MS0621, a novel small-molecule modulator of Ewing sarcoma chromatin accessibility, interacts with an RNA-associated macromolecular complex and influences RNA splicing

Abstract

Ewing sarcoma is a cancer of children and young adults characterized by the critical translocation-associated fusion oncoprotein EWSR1::FLI1. EWSR1::FLI1 targets characteristic genetic loci where it mediates aberrant chromatin and the establishment of de novo enhancers. Ewing sarcoma thus provides a model to interrogate mechanisms underlying chromatin dysregulation in tumorigenesis. Previously, we developed a high-throughput chromatin-based screening platform based on the de novo enhancers and demonstrated its utility in identifying small molecules capable of altering chromatin accessibility. Here, we report the identification of MS0621, a molecule with previously uncharacterized mechanism of action, as a small molecule modulator of chromatin state at sites of aberrant chromatin accessibility at EWSR1::FLI1-bound loci. MS0621 suppresses cellular proliferation of Ewing sarcoma cell lines by cell cycle arrest. Proteomic studies demonstrate that MS0621 associates with EWSR1::FLI1, RNA binding and splicing proteins, as well as chromatin regulatory proteins. Surprisingly, interactions with chromatin and many RNA-binding proteins, including EWSR1::FLI1 and its known interactors, were RNA-independent. Our findings suggest that MS0621 affects EWSR1::FLI1-mediated chromatin activity by interacting with and altering the activity of RNA splicing machinery and chromatin modulating factors. Genetic modulation of these proteins similarly inhibits proliferation and alters chromatin in Ewing sarcoma cells. The use of an oncogene-associated chromatin signature as a target allows for a direct approach to screen for unrecognized modulators of epigenetic machinery and provides a framework for using chromatin-based assays for future therapeutic discovery efforts.

Keywords: Ewing sarcoma (ES); RNA-binding proteins; RNA-processing; SWI/SNF (BAF) complex; chromatin; drug discovery.

Scheme 1

Synthetic route for MS0621 Reagents and conditions: (A) Piperidine, K₂CO₃, NaI, CH₃CN, 80°C, overnight, 78%; (B) Fe, NH₄OAc, EtOAC/H₂O, reflux, overnight, 58%; (c) NaOCN, HOAc/H₂O, rt, overnight; (D) NaOH, MeOH/H₂O, reflux, 3h; (E) POCl₃, PhEt₂, reflux, 7h, 50%; (F) (1-isopropylpiperidin-4-yl)amine, DIEA, THF, rt, overnight, 99%; (G) 1-Ethylhomopiperazine, TFA, i-PrOH, MW, 160°C, 15 min, 80%.

Scheme 2

Synthetic route for MS1360 Reagents and conditions: (A) K₂CO₃, CH₃CN, reflux, 5 h, 56%; (B) TFA, DCM, rt; (C) TFA, i-PrOH, MW, 130°C, 20 min, 40% in 2 steps. (D) azide-PEG2-biotin conjugate, Vc, CuSO₄, t-BuOH, rt, overnight, 40%.

Figure 1

A chromatin-based screen identifies a small molecule modulator of chromatin state at EWS-FLI-bound GGAA microsatellites. (A) Waterfall plot of screen results with rank ordered Relative Chromatin Inhibition values (RCI, FAIRE signal at Ewing sarcoma regions/FAIRE at control regions, log2). Dashed lines indicate +/- 2SDs from the average RCI for vehicle controls. Magnification: Thirty compounds demonstrating the greatest decrease in FAIRE signal. The bar representing UNC0621/MS0621 is indicated in light blue. Dark gray bars indicate structural analogs of MS0621 (discussed further in Figure 2 ). Figure adapted from (9). (B) Chemical structure of MS0621 (C) FAIRE-qPCR at the regions used in the screen in EWS894 cells treated with MS0621 or a control compound for 16 (h) Results are shown as a fraction of input control. Error bars represent the standard error of four biological replicates for MS0621 and two replicates for the control compound. Statistical significance compared to the lowest concentration of each compound was assessed using unpaired student t-tests *p < 0.05, **p < 0.01, ns: not significant. (D) FAIRE-qPCR at EWSR1::FLI1-bound and control loci in EWS894 cells treated with 10 μM MS0621 or vehicle control (DMSO) for 16 hours. Results are shown as a fraction of input control normalized to vehicle control. Error bars represent the standard error of three biological replicates. Statistical significance compared to the vehicle control at each locus was assessed using unpaired student t-tests *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 ns: not significant. (E) FAIRE-qPCR in the indicated Ewing sarcoma cell lines treated with 5 μM MS0621 or vehicle control (PBS) for 16 (h) Results are shown as Log2 RCI scores. Dashed lines indicate –2 SDs from the average RCI for vehicle controls. Error bars represent the standard deviation of two biological replicates. (F) ChIP-qPCR at EWS-FLI-bound (P1, P7, CAV2) and control regions (Neg) in A673 cells treated with 5 μM MS0621 or vehicle control (PBS) for 16 (h) Results are shown as a fraction of input control. Error bars represent the standard deviation of two biological replicates.

Figure 2

MS0621 induces a G1/S arrest in Ewing sarcoma cell lines. (A) Cell proliferation was assayed following 3 days of treatment with MS0621 or a vehicle control (PBS) in the indicated Ewing sarcoma cell lines. Viable cells were assessed by WST-1. Results are shown as the fold change relative to vehicle treated cells. Error bars represent the SD of three biological replicates. (B) Cell proliferation in UMRC2 (clear cell renal cell carcinoma, ccRCC), 786-O (ccRCC), RPTEC (Primary Renal Proximal Tubule Epithelial Cells), HUVEC (Primary Umbilical Vein Endothelial Cells), and SU-CCS-1 (clear cell sarcoma) cells treated with MS0621 or vehicle control for 3 days. Proliferation was assessed on day 3 by WST. Results are shown as the fold change of vehicle treated cells. Error bars represent the SD of three biological replicates. (C) Cell proliferation was assessed by live cell imaging in A673 cells treated with MS0621 or vehicle control over 6 days. Concentrations are in micromolar. Proliferation, assessed by images captured every 2 hours, are shown as the percent confluence. Error bars represent the standard error of the mean for six (MS0621 treatment) technical replicates or two (vehicle control) technical replicates. (D) Soft agar colony formation assays for EWS894 cells treated with the indicated doses of MS0621 or vehicle (DMSO) for 15 days. Viable cell colonies were visualized on day 15 with MTT. Results shown are representative wells of three biological replicates. (E) Quantification of soft agar colony formation. Error bars represent the SD of three biological replicates. (F) Cell cycle analysis in EWS502 cells treated with the indicated doses of MS0621. Following treatment with MS0621 or vehicle (DMSO) for 3 days, cells were stained with propidium iodide (PI) and analyzed by flow cytometry. (G) Quantification of the percent of the population in each stage of the cell cycle for each treatment condition in (F). (H) S-phase quantification by BrdU staining in EWS502 cells treated with 5 μM MS0621or vehicle (DMSO). Following compound treatment for the indicated time, cells were pulsed with BrdU for 1 hour, harvested, fixed, and analyzed by flow cytometry. (I) Quantification of the BrdU-positive population in (H). Results are shown as the percent of cells in S-phase.

Figure 3

MS0621 active analog interacts with a macromolecular complex enriched with RNA-binding proteins (A) Chemical structure of MS2616, alkyne-substituted analog of MS0621. (B) FAIRE-qPCR at the screening loci in EWS894 cells treated with the indicated doses of MS0621 or UNC4151 for 16 hours. Results are shown as Relative Chromatin Inhibition. Error bars represent the standard deviation of 5 technical replicates. (C) Cell proliferation curves for EWS894 cells treated with the indicated doses of MS0621 or MS2616 (twofold dilutions from 5 μM to 0.15625 μM) or vehicle control for 3 days. Proliferation was assessed on day 3 by WST assay. Results are shown as the fold change of vehicle treated cells. Error bars represent the SD of three biological replicates. (D) STRING diagram displaying interactions networks between proteins identified by mass spectrometry in EWS894 cells. (E) KEGG Enriched Biological Processes GO terms for proteins identified by mass spectrometry in EWS894 cells. Results shown are the -log₁₀ FDR for the top 50% of enriched GO terms. (F) Western blot confirmation of proteins identified by mass spectrometry. Proteins chemiprecipitated by 5 μM MS1360 (+) or vehicle control (–) in MNase-digested nuclear extracts of EWS894 cells with or without 100 ug/mL RNase A digestion assessed by western blot. 20% of the MNase-digested nuclear extract input was included for reference (Input). (G) Western blot analyses of proteins immunoprecipitated by EWSR1::FLI1 from nuclear extracts of A673 cells. 6% of the input extract was included for reference. Antibody heavy chain is marked with an asterisk. (H) Western blot analyses of proteins enriched by 5 μM MS1360 (+) or vehicle control (–) in nuclear extracts of A673 cells with or without 100 ug/mL RNase A digestion assessed by western blot. 10% of the nuclear extract input was included for reference. (I) MNase-digested extracts of A673 cells were chemiprecipitated with MS1360 in the presence 150 mM NaCl. Beads were then sequentially washed with the indicated concentrations of NaCl. Beads were removed at each condition for Western blot analyses. 20% of the MNase-digested nuclear extract input was included for reference. (J) Quantification of western blot band intensities for protein enrichment under standard (150 mM NaCl) and high salt (500 mM NaCl) conditions normalized to Input band intensities. Addition of RNase A is indicated by black lines beneath the bars. Results are shown as the fold change of the 150 mM NaCl condition. Error bars represent the standard deviation of three technical replicates. Hist3: Histone H3. (K) Western blot analyses of chemipreciptation of MNase-digested extracts of A673 cells in the presence of decreasing concentrations of MS1360. 20% of the MNase-digested nuclear extract input was included for reference.

Figure 4

hnRNPH1 knockdown recapitulates the effect of MS0621 on Ewing sarcoma cell proliferation and FAIRE-qPCR (A) Western blot analyses of hnRNPH1 and EWSR1::FLI1 protein knockdown in A673-CRISPRi cells. Left: Representative western blots. Right: Quantification of western blot band intensities for hnRNPH1 and EWSR1::FLI1 normalized to Histone H3 band intensities. Results are shown as the fold change of non-specific guide (NS) cells. Error bars represent the standard deviation of three biological replicates. (B) hnRNPH1 and EWSR1::FLI1 RNA abundance measured by RT-qPCR in A673-CRISPRi cells following infection with the indicated sgRNAs for 4 days. Results are shown as the fold change of non-specific guide (sg NS) cells. Error bars represent the standard deviation of three biological replicates. (C) Cell proliferation assayed by live cell imaging for A673-CRISPRi cells infected with the indicated sgRNAs over 4 days. Images were obtained every 2 hours. Results are shown as the percent confluence. Error bars represent the standard error of the mean for two technical replicates where each replicate consists of 16 images per time point. Two biological replicates (A, B) are shown for hnRNPH1 and EWSR1. Nonspecific sgRNA was used as a control (NS). (D) FAIRE-qPCR in A673-CRISPRi cells infected with lentivirus transducing sgRNA for hnRNPH1 for 4 days (Left: EWSR1::FLI1 target regions. Right: Positive control regions). Results are shown as a fraction of input control. Error bars represent the standard error of three biological replicates. (E) FAIRE-qPCR in A673-CRISPRi cells infected with lentivirus transducing sgRNA for EWSR1 for 4 days (Left: EWSR1::FLI1 target regions. Right: Positive control regions). Results are shown as a fraction of input control. Error bars represent the standard error of three biological replicates. Control sgRNA results are duplicated in Figures (D, E). The asterisks indicate the significance of unpaired t-tests where *p < 0.05 and **p < 0.01. ns, nonsignificant.

Figure 5

MS0621 affects EWSR1::FLI1 regulated genes and influences RNA splicing (A) Heatmap displaying row normalized Z-scores for differentially expressed genes in EWS894 cells following 16 hours of treatment with 5 μM MS0621 or vehicle control (DMSO). The right most column indicates genes also differentially regulated by EWSR1::FLI1 in (30). (B) Gene set enrichment analysis of genes differentially expressed in EWS894 cells following treatment for 16 hours with 5 μM MS0621 or vehicle control (DMSO) using gene sets upregulated by EWSR1::FLI1. (C) Enriched Biological Processes GO terms for genes that were downregulated with MS0621 treatment but not regulated by EWSR1::FLI1. Results shown are the -log₁₀ adjusted p-value for the top 20 enriched GO terms. (D) Significant differential alternative splicing events were identified by rMATS from RNA prepared from EWS894 cells treated 16 hours with 5 μM MS0621 or vehicle control (DMSO). Indicated splicing events were supported by at least 20 reads, an Inclusion Level Difference > 10%, and FDR < 0.05. (E) Differential retention of a tumor-characteristic pair of introns in the FUS gene. Gene tracks represent three replicates each for Vehicle- or MS0621- treated EWS894 cells. (F) Box plots of log10 TPM values for genes with an RI event gained or lost with MS0621 or with no RI event. (G) Percentage of transcript-specific SE events identified by rMATS annotated by GENCODE biotype. (H) Sashimi plots of differential retention of the SRSF6 poison exon. Sashimi plots represent three replicates each for Vehicle- or MS0621- treated EWS894 cells. (I) Enrichment of SE events gained with MS0621 and EWSR1::FLI1 was assessed by permuting 1,000 times over spliceable exons with minimum read coverage greater than or equal to 20 reads. ***p-value less than 0.001.

Figure 6

Proposed Model of Perturbation of MS0621. The chromatin landscape at EWSR1::FLI1 targeted sites is likely established and maintained through the activity of SWI/SNF as well as through biophysical properties such as phase separation mediated by proteins and RNA (gold RNA molecule). Likewise, EWSR1::FLI1 alters the function of RNA binding proteins to program Ewing sarcoma-specific alternative patterns (gold and red RNA molecule). EWSR1::FLI1 may depend on the complex to mediate its roles in chromatin and transcription, but this dependence may also represent a vulnerability for Ewing sarcoma cells. MS0621 interacts with and may perturb the complex and inhibit the functions required by EWSR1::FLI1 leading to altered chromatin organization, gene expression, and splicing programs. Thus, perturbation of these functions likely contributes to the cell proliferation defects and cell cycle arrest observed in Ewing sarcoma cell lines upon MS0621 treatment. Created with BioRender.com.