SAM68 is a multifunctional post-transcriptional regulator of cardiomyocyte differentiation.

SAM68 is a multifunctional post-transcriptional regulator of cardiomyocyte differentiation - PMC Skip to main content An official website of the United States government Here's how you know Here's how you know Search Log in Primary site navigation Search Log in Search PMC Full-Text Archive Search in PMC Journal List User Guide As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. Learn more: PMC Disclaimer | PMC Copyright Notice Nucleic Acids Res . 2026 Jun 4;54(10):gkag547. doi: 10.1093/nar/gkag547 SAM68 is a multifunctional post-transcriptional regulator of cardiomyocyte differentiation Laura Broglia 1, c , Alessandro Dasti 2, c , Maria Carla Antonelli 3, c , Guy Aoun 4 , Sabrina D’Agostino 5 , Andrea Vandelli 6 , Alexandros Armaos 7 , Riccardo Delli Ponti 8 , Sarah Wolf 9 , Melina Klostermann 10 , Magdalena Arnal Segura 11 , Tian V Tian 12 , Davide Mariani 13, 14 , Alessio Colantoni 15 , Maria Paola Paronetto 16 , Stefano Gustincich 17 , Kathi Zarnack 18 , Elias Bechara 19, ✉ , Gian Gaetano Tartaglia 20, ✉ Author information Article notes Copyright and License information 1 Center for Human Technologies, RNA Systems Biology, Istituto Italiano di Tecnologia, Genoa 16152, Italy 2 Center for Human Technologies, RNA Systems Biology, Istituto Italiano di Tecnologia, Genoa 16152, Italy 3 Center for Human Technologies, RNA Systems Biology, Istituto Italiano di Tecnologia, Genoa 16152, Italy 4 Center for Human Technologies, RNA Systems Biology, Istituto Italiano di Tecnologia, Genoa 16152, Italy 5 Center for Human Technologies, Non-Coding RNAs and RNA-Based Therapeutics, Istituto Italiano di Tecnologia, Genoa 16152, Italy 6 Center for Human Technologies, RNA Systems Biology, Istituto Italiano di Tecnologia, Genoa 16152, Italy 7 Center for Human Technologies, RNA Systems Biology, Istituto Italiano di Tecnologia, Genoa 16152, Italy 8 Center for Human Technologies, RNA Systems Biology, Istituto Italiano di Tecnologia, Genoa 16152, Italy 9 Theodor Boveri Institute, Biocenter, University of Würzburg, 97074 Würzburg, Germany 10 Theodor Boveri Institute, Biocenter, University of Würzburg, 97074 Würzburg, Germany 11 Center for Human Technologies, RNA Systems Biology, Istituto Italiano di Tecnologia, Genoa 16152, Italy 12 Vall d’Hebron Institute of Oncology, Barcelona 08035, Spain 13 Center for Human Technologies, Non-Coding RNAs in Physiology and Pathology, Istituto Italiano di Tecnologia, Via Enrico Melen 83, 16153 Genoa, Italy 14 Department of Biology and Biotechnologies “C. Darwin”, Sapienza University of Rome, Piazzale Aldo Moro 5, 00185 Rome, Italy 15 Center for Human Technologies, RNA Systems Biology, Istituto Italiano di Tecnologia, Genoa 16152, Italy 16 Department of Movement, Human and Health Sciences, University of Rome “Foro Italico”, Rome 00135, Italy 17 Center for Human Technologies, Non-Coding RNAs and RNA-Based Therapeutics, Istituto Italiano di Tecnologia, Genoa 16152, Italy 18 Theodor Boveri Institute, Biocenter, University of Würzburg, 97074 Würzburg, Germany 19 Center for Human Technologies, RNA Systems Biology, Istituto Italiano di Tecnologia, Genoa 16152, Italy 20 Center for Human Technologies, RNA Systems Biology, Istituto Italiano di Tecnologia, Genoa 16152, Italy ✉ To whom correspondence should be addressed. Email: elias.bechara@iit.it ✉ Correspondence may also be addressed to Gian Gaetano Tartaglia. Email: gian.tartaglia@iit.it c Joint authors. Roles Laura Broglia : Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Validation, Visualization, Writing - original draft, Writing - review & editing Alessandro Dasti : Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Validation, Visualization Maria Carla Antonelli : Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Validation, Visualization Guy Aoun : Formal analysis, Investigation, Validation Sabrina D’Agostino : Formal analysis, Investigation, Methodology, Validation Andrea Vandelli : Formal analysis, Investigation, Methodology, Validation, Visualization Alexandros Armaos : Formal analysis, Investigation, Methodology, Validation, Visualization, Formal analysis, Investigation, Methodology, Validation, Visualization Riccardo Delli Ponti : Data curation, Formal analysis, Investigation Sarah Wolf : Data curation, Formal analysis, Investigation Melina Klostermann : Data curation, Formal analysis, Investigation, Methodology Magdalena Arnal Segura : Formal analysis, Investigation, Validation, Visualization Tian V Tian : Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Visualization Davide Mariani : Formal analysis, Methodology, Visualization Alessio Colantoni : Formal analysis, Methodology Maria Paola Paronetto : Formal analysis, Methodology, Supervision Stefano Gustincich : Investigation, Methodology, Supervision Kathi Zarnack : Data curation, Formal analysis, Methodology Elias Bechara : Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Supervision, Validation, Visualization, Writing - original draft, Writing - review & editing Gian Gaetano Tartaglia : Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Validation, Visualization, Writing - original draft, Writing - review & editing Received 2025 Jun 12; Revised 2026 May 6; Accepted 2026 May 9; Collection date 2026 Jun 10. © The Author(s) 2026. Published by Oxford University Press. This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License ( https://creativecommons.org/licenses/by-nc/4.0/ ), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact reprints@oup.com for reprints and translation rights for reprints. All other permissions can be obtained through our RightsLink service via the Permissions link on the article page on our site—for further information please contact journals.permissions@oup.com . PMC Copyright notice PMCID: PMC13234510  PMID: 42240622 Abstract RNA-binding proteins (RBPs) of the STAR family play key roles in mammalian development, yet their contributions to lineage specification remain incompletely understood. Here, using CRISPR–Cas9 knockout models combined with multi-omics approaches, we investigate the functions of two STAR proteins, SAM68 and QKI, in mouse embryonic stem cells (mESCs). Both RBPs support mESC proliferation, self-renewal, and efficient differentiation into cardiomyocytes. Although SAM68 and QKI belong to the same protein family, they control largely distinct regulatory programs during differentiation. We uncover an unexpected role for SAM68 in cardiomyocyte specification through multiple post-transcriptional mechanisms. SAM68 modulates alternative splicing and promotes the biogenesis of a subset of cardiac-enriched circular RNAs, through binding to intronic regions flanking back-splice junctions and potentially through association with NF90/110. In addition, SAM68 binds untranslated regions of key differentiation-related transcripts, including Gata4 mRNA, and functions in ribonucleoprotein complexes to regulate their translation. Together, these findings identify SAM68 as a multifunctional regulator coordinating multiple layers of RNA metabolism—including splicing, circRNA biogenesis, and translation—during cardiomyocyte differentiation and provide insight into how STAR proteins shape post-transcriptional gene regulatory networks during early development. Graphical Abstract Graphical Abstract. Open in a new tab Introduction RNA-binding proteins (RBPs) are central regulators of gene expression that control multiple stages of the RNA lifecycle, from its synthesis to its degradation [ 1 ]. By directly interacting with their target RNAs, RBPs shape the transcriptome and proteome in response to developmental and environmental cues, thereby adding critical layers of regulation beyond transcription [ 2 , 3 ]. In embryonic stem cells (ESCs), RBPs play central roles in maintaining pluripotency and orchestrating differentiation by coordinating multiple post-transcriptional processes [ 3 ]. A major function of RBPs is the regulation of alternative splicing (AS), a process that contributes to transcriptome complexity by generating multiple messenger RNA (mRNA) isoforms through exon inclusion or skipping, alternative 5′ and 3′ splice site selection, and intron retention. AS plays a central role in tissue- and species-specific gene expression programs and drives cell differentiation [ 4 ]. In ESCs, specific RBPs such as RBFOX2, MBNL1, and SON orchestrate pluripotency-associated splicing networks to regulate lineage-specific isoform expression [ 5 – 10 ]. Alongside linear splicing, RBPs also regulate back-splicing, in which a downstream splice donor is covalently linked to an upstream splice acceptor through a 3′–5′ phosphodiester bond, resulting in the formation of circular RNAs (circRNAs). These covalently closed RNA molecules exhibit highly specific expression patterns across developmental stages, cell types, and tissues [ 11 ], and have emerged as important regulators of pluripotency and differentiation [ 12 – 14 ]. ESRP1 promotes the production of circBIRC6 , which acts as a sponge for miR-34a and miR-145, thereby relieving repression of key pluripotency factors such as OCT4, NANOG, and SOX2 [ 15 ]. circQKI promotes myoblast differentiation, underscoring a critical regulatory axis with its host gene, QKI [ 16 ]. As a member of the Signal Transduction and Activation of RNA (STAR) family of RBPs, QKI plays a dual role in orchestrating both muscle development [ 17 ] and the broader landscape of circRNA biogenesis [ 18 ]. Among the many RBP families, the STAR family stands out for its ability to couple extracellular signals with post-transcriptional control of gene expression. STAR proteins, comprising SAM68 (KHDRBS1), QKI, SLM1, SLM2, and SF1 are characterized by a KH-type RNA-binding domain embedded within regions mediating protein–protein interactions and phosphorylation [ 31 , 32 ]. This modular architecture allows them to act as molecular integrators of signaling pathways and RNA metabolism [ 33 ]. For instance, ERK1/2 signaling activation was shown to promote SAM68-dependent inclusion of CD44 variable exon 5 [ 34 ]. Traditionally viewed as nuclear splicing regulators, STAR proteins are now increasingly recognized for their cytoplasmic roles in RNA localization and stability [ 35 , 36 ]. SAM68 has been extensively studied in the context of male germline development, where it regulates AS and translation of specific target mRNAs during male germ cell differentiation [ 37 – 39 ]. Its genetic ablation results in male infertility due to impaired chromatin condensation during spermatid differentiation [ 38 ]. SAM68 regulates alternative splicing in multiple contexts, including synaptic gene expression in neurons—affecting axon guidance and neurogenesis [ 40 – 42 ]—and adipogenesis [ 43 , 44 ]. Other STAR family members further highlight the developmental relevance of this protein class. Qki -null mice exhibit embryonic lethality due to defective vascular remodeling and heart formation, suggesting a crucial role of QKI in cardiovascular development [ 17 , 45 , 46 ]. More recently, QKI was shown to be indispensable for cardiomyocyte maturation in human embryonic stem cells (hESCs), where it controls the splicing and stability of cardiac-specific transcripts [ 47 ]. Furthermore, QKI is linked to the biogenesis of circRNAs during epithelial-to-mesenchymal transition [ 14 ] and in cardiomyocytes [ 48 ]. SLM1 and SLM2, although less broadly expressed, are highly enriched in the nervous system and regulate neuron-specific alternative splicing programs during cortical development and neuronal differentiation [ 49 ]. Despite these well-characterized tissue specific roles, the function of SAM68 in early embryonic cell fate transitions and impact on the transcriptome, particularly through mechanisms beyond linear splicing, remains poorly defined. In this study, we investigate the roles of SAM68 and QKI during early lineage specification of mouse embryonic stem cells (mESCs). Using CRISPR–Cas9-mediated knockout models, we demonstrate that both proteins contribute to mESC proliferation, self-renewal, and efficient cardiomyocyte differentiation. In agreement with previous reports, our data support a role for QKI in the regulation of cardiac differentiation programs. This work uncovers an unprecedented role for SAM68 in cardiomyocyte specification and highlights different layers of SAM68-mediated post-transcriptional regulation driving cell differentiation. Specifically, we show that SAM68 modulates alternative splicing and promotes the biogenesis of a subset of cardiac-enriched circRNAs through binding to intronic regions flanking circularizing exons. We highlight how SAM68 associates with untranslated regions of fundamental differentiation-related transcripts, including Gata4 mRNA—a key transcription factor regulating cardiogenesis—participating in translation regulation. Lastly, characterization of the SAM68 interactome sheds light on the ribonucleoprotein complexes through which it shapes cell fate regulation. Materials and methods Cell cultures All cell lines used in this study were grown at 37°C, 5% CO 2 , and tested for mycoplasma contamination. E14 mESCs were obtained from the Tissue Engineering Unit of the Center for Genomic Regulation ( Supplementary Table S1 ). For maintenance culture cell were seeded onto plates pre-coated with 0.1% gelatin (ES-006-B, Millipore) under feeder-free conditions at a density of 1.5 × 10 4 cells/cm 2 in knock-out Dulbecco's Modified Eagle Medium (DMEM) supplemented with 15% embryonic stem cells qualified FBS (16141079, Gibco), 1% penicillin/streptomycin (15140122, Gibco), 1% L-glutamine (25030081, Gibco), 1% sodium pyruvate (11360070, Gibco), 1% non-essential amino acids (NEAA) (11140068, Gibco), 50 mM β-mercaptoethanol (31350010, Invitrogen), leukemia inhibitory factor (LIF) (ESG-1106, Millipore), and the 2i inhibitors GSKi (CHIR-99021, Tocris) and MEKi (PD0325901, Tocris). For differentiation culture, the medium was supplemented with 10% FBS. For normal maintenance, cells were passaged any other day by dissociating them with the ACCUTASE ® cell detachment solution (SCR005, Millipore). The ES Rex1-dGFP was obtained by Austin Smith’s laboratory. For maintenance culture cells were plated at a density of 2–3 × 10 4 cells/cm 2 always on pre-coated dishes. Cells were maintained in 2i medium made by mixing Neurobasal (21103048, Gibco) and DMEM/F12 (12634010, Gibco) 1:1 supplemented with N2 factor (17502-048, Gibco) and B27 (17504-047, Gibco), 1% L-glutamine, 1% NEAA, 1% sodium pyruvate, 1% penicillin/streptomycin, β-mercaptoethanol (31350010, Invitrogen), 1 μM PD0325901 (444968, Calbiochem), and 3 μM CHIR99021 (361571, Calbiochem). Cells were passaged any other day. Beyond splicing, RBPs regulate stem cell fate through additional post-transcriptional mechanisms, including control of microRNA biogenesis [ 19 , 20 ], regulation of mRNA translation [ 21 , 22 ], modulation of mRNA stability [ 23 , 24 ], control of RNA export [ 25 , 26 ], and alternative polyadenylation [ 27 ]. Furthermore, emerging layers of regulation—such as m 6 A RNA modifications, RNA granule dynamics, and maturation of chromatin topology—contribute to the precise control of gene expression and mRNA turnover required for cellular plasticity [ 28 – 30 ]. Collectively, these mechanisms highlight the diversity and versatility of post-transcriptional regulation underlying cell fate decisions. For lentiviral production, human embryonic kidney (HEK) 293T cells packaging cell line was used. Cells were cultured in DMEM supplemented with 10% FBS, 1% L-glutamine, and 1% penicillin/streptomycin. CRISPR–Cas9-mediated genome editing Genome-editing was carried out by transfecting E14 cells with the bicistronic plasmid pSpCas9 PX459 (1 μg per well in a six-well plate) encoding for both Cas9 and the single-guide RNAs (sgRNAs) targeting the first exon of either Sam68 or Qki . The transfected cells were then selected with puromycin and Fluorescence-Activated Cell Sorting (FACS) at clonal density in a 96 multi-well plate. Clones that were successfully edited were then Sanger-sequenced and the absence of the protein product verified by western blot. The GFP-SAM68 knock-in in mESC was performed using CRISPR–Cas9-mediated genome editing as well. E14TG2a WT mESCs were transfected with a bicistronic plasmid (pSpCas9 459–sgRNA63) targeting the 5′ end of the Sam68 locus, along with a linearized homology-directed repair (HDR) donor template containing GFP to be inserted in-frame at the N-terminus of the Sam68 coding sequence. Transfected cells were selected with puromycin and screened for GFP fluorescence using an inverted fluorescence microscope. Genomic DNA from fluorescent clones was extracted and analyzed by Polymerase Chain Reaction (PCR) using primers flanking the insertion site. Clones were validated by PCR and Sanger sequencing to confirm precise integration and zygosity. Western blotting confirmed the production of the GFP-SAM68 fusion protein (∼100 kDa). See Supplementary Table S2 for sgRNA sequences. Cell differentiation To set up the embryoid body (EB) assay, E14 mESCs suspended in KO-DMEM supplemented with 10% serum at a density of 5 × 10 4 cells/ml were seeded in hanging drops of 20 μl onto the surface of a 150 mm culture plate filled with PBS to avoid evaporation of the drops. After three days, the hanging drops containing the EBs were flushed and seeded on a low attachment plate. For the generation of EBs collected at day 10 (d10) of differentiation, E14 mESCs were directly plated into Corning spheroid microplates (Corning) at a density of 5 × 10 4 cells/ml. Medium was replenished during differentiation, and EBs were collected at d10 for downstream experiments. In order to derive cardiomyocytes, at day three (d3), the cells were seeded on a gelatin pre-coated 96 multi-well plate and cultured for a maximum of 8 more days. Teratoma formation, proliferation, clonogenicity, and exit-from-pluripotency assays, as well as lentiviral production and subsequent knockdown of Sam68 and Qki , are described in the Supplementary Data under the section Additional Materials and Methods. Analysis of protein and RNA expression Methods for assessing protein expression (i.e. western blot analysis and immunofluorescence) and RNA expression (including circRNAs) are described in detail in the section Additional Materials and Methods. Cloning and mESC plasmid transfection The sequence encoding the SAM68 ORF was PCR-amplified using CloneAmp HiFi PCR Premix (TakaraBio) from complementary DNA (cDNA) generated from RNA extracted from N2a cells, using primers listed in Supplementary Table S2 . The amplified fragment was cloned into the pcDNA3.1 (+)eGFP plasmid (Addgene #129 020) using NEBuilder ® HiFi DNA Assembly (New England Biolabs). Briefly, the eGFP ORF was removed from the plasmid by PCR and replaced with the SAM68 ORF amplified with primers containing 15 nt overlaps with the pcDNA3.1 (+)-N-eGFP backbone at the insertion site. Site-directed mutagenesis was performed to generate the V229F SAM68 mutant using rolling-circle PCR with overlapping primers. Following DpnI digestion, the amplified plasmid was transformed into DH5α competent cells and positive clones were identified by Sanger sequencing. For transfection experiments, pcDNA3.1-SAM68 WT or pcDNA3.1-SAM68 V229F plasmids were introduced into undifferentiated mESCs using Lipofectamine ™ Stem Transfection Reagent (Thermo Fisher Scientific), according to the manufacturer’s instructions. A reverse transfection strategy was employed in 10 cm dishes, seeding 1 × 10⁶ undifferentiated mESCs with 12 µg of plasmid DNA. Complete KO-DMEM medium was added the following day, and cells were collected 48 h post-transfection. The 5′ and 3′ UTRs of Gata4 mRNA (isoform ENSMUST00000118022.8) were PCR-amplified from cDNA using CloneAmp HiFi PCR Premix (Takara Bio) and primers listed in Supplementary Table S2 . The 5′ UTR fragment was cloned into the psiCHECK-2 vector (Promega), upstream of the Rluc CDS, using NEBuilder ® HiFi DNA Assembly (New England Biolabs), as described above. The 3′ UTR was cloned using standard restriction enzyme-based cloning with XhoI and NotI (New England Biolabs), downstream of the Rluc CDS. Deletion of SAM68 binding motifs was performed by rolling-circle PCR using CloneAmp HiFi PCR Premix (Takara Bio), as previously described (see Supplementary Table S2 for primer sequences). For the luminescence assay, 1.5 × 10⁵ mESCs were seeded in six-well plates and transfected with 2 µg of plasmid DNA using Lipofectamine ™ Stem Transfection Reagent (Thermo Fisher Scientific), following the manufacturer’s instructions and applying a reverse transfection strategy. The medium was replaced the next day, and cells were collected 48 h post-transfection for luminescence measurement. Dual luciferase assay Cells were washed with PBS, and Renilla (Rluc) and Firefly (Fluc) luciferase activities were measured using the Dual-Glo ® Luciferase Assay System (Promega) according to the manufacturer’s instructions, using an Infinite 200 PRO microplate reader (Tecan). Rluc activity was first normalized to Fluc to account for transfection efficiency. As a second normalization step, values were normalized to the luminescence of the empty psiCHECK-2 vector (lacking the Gata4 5′ or 3′ UTR) to control for background effects. Experiments were performed with at least four biological replicates. Statistical significance was assessed using an unpaired two-tailed t -test, with P < 0.005 considered significant. RNA-sequencing To perform RNA sequencing, total RNA was extracted from WT, Sam68 −/− , and Qki −/− EBs using the Maxwell RNA Cells Kit (AS1270, Promega), according to the manufacturer’s instructions, with three biological replicates per condition. DNA-free RNA was then quantified using the Nanodrop and quality checked with the Bioanalizer. rRNA depletion and library preparation were performed with the TruSeq stranded total RNA Library Human/Mouse/Rat (Illumina). The sequencing was performed using 2 × 125 bp paired-ends reads on a HiSeq 2500 sequencer with HiSeq v4 chemistry. The raw reads were de-multiplexed, by the bcl2fastq Illumina software. The IlluminaUniversal adaptor was trimmed off the raw reads with Skewer (version 0.2.2), option “-x AGATCGGAAGAG”. The quality of both raw and trimmed reads was assessed with the FastQC tool. Trimmed reads were aligned to the Mus musculus genome (mm10/GRCm38) with the STAR RNA-seq mapper (version 2.5.3a) using option “– quantMode GeneCounts” and the M14 version of the Gencode annotation in order to obtain raw counts per gene and per sample (the 4th column of the resulting counts file that corresponds to a “reverse stranded” protocol was kept). Bam files were converted to bigwig format with Samtools, so as to be loaded into the UCSC genome browser. The DESeq2 R/Bioconductor package was used to assess differential expression at gene level between experimental conditions. A gene was considered differentially expressed if the corresponding FDR-adjusted P -value was < 0.01 and the absolute fold change in log 2 scale was > 0.6. Last, we used vast-tools (version 2.0.0) in order to detect alternative splicing events. MATT was used to filter vast-tools combined output. In this case, a dPSI threshold of 25 (described as stringent by the authors) was applied to filter significant results. Ribosome profiling WT and Sam68 −/− cells were treated with cycloheximide (100 μg/ml; Sigma–Aldrich) for 15 min at 37°C, then lysed in buffer (20 mM Tris–HCl pH 7.4, 150 mM NaCl, 5 mM MgCl₂, 1 mM DTT, 1% Triton X-100, 100 μg/ml cycloheximide, and 25 U/ml Turbo DNase I (Invitrogen) using glass beads (3 × 5 s vortexing). Lysates were incubated on ice for 10 min and clarified by centrifugation (10 000 rpm, 10 min). RNA corresponding to 8–10 OD₂₆₀ units was digested with RNase I (100 U/μl, Invitrogen) for 15 min at 25°C. Digestion was stopped with SUPERase•In (200 U total; Invitrogen), and samples were loaded onto a 34% (w/v) sucrose cushion (Beckman Coulter). Samples were centrifuged at 70 000 rpm for 4 h at 4°C. Pellets were resuspended in the lysis buffer from the Maxwell 16 LEV simplyRNA Cells Kit (Promega), and RNA was extracted according to the manufacturer’s instructions. For size selection, RNA (5 μg) was resolved on a pre-run 15% TBE–urea polyacrylamide gel (200 V, 15 min pre-run; 200 V, 65 min run) following denaturation (80°C, 90 s) in TBE–urea sample buffer. Gels were stained for 5 min with SYBR Green and fragments of 28–32 nt were excised. RNA was recovered by incubation at 70°C for 10 min followed by filtration through a spin column with Whatman paper. RNA was precipitated overnight at −80°C using 40 μl of 3 M sodium acetate, 1 μl of GlycoBlue (Invitrogen), and 500 μl of isopropanol. Samples were centrifuged (20 000 g , 30 min, 4°C), washed with 70% ethanol at max speed, air-dried, and resuspended in RNase-free water. Ribosomal RNA was depleted using the Ribo-Zero kit (Illumina), according to the manufacturer’s instructions. After purification, RNA was dephosphorylated using T4 PNK (New England Biolabs) in the presence of SUPERase•In (37°C, 1 h), followed by heat inactivation at 70°C. A pre-adenylated linker (1.5 μl, 0.5 μg/μl) was denatured (80°C, 90 s), cooled to room temperature, and ligated using T4 RNA ligase (truncated; New England Biolabs) in the presence of PEG 8000 and SUPERase•In (2.5 h, RT). RNA was purified, resolved on a 15% TBE–urea gel, and the ligated fraction was excised and recovered as described above. Reverse transcription was performed using SuperScript III (Invitrogen) with a specific primer ( Supplementary Table S2 ) at 50°C for 30 min. cDNA was purified, size-selected by PAGE, and circularized using 100 units of CircLigase I (CL4111K, Epicentre), 2 μl of circ-ligase buffer, 1 mM of ATP, and 50 mM of MnCl 2 for 1 h at 60°C. Following an additional rRNA depletion step, cDNA was PCR-amplified using Phusion polymerase (New England Biolabs) for 14 cycles. Libraries were size-selected (∼180 nt) on an 8% native polyacrylamide gel, purified, quality-checked, and sequenced (single-end, 50 bp) on an Illumina HiSeq 2500 platform. SAM68 pull-down and mass-spectrometry analysis Pull-down of GFP-SAM68 complexes at different time points (d0, d3, and d10) was performed according to GFP trap protocol. GFP-Trap ® _Magnetic Agarose Beads protocol (gtma-100, Chromotek) for immunoprecipitation of GFP-fusion proteins from mammalian cell extract was performed, according to manufacturer’s instructions. Cell pellets were resuspended in 200 μl of ice-cold lysis buffer by pipetting or using a syringe. Tubes were placed on ice for 30 min with extensive pipetting every 10 min to disrupt membranes. Cell lysates were centrifuged at 20 000 g for 10 min at 4°C and transferred to pre-cooled tubes. Three hundred microliters of dilution buffer was added to lysate. For each sample, 25 μl of GFPTrap ® _MA beads were diluted into 500 μl of ice-cold dilution buffer and magnetically separated until the supernatant was clear. Equilibrated beads were washed twice. Lysate was added to equilibrated GFP-Trap ® _MA and the samples were immunoprecipitated for 2 h at 4°C. Beads were magnetically separated until supernatant was clear and washed in 500 μl of dilution buffer. After two additional washes, beads were resuspended in 100 μl of 2× SDS-sample buffer. Samples were boiled for 10 min at 95°C shaking to dissociate immunocomplexes from the beads and the efficiency of the GFP-trap pull-down was verified by SDS–PAGE and western blot. For proteomic analysis, snap-frozen samples in five biological replicates were sent to the Buchmann Institute of Molecular Life Sciences (Frankfurt, Germany). Samples were processed by on-beads LysC/Trypsin digestion and released peptides were fractionated into 3 SCX fractions. The MS data were analyzed by Dr. Giulia Calloni using the software environment MaxQuant version 1.5.3.30 (Cox et al . 2008). Proteins were identified by searching MS and MS/MS data against the mouse complete proteome sequences from UniProtKB, version of January 2016, containing 50 189 sequences. Carbamido-methylation of cysteines was set as fixed modification. N-terminal acetylation and oxidation of methionines were set as variable modifications. Up to two missed cleavages were allowed. The initial allowed mass deviation of the precursor ion was up to 4.5 ppm and for the fragment masses it was up to 20 ppm. The “match between runs” option was enabled to match identifications across samples within a time window of 2 min of the aligned retention times. The maximum false peptide and protein discovery rate was set to 0.01. Protein matching to the reverse database or identified only with modified peptides were filtered out. Relative protein quantitation was performed using the LFQ algorithm of the Maxquant with a minimum ratio count of 1 [ 50 ]. Bioinformatic data analysis was performed using Perseus (version 1.5.2.6) [ 51 ]. The proteins with minimum four valid values in at least one group (pulldown/background) of five biological replicates were considered as quantified and used for downstream analysis. Proteins enriched in the pull-down over background control at different time points during cell differentiation were identified by two-sample t -test at different permutation-based FDR cutoffs (0.001, 0.01, and 0.05) and s0 = 0.1. Protein co-immunoprecipitation EB pellets collected at day 0, 3, and 10 of differentiation were lysed in ice-cold buffer containing 150 mM NaCl, 50 mM Tris–HCl (pH7.5), 0.5 mM EDTA (pH8.0), 0.5% NP-40, 1 mM DTT, protease inhibitors (Complete EDTA-Free, Merck), and RNase inhibitor (Super RNase-In, Thermo Fisher Scientific). Lysates were incubated on ice for 10 min and centrifuged at 12 000 g for 10 min at 4°C. Protein concentration was determined using the BCA Protein Assay Kit (Thermo Fisher Scientific). Protein G magnetic beads (Dynabeads, Thermo Fisher Scientific) were pre-washed with lysis buffer and incubated with 3 µg of anti-ILF3 antibody (ProteinTech) or control Rabbit IgG (Cell signaling) for 1 h at 4°C on a rotating wheel. In parallel, 750 µg of lysate was treated with Turbo DNase (Thermo Fisher Scientific) for 30 min at 37°C. Ten percent of each lysate was saved as input. The remaining lysate was incubated with the antibody-bound beads for 2 h at 4°C. Beads were then washed three times with lysis buffer and once with cold PBS. Bound proteins were eluted in PBS, mixed with 6× Laemmli sample buffer, and denatured at 95°C for 5 min. Eluted proteins and 1% of input samples were separated on 4%–12% Bis–Tris gels (Thermo Fisher Scientific), transferred to nitrocellulose membranes (Bio-Rad), and blocked with 5% non-fat milk in TBS-T (0.1% Tween-20) for 1 h at room temperature. Membranes were probed with the following primary antibodies: anti-ILF3 (HPA001897-100UL, Merck), anti-ILF2 (HPA007484-100UL, Merck), and anti-SAM68 (PA5-96507, Thermo Fisher Scientific) (see Supplementary Table S1 ), followed by Clean-Blot IP Detection Reagent (Thermo Fisher Scientific). Signal detection was performed using Clarity Max Western ECL substrate (Bio-Rad) and visualized on a ChemiDoc MP imaging system (Bio-Rad). The SAM68 co-immunoprecipitation procedure in the presence or absence of RNase I ( Supplementary Fig. S2F ) is described in the Supplementary Data file, under the section additional materials and methods. Nuclear cross-linking and RNA immunoprecipitation assay Nuclear cross-linking and SAM68 immunoprecipitation were performed in mESCs at d0 and in EBs collected at d3 and d10 of differentiation as previously described [ 52 ]. Cells at d0 were grown on 15 × 15 cm 2 plates, whereas d3 and d10 EBs were dissociated into single cells using TrypLE ™ Express Enzyme (1×). Cells were then irradiated twice with 150 mJ/cm² UVC at 254 nm using an ultraviolet crosslinker (Bio-Link 254, Vilber). Cells were harvested and resuspended in cell lysis buffer (5 mM PIPES pH8.0, 85 mM KCl, and 0.5% NP-40) supplemented with protease inhibitors (Complete EDTA-Free, Merck) and RNase inhibitor (Super RNase-In, Thermo Fisher Scientific). Following 10 min incubation on ice, cells were homogenized with a Dounce homogenizer. Nuclei were pelleted by centrifugation at 4000 rpm for 5 min at 4°C. Isolated nuclei were lysed in RNA immunoprecipitation (RIP) buffer (50 mM Tris–HCl pH 7.4, 100 mM NaCl, 1% Igepal CA-630, 0.1% SDS, and 0.5% sodium deoxycholate) supplemented with protease and ribonuclease inhibitors. Nuclear lysates were sonicated using a Bioruptor Pico (Diagenode; 3 cycles of 30 s ON/OFF at 4°C) and centrifuged at 12 000 rpm for 10 min at 4°C to remove debris. Protein concentration was measured using the BCA Protein Assay Kit (Thermo Fisher Scientific), and 500 µg of protein was used for each immunoprecipitation. Protein G Dynabeads (Thermo Fisher Scientific) were pre-washed in RIP buffer and incubated with 3 µg of anti-SAM68 antibody (Thermo Fisher Scientific) or rabbit IgG isotype control (Cell Signaling Technology) ( Supplementary Table S1 ) for 1 h at 4°C on a rotating wheel. Ten percent of the lysate was reserved as input and divided for RNA (80%) and protein (20%) analysis. The remaining lysate was incubated with antibody-bound beads for 2 h at 4°C. Immunoprecipitates were washed three times with RIP buffer for 5 min and resuspended in PBS. 80% of the sample was used for RNA extraction and the remaining 20% for IP assessment by western blot. Beads were treated with recombinant Proteinase K (PCR Grade, Merck Sigma–Aldrich) at 37°C for 30 min followed by 50°C for 30 min. RNA was extracted with TRIzol, purified using the RNA Clean and Concentrator-5 kit (Zymo, R1016), and analyzed by real-time PCR as previously described ( Supplementary Table S2 ). SAM68 RNA immunoprecipitation assay RIP was performed under native lysis conditions in mESCs differentiated into EBs and collected at d10 of differentiation (Fig. 5D ). Disassembled EBs were lysed in 150 mM NaCl, 25 mM Tris–HCl pH7.4, 5 mM EDTA, 1 mM DTT, and 0.5% NP-40 supplemented with protease and ribonuclease inhibitors. Lysates were incubated on ice for 10 min and centrifuged at 12 000 g for 10 min at 4°C. Protein concentrations were measured using the BCA Protein Assay Kit (Thermo Fisher Scientific). Protein G Dynabeads (Thermo Fisher Scientific) were washed in RIP buffer and incubated with 3 µg of anti-SAM68 antibody (Thermo Fisher Scientific) or control rabbit IgG (Cell Signaling Technology) for 1 h at 4°C. In parallel, 750 µg of clarified lysate was treated with Turbo DNase (Thermo Fisher Scientific) for 20 min at 37°C. Ten percent of the lysate was set aside as input and divided for RNA (80%) and protein (20%) analysis. The remaining lysate was incubated with the antibody-bound beads for 2 h at 4°C with rotation. Beads were washed three times with RIP wash buffer (200 mM NaCl, 25 mM Tris–HCl pH7.4, 5 mM EDTA, 1 mM DTT, and 0.5% NP-40). RNA was isolated by adding TRIzol directly to the beads, followed by purification using the RNA Clean and Concentrator-5 kit (Zymo, R1016). Eluted RNAs were analyzed by real-time PCR as previously described ( Supplementary Table S2 ). Figure 5. Open in a new tab SAM68 contributes to translational regulation of Gata4 and cardiac transcripts during differentiation. ( A ) Comparison of RNA abundance (RNA-seq) and translation efficiency (Ribo-seq) in Sam68 −/− versus WT EBs at d10 of differentiation. Scatter plot shows log₂ fold changes (FC) of Ribo-seq versus RNA-seq. A negative correlation (Pearson correlation coefficient = −0.783) indicates widespread uncoupling between transcript levels and translation in the absence of SAM68. Selected genes are highlighted and classified as cardiac transcription factors, structural proteins, or ion transport-related genes. ( B ) Overlap between SAM68-bound transcripts and translationally downregulated mRNAs in Sam68 −/− EBs. Top, overlap of transcripts bound by SAM68 in their 5′ UTRs or 3′ UTRs with translationally downregulated mRNAs. Bottom, Gene Ontology (GO) analysis of overlapping targets reveals enrichment for biological processes related to cell differentiation and cardiac muscle function (Panther, Fisher’s exact test, FDR < 0.05). ( C ) iCLIP2 profile showing SAM68 binding across the Gata4 mRNA at d10 of differentiation, with peaks detected in both the 5′ UTR and 3′ UTR. ( D ) RIP analysis at d10 of differentiation. Immunoprecipitation was performed using a SAM68-specific antibody or control IgG. RNA enrichment in SAM68 RIP samples is expressed as % input relative to the IgG control. Gata4 mRNA and its 5′ UTR are enriched in SAM68 immunoprecipitates, whereas negative controls ( Actin, Rplp0 ) show no enrichment. P -values were calculated using multiple t -tests relative to the negative control ( n = 4). Representative western blot showing SAM68 immunoprecipitation efficiency is included; VINCULIN and GAPDH are shown as controls. ( E ) Ribo-seq coverage profiles (IGV tracks) across the Gata4 5′ UTR and first Gata4 exon in WT, Sam68 −/− , and Qki −/− EBs at d10 of differentiation, showing reduced ribosome occupancy in the absence of SAM68. Western blot analysis of GATA4 protein levels confirms decreased protein expression in Sam68 −/− cells compared to WT, with tubulin as loading control. ( F ) Schematic representation of SAM68 binding sites in Gata4 UTRs and nearby RBP motifs identified within ±20 nt of SAM68 peaks. Candidate RBPs include factors involved in post-transcriptional regulation, 11 of which (reported in the figure) are also present in the SAM68 interactome, suggesting potential combinatorial regulation at the level of mRNA translation. RNA-sequencing bioinformatic analyses The bioinformatic analysis was performed by the bioinformatics core facility of the CRG. Briefly, the adaptor was trimmed off the raw sequences (fastq files) with skewer (version 0.2.2). The quality of both raw and trimmed reads was assessed with the FastQC tool. The reads that aligned (using bowtie2 version 3.2.0) to the rRNAs or the tRNAs (coordinates from the UCSC table browser) were removed (using bowtie2 with the following parameters: –no-unal –mp 3 -L 10 -N 1 –local -D 10 -R 3). All the remaining reads were size-selected: only reads ranging from 22 to 36 nt were considered for further analysis to capture the ribosome protected fragments (~30 bp). The reads were then aligned to the Gencode M14 version of the genome (mm10/GRCm38) using the STAR mapper (version 2.5.3a) with parameters: –winAnchorMultimapNmax 100 –seedSearchStartLmax 15 – outFilterScoreMinOverLread 0 –outFilterMatchNminOverLread 0 –outFilterMatchNmin 10. Read coverage around the TSS was assessed, for each selected read length separated, to define an offset/shift value of ribosome position (the shifts are then applied to the read mapping position to obtain the approximate position of the translated codon). The reads were finally sorted out according to their alignment to either the coding sequence (CDS) or the 5′- or 3′-untranslated regions. Analysis of circular RNAs (circRNAs) Trimmed fastq files were mapped again to the Mus musculus reference genome using bwa-mem: mapping results were then fed into the CIRI2 (version 2.0.6) pipeline for detection and quantification of circular RNAs [ 53 ]. The “junction_reads” results from CIRI2 were used to assess circular RNA differential expression between experimental groups. The analysis provided the number of counts of the circRNA and the number of counts of the linear counterpart. To specifically select circRNA altered at the biogenesis level, we only considered circRNAs that had a FDRpVal < 0.01 (column “FDR_circ”) whose linear counterparts were not differentially expressed (column FDRlin > 0.01) or that were differentially expressed in the opposite direction of the circRNA (column FDRlin < 0.01 and column logFC_lin has the opposite sign of column logFC_circ). SAM68 individual-nucleotide-resolution CLIP 2 (iCLIP2) SAM68 iCLIP was performed as previously described with minor modifications. Crosslinking and immunoprecipitation followed the updated iCLIP3 protocol [ 54 ], whereas the adapter sequences and PCR amplification of the libraries were performed according to the original iCLIP2 protocol [ 55 ]. In brief, mESCs were collected at d0 or allowed to differentiate in EBs as previously described (above) and collected at d3 and d10 of differentiation. Cells were irradiated with UVC light at 254 nm using a CL-1000 UV crosslinker (Bio-Link 254, Vilber) at a dose of 150 mJ/cm² to induce covalent protein–RNA crosslinking. For the d3 and d10 time points, EBs were dissociated into single cells using TrypLE ™ Express Enzyme (1×), no phenol red (Thermo Fisher Scientific), prior to crosslinking to ensure uniform UV exposure. Crosslinked cells were lysed in 50 mM Tris–HCl pH 7.4, 100 mM NaCl, 1% Igepal CA-630, 0.1% SDS, and 0.5% sodium deoxycholate, supplemented with protease inhibitors, and sonicated using a Bioraptor Pico (Diagenode; 3 cycles, 30 s ON/OFF). Subsequently, RNA was digested with RNase I (1:200 dilution, AM2295, Thermo Fisher Scientific) to generate RNA fragments of 50–200 nt (see Supplementary Fig. S3H for optimization of RNase I digestion). SAM68 protein immunoprecipitation was performed using a SAM68-specific antibody (SAM68 Polyclonal Antibody; PA5-96507; Thermo Fisher Scientific) conjugated to Dynabeads Protein G beads (10004D, Thermo Fisher Scientific) for 1 h at 4°C. Protein–RNA complexes were visualized using 3′-end pCp-IR750 RNA labeling of 10% of the beads, while the remaining 90% underwent the first adapter ligation with the pre-adenylated L3-App oligo ( Supplementary Table S2 ). After SDS–PAGE and transfer to a nitrocellulose membrane, RNA–protein complexes were isolated from the membrane, treated with proteinase K (Merck), and RNA was extracted using the RNA Clean and Concentrator-5 kit (Zymo, R1016). cDNA was generated by reverse transcription using SuperScript III (18080085, Thermo Fisher Scientific); reverse transcription terminates at the crosslinked site, generating truncated cDNAs. A second adapter ( Supplementary Table S2 ), containing a unique molecular identifier (UMI) and an experimental barcode, was ligated to the 3′ OH of the cDNAs. MyONE beads were used to purify the cDNA, and the library was subsequently subjected to two rounds of PCR amplification (6 and 15 cycles, respectively) and size-selected using ProNex chemistry (NG2001, Promega). Library quality was assessed using a TapeStation system with the High Sensitivity D1000 Kit. iCLIP2 libraries were sequenced on an Illumina NovaSeq 6000 with 100 bp single-end reads. Five biological replicates were used for each condition (i.e. d0, d3, and d10). As a negative control, the entire iCLIP2 procedure was performed using normal Rabbit IgG (Cell Signaling #2729) instead of the SAM68-specific antibody until the TapeStation analysis. Since no signal was detected after PCR amplification, these samples were not sequenced. Processing iCLIP2 data To extract RNA-binding signals from raw iCLIP2 sequencing reads and derive single-nucleotide crosslink sites in a single step, the racoon_clip pipeline (version 2.0.11, 10.18129/B9.bioc.txdbmaker Klostermann & Zarnack 2024) was applied in “peak” mode. Initial quality control was performed using FastQC (version 0.12.1, 10.18129/B9.bioc.txdbmaker Andrews 2010). Reads were filtered to retain a minimum base quality of 10 within the barcode region, ensuring accurate assignment of samples and PCR duplicates. Demultiplexing and trimming of 3′ adapter sequences (Solexa_P3: AGATCGGAAGAGCGGTTCAGCAGGAATGCCGAGACCGATCTCGTATGCCGTCTTCTGCTTG) and 5′ unique molecular identifiers (UMIs) were performed using FLEXBAR (version 3.5.0, cite xx Roehr 2017). Each read contained an experimental barcode comprising a 5 nt UMI, a 6 nt sample barcode, and a 4 nt UMI. The barcodes used are listed in Supplementary Table S2 . Reads were aligned to the mouse reference genome GRCm38.p5 using STAR (version 2.7.11a, 10.18129/B9.bioc.txdbmaker) [ 56 ] with GENCODE annotation (version M14, 10.18129/B9.bioc.txdbmaker [ 57 ]). PCR duplicates were removed using umi_tools dedup (version 1.1.6, 10.18129/B9.bioc.txdbmaker) [ 58 ]. Deduplicated BAM files were converted to BED format using BEDTools bamtobed (version 2.31.1, cite 10.18129/B9.bioc.txdbmaker) [ 59 ]. Single-nucleotide crosslink sites were defined as positions located 1 nt upstream of the 5′ end of aligned reads using BEDTools coverage and BEDTools shift. For visualization in genome browsers, 1 nt crosslink BED files were converted to bigWig format using bedGraphToBigWig (version 472, 10.18129/B9.bioc.txdbmaker) [ 60 ]. Peak calling was performed on merged BAM files of sample groups using PureCLIP (version 1.3.1, 10.18129/B9.bioc.txdbmaker) [ 61 ]. Binding site definition SAM68 binding sites at day 0, day 3 and day 10 were defined independently using the R/Bioconductor package BindingSiteFinder (version 2.6.0, 10.18129/B9.bioc.BindingSiteFinder). PureCLIP sites were imported from BED files, restricted to standard chromosomes, and filtered to remove mitochondrial loci. Strand-specific crosslink coverage from five replicates per time point was imported from bigWig files and assembled into BSFDataSet objects. Gene and transcript annotations were generated from the GENCODE vM14 GTF using txdbmaker (version 1.4.2, 10.18129/B9.bioc.txdbmaker) and comprised CDS, intronic, 3′ UTR, and 5′ UTR regions. Binding sites were computed with the BSFind function using cutoff.geneWiseFilter = 0, repro.nReps = 3 and repro.cutoff = 0.05. A binding-site width of 5 nt was selected for all three time points. Final binding sites were then assigned to genes and transcript regions. For gene assignment binding sites that could not be unambiguously assigned to one specific gene type were removed with the assignToGenes function and overlaps = “remove”. The transcript-region assignment was performed hierarchically in the order 5′ UTR, intron, CDS, and 3′ UTR. Overlapping SAM68 binding sites with regions upstream and downstream of circRNAs To quantify SAM68 binding in the vicinity of circRNAs, we counted binding sites located upstream or downstream of both regulated and non-regulated circRNAs. As a negative control set of circRNAs we used randomly selected circRNAs (from CircAtlas v2.0; [ 62 ]) that are not regulated by SAM68, deriving from different host genes than the regulated circRNAs, but expressed at the corresponding time point (average reads in three replicates > 0). For each circRNA, 1000 nt upstream and downstream windows were generated using the flank () function from the GenomicRanges Bioconductor package (version 1.60.0 https://doi.org/10.18129/B9.bioc.GenomicRanges ), and overlaps were quantified with countOverlaps (). Statistical significance was assessed using Fisher’s exact test as implemented in fisher.test () from the stats package (R version 4.5.2). While enrichment was initially evaluated across multiple window sizes ( Supplementary Fig. S3L ), Fig. 3H displays results for the selected 1000 nt window. Figure 3. Open in a new tab SAM68 promotes cardiac-enriched circRNA biogenesis via interaction with flanking introns. ( A ) Venn diagrams illustrating the number of deregulated AS events in Sam68 −/− and Qki −/− cell lines compared to WT across differentiation stages (d0, d3, and d10). ( B ) Heatmap displaying the distribution of AS event types observed in Sam68 −/− and Qki −/- cell lines across differentiation stages. Values represent the percentage of events in each splicing category relative to the total number of AS events detected at each time point. ( C ) Histogram showing the number of circRNAs significantly upregulated (Up) or downregulated (Dw) in Sam68 −/− and Qki −/− cells at d0, d3, and d10 of differentiation. ( D ) Overlap between SAM68-regulated circRNAs (circRNAs with log 2 FC < 0 and FDR < 0.05, whose corresponding linear transcripts are either not significantly regulated, FDR > 0.05, or show an opposite log 2 FC) and cardiac circRNAs at d0, d3, and d10 of differentiation. Venn diagrams show the intersection between circRNAs downregulated in Sam68 −/− cells and a previously defined set of cardiac circRNAs (Tan, W.L.W., et al . Cardiovasc Res 2017). Statistical significance of the overlap was assessed using a hypergeometric test. ( E ) Visualization of SAM68–RNA complexes during differentiation. Schematic representation of the SAM68 RNA interactome capture performed at d0, d3, and d10 of differentiation. Cells were UV crosslinked (254 nm ) and SAM68–RNA complexes were immunoprecipitated and partially digested with RNase I (optimization shown in Supplementary Fig. S3H ). Following immunoprecipitation, 90% of the material was used for adapter ligation and library preparation, while 10% was used for RNA labeling to visualize protein–RNA complexes as described in iCLIP3. pCp-IR750–labeled RNA associated with SAM68 complexes was resolved on nitrocellulose membranes after RNase I digestion (1:200 dilution). Rabbit IgG immunoprecipitation served as a negative control. The same membranes were subsequently used for western blot analysis to verify SAM68 immunoprecipitation. ( F ) Distribution of binding sites across transcript regions at d10 of differentiation (refer to Supplementary Fig. S3J ). ( G ) De novo motifs identified for SAM68 from iCLIP binding sites at day 10 (d10). Binding sites were extended by ± 10 nt to include local sequence context. The top three enriched motifs are shown with corresponding e -values (see Supplementary Fig. S3K for d0 and d3). ( H ) Fraction of SAM68-regulated and control (non-regulated) circRNAs with SAM68 binding in flanking intronic regions (1000 nt upstream or downstream of circularizing exons) across differentiation stages. Statistical significance was assessed using Fisher’s exact test. ( I ) Motif enrichment in circRNAs downregulated in Sam68 −/− cells at d10. Motifs were ranked by frequency of occurrence within 1000 nt upstream and downstream of circRNA junctions. The top motif (UUUUAAAA) is enriched compared to a negative control set of non-regulated circRNAs, as assessed by a chi-squared test ( P = 0.05). ( J ) Association of circRNA-producing pre-mRNAs with SAM68 at d0 of differentiation assessed by RIP–qPCR following UV (254 nm ) crosslinking of nuclear extracts. Representative western blot showing SAM68 immunoprecipitation is shown; H3 and TDP-43 were used as controls for nuclear fraction isolation. Immunoprecipitation was performed using a SAM68-specific antibody or control IgG. RNA enrichment in SAM68 RIP samples is shown as % input relative to the IgG control. P -values were calculated using multiple t -tests relative to the negative pre-mRNA control ( n = 6). Additional pre-mRNA targets analyzed at d0, d3, and d10 are shown in Supplementary Fig. S3 . ( K ) Top, genome browser tracks showing SAM68 iCLIP2 signal and binding sites mapped within introns flanking circRNA-forming exons (≤2000 nt from the first and last circularizing exons). Bottom, reduced production of circFarsa [ 5 – 7 ], circPibf1 (6,7,8L,9) , and circHuwe1 [ 3 – 7 ] in Sam68 −/− cells is rescued by overexpression of SAM68 WT but not by the RNA-binding-defective mutant SAM68 V229F (see Supplementary Fig. S3N ). Expression and localization of SAM68 WT and V229F were verified by western blot following cytoplasmic (c) and nuclear (n) fractionation. circRNA levels were measured by qPCR, normalized to a stable linear RNA (i.e. linear Farsa mRNA), and expressed relative to WT. Statistical significance was assessed using one-way ANOVA ( P ≤ 0.05). No significant difference in circRNA expression was observed between Sam68 −/− cells and Sam68 −/− cells complemented with mutant SAM68 (V229F) for the circRNAs tested. Computational analysis of the binding propensity The cat RAPID algorithm estimates the binding potential through van der Waals, hydrogen bonding and secondary structure propensities of both protein and RNA sequences allowing identification of binding partners with high confidence [ 63 , 64 ]. In Supplementary Fig. S3F , the binding propensities of SAM68 and QKI for circRNA-forming pre-mRNAs were calculated and compared with the binding propensities of the background mouse RBPome. cat RAPID scores for background RBPome, SAM68, and QKI were computed against the RNAs of interest (longest transcript isoform for each gene). An empirical P -value (from 0 to 0.1) between the background and the score of QKI and SAM68 was calculated. In Supplementary Fig. S3G , we computed the cat RAPID binding propensity of SAM68 for intronic regions surrounding circRNAs identified as downregulated in Sam68 −/− mESCs in comparison to WT. To approximate the intronic regions upstream and downstream of the circRNAs, 500 nt preceding the first exon and 500 nt following the last exon were specifically chosen for SAM68 interaction assessment. As control, 1000 intronic regions of 500 nt in the mouse genome were selected. Significance was calculated using the Kolmogorov–Smirnov test. Overlap SAM68-regulated circRNAs with cardiac circRNAs To evaluate the overlap between SAM68-dependent circRNAs and cardiac-associated circRNAs [ 65 ], we first converted the latter’s genomic coordinates from M1 mm9 to GRCm38/mm10 to enable direct comparison with our dataset. Of the 3017 circRNAs identified as expressed in the heart in that study, 1939 were successfully converted and used as the reference set of cardiac circRNAs for downstream analyses. For each time point, we considered (i) the total number of circRNAs detected and tested for differential expression (whose linear counterpart did not vary expression i.e. FDR_lin > 0.05 or Log2FC opposite to the one of the circRNA), (ii) the subset of circRNAs classified as cardiac that were also detected in our dataset, (iii) the number of significantly downregulated circRNAs (based on FDR thresholds), and (iv) the overlap between downregulated circRNAs and cardiac circRNAs. The statistical significance of the overlap was assessed using a hypergeometric test (Fig. 3D ). To assess whether host genes of SAM68-regulated circRNAs are significantly overlapping with host genes giving rise to cardiac circRNAs [ 65 ], we performed a permutation-based approach ( Supplementary Figs S3D and E ). For each differentiation day, a background pool of mouse genes was defined using CircAtlas v2.0. From this pool, random gene sets were generated with a size matching the number of genes associated with downregulated circRNAs at the corresponding time point. Randomly sampled genes were required to be present in our RNA-seq dataset, to show no significant differential expression (FDR > 0.05), and to exclude genes corresponding to downregulated circRNAs at the same time point. For each random gene set, the overlap with the list of cardiac circRNA host genes was computed and recorded. This random sampling procedure was repeated 1000 times per time point to generate a null distribution of overlap values. An empirical P -value was calculated as the fraction of permutations in which the random overlap equaled or exceeded the observed overlap. Motifs analysis The binding sites were extended by ±10 nucleotides to include local sequence context for motif discovery. Corresponding sequences were extracted from the mouse reference genome from the Bioconductor BSgenome.Mmusculus.UCSC.mm10 package (version 1.4.3, DOI: 10.18129/B9.bioc.BSgenome.Mmusculus.UCSC.mm10) using getSeq function from the Bioconductor Biostrings package (version 2.76.0, DOI: 10.18129/B9.bioc.Biostrings). Sequences were exported in FASTA format with writeXStringSet for subsequent analysis. Motif enrichment analysis was performed using the XSTREME module of the MEME Suite webserver (version 5.5.9, [ 66 ]). Input sequences were analyzed against shuffled control sequences using an RNA alphabet and the “Ray2013” RNA-binding protein motif database. Motif discovery was constrained to widths of 4–10 nucleotides with a significance threshold of 0.05. Both STREME and MEME algorithms were applied to identify up to 10 enriched motifs each, assuming zero or one motif occurrence per sequence and using control-sequence-derived background models. Top 3 highest ranked motifs are shown in Fig. 3G for d10 of differentiation, while top 2 are shown in Supplementary Fig. S3K for d0 and d3 of differentiation. Sequence motif enrichment analysis in circRNA flanking regions Sequence motif enrichment analysis was performed on 278 circRNAs downregulated in Sam68 −/− cells at d10 of differentiation. Starting from top five consensus motifs (obtained from the analysis above), all possible motif variants considering degenerate nucleotide positions were generated ( n = 22) and scanned within 1000 nt upstream and downstream of circRNA junctions. Motifs were ranked based on the number of sequences in which they occurred, excluding motifs with zero counts. To assess statistical significance, the occurrence of the top-ranked motif was compared to a negative control dataset of circRNAs (i.e. randomly selected circRNAs from CircAtlas v2.0 [ 62 ] that are not regulated by SAM68, deriving from different host genes than the regulated circRNAs, but expressed at the corresponding time point) using a chi-squared test. To avoid redundancy, sequences sharing identical start or end coordinates were removed. Statistical significance was assessed using a chi-squared test by comparing motif presence versus absence between SAM68-regulated circRNAs and the negative control set. Motif analysis of RNA-binding protein sites in untranslated regions To identify RBP motifs in proximity to SAM68 binding sites, we selected the central nucleotide of each iCLIP-defined binding region and extracted sequences spanning ±20 nt around these positions. A curated collection of mouse RBP binding motifs was obtained from a previously published dataset [ 63 ]. Motif scanning was performed using FIMO from the MEME suite [ 67 ], following the same approach and parameters described in the original study including the optimization of number of instances for small RNA motifs [ 63 ]. Proteins were considered significantly enriched if at least one motif occurrence was detected within the analyzed regions with a P -value < 0.05. Quantification and statistical analysis The distribution, deviation of data, statistical tests used to calculate significant differences, and number of biological replicates of the experiments are denoted in figure legends. Statistical analyses were performed using GraphPad Prism version 10.0.0 for Mac OS X, GraphPad Software, Boston, Massachusetts USA, www.graphpad.com . Results SAM68 and QKI positively regulate mESC cell cycle, self-renewal, and cardiomyocyte differentiation To study the role of SAM68 and QKI in early development, we engineered mESCs (E14TG2a) knocked out (KO) of either Sam68 or Qki using the CRISPR–Cas9 technology (Fig. 1A and Supplementary Fig. S1 ). sgRNAs were designed to target the first exon of each gene, resulting in the complete abolishment of protein production compared to wild-type (WT) cells (Fig. 1A ). Importantly, the absence of any of the two proteins did not significantly affect the expression level of the other (Fig. 1A and Supplementary Fig. S1B ). Figure 1. Open in a new tab SAM68 and QKI promote cell cycle progression, self-renewal, and cardiomyocyte differentiation in mESCs. ( A ) Schematic of the CRISPR–Cas9 design strategy targeting the first exon of either Sam68 or Qki genomic locus to generate E14 mESCs KO lines. On the right: Western blot validating the absence of either protein. ( B ) Count of WT, Sam68 −/- and Qki −/− beating foci . Data derived from a minimum of three biological replicates, and statistical significance was determined using a t -test ( P < 0.01). ( C ) Beating foci assay in WT, Sam68 −/- and Qki −/− . The beating frequency was measured per minute for a minimum of 30 different beating foci . The statistical significance was determined using a t -test ( P < 0.01). ( D ) Proliferation assay for WT, Sam68 −/− , and Qki −/− mESC cell lines. Data derived from a minimum of three biological replicates, and statistical significance was determined using a t -test ( P < 0.01). ( E ) Clonogenic assay of WT, Sam68 −/− , and Qki −/− mESC cell lines, measuring area in μm 2 for at least 60 different colonies. Statistical significance was determined using a t -test ( P < 0.01). ( F ) Count of WT, Sam68 −/− , and Qki −/- colonies. Data derived from nine biological replicates, and statistical significance was determined using a t -test ( P < 0.01). ( G ) Pluripotency assay: the bar plot shows the percentage of residual GFP (pluripotent cells) at different time points after knocking down Sam68 or Qki . Data derived from a minimum of three biological replicates, and statistical significance was determined using a t -test ( P < 0.01). ( H ) Teratoma assay indicating the presence of terminally differentiated cells derived from ectoderm, mesoderm, and endoderm. Given the established role of QKI in cardiac myofibrillogenesis [ 46 – 68 ], we performed in vitro cardiomyocyte differentiation and quantified spontaneous beating foci after 8 days of induction. A significant reduction in beating foci was observed in the absence of either QKI or SAM68 compared to the WT control (Fig. 1B ), suggesting that SAM68, like QKI, plays a critical role in this differentiation pathway. Furthermore, beating foci from Sam68 −/− cells exhibited an increased beat frequency compared to WT (Fig. 1C ). In contrast, the severely reduced number and distorted morphology of foci in Qki −/− cells prevented quantification. To gain a deeper understanding of this phenotype, we conducted a comprehensive characterization of the KO mESCs. Similar to WT mESC, both Sam68 −/− and Qk i −/− mESCs displayed a typical domed-shaped colony morphology with well-defined borders, and the absence of differentiated colonies exhibiting spindle shapes ( Supplementary Fig. S1C ). The pluripotency and proliferation of ESC are strongly linked with cell cycle [ 69 ]. To specifically evaluate the role of SAM68 and QKI in mESC self-renewal, we first synchronized WT, Sam68 −/− , and Qk i −/− mESCs with nocodazole. Nocodazole efficiently synchronizes ESCs, arresting cells at G2/M phase without affecting their pluripotency state [ 70 ]. Cell proliferation was monitored by daily cell counting through 72 h (Fig. 1D ). We found that depletion of either protein resulted in a significant decrease in the proliferation rate (Fig. 1D and Supplementary Fig. S1D ). Consistent with this observation, the KO cells exhibited the formation of smaller colonies with reduced numbers compared to WT mESCs in the clonogenic assay (Fig. 1E and F , and Supplementary Fig. S1E and F ). Together, these results demonstrate that both SAM68 and QKI are essential for sustaining mESC self-renewal. This aligns with prior studies linking SAM68 to neuronal progenitor self-renewal [ 42 ]. mESC self-renewal is maintained by LIF and bone morphogenetic protein (BMP), and removal of these inhibitors induces differentiation into the three germ layers [ 71 ]. To assess whether SAM68 and QKI influence pluripotency exit, we used Rex1-d2GFP reporter cells, in which GFP tags the pluripotency marker Rex1 [ 72 ]. To inhibit the expression of Sam68 and Qki in Rex1-d2GFP cells, we infected cells with lentiviral particles harboring short-hairpin RNA (shRNA) constructs against Sam68 and Qki (Fig. 1G and Supplementary Fig. S1G ). The dynamics of pluripotency exit of these cells were measured by FACS at 24 and 48 h after growing the cells in differentiation-permissive medium through the removal of LIF and BMP inhibitors. As shown in Fig. 1G , there was no significant difference between control and cells transfected with the shRNAs, suggesting that none of the proteins is involved in the process of exit from pluripotency. Finally, to evaluate their involvement in early lineage specification in vivo , we implanted WT, Sam68 −/− , and Qk i −/− cells subcutaneously into nude mice to form teratomas. All teratomas contained cells derived from the three germ layers (Fig. 1H ). Interestingly, teratomas from Sam68 −/- cells showed an elevated presence of mesoderm-derived cells compared to WT, whereas teratomas from Qk i −/− cells displayed a distinct composition ( Supplementary Fig. S1H ). These findings suggest that SAM68 may specifically regulate mesoderm differentiation. The elevated mesodermal signature observed in Sam68 −/− teratomas may reflect a shift in the subtype composition of mesoderm-derived cells, with reduced commitment to the cardiac lineage and compensatory differentiation toward alternative mesodermal fates. SAM68 interacts with circular RNA biogenesis factors To investigate the molecular mechanisms involving SAM68 in early mESC development and cardiac differentiation, we first employed an embryoid body (EB) differentiation system. In this assay, mESCs are cultured in suspension and spontaneously differentiate into tissue-like spheroids over a 10-day period (Fig. 2A ). This system recapitulates early stages of mammalian development, as EBs form complex three-dimensional structures composed of multiple cell types derived from the three germ layers [ 40 , 41 ]. We first assessed the subcellular localization of SAM68 during EB formation and found that SAM68 is predominantly localized in the nucleus, with a smaller fraction detectable in the cytoplasm ( Supplementary Fig. S2A ). No major changes in SAM68 expression levels or subcellular distribution were observed across the analyzed differentiation time points ( Supplementary Fig. S2A ). Figure 2. Open in a new tab SAM68 interacts with circular RNA biogenesis factors NF90/NF110. ( A ) Schematic representation of GFP-SAM68 pulldown using GFP trap beads followed by mass spectrometry analysis to identify SAM68 interactors at the three different time points of differentiation (d0, d3, and d10). ( B ) Immunoprecipitation (IP) of GFP-SAM68 (100 kDa) compared to input (0.75%), with GFP IP as a control. ( C ) Volcano plots depicting SAM68 interactors across distinct differentiation stages (d0, d3, and d10). Statistically significant interactors are highlighted based on FDR thresholds. Changes in the expression levels of SAM68 interactors are shown in Supplementary Fig. S2D . ( D ) Heatmap displaying the dynamic interactors of SAM68 during differentiation at d0, d3, and d10. Proteins are classified into three categories: core interactors (core), transient interactors (tp-trans), and time-specific interactors (tp-spec). Representative proteins from each group are highlighted. ( E ) Venn diagram illustrating the overlap of SAM68 interactors across differentiation time points (d0, d3, and d10) using a FDR < 0.01 threshold. ( F ) Gene Ontology (GO) enrichment analysis of significant interactors. The top two enriched GO terms from each category—Molecular Function (GO:MF), Biological Process (GO:BP), and Cellular Component (GO:CC)—are shown. The x -axis represents the statistical significance of enrichment, expressed as −log10 (adjusted P -value), where higher values indicate stronger enrichment. The analysis was performed using g:Profiler. ( G ) Immunoprecipitation (IP) of NF90/110, analyzed by western blot, comparing 1% input of total protein samples across the three differentiation stages (d0, d3, and d10). Rabbit IgG IP served as a negative control. To identify protein partners associated with SAM68 during differentiation, we generated E14 GFP-Sam68 knock-in cells using CRISPR–Cas9 technology. These cells were subjected to the same EB differentiation protocol described above and used for immunoprecipitation with GFP-trap beads followed by mass spectrometry analysis to characterize the SAM68 protein interactome (Fig. 2B , and Supplementary Fig. S2B and Supplementary Table S3 ). By employing E14 cells that stably express GFP as a background control, we determined the SAM68 interactors across various differentiation time points (d0, d3, and d10) (Fig. 2C and Supplementary Fig. S2C ). To identify the enriched proteins in the GFP-SAM68 pulldown over the background control (GFP), we used a two-sample t -test at different permutation-based false discovery rate (FDR) cutoffs (0.001, 0.01, and 0.05). The expression levels of the identified proteins were then normalized to the expression level of SAM68 across the different time points (Fig. 2C – E ). The number of proteins significantly enriched in the SAM68 pull-down increases during differentiation (Fig. 2C and E), suggesting a progressive remodeling of the SAM68 interactome and an expanding role for SAM68 in post-transcriptional regulation and lineage specification. This increase is not explained by changes in transcript abundance of the interacting proteins across differentiation stages ( Supplementary Fig. S2D ). In total, we identified 194 significant interactors (FDR < 0.01), including known SAM68 interactors such as DROSHA and HNRNPL [ 73 , 74 ]. We also observed a strong enrichment of SAM68 paralogue T-STAR (encoded by Khdrbs3 ); their dimerization was shown to be crucial for their role in RNA splicing [ 75 ] ( Supplementary Fig. S2E ). We analyzed the temporal dynamics of protein interactions and identified three distinct classes of SAM68 interactors (Fig. 2D and Supplementary Fig. S2E ). ( i ) Core components ( core ), which associate with SAM68 throughout differentiation, are predominantly involved in RNA metabolism, including degradation (e.g. UPF1), translation (e.g. PUM2, EIF4G3, and YBX1), and localization and storage (e.g. FMRP, G3BP1, and CAPRIN1), positioning SAM68 within a broader RNA regulatory network ( Supplementary Fig. S2E ). Additionally, SAM68 interacts with ELAVL2 and ELAVL4, key regulators of neural-specific mRNA processing and neuronal differentiation [ 76 , 77 ], as well as CCAR2, a transcriptional regulator implicated in cell proliferation [ 76 ]. ( ii ) Transient interactors ( tp-trans ), detected at two out of three time points, suggest a dynamic association with SAM68 during differentiation (Fig. 2D ). This includes NF45 and NF90/110, which are known to be involved in the regulation of pluripotency and differentiation of mESCs [ 78 ], FXR1P, a regulator of mRNA translation and decay; and RBM14, which contributes to transcriptional regulation and RNA metabolism [ 79 , 80 ], further supporting the role of SAM68 in post-transcriptional control. ( iii ) Time-point specific interactors ( tp-spec ), such as those found at day 10 of differentiation, include factors involved in neural differentiation, RNA metabolism, and translational regulation (Fig. 2D ). These include SOX1 [ 81 ], DCX, and MAP2, which regulate neuronal differentiation and cytoskeletal remodeling [ 82 ]; AGO2, DROSHA [ 83 ], and FUS, linking SAM68 to miRNA processing and RNA metabolism; and PUM1 [ 84 ], YTHDF3, and YTHDF1 [ 85 , 86 ], mediating mRNA stability and translational control. Additionally, SMARCA2 [ 87 ] and ZIC2 [ 88 ] suggest a role in chromatin remodeling and transcriptional regulation at this stage of differentiation. Gene Ontology (GO) analysis of SAM68 interactors identified at d10 of differentiation revealed a significant representation of factors involved in RNA binding and translation, supporting a prominent role for SAM68 in cytoplasmic post-transcriptional regulation (Fig. 2F ). SAM68 is also a regulator of alternative splicing [ 34 , 89 – 91 ], and we observed an enrichment of proteins associated with the exon–exon junction complex, highlighting its role in pre-mRNA processing. Among the tp-trans interactors of SAM68, we identified the immune response factors NF45 and NF90/110 (Fig. 2G and Supplementary Table S3 ). In particular, NF90/110 has been shown to promote back-splicing, a non-canonical splicing event, thereby positively regulating circRNA production [ 92 ]. Previous studies have demonstrated that QKI functions as a circRNA biogenesis factor during the epithelial-to-mesenchymal transition by binding to specific motifs within introns near circRNA-forming splice sites [ 14 ]. Recently, a new role for human SAM68 in the pre-mRNA circularization process has been described for the SMN locus [ 93 ]. To investigate the role of SAM68 in circRNA biogenesis, we first validated its interaction with NF90/110. Immunoprecipitation of NF90/110 at three different stages of differentiation confirmed the expected co-precipitation of NF45, while also revealing interaction with SAM68 (Fig. 2G ). Although proteomic analysis detected the SAM68–NF90/110–NF45 interaction only at d0 and d3—since the d10 interaction did not pass the significance threshold—IP experiments confirmed its presence at d0, d3, and also at d10, where the enrichment appeared yet lower compared to the earlier time points (Fig. 2G ). Further examination of the SAM68–NF90/110–NF45 complex revealed that their interaction is RNA-dependent ( Supplementary Fig. S2F ), suggesting a potential cooperative role in RNA metabolism, including circRNA biogenesis. SAM68 is involved in circRNA biogenesis circRNAs are produced through a back-splicing event in which a downstream 5′ splice site is covalently joined with an upstream 3′ splice site in reverse order [ 94 ]. Besides back-splicing, SAM68 and QKI are known to be involved in the regulation of AS [ 34 , 41 , 46 , 89 – 91 , 95 – 97 ]. To assess the roles of these proteins in AS and circRNA biogenesis, during early development processes, we performed bulk RNA-seq at d0, d3, and d10 of differentiation in WT, Sam68 −/− , and Qki −/− mESC cell lines. We depicted aberrant AS events in S am68 −/− and Qki −/− compared to WT and found that despite the large number of deregulated AS events compared to the WT cell line, at all differentiation time points, the overlaps between Sam68 −/- and Qki −/- cell lines were less than 5% (Fig. 3A and Supplementary Table S4 ). These results suggest that these STAR proteins play a divergent role in AS regulation during mESCs early differentiation. This distinct mode of regulation of RBPs belonging to the same family has been described also for members of the Serine–Arginine [ 98 , 99 ] and the RNA-binding motif (RBM) [ 100 ] families, respectively. While in the absence of SAM68 the majority of the altered AS events belonged to the category of intron retention—in line with what was previously described for specific SAM68 targets [ 91 , 101 ]—exon skipping appeared to prevail in Qki −/− cell line (Fig. 3B and Supplementary Table S4 ). To investigate the role of SAM68 and QKI in circRNA biogenesis, we analyzed RNA-seq data to identify back-spliced junction (BSJ) reads that were differentially expressed in Sam68 −/− and Qki −/− cell lines, while the expression of their corresponding linear transcripts remained unchanged ( Supplementary Fig. S3A ). Strikingly, differential expression analysis revealed a marked reduction in circRNA levels in the absence of SAM68, uncovering a previously unrecognized role for this RBP in regulating circRNA biogenesis during early development (Fig. 3C , and Supplementary Fig. S3B and Supplementary Table S5 ). Analysis of circRNAs host pre-mRNAs revealed SAM68-dependent splicing changes are consistent with those observed across SAM68-regulated transcripts, suggesting coordinated regulation of alternative RNA processing and backsplicing (Fig. 3B and Supplementary Fig. S3C ). Numerous studies have demonstrated that circRNAs are pervasive across all tissues but also exhibited tissue- and cell type-specific expression, undergoing regulation throughout developmental processes [ 13 , 18 , 102 – 106 ]. Given the defects in cardiomyocyte differentiation observed in Sam68 −/− cells, we investigated whether the circRNAs regulated by SAM68 display cardiac specificity. To this end, we compared circRNAs downregulated in Sam68 −/− cells with a previously defined set of cardiac-enriched circRNAs in mouse [ 65 ]. Remarkably, >50% of the downregulated circRNAs overlapped in a significant manner (hypergeometric test) at each differentiation stage with cardiac circRNAs (Fig. 3D ), suggesting that SAM68 may contribute to the generation of cardiac-associated circRNAs. In line with this, host genes of SAM68-regulated circRNAs significantly overlapped with those identified in the cardiac reference dataset, including highly expressed cardiac genes shown to generate a high number of circRNA isoforms ( Supplementary Figs S3D and E ). To investigate whether SAM68 can engage with the precursors of circRNAs downregulated in Sam68 −/− cells, we employed cat RAPID, a computational method that predicts protein–RNA interaction propensity [ 63 ]. We calculated the binding propensity of SAM68 and QKI and compared the cat RAPID scores to the distribution of interaction propensities for 1799 RBPs comprising the mouse RBPome [ 107 ] ( Supplementary Fig. S3F ). Our analysis showed that SAM68 cat RAPID scores fall within the top 10%–20% of the score distribution, indicating a strong predicted propensity for interaction with these RNAs. In contrast, QKI scores cluster around the midpoint of the distribution, suggesting a comparatively lower predicted binding propensity for these targets ( Supplementary Fig. S3F ). Next, we investigated, using cat RAPID, the propensity of SAM68 to interact with the intronic regions flanking the back-splice junctions of circRNAs regulated by SAM68 ( Supplementary Fig. S3G ). We found that SAM68 had significantly higher binding propensity to these intronic regions compared to randomly selected introns of the same length ( Supplementary Fig. S3G ). To experimentally characterize SAM68 RNA-binding profiles and assess its potential involvement in circRNA biogenesis, we performed individual-nucleotide resolution UV crosslinking and immunoprecipitation (iCLIP2) followed by high-throughput sequencing (Fig. 3E , and Supplementary Fig. S3H–J and Supplementary Table S6 ). Genome-wide analysis revealed that SAM68 predominantly binds intronic regions of protein-coding genes (Fig. 3F and Supplementary Fig. S3J ), consistent with its established role in pre-mRNA processing [ 34 , 39 , 41 , 108 ]. Motif analysis revealed that SAM68 preferentially binds A/U-rich motifs, consistent with previous reports identifying a consensus four-nucleotide A/U-rich motif [ 33 , 75 , 109 ], with a less frequent G-rich motif also detected [ 110 , 111 ] (Fig. 3G and Supplementary Fig. S3K ). Given that regulatory elements influencing back-splicing are typically located within intronic regions flanking circularizing exons [ 112 – 115 ], we investigated whether SAM68 binding is enriched in introns adjacent to SAM68-regulated circRNAs. To this end, we analyzed iCLIP2 signal within intronic regions adjacent to the BSJs of circRNAs downregulated in Sam68 −/− cells, using windows of 500, 1000, and 2000 nucleotides from the junction (Fig. 3H and Supplementary Fig. S3L ). As a control, we included a set of circRNAs whose expression was not affected by SAM68 loss. We found that the proportion of deregulated circRNAs with SAM68 binding in proximal intronic regions (1000 nt) was significantly higher than in control circRNAs (Fig. 3H ), suggesting that SAM68 occupancy at flanking introns may contribute to circRNA regulation. We next asked whether intronic regions flanking SAM68-regulated circRNAs are enriched in specific sequence motifs that may promote SAM68 recruitment and facilitate back-splicing (Fig. 3I ). Focusing on d10 of differentiation, where SAM68 binding in flanking regions is most pronounced, we scanned 1000 nt upstream and downstream of back-splicing junctions for SAM68-associated motifs. We found that a A/U-rich motif (AAAAUUUU) is enriched in regulated circRNAs compared to non-regulated circRNAs (Fig. 3I ). This suggests that pre-mRNA regions containing multiple AAAAUUUU motifs may represent preferred binding sites for SAM68 and contribute to circRNA biogenesis. To further validate the association of SAM68 with circRNA precursors, we performed CLIP-based assays in nuclear extracts using primers spanning exon–intron junctions involved in circularization, focusing on targets for which iCLIP2 revealed binding on both sides of the BSJ. Precursors of selected SAM68-regulated cardiac circRNAs were enriched in SAM68 immunoprecipitates compared to negative controls, namely pre-mRNAs not bound by SAM68 and not producing SAM68-regulated circRNAs (Fig. 3J and Supplementary Fig. S3M ). We next selected a subset of cardiac circRNAs downregulated in Sam68 −/− cells at d0 and assessed the effect of ectopic SAM68 expression on their levels, focusing on candidates showing SAM68 binding within flanking intronic regions (Fig. 3J and Supplementary Fig. S3M ). The d0 time point was chosen to enable efficient transfection, which is not feasible at later stages when cells grow forming EBs. Compared to Sam68 −/− cells, ectopic expression of WT SAM68 restored the expression of most tested circRNAs (Fig. 3K , and Supplementary Fig. S3O and P ). In contrast, expression of SAM68 mutant carrying a mutation in the KH domain (V229F), previously shown to impair RNA binding [ 89 , 116 ] ( Supplementary Fig. S3N ), failed to rescue circRNA expression (Fig. 3K and Supplementary Fig. S3P ). Together, these observations are consistent with a role for SAM68 as a positive regulator of circRNA biogenesis, potentially through its interaction with precursor transcripts. SAM68 and QKI regulate cardiac-specific transcripts We next proceeded with analyzing the RNA-seq data to study the impact of SAM68 on linear transcriptome, in particular the cardiac-related RNAs. Principal component analysis (PCA) revealed consistent changes in gene expression at the transcript levels in Sam68 −/− EBs ( Supplementary Fig. S4A and Supplementary Table S7 ), whereas Qki −/− EBs showed a more similar pattern to its WT counterparts (Fig. 4A , and Supplementary Fig. S4B and Supplementary Table S7 ). Differential expression analysis (FDR < 0.01; fold change > 1.5) revealed a more pronounced effect of Sam68 deletion on gene expression compared to Qki deletion, particularly evident in the number of deregulated genes across differentiation stages. SAM68 controlled the expression of protein coding regions throughout differentiation (d0, d3, and d10), whereas QKI exerted its primary influence on protein-coding genes mainly at the d10 of differentiation ( Supplementary Fig. S4C ). Remarkably, the intersection of differentially expressed RNAs for the two proteins at each time point was rather high (Fig. 4B ). However, the percentage of regulated transcripts in opposite directions within the overlapping cluster, increased from 28% to 62% at d0 and d10, respectively (Fig. 4C and D) revealing distinct regulatory functions of these STAR proteins in guiding differentiation toward various lineages. Figure 4. Open in a new tab SAM68 and QKI regulate distinctly cardiac-specific RNAs. ( A ) Violin plot depicting differentially expressed genes (FC > 1.5 and FDR = 0.001) in Sam68 − /− and Qki −/− compared to WT at d0, d3, and d10 of differentiation. ( B ) Venn diagrams illustrating the overlap of differentially expressed (DE) genes in Sam68 − /− and Qki −/− at d0, d3, and d10 of differentiation. The intersections are analyzed in panels (C) and (D). ( C and D ) Bar graphs representing the proportion (C) and directionality (D) of commonly affected genes, categorized as regulated in the same direction (up/up or down/down) or in opposite directions (up/down or down/up). ( E ) Gene Ontology (GO) enrichment analysis of genes most upregulated in Sam68 − /− and most downregulated in Qki −/− at d10, highlighting enrichment for cardiac development–related terms. ( F ) Violin plot depicting mRNA expression of cardiac genes in Sam68 −/− and Qki −/− cells at d10 of differentiation, revealing opposing regulatory patterns by SAM68 and QKI. ( G ) Validation of RNA-sequencing results through real-time PCR, showing altered expression of selected cardiac genes in Sam68 −/- cells at d10 of embryoid body (EB) differentiation. P -values are indicated for each gene. ( H ) Validation of RNA-sequencing results through real-time PCR, demonstrating the altered expression of cardiac-related mRNAs in Qki −/− cells at d10 of EB differentiation. P -values are indicated for each gene. Gene ontology (GO) analysis on the differentially expressed genes at d10 of EB differentiation revealed an overrepresentation of cardiac-related terms implicating both SAM68 and QKI in cardiac development (Fig. 4E and Supplementary Fig. S4D ). While the majority of cardiac-related RNAs were upregulated in Sam68 −/− EBs, a complete opposite pattern was observed in Qki −/- EBs (Fig. 4F , and Supplementary Fig. S4C and Supplementary Table S7 ). In the absence of SAM68, we observed significant upregulation of Gata4, Nkx2-5 , and Tbx18 alongside downregulation of Isl1 at d10 EBs (Fig. 4G ). This change in expression of cardiac progenitor specific transcription factors [ 117 – 122 ] was likely to cause the general upregulation of transcripts encoding for sarcomeric structural protein ( Tnnt2, Myl2, Myl4 , and Actc1 ) [ 123 , 124 ] and ion transport ( Cacnb2 and Ryr2 ) [ 125 ] (Fig. 4G ). In contrast, in Qki −/- EBs, we found a downregulation of genes encoding pivotal cardiomyocyte structural proteins (e.g. Actn2, Mybphl, Nebl, Myh6, Actc1, Tmod1, Tnni3, Tnnc1 , and Tnnt3 ) [ 126 – 131 ], cardiomyocyte functional markers ( e.g . ion channels Ryr2 and Scn5a ) [ 132 – 134 ], and cardiac-specific splicing regulators ( Rbm20 and Rbm24 ) [ 135 – 137 ] (Fig. 4H ). This downregulation can be partially attributed to the decreased expression of key cardiac transcription factors, namely Gata4, Gata6 , and Mef2c observed during the initial stages of differentiation in the Qki −/− cells ( Supplementary Fig. S4E ) [ 117 , 138 , 139 ]. Overall, our data indicate that both SAM68 and QKI are key factors in the regulation of cardiomyocyte differentiation of mESCs through the control of cardiac-related transcripts AS and expression. SAM68 regulates Gata4 translation during cardiomyocyte differentiation To systematically assess SAM68-dependent translational regulation, we performed ribosome profiling (Ribo-seq) in WT, Sam68 −/− , and Qki −/− EBs at d10 of differentiation (Fig. 5A , and Supplementary Fig. S5A and Supplementary Table S8 ). Comparative analysis of RNA-seq and Ribo-seq datasets ( Supplementary Table S9 ) revealed that, in the absence of SAM68, a substantial fraction of transcripts displayed a negative correlation between ribosome occupancy ( i.e . translation efficiency) and RNA abundance (Fig. 5A ). Notably, among transcripts showing increased RNA levels but reduced translation efficiency in Sam68 −/− EBs, we identified key cardiac regulators, including GATA4 and TBX18, as well as cardiomyocyte structural components such as TNNT2 and LAMA1 (Fig. 5A and Supplementary Table S9 ). To investigate whether these effects could be linked to direct SAM68 binding, we focused on transcripts that were translationally downregulated in Sam68 −/− cells (log 2 FC < 0, FDR < 0.05) without a corresponding decrease in RNA abundance, thereby excluding indirect effects driven by transcript depletion. We intersected this subset with the SAM68 binding sites identified in UTRs at d10 of differentiation ( Supplementary Table S6 ). While only a limited number of targets ( n = 9) showed SAM68 binding within their 5′ UTRs, a larger fraction ( n = 422) harbored binding sites in their 3′ UTRs, consistent with the overall distribution of SAM68 binding across transcripts ( Supplementary Fig. S3J ). GO analysis of these SAM68-bound and translationally regulated targets revealed significant enrichment for terms related to differentiation and cardiac muscle development, supporting a role for SAM68 in the translational control of gene programs required for cardiomyocyte differentiation (Fig. 5B ). Interestingly, 7 out of the 9 transcripts bound by SAM68 in their 5′ UTRs were also bound in their 3′ UTRs. Among these, Gata4 encodes a master transcription factor essential for cardiomyocyte differentiation. SAM68 binding to Gata4 mRNA was detected at both the 5′ and 3′ UTRs (Fig. 5C ), and RNA immunoprecipitation experiments confirmed that SAM68 associates with Gata4 mRNA (Fig. 5D ). Importantly, Ribo-seq analysis revealed reduced ribosome occupancy on Gata4 mRNA in Sam68 −/- cells, resulting in reduction of GATA4 protein production (Fig. 5E and Supplementary Figs S5B–E ). The phenotype observed in Sam68 −/− mESCs—characterized by a reduced number of beating foci (Fig. 1B and C)—is consistent with a reduced GATA4 activity [ 140 , 141 ]. Indeed, although Gata4 mRNA levels were increased in Sam68 −/− cells (Fig. 4G ) and its cytoplasmic abundance remained unchanged ( Supplementary Fig. S5F ), GATA4 protein levels were reduced in Sam68 −/- (Figure E and Supplementary Fig. S5B–E ) pointing to a defect at the translational level. To further explore whether SAM68 binding sites within Gata4 UTRs could directly influence translation, we generated reporter constructs in which the Renilla luciferase ( Rluc ) coding sequence was fused to both the 5′ UTR and the 3′ UTR of Gata4 mRNA, respectively. Individual SAM68 binding sites, identified by iCLIP2, were selectively deleted, and reporter activity was assessed following transient transfection of mESC at d0. Under these conditions, we did not observe substantial changes in Rluc translation upon deletion of individual binding sites ( Supplementary Fig. S5G ). Importantly, these experiments were performed at d0 of differentiation, a stage at which Gata4 is not expressed yet, raising the possibility that additional context-dependent factors required for SAM68-mediated regulation may be absent. Therefore, we next explored whether SAM68 may act in coordination with additional RBPs. To this end, we performed motif enrichment analysis within ±20 nucleotides of SAM68 binding sites in UTRs of translationally downregulated transcripts (Fig. 5F ). This analysis identified 63 candidate RBPs significantly enriched in proximity to SAM68 binding sites, whereas this enrichment was not observed in regions located 200 nt away from SAM68 peaks within UTRs (Fisher’s exact test, P = 0.0005), suggesting potential co-binding with SAM68 ( Supplementary Table S9 ). Many of these proteins are implicated in post-transcriptional regulation, including mRNA stability and translation ( Supplementary Fig. S5H ). Notably, 11 of these candidates were also identified as SAM68 interactors in our interactome analysis performed at d10 of differentiation (Fig. 5F and Supplementary Table S3 ), suggesting that SAM68 may function within multiprotein regulatory complexes. Among those, ELAV-like proteins bind A/U-rich elements (AREs) in 3′ UTRs, promoting mRNA stability and/or translation [ 140 – 142 ]. SAM68 may functionally cooperate with these proteins through its ability to bind similar A/U-rich motifs. We also identified proteins with translation repression function, AGO2, TNRC6A, and DDX6 [ 143 – 145 ], raising the possibility that SAM68 may counteract their activity through binding to the target UTR. Discussion ESC differentiation depends on the coordinated execution of gene expression programs that drive maintenance of pluripotency or eventually exit and lineage specification. In addition to transcriptional regulation, post-transcriptional mechanisms have a central role in this transition by controlling RNA processing, abundance, and protein output. Here, we identify the STAR family proteins SAM68 and QKI as important regulators of these processes in mESCs. We show that both factors support mESC proliferation, self-renewal and are required for efficient cardiomyocyte differentiation, as their loss reduces the formation of beating foci . While QKI has previously been implicated in cardiac development [ 46 , 47 , 146 ], our data uncover an unexpected requirement for SAM68 in this context. Proper cardiac development requires precise coordination of gene expression programs, and growing evidence suggests that post-transcriptional regulation plays a critical role in this process [ 147 , 148 ]. Dysregulation of RNA processing, including mRNA splicing, has been associated with cardiac disease, emphasizing the importance of RNA regulatory mechanisms in shaping cardiac function [ 135 , 148 , 149 ]. Despite belonging to the same protein family, SAM68 and QKI displayed largely non-overlapping regulatory effects at the level of AS. Both knockouts resulted in widespread splicing alterations, yet <5% of events were shared, indicating that these RBPs control largely distinct splicing programs during differentiation. Although QKI-dependent RNA processing changes are observed across multiple stages, it appears to act earlier in cardiac differentiation at the level of the transcriptional program, i.e . d3 of EBs formation, by targeting the core transcription factor network [ 150 ]. We observed a dramatic downregulation, at d3, of multiple pivotal cardiac genes including cardiac transcription factors ( i.e . GATA4 and GATA6), cardiac structural proteins and cardiac-specific splicing regulators ( i.e . RBM24 and RBM20). These extensive defects in the very early stages of differentiation might be the reason underlying the lack of beating foci formation as well as the lethality of the Qki −/− embryos. Conversely, in the case of Sam68 −/− EBs, the majority of the defects were observed at d10 of differentiation. We detected, at this time point, altered RNA levels and translation of cardiac transcription factors ( Gata4, Nkx2-5 , and Tbx18 ), which collectively promote the transition of precursor cells toward a cardiomyocyte identity [ 151 , 152 ]. Similarly, the depletion of Sam68 caused aberrant upregulation of structural proteins (TNNT2 and ACTC1) and the ion channels (RYR2 and CACNB2) important for the cardiomyocyte depolarization and their consequent contractions [ 153 , 154 ]. Together, these observations may contribute to the altered beating activity observed upon Sam68 depletion. These findings suggest that QKI primarily functions at the level of early lineage specification by establishing the cardiac transcriptional program, whereas SAM68 acts downstream to modulate post-transcriptional regulatory layers during cardiomyocyte maturation. In addition to its effects on linear RNA processing and RNA abundance, in this study we found an unprecedented role of SAM68 in circRNA expression during differentiation. Loss of SAM68 resulted in a marked reduction of circRNA levels, comprising a substantial fraction of cardiac-enriched circRNAs. Mechanistically, at least for a subset of circRNA our data indicated that SAM68 binds the pre-mRNA of genes giving rise to cardiac circRNAs, specifically targeting A/U rich motifs located in proximity of the circularizing exons. ​​circRNA biogenesis requires the interplay between trans -acting factors and cis -elements ( i.e . complementary sequences and repetitive elements) located in adjacent intronic regions, although the underlying mechanisms remain incompletely understood. For instance, in human cells at the SMN locus, SAM68 was shown to interact near Alu elements and proposed to cooperate with these sequences in facilitating back-splicing [ 93 ]. It remains unclear how SAM68 achieves target specificity to promote back-splicing at selected sites; however, it is likely that the combination of cis -elements located adjacent to circularizing exons and trans -acting factors contributes to SAM68 recruitment and function. We observed an enrichment of A/U-rich motifs in intronic regions flanking SAM68-regulated circRNAs; the presence of multiple A/U intronic binding sites was already shown to favor SAM68 binding, homodimerization and ultimately splicing at the upstream exon/intron junction [ 101 , 155 ]. Consistent with mechanisms proposed for other RBPs such as QKI, MBNL1, ZC3H14, and ESRP1 [ 14 , 112 , 156 , 157 ], we reasoned that SAM68 binds to introns and, through homodimerization, brings the splice sites into proximity, and facilitates back-splicing. This proposed mechanism aligns with SAM68’s known role in AS, where it forms homodimers [ 120 ], binds to UAA-rich motifs [ 2 ] across distant introns and loop out pre-mRNA regions to promote exon exclusion [ 37 ]. Additionally, SAM68 could also act in concert with other factors. As previously shown for alternative splicing, where it cooperates with other RNA-processing factors [ 39 , 155 ], here we find that SAM68 interacts in an RNA-dependent manner with NF90/110, which has been implicated in stabilizing RNA duplexes formed by intronic complementary sequences and promote backsplicing [ 92 ]. SAM68 may function within large ribonucleoprotein complexes, as indicated by our interactome analysis, which reveals SAM68 association with a broad set of RBPs involved in multiple aspects of RNA metabolism, including mRNA stability and translational regulation. Consistent with this, the combined analysis of Ribo-seq and RNA-seq datasets allowed us to assess changes in translation efficiency upon SAM68 loss. To focus on a potential direct role of SAM68 in translational control, we examined transcripts that were translationally downregulated in Sam68 −/− cells despite unchanged or increased mRNA abundance. We next asked whether these transcripts were directly bound by SAM68 at UTRs, in line with previous evidence showing that SAM68 can promote translation during spermatogenesis by binding 3′ UTRs and facilitating polysome loading through interaction with the eIF4F complex [ 38 , 158 ]. We found that a subset of translationally downregulated mRNAs (∼16%) displayed SAM68 binding within their 3′ UTRs, suggesting a potential direct contribution to translational regulation. In contrast, only a limited number of transcripts ( n = 9) showed SAM68 binding at their 5′ UTRs, and most of these were also bound at the 3′ UTR. Among these targets, Gata4 mRNA—encoding a key regulator of cardiogenesis—was found to be bound by SAM68 at both 5′ and 3′ UTRs and to display reduced translation efficiency in Sam68 −/− cells, despite increased mRNA levels. Notably, many cardiac transcription factors are known to be subject to complex translational regulation mediated by elements within their UTRs. For instance, the RNA helicase DHX36 (RHAU) regulates Nkx2-5 expression by resolving G-quadruplex structures in the 5′ UTR and modulating mRNA stability through interactions with the 3′ UTR, highlighting how multiple regulatory elements can cooperate to fine-tune mRNA translation [ 159 ]. Similarly, Gata4 mRNA has been reported to contain regulatory elements within its 5′ UTR, upstream open reading frames (uORFs) [ 160 , 161 ] and IRES-like features [ 162 ], suggesting complex translational regulation. Since direct SAM68 binding alone may not fully explain the observed translational effects, we investigated whether additional RBPs could act in concert with SAM68. We found that SAM68 binding sites within UTRs are significantly enriched in proximity to binding motifs of other RBPs involved in mRNA stability and translational control. This suggests that SAM68 may operate within regulatory hubs along UTRs, where multiple RBPs converge to control RNA stability and translation. Notably, several of these RBPs were also identified as SAM68 interactors at d10 of differentiation, including factors with known roles as translational repressors. In this context, SAM68 may influence translation by modulating the composition or activity of these complexes, potentially counteracting repressive factors or facilitating access of activating ones. Although SAM68 is predominantly localized in the nucleus, with only a minor cytoplasmic fraction, it has been reported to shuttle to the cytoplasm and regulate translation under specific conditions [ 38 , 158 ], or to transiently localize in the cytoplasm to facilitate mRNA delivery [ 163 ]. In addition, SAM68 has been shown to relocalize in response to extracellular cues, for example upon cell adhesion, where it associates with signaling components and modulates kinase activity, highlighting its role as a signal-responsive adaptor [ 164 ]. In our system, we did not observe a clear relocalization of SAM68 during early stages of EB differentiation. However, we do not exclude the possibility that SAM68 remains associated with specific mRNA targets as part of RNP complexes that influence their translational fate, or that it transiently engages cytoplasmic translation machinery. Together, these observations support a model in which SAM68 acts as a context-dependent translational regulator, functioning within multi-protein RNP assemblies to fine-tune protein output of key cardiac regulators. Altogether, our findings support a multifunctional role for SAM68 in cardiomyocyte differentiation. In addition to its established function in alternative splicing, SAM68 contributes to circRNA biogenesis and to post-transcriptional gene regulation by modulating the translation of key cardiac transcripts, including Gata4 . These results establish SAM68 as a multifaceted regulator that coordinates gene expression programs during cardiomyocyte differentiation. Supplementary Material gkag547_Supplemental_Files gkag547_supplemental_files.zip (70.5MB, zip) Acknowledgements The authors are grateful to the members of the Tartaglia group for insightful discussions and valuable feedback. We thank Prof. Julian König for his advice on the iCLIP procedure implemented in this study. We further acknowledge Dr. Vladimir Despic, Dr. Anna Orekhova, Dr. Mikhail Mesitov, and Dr. Miona Ćorović for their help with troubleshooting iCLIP experiments and for providing the L3 adapter for library preparation. Finally, we thank Dr. Dafne Campigli Di Giammartino and Dr. Julien Ouvrard for sharing the nucleus–cytoplasm separation protocol and for providing the CTCF and U1 snRNP70 antibodies. We thank Dr. Diego Vozzi and the IIT Genomics Facility for their assistance with iCLIP2 library sequencing. We acknowledge our collaborator Dr. Martin Vabulas for the proteomic analysis of SAM68 interactors and Thomas Graf for his guidance with the mESC differentiation model. We are grateful to Omar Peruzzo and Ilaria Paganini for technical support in the laboratory. We sincerely thank Ms. Eva Ferri and Dr. Rosa Cossu (IIT) for their assistance. We also acknowledge the members of the RNA Technologies Flagship @IIT for their constructive input. Author contributions : Laura Broglia (Conceptualization [equal], Data curation [equal], Formal analysis [equal], Funding acquisition [equal], Investigation [equal], Methodology [supporting], Validation [equal], Visualization [equal], Writing – original draft [lead], Writing – review & editing [equal]), Alessandro Dasti (Conceptualization [equal], Data curation [equal], Formal analysis [equal], Investigation [equal], Methodology [lead], Validation [equal], Visualization [equal]), Maria Carla Antonelli (Conceptualization [equal], Data curation [equal], Formal analysis [equal], Investigation [equal], Methodology [equal], Validation [equal], Visualization [equal]), Guy Aoun (Formal analysis [equal], Investigation [equal], Validation [equal]), Sabrina D’Agostino (Formal analysis [supporting], Investigation [supporting], Methodology [supporting], Validation [supporting]), Andrea Vandelli (Formal analysis [equal], Investigation [equal], Methodology [equal], Validation [supporting], Visualization [supporting]), Andrea Vandelli (Formal analysis [equal], Investigation [equal], Methodology [equal], Validation [supporting], Visualization [supporting]), Alexandros Armaos (Formal analysis [supporting], Investigation [supporting], Methodology [supporting], Validation [supporting], Visualization [supporting]), Riccardo Delli Ponti (Data curation [equal], Formal analysis [equal], Investigation [equal]), Sarah Wolf (Data curation [equal], Formal analysis [equal], Investigation [equal]), Melina Klostermann (Data curation [equal], Formal analysis [equal], Investigation [equal], Methodology [equal]), Magdalena Arnal Segura (Formal analysis [supporting], Investigation [supporting], Validation [supporting], Visualization [supporting]), Tian V. Tian (Conceptualization [supporting], Data curation [supporting], Formal analysis [supporting], Investigation [supporting], Methodology [supporting], Visualization [supporting]), Davide Mariani (Formal analysis [supporting], Methodology [supporting], Visualization [supporting]), Alessio Colantoni (Formal analysis [supporting], Methodology [supporting]), Maria Paola Paronetto (Formal analysis [supporting], Methodology [supporting], Supervision [supporting]), Stefano Gustincich (Investigation [supporting], Methodology [supporting], Supervision [supporting]), Kathi Zarnack (Data curation [equal], Formal analysis [equal], Methodology [equal]), Elias Bechara (Conceptualization [lead], Data curation [lead], Formal analysis [lead], Investigation [lead], Methodology [lead], Project administration [equal], Supervision [equal], Validation [supporting], Visualization [supporting], Writing – original draft [equal], Writing – review & editing [equal]), and Gian Gaetano Tartaglia (Conceptualization [lead], Data curation [lead], Formal analysis [lead], Funding acquisition [lead], Investigation [lead], Methodology [supporting], Project administration [lead], Resources [lead], Supervision [equal], Validation [supporting], Visualization [supporting], Writing – original draft [equal], Writing – review & editing [equal]) Notes Present address: RNA Biology and Molecular Therapeutics Team, Centre for Cancer Drug Discovery, Institute of Cancer Research, London, SM2 5NG, United Kingdom Present address: Departmental Faculty of Medicine, UniCamillus-Saint Camillus International University of Health Sciences, Rome 00131, Italy Contributor Information Laura Broglia, Center for Human Technologies, RNA Systems Biology, Istituto Italiano di Tecnologia, Genoa 16152, Italy. Alessandro Dasti, Center for Human Technologies, RNA Systems Biology, Istituto Italiano di Tecnologia, Genoa 16152, Italy. Maria Carla Antonelli, Center for Human Technologies, RNA Systems Biology, Istituto Italiano di Tecnologia, Genoa 16152, Italy. Guy Aoun, Center for Human Technologies, RNA Systems Biology, Istituto Italiano di Tecnologia, Genoa 16152, Italy. Sabrina D’Agostino, Center for Human Technologies, Non-Coding RNAs and RNA-Based Therapeutics, Istituto Italiano di Tecnologia, Genoa 16152, Italy. Andrea Vandelli, Center for Human Technologies, RNA Systems Biology, Istituto Italiano di Tecnologia, Genoa 16152, Italy. Alexandros Armaos, Center for Human Technologies, RNA Systems Biology, Istituto Italiano di Tecnologia, Genoa 16152, Italy. Riccardo Delli Ponti, Center for Human Technologies, RNA Systems Biology, Istituto Italiano di Tecnologia, Genoa 16152, Italy. Sarah Wolf, Theodor Boveri Institute, Biocenter, University of Würzburg, 97074 Würzburg, Germany. Melina Klostermann, Theodor Boveri Institute, Biocenter, University of Würzburg, 97074 Würzburg, Germany. Magdalena Arnal Segura, Center for Human Technologies, RNA Systems Biology, Istituto Italiano di Tecnologia, Genoa 16152, Italy. Tian V Tian, Vall d’Hebron Institute of Oncology, Barcelona 08035, Spain. Davide Mariani, Center for Human Technologies, Non-Coding RNAs in Physiology and Pathology, Istituto Italiano di Tecnologia, Via Enrico Melen 83, 16153 Genoa, Italy; Department of Biology and Biotechnologies “C. Darwin”, Sapienza University of Rome, Piazzale Aldo Moro 5, 00185 Rome, Italy. Alessio Colantoni, Center for Human Technologies, RNA Systems Biology, Istituto Italiano di Tecnologia, Genoa 16152, Italy. Maria Paola Paronetto, Department of Movement, Human and Health Sciences, University of Rome “Foro Italico”, Rome 00135, Italy. Stefano Gustincich, Center for Human Technologies, Non-Coding RNAs and RNA-Based Therapeutics, Istituto Italiano di Tecnologia, Genoa 16152, Italy. Kathi Zarnack, Theodor Boveri Institute, Biocenter, University of Würzburg, 97074 Würzburg, Germany. Elias Bechara, Center for Human Technologies, RNA Systems Biology, Istituto Italiano di Tecnologia, Genoa 16152, Italy. Gian Gaetano Tartaglia, Center for Human Technologies, RNA Systems Biology, Istituto Italiano di Tecnologia, Genoa 16152, Italy. Supplementary data Supplementary data is available at NAR online. Conflict of interest The authors declare no competing interests. Funding The research leading to this work was supported by the ERC ASTRA_855923 (to G.G.T.), EIC Pathfinder IVBM4PAP_101098989 (to G.G.T.), National Center for Gene Therapy and Drugs based on RNA Technology (CN00000041), financed by NextGeneration EU PNRR MUR-M4C2-Action 1.4-Call “Potenziamento strutture di ricerca e di campioni nazionali di R&S” (CUP J33C22001130001) (to G.G.T.). L.B. was supported by an EMBO post-doctoral fellowship (ALTF 1149-2021) and by a Marie Skłodowska-Curie post-doctoral fellowship (UNDERPIN – 101063903). Z.K. lab is supported by Cluster for Nucleic Acid Sciences and Technologies – NUCLEATE. Funding to pay the Open Access publication charges for this article was provided by European Research Council ASTRA n. Data availability All RNA-sequencing, ribosome profiling, and iCLIP2 data are available in GEO under accession numbers GSE153800 , GSE153799 , and GSE326251 , respectively. The proteomics data have been uploaded to the ProteomeXchange Consortium under accession number PXD009209. Please refer to Supplementary Table S1 for a comprehensive list of the software and tools used in this study, together with their respective references and locations. References 1. Van Nostrand  EL, Freese  P, Pratt  GA  et al.  A large-scale binding and functional map of human RNA-binding proteins. Nature. 2020;583:711–9. 10.1038/s41586-020-2077-3. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 2. Dvir  S, Argoetti  A, Lesnik  C  et al.  Uncovering the RNA-binding protein landscape in the pluripotency network of human embryonic stem cells. Cell Rep. 2021;35:109198. 10.1016/j.celrep.2021.109198. [ DOI ] [ PubMed ] [ Google Scholar ] 3. Guallar  D, Wang  J. RNA-binding proteins in pluripotency, differentiation, and reprogramming. Front Biol. 2014;9:389–409. 10.1007/s11515-014-1326-y. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 4. Marasco  LE, Kornblihtt  AR. The physiology of alternative splicing. Nat Rev Mol Cell Biol. 2023;24:242–54. 10.1038/s41580-022-00545-z. [ DOI ] [ PubMed ] [ Google Scholar ] 5. Yeo  GW, Coufal  NG, Liang  TY  et al.  An RNA code for the FOX2 splicing regulator revealed by mapping RNA-protein interactions in stem cells. Nat Struct Mol Biol. 2009;16:130–7. 10.1038/nsmb.1545. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 6. Venables  JP, Lapasset  L, Gadea  G  et al.  MBNL1 and RBFOX2 cooperate to establish a splicing programme involved in pluripotent stem cell differentiation. Nat Commun. 2013;4:2480. 10.1038/ncomms3480. [ DOI ] [ PubMed ] [ Google Scholar ] 7. Jangi  M, Boutz  PL, Paul  P  et al.  Rbfox2 controls autoregulation in RNA-binding protein networks. Genes Dev. 2014;28:637–51. 10.1101/gad.235770.113. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 8. Han  H, Irimia  M, Ross  PJ  et al.  MBNL proteins repress ES-cell-specific alternative splicing and reprogramming. Nature. 2013;498:241–5. 10.1038/nature12270. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 9. Chia  N-Y, Chan  Y-S, Feng  B  et al.  A genome-wide RNAi screen reveals determinants of human embryonic stem cell identity. Nature. 2010;468:316–20. 10.1038/nature09531. [ DOI ] [ PubMed ] [ Google Scholar ] 10. Ahn  E-Y, DeKelver  RC, Lo  M-C  et al.  SON controls cell-cycle progression by coordinated regulation of RNA splicing. Mol Cell. 2011;42:185–98. 10.1016/j.molcel.2011.03.014. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 11. You  X, Vlatkovic  I, Babic  A  et al.  Neural circular RNAs are derived from synaptic genes and regulated by development and plasticity. Nat Neurosci. 2015;18:603–10. 10.1038/nn.3975. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 12. Di Timoteo  G, Rossi  F, Bozzoni  I. Circular RNAs in cell differentiation and development. Development. 2020;147:dev182725. 10.1242/dev.182725. [ DOI ] [ PubMed ] [ Google Scholar ] 13. Lin  Z, Tang  X, Wan  J  et al.  Functions and mechanisms of circular RNAs in regulating stem cell differentiation. RNA Biol;. 2021; 18:2136–49. 10.1080/15476286.2021.1913551. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 14. Conn  SJ, Pillman  KA, Toubia  J  et al.  The RNA binding protein quaking regulates formation of circRNAs. Cell. 2015;160:1125–34. 10.1016/j.cell.2015.02.014. [ DOI ] [ PubMed ] [ Google Scholar ] 15. Yu  C-Y, Li  T-C, Wu  Y-Y  et al.  The circular RNA circBIRC6 participates in the molecular circuitry controlling human pluripotency. Nat Commun. 2017;8:1149. 10.1038/s41467-017-01216-w. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 16. Legnini  I, Di Timoteo  G, Rossi  F  et al.  Circ-ZNF609 is a circular RNA that can be translated and functions in myogenesis. Mol Cell. 2017;66:22–37.e9. 10.1016/j.molcel.2017.02.017e9. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 17. Li  Z, Takakura  N, Oike  Y  et al.  Defective smooth muscle development in qkI-deficient mice. Dev Growth Differ. 2003;45:449–62. 10.1111/j.1440-169X.2003.00712.x. [ DOI ] [ PubMed ] [ Google Scholar ] 18. Wang  X, Shan  G. Features and biomedical relevance of circular RNA biogenesis. Trends Genet. 2026;42:282–94. 10.1016/j.tig.2025.11.001. [ DOI ] [ PubMed ] [ Google Scholar ] 19. Heo  I, Joo  C, Cho  J  et al.  Lin28 mediates the terminal uridylation of let-7 precursor MicroRNA. Mol Cell. 2008;32:276–84. 10.1016/j.molcel.2008.09.014. [ DOI ] [ PubMed ] [ Google Scholar ] 20. Melton  C, Judson  RL, Blelloch  R. Opposing microRNA families regulate self-renewal in mouse embryonic stem cells. Nature. 2010;463:621–6. 10.1038/nature08725. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 21. Qiu  C, Ma  Y, Wang  J  et al.  Lin28-mediated post-transcriptional regulation of Oct4 expression in human embryonic stem cells. Nucleic Acids Res. 2010;38:1240–8. 10.1093/nar/gkp1071. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 22. Peng  S, Chen  L-L, Lei  X-X  et al.  Genome-wide studies reveal that Lin28 enhances the translation of genes important for growth and survival of human embryonic stem cells. Stem Cells. 2011;29:496–504. 10.1002/stem.591. [ DOI ] [ PubMed ] [ Google Scholar ] 23. Leeb  M, Dietmann  S, Paramor  M  et al.  Genetic exploration of the exit from self-renewal using haploid embryonic stem cells. Cell Stem Cell. 2014;14:385–93. 10.1016/j.stem.2013.12.008. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 24. Elatmani  H, Dormoy-Raclet  V, Dubus  P  et al.  The RNA-binding protein Unr prevents mouse embryonic stem cells differentiation toward the primitive endoderm lineage. Stem Cells. 2011;29:1504–16. 10.1002/stem.712. [ DOI ] [ PubMed ] [ Google Scholar ] 25. Subramanian  V, Klattenhoff  CA, Boyer  LA. Screening for novel regulators of embryonic stem cell identity. Cell Stem Cell. 2009;4:377–8. 10.1016/j.stem.2009.04.006. [ DOI ] [ PubMed ] [ Google Scholar ] 26. Wang  L, Miao  Y-L, Zheng  X  et al.  The THO complex regulates pluripotency gene mRNA export and controls embryonic stem cell self-renewal and somatic cell reprogramming. Cell Stem Cell. 2013;13:676–90. 10.1016/j.stem.2013.10.008. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 27. Brumbaugh  J, Di Stefano  B, Wang  X  et al.  Nudt21 controls cell fate by connecting alternative polyadenylation to chromatin signaling. Cell. 2018;172:106–120.e21. 10.1016/j.cell.2017.11.023. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 28. Geula  S, Moshitch-Moshkovitz  S, Dominissini  D  et al.  m6A mRNA methylation facilitates resolution of naïve pluripotency toward differentiation. Science. 2015;347:1002–6. 10.1126/science.1261417. [ DOI ] [ PubMed ] [ Google Scholar ] 29. Stefano  BD, Luo  E-C, Haggerty  C  et al.  The RNA helicase DDX6 controls cellular plasticity by modulating P-body homeostasis. Cell Stem Cell. 2019;25:622–638.e13. 10.1016/j.stem.2019.08.018. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 30. Dehingia  B, Milewska-Puchała  M, Janowski  M  et al.  RNA-binding proteins mediate the maturation of chromatin topology during differentiation. Nat Cell Biol. 2025;27:1510–25. 10.1038/s41556-025-01735-5. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 31. Lukong  KE, Richard  S. Sam68, the KH domain-containing superSTAR. Biochim Biophys Acta. 2003;1653:73–86. [ DOI ] [ PubMed ] [ Google Scholar ] 32. Frisone  P, Pradella  D, Di Matteo  A  et al.  SAM68: signal transduction and RNA metabolism in human cancer. Biomed Res Int. 2015;2015:1. 10.1155/2015/528954. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 33. Galarneau  A, Richard  S. The STAR RNA binding proteins GLD-1, QKI, SAM68 and SLM-2 bind bipartite RNA motifs. BMC Mol Biol. 2009;10:47. 10.1186/1471-2199-10-47. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 34. Matter  N, Herrlich  P, König  H. Signal-dependent regulation of splicing via phosphorylation of Sam68. Nature. 2002;420:691–5. 10.1038/nature01153. [ DOI ] [ PubMed ] [ Google Scholar ] 35. Tao  Y, Zhang  Q, Wang  H  et al.  Alternative splicing and related RNA binding proteins in human health and disease. Sig Transduct Target Ther. 2024;9:1–33. 10.1038/s41392-024-01734-2. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 36. Yamagishi  R, Tsusaka  T, Mitsunaga  H  et al.  The STAR protein QKI-7 recruits PAPD4 to regulate post-transcriptional polyadenylation of target mRNAs. Nucleic Acids Res. 2016;44:2475–90. 10.1093/nar/gkw118. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 37. Paronetto  MP, Messina  V, Barchi  M  et al.  Sam68 marks the transcriptionally active stages of spermatogenesis and modulates alternative splicing in male germ cells. Nucleic Acids Res. 2011;39:4961–74. 10.1093/nar/gkr085. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 38. Paronetto  MP, Messina  V, Bianchi  E  et al.  Sam68 regulates translation of target mRNAs in male germ cells, necessary for mouse spermatogenesis. J Cell Biol. 2009;185:235–49. 10.1083/jcb.200811138. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 39. Naro  C, Pellegrini  L, Jolly  A  et al.  Functional Interaction between U1snRNP and Sam68 Insures Proper 3’ End Pre-mRNA Processing during Germ Cell Differentiation. Cell Rep. 2019;26:2929–2941.e5. 10.1016/j.celrep.2019.02.058. [ DOI ] [ PubMed ] [ Google Scholar ] 40. Iijima  T, Wu  K, Witte  H  et al.  SAM68 regulates neuronal activity-dependent alternative splicing of neurexin-1. Cell. 2011;147:1601–14. 10.1016/j.cell.2011.11.028. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 41. Chawla  G, Lin  C-H, Han  A  et al.  Sam68 regulates a set of alternatively spliced exons during neurogenesis. Mol Cell Biol. 2009;29:201–13. 10.1128/MCB.01349-08. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 42. La Rosa  P, Bielli  P, Compagnucci  C  et al.  Sam68 promotes self-renewal and glycolytic metabolism in mouse neural progenitor cells by modulating Aldh1a3 pre-mRNA 3’-end processing. eLife. 2016;5:e20750. 10.7554/eLife.20750. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 43. Li  N, Hébert  S, Song  J  et al.  Transcriptome profiling in preadipocytes identifies long noncoding RNAs as Sam68 targets. Oncotarget. 2017;8:81994–2005. 10.18632/oncotarget.17813. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 44. Song  J, Richard  S. Sam68 regulates S6K1 alternative splicing during adipogenesis. Mol Cell Biol. 2015;35:1926–39. 10.1128/MCB.01488-14. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 45. Neumann  DP, Goodall  GJ, Gregory  PA. The quaking RNA- binding proteins as regulators of cell differentiation. WIREs RNA. 2022;13:e1724. 10.1002/wrna.1724. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 46. Montañés-Agudo  P, Aufiero  S, Schepers  EN  et al.  The RNA-binding protein QKI governs a muscle-specific alternative splicing program that shapes the contractile function of cardiomyocytes. Cardiovasc Res. 2023;119:1161–74. 10.1093/cvr/cvad007. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 47. Chen  X, Liu  Y, Xu  C  et al.  QKI is a critical pre-mRNA alternative splicing regulator of cardiac myofibrillogenesis and contractile function. Nat Commun. 2021;12:89. 10.1038/s41467-020-20327-5. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 48. Montañés-Agudo  P, van der Made  I, Aufiero  S  et al.  Quaking regulates circular RNA production in cardiomyocytes. J Cell Sci. 2023;136:jcs261120. 10.1242/jcs.261120. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 49. Iijima  T, Iijima  Y, Witte  H  et al.  Neuronal cell type-specific alternative splicing is regulated by the KH domain protein SLM1. J Cell Biol. 2014;204:331–42. 10.1083/jcb.201310136. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 50. Cox  J, Hein  MY, Luber  CA  et al.  Accurate proteome-wide label-free quantification by delayed normalization and maximal peptide ratio extraction, termed MaxLFQ. Mol Cell Proteomics. 2014;13:2513–26. 10.1074/mcp.M113.031591. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 51. Tyanova  S, Temu  T, Cox  J. The MaxQuant computational platform for mass spectrometry-based shotgun proteomics. Nat Protoc. 2016;11:2301–19. 10.1038/nprot.2016.136. [ DOI ] [ PubMed ] [ Google Scholar ] 52. Errichelli  L, Dini Modigliani  S, Laneve  P  et al.  FUS affects circular RNA expression in murine embryonic stem cell-derived motor neurons. Nat Commun. 2017;8:14741. 10.1038/ncomms14741. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 53. Gao  Y, Wang  J, Zhao  F. CIRI: an efficient and unbiased algorithm for de novo circular RNA identification. Genome Biol. 2015;16:4. 10.1186/s13059-014-0571-3. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 54. Despic  V, Klostermann  M, Orekhova  A.  et al.  2026; iCLIP3: a streamlined, non-radioactive protocol for mapping protein–RNA interactions in cellular transcripts at single-nucleotide resolution. bioRxiv, https://doi.org/10.64898/2026.03.01.708747, 3 March 2026, preprint: not peer reviewed.. [ DOI ] [ PMC free article ] [ PubMed ] 55. Buchbender  A, Mutter  H, Sutandy  FXR  et al.  Improved library preparation with the new iCLIP2 protocol. Methods. 2020;178:33–48. 10.1016/j.ymeth.2019.10.003. [ DOI ] [ PubMed ] [ Google Scholar ] 56. Dobin  A, Davis  CA, Schlesinger  F  et al.  STAR: ultrafast universal RNA-seq aligner. Bioinformatics. 2013;29:15–21. 10.1093/bioinformatics/bts635. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 57. Frankish  A, Diekhans  M, Ferreira  A-M  et al.  GENCODE reference annotation for the human and mouse genomes. Nucleic Acids Res. 2019;47:D766–73. 10.1093/nar/gky955. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 58. Smith  T, Heger  A, Sudbery  I. UMI-tools: modeling sequencing errors in Unique Molecular Identifiers to improve quantification accuracy. Genome Res. 2017;27:491–9. 10.1101/gr.209601.116. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 59. Quinlan  AR, Hall  IM. BEDTools: a flexible suite of utilities for comparing genomic features. Bioinformatics. 2010;26:841–2. 10.1093/bioinformatics/btq033. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 60. Kent  WJ, Zweig  AS, Barber  G  et al.  BigWig and BigBed: enabling browsing of large distributed datasets. Bioinformatics. 2010;26:2204–7. 10.1093/bioinformatics/btq351. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 61. Krakau  S, Richard  H, Marsico  A. PureCLIP: capturing target-specific protein–RNA interaction footprints from single-nucleotide CLIP-seq data. Genome Biol. 2017;18:240. 10.1186/s13059-017-1364-2. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 62. Wu  W, Ji  P, Zhao  F. CircAtlas: an integrated resource of one million highly accurate circular RNAs from 1070 vertebrate transcriptomes. Genome Biol. 2020;21:101. 10.1186/s13059-020-02018-y. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 63. Armaos  A, Colantoni  A, Proietti  G  et al.  catRAPID omics v2.0: going deeper and wider in the prediction of protein-RNA interactions. Nucleic Acids Res. 2021;49:W72–9. 10.1093/nar/gkab393. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 64. Bellucci  M, Agostini  F, Masin  M  et al.  Predicting protein associations with long noncoding RNAs. Nat Methods. 2011;8:444–5. 10.1038/nmeth.1611. [ DOI ] [ PubMed ] [ Google Scholar ] 65. Tan  WLW, Lim  BTS, Anene-Nzelu  CGO  et al.  A landscape of circular RNA expression in the human heart. Cardiovasc Res. 2017;113:298–309. [ DOI ] [ PubMed ] [ Google Scholar ] 66. Grant  CE, Bailey  TL.  2021; XSTREME: comprehensive motif analysis of biological sequence datasets. 10.1101/2021.09.02.458722. bioRxiv. [ DOI ] 67. Grant  CE, Bailey  TL, Noble  WS. FIMO: scanning for occurrences of a given motif. Bioinformatics. 2011;27:1017–8. 10.1093/bioinformatics/btr064. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 68. Liu  Y, Chen  X, Xu  C  et al.  RNA-binding protein QKI is a critical pre-RNA splicing regulator for cardiac development and function. FASEB J. 2020;34:1–1. [ Google Scholar ] 69. Liu  L, Michowski  W, Kolodziejczyk  A  et al.  The cell cycle in stem cell proliferation, pluripotency and differentiation. Nat Cell Biol. 2019;21:1060–7. 10.1038/s41556-019-0384-4. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 70. Yiangou  L, Grandy  RA, Morell  CM  et al.  Method to synchronize cell cycle of human pluripotent stem cells without affecting their fundamental characteristics. Stem Cell Rep. 2019;12:165–79. 10.1016/j.stemcr.2018.11.020. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 71. Murray  P, Prewitz  M, Hopp  I  et al.  The self-renewal of mouse embryonic stem cells is regulated by cell-substratum adhesion and cell spreading. Int J Biochem Cell Biol. 2013;45:2698–705. 10.1016/j.biocel.2013.07.001. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 72. Wray  J, Kalkan  T, Gomez-Lopez  S  et al.  Inhibition of glycogen synthase kinase-3 alleviates Tcf3 repression of the pluripotency network and increases embryonic stem cell resistance to differentiation. Nat Cell Biol. 2011;13:838–45. 10.1038/ncb2267. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 73. Messina  V, Meikar  O, Paronetto  MP  et al.  The RNA binding protein SAM68 transiently localizes in the chromatoid body of male germ cells and influences expression of select microRNAs. PLoS One. 2012;7:e39729. 10.1371/journal.pone.0039729. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 74. Mannen  T, Yamashita  S, Tomita  K  et al.  The Sam68 nuclear body is composed of two RNase-sensitive substructures joined by the adaptor HNRNPL. J Cell Biol. 2016;214:45–59. 10.1083/jcb.201601024. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 75. Feracci  M, Foot  JN, Grellscheid  SN  et al.  Structural basis of RNA recognition and dimerization by the STAR proteins T-STAR and Sam68. Nat Commun. 2016;7:10355. 10.1038/ncomms10355. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 76. Muthu  M, Cheriyan  VT, Rishi  AK. CARP-1/CCAR1: a biphasic regulator of cancer cell growth and apoptosis. Oncotarget. 2015;6:6499–510. 10.18632/oncotarget.3376. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 77. DeBoer  EM, Azevedo  R, Vega  TA  et al.  Prenatal deletion of the RNA-binding protein HuD disrupts postnatal cortical circuit maturation and behavior. J Neurosci. 2014;34:3674–86. 10.1523/JNEUROSCI.3703-13.2014. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 78. Ye  J, Jin  H, Pankov  A  et al.  NF45 and NF90/NF110 coordinately regulate ESC pluripotency and differentiation. RNA. 2017;23:1270–84. 10.1261/rna.061499.117. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 79. Auboeuf  D, Dowhan  DH, Li  X  et al.  CoAA, a nuclear receptor coactivator protein at the interface of transcriptional coactivation and RNA splicing. Mol Cell Biol. 2004;24:442–53. 10.1128/MCB.24.1.442-453.2004. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 80. Yang  Z, Sui  Y, Xiong  S  et al.  Switched alternative splicing of oncogene CoAA during embryonal carcinoma stem cell differentiation. Nucleic Acids Res. 2007;35:1919–32. 10.1093/nar/gkl1092. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 81. Elkouris  M, Balaskas  N, Poulou  M  et al.  Sox1 maintains the undifferentiated state of cortical neural progenitor cells via the suppression of Prox1-mediated cell cycle exit and neurogenesis. Stem Cells. 2011;29:89–98. 10.1002/stem.554. [ DOI ] [ PubMed ] [ Google Scholar ] 82. Moores  CA, Perderiset  M, Francis  F  et al.  Mechanism of microtubule stabilization by doublecortin. Mol Cell. 2004;14:833–9. 10.1016/j.molcel.2004.06.009. [ DOI ] [ PubMed ] [ Google Scholar ] 83. Ha  M, Kim  VN. Regulation of microRNA biogenesis. Nat Rev Mol Cell Biol. 2014;15:509–24. 10.1038/nrm3838. [ DOI ] [ PubMed ] [ Google Scholar ] 84. Goldstrohm  AC, Hall  TMT, McKenney  KM. Post-transcriptional regulatory functions of mammalian pumilio proteins. Trends Genet. 2018;34:972–90. 10.1016/j.tig.2018.09.006. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 85. Wang  X, Zhao  BS, Roundtree  IA  et al.  N6-methyladenosine modulates messenger RNA translation efficiency. Cell. 2015;161:1388–99. 10.1016/j.cell.2015.05.014. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 86. Wang  S, Zhang  J, Wu  X  et al.  Differential roles of YTHDF1 and YTHDF3 in embryonic stem cell-derived cardiomyocyte differentiation. RNA Biology. 2021;18:1354–63. 10.1080/15476286.2020.1850628. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 87. Kashiwagi  K, Yoshida  J, Kimura  H  et al.  Mutation of the SWI/SNF complex component Smarce1 decreases nucleosome stability in embryonic stem cells and impairs differentiation. J Cell Sci. 2024;137:jcs260467. 10.1242/jcs.260467. [ DOI ] [ PubMed ] [ Google Scholar ] 88. Inoue  T, Ota  M, Mikoshiba  K  et al. Zic2 and Zic3 synergistically control neurulation and segmentation of paraxial mesoderm in mouse embryo. Dev Biol. 2007;306:669–84. 10.1016/j.ydbio.2007.04.003. [ DOI ] [ PubMed ] [ Google Scholar ] 89. Paronetto  MP, Achsel  T, Massiello  A  et al.  The RNA-binding protein Sam68 modulates the alternative splicing of Bcl-x. J Cell Biol. 2007;176:929–39. 10.1083/jcb.200701005. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 90. Paronetto  MP, Sette  C. Role of RNA-binding proteins in mammalian spermatogenesis. Int J of Andrology. 2010;33:2–12. 10.1111/j.1365-2605.2009.00959.x. [ DOI ] [ PubMed ] [ Google Scholar ] 91. Valacca  C, Bonomi  S, Buratti  E  et al.  Sam68 regulates EMT through alternative splicing-activated nonsense-mediated mRNA decay of the SF2/ASF proto-oncogene. J Cell Biol. 2010;191:87–99. 10.1083/jcb.201001073. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 92. Li  X, Liu  C-X, Xue  W  et al.  Coordinated circRNA biogenesis and function with NF90/NF110 in viral infection. Mol Cell. 2017;67:214–227.e7. 10.1016/j.molcel.2017.05.023e7. [ DOI ] [ PubMed ] [ Google Scholar ] 93. Pagliarini  V, Jolly  A, Bielli  P  et al.  Sam68 binds Alu-rich introns in SMN and promotes pre-mRNA circularization. Nucleic Acids Res. 2020;48:633–45. 10.1093/nar/gkz1117. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 94. Kristensen  LS, Andersen  MS, Stagsted  LVW  et al.  The biogenesis, biology and characterization of circular RNAs. Nat Rev Genet. 2019;20:675–91. 10.1038/s41576-019-0158-7. [ DOI ] [ PubMed ] [ Google Scholar ] 95. Wu  JI, Reed  RB, Grabowski  PJ  et al.  Function of quaking in myelination: regulation of alternative splicing. Proc Natl Acad Sci USA. 2002;99:4233–8. 10.1073/pnas.072090399. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 96. Briganti  F, Wang  Z. Alternative splicing in the heart: the therapeutic potential of regulating the regulators. Int J Mol Sci. 2024;25:13023. 10.3390/ijms252313023. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 97. Caines  R, Cochrane  A, Kelaini  S  et al.  The RNA-binding protein QKI controls alternative splicing in vascular cells, producing an effective model for therapy. J Cell Sci. 2019;132:jcs230276. 10.1242/jcs.230276. [ DOI ] [ PubMed ] [ Google Scholar ] 98. Änkö  M-L, Müller-McNicoll  M, Brandl  H  et al.  The RNA-binding landscapes of two SR proteins reveal unique functions and binding to diverse RNA classes. Genome Biol. 2012;13:R17. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 99. Pandit  S, Zhou  Y, Shiue  L  et al.  Genome-wide analysis reveals SR protein cooperation and competition in regulated splicing. Mol Cell. 2013;50:223–35. 10.1016/j.molcel.2013.03.001. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 100. Bechara  EG, Sebestyén  E, Bernardis  I  et al.  RBM5, 6, and 10 differentially regulate NUMB alternative splicing to control cancer cell proliferation. Mol Cell. 2013;52:720–33. 10.1016/j.molcel.2013.11.010. [ DOI ] [ PubMed ] [ Google Scholar ] 101. Huot  M-É, Vogel  G, Zabarauskas  A  et al.  The Sam68 STAR RNA-binding protein regulates mTOR alternative splicing during adipogenesis. Mol Cell. 2012;46:187–99. 10.1016/j.molcel.2012.02.007. [ DOI ] [ PubMed ] [ Google Scholar ] 102. Ji  P, Wu  W, Chen  S  et al.  Expanded expression landscape and prioritization of circular RNAs in mammals. Cell Rep. 2019;26:3444–3460.e5. 10.1016/j.celrep.2019.02.078. [ DOI ] [ PubMed ] [ Google Scholar ] 103. Liu  C-X, Chen  L-L. Circular RNAs: characterization, cellular roles, and applications. Cell. 2022;185:2016–34. 10.1016/j.cell.2022.04.021. [ DOI ] [ PubMed ] [ Google Scholar ] 104. Wang  M, Wu  J, Wu  P  et al.  Emerging roles of circular RNAs in stem cells. Genes Dis. 2023;10:1920–36. 10.1016/j.gendis.2022.05.015. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 105. Rybak-Wolf  A, Stottmeister  C, Glažar  P  et al.  Circular RNAs in the mammalian brain are highly abundant, conserved, and dynamically expressed. Mol Cell. 2015;58:870–85. 10.1016/j.molcel.2015.03.027. [ DOI ] [ PubMed ] [ Google Scholar ] 106. You  X, Vlatkovic  I, Babic  A  et al.  Neural circular RNAs are derived from synaptic genes and regulated by development and plasticity. Nat Neurosci. 2015;18:603–10. 10.1038/nn.3975. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 107. Hentze  MW, Castello  A, Schwarzl  T  et al.  A brave new world of RNA-binding proteins. Nat Rev Mol Cell Biol. 2018;19:327–41. 10.1038/nrm.2017.130. [ DOI ] [ PubMed ] [ Google Scholar ] 108. Iijima  T, Wu  K, Witte  H  et al.  SAM68 regulates neuronal activity-dependent alternative splicing of neurexin-1. Cell. 2011;147:1601–14. 10.1016/j.cell.2011.11.028. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 109. Lin  Q, Taylor  SJ, Shalloway  D. Specificity and determinants of Sam68 RNA binding. Implications for the biological function of K homology domains. J Biol Chem. 1997;272:27274–80. 10.1074/jbc.272.43.27274. [ DOI ] [ PubMed ] [ Google Scholar ] 110. Van Nostrand  EL, Pratt  GA, Shishkin  AA  et al.  Robust transcriptome-wide discovery of RNA-binding protein binding sites with enhanced CLIP (eCLIP). Nat Methods. 2016;13:508–14. 10.1038/nmeth.3810. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 111. Galaxy CLIP-Explorer: a web server for CLIP-Seq data analysis | GigaScience | Oxford Academic. [ DOI ] [ PMC free article ] [ PubMed ] 112. Ashwal-Fluss  R, Meyer  M, Pamudurti  NR  et al.  circRNA biogenesis competes with pre-mRNA splicing. Mol Cell. 2014;56:55–66. 10.1016/j.molcel.2014.08.019. [ DOI ] [ PubMed ] [ Google Scholar ] 113. Zhang  X-O, Wang  H-B, Zhang  Y  et al.  Complementary sequence-mediated exon circularization. Cell. 2014;159:134–47. 10.1016/j.cell.2014.09.001. [ DOI ] [ PubMed ] [ Google Scholar ] 114. Liang  D, Wilusz  JE. Short intronic repeat sequences facilitate circular RNA production. Genes Dev. 2014;28:2233–47. 10.1101/gad.251926.114. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 115. Yoshimoto  R, Rahimi  K, Hansen  TB  et al.  Biosynthesis of circular RNA ciRS-7/CDR1as is mediated by mammalian-wide interspersed repeats. iScience. 2020;23:101345. 10.1016/j.isci.2020.101345. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 116. Rajan  P, Gaughan  L, Dalgliesh  C  et al.  The RNA-binding and adaptor protein Sam68 modulates signal-dependent splicing and transcriptional activity of the androgen receptor. J Pathol. 2008;215:67–77. 10.1002/path.2324. [ DOI ] [ PubMed ] [ Google Scholar ] 117. Paige  SL, Plonowska  K, Xu  A  et al.  Molecular regulation of cardiomyocyte differentiation. Circ Res. 2015;116:341–53. 10.1161/CIRCRESAHA.116.302752. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 118. Begemann  G, Gibert  Y, Meyer  A  et al.  Cloning of zebrafish T-box genes tbx15 and tbx18 and their expression during embryonic development. Mech Dev. 2002;114:137–41. 10.1016/S0925-4773(02)00040-0. [ DOI ] [ PubMed ] [ Google Scholar ] 119. Bruneau  BG. Transcriptional regulation of vertebrate cardiac morphogenesis. Circ Res. 2002;90:509–19. 10.1161/01.RES.0000013072.51957.B7. [ DOI ] [ PubMed ] [ Google Scholar ] 120. Durocher  D, Charron  F, Warren  R  et al.  The cardiac transcription factors Nkx2-5 and GATA-4 are mutual cofactors. EMBO J. 1997;16:5687–96. 10.1093/emboj/16.18.5687. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 121. Luna-Zurita  L, Stirnimann  CU, Glatt  S  et al.  Complex interdependence regulates heterotypic transcription factor distribution and coordinates cardiogenesis. Cell. 2016;164:999–1014. 10.1016/j.cell.2016.01.004. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 122. Moretti  A, Caron  L, Nakano  A  et al.  Multipotent embryonic isl1+ progenitor cells lead to cardiac, smooth muscle, and endothelial cell diversification. Cell. 2006;127:1151–65. 10.1016/j.cell.2006.10.029. [ DOI ] [ PubMed ] [ Google Scholar ] 123. Peng  W, Li  M, Li  H  et al.  Dysfunction of myosin light-chain 4 (MYL4) leads to heritable atrial cardiomyopathy with electrical, contractile, and structural components: evidence from genetically-engineered rats. JAHA. 2017;6:e007030. 10.1161/JAHA.117.007030. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 124. Sheikh  F, Lyon  RC, Chen  J. Functions of myosin light chain-2 (MYL2) in cardiac muscle and disease. Gene. 2015;569:14–20. 10.1016/j.gene.2015.06.027. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 125. Chernyavskaya  Y, Ebert  AM, Milligan  E  et al.  Voltage-gated calcium channel CACNB2 (β2.1) protein is required in the heart for control of cell proliferation and heart tube integrity. Dev Dyn. 2012;241:648–62. 10.1002/dvdy.23746. [ DOI ] [ PubMed ] [ Google Scholar ] 126. Chen  CY, Croissant  J, Majesky  M  et al.  Activation of the cardiac alpha-actin promoter depends upon serum response factor, Tinman homologue, Nkx-2.5, and intact serum response elements. Dev Genet. 1996;19:119–30. 10.1002/(SICI)1520-6408(1996)19:2<119::AID-DVG3>3.0.CO;2-C. [ DOI ] [ PubMed ] [ Google Scholar ] 127. Chu  M, Gregorio  CC, Pappas  CT. Nebulin, a multi-functional giant. J Exp Biol. 2016;219:146–52. 10.1242/jeb.126383. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 128. Jin  J-P, Zhang  Z, Bautista  JA. Isoform diversity, regulation, and functional adaptation of troponin and calponin. Crit Rev Eukar Gene Expr. 2008;18:93–124. 10.1615/CritRevEukarGeneExpr.v18.i2.10. [ DOI ] [ PubMed ] [ Google Scholar ] 129. Li  MX, Hwang  PM. Structure and function of cardiac troponin C (TNNC1): implications for heart failure, cardiomyopathies, and troponin modulating drugs. Gene. 2015;571:153–66. 10.1016/j.gene.2015.07.074. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 130. Matsson  H, Eason  J, Bookwalter  CS  et al.  Alpha-cardiac actin mutations produce atrial septal defects. Hum Mol Genet. 2008;17:256–65. 10.1093/hmg/ddm302. [ DOI ] [ PubMed ] [ Google Scholar ] 131. Sjöblom  B, Salmazo  A, Djinović-Carugo  K. Alpha-actinin structure and regulation. Cell Mol Life Sci. 2008;65:2688–701. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 132. Yang  H-T, Tweedie  D, Wang  S  et al.  The ryanodine receptor modulates the spontaneous beating rate of cardiomyocytes during development. Proc Natl Acad Sci USA. 2002;99:9225–30. 10.1073/pnas.142651999. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 133. Remme  CA. Cardiac sodium channelopathy associated with SCN5A mutations: electrophysiological, molecular and genetic aspects. J Physiol. 2013;591:4099–116. 10.1113/jphysiol.2013.256461. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 134. Sun  B, Wei  J, Zhong  X  et al.  The cardiac ryanodine receptor, but not sarcoplasmic reticulum Ca2+-ATPase, is a major determinant of Ca2+ alternans in intact mouse hearts. J Biol Chem. 2018;293:13650–61. 10.1074/jbc.RA118.003760. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 135. Guo  W, Schafer  S, Greaser  ML  et al.  RBM20, a gene for hereditary cardiomyopathy, regulates titin splicing. Nat Med. 2012;18:766–73. 10.1038/nm.2693. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 136. Beraldi  R, Li  X, Martinez Fernandez  A  et al.  Rbm20-deficient cardiogenesis reveals early disruption of RNA processing and sarcomere remodeling establishing a developmental etiology for dilated cardiomyopathy. Hum Mol Genet. 2014;23:3779–91. 10.1093/hmg/ddu091. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 137. Poon  KL, Tan  KT, Wei  YY  et al.  RNA-binding protein RBM24 is required for sarcomere assembly and heart contractility. Cardiovasc Res. 2012;94:418–27. 10.1093/cvr/cvs095. [ DOI ] [ PubMed ] [ Google Scholar ] 138. Schlesinger  J, Schueler  M, Grunert  M  et al.  The cardiac transcription network modulated by Gata4, Mef2a, Nkx2.5, Srf, histone modifications, and microRNAs. PLoS Genet. 2011;7:e1001313. 10.1371/journal.pgen.1001313. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 139. Xin  M, Davis  CA, Molkentin  JD  et al.  A threshold of GATA4 and GATA6 expression is required for cardiovascular development. Proc Natl Acad Sci USA. 2006;103:11189–94. 10.1073/pnas.0604604103. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 140. Scheckel  C, Drapeau  E, Frias  MA  et al.  Regulatory consequences of neuronal ELAV-like protein binding to coding and non-coding RNAs in human brain. eLife. 2016;5:e10421. 10.7554/eLife.10421. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 141. Quattrone  A, Pascale  A, Nogues  X  et al.  Posttranscriptional regulation of gene expression in learning by the neuronal ELAV-like mRNA-stabilizing proteins. Proc Natl Acad Sci USA. 2001;98:11668–73. 10.1073/pnas.191388398. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 142. Wutikeli  H, Xie  T, Xiong  W  et al.  ELAV/Hu RNA-binding protein family: key regulators in neurological disorders, cancer, and other diseases. RNA Biol. 2025; 22:1–11. 10.1080/15476286.2025.2471133. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 143. Lu  Y, Tao  M, Su  H  et al.  DDX6 interacts with DDX3X to repress translation in microRNA-mediated silencing. Nucleic Acids Res. 2025;53:gkaf868. 10.1093/nar/gkaf868. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 144. Freimer  JW, Hu  TJ, Blelloch  R. Decoupling the impact of microRNAs on translational repression versus RNA degradation in embryonic stem cells. eLife. 2018;7:e38014. 10.7554/eLife.38014. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 145. Welte  T, Goulois  A, Stadler  MB  et al.  Convergence of multiple RNA-silencing pathways on GW182/TNRC6. Mol Cell. 2023;83:2478–2492.e8. 10.1016/j.molcel.2023.06.001. [ DOI ] [ PubMed ] [ Google Scholar ] 146. Chen  X, Yin  J, Cao  D  et al.  The emerging roles of the RNA binding protein QKI in cardiovascular development and function. Front Cell Dev Biol. 2021;9:668659. 10.3389/fcell.2021.668659. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 147. Blech-Hermoni  Y, Ladd  AN. RNA binding proteins in the regulation of heart development. Int J Biochem Cell Biol. 2013;45:2467–78. 10.1016/j.biocel.2013.08.008. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 148. Yang  J, Hung  L-H, Licht  T  et al.  RBM24 is a major regulator of muscle-specific alternative splicing. Dev Cell. 2014;31:87–99. 10.1016/j.devcel.2014.08.025. [ DOI ] [ PubMed ] [ Google Scholar ] 149. Boeckel  J-N, Möbius-Winkler  M, Müller  M  et al.  SLM2 is a novel cardiac splicing factor involved in heart failure due to dilated cardiomyopathy. Genomics Proteomics Bioinform. 2022;20:129–46. 10.1016/j.gpb.2021.01.006. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 150. Sargent  CY, Berguig  GY, McDevitt  TC. Cardiomyogenic differentiation of embryoid bodies is promoted by rotary orbital suspension culture. Tissue Eng Part A. 2009;15:331–42. 10.1089/ten.tea.2008.0145. [ DOI ] [ PubMed ] [ Google Scholar ] 151. Bondue  A, Lapouge  G, Paulissen  C  et al.  Mesp1 acts as a master regulator of multipotent cardiovascular progenitor specification. Cell Stem Cell. 2008;3:69–84. 10.1016/j.stem.2008.06.009. [ DOI ] [ PubMed ] [ Google Scholar ] 152. Chan  SS-K, Shi  X, Toyama  A  et al.  Mesp1 patterns mesoderm into cardiac, hematopoietic, or skeletal myogenic progenitors in a context-dependent manner. Cell Stem Cell. 2013;12:587–601. 10.1016/j.stem.2013.03.004. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 153. Hedley  PL, Jørgensen  P, Schlamowitz  S  et al.  The genetic basis of Brugada syndrome: a mutation update. Hum Mutat. 2009;30:1256–66. 10.1002/humu.21066. [ DOI ] [ PubMed ] [ Google Scholar ] 154. Provan  AB, Majer  RV, Smith  AG  et al.  Chronic myelomonocytic leukaemia associated with T cell receptor delta gene rearrangement. J Clin Pathol. 1991;44:344–5. 10.1136/jcp.44.4.344. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 155. Subramania  S, Gagné  LM, Campagne  S  et al.  SAM68 interaction with U1A modulates U1 snRNP recruitment and regulates mTor pre-mRNA splicing. Nucleic Acids Res. 2019;47:4181–97. 10.1093/nar/gkz099. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 156. Li  Q, Yang  G, Ren  B  et al.  ZC3H14 facilitates backsplicing by binding to exon–intron boundary and 3’ UTR. Mol Cell. 2024;84:4314–4333.e9. 10.1016/j.molcel.2024.10.001. [ DOI ] [ PubMed ] [ Google Scholar ] 157. Liu  D, Dredge  BK, Bert  AG  et al.  ESRP1 controls biogenesis and function of a large abundant multiexon circRNA. Nucleic Acids Res. 2024;52:1387–403. 10.1093/nar/gkad1138. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 158. Paronetto  MP, Zalfa  F, Botti  F  et al.  The nuclear RNA-binding protein Sam68 translocates to the cytoplasm and associates with the polysomes in mouse spermatocytes. MBoC. 2006;17:14–24. 10.1091/mbc.e05-06-0548. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 159. Nie  J, Jiang  M, Zhang  X  et al.  Post-transcriptional regulation of Nkx2-5 by RHAU in heart development. Cell Rep. 2015;13:723–32. 10.1016/j.celrep.2015.09.043. [ DOI ] [ PubMed ] [ Google Scholar ] 160. Davuluri  RV, Suzuki  Y, Sugano  S  et al.  CART classification of human 5’ UTR sequences. Genome Res. 2000;10:1807–16. 10.1101/gr.GR-1460R. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 161. Hedaya  OM, Venkata Subbaiah  KC, Jiang  F  et al.  Secondary structures that regulate mRNA translation provide insights for ASO-mediated modulation of cardiac hypertrophy. Nat Commun. 2023;14:6166. 10.1038/s41467-023-41799-1. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 162. Sharma  A, Masri  J, Jo  OD  et al.  Protein kinase C regulates internal initiation of translation of the GATA-4 mRNA following vasopressin-induced hypertrophy of cardiac myocytes*. J Biol Chem. 2007;282:9505–16. 10.1074/jbc.M608874200. [ DOI ] [ PubMed ] [ Google Scholar ] 163. Rekad  Z, Ruff  M, Radwanska  A  et al.  Coalescent RNA-localizing and transcriptional activities of SAM68 modulate adhesion and subendothelial basement membrane assembly. eLife. 2023;12:e85165. 10.7554/eLife.85165. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 164. Huot  M-É, Brown  CM, Lamarche-Vane  N  et al.  An adaptor role for cytoplasmic Sam68 in modulating Src activity during cell polarization. Mol Cell Biol. 2009;29:1933–43. 10.1128/MCB.01707-08. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Associated Data This section collects any data citations, data availability statements, or supplementary materials included in this article. Data Citations Galaxy CLIP-Explorer: a web server for CLIP-Seq data analysis | GigaScience | Oxford Academic. [ DOI ] [ PMC free article ] [ PubMed ] Supplementary Materials gkag547_Supplemental_Files gkag547_supplemental_files.zip (70.5MB, zip) Data Availability Statement All RNA-sequencing, ribosome profiling, and iCLIP2 data are available in GEO under accession numbers GSE153800 , GSE153799 , and GSE326251 , respectively. The proteomics data have been uploaded to the ProteomeXchange Consortium under accession number PXD009209. Please refer to Supplementary Table S1 for a comprehensive list of the software and tools used in this study, together with their respective references and locations. Articles from Nucleic Acids Research are provided here courtesy of Oxford University Press ACTIONS View on publisher site PDF (4.0 MB) Cite Collections Permalink RESOURCES Similar articles Cited by other articles Links to NCBI Databases Follow NCBI NCBI on X (formerly known as Twitter) NCBI on Facebook NCBI on LinkedIn NCBI on GitHub NCBI RSS feed Connect with NLM NLM on X (formerly known as Twitter) NLM on Facebook NLM on YouTube National Library of Medicine 8600 Rockville Pike Bethesda, MD 20894 Web Policies FOIA HHS Vulnerability Disclosure Help Accessibility Careers NLM NIH HHS USA.gov Back to Top

Related documents

• Mechanobiology-Driven Metabolic Reprogramming: Integrative Roles of YAP/TAZ Signaling and Extracellular Matrix Dynamics.#364690
• A temporal in vivo catalog of chromatin accessibility and expression profiles in pineoblastoma reveals a prevalent role for repressor elements#311939
• From the unfolded protein response to metabolic diseases – lipids under the spotlight#1133788
• Improving Metabolism With HDL Cholesterol#897010
• hnRNPC facilitates coronavirus replication by directly binding the frameshift-stimulatory element of viral genomic RNA.#439300