The Antitumor Drugs Trabectedin And

Published OnlineFirst December 14, 2018; DOI: 10.1158/1541-7786.MCR-18-0575

Genome Maintenance Molecular Cancer Research The Antitumor Drugs Trabectedin and Lurbinectedin Induce Transcription-Dependent Replication Stress and Genome Instability Emanuela Tumini1, Emilia Herrera-Moyano1, Marta San Martín-Alonso1, Sonia Barroso1, Carlos M. Galmarini2, and Andres Aguilera1

Abstract

R-loops are a major source of replication stress, DNA tedin and lurbinectedin increased the presence of Rad52 foci, a damage, and genome instability, which are major hallmarks marker of DNA damage, in an R-loop–dependent manner. In of cancer cells. Accordingly, growing evidence suggests that R- addition to providing new insights into the mechanisms of loops may also be related to cancer. Here we show that R-loops action of these drugs, our study reveals that R-loops could be play an important role in the cellular response to trabectedin targeted by anticancer agents. Given the increasing evidence (ET743), an anticancer drug from marine origin and its deriv- that R-loops occur all over the genome, the ability of lurbi- ative lurbinectedin (PM01183). Trabectedin and lurbinecte- nectedin and trabectedin to act on them may contribute to din induced RNA–DNA hybrid-dependent DNA damage in enhance their efﬁcacy, opening the possibility that R-loops HeLa cells, causing replication impairment and genome insta- might be a feature shared by speciﬁc cancers. bility. We also show that high levels of R-loops increase cell sensitivity to trabectedin. In addition, trabectedin led to Implications: The data presented in this study provide the new transcription-dependent FANCD2 foci accumulation, which concept that R-loops are important cellular factors that con- was suppressed by RNase H1 overexpression. In yeast, trabec- tribute to trabectedin and lurbinectedin anticancer activity.

Introduction (TAM). Indeed, both drugs inhibit the transcription of selected cytokines (e.g., CCL2, IL6, IL8, PTX3) by TAMs abrogating their Cancer therapy greatly relies on the use of small molecules that protumoral properties and modifying the tumor microenviron- directly perturb the DNA metabolism. Trabectedin (ET743) is one ment (9, 10). of such molecules. It has been initially derived from the sea squirt Tumor cells are commonly associated with genome instability. Ecteinascidia turbinata and is now produced synthetically, being One important natural source of instability is represented by R- currently used in therapy of patients with advanced soft tissue loops, a three-stranded nucleic acid structure consisting of an sarcoma (1) and platinum-sensitive ovarian cancer (2). Different RNA–DNA hybrid and the displaced single-stranded DNA of the derivatives of trabectedin have been synthesized, among them original DNA duplex (11). This structure could fulﬁll physiologic lurbinectedin (PM01183), which is presently under evaluation in roles as it occurs during immunoglobulin class switching recom- phase II and III clinical trials. bination, plasmid or mitochondrial DNA replication, and tran- One of the most important features of the mechanism of action scription regulation. However, they are an important source of of trabectedin and lurbinectedin is their inhibitory effect on DNA damage, genome instability, and replication stress, which transcription of protein-coding genes, which is mediated by relates R-loops with a number of neurodegenerative disorders and different means. First, they preferentially bind to speciﬁc DNA cancer (12, 13). Interestingly, the loss of function of the breast triplets that are often present on transcription recognition sites. In cancer susceptibility genes BRCA1 and BRCA2, which have been this way they could directly prevent loading of transcription shown to increase R-loops and R-loop-dependent DNA damage factors onto chromatin (3). Moreover, both drugs were shown (14, 15), is also associated with an increased sensitivity to tra- to induce degradation of the RNA polymerase II (RNAPII) bectedin (16). In a similar way, dysfunctions of the Fanconi through the ubiquitin–proteasome pathway (4–8). Of note, these anemia (FA) pathway that increased sensitivity to such agent effects are also observed in tumor-associated macrophages (5) also cause accumulation of R-loops (17, 18). In addition, there are some similarities between R-loops and trabectedin adducts worth being considered. The head-to-tail 1 Centro Andaluz de Biología Molecular y Medicina Regenerativa-CABIMER, binding of three trabectedin molecules to three adjacent opti- CSIC-Universidad Pablo de Olavide-Universidad de Sevilla, Seville, Spain. 2PharmaMar, Colmenar Viejo, Spain. mal DNA binding sites changes the DNA duplex conformation from the B-form to an intermediate between the A- and the B- Note: Supplementary data for this article are available at Molecular Cancer form, which strongly resembles the conformation of an RNA– Research Online (http://mcr.aacrjournals.org/). DNA hybrid (19). Moreover, R-loops that accumulate in the Corresponding Author: Andres Aguilera, Universidad de Sevilla; Av. Americo absence of the putative RNA–DNA helicases AQR or SETX Vespucio 24, Seville 41092, Spain. Phone: 34954468372; E-mail: [email protected] require the action of nucleotide excision repair (NER) to be doi: 10.1158/1541-7786.MCR-18-0575 processed into double-strand breaks (DSB; ref. 20) as well as 2018 American Association for Cancer Research. trabectedin adducts (21).

www.aacrjournals.org 773

Tumini et al.

For these reasons, in this study we have further explored the Immunofluorescence and EdU labeling mechanism of action of trabectedin and lurbinectedin, in partic- Immunofluorescence was performed as previously described ular whether it could be partially mediated or affected by R-loops. (25). Briefly, cells were fixed with 3.7% formaldehyde in PBS for We show that the DNA damage and replicative stress caused by 15 minutes, permeabilized with 0.5% Triton X-100, and blocked trabectedin and lurbinectedin is indeed partially dependent on R- with 3% BSA in PBS for 1 hour. The coverslips were then incubated loops, as demonstrated by molecular and genetic assays using with primary antibodies diluted 1:500 in 3% BSA in PBS for both human cell lines and the yeast Saccharomyces cerevisiae as a 2 hours followed by 1 hour incubation with 1:1,000 diluted model eukaryotic system. Our findings suggest that R-loops could secondary antibodies and nuclei counterstaining with 1 mg/mL be targets of anticancer agents and given the evidence that R-loops DAPI in PBS. For FANCD2 immunolabeling, before fixation cells occur all over the genome (22), this opens the possibility that were pre-permeabilized with 0.25% Triton X-100 in PBS for 1 tumoral cells that accumulate R-loops may be more suitable for minute on ice. S9.6 immunofluorescence was performed as trabectedin and lurbinectedin treatment. previously described (20). For the EdU labeling, cells were pulse-labeled with 10 mmol/L EdU for 20 minutes before fixation. Materials and Methods EdU staining was performed with a Click-iT EdU Alexa Fluor 555 Imaging Kit (Invitrogen) according to manufacturer's instruc- Cell culture and treatments tions. Random images were acquired with a 63 or 40 objective HeLa cells were purchased from European Collection of Cell and foci or nuclear intensity were scored using the MetaMorph Cultures (ECACC), HeLa-TR (23), and HeLa HB-GFP (14) software. At least 100 cells were scored for each condition. were previously generated in our laboratory. All cell lines were maintained in DMEM medium, supplemented with 10% heat- DNA combing inactivated FCS, cultured at 37C in a humidified atmosphere DNA combing was performed as previously described (23). containing 5% CO , and routinely tested for mycoplasma using 2 Briefly, DNA fibers were extracted from cells in agarose plugs MycoAlert Mycoplasma Detection Kit (Lonza). HeLa HB-GFP immediately after CldU labeling and were stretched on silanized cells medium was supplemented with doxycycline at the final coverslips. DNA molecules were counterstained with an autoanti- concentration of 2 mg/mL to induce HB-GFP expression. Specific ssDNA antibody (DSHB, 1:500) and an anti-mouse IgG coupled genes were knocked down using ON-TARGET SMARTpool siRNA to Alexa Fluor 647 (A21241; Molecular Probes; 1:50). CldU and from Dharmacon. Transient transfection of siRNA was performed IdU were detected with BU1/75 (AbCys; 1:20) and an anti-rat IgG using DharmaFECT 1 (Dharmacon), according to the manufac- coupled to Alexa Fluor 488 (A21470; Molecular Probes; 1:50) or turer's instructions. The following plasmids were used for trans- B44 (Becton Dickinson; 1:20) anti-BrdU antibodies and an anti- fection: pcDNA3 (Invitrogen) (RNH1) and pcDNA3-RNaseH1 mouse IgG coupled to Alexa Fluor 546 (A21123; Molecular (RNH1þ), containing the full-length RNase H1 cloned into Probes; 1:50), respectively. DNA fibers were analyzed on a Leica pcDNA3 (24). Lipofectamine 2000 (Invitrogen) was used for DM6000 microscope equipped with a DFC390 camera (Leica). plasmid transfection. Assays were performed 48 or 72 hours after Data acquisition was performed with LAS AF (Leica). Fork velocity transfection. Trabectedin (ET743) and lurbinectedin (PM01183) was calculated as previously described (26). Replication asym- were provided from Pharmamar. Cordycepin (C3394) and metry was calculated by dividing (longest green tract –shortest neocarzinostatin (NCZ; N9162) were purchased from Sigma. green tract) by the longest tract in divergent CIdU tracks. Single-cell electrophoresis Single-cell electrophoresis or comet assays were performed as Yeast "halo" assay described using a commercial kit (Trevigen) following the man- Yeast cells of the wild-type W303 or the rad52D strains were ufacturer's instructions. Comet tail moments were analyzed using plated in presence of 0.003% SDS in order to increase yeast cell Open Comet software. At least 100 cells were scored in each permeability. Afterwards, paper filter disks were placed on the agar experiment to calculate the median of the tail moment, as plates and 3 mL of increasing amounts of trabectedin or lurbi- reported (23). nectedin (or DMSO as carrier control) were spotted onto them. After 3 to 6 days cells growth was monitored, plates were scanned, Cell proliferation and the diameter of the growth inhibition halo was measured. HeLa cells were siRNA transfected to knock down the specified genes and plated in 96-well plate. Following 72 hours, cells were Rad52 foci analysis in yeast cells treated with increasing doses of trabectedin (ET743) for 24 hours Wild-type W303 yeast cells were transformed in plates contain- and cell proliferation was measured by the WST-1 reagent (Roche) ing doxycycline (at the final concentration of 10 mg/mL in order to following the manufacturer's instructions. Absorbance was mea- repress the expression of ectopic RNase H1) with pWJ1344 sured using microplate (ELISA) reader VARIOSKAN FLASH plasmid coding the RAD52–YFP fusion protein (27) and with (Thermo) at 450 nm with a reference wavelength of 690 nm. pCM189-RNH1 with the RNH1 gene under the tet promoter (28) Absorbance values were normalized to the value of the untreated or the empty vector pCM189 (29). Cells were grown overnight and represented as arbitrary units (A.U.). with medium SC-L-U (Complete medium SC: 0.17% yeast nitro- gen base and 0.5% ammonium sulfate, 2% glucose, supplemen- Antibodies ted with amino acids; –L and –U indicate the absence of leucine Antibodies anti-RNASEH1 (15606-1-AP) and anti-FANCD2 and uracil, respectively) containing 0.003% SDS and without (sc-20022) anti-53BP1 (NB100-304) were purchased from doxycyclin. Mid-log cultures overexpressing RNase H1 or not, Proteintech, Santa Cruz Biotechnology, and Novus Biologicals, were treated with increasing doses of trabectedin or lurbinectedin respectively. The S9.6 antibody was purified from the hybrid- during 2 hours at 30C with 200 rpm shaking. Following fixation oma cell line HB-8730. with 2.5% formaldehyde in 0.1M KHPO at pH 6.4 for 10 minutes,

774 Mol Cancer Res; 17(3) March 2019 Molecular Cancer Research

ET743 and PM01183 and RNA-Dependent DNA Damage

the cells were washed twice with KHPO 0.1M pH 6.6 and once tedin, we analyzed the effect of a different and well-establish with KHPO 0.1M pH 7.4. Cells were permeabilized with ethanol genotoxic and antitumoral drug, NCZ (30), and we found that it 80% for 10 minutes and after centrifugation were resuspended in also increased FANCD2 foci but this phenotype was not sup- water containing DAPI at the final concentration of 1 mg/mL to pressed by RNase H1 overexpression (Supplementary Fig. S2). counterstain the nuclei. Rad52–YFP foci were scored in more than Given the observed connection of RNA–DNA hybrids and DNA 200 nuclei from S-G2 mid-log cells, using a Leica DC 350F damage with trabectedin and lurbinectedin treatment, we tested microscope. directly for the presence of RNA–DNA hybrids using the S9.6 antibody, which specifically recognizes these structures, by in situ Statistical analysis immunofluorescence (IF). As can be seen in Supplementary Fig. Statistical significance was determined by t tests or the Mann– S3, trabectedin and lurbinectedin did not cause a significant Whitney test as specified. All tests performed were two-tailed increase in the nuclear intensity of the S9.6 signal. Because these unless otherwise specified and three levels of statistical signifi- drugs significantly decrease transcription and therefore the overall cance were considered: , P 0.05; , P 0.01 and , P 0.001. RNA levels (8, 31) it is unlikely that they concomitantly increase R-loop levels. Nonetheless, they could act by exacerbating their impact on replication and genome integrity. Consistently, both Results drugs affected nucleolus integrity, as deduced from a perturbed Trabectedin and lurbinectedin treatment increase localization of the nucleolar marker nucleolin and by a smaller or transcription-dependent DNA damage fragmented appearance of nucleoli detected with the S9.6 Trabectedin and lurbinectedin covalently bind to DNA, phys- antibody. ically impeding transcription and other DNA metabolic processes. To explore the possibility that the DNA adducts formed by these Trabectedin enhances R-loop–dependent replication drugs could induce transcription-dependent genome instability, impairment we analyzed DNA damage foci accumulation in HeLa cells in Trabectedin directly interacts with DNA, forming adducts and which transcription was abrogated by treatment with cordycepin, producing distortions of the double helix. Because R-loops impair a potent transcription inhibitor that halts RNA chain elongation. replication progression (32–36), we analyzed the impact of HeLa cells treated during 4 hours with 25 nmol/L of trabectedin or trabectedin on DNA replication and whether this would be lurbinectedin accumulated 53BP1 foci, a DSB marker (Fig. 1A). associated with the presence of RNA–DNA hybrids. For this, we However, when transcription was previously shut off with cordy- first assessed DNA replication levels after trabectedin treatment, cepin (added 1 hour earlier), 53BP1 foci were significantly by measuring the amount of incorporated EdU, a thymidine reduced, suggesting an important role of transcription in the analog. Trabectedin treatment strongly affected the rate of DNA mechanism of action of these drugs. synthesis in HeLa cells, leading to lower levels of EdU incorpo- Next, we assayed whether RNA–DNA hybrids could be ration compared with untreated cells (Fig. 2A). However, the involved in the transcription-dependent DNA damage observed. percentage of cells in S phase (EdU-incorporating cells) was not Indeed, overexpression of RNase H1, which removes the RNA affected (Supplementary Fig. S4A). Importantly, overexpression moiety of the hybrids, partially reduced the 53BP1 foci observed of RNase H1, which by itself caused a slight decrease in EdU after trabectedin treatment, suggesting that part of the role of incorporation, partially rescued the phenotype of reduced DNA transcription in the mechanism of action of these drugs was synthesis (Fig. 2A). associated with R-loops (Supplementary Fig. S1). In this case, we Next, we used DNA combing to check for impairment of exposed cells to a lower dose of trabectedin for longer (10 nmol/L replication fork progression in single molecules. Using low tra- during 6 hours) to better see the increase of 53BP1 foci. This bectedin concentrations at which the replication fork velocity was longer treatment was not possible with cordycepin-mediated not perturbed, the drug caused an increase in fork stalling, as transcription shut off. determined by fork asymmetry, which was significantly higher in Provided the known negative effect of transcription and RNA– trabectedin-treated versus nontreated cells (Fig. 2B). To know DNA hybrids on replication fork (RF) progression, we decided to whether this effect was dependent on transcription, required for analyze the accumulation of FANCD2 foci, a central player of the the cotranscriptional formation of RNA–DNA hybrids, we inhib- FA DNA repair pathway, as a way to measure repair events ited transcription using cordycepin. With this approach, all cells occurring at potential stalled forks. FA is a DNA repair pathway were exposed to the same treatment, something that was not that works on stalled RFs and cells deficient in this pathway have possible after transfection with an RNase H1-overexpressing also been shown to be more sensitive to trabectedin (5). The plasmid. We observed that treatment of cells with the transcrip- experiment was performed as described above for 53BP1 foci. tion inhibitor cordycepin abolished the difference in fork asym- Trabectedin and lurbinectedin caused a significant increase metry between trabectedin-treated and untreated samples. This of FANCD2 foci that was completely reversed by cordycepin result suggests that fork-stalling caused by trabectedin is linked to treatment (Fig. 1B), suggesting a key role of these drugs in transcription. Inhibition of transcription by cordycepin was det- transcription–replication (T–R) conflicts. Because both drugs rimental for replication at higher doses of trabectedin, which per se caused very similar phenotypes, consistent with an expected provoked a reduction in the replication rate (Supplementary Fig. common mechanism of action, we focused the rest of the study S4B). Thus, it was not surprising that the concomitant treatment mainly on trabectedin. with the two chemicals reduced further the DNA replication rate FANCD2 foci accumulation approximately doubled after a as assessed by EdU incorporation. However, because trabectedin longer exposure to trabectedin at lower dose and this increase can also inhibit transcription (31), it is possible that the effect on was entirely rescued by RNase H1 overexpression (Fig. 1C). To DNA replication has multiple causes that would require further assure that this suppression by RNase H1 was specific for trabec- investigation.

www.aacrjournals.org Mol Cancer Res; 17(3) March 2019 775

Tumini et al.

Figure 1. Trabectedin and lurbinectedin treatment increased transcription-dependent 53BP1 and FANCD2 foci. A and B, Immunodetection of 53BP1 and FANCD2 foci in HeLa cells treated with trabectedin (ET743 or ET), lurbinectedin (PM01183 or PM) or left untreated (Unt) as indicated. Cells were or were not pretreated with cordycepin 50 mmol/L (CORD or CORDþ) during 1 hour and during the following 4 hours of treatment. Representative images and quantification are shown. Bars represent the averages SEM of the percentage of cells with more than five foci from at least three independent experiments. C, HeLa cells were transfected with the RNase H1 overexpression plasmid (RNH1þ) or with the empty vector (RNH1). Following 24 hours after transfection cells were treated with trabectedin (ET743) as indicated. After the incubation with ET743 cells were fixed and immunodetection of FANCD2 was performed. Representative images and quantification of cells with more than five FANCD2 foci in cells positive for RNase H1 overexpression are shown. Bars represent the averages SEM of the percentage of cells with more than five foci from five independent experiments. Statistical significance was assessed by Student t test.

776 Mol Cancer Res; 17(3) March 2019 Molecular Cancer Research

ET743 and PM01183 and RNA-Dependent DNA Damage

Figure 2. Trabectedin treatment caused R- loop–dependent replication impairment. A, HeLa cells were transfected with a plasmid built to overexpress RNase H1 (RNH1þ)or with the empty vector (RNH1). Following 24 hours after transfection cells were treated with trabectedin (ET743) as indicated and pulse labeled with EdU before ﬁxation. Representative images from the cells labeled for EdU and RNase H1 are shown. Distribution of EdU nuclear intensity in cells positive for EdU (in RNH1 and in RNH1þ cells) and for RNase H1 overexpression (in RNH1þ cells) presented by a scatter dot plot where the median value is shown. The data from three independent experiments were pulled together. Differences between distributions were assessed by the Mann– Whitney test. B, Outline of the experimental procedure followed before harvesting and processing HeLa cells for DNA combing. Distribution of replication fork velocity and asymmetry are presented by a scatter dot plot where the median value is shown. "Cord" indicates cordycepin. The data from four or two independent experiments were pulled together for fork velocity and asymmetry, respectively. Differences between distributions were assessed by the Mann–Whitney test.

Based on these observations, we infer that the replication cellular sensitivity to trabectedin both in the absence of HB-GFP obstacles posed by trabectedin treatment are partially dependent and in cells expressing HB–GFP (Fig. 3, middle and right), having on RNA–DNA hybrids. THOC1 depletion a stronger effect.

RNA–DNA hybrid-accumulating cells are hyper-sensitive to Trabectedin and lurbinectedin induce DNA breaks that are trabectedin partially mediated by RNA–DNA hybrids in human cells To better establish a functional link between RNA–DNA Next, we analyzed the effect of trabectedin treatment in cell hybrids and trabectedin, we assayed the sensitivity to trabectedin proliferation and genome integrity in human cells in relation to R- of cells in which RNA–DNA hybrids were stabilized by a hybrid- loop accumulation. For this, HeLa cells were ﬁrst transfected with binding protein. For this purpose we took advantage of a stable a plasmid that leads to the overexpression of RNase H1 before cell line that expresses the chimeric protein HB-GFP, a fusion being treated with trabectedin or lurbinectedin and analyzed by protein constituted by the hybrid binding (HB) domain of the alkaline single-cell electrophoresis (comet assay) for the accumu- RNase H1 enzyme and the GFP. Under these conditions RNA– lation of DNA breaks. As shown in Fig. 4, trabectedin and DNA hybrids are likely "stabilized" by the binding of the HB lurbinectedin increased DNA breaks, as expected. However, when domain, thus retaining the R-loops in the cell, consistent with cells overexpressed RNase H1, breaks were signiﬁcantly reduced in previously reported results from BRCA2-depleted cells (14). cells treated with 50 nmol/L lurbinectedin for 2 hour. We eval- Interestingly, cells expressing the HB–GFP fusion protein were uated gH2AX foci formation, a molecular marker that in addition more sensitive to trabectedin as assessed by cell proliferation to DNA damage is also associated with replicative stress signaling. assays (Fig. 3, left), suggesting that cells with higher levels of R- Trabectedin treatment increased gH2AX foci accumulation (Sup- loops can be made more susceptible to this anticancer agent. We plementary Fig. S5A). In this case, however, RNase H1 treatment next knocked down two cellular factors, THOC1 and BRCA2, increased further the gH2AX foci, suggesting that a double treat- previously shown to enhance RNA–DNA hybrids by either pre- ment with trabectedin and RNase H1 leads to an additive repli- venting their formation or by promoting their removal. Depletion cative stress and DNA damage. Indeed, RNA–DNA hybrids are of any of the two genes is associated with an increase in RNA–DNA also important intermediates of normal DNA replication (for hybrids formation (14). The lack of BRCA2 or THOC1 increased Okazaki fragment synthesis) and, consistently, a clonogenic assay

www.aacrjournals.org Mol Cancer Res; 17(3) March 2019 777

Tumini et al.

Figure 3. HB-GFP cells showed increased sensitivity to trabectedin treatment. HB-GFP cells (HBGFPþ) expressing the fusion protein HB-GFP or the parental cell line (HBGFP) were knocked down for the speciﬁed genes for 72 hours. After treatment with increasing doses of trabectedin (ET743) for 24 hours cell proliferation was measured by the WST-1 reagent. The graphs show the relative values of cell proliferation referred to the untreated sample at increasing doses of trabectedin. Averages SEM from 10 independent experiments is presented. Statistical signiﬁcance was assessed by paired t test comparing for each siRNA the parental cells with the HB-GFP–expressing cell line. In the central and left graphs, the dotted grey line represents the parental cells transfected by siC.

revealed that survival of cells overexpressing RNase H1 at the time foci compared with the untreated control (Fig. 5B). To assay of treatment with increasing doses of trabectedin was reduced whether Rad52 foci were dependent on RNA–DNA hybrids, (Supplementary Fig. S5B). RNAse H1 was overexpressed in yeast cells by transformation with an RNase H1-overexpressing plasmid. Notably, the increase The RNA–DNA hybrid-dependent genome instability in Rad52 foci was suppressed by RNase H1 overexpression associated with trabectedin and lurbinectedin is evolutionary (Fig. 5B; Supplementary Fig. S6A). These results indicate that conserved trabectedin-induced breaks are linked to the accumulation of To investigate whether trabectedin and lurbinectedin interact RNA–DNA hybrids also in yeast cells. with RNA–DNA hybrids regardless of cell type and organism as a Next, we assayed the effect of lurbinectedin on cell growth and way to confirm that they target directly DNA metabolism, we genome instability using the assays described above for trabecte- asked whether similar DDR phenotypes could be reproduced in din. As can be seen in Fig. 5C, lurbinectedin affected yeast growth the eukaryotic model organism Saccharomyces cerevisiae. For this as determined by the size of the growth inhibition halo formed we first analyzed the effect of trabectedin. Replication impairment around the increasing concentrations of the chemical spotted on causes genome instability and Rad52 is used as a marker to detect Whatman filter papers. Again, this effect was significantly instability in the form of the accumulation of DNA damage. To enhanced in rad52D strains, consistent with the idea that cell assess sensitivity to trabectedin and lurbinectedin, we used the growth was impaired due to the accumulation of unrepaired "halo" assay. Different yeast strains were homogenously plated on DSBs. Accordingly, lurbinectedin treatment caused a two- to YEPD-rich medium plates, and then Whatman-paper disks three-fold increase of cells with Rad52 foci compared with soaked with increasing concentrations of the tested drug were untreated cells and this phenotype was fully suppressed by RNase placed on top of the plate. After growth, halos of growth inhibi- H1 overexpression (Fig. 5D; Supplementary Fig. S6A). The tion around the disks were formed with a diameter that correlates decreased amount of Rad52 foci after overexpression of RNase with the levels of toxicity of the drug. As can be seen in Fig. 5A, H1 in trabectedin- or lurbinectedin-treated cells was not due to the increasing concentrations of trabectedin lead to increased yeast reduction of the protein levels, as confirmed by Western blot sensitivity, as detected by the size of the halo of growth inhibition analysis (Supplementary Fig. S6B). around the trabectedin spot. Importantly, this halo was largely Consequently, lurbinectedin, as well as trabectedin, caused R- increased in rad52D yeast strains inactivated for homologous loop–dependent genome instability both in yeast and in human recombination DSB repair in comparison with the wild-type cells. This finding suggests that their action is most likely common strain, even at concentrations at which trabectedin had no effect to many different cell types, because they target molecular path- on wild-type cells (Fig. 5A). These result suggest that recombino- ways that are well conserved across evolution. genic DNA breaks accumulated after trabectedin treatment, consistent with previous reports (21). To confirm this, we analyzed recombinogenic DSBs by Rad52 foci, used as a way to detect DSB Discussion repair centers (27). Consistently, trabectedin caused a significant Increasing evidence points out the role of RNA–DNA hybrids in increase of Rad52 foci. Cells treated with 250 mmol/L trabectedin, the boost of genome instability and replication stress that are showed a three-fold increase in the percentage of cells with Rad52 major hallmarks of cancer cells. This opens the possibility that

778 Mol Cancer Res; 17(3) March 2019 Molecular Cancer Research

ET743 and PM01183 and RNA-Dependent DNA Damage

Figure 4. Trabectedin and lurbinectedin caused R-loop–dependent DNA damage. HeLa cells were transfected with the RNase H1 overexpression plasmid (RNH1þ) or with the empty vector (RNH1). Following 48 hours after transfection cells were treated with trabectedin (ET743; A) or lurbinectedin (PM01183; B) as indicated and DNA damage levels were measured. Representative images from alkaline comet assays and quantification of the tail moment are shown. Bars represent the averages SEM of the median from at least three independent experiments. Statistical significance was assessed by the one-tailed Mann–Whitney test. such hybrids could occur at high levels in some tumor cells and trabectedin and lurbinectedin activities are favored by transcrip- consequently could be used as therapeutic targets. For this reason, tion and, at least partially, they might be enhanced by R-loops. it is of growing importance to address the relationship between R- Trabectedin and lurbinectedin treatment enhanced transcription- loops and the action of anticancer agents. Trabectedin and lurbi- dependent genome instability and replication stress (Figs. 1 nectedin are two powerful drugs used in cancer treatment whose and 2). Exposure to trabectedin or lurbinectedin did not cause ability to block cell proliferation is linked to its potential to cause a global increase in R-loops as assayed by S9.6 immunofluores- DNA breaks and inhibit transcription. In this study, we show that cence (Supplementary Fig. S3). However, these drugs caused

www.aacrjournals.org Mol Cancer Res; 17(3) March 2019 779

Tumini et al.

Figure 5. Trabectedin or lurbinectedin treatment caused R-loop– dependent genome instability in yeast cells. A, Representative images of halo assay performed in W303 and rad52D yeast cells and quantification of the inhibition halo diameter after the indicated trabectedin (ET743) treatments. Bars represent the averages SEM from four independent experiments. Statistical significance was assessed by Student t test. B, Yeast cells W303 were transformed with Rad52-GFP plasmid and the RNase H1 overexpression plasmid (RNH1þ) or with the empty vector (RNH1) and treated with increasing amount of trabectedin. Bars represent the averages SEM of the percentage of cells with one or more Rad52 foci from three independent experiments. Statistical significance was assessed by Student t test. C, Representative image of halo assay performed in W303 and rad52D yeast cells and quantification of the inhibition halo diameter after the indicated lurbinectedin (PM01183) treatments. Bars represent the averages SEM from four independent experiments. Statistical significance was assessed by Student t test. D, Yeast cells W303 were transformed with Rad52-GFP plasmid and RNase H1 overexpression plasmid (RNH1þ)or with the empty vector (RNH1)and treated with increasing amount of lurbinectedin. Bars represent the averages SEM of the percentage of cells with one or more Rad52 foci from three independent experiments. Statistical significance was assessed by Student t test.

genome instability and replicative stress that were dependent on RNA–DNA hybrids therefore, overexpression of this endoribo- transcription and RNA–DNA hybrids both in yeast and human nuclease counteracts the accumulation of RNA–DNA hybrids. cells. These ﬁndings suggest that trabectedin and lurbinectedin However, the rescue of the EdU incorporation defect by RNase toxicity is higher at sites undergoing active transcription and R- H1 overexpression was mild, which suggests that only a subset of loop accumulation. the trabectedin-induced replicative problems were mediated by It was previously demonstrated that trabectedin and lurbinec- RNA-DNA hybrids. Analysis of DNA replication at the single tedin directly bind DNA, forming adducts (6) and distort the molecule level by DNA combing revealed that transcription double helix. Here we showed that these DNA adducts compro- inhibition with cordycepin reduces the increase in fork asymmetry mise replication fork progression; indeed, we found that trabec- provoked by trabectedin exposure (Fig. 2B), consistent with the tedin caused a profound impairment of DNA synthesis (Fig. 2A). conclusion that R-loops contribute to fork stalling in trabectedin To some extent, we can attribute this effect to R-loop formation, treated cells. In any case, we cannot exclude the possibility that the because RNase H1 overexpression partially suppressed this DNA inhibition of transcription by trabectedin could contribute to RF synthesis defect. RNase H1 speciﬁcally cleaves the RNA strand of asymmetry.

780 Mol Cancer Res; 17(3) March 2019 Molecular Cancer Research

ET743 and PM01183 and RNA-Dependent DNA Damage

R-loop–dependent replication problems caused by trabectedin zaki fragment dynamics. Accordingly, DNA synthesis was are possibly counteracted by the FA pathway. The FA pathway has impaired in cells overexpressing RNase H1 (Fig. 2A). The double a key role in balancing replication stress, in particular when treatment with RNase H1 overexpression and trabectedin might interstrand crosslinking agents such as Mitomycin C (MMC) cause have an additive effect by their impact on replication, enhancing replication fork blockage (37) and has been shown to limit R-loop the probability of ssDNA gaps or breaks that would lead to accumulation and R-loop–dependent genome instability (17, increased cell death (Supplementary Fig. S5B). 18). Interestingly, FA is critical for the integrity of common fragile The novel finding of an interplay between R-loops and trabec- sites (CFS), because FANCD2, a central player of the FA pathway, tedin in genome instability induction is of particular interest is required for a safe replication through R-loop–accumulating because it could be exploited for cancer cells in which R-loops CFSs (38). Here, we found an increased accumulation of FANCD2 might be a common feature. The observation that trabectedin foci after trabectedin treatment that, importantly, was fully res- toxicity is augmented by the dysfunction in pathway that have cued by RNase H1 overexpression (Fig. 1C). This phenotype was been shown to protect both from cancer and from R-loops different to that caused by other genotoxic agents. Thus, contrary accumulation, such as the FA pathway and the BRCA1 and BRCA2 to trabectedin, overexpression of RNase H1 in cells treated with tumor suppressor genes (14, 15, 17, 18), supports the idea that R- the radiomimetic agent NCZ exacerbated FANCD2 foci accumu- loops might be a hallmark of specific types of cancer (40). It has lation (Supplementary Fig. S2). This may be explained if FANCD2 been shown that cancer patients with BRCA mutations exhibit foci induced by NCZ are not RNA–DNA hybrid-dependent, and improved clinical response to trabectedin (16). It would be RNase H1 overexpression causes an additional type of replication interesting in the future to check if R-loops are increased in stress. This finding adds further support to the conclusion that tumoral cells derived from such patients to strengthen our trabectedin has a prominent and specific action on R-loops where hypothesis that a subset of tumors characterized by an higher it preferentially causes DNA damage and breaks able to block accumulation of R-loops are more suitable to be treated with replication fork progression. Consistently, using a cell line system trabectedin. where R-loops were stabilized by the ectopic expression of the RNA–DNA hybrid binding domain of the RNase H1 enzyme Disclosure of Potential Conflicts of Interest fused to GFP, and additionally increasing the basal levels of R- No potential conflicts of interest were disclosed. loop through the siRNA-mediated depletion of THOC1 or BRCA2, we determined that cells with higher levels of R-loops Authors' Contributions were more sensitive to trabectedin treatment (Fig. 3). Conception and design: E. Tumini, C.M. Galmarini, A. Aguilera Importantly, we were able to recapitulate the phenotype of R- Development of methodology: E. Tumini, E. Herrera-Moyano, M.S. Martín- Alonso, S. Barroso loop–dependent genome instability caused by trabectedin and Analysis and interpretation of data (e.g., statistical analysis, biostatistics, lurbinectedin in yeast cells. Rad52 foci induced after treatment of computational analysis): E. Tumini, E. Herrera-Moyano, M.S. Martín-Alonso, either trabectedin or lurbinectedin were rescued by RNase H1 S. Barroso, overexpression (Fig. 5). This result strongly suggests that the Writing, review, and/or revision of the manuscript: E. Tumini, C.M. Galmarini, preferred action of these antitumoral agents on DNA is indepen- A. Aguilera dent of the eukaryotic system or cell type, arguing strongly in favor Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): A. Aguilera of a preferential action on the R-loop–containing chromatin. Study supervision: A. Aguilera In HeLa cells exposed to these agents, the RNA–DNA hybrid- dependent DNA damage phenotype was not so clear as it was in Acknowledgments yeast cells. However, we were able to show by comet assay and The authors thank Helene Gaillard and Oksana Brehey for assistance in 53BP1 immunofluorescence that the DNA damage induced by setting up yeast experiments. This work was supported with funding by the trabectedin or lurbinectedin was partially rescued by RNase H1 European Research Council (ERC2014 AdG669898 TARLOOP), the Spanish overexpression (Fig. 4; Supplementary Fig. S1). Besides, we found Ministry of Economy and Competitiveness (BFU2016-75058-P), the European that gH2AX foci accumulation after trabectedin treatment was Union (FEDER; BFU2016-75058-P), and PharmaMar (PRJ201402213). further increased in cells overexpressing RNase H1 (Supplemen- The costs of publication of this article were defrayed in part by the tary Fig. S5A). However, even if phosphorylated gH2AX on Ser139 payment of page charges. This article must therefore be hereby marked is used as a DSB marker, it also denotes replication stress, because advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate it is extended through a large chromatin domain as an early this fact. response to replication fork stalling in an ATR-dependent process (39). RNase H1 overexpression could cause replication stress and Received June 1, 2018; revised October 19, 2018; accepted November 30, instability, likely because it hampers replication by altering Oka- 2018; published first December 14, 2018.

References 1. Gordon EM, Sankhala KK, Chawla N, Chawla SP. Trabectedin for soft 4. Aune GJ, Takagi K, Sordet O, Guirouilh-Barbat J, Antony S, Bohr VA, et al. tissue sarcoma: current status and future perspectives. Adv Ther 2016; Von Hippel-Lindau-coupled and transcription-coupled nucleotide exci- 33:1055–71. sion repair-dependent degradation of RNA polymerase II in response to 2. Teplinsky E, Herzog TJ. The efﬁcacy of trabectedin in treating ovarian trabectedin. Clin Cancer Res 2008;14:6449–55. cancer. Expert Opin Pharmacother 2017;18:313–23. 5. Casado JA, Rio P, Marco E, Garcia-Hernandez V, Domingo A, Perez L, et al. 3. Minuzzo M, Marchini S, Broggini M, Faircloth G, D'Incalci M, Relevance of the Fanconi anemia pathway in the response of human cells to Mantovani R. Interference of transcriptional activation by the anti- trabectedin. Mol Cancer Ther 2008;7:1309–18. neoplastic drug ecteinascidin-743. Proc Natl Acad Sci U S A 2000; 6. Leal JF, Martinez-Diez M, Garcia-Hernandez V, Moneo V, Domingo A, 97:6780–4. Bueren-Calabuig JA, et al. PM01183, a new DNA minor groove covalent

www.aacrjournals.org Mol Cancer Res; 17(3) March 2019 781

Tumini et al.

binder with potent in vitro and in vivo anti-tumour activity. Br J Pharmacol 23. Dominguez-Sanchez MS, Barroso S, Gomez-Gonzalez B, Luna R, Aguilera 2010;161:1099–110. A. Genome instability and transcription elongation impairment in human 7. Romano M, Frapolli R, Zangarini M, Bello E, Porcu L, Galmarini CM, et al. cells depleted of THO/TREX. PLoS Genet 2011;7:e1002386. Comparison of in vitro and in vivo biological effects of trabectedin, 24. ten Asbroek AL, van Groenigen M, Nooij M, Baas F. The involvement of lurbinectedin (PM01183) and Zalypsis(R) (PM00104). Int J Cancer human ribonucleases H1 and H2 in the variation of response of cells to 2013;133:2024–33. antisense phosphorothioate oligonucleotides. Eur J Biochem 2002; 8. Santamaria Nunez G, Robles CM, Giraudon C, Martinez-Leal JF, Compe E, 269:583–92. Coin F, et al. Lurbinectedin specifically triggers the degradation of phos- 25. Tumini E, Barroso S, Calero CP, Aguilera A. Roles of human POLD1 and phorylated RNA polymerase II and the formation of DNA breaks in cancer POLD3 in genome stability. Sci Rep 2016;6:38873. cells. Mol Cancer Ther 2016;15:2399–412. 26. Bianco JN, Poli J, Saksouk J, Bacal J, Silva MJ, Yoshida K, et al. Analysis of 9. Germano G, Frapolli R, Belgiovine C, Anselmo A, Pesce S, Liguori M, et al. DNA replication profiles in budding yeast and mammalian cells using Role of macrophage targeting in the antitumor activity of trabectedin. DNA combing. Methods 2012;57:149–57. Cancer Cell 2013;23:249–62. 27. Lisby M, Rothstein R, Mortensen UH. Rad52 forms DNA repair and 10. Belgiovine C, Bello E, Liguori M, Craparotta I, Mannarino L, Paracchini L, recombination centers during S phase. Proc Natl Acad Sci U S A 2001; et al. Lurbinectedin reduces tumour-associated macrophages and 98:8276–82. the inflammatory tumour microenvironment in preclinical models. Br J 28. Castellano-Pozo M, Santos-Pereira JM, Rondon AG, Barroso S, Andujar E, Cancer 2017;117:628–38. Perez-Alegre M, et al. R loops are linked to histone H3 S10 phosphorylation 11. Aguilera A, Garcia-Muse T. Causes of genome instability. Annu Rev Genet and chromatin condensation. Mol Cell 2013;52:583–90. 2013;47:1–32. 29. Gari E, Piedrafita L, Aldea M, Herrero E. A set of vectors with a 12. Santos-Pereira JM, Aguilera A. R loops: new modulators of genome tetracycline-regulatable promoter system for modulated gene expres- dynamics and function. Nat Rev Genet 2015;16:583–97. sion in Saccharomyces cerevisiae. Yeast 1997;13:837–48. 13. Skourti-Stathaki K, Proudfoot NJ. A double-edged sword: R loops as threats 30. Kim KH, Kwon BM, Myers AG, Rees DC. Crystal structure of neocarzinos- to genome integrity and powerful regulators of gene expression. Genes Dev tatin, an antitumor protein-chromophore complex. Science 1993;262: 2014;28:1384–96. 1042–6. 14. Bhatia V, Barroso SI, Garcia-Rubio ML, Tumini E, Herrera-Moyano E, 31. D'Incalci M, Galmarini CM. A review of trabectedin (ET-743): a unique Aguilera A. BRCA2 prevents R-loop accumulation and associates with mechanism of action. Mol Cancer Ther 2010;9:2157–63. TREX-2 mRNA export factor PCID2. Nature 2014;511:362–5. 32. Gan W, Guan Z, Liu J, Gui T, Shen K, Manley JL, et al. R-loop-mediated 15. Hill SJ, Rolland T, Adelmant G, Xia X, Owen MS, Dricot A, et al. genomic instability is caused by impairment of replication fork progres- Systematic screening reveals a role for BRCA1 in the response to sion. Genes Dev 2011;25:2041–56. transcription-associated DNA damage. Genes Dev 2014;28:1957–75. 33. Wellinger RE, Prado F, Aguilera A. Replication fork progression is impaired 16. Monk BJ, Lorusso D, Italiano A, Kaye SB, Aracil M, Tanovic A, et al. by transcription in hyperrecombinant yeast cells lacking a functional Trabectedin as a chemotherapy option for patients with BRCA deficiency. THO complex. Mol Cell Biol 2006;26:3327–34. Cancer Treat Rev 2016;50:175–82. 34. Salas-Armenteros I, Perez-Calero C, Bayona-Feliu A, Tumini E, Luna R, 17. Garcia-Rubio ML, Perez-Calero C, Barroso SI, Tumini E, Herrera-Moyano E, Aguilera A. Human THO-Sin3A interaction reveals new mechanisms Rosado IV, et al. The Fanconi anemia pathway protects genome integrity to prevent R-loops that cause genome instability. EMBO J 2017;36: from R-loops. PLoS Genet 2015;11:e1005674. 3532–47. 18. Schwab RA, Nieminuszczy J, Shah F, Langton J, Lopez Martinez D, Liang 35. Hamperl S, Bocek MJ, Saldivar JC, Swigut T, Cimprich KA. Transcription- CC, et al. The Fanconi Anemia pathway maintains genome stability by replication conflict orientation modulates R-loop levels and activates coordinating replication and transcription. Mol Cell 2015;60:351–61. distinct DNA damage responses. Cell 2017;170:774–86. 19. Marco E, Garcia-Nieto R, Mendieta J, Manzanares I, Cuevas C, Gago F. A 3. 36. Lang KS, Hall AN, Merrikh CN, Ragheb M, Tabakh H, Pollock AJ, et al. (ET743)-DNA complex that both resembles an RNA-DNA hybrid and Replication-transcription conflicts generate R-loops that orchestrate mimicks zinc finger-induced DNA structural distortions. J Med Chem bacterial stress survival and pathogenesis. Cell 2017;170:787–99. 2002;45:871–80. 37. Ceccaldi R, Sarangi P, D'Andrea AD. The Fanconi anaemia pathway: 20. Sollier J, Stork CT, Garcia-Rubio ML, Paulsen RD, Aguilera A, Cimprich KA. new players and new functions. Nat Rev Mol Cell Biol 2016;17: Transcription-coupled nucleotide excision repair factors promote R-loop- 337–49. induced genome instability. Mol Cell 2014;56:777–85. 38. Madireddy A, Kosiyatrakul ST, Boisvert RA, Herrera-Moyano E, Garcia- 21. Herrero AB, Martin-Castellanos C, Marco E, Gago F, Moreno S. Cross-talk Rubio ML, Gerhardt J, et al. FANCD2 facilitates replication through between nucleotide excision and homologous recombination DNA repair common fragile sites. Mol Cell 2016;64:388–404. pathways in the mechanism of action of antitumor trabectedin. Cancer Res 39. Sirbu BM, Couch FB, Feigerle JT, Bhaskara S, Hiebert SW, Cortez D. Analysis 2006;66:8155–62. of protein dynamics at active, stalled, and collapsed replication forks. 22. Wongsurawat T, Jenjaroenpun P, Kwoh CK, Kuznetsov V. Quantitative Genes Dev 2011;25:1320–7. model of R-loop forming structures reveals a novel level of RNA-DNA 40. Gaillard H, Garcia-Muse T, Aguilera A. Replication stress and cancer. interactome complexity. Nucleic Acids Res 2012;40:e16. Nat Rev Cancer 2015;15:276–89.

782 Mol Cancer Res; 17(3) March 2019 Molecular Cancer Research

The Antitumor Drugs Trabectedin and Lurbinectedin Induce Transcription-Dependent Replication Stress and Genome Instability

Emanuela Tumini, Emilia Herrera-Moyano, Marta San Martín-Alonso, et al.

Mol Cancer Res 2019;17:773-782. Published OnlineFirst December 14, 2018.

Updated version Access the most recent version of this article at: doi:10.1158/1541-7786.MCR-18-0575

Supplementary Access the most recent supplemental material at: Material http://mcr.aacrjournals.org/content/suppl/2018/12/13/1541-7786.MCR-18-0575.DC1

Cited articles This article cites 40 articles, 14 of which you can access for free at: http://mcr.aacrjournals.org/content/17/3/773.full#ref-list-1

Citing articles This article has been cited by 1 HighWire-hosted articles. Access the articles at: http://mcr.aacrjournals.org/content/17/3/773.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://mcr.aacrjournals.org/content/17/3/773. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.