Dysregulation of the Let-7/HMGA2 Axis with Methylation of p16 Promoter As a Possible Target of Histone Deacetylase Inhibitor in Myeloproliferative Neoplasms

Myeloproliferative neoplasms (MPNs), including polycythemia vera (PV), essential thrombocythemia (ET) and primary myelofibrosis (PMF), are clonal hematological disorders characterized by proliferation of mature blood cells. Recently, several agents that influence epigenetic modifications, such as histone deacetylase inhibitors (HDACi), as well as JAK2 inhibitors, have been investigated for high-risk MPNs. For example, an HDACi, panobinostat has shown significant efficacy including nearly complete response in PMF (Mascarenhas et al, BJH, 2013), but molecular targets of HDACi remain largely unknown. The High Mobility Group AT-hook 2 (HMGA2) is a non-histone chromatin protein that modulates transcriptions of various genes and contributes to chromatin modification and epigenetic regulation including DNA methylation (Fusco et al, Nat Rev Cancer, 2007; Sun et al, PNAS 2013). Let-7 micro RNAs (miRNAs) negatively regulate expression of HMGA2 through 3’UTR of HMGA2 mRNA, although HMGA2 mRNA consists of both the major variant containing 3’UTR with let-7-specific sites (variant 1) and some minor variants without 3’UTR. We previously reported that overexpression of HMGA2 due to transgenic expression of Hmga2 cDNA without 3’UTR caused proliferative hematopoiesis with providing a clonal advantage to hematopoietic stem cells in mice (Ikeda et al, Blood, 2011). We also showed a deregulation of HMGA2 mRNA expression due to reduced let-7 miRNA level in granulocytes from patients with almost all of PMF and over 20% of PV and ET (Harada-Shirado et al, Blood [Abst], 2013), being associated with splenomegaly, elevated serum LDH values, and methylation of p16 promoter. Thus, we hypothesized that HMGA2 may be a candidate gene as a therapeutic target in MPNs. Since association of HDAC with HMGA2 has been reported in cord blood-derived cells (Lee et al, Cell Mol Life Sci, 2011), we here studied effects of the panobinostat on expressions of HMGA2 and let-7 in HMGA2-expressing myeloid cells including PMF-derived CD34⁺ cells. First, we found significantly higher HMGA2 mRNA levels in CD34⁺ cells from 2 PMF patients compared with CD34⁺ cells from 2 healthy individuals (P<0.001), as well as U937 cells compared with HL60 cells (P<0.001). Thus, we used CD34⁺ cells from one of these 2 PMF patients and U937 cells for further experiments. Interestingly, treatment with panobinostat at the concentration of 40 nM for 8 hours significantly increased expressions of let-7a (P<0.001 and P=0.003, respectively), -7b (P<0.001 and P<0.001), and -7c (P<0.001 and P=0.06) in U937 cells and PMF CD34⁺ cells, compared with samples without the treatment. In contrast, Western blotting showed clearly reduced expression of HMGA2 protein in U937 cells after the treatment with panobinostat. Moreover, we found that variant 1 of HMGA2 mRNA with 3’UTR was significantly reduced by the treatment with panobinostat, compared with samples without the treatment in both U937 cells and PMF CD34⁺ cells (P<0.001 and P<0.001, respectively), while expression levels of variant 2 lacking let-7-specific sites were not changed by the treatment. These findings strongly suggested that panobinostat decreased expression of HMGA2 through 3’UTR of HMGA2 mRNA by increasing expressions of let-7 miRNAs. Of note, we next found much higher expression of variant 1 of HMGA2 mRNA than variant 2 in granulocytes from 15 of 17 (88%) MPN patients whose HMGA2 mRNA levels were higher than controls in our previous study (Blood [Abst], 2013). We also assessed if treatment by panobinostat for the dysregulated let-7/HMGA2 axis may be a therapeutic option for MPNs with respect to DNA methylation. Panobinostat treatment substantially reduced the expressions of DNMT1 and DNMT3a as well as HMGA2 proteins, significantly demethylated the p16 promoter (P<0.001), and decreased survival (P<0.001) in U937 cells. Moreover, knocking-down of HMGA2 with small interfering RNA in U937 cells significantly increased expression of TET3 mRNA and demethylated the p16 promoter, suggesting that HMGA2 expression may contribute to methylation of the p16 promoter. In conclusion, deregulated expression of HMGA2 due to downregulation of let-7 miRNAs, which may lead to some epigenetic modifications such as methylation of the p16 promoter, is a possible therapeutic target of HDACi in MPNs.