SF3B1 Mutation Causes Aberrant Splicing and Downregulation of PPP2R5A in Chronic Lymphocytic Leukemia

Qimei Han,Libin Deng,Lirong Pei,Jeong-Hyeon Choi,Victor Jin,Huidong Shi
DOI: https://doi.org/10.1182/blood.v130.suppl_1.2477.2477
IF: 20.3
2017-01-01
Blood
Abstract:The splicing factorSF3B1 is mutated in 10-20% of patients with chronic lymphocytic leukemia (CLL). SF3B1 mutation is associated with faster disease progression, shorter overall survival and fludarabine resistance. However, the molecular mechanism of how SF3B1 mutation contributes to leukemogenesis and the drug resistance phenotype in CLL remains elusive. Comparison of the transcriptome sequencing data of SF3B1 -mutated cancer patients with that of un-mutated samples identified a subset of genes that were aberrantly spliced, mainly at alternative 3‘ splicing sites, by the usage of cryptic 3‘ splice and branchpoint sites in CLL. In this study, we performed an integrated analysis on the transcriptome sequencing data of primary CLL patients with and without SF3B1 mutation available in our lab and those downloaded from Gene Ominibus (GEO) database, as well as those generated from a human B cell precursor leukemia cell line with a knock-in SF3B1K700E mutation. PPP2R5A ,encoding one of the regulatory subunits of tumor protein phosphatase-2A (PP2A B56α), stands out as one of the key genes that were most consistently affected by an aberrant spliced junction, and significantly down-regulated (>2 fold) in CLL patient samples with various SF3B1 hotspot mutations across several CLL transcriptome sequencing data sets. Splicing analyses by RT-PCR and Sanger sequencing confirmed that a 13-nucleotides sequence was added before the 5th exon of PPP2R5A via alternative splicing in CLL patients with SF3B1 mutation. Due to this 13-nucleotides addition, three consecutive premature stop codons were created by frame-shift at a position more than 55 bases upstream of an exon-exon junction, which is a canonical feature of the mRNAs degraded subsequently through non-sense mediated mRNA decay (NMD). The down-regulation of PPP2R5A was confirmed by quantitative PCR (p <0.05) and immunoblot at mRNA and protein levels in CLL patients with SF3B1 mutation as compared to CLL patients without SF3B1 mutation and normal CD19 B+ cell samples. Furthermore, overexpression of mutant SF3B1K700E in HEK 293T cells resulted in the same aberrant splicing pattern observed in CLL patients with SF3B1 mutation and downregulation of PPP2R5A (p <0.05) as compared to HEK 293T cells overexpressing wild type SF3B1. Treatment with cycloheximide, a translation inhibitor known to inhibit NMD, resulted in an increased abundance of alternatively spliced PPP2R5A transcript, suggesting that the aberrant PPP2R5A transcript was degraded by NMD. PP2A, when incorporated with PPP2R5A as the regulatory subunit, dephosphorylates BCL2 and MYC, and results in a decreased anti-apoptotic activity of BCL2 and the degradation of MYC and beta-catenin. Our results demonstrate that SF3B1 mutation causes aberrant splicing and downregulation of tumor suppressor protein PPP2R5A in vitro and in vivo, and reveal one of the consequential roles of SF3B1 mutation in leukemogenesis and the development of drug resistance phenotype.
What problem does this paper attempt to address?