Mutations in the spliceosome machinery, a novel and ubiquitous pathway in leukemogenesis

Abstract

Myelodysplastic syndromes (MDSs) are chronic and often progressive myeloid neoplasms associated with remarkable heterogeneity in the histomorphology and clinical course. Various somatic mutations are involved in the pathogenesis of MDS. Recently, mutations in a gene encoding a spliceosomal protein, SF3B1, were discovered in a distinct form of MDS with ring sideroblasts. Whole exome sequencing of 15 patients with myeloid neoplasms was performed, and somatic mutations in spliceosomal genes were identified. Sanger sequencing of 310 patients was performed to assess phenotype/genotype associations. To determine the functional effect of spliceosomal mutations, we evaluated pre-mRNA splicing profiles by RNA deep sequencing. We identified additional somatic mutations in spliceosomal genes, including SF3B1, U2AF1, and SRSF2. These mutations alter pre-mRNA splicing patterns. SF3B1 mutations are prevalent in low-risk MDS with ring sideroblasts, whereas U2AF1 and SRSF2 mutations are frequent in chronic myelomonocytic leukemia and advanced forms of MDS. SF3B1 mutations are associated with a favorable prognosis, whereas U2AF1 and SRSF2 mutations are predictive for shorter survival. Mutations affecting spliceosomal genes that result in defective splicing are a new leukemogenic pathway. Spliceosomal genes are probably tumor suppressors, and their mutations may constitute diagnostic biomarkers that could potentially serve as therapeutic targets.

Figure 1

Somatic spliceosomal gene (U2AF1, SF3B1, SRSF2, LUC7L2, PRPF8, and ZRSR2) mutations as detected by next-generation sequencing (NGS) and Sanger sequencing technologies. (A) With the use of an NGS-based whole exome sequencing analysis of whole BM DNA from a patient with refractory cytopenia with unilineage dysplasia (left), a mutation of U2AF1 (21q22.3) at position 44 514 777 (T > C) was detected in 13 of 18 reads. Analysis of DNA from CD3⁺ cells showed a much lower frequency of the base change (2 of 15 reads; right), highlighting the somatic nature of this alteration. The finding was confirmed by Sanger sequencing. Arrows and bars indicate the specific nucleotide and predicted codon, respectively. It should be noted that U2AF1 is expressed from the minus strand; therefore, the NGS presentation (top panels) is complementally reversed in comparison to the Sanger sequencing results (middle panels). This heterozygous somatic mutation results in the predicted nucleotide change 470 A > G in exon 6 of the coding region, which lead to the amino acid change Q157R in the second zinc finger domain. In the entire cohort, 27 mutations were observed in 26 patients, including a whole gene deletion. All 26 missense mutations were located in 1 of the 2 zinc finger domains (ZNFs); 2 residues, S34 or Q157, were frequently affected (bottom panels). RRM indicates RNA recognition motif. (B) With the use of an NGS analysis of a patient with CMML (middle left), a mutation of SF3B1 (2q33.1) at position 198 267 491(C > G) was detected in 9 of 24 reads. The somatic nature of this alteration was confirmed by an analogous analysis of the CD3⁺ fraction, with the change being less frequent (2 of 23; middle right). The mutation was confirmed by Sanger sequencing (bottom). This heterozygous somatic mutation results in the nucleotide change 1866 G > T in exon 14 of SF3B1, resulting in the amino acid change E622D in the HSH155 domain. Analysis of the entire cohort identified mutations in 33 patients, including a case with a whole gene deletion. (C) Further screening by NGS led to the detection of a nonsense mutation (R27X) in LUC7L2 (7q34; top) which participates in the recognition of splice donor sites in association with the U1 snRNP spliceosomal subunit, and a missense mutation (M1307I) in PRPF8 (17p13.3; bottom) which is a large U5 snRNP-specific protein essential for pre-mRNA splicing. RS indicates serine/arginine-rich domain; U5 2-snRNA bdg, U5-snRNA binding site 2; and MPN, Mpr1p, Pad1p N-terminal domain. (D) Mutations of SRSF2, an arginine/serine-rich splicing factor, were detected in 29 cases among the entire cohort, including 2 whole gene deletions and a microdeletion within the gene (top). All mutations were heterozygous and affected P95. The somatic nature of the P95R mutation was confirmed with whole BM and T-cell rich fraction DNAs (bottom). (E) A nonsense mutation (W153X) was found in ZRSR2, another arginine/serine-rich splicing regulatory factor, in a case of CMML. ZRSR2 is located at Xp22.2, and the nonsense mutation was hemizygous in this male case (BM).

Figure 2

Frequency and phenotypic association of spliceosomal mutations in myeloid malignancies. (A) In the entire cohort (n = 310), a total of 88 mutations in the spliceosome pathway components U2AF1, SF3B1, and SRSF2 were observed in every subtype of myeloid malignancies, except for MPN. In low-risk MDS, SF3B1 mutations were most frequent among the 3 genes. In particular, SF3B1 was mutated in 15 of 20 cases of RARS (60%). In the high-risk MDS and AML group, U2AF1 mutations were most frequent (15 of 139; 10.8%). In the MDS/MPN group, SRSF2 was most frequently mutated (13 of 46; 28.2%), whereas SF3B1 is mutated at a high frequency in RARS-T (10 of 11; 90.1%). (B) Effect of spliceosomal mutations on clinical outcomes. In the entire cohort, patients with U2AF1 mutations (MT) had worse OS, compared with WT, but SF3B1 mutations made OS significantly shorter. In low-risk MDS, mutation of SF3B1 was a good prognostic factor, but SRSF2 mutations are associated with worse prognosis. In MDS/MPN, patients with mutated U2AF1 had a shorter OS, but SF3B1 mutations were associated with significantly better prognosis. In addition, SRSF2 mutations did not affect outcomes.

Figure 3

Unsplicing of specific genes because of spliceosomal mutations as detected by deep RNA sequencing. Next-generation–based RNA deep sequencing was used to quantitatively study splicing patterns. (A) The top panel shows the intron 5 and exon 6 boundary of TET2 (dotted line). Five reads correspond to transcripts that were not spliced (unspliced) and 4 were spliced at this boundary. The bottom panel shows read counts at the 5′ and 3′ splice sites of each intron (3-10) of TET2. White and black bars indicate the number of spliced and unspliced reads, respectively. In a case of AML with a U2AF1 mutation, more unspliced than spliced reads were observed at the 3′ splice site of intron 5 (left panel), probably because of a loss of spliceosome function. However, unspliced RNAs were less frequent than spliced RNAs in WT RNA sequencing (right panel). (B) At both the 3′ and 5′ splice sites of RUNX1 intron 6, unspliced reads were more frequent than spliced reads in AML cases with U2AF1 and SRSF2 mutations. However, there were fewer unspliced transcripts at the same site in WT and SF3B1 mutant samples. Splicing abnormalities in the selected genes are summarized (bottom right), including the results presented in detail in supplemental Figures 7 and 8.