Hotspot U2AF1 Mutations Determine Missplicing Selectivity: Novel Mechanisms Altering Splicing Factors

• Published in: Blood Vol. 134; no. Supplement_1; p. 2985
• Main Authors: Adema, Vera, Hershberger, Courtney E, Walter, Wencke, Kerr, Cassandra M, Hutter, Stephan, Nagata, Yasunobu, Awada, Hassan, Kongkiatkamon, Sunisa, Snider, Christina, Co, Milo, Meggendorfer, Manja, Nazha, Aziz, Carraway, Hetty E., Sekeres, Mikkael A., Sole, Francesc, Padgett, Richard A, Haferlach, Torsten, Maciejewski, Jaroslaw P., Visconte, Valeria
• Format: Journal Article
• Language: English
• Published: Elsevier Inc 13.11.2019
• ISSN: 0006-4971; 1528-0020
• DOI: 10.1182/blood-2019-129367

Mutations in splicing factor genes (SF3B1, SRSF2, U2AF1, and ZRSR2) are identified in over 50% of patients diagnosed with myelodysplastic syndrome (MDS). U2AF1 is a U2 auxiliary factor that forms a heterodimer with U2AF2 for the recognition of the 3' splice site (SS) and results in the subsequent recruitment of U2snRNPs during pre mRNA splicing. U2AF1 mutations are present in 11% of MDS and its presence is correlated with an increased risk of progression to AML. Non-canonical mutations are rarely seen in U2AF1 but two highly conserved hotspots (S34, Q157) are frequently seen and result in distinct downstream effects. We performed NGS analysis of a large cohort of patients with myeloid neoplasms (MN; n=3,328) and identified 238 patients with U2AF1 mutations (7%). We analyzed the somatic mutational profile, clonal hierarchy, and splicing profile of patients with U2AF1S34 (n=99), U2AF1Q157 (n=119), and others (n=20; M1/*, A26T/V, R35Q, R118C, E124K, F150L, E152G, C154S, R156H, M172L). The mutational spectrum of U2AF1S34 and U2AF1Q157 was equally dismal but pretty distinct. U2AF1S34 cases were mostly co-mutated for DNMT3A (5%), TET2 (4%), RUNX1 (2%), ASXL1, CBL, ETV6, KRAS, NRAS (1.3%), STAG2 (1%), CUX1, and TP53 (<1%) while U2AF1Q157 had higher numbers of mutations in ASXL1 (10%), SETBP1, TET2 (3%), NRAS (2%), DNMT3A, PHF6 (2%), JAK2 (2%), CBL, EZH2, TP53 (2%), IDH2, RUNX1, STAG2 (2%), KRAS (1%), and IDH1 (<1%). Mutational rank showed: U2AF1S34 was ancestral in 38% of the cases followed by secondary DNMT3A and NRAS (5%, both). Ancestral U2AF1Q157 was found in 35%, with ASXL1 (19%) being the most common secondary hit. Subclonal U2AF1S34 (44%) was most commonly preceded by DNMT3A (21%), while secondary U2AF1Q157 (40%) had ASXL1 (28%) as the most common first hit. U2AF1S34 and U2AF1Q157 were co-dominant to a miscellanea of mutations. U2AF1S34 cases had a shorter OS than U2AF1Q157 cases (n=82 vs. n=101; 20 vs. 25 mo.; P=.002). Ancestral U2AF1S34 or U2AF1Q157 defined a dismal prognosis compared to secondary U2AF1S34 or U2AF1Q157 (OS: n=63 vs. n=86; 29 vs. 37 mo.; P=.03). To investigate the effects of both mutations on splicing patterns, we analyzed RNA-Seq profiles, followed by the rMATS bioinformatics pipeline to determine alternative splicing (AS) events that were classified as skipping exon (SE), retained intron (RI), and 3‘ or 5‘ alternative SS (A3SS, A5SS) (Hershberger, ASH 2019). Overall, 675 AS events in 430 genes were scored significant (pFDR<.05) when the splicing inclusion/exclusion difference was ±10%. U2AF1S34 and U2AF1Q157 caused an equal fraction of SE (79% U2AF1S34; 72% U2AF1Q157). Only 4% of the genes were commonly misspliced by both mutations, while the rest of the genes were uniquely spliced according to each mutation. Some exemplary genes misspliced by both mutations were DDX3X (an RNA helicase) showing a consistent SE at exon 5 and an RI 3-5a and CCNG1 (cell growth regulation) which was enriched for RI 7-7. Among others, U2AF1S34 uniquely affected the splicing of TET2 (SE e3); cell cycle regulators, CDC37L1 (SE), CCNC (SE e8), and HDAC3 (SE e6). We then investigated whether U2AF1 mutations might affect the splicing of other RNA splicing components. This mechanism would lead to the loss of regulation of the spliceosome complex. U2AF1S34 produces selective RIs in SRSF2 (4-4a) and A5SS in SRSF6 (7a and 7). Tumor suppressors and proto-oncogenes were also found to be misspliced by U2AF1S34 including PTEN (RI 3-Ua), CTNNB1 (RI 15-19; 3'SS (19 and 19a), and CCNL1 (RI 4-U). Major regulators of splicing factor activity are phosphatases like PP1R12A and PPP1R12B, which showed an RI 8-7 and an A3SS, respectively. Among genes exclusively misspliced by U2AF1Q157, we identified DEAD-Box helicases [DDX17 (RI), DDX59 (SE e8), DHX29 (RI)], ALAS family members (ALAS1, SE e6; ALAS2, SE e5) and UTX (KDM6A; SE e16). U2AF1Q157 affected the missplicing of DYRK1A (SE e7), a kinase known to phosphorylate SF3B1 at T434 site. In sum, our study suggests that while concurrent mutations in splicing factors lead to lethality, the presence of mutations (as the case of U2AF1S34 and U2AF1Q157) and consequent missplicing of other splicing factors are permissive events in MN and might represent novel mechanisms of disease pathogenesis. Walter:MLL Munich Leukemia Laboratory: Employment. Hutter:MLL Munich Leukemia Laboratory: Employment. Meggendorfer:MLL Munich Leukemia Laboratory: Employment. Nazha:Abbvie: Consultancy; Daiichi Sankyo: Consultancy; Jazz Pharmacutical: Research Funding; Incyte: Speakers Bureau; Novartis: Speakers Bureau; Tolero, Karyopharma: Honoraria; MEI: Other: Data monitoring Committee. Sekeres:Syros: Membership on an entity's Board of Directors or advisory committees; Millenium: Membership on an entity's Board of Directors or advisory committees; Celgene: Membership on an entity's Board of Directors or advisory committees. Haferlach:MLL Munich Leukemia Laboratory: Employment, Equity Ownership. Maciejewski:Alexion: Consultancy; Novartis: Consultancy.