アクティブボード・２０１８年　１月
　　　　　・・・・・２０１８年　１月　５日作成・・・・・
研究発表を行った学会；
　　　59th ASH Annual Meeting
　　　2017年12月 9日〜12日（Atlanta, USA）
タイトル；RUNX3 Promotes the Development of MDS/MPN Overlap Syndrome Via Enhancing Expression of Myc in the Absence of Tet2.
発表者；横溝　貴子　氏
　　　（熊本大学　国際先端医学研究機構　指田研究室）
要旨；
The RUNX transcription factors have a critical role in normal and malignant hematopoiesis including myeloproliferative neoplasms (MPN) and myelodysplastic syndromes (MDS). RUNX3 is expressed in various hematopoietic cells, and the reduced expression of RUNX3 due to promoter DNA hyper-methylation is seen in patients with acute lymphoid leukemia, indicating a tumor suppressive role of RUNX3. The loss of Runx3 promotes the development of MPN-like disease in mice. However, loss-of-function mutations and/or a deletion of RUNX3 are rarely observed in patients with myeloid malignancies. In fact, given that enhanced levels of RUNX3 expression are found in a part of patients with AML and MDS, RUNX3 may play an oncogenic role for myeloid malignancies.
To determine how RUNX3 over-expression promotes the development of myeloid malignancies, we generated a mouse model of RUNX3-induced MDS/MPN by utilizing Tet2-deficient hematopoietic stem cells (HSCs). Since TET2 loss-of-function mutations initiate the transformation of HSCs, we harvested bone marrow CD150+CD34-LSK HSCs from tamoxifen-treated Cre-ERT2 (WT) and Cre-ERT2;Tet2Δ/Δ (Tet2 KO) mice, and transduced expression of 3xFLAG-tagged RUNX3 by using a retrovirus vector in vitro. Then, we transplanted them into lethally-irradiated Ly5.1+ recipient mice together with wild-type competitor cells. We observed that Tet2 KO mice died dominantly due to MPN-like disease, whereas RUNX3-expressing Tet2 wild-type (RUNX3/WT) HSCs neither competed wild-type bone marrow cells nor developed hematological malignancies in mice by 12 months post transplantation. In contrast, RUNX3-expressing Tet2 KO (RUNX3/Tet2 KO) HSCs preferentially generated myeloid cells, but not lymphoid cells, and subsequently competed wild-type cells in the PB and the BM. We found that 7 out of 18 RUNX3/Tet2 KO mice developed lethal MDS/MPN diseases despite a longer latency. At the time of sacrifice, RUNX3-Tet2 KO mice showed severe pancytopenia accompanied with dyplastic blood cells including pseudo-Pelger neutrophils and micro megakaryocytes. In addition, these moribund mice significantly enlarged spleen due to the expansion of myeloid cells. FACS analysis demonstrated that the expansion of LSK hematopoietic stem/progenitor cells (HSPCs) in the BM and the spleen, consistent with the features of extramedullary hematopoiesis. Thus, RUNX3/Tet2 KO mice recapitulated the phenotypic features of human MDS/MPN overlap syndrome, indicating that RUNX3 over-expression is prerequisite to the development of MDS/MPN in the absence of Tet2, although it has an adverse effect on the function of Tet2-present HSCs, reminiscent with a tumor suppressive function of RUNX3 as seen in Runx3 KO mice.
To understand the underlying molecular mechanism in the pathogenesis of RUNX3-induced MDS/MPN, we performed gene expression analysis in HSPCs isolated from WT, RUNX3/WT, Tet2 KO and RUNX3/Tet2 KO mice at pre-disease and MDS/MPN stages. Gene Set Enrichment Analysis (GSEA) demonstrated that RUNX3/Tet2 KO MDS/MPN cells significantly enhanced expression of c-Myc target genes, compared to either WT cells or RUNX3/Tet2 KO cells at the pre-disease stage. While RUNX3/Tet2 KO cells showed negative enrichments of differentiation-related pathway genes, compared to WT cells, we found that those MDS/MPN cells enhanced expression of cell cycle, ribosome, and mitochondria-related biological pathways, in which c-Myc may regulate target genes and contribute to the transformation. In order to determine whether RUNX3 directly activates expression of c-Myc, we performed ChIP-sequencing by using an anti-FLAG antibody in RUNX3/Tet2 KO bone marrow cells, and found that c-Myc 3’ distal enhancer regions showed conserved binding sites for Runx transcription factors. In addition, we found that RUNX3-expressing human AML cells showed higher sensitivity to JQ1, a BRD4 inhibitor, following the significant reduction of both RUNX3 and C-MYC expression. The growth inhibition by JQ1 was rescued by ectopic expression of RUNX3.
In conclusion, RUNX3 and c-Myc collaborated to activate the function of oncogenic enhancers, in which Tet2 deficiency may alter the epigenetic status for allowing the bindings of these transcription factors, to an extent not seen in RUNX3-expressing alone. We demonstrated that RUNX3 functions as either an oncogene or a tumor suppressor for myeloid malignancies in the context dependent manner.