Abnormal Expression Of WT1-As, MEG3 and ANRIL Long Non-Coding RNAs In Primary Myelofibrosis and Their Clinical Correlates

A growing body of evidence suggests that long non-coding RNAs (lncRNAs) are involved in the regulation of gene expression in normal and cancer cells by recruiting chromatin remodeling complexes to specific genomic loci [Rinn JL, et al. Cell. 2007; Mattick JS, et al. Bioessays 2009]. For this reason, lncRNAs deregulation can play key roles in malignant transformation and cancer cell behavior. [Kogo R, et al. Cancer Res 2011; Yang F, et al. Hepatology 2011]. Despite the increasing number of studies on lncRNAs expression and their involvement in some solid tumors, lncRNAs are far from being extensively characterized in malignant hematopoiesis [Heuston EF, Front Genet 2011]. In particular, lncRNAs expression has never been investigated in cells from primary myelofibrosis (PMF) patients. PMF is a Philadelphia negative chronic Myeloproliferative Neoplasm (MPN) that originates from deregulated clonal proliferation of hematopoietic stem cell associated with overproduction of mature blood cells. Molecular mechanisms underlying MPN pathogenesis were partially unraveled in 2005-2006 with the identification of somatic gain-of-function mutations of JAK2 [James C, et al. Nature 2005; Kralovics R, et al. N Engl J Med 2005; Levine RL, et al. Cancer Cell 2005] and MPL [Pardanani AD, et al. Blood 2006] after which many other mutations were identified. Recently, several new molecular pathogenetic mechanisms were proposed, such as the aberrant expression of coding and non-coding RNAs. After extensive screening of the long non-coding RNA database (lncRNAdb), which is a database providing comprehensive annotations of eukaryotic lncRNAs (http://www.lncrnadb.org) [Amaral PP, Nucleic Acids Res 2011], we identified some previously described lncRNAs as potentially involved in hematological malignancies, such as CDKN2B-antisense (ANRIL), MEG3 and WT1-antisense lncRNAs. With the aim to describe possible abnormalities of lncRNAs in MPN, we investigated the expression of those lncRNAs in CD34+ cells from PMF patients. Since the perturbation of the antisense RNA could alter the expression of the sense gene, we also analyzed the expression of the ANRIL and WT1-as coding genes, called CDKN2B and WT1, to evaluate the correlation of each sense/antisense couple. This study included a total of 26 PMF patients and 16 healthy donors. The diagnosis of PMF was made according to World Health Organization (WHO) criteria. The results evidenced that the majority of PMF samples (74%) displayed a co-upregulation of WT1 and its antisense RNA compared to healthy controls. We found that the large part (61%) of patients harboring concordant upregulation of WT1 sense and antisense belongs to the intermediate-2 and high risk categories of DIPSS-plus scoring system (57.4% and 43.1% in the intermediate-1 and -2, respectively versus 0% in the high-risk category; P = .006) and presents low Hb level, higher percentage of circulating CD34+ cells and a blast count > 1% . Noteworthy, almost 90% of patients with increased expression of WT1 and its antisense showed concurrent upregulation of MEG3 and shared the same clinical variables associated with poor outcome. In fact, also patients harboring higher levels of MEG3 cluster in higher DIPSS-plus scoring system categories, suffer from anemia, have high levels of circulating CD34+ cells and a blast count >1%. These data suggest that not only high levels of WT1-as we previously described [Guglielmelli P, Stem Cells 2007], but also its antisense and MEG3 could identify PMF patients with more severe disease. Finally, we could observe a correlation between CDKN2B upregulation and the grade of BM fibrosis. In fact, the majority of samples having CDKN2B upregulation clustered in categories 2 and 3 (87.3%) as compared to only 4.5% in case of samples showing CDKN2B downregulation (P = .043). Noteworthy, 82% of patients overexpressing CDKN2B were JAK2V617F positive as compared to 22.1% among those with CDKN2B downregulation (P = .004) To our knowledge, this is the first study describing the expression profiles of human lncRNAs in CD34+ cells of PMF patients. Our data demonstrate that WT1-as, MEG3 and ANRIL lncRNAs are aberrantly expressed in PMF CD34+ cells compared to healthy controls. Moreover, different patterns of expression are correlated with some patient\u2019 clinical features. Therefore, our results suggest that a deregulated expression of these lncRNAs could play a role in PMF pathogenesis and progression