Trabectedin triggers direct and NK-mediated cytotoxicity in multiple myeloma
Genomic instability is a feature of multiple myeloma (MM), and impairment in DNA damaging response (DDR) has an established role in disease pathobiology. Indeed, a deregulation of DNA repair pathways may contribute to genomic instability, to the establishment of drug resistance to genotoxic agents, and to the escape from immune surveillance. On these bases, we evaluated the role of different DDR pathways in MM and investigated, for the first time, the direct and immune-mediated anti-MM activity of the nucleotide excision repair (NER)-dependent agent trabectedin.
Gene-expression profiling (GEP) was carried out with HTA2.0 Affymetrix array. Evaluation of apoptosis, cell cycle, and changes in cytokine production and release have been performed in 2D and 3D Matrigel-spheroid models through flow cytometry on MM cell lines and patients-derived primary MM cells exposed to increasing nanomolar concentrations of trabectedin. DNA-damage response has been evaluated through Western blot, immunofluorescence, and DNA fragmentation assay. Trabectedin-induced activation of NK has been assessed by CD107a degranulation. miRNAs quantification has been done through RT-PCR.
By comparing GEP meta-analysis of normal and MM plasma cells (PCs), we observed an enrichment in DNA NER genes in poor prognosis MM. Trabectedin triggered apoptosis in primary MM cells and MM cell lines in both 2D and 3D in vitro assays. Moreover, trabectedin induced DDR activation, cellular stress with ROS production, and cell cycle arrest. Additionally, a significant reduction of MCP1 cytokine and VEGF-A in U266-monocytes co-cultures was observed, confirming the impairment of MM-promoting milieu. Drug-induced cell stress in MM cells led to upregulation of NK activating receptors ligands (i.e., NKG2D), which translated into increased NK activation and degranulation. Mechanistically, this effect was linked to trabectedin-induced inhibition of NKG2D-ligands negative regulators IRF4 and IKZF1, as well as to miR-17 family downregulation in MM cells.
Taken together, our findings indicate a pleiotropic activity of NER-targeting agent trabectedin, which appears a promising candidate for novel anti-MM therapeutic strategies.
KeywordsMyeloma 3D-models Natural killer Micro-RNAs Trabectedin
Alternative non-homologous and joining
Base excision repair
Classical non-homologous and joining
Relating Clinical Outcomes in MM to Personal Assessment of Genetic Profile
DNA damage response
Fetal bovine serum
Gene expression profiling
Gene set enrichment analysis
Mitochondrial membrane potential
Nucleotide excision repair
Normalized enrichment score
Peripheral blood mononuclear cells
Reactive oxygen species
Real-time polymerase chain reaction
Multiple myeloma (MM) is still an incurable hematologic malignancy characterized by clonal proliferation of malignant plasma cells (PCs) within the bone marrow (BM). Current MM therapy includes triple- or double-drug combination, based on proteasome inhibitors (PIs) and/or immune-modulatory drugs (IMiDs) plus dexamethasone, with or without chemotherapeutic agents . Autologous stem cell transplant is reserved to selected patients as consolidation following induction treatment. However, despite recent advancements that significantly improved clinical outcome, patients invariably progress to drug resistance.
DNA repair mechanisms have a crucial role for the maintenance of the genome integrity, and their activation is fine tuned to resolve specific DNA damages. Currently, at least seven DNA repair active systems have been described in MM as protection from different DNA lesions . Specifically, base excision repair (BER), nucleotide excision repair (NER), and mismatch repair (MMR) pathways are involved in the repair of single-strand DNA damages; homologous recombination (HR), classical non-homologous end joining (c-NHEJ), and alternative NHEJ (a-NHEJ) pathways are conversely involved in double-strand breaks (DSBs), while Fanconi anemia pathway (together with NER and HR) is involved in the repair of interstrand crosslinks [2, 3]. Dysregulation of these systems has been found to promote tumor progression, cell survival, and development of drug resistance [2, 3, 4]. Furthermore, activation of DNA damage response (DDR) has been involved in the upregulation of ligands for activating receptors of natural killer (NK) lymphocytes. Indeed, besides participating in cell cycle control and induction of apoptosis, DDR works as a sensor for cellular stress or transformation, inducing recognition by the immune system [5, 6].
Genomic instability is a major hallmark of MM and most of the drugs currently used in the treatment of MM have direct genotoxic activity (i.e., melphalan, doxorubicin, cyclophosphamide) or interfere with the DNA repair machinery (PIs or IMiDs) . Accordingly, these drugs have been reported to trigger the expression of DNAM-1 and NKG2D ligands on MM cells and to induce NK cells activation [7, 8].
Herein, the expression and prognostic relevance of genes of DNA repair pathways in MM has been investigated. Since overexpression of NER pathway has been found, evaluation of the direct and immune-mediated anti-MM activity of the NER-targeting agent trabectedin in 2D and 3D experimental models of MM has been performed.
Cell lines, MM primary cells, and drugs
Multiple myeloma cell lines were cultured at 37 °C with 5% CO2. AMO-1, U266, and NCI-H929, SKMM1 were purchased from DSMZ (Braunschweig, Germany). AMO-BZB and AMO-CFZ were kindly provided by Dr. Christoph Driessen (Eberhand Karls University, Tübingen Germany), MM1S and RPMI-8226 were purchased from ATCC (Manassas, VA, USA), and OPM2 and RPMI-8226 DOX40 were kindly provided by Dr. K.C. Anderson (Dana-Farber Cancer Institute, Harvard Medical School, Boston, MA, USA). All these cells were cultured in RPMI-1640 medium (Gibco, Life Technologies) supplemented with 10% heat-inactivated fetal bovine serum (FBS) and 100 U/mL penicillin, and 100 μg/mL streptomycin (GIBCO; Thermo Fischer, Carlsbad, CA). U266 and U266 LR7 (kindly provided by Dr. A. Pandiella, Instituto de Biología Molecular y Celular del Cáncer, CSIC-Universidad de Salamanca, Salamanca, Spain) were cultured in RPMI-1640 with 20% FBS and 100 U/mL penicillin, and 100 μg/mL streptomycin. JJN3 were purchased from DSMZ and were cultured in Dulbecco’s modified Eagle medium supplemented with 20% FBS and 100 U/mL penicillin, and 100 μg/mL streptomycin.
NK-92 CI were obtained from NantKwest (Dr. Kerry S. Campbell) and cultured in alpha-MEM medium with ribonucleosides and deoxyribonucleosides (Gibco, Thermo Scientific) supplemented with 10% horse serum, 10% FBS, 0.2 mM myo-inositol (Sigma), 2 mM l-glutamine, 0.1 mM β-mercaptoethanol (Sigma), 0.002 mM folic acid (Fisher Scientific), 1x NEAA (Gibco, Thermo Scientific), 1 mM Na pyruvate (Gibco, Thermo Scientific), 100 U/mL penicillin, and 100 μg/mL streptomycin (Life Technologies) and 100 IU/mL of recombinant human IL-2 (IL-2 improved sequence, Miltenyi).
Primary MM cells were immune-magnetically sorted by using CD138 MicroBeads (MACS, Miltenyi, according to producers’ guidelines) from leftover samples of three MM patients’ bone marrow aspirates after all diagnostic procedures. All patients had provided the informed consent according to institutional bioethical standards, and all the samples have been anonymized before use (institutional approval n. 120/2015, within the project Innovative immunotherapeutic treatments of human cancer, MultiUnit—Multi Unit Regional n.16695).
Human monocytes were obtained by immune magnetical separation from healthy donor peripheral blood mononuclear cells (PBMCs) after Ficoll-Paque density-gradient separation. Specifically, BD IMag™ anti-human CD14 magnetic particles were used for positive selection of CD14+ monocytes according to producers’ guidelines.
Ascorbic acid, as antioxidant agent, was purchased from Sigma-Aldrich (Saint Louis, USA).
Trabectedin (PharmaMar, Madrid, Spain) was reconstituted in DMSO.
Virus generation and transduction of MM cells
MM cells stably expressing miR-17-92 cluster have been gently provided by Dr. E. Morelli. These cells have been obtained by using a PMIRH17-92PA-1 lenti-vector (System Biosciences, Palo Alto, CA, USA) through a methodology already described elsewhere [9, 10]. Briefly, packaging of the miR-17-92 cluster constructs in pseudoviral particles was performed in 293Ta cells using the Lenti-Pac FIV Expression Packaging Kit (FPKLvTR-20), according to the manufacturer’s instructions (Genecopoeia, Rockville, MD, USA). After transfection of 293 T cells, supernatants containing miR-17-92 lentivirus were collected at 8-h intervals, filtered, and used for two rounds of transduction of U266 cells (1 × 106) in the presence of 8 mg/mL of polybrene (Sigma-Aldrich). Two days after transduction, selection with 1 μg/ml puromycin for 3 days was performed to achieve almost 100% transduced cells. Empty lentivirus transduced cells were used as a control for the experiments.
MM cells (2 × 105) were treated with dose escalation of trabectedin (0-0.1-0.25-0.5-1-2.5 nM) and analyzed for apoptosis after 24, 48, and 72 h through Annexin V/7-AAD flow cytometry assay (Becton Dickinson). All experiments have been performed at least three times. Primary MM cells were exposed to 2.5 nM of Trabectedin and apoptosis was evaluated after 24 h. Apoptosis was further investigated at molecular level, analyzing the cleaved/total levels of caspase 3 and PARP by Western blot.
Cell cycle analysis
U266 and MM1S (1X106) were cultured in the presence of trabectedin (1 nM and 0.1 nM, respectively) in 6-wells plate for 48 h. Cells were then collected and washed twice with PBS 1X. Subsequently, 1 mL of 70% ice cold ethanol for each sample was added. Cells were stored at − 20 °C until used and then centrifuged and washed twice with PBS 1X. Cells were then resuspended in 1 mL of PI staining solution (100 μg/mL of ribonuclease A, 50 μg/mL of propidium iodide, and 0.01% of NP-40) and incubated for 1 h at room temperature, in the dark. Analysis was performed with flow cytometer and repeated three times.
We established a 3D in vitro model of MM cell lines alone or in the presence of human monocytes using Matrigel® matrix (Corning). Briefly, 1 × 105 MM cells (U266, OPM2, MM1S) alone or in co-culture with 0.5 × 105 CD14+ monocytes (2:1 MM/monocytes ratio) were resuspended in ice-cold matrigel and a matrigel drop of 35 μL was placed in 24-wells plate coated with a sterile parafilm dish to form a Matrigel-spheroid. After 30 min of incubation at 37 °C, 500 μL of medium with different concentrations of trabectedin was added to each well and the spheroids were incubated for 72 h. Matrigel-spheroids were then resuspended in Dispase (Sigma-Aldrich) and the recovered cells were stained with annexin-V/ 7AAD for analyzing apoptosis induction by flow-cytometry. Supernatants were collected to analyze cytokines expression. Alternatively, Matrigel-spheroids were stored for immunohistochemistry evaluation. All experiments have been performed at least three times.
Matrigel-spheroids of either tumor cells alone or co-cultured with CD14+ monocytes isolated from healthy donors were fixed in 0.3% glutaraldehyde, then in 4.21% formaldehyde, and subsequently paraffin-embedded. Serial section of 4-μm-thick were cut and mounted on acid-cleaned glass slides, which were dewaxed with xylene, and processed for hematoxylin-eosin staining and immunohistochemistry.
Slides were incubated overnight at 4 °C with anti-g-H2ax monoclonal antibody (Cell Signaling) and anti-cleaved caspase 3 (Santa Cruz Technologies) primary antibodies, washed with PBS three times and incubated with appropriate chromogen-conjugated secondary antibody for 1 h at room temperature. After washings using PBS, samples were observed by an optical microscope and images were acquired.
Single-cell gel electrophoresis (Comet) assay
Comet assay (Trevigen) was performed according to manufacturer’s instructions. Briefly, cells were harvested (1 × 105 cells per pellet), mixed with 200 mL low-melting agarose, and layered onto agarose-coated glass slides. The slides were immersed in lysis solution, and then placed into a horizontal electrophoresis apparatus filled with fresh alkaline or neutral electrophoresis buffer. After electrophoresis (30 min at 1 V/cm tank length), air-dried and neutralized slides were stained with Dapi and kept in a moist chamber in the dark at 4 °C. Images were acquired at × 63 oil immersion with an SP2 Leica Zeiss confocal laser-scanning microscope.
Mitochondrial membrane potential and ROS/superoxide analysis
MM cells (5 × 105) were seeded in 12-wells plate and were incubated for 24 h, untreated or treated with sub-lethal doses of trabectedin (depending on cell line), in the presence or absence of ascorbic acid (25 μM), as antioxidant agent. Trabectedin-induced changes in the production of mitochondrial membrane potential (MMP) and radical oxygen species (ROS) were evaluated by MitoScreen assay (Becton Dickinson) and Total ROS/Superoxides Detection kit (ENZO Life Sciences) respectively, by flow cytometry according to producer’s guidelines.
Flow cytometry and degranulation assay
The expression of the NKG2D and DNAM-1 ligands on different MM cells was evaluated, after 48 h of culture in the presence of trabectedin, by using fluorochrome-conjugated antibodies against MIC-A/B (Becton Dickinson), ULBP 1 (R&D Systems), ULBP 2-5-6 (R&D Systems), PVR (R&D Systems), and NECTIN-2 (Becton Dickinson) according to producer’s guidelines.
NK cell degranulation was evaluated using the CD107a staining. Specifically, trabectedin-treated MM cell lines were washed twice in complete medium and incubated with NK-92 CI cell line at effector/target (E:T) ratio of 1:1, in a U-bottom 96-well plate in complete medium at 37 °C and 5% CO2 in the presence of anti-CD107a/PE (Becton Dickinson) for 2 h. Cells were then stained with anti-CD3/PcP and anti-CD56/APC to identify NK cell population. NK cells positive for CD107a were considered as degranulating/activated cells able to induce cytotoxicity.
All experiments were acquired by an ATTUNE Nxt (Thermo Scientific) flow cytometer. For each sample, at least 1 × 104 events in the gate of interest were acquired.
RNA extraction and quantitative real-time-PCR
Total RNA from MM cells was prepared with TRIzol® Reagent (Life Technologies) according to manufacturer’s instructions. The integrity and quantity of total RNA was assessed using the NanoDrop Spectrophotometer (Thermo Scientific). The single-tube TaqMan miRNA assay (Life Technologies) was used to detect and quantify mature miR-17, miR-18a, miR-19a, miR-19b, miR-20a, miR-92a, performing a real-time polymerase chain reaction (RT-PCR) using TaqMan®Fast Universal PCR Master Mix on a ViiA7 RT reader (Life Technologies). MiRNAs expression was normalized on the RNU44 snoRNA (Life Technologies). Comparative RT–PCR was performed in triplicate, including no-template controls. Relative expression was calculated by using the ∆∆-cycle threshold (CT) method .
U226 MM cells (3 × 106), obtained from two different experiments, were treated with PBS or 2.5 μM of trabectedin for 24 h. Gene expression profiling was performed as described elsewhere . Briefly, total RNA (tRNA) was extracted through column purification with RNeasy kit (Qiagen, Hilden, Germany). A total of 300 ng RNA was used as starting material for preparing the hybridization target by using the Ambion® WT Expression Kit (Ambion, Life Technologies). The integrity, quality, and quantity of tRNA were assessed by the Agilent Bioanalyzer 2100 (Agilent Technologies, Santa Clara, CA) and NanoDrop 1000 Spectrophotometer (Thermo Scientific, Wilmington, DE). The amplification of cRNA, the cleanup, and the fragmentation were performed according to the Affymetrix’s procedures. Microarray data were generated by Human transcriptom array 2.0 ST (Affymetrix Inc., Santa Clara, Ca). Arrays were scanned with an Affymetrix GeneChip Scanner 3000. Raw data produced by the Affymetrix Platform (i.e., CEL files) were processed and RMA normalized using Affymetrix Expression Console (EC). Data set has been deposited under the GEO accession number GSE128020.
Gene set enrichment analysis and gene ontology
We used the gene set enrichment analysis (GSEA)  tool to enrich the target pathways with statistically significant differences between trabectedin treated versus untreated cells. Indeed, given a specific gene expression profile sorted by the expression ratio between the two conditions, the target pathway is considered significantly enriched if members are enriched in the top (up-regulated) or bottom (down-regulated) region of the profile. GSEA software then calculates an enrichment score (ES) by using a Kolmogorov-Smirnov test, to measure the degree to which the pathway is enriched in the top-ranked or bottom-ranked region of the profile. Next, ES is normalized (NES) according to the number of genes belonging to the pathways, in order to make comparable pathways with different size. Of note, this process takes into account the contribution of all genes included into the analysis, including those with minimum fold change. GSEA analysis parameters have been set as follows: number of permutations: 1000; permutation type: gene_set; metric for ranking genes: log2 ratio of Classes; size of genesets: 25–500 genes; gene sets evaluated: Hallmark gene sets and C2 curated gene sets from MSigDB.
Additionally, genes upregulated or downregulated with a fold change of at least 1.5 where analyzed with ClueGO, a Cytoscape plug in app that visualizes non-redundant biological terms for large clusters of gene sets in a functionally grouped network [14, 15].
Proteins were extracted from MM cells after lysing in NP40 CellLysis Buffer (Novex) containing a cocktail of protease and phosphatase inhibitors (Thermo Scientific, Waltham, MA). Whole cells lysates (20–30 μg/well) were loaded and separated on 4–12% NovexBis–Tris SDS–acrylamide gels or 3–8% Tris-Acetate Protein Gels (Gibco, Life Technologies). Proteins were then transferred on nitrocellulose membranes by Trans-Blot Turbo Transfer Starter System (Bio-Rad, Berkeley, CA). Subsequently, membranes were blotted with the following primary antibodies: anti-P21, anti-BCL2, anti-RAD51, anti-MCL1, anti-PRO-CASP3, anti-C-CASP3, anti-CDK6, anti-cyclinD1, anti-E2F1, anti-IKZF1, anti-XRCC1, anti-RPA32 (Cell Signaling) and anti-cPARP, anti-PARP, anti-BCL2, anti-IRF4, anti-PTEN, anti-XPF, anti-DDB2, anti-ERCC1, anti-actin, anti-vinculin, and GAPDH (Santa Cruz). To study major signaling checkpoints in response to active DNA damage, a specific commercial kit from Cell Signaling (DNA Damage Antibody Sampler Kit #9947) has been used, as already described in previous work by us and others [16, 17, 18]. The kit includes the following primary antibodies: anti-phospho (p)P53 (Ser15), anti-gammaH2AX (Ser139), anti-pATM (Ser1981), anti-pATR (Ser428), anti-pCHEK1 (Ser345), anti-pCHEK2 (Thr68), and anti-pBRCA1 (Ser1524). All these forms are virtually absent in normal conditions and are activated in response to DNA damage to induce an attempt to DNA repair and block cell cycle progression (through, for example, p21). GAPDH expression has been used as a protein loading control for this kit. Blots were then incubated with goat anti-mouse or goat anti-rabbit HRP-conjugated antibodies (Santa Cruz Biotechnology); immunoreactive bands were detected by use of enhanced chemiluminescence (ECL) method, acquired through the C-DIGIT scanner (LI-COR) and quantified by Image Studio Lite 5.0 (LI-COR).
Immunostaining for confocal microscopy
Trabectedin-treated and control MM cells were seeded onto glass coverslips and cytospin for 5 min at 800 rpm was performed. Cells were then washed in PBS, fixed in 4% paraformaldehyde for 12 min, washed three times with PBS, followed by permeabilization with 0.01% Triton-X for 15 min, and again washed in PBS containing 0.5% BSA. Cells were then incubated with anti-g-H2ax monoclonal antibody (cell signaling) overnight at 4 °C, washed with PBS three times, and incubated with Alexa-fluor 488-conjugated secondary antibody (Molecular Probes, Life Technologies, NY) for 1 h at room temperature. Glass coverslips were then washed three times with PBS and mounted with Vecta-Shield mounting media containing DAPI. Samples were visualized and images captured using a Leica microscope. Images were acquired at × 63 oil immersion with an SP2 Leica Zeiss confocal laser-scanning microscope.
A panel of different cytokines including IL1b, IL4, IL6, IL8, IL23, TNF, IFNg, G-CSF, IP10, MCP1, IL10, and VEGF were detected in the supernatants of CD14+ cells alone obtained by PBMCs of healthy donors and co-cultured with MM cells in 3D Matrigel-spheroids, in the presence or absence of Trabectedin, using BD CBA Human Soluble Protein Flex Set system (Becton Dickinson, Heidelberg, Germany). Samples from three different experiments were analyzed with an Attune Nxt Thermo Scientific flow cytometer.
Tube assay formation
A drop of 50 μL of Matrigel (CORNING) were used to coat 96-wells plates and allowed to polymerize at 37 °C for 30 min. Then, 15 × 103 HUVECs were seeded in each well and then 50 μL of conditioned medium from trabectedin-treated cells were added. After 1-h incubation at 37 °C, at least pictures of three representative fields per well were taken using phase contrast microscopy. The tubulogenic potential was quantified by estimating the total tube length and the number of nodal branchpoints (a single pixel connected to three or more pixels), through the “Pipeline 1.4” tool  (https://sourceforge.net/projects/pipelinetfaanalysis/). All experiments have been performed at least three times.
Gene expression datasets analysis
Datasets of gene expression profiling of MM retrieved from GEO database
Affymetrix Human Gene 1.0 ST Array (GPL6244)
Centro de Investigación del Cáncer de Salamanca
Analysis of plasma cells from patients with monoclonal gammopathy of undetermined significance (MGUS) (n = 20), smoldering multiple myeloma (sMM) (n = 33), symptomatic MM (n = 41), and healthy donors (n = 5).
Affymetrix Human Exon 1.0ST Array(GPL5175)
Dana-Farber Cancer Institute
Gene expression microarray datasets from CD138 purified plasma cells isolated from 170 patients with newly diagnosed MM and 6 healthy subjects. All patients received triple drug regime—Vincristine, Adriamycin, and Dexamethasone (VAD)—as induction therapy followed by autologous stem cell transplant (ASCT) as a maintenance therapy.
Affymetrix Human Genome U133A Array (GPL96)
Gene expression profile of CD138 purified plasma cells from 22 MGUS, 24 sMM, 69 newly diagnosed MM, 32 relapsed MM, and 15 healthy subjects. Each sample has been further characterized by FISH for the identification of hyperdiploidy.
Affymetrix Human Genome
U133A Array (GPL96)
University of Milan—Fondazione IRCCS Ospedale Maggiore Policlinico
This series of microarray experiments contains the gene expression profiles of immunomagnetically purified CD138+ plasma cells obtained from 5 normal donors, 11 MGUS, 133 MM, and 9 plasma cell leukemia at diagnosis
The CoMMpass (Relating Clinical Outcomes in MM to Personal Assessment of Genetic Profile) Trial (NCT0145429), a longitudinal study in MM, relating clinical outcomes to genomic and immune-phenotypic profiles of CD138+ selected plasma cells from the BM of newly diagnosed MM patients (in the release used in this work (interim analysis 8, IA8), RNA-seq, together with clinical data, was available for 549 MM patients).
Datasets including MM cell lines gene expression profiling were retrieved from GEO database with the accession code GSE68379 and GSE6205. These data were normalized in Transcription analysis console (TAC, Thermo Scientific) software and result table processed through R Studio (R version: 3.5).
Differences between means were analyzed by using GraphPad statistical package. The results were expressed as the mean ± SD of at least three different experiments, and the significance assessed by the two-tailed Student t test or Mann-Whitney test according to samples distribution. A p value of 0.05 or less was considered statistically significant. Overall survival (OS) and progression-free survival (PFS) analyses (Kaplan-Meier curves and log-rank test) have been performed by using SPSS statistical software on data retrieved by the CoMMpass database.
NER genes are highly expressed in MM patients and correlate with prognosis
The finding that upregulation of different NER genes in MM PCs was correlated to poor prognosis suggested a major role for NER in MM pathobiology. On this basis, the efficacy of NER-targeting agents in this setting was investigated. Therefore, the anti-MM activity of trabectedin (Ecteinascidin 743), whose mechanism of action relies on NER system expression , was studied. Specifically, this drug binds to the minor groove of DNA and traps the NER machinery as it attempts to repair DNA, leading to the generation of lethal DNA double strand breaks. To our knowledge, the antitumor activity of trabectedin in MM has not been described so far.
Trabectedin exerts potent anti-MM activity in vitro and in 3D models
To strengthen our findings, trabectedin was further studied by a new relevant 3D model in which MM cells grow in matrigel-based spheroids. In this model, U266 cells form multicellular niches recapitulating the bone marrow structure of MM patients. The addition of trabectedin to Matrigel-spheroids impaired the development of MM niches (Fig. 2d upper panel). The 3D architecture mimics the protective effect of BM tissue structure on MM cells from drug-induced cytotoxicity and, indeed, a slight reduction of trabectedin activity in this model, as compared to 2D classic cultures of several MM cell lines (U266, Fig. 2d; MM1S and OPM2, Additional file 3: Figure S2D), was observed. However, this model does not fully recapitulate the BM microenvironment (BMM). To overcome this limitation, taking into account that macrophages infiltrate the BMM and protect MM cells from DNA-damaging [26, 27], Matrigel-spheroids in which MM cell lines (U266, MM1S, and OPM2) were co-cultured with healthy donors-derived monocytes were generated. As expected, after 72 h, monocytes/macrophages greatly improved the proliferative activity and viability of MM cells, and strongly promoted the generation of multi-cellular “microenvironmental” niches. Nevertheless, trabectedin overcame monocyte-dependent protective effects and significantly induced apoptosis in all MM cell lines tested (U266, MM1S, and OPM2) (Fig. 2d lower panel and Additional file 3: Figure S2D).
Trabectedin modulates the transcriptome and induces cell cycle arrest, cellular stress, and DNA damage in MM cells
We then performed an in vitro validation of these bioinformatics findings.
It is well known that in other malignancies, trabectedin-induced apoptosis may occur in a P53-dependent or independent manner . Accordingly, in our experimental setting, an increase of markers of late apoptosis, cleaved (c)-PARP and c-caspase3, in both P53-WT MM1S and P53-mutated U266 cell lines after 48 h treatment (Fig. 3b left panel) was observed. By immunohistochemistry, an increase of c-caspase 3-positive U266 cells from our 3D model after trabectedin treatment (Fig. 3b right panel) was further confirmed. Moreover, in U266 cells, a stronger reduction in the anti-apoptotic protein BCL-2 than in MM1S was detected (Additional file 5: Figure S4A). This effect was paralleled by a deepest alteration in mitochondrial membrane polarization (Fig. 3c). Additionally, in both cell lines, ROS and superoxide production significantly increased, confirming the predicted activation of stress response pathways (Fig. 3d and Additional file 5: Figure S4B). Then, to investigate whether DNA damage and apoptosis were correlated with ROS production, both U266 and MM1S cell lines were co-treated with trabectedin and ascorbic acid as antioxidant. Notably, despite a significant reduction in ROS levels, the antioxidant did not protect MM cells from trabectedin-induced apoptosis and DNA damage, evaluated as H2AX protein phosphorylation (γ-H2AX) (Fig. 3d), demonstrating both effects (DNA damage and apoptosis) to be independent from ROS production (representative dot plots of U266 and MM1S MM cells reporting trabectedin activity in term of apoptosis in presence or absence of trabectedin and/or ascorbic acid, were showed in Additional file 5: Figure S4B). Unfortunately, the exact mechanism of trabectedin-associated ROS production (already observed in other cellular systems ) is currently not fully understood. We may hypothesize that, at least in our settings, ROS are produced as a consequence of DNA double-strand breaks, DNA damage response activation, and mitochondrial membrane potential alteration.
Next, cell cycle alterations upon trabectedin were evaluated. In line with bioinformatics analysis, a trabectedin-dependent S-phase arrest with progressive decrease of cells in G0/G1 phases was observed (Fig. 3e left panel). The scenario depicted by these results was further empowered by an increase in the expression of the cell-cycle inhibitor p21WAF1 in MM1S, and by downregulation of Cyclin D1 and CDK6 in U266 (Fig. 3e right panel).
Lastly, the DNA-damaging activity of trabectedin was studied. As a surrogate marker of DSBs, the expression of γ-H2AX, a variant of histone 2A that is phosphorylated in the presence of DNA DSBs, was evaluated. Specifically, H2AX has a key role in DNA repair machinery and its phosphorylation represents the first step to recruit DNA repair proteins . Additionally, g-H2AX is organized in foci that are associated to DNA fragmentation during apoptosis and DNA damage . By Western blot, a strong increase of phosphorylated H2AX in the presence of trabectedin (results for U266 and MM1S cell lines are reported in Fig. 3f, and for OPM2 in Additional file 5: Figure S4C) was observed. These findings were empowered by demonstration of γ-H2AX foci within the nuclei of U266 and MM1S cells by confocal microscopy and by immunohistochemistry in the 3D model of MM (Fig. 3f and Additional file 5: Figure S4D). Furthermore, DNA fragmentation by a COMET assay demonstrated that trabectedin induces DNA damage and loss of chromatin organization in both U266 and MM1S (Fig. 3f). Accordingly, in both cell lines, trabectedin induced increased expression of proteins involved in DDR, including pATM and pATR, their downstream mediators pChk1 and pChk2, RAD51, pBRCA1, and pP53 (the latter only in MM1S) (Fig. 3f).
Trabectedin reduces MM-macrophages-induced angiogenesis
Trabectedin increases the expression of NKG2D ligands in MM cells and induces NK activation
Trabectedin activity is dependent from perturbation of transcription factors and microRNAs networks
To investigate the transcriptional mechanisms underlying trabectedin-induced MICA/B upregulation in MM cells, we explored in our GEP analysis of U266 cells the mRNA expression of known transcription factors involved in MICA/B transcriptional modulation . Among MICA/B-positive regulators, only E2F1 was significantly upregulated, while, among negative regulators, IRF4 and IKZF1 were downregulated in MM cell lines treated with trabectedin as compared to control (Fig. 5c). Additionally, we further observed a trabectedin-dependent increase of the transcription of CRBN, CUL4, and DDB2 genes, all involved in the ubiquitination of IKZF1 protein (Additional file 5: Figure S4G).
E2F1, IRF4, and IKZF1 modulations were confirmed at protein level in U266 and OPM2 cell lines after 48-h exposure to trabectedin (Fig. 5c). Interestingly, while statistically significant, IKZF1 downregulation is far less pronounced in OPM2 than in U266 cell line, while E2F1 upregulation is stronger in OPM2. This could reflect a different weight of the two signaling in NKG2D ligands upregulation.
Moreover, to explore the post-transcriptional mechanisms correlated to trabectedin exposure in the regulation of MICA/B expression, we sought to examine whether trabectedin could regulate the expression of miRNAs. Indeed, by targeting the 3′ untranslated region (3′-UTR) of different mRNAs at the same time and, consequently, modulating their translation into proteins, miRNAs have a pivotal role in regulating almost all cellular function, including immune response. Indeed, increasing evidence reports on their capability of modulating, among others, the susceptibility of tumor cells to NK cell-mediated lysis [35, 36, 37]. On this basis, the miRNA target prediction tool mirDIP (http://ophid.utoronto.ca/mirDIP) has been used to identify miRNAs predicted to target, at the same time, MICA, MICB, and E2F1 mRNAs (which we previously demonstrated, in our experiments, to be upregulated after trabectedin treatment). As shown in Fig. 5d, all these miRNAs belong to well-characterized families, namely miR-17, miR-302, miR-372, and C19MC, all part of an oncogenic signaling network characterized by the presence of an AAGUGC sequence . Interestingly, all these miRNAs modulate DNA damage response and cell cycle progression.
Taking into account the prominent role of miR-17-92 cluster in MM [10, 39, 40, 41] and the already reported capability of miR-17 family to modulate the expression of NKG2D ligands [42, 43, 44], the effects of trabectedin on the expression levels of miR-17 and miR-20a (the two miRNAs of the miR-17-92 cluster belonging to the miR-17 family) has been evaluated. As described in Fig. 5e and in Additional file 5: Figure S4H, trabectedin exposure significantly downregulated pri-miR-17-92, miR-17, and miR-20a in both U266 and OPM-2 cell lines. Importantly, as a control for the selectivity of trabectedin activity, two additional mature miRNAs of the miR-17-92 cluster (miR-19b and miR-92a), but not belonging to the miR-17 family, were not modulated after treatment (Additional file 5: Figure S4I).
Next, to assessed the role of these miRNAs in the regulation of NKG2DL expression, U266 cells were lentivirally transduced to generate a pri-MiR-17-92 overexpressing MM cell line (17-92-U266) (Additional file 5: Figure S4J, reports expression levels of mature miRNAs after transduction). Interestingly, 17-92-U266 cells demonstrated a lower (but not significantly) baseline expression of MICA/B respect to empty vector U266 cells (EV-U266), which levels increased significantly less after trabectedin treatment (Fig. 5f). We further observed an increased resistance to trabectedin-induced direct cytotoxicity in 17-92-U266 cells as compared to empty vector: as shown in Fig. 5g and in Additional file: Figure S4K, 17-92-U266 cells were significantly more resistant to trabectedin-induced apoptosis as compared to EV-U266 cells. Overall, these findings indicate a role for miR-17 family in trabectedin-dependent NKG2D ligands upregulation in MM cells.
In this work, through a gene expression dataset meta-analysis, we demonstrated that among DNA repair systems, NER is the most upregulated in MM and it is strongly associated with patients’ prognosis. On these bases, for the first time, we decided to evaluate the activity of NER-targeting agent trabectedin in this setting. Trabectedin has been granted approval for the treatment of advanced soft tissue sarcoma and for relapsed ovarian cancer, while several studies are ongoing to evaluate its activity in other malignancies . In our experimental models, it exerts strong anti-myeloma activity on cell lines and primary cells at nanomolar concentrations, both in conventional 2D and in advanced 3D models. Interestingly, a quick response to trabectedin was associated to a high expression of ERCC1 protein, thus confirming its already known role in trabectedin mechanism of action [22, 23]. Additionally, trabectedin showed a pleiotropic activity in MM, which includes DNA DSBs generation, cell cycle arrest, exacerbation of cellular stress, reduction of angiogenesis, and immunomodulation (Fig. 5h). Regarding the latters, two different aspects have been of our interest. Several studies showed that trabectedin has significant effects on tumor microenvironment by impairing tumor-associated macrophages [32, 33]. Accordingly, in 3D Matrigel-spheroids, monocytes promoted MM cells viability and proliferation, which are strongly reduced by trabectedin. In the same setting, trabectedin modulated the pro-inflammatory cytokine/chemokine network by reducing MCP1, VEGF-A, and IL-10 which functionally translated into a decrease of pro-angiogenic potential. These effects are of particular relevance taking into account the role of inflammation and angiogenesis in MM [12, 45]. A relevant finding is, in our opinion, the triggering activity of trabectedin on the innate immune response against MM by increased expression of NKG2D ligands MICA/B and ULBP1. The relevance of NK response in MM pathogenesis has been deeply investigated [37, 46, 47], and in patients, a downregulation of surface expression of MICA on malignant plasma cells or a decline in NK-dependent immune-surveillance has been observed when MGUS progresses to symptomatic MM [48, 49]. Furthermore, the expression of killer cell inhibitory receptor (KIR) ligands, such as MHC class I, increases in advanced disease, impairing the balance between stimulatory and inhibitory signaling to promotion of NK inactivation . Moreover, most of the anti-MM activity of immune-modulatory agents (IMiDs) has been attributed to their capability to induce proliferation and activation of NK cells . Along this line, our study provides novel evidence that trabectedin-treated MM cells are more recognizable by innate immune effectors and that the drug strongly induces NK cell activation. These results are consistent with previous evidence reporting that low doses of doxorubicin or melphalan are able to increase NK activating receptors ligands with a mechanism dependent on DNA damage response .
In our work, we further investigated the regulatory network underlying trabectedin-mediated NKG2D ligands upregulation in MM. To this aim, the expression levels of all known MICA/B regulatory factors were evaluated by GEP and validated at protein level. Our results showed that trabectedin increased the expression level of the MICA/B-positive regulator E2F1 and reduced the expression of the negative regulators IRF4 and IKZF1. These findings are in line with recent reports where the phosphorylation of the kinases ATM/ATR, together with the production of ROS, induced the activation of E2F1 that, in turn, could promote MICA, MICB, and PVR transcription . Additionally, the inhibition of IRF4 and IKZF1/3 by several drugs such as bromodomain and extra terminal domain inhibitors (BETi) and IMiDs has been found to induce MICA and PVR transcription [7, 52].
The mechanism through which trabectedin could affects E2F1, IRF4, and IKZF1 expression still remains elusive. Several reports in this field suggest the existence of possible networks between all these molecules that could account for the fine and reciprocal regulation we observed in this study [53, 54, 55, 56]. Specifically, CUL4A, one of the four components (together with DDB1, CRBN, and DDB2) of a complex ubiquitination machinery responsible for lenalidomide-induced IKZF1 and IKZF3 degradation, has been demonstrated to be necessary for trabectedin activity . Additionally, we observed an upregulation of three of the four genes of the ubiquitination machinery after trabectedin treatment. However, the strong activation of caspase 3 quickly induced by trabectedin treatment may account for the degradation of IKZF1 , with a mechanism that is independent from CRBN expression (differently from what observed with IMiDs). Both mechanisms could than contribute to IKZF1 downregulation that in turn inhibits IRF4 expression at transcriptional level , reducing the repressive regulation on MICA/B promoter.
On the other side, E2F1 has been demonstrated to be induced in response to several DNA-damaging agents, including UV radiation and a number of chemotherapeutic drugs . This translates in an increase in protein stability and in some cases apoptosis . Furthermore, taking into account that reprogramming the immune response requires rapid changes at both transcriptional and post-transcriptional level, we hypothesized a role for miRNAs in finely tuning this regulatory network. By using miRNA target prediction tools, miR-17 family has been identified as the most relevant in MM biology predicted to target at the same time MICA, MICB, and E2F1. We then confirmed that trabectedin downregulates miR-17 and miR-20a in MM cells and that miR-17-92 stable overexpression produced by a lentiviral construct reduces trabectedin-dependent upregulation of NKG2D ligands, making MM cells resistant to drug-induced apoptosis. The latter result appears to be in line with recent findings where miR-17-92 upregulation was found to be associated with resistance to trabectedin . Additionally, the downregulation of miR-17 and miR-20a may further increase the upregulation of E2F1 , contributing to cell cycle arrest, and MICA/B surface expression. Thus, trabectedin could induce E2F1 upregulation through both DNA damage and miR-17 and miR-20a downregulation. However, the mechanism by which trabectedin reduces miR-17-92 transcription and impairs the miR-17-92/E2F1 auto-regulatory loop [61, 62] is currently under investigation.
Altogether, our results demonstrated a potent and pleiotropic preclinical activity of trabectedin in multiple myeloma. Specifically, we here demonstrated an overexpression of NER genes in malignant MM plasma cells that strongly correlates with patients’ prognosis. Accordingly, we found that trabectedin exerts cytotoxicity on both MM cell lines (including drug-resistant derivatives) and primary MM patients derived cells at nanomolar concentrations, by inducing apoptosis, as confirmed by upregulation of caspase3 and downregulation of BCL-2, and cell cycle arrest increasing S-phase. Trabectedin also induces ROS production, with activation of stress response pathways, and DNA damage, enhancing the cytotoxic effect on MM cells. Importantly, trabectedin overcomes microenvironment-induced resistance, impairs MM-macrophages-mediated neo-angiogenesis, and induces NKG2D ligands upregulation enhancing NK-mediated killing. On the basis of these findings, trabectedin emerges as a new potential agent for the treatment of MM that deserves further translational and clinical investigation.
We thank Dr. Ivana Criniti for her technical and editorial support.
This work has been supported by the Italian Association for Cancer Research (AIRC), Special Program Molecular Clinical Oncology-5 per mille n. 9980, 2010/15 PI: PT and 5 × 1000 Molecular Clinical Oncology Extension Program, n.9980, 2016/18 PI: PT and partially by Innovative immunotherapeutic treatments of human cancer, MultiUnit–Multi Unit Regional n.16695 (co-financed by AIRC and CARICAL foundation), 2015/18 PI: PT.
Availability of data and materials
All the datasets used to support the conclusions of this article are available from the corresponding author upon request. Gene expression profiling dataset has been deposited under the GEO accession number GSE128020.
MC, MEGC, P Tagliaferri, P Tassone, and CB conceived and designed experiments. MC, MEGC, MAS, CR, DC, FS, KG, VZ, LMADV, and CB performed the experiments. NS, NA, VB, MA, MTDM, and MR provided crucial samples and technical advice. MC, MEGC, and CB analyzed the data, with conceptual input from P Tagliaferri, P Tassone, MR, DC, NA, NS, CR, MA, and analytical advice from KG, MA, and MTDM. MC, MEGC, P Tagliaferri P Tassone, and CB wrote the manuscript. All authors discussed the results and commented on the manuscript. All authors read and approved the final manuscript.
Ethics approval and consent to participate
Consent for publication
The authors declare that they have no competing interests.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
- 3.Kassambara A, Gourzones-Dmitriev C, Sahota S, Reme T, Moreaux J, Goldschmidt H, Constantinou A, Pasero P, Hose D, Klein B. A DNA repair pathway score predicts survival in human multiple myeloma: the potential for therapeutic strategy. Oncotarget. 2014;5(9):2487–98.PubMedPubMedCentralCrossRefGoogle Scholar
- 7.Fionda C, Abruzzese MP, Zingoni A, Cecere F, Vulpis E, Peruzzi G, Soriani A, Molfetta R, Paolini R, Ricciardi MR, et al. The IMiDs targets IKZF-1/3 and IRF4 as novel negative regulators of NK cell-activating ligands expression in multiple myeloma. Oncotarget. 2015;6(27):23609–30.PubMedPubMedCentralCrossRefGoogle Scholar
- 8.Soriani A, Zingoni A, Cerboni C, Iannitto ML, Ricciardi MR, Di Gialleonardo V, Cippitelli M, Fionda C, Petrucci MT, Guarini A, et al. ATM-ATR-dependent up-regulation of DNAM-1 and NKG2D ligands on multiple myeloma cells by therapeutic agents results in enhanced NK-cell susceptibility and is associated with a senescent phenotype. Blood. 2009;113(15):3503–11.PubMedCrossRefGoogle Scholar
- 10.Morelli E, Biamonte L, Federico C, Amodio N, Di Martino MT, Gallo Cantafio ME, Manzoni M, Scionti F, Samur MK, Gulla A, et al. Therapeutic vulnerability of multiple myeloma to MIR17PTi, a first-in-class inhibitor of pri-miR-17-92. Blood. 2018;132(10):1050–63.PubMedPubMedCentralCrossRefGoogle Scholar
- 12.Botta C, Cuce M, Pitari MR, Caracciolo D, Gulla A, Morelli E, Riillo C, Biamonte L, Gallo Cantafio ME, Prabhala R, et al. MiR-29b antagonizes the pro-inflammatory tumor-promoting activity of multiple myeloma-educated dendritic cells. Leukemia. 2018;32(4):1003–15.PubMedPubMedCentralCrossRefGoogle Scholar
- 13.Subramanian A, Tamayo P, Mootha VK, Mukherjee S, Ebert BL, Gillette MA, Paulovich A, Pomeroy SL, Golub TR, Lander ES, et al. Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc Natl Acad Sci U S A. 2005;102(43):15545–50.PubMedPubMedCentralCrossRefGoogle Scholar
- 16.Caracciolo D, Di Martino MT, Amodio N, Morelli E, Montesano M, Botta C, Scionti F, Talarico D, Altomare E, Gallo Cantafio ME, Zuccalà V, Maltese L, Todoerti K, Rossi M, Arbitrio M, Neri A, Tagliaferri P, Tassone P. miR-22 suppresses DNA ligase III addiction in multiple myeloma. Leukemia. 2019;33(2):487-498. https://doi.org/10.1038/s41375-018-0238-2. Epub 2018 Aug 17. PubMed PMID: 30120376; PubMed Central PMCID: PMC6365379.
- 20.Palumbo A, Avet-Loiseau H, Oliva S, Lokhorst HM, Goldschmidt H, Rosinol L, Richardson P, Caltagirone S, Lahuerta JJ, Facon T, et al. Revised international staging system for multiple myeloma: a report from international myeloma working group. J Clin Oncol. 2015;33(26):2863–9.PubMedPubMedCentralCrossRefGoogle Scholar
- 23.Schoffski P, Taron M, Jimeno J, Grosso F, Sanfilipio R, Casali PG, Le Cesne A, Jones RL, Blay JY, Poveda A, et al. Predictive impact of DNA repair functionality on clinical outcome of advanced sarcoma patients treated with trabectedin: a retrospective multicentric study. Eur J Cancer. 2011;47(7):1006–12.PubMedCrossRefGoogle Scholar
- 25.Lombardi L, Poretti G, Mattioli M, Fabris S, Agnelli L, Bicciato S, Kwee I, Rinaldi A, Ronchetti D, Verdelli D, et al. Molecular characterization of human multiple myeloma cell lines by integrative genomics: insights into the biology of the disease. Genes Chromosomes Cancer. 2007;46(3):226–38.PubMedCrossRefGoogle Scholar
- 36.Caracciolo D, Montesano M, Altomare E, Scionti F, Di Martino MT, Tagliaferri P, Tassone P. The potential role of miRNAs in multiple myeloma therapy. Expert Rev Hematol. 2018;11(10):793-803. https://doi.org/10.1080/17474086.2018.1517041. Epub 2018 Sep 12. PubMed PMID: 30148649.
- 51.Soriani A, Iannitto ML, Ricci B, Fionda C, Malgarini G, Morrone S, Peruzzi G, Ricciardi MR, Petrucci MT, Cippitelli M, et al. Reactive oxygen species- and DNA damage response-dependent NK cell activating ligand upregulation occurs at transcriptional levels and requires the transcriptional factor E2F1. J Immunol. 2014;193(2):950–60.PubMedCrossRefGoogle Scholar
- 52.Abruzzese MP, Bilotta MT, Fionda C, Zingoni A, Soriani A, Vulpis E, Borrelli C, Zitti B, Petrucci MT, Ricciardi MR, et al. Inhibition of bromodomain and extra-terminal (BET) proteins increases NKG2D ligand MICA expression and sensitivity to NK cell-mediated cytotoxicity in multiple myeloma cells: role of cMYC-IRF4-miR-125b interplay. J Hematol Oncol. 2016;9(1):134.PubMedPubMedCentralCrossRefGoogle Scholar
- 53.Dubrez L. Regulation of E2F1 transcription factor by ubiquitin conjugation. Int J Mol Sci. 2017;18(10)Google Scholar
- 56.Garcia MJ, Saucedo-Cuevas LP, Munoz-Repeto I, Fernandez V, Robles MJ, Domingo S, Palacios J, Aracil M, Nieto A, Tercero JC, et al. Analysis of DNA repair-related genes in breast cancer reveals CUL4A ubiquitin ligase as a novel biomarker of trabectedin response. Mol Cancer Ther. 2013;12(4):530–41.PubMedCrossRefGoogle Scholar
- 60.Uboldi S, Calura E, Beltrame L, Fuso Nerini I, Marchini S, Cavalieri D, Erba E, Chiorino G, Ostano P, D'Angelo D, et al. A systems biology approach to characterize the regulatory networks leading to trabectedin resistance in an in vitro model of myxoid liposarcoma. PLoS One. 2012;7(4):e35423.PubMedPubMedCentralCrossRefGoogle Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.