U2AF1 Mutations in Chinese Patients with Acute Myeloid Leukemia and Myelodysplastic Syndrome

Somatic mutations of U2AF1 gene have recently been identified in myelodysplastic syndrome (MDS) and acute myeloid leukemia (AML). In this study, we analyzed the frequency and clinical impact of U2AF1 mutations in a cohort of 452 Chinese patients with myeloid neoplasms. Mutations in U2AF1 were found in 2.5% (7/275) of AML and 6.3% (6/96) of MDS patients, but in none of 81 CML. All mutations were heterozygous missense mutations affecting codon S34 or Q157. There was no significant association of U2AF1 mutation with blood parameters, FAB subtypes, karyotypes and other gene mutations in AML. The overall survival (OS) of AML patients with U2AF1 mutation (median 3 months) was shorter than those without mutation (median 7 months) (P = 0.035). No difference in the OS was observed between MDS patients with and without U2AF1 mutations. Our data show that U2AF1 mutation is a recurrent event at a low frequency in AML and MDS.


Introduction
Acute myeloid leukemia (AML) is characterized by autonomous proliferation and impaired differentiation of hematopoietic progenitors but is a genetically and phenotypically heterogeneous disease. The development of AML is associated with accumulation of acquired genetic alterations and epigenetic changes in hematopoietic progenitor cells that induces normal hematopoietic progenitor cell to loss the ability of self-renewal and differentiation to various mature cell lineages, to transform into a leukemic stem cell, and to accumulate in bone marrow [1,2]. In recent years, an increasing number of gene mutations have been identified involved in the pathogenesis of the disease and have been shown to be correlated with prognosis of AML patients [3]. Some gene mutations have been further introduced into the current World Health Organization (WHO) classification [4]. U2AF1 (U2AF35), an essential component of the U2 small nuclear ribonucleoprotein auxiliary factor (U2AF), plays an important role in the splicing process in which functional mRNA is generated from pre-mRNA [5,6]. The disruption of interactions of several factors involved in the splicing process can cause various types of mutations in an ever increasing number of genes [7]. Recently, somatic mutations in U2AF1 were discovered in myelodysplastic syndrome (MDS) and mainly occurred in two codons (Ser34 and Q157) [8,9]. Patients with U2AF1 mutations had an increased probability of progression from MDS to AML, however, they had similar overall survival as those with wild-type U2AF1 [8]. The prognostic impact and clinical characteristics of patients with U2AF1 mutations in AML are unknown. Thus, in this study we investigated the frequency and prognostic influence of U2AF1 mutations in a cohort of patients with AML, MDS and chronic myeloid leukemia (CML) patients.

Patients' samples
This study was approved by the Ethics Committee Board of Affiliated People's Hospital of Jiangsu University. Bone marrow aspirates or peripheral blood samples of patients with various hematologic malignancies were collected after informed consent written. The patients included 275 primary AML, 96 primary MDS, 81 CML (61 at chronic phase, 4 at accelerated phase, 16 at blast crisis). These hematological malignancies were diagnosed according to the French-American-British Cooperative Group Criteria and the 2008 World Health Organization proposal [4,10]. Karyotypes were classified according to reported previously [11,12]. Bone marrow specimens obtained at the time of complete hematologic remission from three patients (one AML, one RAEB-1 and one RAEB-2) with U2AF1 mutations at initial diagnosis and peripheral blood from 103 healthy individuals were used as control. The mononuclear cells were separated by density-gradient centrifugation using Ficoll. Subsequently, genomic DNAs were extracted using the Genomic DNA Purification Kit (Gentra, USA) according to the manufacturer's instructions.

Mutation scanning
Primers to amplify the coding sequences of U2AF1 (GenBank NC_000021.8) are listed in Table S1. genomic DNA samples were amplified in a final volume of 25 ml containing 16 PCR buffer (Invitrogen, Merelbeke, Belgium), 0.2 mmol/L of each dNTP, 2.5 mmol/L of MgCl 2 , 0.4 mmol/L of both forward and reverse primers, 0.8 mmol/L of oligonucleotide calibrators [13], 16 LCgreen Plus (Idaho Technology Inc. Salt Lake City, Utah), and 1 U Taq polymerase (MBI Fermentas, Canada). All PCR amplicons were generated on a 7300 Thermo cycler (Applied Biosystems, Foster City, CA, USA) using the PCR program as follows: an initial denaturation step was started at 95uC for 5 minutes, followed by 40 cycles of denaturation at 94uC for 30 seconds, annealing at melting temperature for 30 seconds, and an extension at 72uC for 30 seconds. Mutation scanning was performed for PCR products using high-resolution melting analysis (HRMA) with the LightScanner TM platform (Idaho Technology Inc. Salt Lake City, Utah) according to the protocol reported previously [14].

DNA sequencing
To confirm HRMA results, DNA sequencing was also performed in all samples identified by HRMA. PCR products were directly sequenced on both strands using an ABI 3730 automatic sequencer.

Statistics
Statistical analysis was performed using the SPSS 13.0 software package (SPSS, Chicago, IL, USA). Pearson Chi-square analysis and Fisher exact test were carried out to compare the difference of categorical variables between patient groups. Mann-Whitney's Utest was carried out to compare the difference of continuous variables between patient groups. Survival was analyzed according to the Kaplan-Meier method. For all analyses, a P-value of less than 0.05 (two-tailed) was considered statistically significant.

Results and Discussion
HRMA could easily distinguish U2AF1 mutations (S34Y and Q157R) with the sensitivity of 5% in a background of wild-type DNA, higher than that obtained by direct DNA sequencing (10%) (Figures S1, S2, S3 and S4).
In the cohort of 452 patients with myeloid malignancies, a heterozygous U2AF1 mutant was found to in 13 cases with AML or MDS, but in none with CML. The representative results of HRMA and directing sequencing of U2AF1 mutations in AML and MDS were presented in Figures S5, S6, S7 and S8. The clinical characteristics of all patients with U2AF1 mutations were shown in Table 1. U2AF1 mutation, positive in the bone marrow samples from three individuals (case 4,11 and 12, Table 1) at initial diagnosis, disappeared after the first complete remission. Furthermore, U2AF1 mutation was not present in all healthy controls. These results support the somatic nature of U2AF1 mutations.
Follow-up data were obtained for 150 AML patients. There was no significant difference in complete remission rate between patients with and without U2AF1 mutation (57.1% vs 72.9%, P = 0.398).  The median follow-up duration of the patients was 7 months (range, 1-73 months). M3 subtype was excluded from survival analysis due to different therapy and prognosis. The estimated 50% survival time of the remaining 126 patients was 7 months. The overall survival (OS) of AML patients with U2AF1 mutation (median 3 months, 95% confidence interval 0-7.8 months) was shorter than those without mutation (median 7 months, 95% confidence interval 4.8-9.2 months) (P = 0.035, Figure S9). However, there was no difference in disease-free survival between the patients with and without U2AF1 mutation. Two patients with U2AF1 mutation died early after initial diagnosis due to central nervous system involvement and sepsis respectively. Furthermore, a multivariate analysis for outcomes could not be performed because of the small sample size of patients with mutations. A larger cohort from a clinical trial will be needed to definitively address the effect of U2AF1 mutations on outcomes. 6 (6.3%) MDS cases were identified with heterozygous U2AF1 mutations (3 S34F, 2 S34Y, and 1Q157P) ( Table 1). No significance in sex, age, blood parameters, WHO subtypes, and IPSS classification was observed between MDS patients with and without U2AF1 mutations (P.0.05, Table 2). 7.3%-8.8% of U2AF1 mutations have been reported in primary MDS recently [8,9,[19][20][21]. All 72 U2AF1 mutations in MDS including ours, which could be almost found in each FAB or WHO subtype, exclusively occurred at the highly conserved sites of exon S34 and Q157 with a rare exception of A26V, E159, or R156H mutation in three subjects [9,17,18]. Significant association of U2AF1 mutations has not been identified with specific karyotypes. Although Damm et al [21] found that the association of del20q with U2AF1 mutation, no U2AF1 mutation was observed in all four cases with del20q in our group. Survival analysis was performed in 76 MDS cases with follow-up information. No difference in OS was observed between patients with and without U2AF1 mutations (P = 0.821, Figure S9). The impact of U2AF1 mutations on clinical outcome has been controversial in MDS [8,19,20]. Although Makishima et al considered U2AF1 mutation as a factor predictive for shorter survival [19], other three studies did not find the association of U2AF1 mutation with prognosis [8,20,21]. More cases with U2AF1 mutations should be further studied to determine its prognostic relevance.
The definite role of U2AF1 in the cancer pathogenesis has not been known. U2AF1 mutation induces abnormal global RNA splicing which has been described in a wide variety of cancers [22][23][24]. Reduced expression of U2AF1 was found in pancreatic cancer cells and correlated with mis-splicing of the cholecystokinin-B/ gastrin receptor mRNA [25]. Additionally, knockdown of U2AF1 reduces cell proliferation, induces G2/M arrest, and enhances apoptosis [26]. Moreover, a recent study demonstrated that S34F mutant had the same effect as U2AF1 downregulation [9]. These results suggest that U2AF1 mutant leads to loss of function and contributes to ineffective hematopoiesis and the cytopenias seen in MDS. Furthermore, subjects with U2AF1 mutations were not restricted to a particular WHO subtypes, which indicates U2AF1 mutation should be an early, initial genetic event in MDS.
In summary, mutations in U2AF1 occur in patients with AML at a low frequency and are associated with a negative prognosis in AML which will require confirmation in a larger cohort. Figure S1 Results of a dilution series of S34Y U2AF1 mutant in a background of wild-type DNA detected by HRMA.