- Pacholewska, Alicja
- Lienhard, Matthias
- Brüggemann, Mirko
- Hänel, Heike
- Bilalli, Lorina
- Königs, Anja
- Heß, Felix
- Becker, Kerstin
- Köhrer, Karl
- Kaiser, Jesko
- Gohlke, Holger
- Gattermann, Norbert
- Hallek, Michael
- Herling, Carmen D.
- König, Julian
- Grimm, Christina
- Herwig, Ralf
- Zarnack, Kathi
- Schweiger, Michal R.
Mutations in splicing factor 3B subunit 1 (SF3B1) frequently occur in patients with chronic lymphocytic leukemia (CLL) and myelodysplastic syndromes (MDSs). These mutations have different effects on the disease prognosis with beneficial effect in MDS and worse prognosis in CLL patients. A full-length transcriptome approach can expand our knowledge on SF3B1 mutation effects on RNA splicing and its contribution to patient survival and treatment options. We applied long-read transcriptome sequencing (LRTS) to 44 MDS and CLL patients, as well as two pairs of isogenic cell lines with and without SF3B1 mutations, and found >60% of novel isoforms. Splicing alterations were largely shared between cancer types and specifically affected the usage of introns and 3' splice sites. Our data highlighted a constrained window at canonical 3' splice sites in which dynamic splice-site switches occurred in SF3B1-mutated patients. Using transcriptome-wide RNA-binding maps and molecular dynamics simulations, we showed multimodal SF3B1 binding at 3' splice sites and predicted reduced RNA binding at the second binding pocket of SF3B1K700E Our work presents the hitherto most-complete LRTS study of the SF3B1 mutation in CLL and MDS and provides a resource to study aberrant splicing in cancer. Moreover, we showed that different disease prognosises result most likely from the different cell types expanded during carcinogenesis rather than different mechanisms of action of the mutated SF3B1. These results have important implications for understanding the role of SF3B1 mutations in hematological malignancies and other related diseases., We acknowledge the IMB Genomics Core Facility and its NextSeq 500 sequencer (funded by the Deutsche Forschungsgemeinschaft [DFG, German Research Foundation] INST 247/870-1 FUGG). We thank Angelos Constantinou (IGH-Institute of Human Genetics, France) and Marc-Henri Stern (Institut Curie, Paris, France) for sharing the SF3B1 plasmids, Anke Busch (IMB Mainz, Germany) for iCLIP data preprocessing, and Elena Wasserburger-Zichel (University Hospital Cologne, Germany) and Bernd Timmermann (Sequencing Core Unit, Max Planck Institute for Molecular Genetics, Berlin, Germany) for their technical assistance. We furthermore thank the Regional Computing Center of the University of Cologne (RRZK) for providing computing time on the DFG-funded (funding number INST 216/512/1FUGG) high-performance computing (HPC) system CHEOPS as well as IT support. In addition, we acknowledge computational support of the Center for Information and Media Technology, especially the HPC team at the Heinrich Heine University, as well as the computing time provided by the John von Neumann Institute for Computing on the supercomputer JUWELS at the Jülich Supercomputing Centre (user IDs: VSK33, DNAzyme). This work was supported by the DFG Research Infrastructure West German Genome Center (407493903) as part of the Next Generation Sequencing Competence Network (project 423957469). High-throughput sequencing was carried out at the West German Genome Center, and production sites in Cologne and Düsseldorf. The study was funded by the German Research Foundation: KFO286-RP8/SCHW1605/1-1, SCHW1605/4-1 (GO-LONG), SFB1399 and SFB1530 to M.R.S., KFO-286-RP6 to M.H., KFO-286-CP to C.D.H., SFB1530 to M.H., the Volkswagen Stiftung Lichtenberg program to M.R.S., the Center for Molecular Medicine Cologne, CMMC (A12 to M.R.S.), and the EU Horizon 2021 LongTREC (no. 101072892) to R.H., Peer reviewed