAMTOR1 MRPS27 MRPL28 GEMIN5 SLBP TUFM CTU1 RBPJ MRPS31 PPP1CB PSMD3 MRPS18B Cluster A CTU2 PSMD8 PSMD11 MRPS30 MRPL22 CASP8 TRIM21 PAN2 SSU72 PSMC2 CDC27 Cluster B TIMM44 PSMC4 KLHL22 MRPS6 EIF4E TSC1 MRPL18 NUP62 MRPS15 HSPA9 PSMD5 PSMD14 RBL2 MRPL44 Cluster C HSCB NXF1 CUL7 CCND1 KEAP1 NT5C3 RAB9A ARHGAP5 UMPS STX10 CPSF1 GSK3A CDK2 GMPS EPS15L1 EIF4A3 PTBP1 GSK3B PKN1 CMPK1 NCOA3 SMAD2 GUK1 MAGOH SF3B1 ARID2 EPS15 CCDC97 DHX38 SMARCB1 SNX4 PKN2 RBM8A PCBP2 SAR1A RRM1 EPN1 SF3B2 PHKG2 SNRPC SRGAP2 PFN2 SUGP1 POLR2K CCNT1 EIF5A PACSIN1 SF3B4 DOHH REPS1 DYNC1LI2 PAPOLA POLR2L DYNC1H1 RBM22 RDBP HAUS4 PTPN6 ELAVL1 HNRNPM POLE4 RAD50 ARRB2 POLR2A BLM CEP250 SYK EYA3 IQGAP1 KIF3A KIF5B CSTF1 SF1 NABP2 BCAR1 NBN PRKAR1A MAPKAPK2 PRKCD TIAL1 CRCP RFC2 RFC5 TRIP6 IQGAP2 KIF13B DYNLL1 HAUS6 RING1 CSTF2T TIPIN TUBA1C RFC4 BAIAP2 ALDH7A1 RABGEF1 ARF4 NCK1 CHMP4C URI1 POLR3G DTL RABEP2 CHMP5 GNB2 PAK4 IFT81 RABL5 NUDT2 CCT5 GIT1 HADH DCAF11 CHTF18 AKTIP CHMP7 HIBCH EPHB4 -Log (P) HOOK1 UBAP1 HADHA 0510 20 30 40 0 HADHB RNA splicing, via transesterificaonreacons TRMT10C G mRNA splicing, via spliceosome RNA splicing HSD17B10 mRNA processing mRNA metabolic process F RITA AF Onc-1 150 Mitochondrial translaonal elongaon Mitochondrial translaon Translaonal elongaon 100 Mitochondrial translaonal terminaon Mitochondrial gene expression -Log (P) Protein modificaon by small protein removal Cluster A 50 Protein deubiquinaon Cluster B Post-translaonal protein modificaon Cluster C 0 Mitoc DNA damage checkpoint Percentage % Proteins in the cluster% Proteins in the Cluster A BC Mitoc DNA integrity checkpoint 0251510 20 5 Percentage

Figure 3. TPP identiﬁes RNA processing as the common target of RITA, AF, and Onc-1. A, Volcano plots show all thermal shifts detected by TPP upon treatment. TPP hits (red) were deﬁned as proteins with thermal stability shift |DTm| > 0.5 C between the treatment and the control in both replicates, with P < 0.05. Proteins involved in RNA processing are indicated, and common hits upon at least two treatments are highlighted. B, Overlap of TPP hits upon treatment by RITA, AF, and Onc-1. C, GO

biological process terms significantly enriched (Padj < 00.1) in TPP hits upon treatment with RITA, AF, and Onc-1. The percentage indicates the number of hits relative to the GO group size. D, Overlap of GO biological process terms significantly enriched in TPP hits upon treatment with RITA, AF, and Onc-1. E, STRING network analysis of the combination of TPP hits from the three compounds. Each ball represents a protein with significantly altered thermal stability upon treatment with at least one of the compounds. Only proteins with connections in the network are displayed. The three main clusters of protein interactions in the network identified by unbiased k- means are indicated in different colors. F, Relative proportion of TPP hits in A, B, and C clusters affected by RITA, AF, or Onc-1. G, Top five GO biological process terms significantly enriched in TPP hits in each cluster. The percentage indicates the number of hits relative to the GO group size.

AACRJournals.org Cancer Res; 80(7) April 1, 2020 1543

Peuget et al.

(clusters A, B, and C), containing 34, 20, and 16 proteins, respectively. described upon RNA Pol II or splicing inhibition (Fig. 4F; Supple- The proportion of proteins in the clusters affected by each treatment is mentary Fig. S5A; ref. 25). shown in Fig. 3F. To identify the biological function of each cluster, we Not only RNA Pol II but also Pol I was inhibited by the compounds, performed a GO analysis (Fig. 3G). Cluster A represents proteins as evidenced by shuttling of the RNA helicase DDX21 from the involved in RNA processing, including multiple splicing factors and nucleoli to the nucleoplasm (Fig. 4G), which is known to occur upon auxiliary splicing regulatory proteins (MAGOH, RBM8A, DHX38, RNA Pol I inhibition (26). Consistently, protein level of POLR1A, the SF1, SF3B1, SF3B2, SF3B4, RBM22, SUGP1, PTPB1, HNRNPM, and main subunit of RNA Pol I, was decreased upon treatment to the same SNRPC), RNA Pol II subunits (POLR2A, POLR2L, and POLR2K), extent as upon specific Pol I inhibitor, BMH21 (Fig. 4H). Finally, we transcriptional elongation regulators (RDBP), and polyadenylation compared the inhibition of both Pol II and Pol I by testing the factors (CPSF1, CSTF1, and PAPOLA). Importantly, this cluster was expression of RNA Pol II- and RNA Pol I–regulated RNAs, MYC affected evenly by the three treatments. Cluster C has a less defined mRNA, and 45S preribosomal RNA, respectively, upon a broad range biological function, including proteasome subunits, cell-cycle regula- of compounds’ concentrations in MCF7 cells (Fig. 4I). Expression of tors (CCND1 and CDK2), and DNA damage checkpoints factors both genes was substantially decreased by the three compounds, (BLM and RAD50). indicating the block of both RNA polymerases. To exclude the effect Cluster B contains mostly mitochondrial ribosomal proteins affect- of p53-mediated MYC repression (27), we used MCF7 p53KO cells and ed by AF treatment, although some (such as MRPL10) were also got a similar result (Supplementary Fig. S5H). RNA Pol I inhibition affected by RITA and Onc-1. Using mitochondrial membrane per- (i.e., upon low doses of ActD) and nucleolar stress activate p53 through meabilization assay, we found that only AF caused a small, but the release of MDM2-binding protein, RPL11 (28), therefore we tested significant change in the mitochondrial membrane potential (Supple- whether RITA had the same effect (Supplementary Fig. S5I). SiRNA- mentary Fig. S4C). We also assessed mitochondrial morphology by mediated RPL11 knockdown rescued p53 stabilization by 5 nmol/L electron microscopy but did not observe changes upon treatment ActD but not upon RITA, demonstrating that nucleolar stress is not (Supplementary Fig. S4D). involved in p53 activation upon RITA. In summary, TPP data assessed by both GO and network analysis FACS analysis of synchronized cells using bromodeoxyuridine/ approaches revealed RNA processing as a common pathway affected propidium iodide staining showed that RITA-induced cell death by the three compounds. occurs mainly in early S-phase (Supplementary Fig. S6A), suggesting replication stress, in-line with previous study (29). Thus, we addressed RITA, AF, and Onc-1 inhibit transcription in cancer cells, but not the question whether RNA Pol II inhibition was a consequence of in normal cells replication stalling. Using cells synchronized in early G1-phase by To investigate the effect of RITA, AF, and Onc-1 on RNA proces- lovastatin and treated with RITA before they entered S-phase, we sing, we visualized the incorporation of EU. As shown in Fig. 4A and found that the inhibition of transcription occurred in G1-phase Supplementary Fig. S5A, RITA, AF, and Onc-1 completely abolished and was therefore independent of replication stress. Furthermore, RNA synthesis, similar to known RNA Pol II inhibitors actinomycin we demonstrated that early S-phase arrest and replication stress D (ActD) and a-amanitin. Notably, the levels of total, phospho-S5, (as manifested by gH2AX induction) induced by double thymidine and phospho-S2 (elongating) RNA Pol II strongly decreased upon block were not sufficient to inhibit transcription (Supplementary treatment in different cancer lines (Fig. 4B; Supplementary Fig. S5B Fig. S6B and S6C). and S5C). RNA Pol II level was rescued by the proteasome inhibitor To investigate the mechanism of RNA polymerase inhibition, we MG132, suggesting proteasomal degradation after stalling (Fig. 4C). tested whether RITA, AF, and Onc-1 can intercalate DNA by perform- Accordingly, cycloheximide chase experiment showed a clear decrease ing a thiazole orange displacement assay. No DNA intercalation was in RNA Pol II half-life upon RITA treatment (Fig. 4D). Then, we observed at doses 12–60 times higher than those used in cells (Sup- investigated whether the inhibition of RNA synthesis was dependent plementary Fig. S7A), demonstrating that these compounds are not on p53 activation by RITA, AF, and Onc-1. RNA synthesis was also DNA intercalating agents. Interestingly, gene set enrichment analysis abolished in p53 KO MCF7 cells (Supplementary Fig. S5D). of our microarray data revealed an enrichment of UV response genes Moreover, comparison of expression profiles upon ActD and among the DEGs induced by all three compounds (Supplementary CDK12/CDK13 inhibitor THZ531 revealed similarities with the tran- Fig. S7B). This correlates with the UV-like (30), pan-nuclear gH2AX scriptional response to RITA, AF, and Onc-1, even though different staining, different from the discrete nuclear foci associated with DNA cancer cell lines were used (Supplementary Fig. S5E and S5F). We breaks (Supplementary Fig. S7C). To investigate the possibility that Pol confirmed that the oncogenes involved in RITA-induced cell death II stalling might be caused by DNA adducts, we performed ASCC2 foci were also downregulated at protein level by THZ531 and ActD formation analysis in MCF7 cells, a marker for cell response to DNA treatment (Supplementary Fig. S5G). alkylation (31). Although we indeed observed the formation of ASCC2 Strikingly, RITA, AF, and Onc-1 did not prevent EU incorporation foci (Supplementary Fig. S7D and S7E), they were induced to a lesser and RNA synthesis in normal human fibroblasts (Fig. 4E). In contrast, extent than foci formed upon low dose of alkylating agent, methyl ActD and a-amanitin completely shut down transcription in non- methanesulfonate. transformed cells as well. Thus, the inhibition of transcription by Altogether, our data confirm the results obtained via TPP analysis RITA, AF, and Onc-1 occurs selectively in cancer cells, contrary to and demonstrate that RITA, AF, and Onc-1 inhibit RNA Pol I- and Pol other RNA Pol II inhibitors. II–mediated transcription and interfere with RNA processing such as Several splicing/RNA-binding factors were among the top common splicing. TPP hits, including CIRBP and RBM8A for all three compounds and SUGP1, SF3B2, and GRSF1 for any two of them. Therefore, we RITA, AF, and Onc-1 cancer-specific effects are dependent on reasoned that RITA, AF, and Onc-1 might affect RNA splicing. Indeed, oxidative stress all three compounds induced the formation of enlarged sc35 speckle As discussed previously, RITA, AF, and Onc-1 induce a strong domains (splicing factor compartments, SFC), which were previously oxidative stress (Fig. 1F). This was associated with oxidative DNA

1544 Cancer Res; 80(7) April 1, 2020 CANCER RESEARCH

TPP Reveals the Mechanism of Selective Anticancer Compounds

A DMSO 8 h RITA 8 h AF 8 h Onc-1 24 h DMSO 24 h ActD 24 h α-amanin 24 h

DAPI

B RITA AF Onc-1 C DMSO RITA AF Onc-1 0816 24 0816240816 24 h MG132 −+−+ −+−+ RNA Pol II RNA Pol II p-RNA Pol II (S2) β-Acn β-Acn 1 D 0.5 DMSO RITA

Relave 0 CHX 0246 8 10 0 2 4168 0h

protein level RNA Pol II (total) p-RNA Pol II (S2) RNA Pol II β-Acn BJ-hTert 1 E 0.8 DMSO RITA AF Onc-1 ActD α-amanin 0.6 0.4 DMSO EU 0.2 RITA 0 Relave RNA Pol II 0 2 468 10 Time of CHX (h) DAPI

G H F DMSO 8 h RITA 8 h AF 8 h Onc-1 24 h DMSO 24 h DMSO RITA AF POLR1A β-Acn EU DDX21 1 0.8 * 0.6 * * DAPI * sc35 0.4 0.2

Relave POLR1A 0 DAPI

I 45S pre-rRNA 1.5 45S pre-rRNA 1.5 45S pre-rRNA 1.5 MYC MYC mRNA MYC mRNA mRNA RPL13A mRNA RPL13A mRNA RPL13A mRNA

1.0 1.0 1.0

0.5 0.5 0.5 Relave RNA level Relave

0.0 0.0 0.0 −5 −4 −3 −2 −1 0 1 −5 −4 −3 −2 −1 0 1 −5 −4 −3 −2 −1 0 1 Log ([RITA]) Log ([RITA]) Log ([RITA])

Figure 4. RITA, AF, and Onc-1 inhibit transcription by RNA Pol I and RNA Pol II. A, RNA synthesis monitored by EU incorporation in MCF7 cells treated with RITA, AF, Onc-1, ActD (2.5 mmol/L), or a-amanitin (10 mmol/L), then incubated for 1 hour with 1 mmol/L EU. EU was labeled by click chemistry and visualized by fluorescent microscopy. B, Total and phospho-Ser2 RNA Pol II CTD protein levels in MCF7 cells treated with RITA, AF, or Onc-1. Densitometric quantification of the bands normalized to b-actin (bottom). C, RNA Pol II CTD protein level in MCF7 cells pretreated with 20 mmol/L MG132 for 2 hours and then treated with RITA, AF, or Onc-1 for 24 hours. D, Cycloheximide (CHX) chase analysis of RNA Pol II degradation. RNA Pol II level in cycloheximide-treated (10 mmol/L) MCF7 cells upon RITA treatment (top). Densitometric quantification of the bands normalized to b-actin (bottom). E, RNA synthesis of immortalized BJ-hTert fibroblasts treated with RITA, AF, Onc-1, ActD (2.5 mmol/L), or a-amanitin (10 mmol/L) for 24 hours analyzed as in A. F, Modifications of SFCs detected by sc35 immunostaining in MCF7 cells treated with RITA, AF, or Onc-1. G, DDX21 subcellular localization detected by immunofluorescence. H, POLR1A level in MCF7 upon 24-hour treatment with BMH21 (0.5 mmol/L), RITA, AF, or Onc-1. Densitometric quantification of POLR1A bands normalized to b-actin from two independent experiments (, P < 0.05; bottom). I, 45S pre-rRNA or MYC mRNA detected by qPCR in MCF7 cells upon RITA, AF, or Onc-1 (8 hours) treatment. RNA level (DCt) was relative to untreated control and RPL13A was used as housekeeping gene.

AACRJournals.org Cancer Res; 80(7) April 1, 2020 1545

Downloaded from cancerres.aacrjournals.org on September 23, 2021. © 2020 American Association for Cancer Research. 1546 Downloaded from acrRs 07 pi ,2020 1, April 80(7) Res; Cancer pn4dy fRT,A,o n- ramn nteasneo rsneo ciae Ras. activated of presence or absence the in treatment Onc-1 or AF, HRasV12 RITA, of days 4 upon egte al. et Peuget ihRT,A,o n- o 4husi rsneo evrto rnriyrgaaei acid. nordihydroguaiaretic or resveratrol of presence in hours 24 for resveratrol. Onc-1 of or AF, presence RITA, in with hours 24 for Onc-1 or AF, RITA, (1 resveratrol antioxidants of presence in hours 24 for Onc-1 dependent. or stress AF, oxidative RITA, with are effects Onc-1 and AF, RITA, 5. Figure ihRT,A,o n- o 4husi rsneo evrto rnriyrgaaei acid. nordihydroguaiaretic or resveratrol of presence in hours 24 for Onc-1 or AF, RITA, with omlzdto normalized C7clsuo IA F rOc1i rsneo resveratrol. of presence in Onc-1 or AF, RITA, upon cells MCF7 Cleav. PARP ER-Tam β-Acn γ PARP p-RNA PolII (S2) Relave expression level (log *FC) H2AX A

2 E −3 −2 −1 p53 0 1 2 3 4 5 F b el pn4dy fRT,A,o n- ramn nteasneo rsneo ciae a.Dnioercquanti Densitometric Ras. activated of presence or absence the in treatment Onc-1 or AF, RITA, of days 4 upon cells atn rmtoidpneteprmns(bottom). experiments independent two from -actin, RNA Pol II RNA Pol Relave viability β-Acn 0.2 0.4 0.6 0.8 1.2 C cancerres.aacrjournals.org B3PAP DNAMM P1 C1MDM4 MCL1 PPM1D MDM2 CDKN2A PMAIP1 BBC3 Onc-1 AF RITA 0 1 Published OnlineFirstFebruary4,2020;DOI:10.1158/0008-5472.CAN-19-2069 4− 2− 1 0 −1 −2 −3 −4 Log [compound] (μmol/L) Log [compound] JhetHRasV12 BJ-hTert Resveratrol AF +resveratrol Onc-1 +resveratrol Onc-1 RITA +resveratrol ER-Tam ER-Tam Resveratrol D, on September 23, 2021. © 2020American Association for Cancer Research. rti eeso eetdRT-oneuae noee n rsria atr nMF el treated cells MCF7 in factors prosurvival and oncogenes RITA-downregulated selected of levels Protein A, F, nuto faotss(APcevg)adDAdmg epne( response damage DNA and cleavage) (PARP apoptosis of Induction elvaiiyo omlimmortalized normal of viability Cell NDGA HRas HRas Onc-1 AF HRas RITA , P < V12 V12 V12 m 0.05; o/)o odhdouirtcai (10 acid nordihydroguaiaretic or mol/L) + Onc-1 + AF + RITA β-Acn MDM2 MDMX , B D MCL1 P Wip1 p21 < E, C, Relave cell viability 0.01. 1.2 0.2 0.4 0.6 0.8 RAlvlo 5 agtgnsadRT-oneuae noee in oncogenes RITA-downregulated and genes target p53 of level mRNA N o ICDttladpopoSr eesi C7clstetdwith treated cells MCF7 in levels phospho-Ser2 and total CTD II Pol RNA G, 0 1 oa n hsh-e2RAPlI T rti eesi BJ-hTert in levels protein CTD II Pol RNA phospho-Ser2 and Total p-RNA Pol II (S2) p-RNA Pol ﬁ rbat ihidcbeRssse B-Tr HRasV12 (BJ-hTert system Ras inducible with broblasts RNA Pol II RNA Pol G * Relave protein level β-Acn 0.2 0.4 0.6 0.8 1.2 0 1 * * m mol/L). N o Ip-RNAPolII (S2) II RNA Pol Resveratrol BJ-hTert Resveratrol B, elvaiiyo C7clstreated cells MCF7 of viability Cell g 2X nMF el treated cells MCF7 in H2AX) ﬁ aino N o Ibands II Pol RNA of cation Jhet+HRas BJ-htert ACRRESEARCH CANCER * * * NDGA * * NDGA V12 * * * ER-Tam ) Published OnlineFirst February 4, 2020; DOI: 10.1158/0008-5472.CAN-19-2069

TPP Reveals the Mechanism of Selective Anticancer Compounds

damage as assessed by immunostaining with 8-oxo-7,8-dihydro-20- clinical trials). Remarkably, we observed synthetic lethality with both deoxyguanosine (8-oxodG) antibody (Supplementary Fig. S8A). inhibitors in MCF7 (Fig. 6E and F; Supplementary Fig. S9F) and Oxidative stress–dependent growth suppression has been previous- OVCAR-3 cells (Supplementary Fig. S9G). We tested the combination ly described for all three compounds (22, 32, 33) and we confirmed ex vivo in primary ascites–derived cells from patients with HGSOC and in our cell models that two antioxidants, resveratrol and nordihy- observed similar effects (Fig. 6G), demonstrating the therapeutic droguaiaretic acid, rescue different cancer lines from apoptosis and potential of a combination of RITA, AF, or Onc-1 with PARP growth suppression induced by RITA, AF, and Onc-1 (Fig. 5A inhibition both in vitro and ex vivo. Furthermore, we observed and B; Supplementary Fig. S8B and S8C). Notably, antioxidants also synthetic lethality of RITA combination with PARP inhibitors in prevented RNA Pol II stalling and degradation in a p53-indepen- samples derived from patients with breast cancer in 3D conditions dent way (Fig. 5C; Supplementary Fig. S8D), as well as inhibition of ex vivo. Notably, we found that the beneficial effect of this combination prosurvival genes (Fig. 5D and E; Supplementary Fig. S8E). was evident in patient samples with high Ki67 index and tumor grade, As cancer cells have increased ROS levels and therefore they are that is, highly malignant tumors (Fig. 6H). Similarly, synthetic lethal- more sensitive to further oxidative stress (34), we hypothesized that ity was significantly enhanced upon activation of oncogene in cMyc- induction of ROS could be the reason for cancer selectivity of our inducible U2OS cells (U2OS cMycER-Tam; Supplementary Fig. S9H). compounds. To test this idea, we used a model of increased ROS upon These findings might provide opportunities for novel therapeutic HRasV12ER-TAM induction in BJ-hTert fibroblasts. We found that interventions in cancer. RITA, AF, and Onc-1 suppressed the growth of fibroblasts upon HRasV12ER-TAM activation by tamoxifen (Fig. 5F) followed by the increase of the basal ROS level (Supplementary Fig. S8F). Moreover, we Discussion observed a significant downregulation of RNA Pol II CTD phosphor- Identification of small molecules’ MoA has always been a chal- ylation only upon HRasV12 induction (Fig. 5G). lenge in drug development. It became particularly important with Taken together, our results suggest that the cancer selectivity of the rise of personalized medicine and targeted therapies. TPP has transcription inhibition and growth suppression by the three com- been developed as a novel approach for the identification of small pounds is dependent on oncogene-induced oxidative stress. molecules’ targets. However, finding the correct target among the TPP hits is challenging given the number of proteins with thermal RITA, AF, and Onc-1 inhibit homologous recombination– stability shifts due to secondary events. TPP is also limited to mediated DNA repair soluble proteins; additional restrictions inherent to quantitative It has been suggested that DNA repair factors are among the most proteomics may lead to a number of false negatives. For example, effected genes upon inhibition of transcription elongation (4). Indeed, p53 and TrxR1, two previously described direct binding targets of upon treatments the DNA repair factors CtIP, RPA32, WRAP53, RITA (11, 23), were not found in our TPP screen. However, our RNF168, RNF8, and RAD51 (Supplementary Fig. S9A) were down- study demonstrates how comprehensive analysis of the global regulated, the HR-associated protein RAD51 and its partners RNF8 modulation of protein interactions at the whole-proteome level by and RNF168 being the most prominently decreased (Fig. 6A; Sup- TPP can help to decipher therapeutic MoA. plementary Fig. S9B). The observed decrease of these proteins was Our TPP analysis shows that mRNA maturation machinery is associated with transcriptional repression (Fig. 6B), but not with significantly affected by RITA, AF, and Onc-1 treatment. Applying protein stability, as measured by cycloheximide chase experiment GO biological processes terms and a more integrated view of protein– (Supplementary Fig. S9C). protein interaction network revealed a striking overlap between the We confirmed that HR was impaired upon treatment by RITA, three treatments, with RNA processing being the common process. AF, and Onc-1 using U2OS cells harboring stably integrated Validation of these findings confirmed the inhibition of RNA Pol I- reporter construct for HR (DR-GFP), where repair of I-SceI– and Pol II–mediated transcription selectively in cancer cells. This induced break by HR leads to a functional GFP gene. As shown transcription blockade leads to the loss of expression of dozens of vital in Fig. 6C, treatment with RITA, AF, or Onc-1 leads to a stable oncogenes, involved in major oncogenic pathways, such as PDGF, decrease of GFP-positive cells, indicating impairment of the HR EGFR, VEGF, insulin/IGF/MAPKK, FGF, Hedgehog, TGFb, and PI3K repair system. signaling pathways. Overexpression of RAD51, RNF8, and RNF168 did not rescue Targeting transcription for anticancer therapy has been consid- RNA Pol II inhibition or downregulation of prosurvival genes ered to be too toxic to normal cells. However, recent studies have (Fig. 6D; Supplementary Fig. S9D), but we observed a partial challenged this view by showing that the expression of oncogenes rescue of gH2AX when overexpressing the three repair factors required for the survival of cancer cells is highly dependent on simultaneously, indicating that inhibition of RAD51 and its partners uninterrupted active transcription. For instance, blocking RNA Pol plays a role in the replication stress observed upon RITA (Fig. 6D). II function affects more c-Myc–transformed cells than their normal However, the rescue of gH2AX was not sufficient to increase cell counterparts (35). This “transcriptional addiction” of cancer cells viability (Supplementary Fig. S9E), highlighting the importance of the creates a therapeutic window allowing selective anticancer growth downregulation of prosurvival factors for apoptosis induction. suppression (3). Our results are in-line with these ideas. We show that inhibition of transcription in cancer cells by RITA, AF, and RITA, AF, and Onc-1 sensitize to PARP inhibition Onc-1 leads to the preferential loss of a crucial set of oncogenes and As DNA repair, and more specifically HR, is impaired because of the therefore target the “Achilles’ heel” of cancer cells, exerting a highly transcription inhibition of HR factors by RITA, AF, and Onc-1, we selective effect on cancer, but not normal cells. hypothesized that these compounds could synergize with PARP How to achieve selectivity for tumor cells without affecting normal inhibition to efficiently kill cancer cells. Therefore, we used sublethal tissues is one of the main challenges for anticancer drug discovery. One doses of RITA, AF, and Onc-1 in combination with two PARP of the cancer cells vulnerabilities is a high basal level of ROS, leading to inhibitors, olaparib (FDA approved) and talazoparib (currently in the dependence on high expression of their antioxidant systems.

AACRJournals.org Cancer Res; 80(7) April 1, 2020 1547

Downloaded from cancerres.aacrjournals.org on September 23, 2021. © 2020 American Association for Cancer Research. 1548 Downloaded from acrRs 07 pi ,2020 1, April 80(7) Res; Cancer 1 ascites primary of ( olaparib inhibitor PARP of presence egte al. et Peuget eesi C7tasetyoeepesn A5-F,RF-A n N18GPtetdwt IAfr1 or.Dnioercquanti Densitometric hours. 16 for RITA with treated RNF168-GFP and RNF8-HA, RAD51-GFP, to normalized overexpressing transiently MCF7 in levels hours. 16 inhibition. treatment. PARP 1 to sensitize and HR inhibit Onc-1 and AF, RITA, 6. Figure IAi rsneo lprbo aaoai (n talazoparib or olaparib of presence in RITA h estvt otecmiaino IAwt APihbtr.N,nonsigni NS, inhibitors. PARP with RITA of combination the to sensitivity the m o/ ptet1 r0.1 or 1) (patient mol/L D A RNF168 β-Acn RAD51 RAPlI S2 II Pol pRNA RAD51 RNF168 C, RNF8 RNF168-GFP Rad51-GFP Ref HR N o II Pol RNA B, RNF8- Olaparib Olaparib G b endo. endo. β-Acn -GFP -GFP γ PRfrmN ee fDArpi factors repair DNA of level mRNA for qPCR atnfo w needn xeiet ( experiments independent two from -actin Relave cell viability Relave cell viability Relave H2AX RNF8 8 0 ﬁ – 0.2 0.4 0.6 0.8 1.4 1.2 0.2 0.4 0.6 p53 0.8 1.2 1.4

γ HA inyuo ramn ihRT,A,o n- oioe sn Rrpre 2S(RGP cells. (DR-GFP) U2OS reporter HR using monitored Onc-1 or AF, RITA, with treatment upon ciency

eie el rmptet ihHSC rae ihRT (0.05 RITA with treated HGSOC, with patients from cells derived H2AX level 1 0 0 1 0 2 4 6 RITA cancerres.aacrjournals.org +– –+ +– –+ Paent HGSOC1 aetHGSOC2 Paent –+ –+ –+ m 624 16 o/ ptet2; (patient mol/L Published OnlineFirstFebruary4,2020;DOI:10.1158/0008-5472.CAN-19-2069 IAA Onc-1 AF RITA IAA Onc-1 AF RITA –+ –+ *** + – * DMSO + + – – – 08 ++ ++ E –– –– rtlzprb( talazoparib or ) *** * AF 624 16 n ¼ ¼ –+ –+ –+ 3; 3; , + – – P , H P 08 < < RITA F 0.05; +– –+ * 0.05 μmol/L 0.05; ﬁ quanti Relative staining. violet crystal by assessed as ), 0.1 μmol/L Relave cell viability

Relave cell viability RITA RITA Onc-1 0.6 0.2 0.4 0.8 1.2 –– , 0.2 0.6 0.4 0.8 1.2 on September 23, 2021. © 2020American Association for Cancer Research. N8 RNF168 RNF8, 624 16 0 1 P 0 1 A, < , – aetBRC5 Paent P .5 otm.Snhtcltaiyi C7clstetdwt ulta oe fRT,A,o n- in Onc-1 or AF, RITA, of doses sublethal with treated cells MCF7 in lethality Synthetic bottom). 0.05; , etr lt o N earfcosRF,RF6,adRD1i C7clsuo IA F rOnc- or AF, RITA, upon cells MCF7 in RAD51 and RNF168, RNF8, factors repair DNA for blots Western – aetBRC1 Paent P i7=39% = Ki67 < NS +–+ i7=98% = Ki67 NS < 0.01). ++ 10 μmol/L h 1 μmol/L .1.Darm(otm ih)sostecreainbtenteK6 ne/uo rd and grade index/tumor Ki67 the between correlation the shows right) (bottom, Diagram 0.01). NS Olap. Olap. NS – * NS E

H, Olaparib (μmol/L) ,and B 10 NS 5 0 0.1 μmol/L elvaiiyo ain-eie rmr ratcne el rae ihsbehlds of dose sublethal with treated cells cancer breast primary patient-derived of viability Cell – Talaz. 0.1 μmol/L Relave expression level (log2*FC) RITA * ﬁ DMSO –0.5 –1.5 + 0.6 0.2 0.4 0.8 1.2 cant. 0.5 RAD51 1.11 1.10 1.00 –1 0 1 0 m aetBRC2 Paent +–+ –+ . μmol/L 0.5 A5 N8RNF168 RNF8 RAD51 o/) F(1 AF mol/L), 0.05 μmol/L i7=75% = Ki67 RITA 0.6 0.2 0.4 0.8 1.2 NS fe ramn fMF el ihRT rA o or rwt n- for Onc-1 with or hours 8 for AF or RITA with cells MCF7 of treatment after RITA 0.71 0.87 1.06 0 1 1 μmol/L Olap. NS aetBRC6 Paent +–+ –+ i7=4% = Ki67 * NS 0.5 μmol/L 1 μmol/L Olap. .61 0. NS 0.73 0.95 AF 0.1 μmol/L RITA m 0.6 0.2 0.4 0.8 1.2 NS o/) rOc1(1 Onc-1 or mol/L), 0 1 0.1 μmol/L NS Talaz. –+ 2.5 μmol/L aetBRC3 Paent +–+ –+ Onc-1 NS 0.71 0.91 1.05 i7=65% = Ki67 NS Olap. NS 1 μmol/L Onc1 AF RITA aini niae eo h pictures. the below indicated is cation

Relave cell viability * obnno IAadPARP and RITA of Combinaon 0.5 0.8 0.7 0.9 0.6 0.05 μmol/L C D, 1 niio tsbehldose sublethal at inhibitor

Talazoparib (μmol/L) F m RITA 0 0.1 o/)fr4 or npeec folaparib of presence in hours 48 for mol/L) 0.6 0.2 0.4 0.8 1.2 oa n hsh-e2N o Iprotein II Pol phospho-Ser2RNA and Total Olaparib 0 1

0 1 % Homologous recombinaon 20 uo i7(%) Ki67 Tumor 100 aetBRC4 Paent DMSO 20 40 60 80 +–+ –+ 0.89 0.96 1.00 0 i7=45% = Ki67 NS 1 μmol/L Olap. NS 0.05 μmol/L uo grade Tumor 123 Talazoparib * * * RITA 0.35 0.60 0.96 * 806040 R² NS P =0.0197 – 10 nmol/L Talaz. Pearson 100 ﬁ =0.5639 ACRRESEARCH CANCER * * + ainof cation 0.5 μmol/L 0.44 0.68 1.00 AF * G, 2.5 μmol/ g Onc-1 2Xbands H2AX 0.55 0.65 1.03 elviability Cell L Published OnlineFirst February 4, 2020; DOI: 10.1158/0008-5472.CAN-19-2069

TPP Reveals the Mechanism of Selective Anticancer Compounds

Therefore, compounds increasing the oxidative stress may induce Disclosure of Potential Conflicts of Interest cancer cell death while sparing normal cells, which have lower ROS T. Helleday is a SAB (paid consultant) at Sierra Oncology, has ownership interest levels and therefore can cope with ROS increase. Indeed, it has been (including patents) in PARP inhibitor sales (royalty), Shareholder Oxcia AB, and has shown that normal cells keep a low level of ROS upon exposure to an unpaid consultant/advisory board relationship with Chair Helleday Foundation. No potential conflicts of interest were disclosed by the other authors. RITA (23). RITA, AF, and Onc-1 all increase oxidative stress in cells and different antioxidants can prevent their cytotoxic effect (22, 32, 33). Authors’ Contributions Moreover, head and neck cancer cells with high level of antioxidant Conception and design: S. Peuget, R. Zubarev, G. Selivanova systems are more resistant to RITA (36). Here, we show that high level Development of methodology: M. Singh, L. Moyano-Galceran, K. Lehti, R. Zubarev, of intracellular ROS is important not only for apoptosis induced by G. Selivanova RITA, AF, and Onc-1 but also for their activity as RNA Pol inhibitors. Acquisition of data (provided animals, acquired and managed patients, provided We detected oxidative DNA damage manifested as 8-oxodG upon facilities, etc.): S. Peuget, J. Zhu, M. Singh, M. Gaetani, X. Chen, A.A. Saei, T. Visnes, € RITA, AF, and Onc-1, in-line with the report about AF (32). Impor- M.S. Lindstrom, L. Moyano-Galceran, J.W. Carlson, E. Hjerpe, U. Joneborg, K. Lehti, tantly, a widespread transcriptional silencing is induced upon base J. Hartman, R. Zubarev Analysis and interpretation of data (e.g., statistical analysis, biostatistics, excision repair (BER) of oxidative DNA damage (37). Taken together, computational analysis): S. Peuget, J. Zhu, G. Sanz, M. Singh, M. Gaetani, these data suggest that the inhibition of transcription by the three X. Chen, Y. Shi, A.A. Saei, A. Rihani, E. Hjerpe, G. Selivanova compounds could be caused by BER of oxidized DNA bases. In Writing, review, and/or revision of the manuscript: S. Peuget, G. Sanz, M. Singh, addition, we cannot exclude the possibility that the three compounds M. Gaetani, T. Visnes, E. Hjerpe, U. Joneborg, J. Hartman, T. Helleday, G. Selivanova may cause nucleotide imbalance, for instance, through oxidation of Administrative, technical, or material support (i.e., reporting or organizing data, guanine, which might lead to a disruption of RNA synthesis. This constructing databases): S. Peuget, X. Chen, R. Zubarev, G. Selivanova Study supervision: S. Peuget, T. Helleday, G. Selivanova hypothesis will be investigated in the future. Overall, our data suggest that RITA, AF, and Onc-1 high Acknowledgments selectivity for cancer cells can be explained by their simultaneous We thank all our colleagues for sharing precious reagents and material with targeting of multiple cancer vulnerabilities, including oncogene us,especiallyProf.J.Bartek(DanishCancer Society Research Center and addiction, increased ROS, and replication stress (as summarized Karolinska Institute, Stockholm, Sweden) and Dr. M. Farnebo (Karolinska in Supplementary Fig. S10). Our study paves a way to rational Institute, Stockholm, Sweden). This study was supported by the Swedish Cancer treatment designs with RITA, AF, and Onc-1. We found that DNA Society, the Swedish Research Council, and Knut and Alice Wallenberg Foun- repair factors, including HR factors RAD51, RNF8, and RNF168 are dation. S. Peuget has been supported by a fellowship from Wenner Gren strongly repressed upon RITA, AF,andOnc-1,similartoCDK12/ Foundation. CDK13 inhibitors (4). HR-deficient cells, such as BRCA2-mutant The costs of publication of this article were defrayed in part by the payment of cells, are known to be particularly sensitive to PARP inhibitors (38). page charges. This article must therefore be hereby marked advertisement in Thus, compounds such as RITA, AF, and Onc-1, which simulta- accordance with 18 U.S.C. Section 1734 solely to indicate this fact. neously cause replication stress and inhibit HR, are very attractive for combination therapy with PARP inhibitors to achieve synthetic Received July 5, 2019; revised November 11, 2019; accepted January 23, 2020; lethality in cancer cells with functional HR. published first February 4, 2020.

References 1. Schenone M, Dancík V, Wagner BK, Clemons PA. Target identification and 10. Pierce AJ, Johnson RD, Thompson LH, Jasin M. XRCC3 promotes homology- mechanism of action in chemical biology and drug discovery. Nat Chem Biol directed repair of DNA damage in mammalian cells. Genes Dev 1999;13:2633–8. 2013;9:232–40. 11. Issaeva N, Bozko P, Enge M, Protopopova M, Verhoef LGGC, Masucci M, et al. 2. Savitski MM, Reinhard FBM, Franken H, Werner T, Savitski MF, Eberhard D, Small molecule RITA binds to p53, blocks p53-HDM-2 interaction and activates et al. Tracking cancer drugs in living cells by thermal profiling of the proteome. p53 function in tumors. Nat Med 2004;10:1321–8. Science 2014;346:1255784. 12. Grinkevich VV, Nikulenkov F, Shi Y, Enge M, Bao W, Maljukova A, et al. 3. Bradner JE, Hnisz D, Young RA. Transcriptional addiction in cancer. Cell 2017; Ablation of key oncogenic pathways by RITA-reactivated p53 is required for 168:629–43. efficient apoptosis. Cancer Cell 2009;15:441–53. 4. Zhang T, Kwiatkowski N, Olson CM, Dixon-Clarke SE, Abraham BJ, Greifen- 13. Wanzel M, Vischedyk JB, Gittler MP, Gremke N, Seiz JR, Hefter M, et al. berg AK, et al. Covalent targeting of remote cysteine residues to develop CDK12 CRISPR-Cas9-based target validation for p53-reactivating model compounds. and CDK13 inhibitors. Nat Chem Biol 2016;12:876–84. Nat Chem Biol 2016;12:22–8. 5. Pagliarini R, Shao W, Sellers WR. Oncogene addiction: pathways of therapeutic 14. Saei AA, Beusch CM, Chernobrovkin A, Sabatier P, Zhang B, Tokat UG,€ et al. response, resistance, and road maps toward a cure. EMBO Rep 2015;16:280–96. ProTargetMiner as a proteome signature library of anticancer molecules for 6. Reinhold WC, Sunshine M, Liu H, Varma S, Kohn KW, Morris J, et al. CellMiner: functional discovery. Nat Commun 2019;10:1–13. a web-based suite of genomic and pharmacologic tools to explore transcript and 15. Akama T, Shida Y, Sugaya T, Ishida H, Gomi K, Kasai M. Novel 5-aminoflavone drug patterns in the NCI-60 cell line set. Cancer Res 2012;72:3499–511. derivatives as specific antitumor agents in breast cancer. J Med Chem 1996;39: 7. Franken H, Mathieson T, Childs D, Sweetman GMA, Werner T, Togel€ I, et al. 3461–9. Thermal proteome profiling for unbiased identification of direct and indirect 16. Meng L, Kohlhagen G, Liao Z, Antony S, Sausville E, Pommier Y. DNA-protein drug targets using multiplexed quantitative mass spectrometry. Nat Protoc 2015; cross-links and replication-dependent histone H2AX phosphorylation induced 10:1567–93. by aminoflavone (NSC 686288), a novel anticancer agent active against human 8. Mi H, Huang X, Muruganujan A, Tang H, Mills C, Kang D, et al. PANTHER breast cancer cells. Cancer Res 2005;65:5337–43. version 11: expanded annotation data from gene ontology and reactome path- 17. Guo W, Wu S, Liu J, Fang B. Identification of a small molecule with synthetic ways, and data analysis tool enhancements. Nucleic Acids Res 2017;45:D183–9. lethality for K-ras and protein kinase C iota. Cancer Res 2008;68:7403–8. 9. Szklarczyk D, Franceschini A, Wyder S, Forslund K, Heller D, Huerta-Cepas J, 18. Wu S, Wang L, Guo W, Liu X, Liu J, Wei X, et al. Analogues and derivatives of et al. STRING v10: protein-protein interaction networks, integrated over the tree oncrasin-1, a novel inhibitor of the C-terminal domain of RNA polymerase II, of life. Nucleic Acids Res 2015;43:D447–452. and their antitumor activities. J Med Chem 2011;54:2668–79.

AACRJournals.org Cancer Res; 80(7) April 1, 2020 1549

Peuget et al.

19. Guo W, Wu S, Wang L, Wei X, Liu X, Wang J, et al. Antitumor activity of a novel 29. Ahmed A, Yang J, Maya-Mendoza A, Jackson DA, Ashcroft M. Pharmacological oncrasin analogue is mediated by JNK activation and STAT3 inhibition. activation of a novel p53-dependent S-phase checkpoint involving CHK-1. PLoS One 2011;6:e28487. Cell Death Dis 2011;2:e160. 20. Guo W, Wu S, Wang L, Wang R, Wei X, Liu J, et al. Interruption of RNA 30. de Feraudy S, Revet I, Bezrookove V, Feeney L, Cleaver JE. A minority of foci processing machinery by a small compound 1-[(4-chlorophenyl) methyl]-1H- or pan-nuclear apoptotic staining of gH2AXintheSphaseafterUVdamage indole-3-carboxaldehyde (oncrasin-1). Mol Cancer Ther 2009;8:441–8. contain DNA double-strand breaks. Proc Natl Acad Sci U S A 2010;107: 21. McLean LS, Brantley E. Apoptotic mechanism of novel anticancer agents is 6870–5. mediated by MAPKs in breast cancer cells. FASEB J 2013;27:1105.3. 31. Brickner JR, Soll JM, Lombardi PM, Vagbø CB, Mudge MC, Oyeniran C, et al. 22. Shi Y, Nikulenkov F, Zawacka-Pankau J, Li H, Gabdoulline R, Xu J, et al. ROS- A ubiquitin-dependent signalling axis specific for ALKBH-mediated DNA dependent activation of JNK converts p53 into an efficient inhibitor of oncogenes dealkylation repair. Nature 2017;551:389–93. leading to robust apoptosis. Cell Death Differ 2014;21:612–23. 32. McLean L, Soto U, Agama K, Francis J, Jimenez R, Pommier Y, et al. 23. Hedstrom€ E, Eriksson S, Zawacka-Pankau J, Arner ESJ, Selivanova G. p53- Aminoflavone induces oxidative DNA damage and reactive oxidative dependent inhibition of TrxR1 contributes to the tumor-specific induction of species-mediated apoptosis in breast cancer cells. Int J Cancer 2008;122: apoptosis by RITA. Cell Cycle 2009;8:3584–91. 1665–74. 24. Enge M, Bao W, Hedstrom€ E, Jackson SP, Moumen A, Selivanova G. MDM2- 33. Wei X, Guo W, Wu S, Wang L, Huang P, Liu J, et al. Oxidative dependent downregulation of p21 and hnRNP K provides a switch between stress in NSC-741909-induced apoptosis of cancer cells. J Transl Med apoptosis and growth arrest induced by pharmacologically activated p53. 2010;8:37. Cancer Cell 2009;15:171–83. 34. Gorrini C, Harris IS, Mak TW. Modulation of oxidative stress as an anticancer 25. Girard C, Will CL, Peng J, Makarov EM, Kastner B, Lemm I, et al. Post- strategy. Nat Rev Drug Discov 2013;12:931–47. transcriptional spliceosomes are retained in nuclear speckles until splicing 35. Koumenis C, Giaccia A. Transformed cells require continuous activity of completion. Nat Commun 2012;3:994. RNA polymerase II to resist oncogene-induced apoptosis. Mol Cell Biol 26. Calo E, Gu B, Bowen ME, Aryan F, Zalc A, Liang J, et al. Tissue–selective effects of 1997;17:7306–16. nucleolar stress and rDNA damage in developmental disorders. Nature 2018; 36. Shin D, Kim EH, Lee J, Roh J-L. RITA plus 3-MA overcomes chemoresistance of 554:112–7. head and neck cancer cells via dual inhibition of autophagy and antioxidant 27. Ho JSL, Ma W, Mao DYL, Benchimol S. p53-Dependent transcriptional systems. Redox Biol 2017;13:219–27. repression of c-myc is required for G1 cell cycle arrest. Mol Cell Biol 2005;25: 37. Allgayer J, Kitsera N, Bartelt S, Epe B, Khobta A. Widespread transcriptional gene 7423–31. inactivation initiated by a repair intermediate of 8-oxoguanine. Nucleic Acids 28. Fumagalli S, Di Cara A, Neb-Gulati A, Natt F, Schwemberger S, Hall J, et al. Res 2016;44:7267–80. Absence of nucleolar disruption after impairment of 40S ribosome biogenesis 38. Bryant HE, Schultz N, Thomas HD, Parker KM, Flower D, Lopez E, et al. Specific reveals an rpL11-translation-dependent mechanism of p53 induction. Nat Cell killing of BRCA2-deficient tumours with inhibitors of poly(ADP-ribose) poly- Biol 2009;11:501–8. merase. Nature 2005;434:913–7.

1550 Cancer Res; 80(7) April 1, 2020 CANCER RESEARCH

Thermal Proteome Profiling Identifies Oxidative-Dependent Inhibition of the Transcription of Major Oncogenes as a New Therapeutic Mechanism for Select Anticancer Compounds

Sylvain Peuget, Jiawei Zhu, Gema Sanz, et al.

Cancer Res 2020;80:1538-1550. Published OnlineFirst February 4, 2020.

Updated version Access the most recent version of this article at: doi:10.1158/0008-5472.CAN-19-2069

Supplementary Access the most recent supplemental material at: Material http://cancerres.aacrjournals.org/content/suppl/2020/02/01/0008-5472.CAN-19-2069.DC2 http://cancerres.aacrjournals.org/content/suppl/2020/01/31/0008-5472.CAN-19-2069.DC1 http://cancerres.aacrjournals.org/content/suppl/2020/02/04/0008-5472.CAN-19-2069.DC3

Cited articles This article cites 37 articles, 8 of which you can access for free at: http://cancerres.aacrjournals.org/content/80/7/1538.full#ref-list-1

Citing articles This article has been cited by 1 HighWire-hosted articles. Access the articles at: http://cancerres.aacrjournals.org/content/80/7/1538.full#related-urls

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

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at Subscriptions [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/80/7/1538. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.