Advertisement
Review

The Changing Mutational Landscape of Acute Myeloid Leukemia and Myelodysplastic Syndrome

Connie A. Larsson, Gilbert Cote and Alfonso Quintás-Cardama
DOI: 10.1158/1541-7786.MCR-12-0695 Published August 2013

Abstract

Over the past few years, large-scale genomic studies of patients with myelodysplastic syndrome (MDS) and acute myelogenous leukemia (AML) have unveiled recurrent somatic mutations in genes involved in epigenetic regulation (DNMT3A, IDH1/2, TET2, ASXL1, EZH2 and MLL) and the spliceosomal machinery (SF3B1, U2AF1, SRSF2, ZRSR2, SF3A1, PRPF40B, U2AF2, and SF1). The identification of these mutations and their impact on prognostication has led to improvements in risk-stratification strategies and has also provided new potential targets for the treatment of these myeloid malignancies. In this review, we discuss the most recently identified genetic abnormalities described in MDS and AML and appraise the current status quo of the dynamics of acquisition of mutant alleles in the pathogenesis of AML, during the transformation from MDS to AML, and in the context of relapse after conventional chemotherapy.

Implications: Identification of somatic mutations in AML and MDS suggests new targets for therapeutic development. Mol Cancer Res; 11(8); 815–27. ©2013 AACR.

Introduction

Acute myelogenous leukemia (AML) and myelodysplastic syndrome (MDS) are hematologic disorders caused by defects in programs that regulate the differentiation potential and self-renewing capacity of myeloid cells originating from multipotent hematopoietic stem cells (HSC). AML is characterized by an abnormal expansion of hematopoietic precursor cells with limited or abnormal differentiation. MDS represents a heterogeneous group of clonal disorders characterized by an expansion of poorly differentiated hematopoietic precursors with limited self-renewing capacity and ineffective hematopoiesis, dysplastic changes, enhanced apoptosis, and a propensity to transform into AML. Recurrent chromosomal structural aberrations have been linked with distinct outcomes and continue to represent one of the most important risk factors when patients are stratified by level of risk before therapy. However, approximately 50% of patients with AML or MDS are cytogenetically normal and lack recurrent structural abnormalities, which suggests other molecular events in the pathogenesis of these diseases. With the application of global DNA sequencing, somatic gene mutations have been found to be more common than previously expected. For example, in cytogenetically normal AML (CN-AML) the discovery of recurrent mutations in three different genes, NPM1, FLT3, and CEBPA, has led to improvements in prognostication, minimal residual disease monitoring, and molecular characterization within these subsets (1–3). As a result, these mutations are thought to be primary genetic events contributing to the pathogenesis of AML (4). Among these, mutations to NPM1 and CEBPA were included as provisional entities in the 2008 World Health Organization (WHO) classification of “AML with recurrent genetic abnormalities” and account for more than 50% of AML patients with normal karyotype (5). The literature covering mutations to these genes and their implication in leukemogenesis is extensive. Therefore, this review does not focus on NPM1, CEBPA, and FLT3 mutations except in the context of mutations to genes in this review. We focus more on recently identified novel genetic alterations and provide an updated report on mutated genes that are well described in MDS and AML.

Recently, the use of high-throughput massive parallel sequencing (i.e., next-generation sequencing) platforms has led to the identification of novel somatic mutations that also have important prognostic value and/or potential as therapeutic targets. Several reports have been published in the past 4 years describing the sequence of 26 AML genomes (12 M1-AML, 13 M3-AML, and 1 therapy-related AML with complex karyotype) as well as exome sequencing of 14 cases of M5-AML (6–12). These reports have unveiled mutations in epigenetic regulator genes such as TET2, IDH1, IDH2, DNMT3a, and EZH2. Importantly, sequence analysis of patients with myeloid malignancies revealed a considerable overlap of mutated genes between patients with cytogenetically normal MDS (CN-MDS) and CN-AML. These mutations may have important therapeutic consequences because drugs that influence the epigenetic regulation of genes [i.e., DNA hypomethylating agents and histone deacetylase (HDAC) inhibitors] are widely used for the treatment of patients with AML or MDS. In addition, recent studies have reported the presence of mutations in genes that encode components of the splicing machinery in patients with MDS and AML (Table 1). Here, we discuss the most exciting recent developments about recurrent mutations in AML and MDS, the prognostic impact of such mutations (Table 2), and their potential as therapeutic targets as well as the dynamics of mutant alleles during the progression from MDS to AML and between genomes at diagnosis and after failure to standard therapy.

View this table:
Table 1.

Frequency of gene mutations in patients with MDS or AML

View this table:
Table 2.

Prognostic implications of gene mutations in patients with MDS or AML

Epigenetic Deregulation in the Pathogenesis of MDS and AML

Regulation of gene expression through epigenetic reprogramming is fundamental to development and cellular differentiation of higher eukaryotes. DNA methylation and histone modifications that change the conformation of chromatin stably alter the gene expression potential through mechanisms that do not change the sequence of the genome. Notably, many malignancies, including AML and MDS, exhibit aberrant DNA methylation and altered histone modifications that result in the alteration of gene expression, such as silencing of tumor suppressors and/or activation of oncogenes. The main epigenetic modification in humans is the methylation of DNA, in which a methyl group is covalently attached to the 5′ carbon of cytosine [5′-methylcytosine (5mC)] within a CpG dinucleotide. There are regions of the genome that are rich in CpG dinucleotides (i.e., CpG islands), which generally span the 5′ regulatory region of genes and are present in approximately half of all human genes. The methylation status of promoter CpG islands is modulated by DNA methyltransferases (DNMT) and dictates the outcome of gene expression (Fig. 1). Under normal conditions, CpG islands are unmethylated, which is characteristic of active gene expression. Conversely, promoter CpG methylation is a hallmark of gene repression and, under normal conditions, is generally restricted to X-chromosome inactivation in women and imprinted germline-specific and tissue-specific gene (13). Using mouse models, the specific roles of different DNMTs have been elucidated and well characterized. Dnmt3a−/− mice displayed normal levels of DNA methylation and were born normal but eventually succumbed to developmental defects and died by 4 weeks of age. Dnmt3b−/− mice exhibited slightly lower levels of DNA methylation and were not viable. However, de novo DNA methylation was severely impaired in Dnmt3a−/−; Dnmt3b−/− double mutants, showing that DNMT3a and DNMT3b have overlapping functions in de novo DNA methylation and embryonic development (14). Dnmt1−/− mice are embryonic lethal and displayed a global reduction in DNA methylation levels, which is consistent with the notion that DNMT1 is essential for maintaining hemi-methylated DNA during replication (13). 5-mC, a mark of DNA methylation, can be further modified by the α-ketoglutarate (α-KG)–dependent oxygenases TET1, TET2, and TET3, which catalyze the conversion of 5-mC to 5′-hydroxymethylcytosine (5-hmC; ref. 15). α-KG is obtained from isocitrate in a process catalyzed by isocitrate dehydrogenase 1 (IDH1) in the cytoplasm and by IDH2 in the mitochondria. Sequencing analysis of MDS and AML patient samples revealed mutations in genes important for DNA methylation such as TET2, IDH1/2, and DNMT3A. Not surprisingly, these mutations are consistent with aberrant DNA methylation patterns being a common feature of MDS and AML (7, 16).

Figure 1.
Figure 1.

Role of proteins encoded by genes mutated in MDS or AML in DNA methylation. DNMTs catalyze the addition of a methyl group to cytosine. IDH enzymes are responsible for catalyzing the reversible conversion of isocitrate to α-KG and CO2 in a two-step reaction. In turn, α-KG is required by the TET proteins to oxidize 5-mC to 5-hmC, an intermediate in DNA demethylation. Mutant IDH proteins metabolize α-KG to 2-HG, an oncometabolite that competitively inhibits the enzymatic activity of TET proteins.

Mutations in genes involved in DNA methylation

DNMT3A.

DNMT3A, which encodes a DNMT and is related to the DNMT1 and DNMT3b genes, has been found mutated in 18% to 25% of patients with AML and approximately 8% of those with MDS. The most common mutation observed is a missense mutation that results in a substitution of arginine-882 to histidine, cysteine, proline, or serine (7, 17). Patients with DNMT3A R882 missense mutations have reduced DNA methylation compared with matched AML patients wild-type for DNMT3A (18). This reduction in DNA methylation suggests that the R882 mutation reduces the methyltransferase activity of the enzyme in a dominant-negative manner, which is further supported by the fact that all R882 mutations are heterozygous (7). Surprisingly, however, differences in gene expression, methylation patterns, or altered total 5-mC content could not be robustly linked to DNMT3A mutational status (7, 19). Serial transplantations of Dnmt3a-null HSCs, which were derived and purified from a conditional Dnmt3a knockout mouse, into wild-type recipient mice resulted in a competitive advantage of mutant cells over the wild-type ones, suggesting that loss of Dnmt3a may contribute to clonal dominance. Reduction in DNA methylation at sites that correlated with increased expression of multipotency genes and downregulation of differentiation factors was also observed in the differentiated progeny of Dnmt3a null HSCs (19). This suggests that loss of de novo DNA methylation impairs the differentiation potential of HSCs, providing a possible explanation of how DNMT3A mutations contribute to the pathogenesis of MDS and AML. DNMT3A mutations also tend to cluster in patients carrying NPM1, FLT3, and/or IDH1 mutations (20). In a study including AML patients under the age of 60 years, DNMT3A mutations were observed in 23% of patients and were associated with worse overall survival (OS) and relapse-free survival than that observed in patients with wild-type DNMT3A alleles (21). Similar results were observed in a cohort of 1185 Chinese patients with AML (22). DNMT3A mutations, typically R822H, were also noted in patients with all subtypes of MDS. Similar to AML, DNMT3A mutations correlated with shorter survival and increased risk of transformation to AML (23).

TET2.

Uniparental disomy and deletions spanning chromosome 4q24 were first described in patients with myeloid malignancies (16). Sequencing of the commonly deleted region unveiled acquired somatic mutations to TET2 (the ten-eleven-translocation gene 2) in patients with myeloid cancers, including patients with MDS (19%–26%) or AML (24%–27%; refs. 16, 24, 25). When they occur, TET2 mutations are present in the vast majority of bone marrow precursor hematopoietic cells (including CD34+ progenitors; ref. 25). TET2 mutations can be observed concomitantly with NPM1, FLT3, JAK2, RUNX1, CEBPA, CBL, and KRAS mutations but are mutually exclusive with IDH1/2 mutations (24, 26). Frequently, patients will carry more than one TET2 mutation (16). Missense mutations frequently occur in regions that are evolutionary conserved and are typically C-terminus truncation mutations that result in loss of TET2 protein function (25). TET2 catalyzes the oxidation of 5-mC to 5-hmC, which has been shown to be a demethylation intermediary (27, 28). The role of TET2 mutations in the pathogenesis of MDS and AML, however, is not well understood, and published reports are often conflicting. In general, loss of TET2 function leads to a global decrease in 5-hmC levels and accumulation of 5-mC, which is a feature of DNA hypermethylation that results in gene repression (28). Surprisingly, an analysis of 88 patients with different myeloid malignancies harboring TET2 mutations revealed that a reduction in 5-hmC levels was associated with CpG hypomethylation, contradicting the paradigm that loss of TET2 function results in hypermethylation of differentially methylated CpG sites (27). Furthermore, the dysplastic phenotype of MDS cannot be attributed to TET2 mutations alone. Experiments using mouse models have shown that loss of Tet2 in the hematopoietic compartment results in an increase in HSC self-renewal with enhanced HSC function and expression of myeloid-specific and self-renewal gene programs, leading to progressive myeloproliferation and extramedullary hematopoiesis (Table 3; ref. 29). In addition, TET2 mutations are relatively frequent in patients with myeloproliferative neoplasms, in whom dysplastic features are very infrequently present. The impact of TET2 mutations on the outcomes of patients with AML or MDS remains controversial. In one study, among patients with AML who had achieved complete remission after induction chemotherapy, the incidence of TET2 mutations was 17% but had no impact on survival. This finding is further supported by another study in 783 younger adult patients with AML, in which TET2 mutations were identified in 7% of cases but had no impact on response or OS (30). Another study found that 20% of patients with MDS had TET2 mutations, clustering in cases of CN-MDS, and not affecting survival (26). However, other groups have reported a favorable impact of TET2 mutations on survival in MDS (31). Yet, other groups have reported that TET2 mutations may have a detrimental effect on survival in AML, particularly among patients with favorable molecular risk (i.e., carriers of CEPBA and NPM1 mutations; ref. 24). Intriguingly, the presence of TET2 mutations seems to improve the response rates of patients with high-risk MDS or AML to azacitidine, although no impact on OS has been yet observed (32). Because of the inconsistent reports on the prognosis of patients with MDS or AML harboring TET2 mutations, the prognostic value of TET2 mutations remains controversial and therefore it is not currently used as a prognostic indicator in clinical practice.

View this table:
Table 3.

Mouse models genetically engineered to express gene mutations encountered in patients with MDS or AML

IDH1 and IDH2.

IDH1 and IDH2 are metabolic enzymes that catalyze the interconversion of α-KG and isocitrate. Mutations to either IDH1 or IDH2 share a common feature of acquiring a neomorphic enzymatic phenotype where α-KG is reduced to 2-hydroxyglutarate (2-HG). 2-HG is presumed to be an oncometabolite that contributes to the tumorigenic process by competitively inhibiting the enzymatic activity of dioxygenases that require α-KG as a cofactor (refs. 15; Fig. 1). Notably, the TET proteins are dioxygenases that metabolize α-KG, and high levels of 2-HG inhibit the enzymatic activity of the TET proteins. Mutations mainly map to codons R132 of IDH1 and R140 or R172 of IDH2. All IDH mutations reported thus far occur in the active catalytic site and participate in isocitrate binding (8, 15). An analysis of 385 patients with de novo AML revealed that those harboring an IDH1 or IDH2 mutation displayed global DNA hypermethylation and impaired hematopoietic differentiation, correlating with an increase in stem and progenitor cell markers (15). Mutations to IDH1 are always heterozygous and exclusively occur at residue 132, with an arginine to histidine being the most common amino acid substitution. It is important to note that IDH mutations are mutually exclusive with TET2 mutations, suggesting a common pathogenetic nexus leading to TET2 inactivation through alterations in either enzyme (24). Although the mechanism for the hypermethylation and leukemogenesis in IDH-mutant AML is not entirely known, it has been linked to downregulation of α-KG levels, which are needed by a variety of enzymes for their function, such as TET proteins, thus affecting differentiation of myeloid cells, or the Jumonji family of histone lysine demethylases (which demethylate H3K9 and H3K36), thus resulting in increased methylation of lysine residues (33). Recently, a mouse model that specifically expresses the Idh1R132H missense mutation in the hematopoietic or in the myeloid compartment was generated to gain a better understanding of how IDH1/2 mutations promote leukemogenesis. These mice showed an increase in hematopoietic progenitors and developed splenomegaly and anemia with extramedullary hematopoiesis, which occurs in 10% to 15% of patients with MDS bearing an IDH1/2 mutation who progress to AML. Furthermore, expression of the mutant in the myeloid compartment led to hypermethylated histones and patterns of aberrant DNA methylation that are consistent with those observed in patients with AML with IDH1/2 mutations (34).

The impact of IDH mutations in the prognosis of AML was investigated in patients with CN-AML, with frequencies of 14% and 19% for IDH1 and IDH2, respectively (35). IDH mutations tend to cluster in cases with mutant NPM1 but wild-type FLT3. In that setting, IDH1 mutations were associated with worse disease-free survival (DFS; ref. 35). Large studies from the United Kingdom, France, and Germany reported IDH mutations in 8% to 16% of patients, which were associated with worse complete remission rates and shorter survival among patients with CN-AML carrying NPM1 mutations (36–38). An analysis of 398 patients younger than 60 years randomly assigned to receive standard-dose or high-dose daunorubicin in the Eastern Cooperative Oncology Group (ECOG) E1900 trial revealed that the favorable effect of NPM1 mutations was limited to patients with simultaneous expression of NPM1 and IDH1 or IDH2 mutations (20). IDH1/IDH2 mutations are less frequent in MDS (4%–11%), compared with AML, but have been linked to shorter OS compared with unmutated patients (39). However, as with TET2 mutations, the prognostic value of IDH mutations remains controversial and needs to be further evaluated both experimentally and in additional cohorts. Importantly, given that mutant IDH proteins acquire neomorphic enzymatic activity, they might lend themselves to inhibition by molecules designed to inhibit the synthesis of 2-HG.

Mutations in chromatin remodeling genes

Although DNA methylation is the best characterized epigenetic modification, a combination of different high-throughput sequencing applications has recently revealed recurrent mutations to chromatin remodelers in MDS and AML, suggesting a role in neoplastic transformation (Table 1). This led to the identification of recurrent mutations to genes that encode histone modifiers, such as ASXL1 (additional sex-comb like-1), and EZH2. ASXL1, and EZH2 encode proteins that are members of the polycomb-group (PcG) family. PcG family members form protein complexes that function to maintain the transcriptionally repressive state of genes, such as the clustered (class I) homeobox genes that encode HOX proteins, which are frequently overexpressed in AML and confer a poor prognosis. For instance, the H2.0-like homeobox (HLX) gene is overexpressed by 2- to 6.8-fold in 87% of patients with AML compared with CD34+ cells of healthy donors, and increased HLX expression correlated with inferior survival (40).

ASXL1.

ASXL1 has opposing functions in maintaining both the activation and suppression of HOX genes, depending on the histone-modifying protein that interacts with ASXL1 (41). Mutations in ASXL1 that map to the C-terminus of the protein have been identified in 3% to 22% of patients with MDS or AML, and correlate with poor OS (Table 2; ref. 20, 26). In vivo studies using Asxl1 knockout mice showed that loss of ASXL1 resulted in a mild phenotype characterized by differentiation defects in the myeloid and lymphoid progenitors without ever progressing to a more severe disease state that is typically observed in patients with MDS and AML harboring an ASXL1 mutation (42). It is possible that loss of ASXL1 is compensated for by homologous proteins with redundant functions, such as ASXL2 and ASXL3, thus resulting in a milder phenotype. However, mutations conferring loss of both ASXL1 loci are rare events. In fact, 90% of ASXL1 mutations are truncating mutations that result in the loss of the plant homeodomain (PHD) of the protein, yet the function of this mutant protein remains unclear (26). It is possible that a truncating mutation in ASXL1 results in gain-of-function or dominant-negative activity of the mutated protein. Recently, using a panel of human myeloid leukemia cell lines, it was shown that leukemia cells with homozygous ASXL1 frameshift/nonsense mutations had undetectable ASXL1 protein levels, whereas leukemia cells with heterozygous ASXL1 mutations showed a reduction in ASXL1 protein levels. Furthermore, knockdown of ASXL1 in these cell lines is associated with global loss of the trimethylation mark of H3K27 and upregulation of HOXA gene expression (43). Interestingly, ASXL1 is a cofactor essential for the enzymatic activity of BAP1, and coimmunoprecipitation assays showed that ASXL1 forms a complex with BAP1 in human myeloid leukemia cells that are wild-type for ASXL1. Selective BAP1 haploinsufficiency in hematopoietic cells in mice induces a myelodysplastic phenotype reminiscent of chronic myelomonocytic leukemia (CMML), suggesting than the ASXL1–BAP1 axis may play an important role in MDS (44). Contrary to this, loss of BAP1 in human leukemia cell lines with wild-type ASXL1 did not lead to an increase in HOXA gene expression, suggesting that leukemic transformation as a result of ASXL1 loss occurs through a mechanism that is independent of BAP1 (43). In a study involving 439 patients with MDS, ASXL1 mutations were found in 14% of patients and were independent prognostic factors for survival [HR for death from any cause, 1.38; 95% confidence interval (CI), 1.00–1.89; ref. 26). ASXL1 mutations were found in 5% of patients with AML and were inversely associated with FLT3-ITD and mutually exclusive with NPM1 mutations, and represented an unfavorable prognostic factor for survival (45).

EZH2.

EZH2 makes up the catalytic component of the polycomb repressive complex 2 (PRC2), where it interacts with EED and SUZ12. Mutations to any of these genes have been described in early T-cell precursor acute lymphoblastic leukemia (T-ALL; ref. 46). The SET domain of EZH2 catalyzes the progressive trimethylation of H3K27 (H3K27me3) to induce the repression of target genes (47). EZH2 is overexpressed in many different epithelial cancers and lymphomas, whereas inactivating mutations are observed in myeloid malignancies, suggesting that H3K27me3 alterations have biologic consequences that are tissue specific (48). In patients with MDS, mutations to EZH2 are generally missense or truncating mutations that disrupt or delete the SET domain, leading to a reduction of H3K27me3 levels (48, 49). Another study reported that biallelic deletion of Ezh2 in the HSCs of mice was sufficient to cause T-ALL, which occurred with almost complete penetrance in less than 1 year (50). PRC proteins play a significant role in regulating HOX gene expression, providing a possible causal link between inactivating mutations to EZH2 and its implication in myeloid malignancies. Certain HOX proteins, such as HOXA7-11, are frequently overexpressed in AML, and in vivo studies showed that overexpression of HOXA10 in HSCs was sufficient to cause AML in mice (51, 52). One study showed that repression of certain HOX genes involves both promoter CpG methylation and NSPc1-mediated H2A ubiquitination in a process that requires the cooperation of EZH2, DNMT3A, and NSPc1, a component of PRC1. Trimethylation of H3K27 is required for the recognition and binding of DNMT3A and NSPc1, and loss of H3K27 trimethylation, as a result of EZH2 depletion, led to impairments in NSPc1-mediated H2A ubiquitination and CpG methylation in HOXA gene clusters (53). EZH2 mutations are observed in 6% of patients with MDS and are associated with worse OS compared with patients with wild-type EZH2 (26, 48, 49). This is not surprising, as EZH2 maps to chromosome 7, which is a genomic area frequently lost in MDS and associated with very poor prognosis. Interestingly, EZH2 mutations are infrequently found in patients with AML (20). In fact, loss of Ezh2 perturbs the progression of AML by promoting the differentiation of leukemic cells in mice transplanted with MLL-AF9–transformed granulocyte-macrophage progenitors (GMP), suggesting that EZH2 is oncogenic in AML (54).

MLL.

MLL (mixed lineage leukemia) is another SET domain-containing protein that represents one of the first identified epigenetic modifiers that is frequently altered in leukemia. MLL is translocated in 10% of AML cases and in 30% of secondary AML. MLL is recruited to many promoters where it mediates H3K4 methyltransferase activity, resulting in activated gene expression (55). It has been speculated that mislocalized activity of the H3K79 histone methyltransferase, DOT1L, is the main leukemogenic driver in cases of AML carrying MLL rearrangements. Importantly, the novel DOT1L inhibitor EPZ004777 has been shown to selectively inhibit H3K79 methylation, resulting in blocked expression of leukemogenic genes with negligible impact on non-MLL–translocated cells (56). These data suggest that pharmacologic inhibition of DOT1L may represent a potential therapeutic option for patients with MLL rearrangements.

BCOR.

The genome of a female patient with AML and devoid of mutations to FLT3, NPM1, and CEBPA was sequenced as a means to gain a better understanding of the molecular characterization of the subgroup of patients with CN-AML who lack NPM1 and CEBPA mutations. Further mutational screenings and additional sequencing of CN-AML cases identified recurrent mutations to BCOR, which encode a corepressor of BCL6. Mutations to BCOR occur at a frequency of 3.8% in unselected CN-AML cases. However, in the subgroup of CN-AML that is least characterized, BCOR mutations are observed in 17% of cases. This subgroup includes patients with no MLL-PTD and lacking mutations to CEBPA, NMP1, FLT3, or IHD1/2 (4). In addition, a single BCOR mutation has been observed in a small MDS examination that included 26 patient samples (57). BCOR has been shown to interact with HDACs, and mutations to BCOR are associated with DNMT3A mutations, implicating a cooperative epigenetic mechanism for AML pathogenesis. Furthermore, this study showed that BCOR mutations portend inferior survival (4).

Mutations in spliceosomal genes

The spliceosome plays a fundamental role in the processing of nascent RNA transcripts. Spliceosomes are complexes of small nuclear RNAs and proteins that remove (splice) introns and unconventional exons from primary transcripts (pre-mRNA) to assemble recognized exons (splicing) into the mature mRNA (Fig. 2). Disease-associated aberrant RNA splicing is a well-described paradigm that is primarily gene specific and due to mutations in cis-regulatory sequences that guide spliceosomal recognitions of exons (58). However, a growing number of mutations are known to target components of the spliceosome, and thus potentially to impart global rather than gene-specific RNA splicing changes. In patients with myeloid malignancies, somatic mutations have been described in eight core spliceosomal genes—SF3B1, SRSF2, U2AF35 (U2AF1), ZRSR2, SF3A1, PRPF40B, U2AF65 (U2AF2), and SF1. Mutations were highly recurrent in SF3B1, SRSF2, ZRSR2, and U2AF1, whereas mutations in the other spliceosomal genes (U2AF2, SF1, PRPF40B, and SF3A1) occurred with a frequency of 1% or less (57, 59–61). Mutations to spliceosomal genes are highly specific for patients with refractory anemia with ring sideroblasts (RARS; 84.9%), or MDS without ringed sideroblasts (43.9%), and CMML (54.5%) compared with patients with de novo AML (6.6%) or myeloproliferative neoplasms (9.6%). The spectrum of spliceosomal mutations also seems to be specific to disease type (57). As mutations to U2AF2, SF1, PRPF40B, and SF3A1 are rare events in MDS and AML, there will be no further discussion in regards to these genes within this review. This review focuses on the four most frequently mutated splicing genes, SF3B1, SRSF2, ZRSR2, and U2AF1, occurring in all subtypes of MDS. It is important to note that with the exception of ZRSR2, the majority of the mutations described to date are targeted missense rather than nonsense or frameshift mutations, suggesting a gain-of-function rather than a reduction in splicing factor levels. Because these genes play a fundamental role in the recognition of the 3′ splice site during early spliceosome formation, functional changes in these proteins, such as in their ability to stabilize U2 interactions, have the potential to differentially affect both alternative and constitutive splicing decisions.

Figure 2.
Figure 2.

MDS and AML mutations target spliceosomal genes associated with initial 3′ splice site recognition. Somatic mutations have been found in all major genes involved in 3′ splice site recognition and spliceosome commitment. This includes SF1, U2AF2, and U2AF1, which are critical for commitment complex E formation in U2-dependent splicing, as well as SF3A1, SF3B1, and ZRSR2, proteins that are essential for stabilization of splicing complex A. It is important to note that SRSF2 serves as a generalized enhancer of 3′ splice recognition facilitating splice site recognition through stabilization of U2AF1 interaction with the AG dinucleotide. Because SRSF2 functions through binding of exonic splicing enhancer (ESE) sequences it acts on only a subset of exons. ZRSR2, which is the only spliceosomal gene predominately targeted by inactivating nonsense mutations, is essential for U2AF-independent recognition of U12-mediated RNA splicing.

SF3B1.

The SF3B1 gene encodes the 155-kDa subunit of the SF3B complex, which serves to target the U2snRNP complex to the branch point during complex A formation (Fig. 2). Mutations are found in more than 80% of patients with RARS, whereas it is mutated in only 6% of patients with MDS without ringed sideroblasts, specifically implicating SF3B1 mutation in the pathogenesis of this subtype of MDS (57, 62). Mutations to SF3B1 confer a more favorable prognosis in MDS, MDS/MPN, and RARS cases and are associated with lower risk of progression to AML compared with patients with wild-type SF3B1 (59, 62). SF3B1 mutations are predominantly heterozygous missense mutations at residues K700, K666, R625, and H662, but the specific effect of these mutations on protein function remains unknown. It is hypothesized that mutations in SF3B1 alter branch point and 3′ splice site recognition, which contributes to changes in the mature mRNA pool. Indeed, a recent examination of splicing of 81,564 exons in 9,069 genes revealed that 423 exons in 350 genes were differentially used in mutants compared with a representative healthy donor. Notable genes with aberrant RNA splicing included ASXL1, CBL, EZH, and RUNX. Some consequences of the aberrant splicing pathways were structural differences in iron distribution contributing to the pathogenesis of RARS (63). Importantly, SF3B1 is involved in repressing HOX genes by physically interacting with PcG proteins, suggesting a potential role for SF3B1 in epigenetic regulation. Loss of one Sf3b1 allele leads to developmental defects characterized by various skeletal alterations as a result of impaired Hox gene silencing in murine models. Furthermore, Mll mutations rescue the Sf3b1+/− phenotype (64). It is important to emphasize that translocations or insertions involving MLL, resulting in an MLL gene fusion, occur in about 10% of adult AML and confer a poor prognosis.

SRSF2.

SRSF2 is a member of the SR family of RNA-binding proteins that predominantly function to enhance exon recognition by binding regulatory splicing sequences. At the 3′ splice site region, SRSF2 facilitates the recruitment of both U2AF1 and SF3A2, which is consistent with the model whereby gene mutation alters regulation of exon recognition (Fig. 2). For the myeloid diseases, SRSF2 is currently the second most mutated splicing gene. Mutations to SRSF2 are almost always heterozygous missense mutations that specifically occur at P95 and are more prevalent in CMML cases (28%) when compared with patients with RARS (1.5%), MDS without ringed sideroblasts (11.6%), and AML (∼7%; ref. 57). MDS harboring an SRSF2 mutation is associated with a higher rate of transformation to AML with shorter progression time and lower OS compared with MDS with wild-type SRSF2 (65). Depletion of SRSF2 leads to an accumulation of DNA damage and genomic instability and to induction of G2–M cell-cycle arrest, uncovering pathways that may be targeted as a result of aberrant RNA splicing (66). Thus, gain-of-function mutations could provide a compelling reason as to why mutations to SRSF2 confer a poor prognosis. Furthermore, SRSF2 mutations are associated with concomitant mutations in IDH1 and RUNX1, a transcription factor that is frequently mutated and/or misspliced in MDS (61). Both mutations and missplicing of RUNX1 are associated with inferior event-free survival (EFS) and OS. The fact that SRSF2 mutations frequently co-occur with mutations in RUNX1, which also happens to be abnormally spliced when SRSF2 is mutated, may explain why deregulations in these two genes lead to a similar phenotype and prognosis.

ZRSR2.

ZRSR2 is a member of the U2AF1-related protein family, which functions in early spliceosome assembly through direct interactions with the U2AF 65-kDa subunit (U2AF2). Unlike the other members of this group, ZRSR2 does not seem capable of functionally replacing U2AF1, and while required its precise role in U2-dependent splicing is unclear (67). An important consideration, however, is that ZRSR2 plays a different role in assembly of the minor U12-dependent spliceosome, where it is capable of completely replacing the U2AF complex (68). Mutations of the ZRSR2 gene occur at similar frequency in MDS (∼8%) and CMML (∼8), and unlike other spliceosomal gene mutations they do not occur in select “hot spots” (60). Instead, nearly all reported mutations are nonsense or frameshift mutations, clearly establishing a loss-of-function role. In one study, patients with ZRSR2 mutations seemed to cluster with RAEB-1 and RAEB-2, MDS subtypes with further associated pronounced thrombocytopenia (60). Furthermore, these mutations were associated with a higher AML transformation rate and poor OS. A specific link to U12-dependent gene splicing remains to be established.

U2AF1.

The U2AF1 gene encodes the 35-kDa subunit of U2AF and functions to recognize the AG dinucleotide marking the end of the intron. For introns with a weak U2AF2-binding sequence, U2AF1 is critical for splice site recognition (69). U2AF1-related proteins, ZRSR1, ZRSR2, and U2AF1L4, substitute for U2AF1 in some spliceosomal reactions, clearly defining the potential to affect a subset of genes (68). Mutations to U2AF1 were not observed in patients with RARS but occurred with relatively equal frequencies in MDS without ringed sideroblasts (11.6%), CMML (23%), and AML (11.0%) cases (57, 59, 62, 70). Mutations to U2AF1 occur exclusively at codons S34 or Q157 and correlate with a more rapid transformation from MDS to AML, although their impact on OS remains unclear (65). Expression of mutant U2AF1 in HeLa cells leads to a reduction in cell proliferation and enhanced apoptosis, which are common features of MDS (57). Much like SRSF2 mutations, mutations to U2AF1 are implicated in abnormal splicing of certain genes, namely RUNX1 and TET2, and are also associated with mutations in ASXL1 and TET2. Furthermore, expression of mutant U2AF1 resulted in upregulation of nonsense-mediated RNA decay (NMD) pathway genes, indicating that U2AF1 mutations lead to increases in RNA splicing defects that require targeting for degradation (61).

Overall, mutations to SF3B1, SRSF2, ZRSR2, and U2AF1 are the most frequent gene aberrations occurring in all subtypes of MDS and are mutually exclusive (57, 59–62).

TP53 mutations in the pathogenesis of MDS/AML

TP53 is the tumor suppressor more frequently mutated in human cancer. Mutations to TP53 occur in approximately 10% of de novo MDS and AML cases, whereas it is mutated in more than 25% of therapy-related MDS (t-MDS) and AML (t-AML) cases. Although the presence of TP53 mutations has long been recognized as a poor risk factor for survival, novel aspects of TP53 biology have been linked to the pathogenesis of myeloid malignancies. It is now widely recognized that patients with TP53 mutations typically have a complex karyotype (CK; ref. 71). It has been recently shown that the events leading to complex genomic rearrangements in CK-AML are orchestrated by a one-step catastrophic event termed chromothripsis (Fig. 3; ref. 72). Genomic rearrangements caused by chromothripsis are characterized by complex somatic rearrangements with alternating copy-number states affecting one or a few chromosomes. Two scenarios describing the events leading to p53-mediated chromothripsis have been discussed. One scenario describes critical telomere shortening, followed by chromosomal end-to-end fusion, and subsequent breakage as the events leading to complex interchromosomal rearrangements. The fact that progressive telomere shortening occurs as a function of patient age may also explain why chromothripsis is more frequent in advanced aged TP53-mutated AML and why the frequency of CK-AML increases with age. The other scenario postulates that chromothripsis occurs when double-stranded DNA breaks, caused by ionizing radiation or DNA-damaging agents, are subjected to repair by the error-prone nonhomologous end joining (NHEJ) pathway, the dominant repair pathway in p53-deficient (e.g., mutated) cells (72).

Figure 3.
Figure 3.

Schema representing a potential path of acquisition of mutant disease alleles leading to MDS, transformation to AML, and relapse after failure to respond to standard chemotherapy. Normal HSCs randomly acquire hundreds of nonpathogenetic mutations in a time-dependent fashion. At some point, a “driver” mutation is acquired (e.g., TET2) that confers a growth advantage to a “founding clone.” Previously acquired background mutations (i.e., “passenger”) migrate with the driver mutation and are present in all its progeny. Driver mutations disturb normal myeloid maturation and induce dysplastic features clinically compatible with MDS. Subsequent acquisition of mutant alleles in the founding clone (e.g., FLT3) may result in leukemic transformation (i.e., secondary AML). Patients with AML receive induction to remission cytotoxic chemotherapy followed by consolidation chemotherapy. A fraction of patients are cured upon eradication of the founding clone as well as “subclones” with a more complex mutational make-up than the former. However, a significant fraction of patients relapse after chemotherapy due to resistance and subsequent reexpansion of resistant clones. This sequence of events is driven by the acquisition of new mutations (e.g., RUNX1, ASXL1, and TP53) either in the founding clone or in a subclone of the founding clone as a consequence of therapy-induced selection pressure or as a result of DNA damage directly induced by cytotoxic chemotherapy. TP53 mutations have been associated with a phenomenon termed chromothripsis, which may occur at any point during the pathogenesis of AML. The latter induces local chromosome fragmentation resulting from DNA double-strand breaks, likely repaired by NHEJ. The net result is a conglomerate of complex chromosomal rearrangements that adds a new layer of genomic complexity to that provided by the numerous point mutations present in AML genomes and provides the genetic basis for tumor growth and resistance to therapy.

TP53 function is also implicated in the pathogenesis of 5q− syndrome. In vitro studies showed that downregulation of RPS14, a 40S ribosomal subunit, in human bone marrow cells phenocopied the erythroid failure observed in patients with 5q− syndrome (73). Deletion of the Cd74-Nid67 interval (syntenic to the 5q− syndrome commonly deleted region that contains RPS14) in the bone marrow of mice resulted in elevated levels of p53 and increased bone marrow apoptosis. Crossing Cd74-Nid67+/− mice with p53-null mice (74) or pharmacologic inhibition of p53 rescues the erythroid defect (75), thereby definitely implicating p53 in the pathogenesis of this syndrome.

Dynamics of mutated alleles in MDS and AML

Although the use of next-generation sequencing platforms has shed invaluable new light into the mutational landscape in MDS and AML, several questions remain to be answered. First, available studies only provide a snapshot of the mutational make-up of MDS or AML but do not address the mutational clonal hierarchy within the bulk of leukemic cells or aid in establishing which mutations bear real weight in the pathogenesis of the disease (i.e., “driver” vs. “passenger” mutations). Second, serial mutational screenings will be necessary to establish the genetic lesions involved in the progression from MDS to AML. Third, extensive work is warranted to exploit the available wealth of information for diagnostic, prognostic, and therapeutic purposes.

A recent study has shown that hematopoietic stem/progenitor cells (HSPC) of normal individuals contain hundreds of random mutations, that the number of mutations is positively correlated with age, and that the mutational spectrum between normal healthy HSPCs and that of AML genomes is very similar. This suggests that normal HSPCs accumulate random mutations over time that are irrelevant for AML pathogenesis but will be transmitted (i.e., passengers) to all AML clones if and when a driver mutation is acquired and provides a growth advantage for an HSPC. Furthermore, a comparative analysis of 12 FAB-M1 CN-AML and 12 FAB-M3 AML genomes identified 13 recurring mutations in M1-AML, including NPM1, DNMT3a, and IDH1, which only coexisted in M1-AML, suggesting that the latter are likely M1-AML-initiating mutations rather than cooperating mutations (12). Not surprisingly, an analysis of the dynamics of mutated alleles during the progression from MDS to AML in 7 patients by high-throughput sequencing, DNA copy-number analysis, and microarray-based gene-expression studies found that the MDS clones also contained hundreds of acquired mutations. More importantly, the AML arising from them were derived from at least one subclone acquiring new driver mutations or genomic rearrangements and that 85% to 90% of MDS and secondary AML unfractionated bone marrow cells were clonal, even with a myeloblast count of less than 5% (76). In this group of patients, 11 genes were found mutated recurrently: CDH23, NPM1, PTPN11, RUNX1, SMC3, STAG2, TP53, U2AF1, UMODL1, WT1, and ZSWIM4 (76). It is also important to consider that mutation of spliceosomal genes may serve as complete genetic drivers of tumorigenesis, a concept originating from the observation that SRSF1 is capable of inducing tumor formation (77). It is clear that mutations involving spliceosomal genes specifically affect genes known to be sequentially targeted by individual mutations.

Although most patients with AML respond to initial induction chemotherapy, a large number of them will relapse and exhibit refractoriness to subsequent therapy that almost invariably leads to the patient's death. It is therefore important to understand the genetics underlying primary tumorigenesis as well as disease recurrence. The genome sequences of 8 patients with AML at diagnosis have been compared from the same patients at the time of relapse by deep sequencing (44). Such analysis unveiled novel mutations in AML such as those in WAC, SMC3, DIS3, DDX41, and DAXX. More importantly, two patterns of clonal evolution at relapse were identified. These include one in which the founding clone at diagnosis acquired new mutations that were later identified in clones at relapse, and another one in which a subclone of the founding clone withstood initial therapy and acquired additional mutations that led to clonal expansion and relapse (Fig. 3). In both instances, initial therapy failed to eradicate all or specific subclones from the founding clone. Interestingly, some of the mutations observed at relapse were found to be due to DNA damage secondary to cytotoxic chemotherapy, highlighting potential deleterious effects of current AML therapeutic regimens (44).

Perspective and Conclusions

Available mutational data challenge the conceptual “two-hit” dictum in AML and MDS, whereby the pathogenesis of these diseases is driven by a mutation in a gene-regulating cell growth or proliferation that is complemented by a second mutation in a transcription factor regulation myeloid maturation and depicts a much more complex picture about the pathogenesis of these malignancies. It is now becoming more recognized that mutations driving neoplastic transformation and progression are not restricted to tumor suppressors, oncogenes, or pathway-specific genes involved in growth signaling, proliferation, migration, and survival. Regulatory proteins that are not defined to a specific pathway but have functional roles in the regulation of global processes (i.e., transcription and translation) are frequently targeted to functionally deregulating specific mutations in many different blood disorders, including AML and MDS. Cross-examinations of the epigenetic profile in patients with myeloid cancers harboring mutations to epigenetic genes suggest that aberrations to the epigenome play a significant role in the pathogenesis of these malignancies. This is not surprising considering that many of these epigenetic regulators are important for maintaining the repressive state of genes involved in growth, stem cell renewal, and survival. Importantly, epigenetic-modifying agents such as HDAC inhibitors and hypomethylating agents have already shown clinical activity in patients with AML and MDS. More importantly, the number of different genetic lesions in epigenetic regulators provides researchers with more potential therapeutic targets and currently remains an area of focus in cancer drug discovery. Furthermore, a high proportion of patients with MDS carry somatic mutations in several different spliceosome genes, namely SF3B1, SRSF2, and U2AF1. Although the roles of these genes in RNA splicing are well described, little is known about their roles in the context of hematopoietic function or how splicing mutations contribute to the pathogenesis of MDS and in some cases, the transformation to AML. With the accumulating data correlating spliceosome mutations with mutations in epigenetic regulators, it is arguable that abnormal RNA splicing parallels aberrant epigenetic reprogramming in the pathogenesis of MDS or that a direct link may exist between RNA splicing and epigenetic regulation. The overlap and frequency of mutations to these genes and the data correlating splicing mutations with the missplicing of epigenetic regulators recurrently mutated in myeloid malignancies further supports this notion. Studying the interplay between concomitantly mutated genes will undoubtedly shed new light into the mechanisms whereby normal myeloid cells become malignant and will aid in developing more biologically informed therapeutic approaches.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.

Authors' Contributions

Conception and design: C.A. Larsson, A. Quintás-Cardama

Development of methodology: A. Quintás-Cardama

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): A. Quintás-Cardama

Writing, review, and/or revision of the manuscript: C.A. Larsson, G. Cote, A. Quintás-Cardama

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): A. Quintás-Cardama

Study supervision: A. Quintás-Cardama

Grant Support

Work in the authors' laboratory is supported by the NIH (grant 5T32CA009299-32) and the American Legion Auxiliary Fellowship.

Acknowledgments

The authors thank Amanda Wasylishen for critical review of the article.

  • Received December 18, 2012.
  • Revision received March 12, 2013.
  • Accepted April 11, 2013.

References

  1. 1.↵
    1. Falini B,
    2. Mecucci C,
    3. Tiacci E,
    4. Alcalay M,
    5. Rosati R,
    6. Pasqualucci L,
    7. et al.
    Cytoplasmic nucleophosmin in acute myelogenous leukemia with a normal karyotype. N Engl J Med 2005;352:254–66.
    OpenUrlCrossRefPubMed
  2. 2.↵
    1. Nakao M,
    2. Yokota S,
    3. Iwai T,
    4. Kaneko H,
    5. Horiike S,
    6. Kashima K,
    7. et al.
    Internal tandem duplication of the flt3 gene found in acute myeloid leukemia. Leukemia 1996;10:1911–8.
    OpenUrlPubMed
  3. 3.↵
    1. Pabst T,
    2. Mueller BU,
    3. Zhang P,
    4. Radomska HS,
    5. Narravula S,
    6. Schnittger S,
    7. et al.
    Dominant-negative mutations of CEBPA, encoding CCAAT/enhancer binding protein-alpha (C/EBPalpha), in acute myeloid leukemia. Nat Genet 2001;27:263–70.
    OpenUrlCrossRefPubMed
  4. 4.↵
    1. Grossmann V,
    2. Tiacci E,
    3. Holmes AB,
    4. Kohlmann A,
    5. Martelli MP,
    6. Kern W,
    7. et al.
    Whole-exome sequencing identifies somatic mutations of BCOR in acute myeloid leukemia with normal karyotype. Blood 2011;118:6153–63.
    OpenUrlAbstract/FREE Full Text
  5. 5.↵
    1. Vardiman JW,
    2. Thiele J,
    3. Arber DA,
    4. Brunning RD,
    5. Borowitz MJ,
    6. Porwit A,
    7. et al.
    The 2008 revision of the World Health Organization (WHO) classification of myeloid neoplasms and acute leukemia: rationale and important changes. Blood 2009;114:937–51.
    OpenUrlAbstract/FREE Full Text
  6. 6.↵
    1. Ley TJ,
    2. Mardis ER,
    3. Ding L,
    4. Fulton B,
    5. McLellan MD,
    6. Chen K,
    7. et al.
    DNA sequencing of a cytogenetically normal acute myeloid leukaemia genome. Nature 2008;456:66–72.
    OpenUrlCrossRefPubMed
  7. 7.↵
    1. Ley TJ,
    2. Ding L,
    3. Walter MJ,
    4. McLellan MD,
    5. Lamprecht T,
    6. Larson DE,
    7. et al.
    DNMT3A mutations in acute myeloid leukemia. N Engl J Med 2010;363:2424–33.
    OpenUrlCrossRefPubMed
  8. 8.↵
    1. Mardis ER,
    2. Ding L,
    3. Dooling DJ,
    4. Larson DE,
    5. McLellan MD,
    6. Chen K,
    7. et al.
    Recurring mutations found by sequencing an acute myeloid leukemia genome. N Engl J Med 2009;361:1058–66.
    OpenUrlCrossRefPubMed
  9. 9.↵
    1. Welch JS,
    2. Westervelt P,
    3. Ding L,
    4. Larson DE,
    5. Klco JM,
    6. Kulkarni S,
    7. et al.
    Use of whole-genome sequencing to diagnose a cryptic fusion oncogene. JAMA 2011;305:1577–84.
    OpenUrlCrossRefPubMed
  10. 10.↵
    1. Link DC,
    2. Schuettpelz LG,
    3. Shen D,
    4. Wang J,
    5. Walter MJ,
    6. Kulkarni S,
    7. et al.
    Identification of a novel TP53 cancer susceptibility mutation through whole-genome sequencing of a patient with therapy-related AML. JAMA 2011;305:1568–76.
    OpenUrlCrossRefPubMed
  11. 11.↵
    1. Zalzman M,
    2. Falco G,
    3. Sharova LV,
    4. Nishiyama A,
    5. Thomas M,
    6. Lee SL,
    7. et al.
    Zscan4 regulates telomere elongation and genomic stability in ES cells. Nature 2010;464:858–63.
    OpenUrlCrossRefPubMed
  12. 12.↵
    1. Welch JS,
    2. Ley TJ,
    3. Link DC,
    4. Miller CA,
    5. Larson DE,
    6. Koboldt DC,
    7. et al.
    The origin and evolution of mutations in acute myeloid leukemia. Cell 2012;150:264–78.
    OpenUrlCrossRefPubMed
  13. 13.↵
    1. Esteller M
    . CpG island hypermethylation and tumor suppressor genes: a booming present, a brighter future. Oncogene 2002;21:5427–40.
    OpenUrlCrossRefPubMed
  14. 14.↵
    1. Okano M,
    2. Bell DW,
    3. Haber DA,
    4. Li E
    . DNA methyltransferases Dnmt3a and Dnmt3b are essential for de novo methylation and mammalian development. Cell 1999;99:247–57.
    OpenUrlCrossRefPubMed
  15. 15.↵
    1. Figueroa ME,
    2. Abdel-Wahab O,
    3. Lu C,
    4. Ward PS,
    5. Patel J,
    6. Shih A,
    7. et al.
    Leukemic IDH1 and IDH2 mutations result in a hypermethylation phenotype, disrupt TET2 function, and impair hematopoietic differentiation. Cancer Cell 2010;18:553–67.
    OpenUrlCrossRefPubMed
  16. 16.↵
    1. Delhommeau F,
    2. Dupont S,
    3. Della Valle V,
    4. James C,
    5. Trannoy S,
    6. Masse A,
    7. et al.
    Mutation in TET2 in myeloid cancers. N Engl J Med 2009;360:2289–301.
    OpenUrlCrossRefPubMed
  17. 17.↵
    1. Yan XJ,
    2. Xu J,
    3. Gu ZH,
    4. Pan CM,
    5. Lu G,
    6. Shen Y,
    7. et al.
    Exome sequencing identifies somatic mutations of DNA methyltransferase gene DNMT3A in acute monocytic leukemia. Nat Genet 2011;43:309–15.
    OpenUrlCrossRefPubMed
  18. 18.↵
    1. Yamashita Y,
    2. Yuan J,
    3. Suetake I,
    4. Suzuki H,
    5. Ishikawa Y,
    6. Choi YL,
    7. et al.
    Array-based genomic resequencing of human leukemia. Oncogene 2010;29:3723–31.
    OpenUrlCrossRefPubMed
  19. 19.↵
    1. Challen GA,
    2. Sun D,
    3. Jeong M,
    4. Luo M,
    5. Jelinek J,
    6. Berg JS,
    7. et al.
    Dnmt3a is essential for hematopoietic stem cell differentiation. Nat Genet 2012;44:23–31.
    OpenUrlCrossRefPubMed
  20. 20.↵
    1. Patel JP,
    2. Gonen M,
    3. Figueroa ME,
    4. Fernandez H,
    5. Sun Z,
    6. Racevskis J,
    7. et al.
    Prognostic relevance of integrated genetic profiling in acute myeloid leukemia. N Engl J Med 2012;366:1079–89.
    OpenUrlCrossRefPubMed
  21. 21.↵
    1. Ribeiro AF,
    2. Pratcorona M,
    3. Erpelinck-Verschueren C,
    4. Rockova V,
    5. Sanders M,
    6. Abbas S,
    7. et al.
    Mutant DNMT3A: a marker of poor prognosis in acute myeloid leukemia. Blood 2012;119:5824–31.
    OpenUrlAbstract/FREE Full Text
  22. 22.↵
    1. Shen Y,
    2. Zhu YM,
    3. Fan X,
    4. Shi JY,
    5. Wang QR,
    6. Yan XJ,
    7. et al.
    Gene mutation patterns and their prognostic impact in a cohort of 1185 patients with acute myeloid leukemia. Blood 2011;118:5593–603.
    OpenUrlAbstract/FREE Full Text
  23. 23.↵
    1. Walter MJ,
    2. Ding L,
    3. Shen D,
    4. Shao J,
    5. Grillot M,
    6. McLellan M,
    7. et al.
    Recurrent DNMT3A mutations in patients with myelodysplastic syndromes. Leukemia 2011;25:1153–8.
    OpenUrlCrossRefPubMed
  24. 24.↵
    1. Weissmann S,
    2. Alpermann T,
    3. Grossmann V,
    4. Kowarsch A,
    5. Nadarajah N,
    6. Eder C,
    7. et al.
    Landscape of TET2 mutations in acute myeloid leukemia. Leukemia 2012;26:934–42.
    OpenUrlCrossRefPubMed
  25. 25.↵
    1. Langemeijer SM,
    2. Kuiper RP,
    3. Berends M,
    4. Knops R,
    5. Aslanyan MG,
    6. Massop M,
    7. et al.
    Acquired mutations in TET2 are common in myelodysplastic syndromes. Nat Genet 2009;41:838–42.
    OpenUrlCrossRefPubMed
  26. 26.↵
    1. Bejar R,
    2. Stevenson K,
    3. Abdel-Wahab O,
    4. Galili N,
    5. Nilsson B,
    6. Garcia-Manero G,
    7. et al.
    Clinical effect of point mutations in myelodysplastic syndromes. N Engl J Med 2011;364:2496–506.
    OpenUrlCrossRefPubMed
  27. 27.↵
    1. Ko M,
    2. Huang Y,
    3. Jankowska AM,
    4. Pape UJ,
    5. Tahiliani M,
    6. Bandukwala HS,
    7. et al.
    Impaired hydroxylation of 5-methylcytosine in myeloid cancers with mutant TET2. Nature 2010;468:839–43.
    OpenUrlCrossRefPubMed
  28. 28.↵
    1. Tahiliani M,
    2. Koh KP,
    3. Shen Y,
    4. Pastor WA,
    5. Bandukwala H,
    6. Brudno Y,
    7. et al.
    Conversion of 5-methylcytosine to 5-hydroxymethylcytosine in mammalian DNA by MLL partner TET1. Science 2009;324:930–5.
    OpenUrlAbstract/FREE Full Text
  29. 29.↵
    1. Moran-Crusio K,
    2. Reavie L,
    3. Shih A,
    4. Abdel-Wahab O,
    5. Ndiaye-Lobry D,
    6. Lobry C,
    7. et al.
    Tet2 loss leads to increased hematopoietic stem cell self-renewal and myeloid transformation. Cancer Cell 2011;20:11–24.
    OpenUrlCrossRefPubMed
  30. 30.↵
    1. Gaidzik VI,
    2. Paschka P,
    3. Spath D,
    4. Habdank M,
    5. Kohne CH,
    6. Germing U,
    7. et al.
    TET2 mutations in acute myeloid leukemia (AML): results from a comprehensive genetic and clinical analysis of the AML study group. J Clin Oncol 2012;30:1350–7.
    OpenUrlAbstract/FREE Full Text
  31. 31.↵
    1. Kosmider O,
    2. Gelsi-Boyer V,
    3. Cheok M,
    4. Grabar S,
    5. Della-Valle V,
    6. Picard F,
    7. et al.
    TET2 mutation is an independent favorable prognostic factor in myelodysplastic syndromes (MDSs). Blood 2009;114:3285–91.
    OpenUrlAbstract/FREE Full Text
  32. 32.↵
    1. Itzykson R,
    2. Kosmider O,
    3. Cluzeau T,
    4. Mansat-De Mas V,
    5. Dreyfus F,
    6. Beyne-Rauzy O,
    7. et al.
    Impact of TET2 mutations on response rate to azacitidine in myelodysplastic syndromes and low blast count acute myeloid leukemias. Leukemia 2011;25:1147–52.
    OpenUrlCrossRefPubMed
  33. 33.↵
    1. Xu W,
    2. Yang H,
    3. Liu Y,
    4. Yang Y,
    5. Wang P,
    6. Kim SH,
    7. et al.
    Oncometabolite 2-hydroxyglutarate is a competitive inhibitor of alpha-ketoglutarate-dependent dioxygenases. Cancer Cell 2011;19:17–30.
    OpenUrlCrossRefPubMed
  34. 34.↵
    1. Sasaki M,
    2. Knobbe CB,
    3. Munger JC,
    4. Lind EF,
    5. Brenner D,
    6. Brustle A,
    7. et al.
    IDH1(R132H) mutation increases murine haematopoietic progenitors and alters epigenetics. Nature 2012;488:656–9.
    OpenUrlCrossRefPubMed
  35. 35.↵
    1. Marcucci G,
    2. Maharry K,
    3. Wu YZ,
    4. Radmacher MD,
    5. Mrozek K,
    6. Margeson D,
    7. et al.
    IDH1 and IDH2 gene mutations identify novel molecular subsets within de novo cytogenetically normal acute myeloid leukemia: a Cancer and Leukemia Group B study. J Clin Oncol 2010;28:2348–55.
    OpenUrlAbstract/FREE Full Text
  36. 36.↵
    1. Green CL,
    2. Evans CM,
    3. Hills RK,
    4. Burnett AK,
    5. Linch DC,
    6. Gale RE
    . The prognostic significance of IDH1 mutations in younger adult patients with acute myeloid leukemia is dependent on FLT3/ITD status. Blood 2010;116:2779–82.
    OpenUrlAbstract/FREE Full Text
  37. 37.↵
    1. Boissel N,
    2. Nibourel O,
    3. Renneville A,
    4. Gardin C,
    5. Reman O,
    6. Contentin N,
    7. et al.
    Prognostic impact of isocitrate dehydrogenase enzyme isoforms 1 and 2 mutations in acute myeloid leukemia: a study by the Acute Leukemia French Association group. J Clin Oncol 2010;28:3717–23.
    OpenUrlAbstract/FREE Full Text
  38. 38.↵
    1. Paschka P,
    2. Schlenk RF,
    3. Gaidzik VI,
    4. Habdank M,
    5. Kronke J,
    6. Bullinger L,
    7. et al.
    IDH1 and IDH2 mutations are frequent genetic alterations in acute myeloid leukemia and confer adverse prognosis in cytogenetically normal acute myeloid leukemia with NPM1 mutation without FLT3 internal tandem duplication. J Clin Oncol 2010;28:3636–43.
    OpenUrlAbstract/FREE Full Text
  39. 39.↵
    1. Thol F,
    2. Damm F,
    3. Wagner K,
    4. Gohring G,
    5. Schlegelberger B,
    6. Hoelzer D,
    7. et al.
    Prognostic impact of IDH2 mutations in cytogenetically normal acute myeloid leukemia. Blood 2010;116:614–6.
    OpenUrlAbstract/FREE Full Text
  40. 40.↵
    1. Kawahara M,
    2. Pandolfi A,
    3. Bartholdy B,
    4. Barreyro L,
    5. Will B,
    6. Roth M,
    7. et al.
    H2.0-like homeobox regulates early hematopoiesis and promotes acute myeloid leukemia. Cancer Cell 2012;22:194–208.
    OpenUrlCrossRefPubMed
  41. 41.↵
    1. Park UH,
    2. Yoon SK,
    3. Park T,
    4. Kim EJ,
    5. Um SJ
    . Additional sex comb-like (ASXL) proteins 1 and 2 play opposite roles in adipogenesis via reciprocal regulation of peroxisome proliferator-activated receptor {gamma}. J Biol Chem 2011;286:1354–63.
    OpenUrlAbstract/FREE Full Text
  42. 42.↵
    1. Fisher CL,
    2. Pineault N,
    3. Brookes C,
    4. Helgason CD,
    5. Ohta H,
    6. Bodner C,
    7. et al.
    Loss-of-function additional sex combs like 1 mutations disrupt hematopoiesis but do not cause severe myelodysplasia or leukemia. Blood 2010;115:38–46.
    OpenUrlAbstract/FREE Full Text
  43. 43.↵
    1. Abdel-Wahab O,
    2. Adli M,
    3. Lafave LM,
    4. Gao J,
    5. Hricik T,
    6. Shih AH,
    7. et al.
    ASXL1 mutations promote myeloid transformation through loss of PRC2-mediated gene repression. Cancer Cell 2012;22:180–93.
    OpenUrlCrossRefPubMed
  44. 44.↵
    1. Ding L,
    2. Ley TJ,
    3. Larson DE,
    4. Miller CA,
    5. Koboldt DC,
    6. Welch JS,
    7. et al.
    Clonal evolution in relapsed acute myeloid leukaemia revealed by whole-genome sequencing. Nature 2012;481:506–10.
    OpenUrlCrossRefPubMed
  45. 45.↵
    1. Pratcorona M,
    2. Abbas S,
    3. Sanders MA,
    4. Koenders JE,
    5. Kavelaars FG,
    6. Erpelinck-Verschueren CA,
    7. et al.
    Acquired mutations in ASXL1 in acute myeloid leukemia: prevalence and prognostic value. Haematologica 2012;97:388–92.
    OpenUrlAbstract/FREE Full Text
  46. 46.↵
    1. Zhang J,
    2. Ding L,
    3. Holmfeldt L,
    4. Wu G,
    5. Heatley SL,
    6. Payne-Turner D,
    7. et al.
    The genetic basis of early T-cell precursor acute lymphoblastic leukaemia. Nature 2012;481:157–63.
    OpenUrlCrossRefPubMed
  47. 47.↵
    1. Carrington EA,
    2. Jones RS
    . The Drosophila enhancer of zeste gene encodes a chromosomal protein: examination of wild-type and mutant protein distribution. Development 1996;122:4073–83.
    OpenUrlAbstract
  48. 48.↵
    1. Ernst T,
    2. Chase AJ,
    3. Score J,
    4. Hidalgo-Curtis CE,
    5. Bryant C,
    6. Jones AV,
    7. et al.
    Inactivating mutations of the histone methyltransferase gene EZH2 in myeloid disorders. Nat Genet 2010;42:722–6.
    OpenUrlCrossRefPubMed
  49. 49.↵
    1. Nikoloski G,
    2. Langemeijer SM,
    3. Kuiper RP,
    4. Knops R,
    5. Massop M,
    6. Tonnissen ER,
    7. et al.
    Somatic mutations of the histone methyltransferase gene EZH2 in myelodysplastic syndromes. Nat Genet 2010;42:665–7.
    OpenUrlCrossRefPubMed
  50. 50.↵
    1. Simon C,
    2. Chagraoui J,
    3. Krosl J,
    4. Gendron P,
    5. Wilhelm B,
    6. Lemieux S,
    7. et al.
    A key role for EZH2 and associated genes in mouse and human adult T-cell acute leukemia. Genes Dev 2012;26:651–6.
    OpenUrlAbstract/FREE Full Text
  51. 51.↵
    1. Cao R,
    2. Wang H,
    3. He J,
    4. Erdjument-Bromage H,
    5. Tempst P,
    6. Zhang Y
    . Role of hPHF1 in H3K27 methylation and Hox gene silencing. Mol Cell Biol 2008;28:1862–72.
    OpenUrlAbstract/FREE Full Text
  52. 52.↵
    1. Thorsteinsdottir U,
    2. Sauvageau G,
    3. Hough MR,
    4. Dragowska W,
    5. Lansdorp PM,
    6. Lawrence HJ,
    7. et al.
    Overexpression of HOXA10 in murine hematopoietic cells perturbs both myeloid and lymphoid differentiation and leads to acute myeloid leukemia. Mol Cell Biol 1997;17:495–505.
    OpenUrlAbstract/FREE Full Text
  53. 53.↵
    1. Wu X,
    2. Gong Y,
    3. Yue J,
    4. Qiang B,
    5. Yuan J,
    6. Peng X
    . Cooperation between EZH2, NSPc1-mediated histone H2A ubiquitination and Dnmt1 in HOX gene silencing. Nucleic Acids Res 2008;36:3590–9.
    OpenUrlAbstract/FREE Full Text
  54. 54.↵
    1. Tanaka S,
    2. Miyagi S,
    3. Sashida G,
    4. Chiba T,
    5. Yuan J,
    6. Mochizuki-Kashio M,
    7. et al.
    Ezh2 augments leukemogenicity by reinforcing differentiation blockage in acute myeloid leukemia. Blood 2012;120:1107–17.
    OpenUrlAbstract/FREE Full Text
  55. 55.↵
    1. Slany RK
    . The molecular biology of mixed lineage leukemia. Haematologica 2009;94:984–93.
    OpenUrlAbstract/FREE Full Text
  56. 56.↵
    1. Daigle SR,
    2. Olhava EJ,
    3. Therkelsen CA,
    4. Majer CR,
    5. Sneeringer CJ,
    6. Song J,
    7. et al.
    Selective killing of mixed lineage leukemia cells by a potent small-molecule DOT1L inhibitor. Cancer Cell 2011;20:53–65.
    OpenUrlCrossRefPubMed
  57. 57.↵
    1. Yoshida K,
    2. Sanada M,
    3. Shiraishi Y,
    4. Nowak D,
    5. Nagata Y,
    6. Yamamoto R,
    7. et al.
    Frequent pathway mutations of splicing machinery in myelodysplasia. Nature 2011;478:64–9.
    OpenUrlCrossRefPubMed
  58. 58.↵
    1. Singh RK,
    2. Cooper TA
    . Pre-mRNA splicing in disease and therapeutics. Trends Mol Med 2012;18:472–82.
    OpenUrlCrossRefPubMed
  59. 59.↵
    1. Papaemmanuil E,
    2. Cazzola M,
    3. Boultwood J,
    4. Malcovati L,
    5. Vyas P,
    6. Bowen D,
    7. et al.
    Somatic SF3B1 mutation in myelodysplasia with ring sideroblasts. N Engl J Med 2011;365:1384–95.
    OpenUrlCrossRefPubMed
  60. 60.↵
    1. Damm F,
    2. Kosmider O,
    3. Gelsi-Boyer V,
    4. Renneville A,
    5. Carbuccia N,
    6. Hidalgo-Curtis C,
    7. et al.
    Mutations affecting mRNA splicing define distinct clinical phenotypes and correlate with patient outcome in myelodysplastic syndromes. Blood 2012;119:3211–8.
    OpenUrlAbstract/FREE Full Text
  61. 61.↵
    1. Makishima H,
    2. Visconte V,
    3. Sakaguchi H,
    4. Jankowska AM,
    5. Abu Kar S,
    6. Jerez A,
    7. et al.
    Mutations in the spliceosome machinery, a novel and ubiquitous pathway in leukemogenesis. Blood 2012;119:3203–10.
    OpenUrlAbstract/FREE Full Text
  62. 62.↵
    1. Visconte V,
    2. Makishima H,
    3. Maciejewski JP,
    4. Tiu RV
    . Emerging roles of the spliceosomal machinery in myelodysplastic syndromes and other hematological disorders. Leukemia 2012;26:2447–54.
    OpenUrlCrossRefPubMed
  63. 63.↵
    1. Visconte V,
    2. Rogers HJ,
    3. Singh J,
    4. Barnard J,
    5. Bupathi M,
    6. Traina F,
    7. et al.
    SF3B1 haploinsufficiency leads to formation of ring sideroblasts in myelodysplastic syndromes. Blood 2012;120:3173–86.
    OpenUrlAbstract/FREE Full Text
  64. 64.↵
    1. Isono K,
    2. Mizutani-Koseki Y,
    3. Komori T,
    4. Schmidt-Zachmann MS,
    5. Koseki H
    . Mammalian polycomb-mediated repression of Hox genes requires the essential spliceosomal protein Sf3b1. Genes Dev 2005;19:536–41.
    OpenUrlAbstract/FREE Full Text
  65. 65.↵
    1. Thol F,
    2. Kade S,
    3. Schlarmann C,
    4. Loffeld P,
    5. Morgan M,
    6. Krauter J,
    7. et al.
    Frequency and prognostic impact of mutations in SRSF2, U2AF1, and ZRSR2 in patients with myelodysplastic syndromes. Blood 2012;119:3578–84.
    OpenUrlAbstract/FREE Full Text
  66. 66.↵
    1. Xiao R,
    2. Sun Y,
    3. Ding JH,
    4. Lin S,
    5. Rose DW,
    6. Rosenfeld MG,
    7. et al.
    Splicing regulator SC35 is essential for genomic stability and cell proliferation during mammalian organogenesis. Mol Cell Biol 2007;27:5393–402.
    OpenUrlAbstract/FREE Full Text
  67. 67.↵
    1. Tronchere H,
    2. Wang J,
    3. Fu XD
    . A protein related to splicing factor U2AF35 that interacts with U2AF65 and SR proteins in splicing of pre-mRNA. Nature 1997;388:397–400.
    OpenUrlCrossRefPubMed
  68. 68.↵
    1. Shen H,
    2. Zheng X,
    3. Luecke S,
    4. Green MR
    . The U2AF35-related protein Urp contacts the 3′ splice site to promote U12-type intron splicing and the second step of U2-type intron splicing. Genes Dev 2010;24:2389–94.
    OpenUrlAbstract/FREE Full Text
  69. 69.↵
    1. Wu S,
    2. Romfo CM,
    3. Nilsen TW,
    4. Green MR
    . Functional recognition of the 3′ splice site AG by the splicing factor U2AF35. Nature 1999;402:832–5.
    OpenUrlCrossRefPubMed
  70. 70.↵
    1. Abdel-Wahab O,
    2. Levine R
    . The spliceosome as an indicted conspirator in myeloid malignancies. Cancer Cell 2011;20:420–3.
    OpenUrlCrossRefPubMed
  71. 71.↵
    1. Mrozek K
    . Cytogenetic, molecular genetic, and clinical characteristics of acute myeloid leukemia with a complex karyotype. Semin Oncol 2008;35:365–77.
    OpenUrlCrossRefPubMed
  72. 72.↵
    1. Rausch T,
    2. Jones DT,
    3. Zapatka M,
    4. Stutz AM,
    5. Zichner T,
    6. Weischenfeldt J,
    7. et al.
    Genome sequencing of pediatric medulloblastoma links catastrophic DNA rearrangements with TP53 mutations. Cell 2012;148:59–71.
    OpenUrlCrossRefPubMed
  73. 73.↵
    1. Ebert BL,
    2. Pretz J,
    3. Bosco J,
    4. Chang CY,
    5. Tamayo P,
    6. Galili N,
    7. et al.
    Identification of RPS14 as a 5q− syndrome gene by RNA interference screen. Nature 2008;451:335–9.
    OpenUrlCrossRefPubMed
  74. 74.↵
    1. Barlow JL,
    2. Drynan LF,
    3. Hewett DR,
    4. Holmes LR,
    5. Lorenzo-Abalde S,
    6. Lane AL,
    7. et al.
    A p53-dependent mechanism underlies macrocytic anemia in a mouse model of human 5q− syndrome. Nat Med 2010;16:59–66.
    OpenUrlCrossRefPubMed
  75. 75.↵
    1. Dutt S,
    2. Narla A,
    3. Lin K,
    4. Mullally A,
    5. Abayasekara N,
    6. Megerdichian C,
    7. et al.
    Haploinsufficiency for ribosomal protein genes causes selective activation of p53 in human erythroid progenitor cells. Blood 2011;117:2567–76.
    OpenUrlAbstract/FREE Full Text
  76. 76.↵
    1. Walter MJ,
    2. Shen D,
    3. Ding L,
    4. Shao J,
    5. Koboldt DC,
    6. Chen K,
    7. et al.
    Clonal architecture of secondary acute myeloid leukemia. N Engl J Med 2012;366:1090–8.
    OpenUrlCrossRefPubMed
  77. 77.↵
    1. Karni R,
    2. de Stanchina E,
    3. Lowe SW,
    4. Sinha R,
    5. Mu D,
    6. Krainer AR
    . The gene encoding the splicing factor SF2/ASF is a proto-oncogene. Nat Struct Mol Biol 2007;14:185–93.
    OpenUrlCrossRefPubMed
  78. 78.
    1. Abbas S,
    2. Lugthart S,
    3. Kavelaars FG,
    4. Schelen A,
    5. Koenders JE,
    6. Zeilemaker A,
    7. et al.
    Acquired mutations in the genes encoding IDH1 and IDH2 both are recurrent aberrations in acute myeloid leukemia: prevalence and prognostic value. Blood 2010;116:2122–6.
    OpenUrlAbstract/FREE Full Text
Back to top
Molecular Cancer Research: 11 (8)
August 2013
Volume 11, Issue 8

Sign up for alerts

View this article with LENS

Open full page PDF
Article Alerts
Email Article
Citation Tools
The Changing Mutational Landscape of Acute Myeloid Leukemia and Myelodysplastic Syndrome
Connie A. Larsson, Gilbert Cote and Alfonso Quintás-Cardama
Mol Cancer Res August 1 2013 (11) (8) 815-827; DOI: 10.1158/1541-7786.MCR-12-0695
del.icio.us logo Digg logo Reddit logo Twitter logo CiteULike logo Facebook logo Google logo Mendeley logo

Jump to section

Advertisement