Depletion of TNS3 in EWS cells expressing ETS1 results in changes in the distribution of vinculin and F-actin

Ectopic expression of ETS1 has transcriptome-wide effects.

The genome-wide binding of EWSR1::FLI1 and associated epigenetic marks.

Supplementary Materials and Methods - RNA sequencing, Chromatin Immunoprecipitation (ChIP) and CUT&RUN analysis, Immunoblotting and Immunofluorescence

Intron retention (IR) is increasingly recognized as a feature of long noncoding RNAs (lncRNAs), yet the mechanisms that shape IR in lncRNAs and the functional consequences of this process remain largely unexplored. To investigate how IR contributes to lncRNA regulation, we performed a genome-wide screen to identify factors controlling IR in the lncRNA PURPL. This approach uncovered a prominent role for U2AF2, which promotes retention of a specific intron in PURPL through a weak polypyrimidine tract. IR of this intron drives nuclear enrichment of PURPL and enhances cell proliferation, revealing biological relevance. Transcriptome-wide analyses showed that although U2AF2 broadly supports canonical splicing consistent with its well-established function in promoting splicing, it also facilitates IR within a distinct subset of RNAs, including the nuclear speckle-associated lncRNA MALAT1. Loss of U2AF2 disrupts MALAT1 speckle localization and using MALAT1 knockout cells reconstituted with wild-type or intron deleted variants, we identified a single intron critical for MALAT1's speckle localization. Deletion of this intron from endogenous MALAT1 impaired speckle localization and reduced cell migration, phenocopying the loss of MALAT1. Together, these findings reveal IR as a key regulatory mechanism governing lncRNA localization and function and uncover an unexpected role for U2AF2 in promoting IR within specific lncRNA contexts.

EWSR1::FLI1-regulated transcription factor genes

ETS1 promotes the migration of EWS cells and regulates the expression of TENSIN3

Intron retention (IR) is a form of alternative splicing in which an intron that is typically spliced out is retained in the mature RNA. Despite emerging evidence of widespread IR in protein-coding genes and lncRNAs, the underlying molecular mechanisms remain unclear. Here, we developed a genome-wide screen termed CRASP-seq, to investigate the mechanisms underlying IR in the lncRNA PURPL. Unexpectedly, the top hit was the essential splicing activator U2AF2 that promoted IR in PURPL by directly binding to a weak polypyrimidine tract. U2AF2 promoted IR in additional transcripts, including the nuclear speckle-localized MALAT1 whose localization to nuclear speckles was disrupted upon U2AF2 depletion. Importantly, retention of a specific endogenous MALAT1 intron was critical for its nuclear speckle localization and promoting cell migration. These findings uncover a non-canonical function of U2AF2 in promoting IR and reveal how IR contributes to the subcellular localization and functions of PURPL and MALAT1 .

EWS cells expressing ETS1 exhibit TENSIN3-dependent changes in the organization of vinculin

The tumor suppressor p53 is a transcription factor that controls the expression of hundreds of genes. Emerging evidence indicates that the p53-induced RNA-binding protein ZMAT3 acts as a key splicing regulator that contributes to p53-dependent tumor suppression in vitro and in vivo. However, the mechanism by which ZMAT3 functions within the p53 pathway remains largely unclear. Here, we discovered a function of ZMAT3 in inhibiting transcription of HKDC1, a hexokinase that regulates glucose metabolism and mitochondrial respiration in human cancer cells. Quantitative proteomics revealed HKDC1 as the most significantly upregulated protein in ZMAT3-depleted colorectal cancer cells. ZMAT3 depletion resulted in increased mitochondrial respiration, which was rescued by simultaneous depletion of HKDC1, suggesting that HKDC1 is a critical downstream effector of ZMAT3. Unexpectedly, ZMAT3 did not bind to HKDC1 RNA or DNA; however, proteomic analysis of the ZMAT3 interactome identified its interaction with the oncogenic transcription factor JUN. ZMAT3 depletion enhanced JUN binding to the HKDC1 locus, leading to increased HKDC1 transcription that was rescued upon JUN depletion, suggesting that JUN activates HKDC1 transcription in ZMAT3-depleted cells. Collectively, these findings uncover a mechanism by which ZMAT3 regulates transcription through JUN and demonstrate that HKDC1 is a key component of the ZMAT3-regulated transcriptome in the context of mitochondrial respiration regulation.

RNASE1 is a ribonuclease secreted by cells and degrades extracellular RNAs. Here, we unexpectedly found that RNASE1, in addition to being secreted, is predominantly localized to the nucleus and functions to inhibit gene expression in human colorectal cancer (CRC) cells. RNASE1 expression is highly cell type-specific and is restricted to well-differentiated CRC cells where its transcription is activated by the pioneer transcription factor FOXA1. Using CRISPR interference utilizing three independent sgRNAs targeting the RNASE1 locus followed by RNA-seq, we found that upon depletion of RNASE1, most of the differentially expressed RNAs are modestly but significantly upregulated suggesting that RNASE1 predominantly functions to inhibit gene expression. In CRC patients, RNASE1 is significantly downregulated and high RNASE1 expression is associated with better patient survival, indicating a potential tumor suppressive function. Consistent with this, RNASE1 depletion results in increased proliferation and clonogenicity indicating that RNASE1 inhibits the growth of CRC cells. Finally, a promising RNASE1 target among the most significantly upregulated mRNAs upon RNASE1 depletion is DKK1 (Dickkopf inhibitor 1) which is upregulated in CRC and negatively regulated by RNASE1. Collectively, this initial characterization of endogenous RNASE1 uncovers a function of RNASE1 in inhibition of gene expression and CRC cell proliferation.

HNRNPH1 is a regulator of alternative splicing, but few studies have defined the splicing events it mediates. Here, we used short- and long-read RNA sequencing to interrogate the transcriptome-wide effects of HNRNPH1 depletion and its regulation of specific splicing events. Differential alternative splicing analysis revealed effects on the transcriptome that involved all splice event categories. We confirmed HNRNPH1's regulation of a splicing event involving TCF3-exons 18a and 18b that encode distinct TCF3 transcription factor isoforms. Extending this finding, we present evidence that in neuroblastoma, HNRNPH1 is a MYCN target, potentially explaining the higher levels of HNRNPH1 and TCF3-exon 18a transcript variants in this tumor type. Analysis of two skipped exon events determined that HNRNPH1 regulates the splicing of exons encoding part of the EIF4G1 translation initiation factor's N-terminus and an exon included in the 5'UTR of specific transcript variants encoding the mitotic kinase AURKA. Using reporter constructs, we show this AURKA 5'UTR exon enhances expression, suggesting HNRNPH1 could contribute to regulating AURKA protein levels. Our findings highlight HNRNPH1's roles in regulating the expression of proteins with diverse cellular functions.

Abstract The RNA binding protein HNRNPH1 is a critical regulator of alternative splicing (AS). We have previously demonstrated that HNRNPH1 mediates an exon exclusion event required for the expression of the mRNA encoding the fusion oncoprotein expressed in a subset (∼37%) of Ewing sarcomas. Furthermore, studies by others have shown that HNRNPH1 regulates a mutually exclusive exon (MXE) AS event (exon-18a versus exon-18b) that defines the expression of TCF3 transcript variants, which, when dysregulated, promotes tumorigenesis in Burkitt’s lymphoma. In this study, we have interrogated the transcriptome-wide effects of depleting HNRNPH1 by RNAi to further explore its regulation of transcripts expressed by cancer-associated genes. We performed whole transcriptome RNA sequencing (RNAseq; short-read, Illumina; long-read, PacBio IsoSeq) of two cell lines (HEK-293T and HT-1080) post 48-hour transfection with a control siRNA (siNeg) or an siRNA targeting HNRNPH1 (siHNRNPH1). We then assembled a workflow to quantify changes in individual AS events (percent-spliced-in, ΔPSI=±0.1) using rMATS and MAJIQ, and differential transcript expression (DTE) using RSEM (pValue≤0.05) and EBSeq (FDR≤0.05). RNAseq analyses were validated using PCR assays. The rMATS analysis revealed skipped exons (SE) as the most frequent event effected in HEK-293T and HT-1080 cells following HNRNPH1 silencing, with 1069 SE or MXE events from 607 genes common to both. Gene annotation (Metascape) of these 607 genes identified 49 as cancer-associated. Consistent with previous studies, we detected a significant change in the TCF3-exon-18a/18b MXE event using rMATS (decreased PSI for 18a) and MAJIQ (negative ΔPSI for 18a and positive ΔPSI for 18b) following HNRNPH1 silencing in both cell lines. Furthermore, DTE revealed a corresponding decrease in 18a-containing TCF3-transcripts and an increase in 18b-containing TCF3-transcripts, validating our analytical pipeline. Next, we extended our analysis to uncharacterized HNRNPH1-regulated AS events, beginning with examination of transcripts encoding the mitotic kinase AURKA. Previous studies have shown that AS of AURKA transcripts localizes to the untranslated regions (UTRs), particularly the 5’UTR. However, regulators of these AS events are unknown, and it is unclear how these UTR variants affect AURKA protein expression. Long-read RNAseq of AURKA mRNAs from both cell lines revealed extensive 5’UTR AS, and the rMATS and MAJIQ analyses showed decreased PSI and negative ΔPSI for a specific 5’UTR-exon following HNRNPH1 silencing. DTE revealed corresponding decreases in abundance of AURKA transcripts containing this exon (AURKA-203, -205, -209, and -213) upon HNRNPH1 depletion, which we confirmed with variant-specific RT-PCR. Furthermore, immunoblot analysis showed a decrease in AURKA protein expression following the silencing of HNRNPH1. Our assessment of AURKA AS suggests HNRNPH1 mediates the inclusion of a specific AURKA-5’UTR-exon, providing evidence for a previously uncharacterized post-transcriptional regulatory mechanism of AURKA expression. Citation Format: Tayvia Brownmiller, Patricio Pichling, Tamara L Jones, Ioannis Grammatikakis, Soumya Sundara Rajan, Erica C Pehrsson. HNRNPH1 regulates the alternative splicing of transcripts expressed by cancer-associated genes [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: RNAs as Drivers, Targets, and Therapeutics in Cancer; 2024 Nov 14-17; Bellevue, Washington. Philadelphia (PA): AACR; Mol Cancer Ther 2024;23(11_Suppl):Abstract nr A001.

Abstract Intron retention (IR) is a form of alternative splicing in which an intron that should be spliced out from a precursor transcript, is retained in the mature RNA after the splicing is completed. Although there is emerging evidence of widespread IR in protein-coding genes and long noncoding RNAs (lncRNAs), the underlying molecular mechanisms remain largely unclear. Here, we report the discovery of novel transcripts from the p53-induced lncRNA PURPL, in which intron 2 is retained. To determine the molecular mechanism(s) of IR in PURPL, we conducted a CRISPR-based screen in 3 different cell lines. For this purpose, we utilized a genome-wide guide RNA library expressing a reporter minigene containing the sequence of PURPL intron 2 and its flanking exons. Considering the Percent Intron Retention as readout, we unexpectedly identified proteins of the basal splicing machinery as potential promoting factors of PURPL IR, including the U2-Auxiliary Factor 2 (U2AF2) splicing factor which was one of the top hits of the screen. We next analyzed ENCODE eCLIP-seq datasets to identify RNA binding proteins that could regulate intron 2 and decided to focus on U2AF2 which showed the highest number of binding sites and strongest eCLIP signals within PURPL intron 2 sequence. We validated the effect of U2AF2 by knocking it down, confirming that U2AF2 is a positive regulator of PURPL intron 2 retention as opposed to its canonical function. We also identified the RNA Binding Protein SON as a splicing regulator that acts as an antagonist of U2AF2 in regulating PURPL intron 2 retention. To determine the global impact of U2AF2 knockdown, we performed Iso-Seq and RNA-seq upon U2AF2 depletion. We found that although U2AF2 predominantly acts to promote the splicing of introns in most cellular transcripts, it promotes intron retention in a subset of transcripts. One of these targets is MALAT1, a lncRNA known to play role in splicing by interacting with splicing factors in nuclear speckles. Interestingly, U2AF2 depletion results in MALAT1 exclusion from nuclear speckles. We are currently in the process of characterizing the effect of U2AF2 in the function of the two lncRNAs. These data provide mechanistic insights on PURPL and MALAT1 splicing and function and reveal a previously unrecognized non-canonical function of U2AF2 in promoting intron retention. Citation Format: Ioannis Grammatikakis, Amit Behera, Corrine Corrina R Hartford, Erica C Pehrsson, XiaoLing Li, Yongmei Zhao, Biraj Shrethsa, Tayvia Brownmiller, Natasha J Caplen, Kannanganattu V Prasanth, Thomas Gonatopoulos-Pournatzis, Ashish Lal. Molecular mechanisms of intron retention in Long Non-Coding RNAs [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: RNAs as Drivers, Targets, and Therapeutics in Cancer; 2024 Nov 14-17; Bellevue, Washington. Philadelphia (PA): AACR; Mol Cancer Ther 2024;23(11_Suppl):Abstract nr A013.

Ewing sarcoma (EWS) is an aggressive bone and soft tissue tumor that affects children and young adults. Ewing sarcomas harbor few mutations beyond the chromosomal translocation that initiates disease, and the mechanistic basis for the metastasis of these tumors remains poorly understood. The epigenome of EWS cells reflects the regulatory state of genes associated with the DNA binding activity of the fusion oncoproteins EWSR1::FLI1 or EWSR1::ERG. In this study, we examined the EWSR1::FLI1/ERG’s repression of transcription factor genes, particularly those encoding transcriptional regulators of cell differentiation. To examine EWSR1::FLI1/ERG’s regulation of gene expression, we assayed DNA binding using ChIP-seq or CUT&RUN, RNA using RNA-sequencing (RNA-seq) or qRT-PCR, and protein using immunoblotting or immunofluorescence. We depleted the expression of EWSR1::FLI1/ERG proteins using RNAi and overexpressed ETS1 using a full-length cDNA or CRISPR activation.RNA-seq analysis of control and EWSR1::FLI1-silenced TC-32, TC-71, and A673 EWS cells identified 67 genes encoding proteins with DNA binding activity that exhibited significant increases in expression following depletion of EWSR1::FLI1. Comparison of the expression of these 67 transcription factor genes in EWS cell lines (n=41) and tumors (n=79) showed 37 expressed at a wider range in tumors relative to cell lines (p <0.01). This more variable expression in EWS tumors suggests that one or more of these transcription factors could exert a phenotypic effect. Focusing on one of the EWSR1::FLI1-repressed target genes, ETS1, we detected EWSR1::FLI1 binding and an H3K27me3 repressive mark at this locus. Depletion of EWSR1::FLI1 results in ETS1’s binding of promoter regions. Analysis of EWS cells in which we profiled ETS1 binding and assessed the effects of ectopically expressed ETS1 defined 265 genes as positively regulated by ETS1 of which 103 also exhibited an increase in expression following depletion of EWSR1::FLI1. Of these 103 genes, we focused on ETS1’s regulation of TNS3, as assessment of multiple EWS tumor expression profiles, showed positive correlation of ETS1 and TNS3 expression. TNS3 encodes TENSIN3, a focal adhesion protein that regulates cytoskeletal reorganization and contributes to cell migration by connecting the cytoplasmic tail of integrins to the actin cytoskeleton. EWS cell lines (SK-N-MC (EWSR1::FLI1) and ES-5838 (EWSR1::ERG), in which we activated ETS1 expression (CRISPRa) exhibited a migratory phenotype and increased TNS3 expression. Critically, the activated ETS1 EWS cell lines show TNS3 accumulation at leading cell edges, with F-actin cytoskeletal reorganization, a phenotype associated with cell migration. Our study demonstrates that EWS cells expressing ETS1 exhibit a more migratory phenotype, which, has the potential to promote the dissemination of cells and, thus, metastasis. Citation Format: Vernon Justice Ebegboni, Tamara L. Jones, Tayvia Brownmiller, Erica Pehrsson, Patrick Zhao, Soumya Sundara Rajan, Natasha J. Caplen. ETS1, a target gene of the EWSR1::FLI1 fusion oncoprotein, regulates the expression of the focal adhesion protein TENSIN3 [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 2855.

Abstract The mechanistic basis for the metastasis of Ewing sarcomas remains poorly understood, as these tumors harbor few mutations beyond the chromosomal translocation that initiates the disease. Instead, the epigenome of Ewing sarcoma cells reflects the regulatory state of genes associated with the DNA-binding activity of the fusion oncoproteins EWSR1::FLI1 or EWSR1::ERG. In this study, we examined the EWSR1::FLI1/ERG's repression of transcription factor genes, concentrating on those that exhibit a broader range of expression in tumors than in Ewing sarcoma cell lines. Focusing on one of these target genes, ETS1, we detected EWSR1::FLI1 binding and an H3K27me3-repressive mark at this locus. Depletion of EWSR1::FLI1 results in ETS1’s binding of promoter regions, substantially altering the transcriptome of Ewing sarcoma cells, including the upregulation of the gene encoding TENSIN3 (TNS3), a focal adhesion protein. Ewing sarcoma cell lines expressing ETS1 (CRISPRa) exhibited increased TNS3 expression and enhanced movement compared with control cells. Visualization of control Ewing sarcoma cells showed a distributed vinculin signal and a network-like organization of F-actin; in contrast, ETS1-activated Ewing sarcoma cells showed an accumulation of vinculin and F-actin toward the plasma membrane. Interestingly, the phenotype of ETS1-activated Ewing sarcoma cell lines depleted of TNS3 resembled the phenotype of the control cells. Critically, these findings have clinical relevance as TNS3 expression in Ewing sarcoma tumors positively correlates with that of ETS1. Implications: ETS1’s transcriptional regulation of the gene encoding the focal adhesion protein TENSIN3 in Ewing sarcoma cells promotes cell movement, a critical step in the evolution of metastasis.

Abstract The mechanisms that regulate Ewing sarcoma (EWS) cell dissemination are poorly characterized. We have reported that though cell line studies show the EWSR1::FLI1/ERG fusion oncoproteins represses ETS1 expression, there is inter-tumoral variations in its mRNA levels and higher expression of ETS1 in tumors correlate with reduced overall patient survival. ETS1 encodes a transcription factor that regulates the movement of several early progenitor cell sub-types, and activation of ETS1 in EWS cells enhanced their movement and migration. This suggests that ETS1 expression could promote cell dissemination. We also showed that ETS1 expression in EWS cells upregulated the expression of the focal adhesion (FA) protein TENSIN3 (TNS3). Critically, depletion of TNS3 reverted the enhanced formation of FAs induced by ETS1 and partially rescued the increased migratory phenotype. In this study, we investigated the ETS1-induced changes in the phenotype of EWS cells by assessing TNS3 function in the context of other FA proteins. We generated epigenetic and transcriptomic profiles of EWS cells using ChIP-seq, CUT&RUN-seq, or RNA-seq in unperturbed, siRNA-treated or CRISPRa-EWS cells. We used dCas9-VP64 (control) or ETS1-activated (ETS1a) SK-N-MC cells and performed immunofluorescence and super resolution microscopy employing antibodies against TNS3, vinculin (VCL), paxillin (PXN), pan-talin (TLN), and focal adhesion kinase (FAK). Previously, we observed TNS3 and the FA scaffold protein VCL at the membrane projections of ETS1a EWS cells, however, we did not determine whether this reflected a change in VCL expression and/or protein redistribution. Epigenetic analyses showed EWSR1::FLI1/ERG, and ETS1 binding (following depletion of EWSR1::FLI1) at VCL and its expression increased 6- and 3-fold following EWSR1::FLI1 and EWSR1::ERG depletion, respectively. In contrast, we observed only a slight increase in VCL expression (RNA or protein) in ETS1a EWS cells. This suggests that our visualization of VCL in the ETS1a EWS cells reflected a change in its localization, not expression. Consistent with this finding, in the ETS1a EWS cells, we observed an increase in VCL’s non-nuclear intensity relative to control cells. This observation reflected VCL’s recruitment to the plasma membrane, a phenotype that TNS3-depletion reversed. To further study whether TNS3 is the predominate determinant of FA formation in ETS1a EWS cells, we examined other FA components, specifically PAX, TLN1, TLN2, and FAK. We observed no evidence of EWSR1::FLI1/ERG or ETS1 directly regulating these genes. Nevertheless, PXN, TLN, and FAK, all showed an increase in both total and non-nuclear intensities following ETS1 activation, suggesting that ETS1 indirectly activates the expression of other FA proteins. Nevertheless, depletion of TNS3 decreased the non-nuclear intensities of these FA proteins, confirming the importance of TNS3 in the formation FAs in the ETS1-modfied lines. We thus conclude that if expressed in an EWS cell, ETS1’s regulation of TNS3 that will enhance FA formation and cell movement. Citation Format: Vernon J. Ebegboni, Tamara J. Jones, Shaoli Das, Erica C. Pehrsson, Soumya Sundara Rajan, Natasha J. Caplen. The ETS1-driven expression of TNS3 in Ewing sarcoma cells enhances the formation of the focal adhesions that promote cell movement [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Advances in Pediatric Cancer Research; 2024 Sep 5-8; Toronto, Ontario, Canada. Philadelphia (PA): AACR; Cancer Res 2024;84(17 Suppl):Abstract nr A004.