Yuka Nakamura

Aberrant expression of MIR9 indicates poor prognosis in acute myeloid leukaemia.

microRNAs (miRNA) play vital roles in each step of normal haematopoiesis starting at the level of haematopoietic stem cell and continuing during the differentiation process of both myeloid and lymphoid lineages (Bhagavathi & Czader, 2010; Havelange & Garzon, 2010). On the other hand, dysregulated expression of miRNA results in the development of haematological malignancies including leukaemia, lymphoma and myeloma. Considering that inactivation of RUNX1 is one of the major roles in the development of acute leukaemia, we screened expression levels of several RUNX1-inhibiting miRNAs in leukaemic bone marrow samples for preliminary study. Among them, only the amount of mature MIR9, transcribed from either MIR9-1, MIR9-2 or MIR9-3 genes, was extraordinarily increased in a certain fraction of patients (data not shown). We thus conducted a further study on mature MIR9 expression by real-time polymerase chain reaction (PCR) methods with a total of 124 bone marrow samples from 101 adult acute myeloid leukaemia (AML) patients at diagnosis and 23 controls, all of whom gave written informed consent for this study. Expression of MIR9 was normalized to the mean expression of three control small nuclear RNAs (SNORD7, RNU6-2 and SNORA74A). The patient cohort consisted of 59 men and 42 women with a median age of 62 years (range; 20–85 years). Eight, 11, 35, 17, 10, 13, 6 and 1 patients were diagnosed as AML-M0, 1, 2, 3, 4, 5, 6 and 7, respectively. Thirty-eight had normal karyotype. Twelve patients were classified as core binding factor (CBF) leukaemia including eight with t (8;21) and four with inv(16)/t(16;16). All of the 17 AMLM3 (acute promyelocytic leukaemia; APL) patients carried t (15;17). Notably, MIR9 expression was hardly detected in normal bone marrow cells. In contrast, some fraction of bone marrow samples from AML patients showed significantly increased levels of MIR9. Thus, we set a cut-off value at the highest level of MIR9 observed in normal bone marrow cells, and divided the AML patients into two groups; MIR9 (+) and MIR9 ( ) groups. Nineteen patients (19%) belonged to the positive group and 82 (81%) to the negative group. Demographic and laboratory characteristics were not significantly different between these two groups. On the other hand, the distribution of French-AmericanBritish subtypes (P = 0·008) and chromosomal findings (P = 0·017) was statistically different between the two groups. No patients in the MIR9 (+) group were diagnosed as AML-M3/t(15;17) or CBF leukaemia/t(8;21) or inv(16)/t (16;16), while three of them had 11q23 abnormalities. We observed FLT3, NPM1 and CEBPA mutations at a frequency of 19%, 8% and 33% in the entire patient cohort. There was no association with aberrant expression of MIR9 and FLT3, NPM1 or CEBPA mutations (P = 0·353, P = 0·641 or P = 0·082). Although patients received various kinds of induction chemo and/or all-trans retinoic acid (ATRA) therapy depending on the subtype of leukaemia and their age, differences in treatment regimens between the MIR9 (+) and ( ) groups in the entire and non-APL cohort were not statistically significant (P = 0·173 and P = 0·699, respectively). Patients with aberrant MIR9 expression had a lower rate of complete remission (44%) compared with those without the expression (59%), but this difference was not statistically significant (P = 0·246). We then compared overall survival (OS) and relapse-free survival (RFS) between the MIR9 (+) and MIR9 ( ) groups by Kaplan–Meier method and LogRank test. The median observation period was 513 d. The MIR9 (+) group was associated with significantly inferior OS (P = 0·005: Fig 1A) and RFS (P < 0·001: Fig 1B) compared with the MIR9 ( ) group. Even when excluding those APL patients who showed a better prognosis with ATRA therapy, the presence of MIR9 expression was significantly associated with unfavourable OS (P = 0·028:Fig 1C) and RFS (P = 0·009: Fig 1D). We performed univariate Cox regression analyses to evaluate the status of aberrant MIR9 expression as well as known prognostic factors as predictors of OS and RFS. As a result, age over 65 years (hazard ratio [HR]=1·992, P=0·035), white blood cell count greater than 50 9 10/l (HR = 2·434, P = 0·011), lactate dehydrogenase (LDH) greater than 600 iu/l (HR = 2·606, P = 0·004), and the presence of aberrant MIR9 expression (HR = 2·602, P = 0·006) were found to be significant predictors for poor OS (data not shown). On the other hand, the presence of aberrant MIR9 expression was the sole predictor of increased risk of relapse in the same group (HR = 3·560, P = 0·001). When the analyses were restricted to non-APL patients, LDH greater than 600 iu/l (HR = 2·088, P = 0·034), the presence of FLT3-internal tandem duplication mutation (HR = 3·397, P = 0·003) and MIR9 expression (HR = 2·153, P = 0·032)

Enhanced Angpt1/Tie2 signaling affects the differentiation and long-term repopulation ability of hematopoietic stem cells

Angiopoietin-1 (Angpt1) signaling via the Tie2 receptor regulates vascular and hematopoietic systems. To investigate the role of Angpt1-Tie2 signaling in hematopoiesis, we prepared conditionally inducible transgenic (Tg) mice expressing a genetically engineered Angpt1, cartridge oligomeric matrix protein (COMP)-Angpt1. The effects of COMP-Angpt1 overexpression in osteoblasts on hematopoiesis were then investigated by crossing COMP-Angpt1 Tg mice with Col1a1-Cre Tg mice. Interestingly, peripheral blood analyses showed that 4 week (wk)-old (but not 8 wk-old) Col1a1-Cre+/COMP-Angpt1+ mice had a lower percentage of circulating B cells and a higher percentage of myeloid cells than Col1a1-Cre-/COMP-Angpt1+ (control) mice. Although there were no significant differences in the immunophenotypic hematopoietic stem and progenitor cell (HSPC) populations between Col1a1-Cre+/COMP-Angpt1+ and control mice, lineage(-)Sca-1(+)c-Kit(+) (LSK) cells isolated from 8 wk-old Col1a1-Cre+/COMP-Angpt1+ mice showed better long-term bone marrow reconstitution ability. These data indicate that Angpt1-Tie2 signaling affects the differentiation capacity of hematopoietic lineages during development and increases the stem cell activity of HSCs.

Overexpression of MIR9 indicates poor prognosis in acute lymphoblastic leukemia

Abstract Aberrant expression of MIR9 predicts a poor prognosis in acute myelogenous leukemia. To evaluate its clinical significance in acute lymphoblastic leukemia, we analyzed expression levels of MIR9 in bone marrow samples from patients with acute lymphoblastic leukemia and compared them to those in normal bone marrow cells. Approximately 20% of them showed higher expression compared with controls. There was a tendency that patients who showed overexpression of MIR9 underwent worse clinical courses, but without statistical significance. However, when the analyses were restricted to patients who did not receive a stem cell transplant, overexpression of MIR9 was significantly associated with worse overall survival. Interestingly, exaggerated MIR9 expression and higher white blood cell count at presentation were independent unfavorable prognostic factors in all patients for overall survival by multivariate analysis. The presence of higher MIR9 expression could be a useful indicator for treatment stratification.

Acute myeloid leukemia with t(7;21)(q11.2;q22) expresses a novel, reversed-sequence RUNX1―DTX2 chimera

The RUNX1 gene is frequently rearranged in acute leukemia. We cloned a novel RUNX1 chimeric gene generated by t(7;21)(q11.2;q22) in a patient with acute myeloid leukemia. 3′-rapid amplification of cDNA ends analysis showed a tail-to-tail fusion between RUNX1 on 21q22 and DTX2 on 7q11.2, with an insertion of short complementary sequence from UPK3B adjacent to DTX2. DTX2 encodes a putative E3-ubiquitin ligase with no known biological function. There are two possible functions of RUNX1-reversed UPK3B–DTX2: one from aberrant RUNX1 chimeric protein and the other from the reversed sequence of DTX2. The predicted aberrant protein expressed under the RUNX1 promoter was highly structurally similar to RUNX1a. In a reporter assay, the aberrant protein inhibited the trans-activation function of RUNX1 in a dominant-negative manner, similar to RUNX1a. In contrast, the DTX2 reversed sequence may degrade wild-type DTX2 transcript or suppress its translation. In conclusion, we identified a novel fusion RUNX1 partner, DTX2, which chimerize in a reverse direction. This is the first example of RUNX1 chimera in an opposing direction generated by chromosomal translocation in leukemia. In addition to the aberrantly truncated RUNX1 protein, the DTX2 antisense sequence may play some role in the development of leukemia carrying the t(7;21) translocation.

RUNX1-EVI1 induces dysplastic hematopoiesis and acute leukemia of the megakaryocytic lineage in mice

The RUNX1-EVI1 gene generated by the t(3;21) translocation encodes a chimeric transcription factor and is a causative gene in the development of de novo acute megakaryoblastic leukemia and leukemic transformation of hematopoietic stem cell tumors. Heterozygous RUNX1-EVI1 knock-in mice die in utero due to hemorrhage in the central nervous system and spinal cord and complete abolishment of definitive hematopoiesis in the fetal liver. On the other hand, the chimeric knock-in mouse develops acute megakaryoblastic leukemia. We created another mouse model of RUNX1-EVI1 using transplantation of retrovirus-infected bone marrow cells. Some mice transplanted with RUNX1-EVI1-expressing bone marrow cells developed acute megakaryoblastic leukemia within eight months, and the other non-leukemic mice showed thrombocytosis at around a year. In the non-leukemic mice, dysplastic megakaryocytes proliferated in the bone marrow and frequently infiltrated into the spleen, which was not associated with marrow fibrosis. In the leukemic mice, their tumor cells were positive for c-kit and CD41, and negative for TER119. Although they were negative for platelet peroxidase in the electron microscopic analysis, they had multiple centrioles in the cytoplasm, which are characteristic of megakaryocytes that undergo endomitosis. The leukemic cells were serially transplantable, and gene-expression analyses using quantitative RT-PCR arrays revealed that they showed significantly elevated expression of stem cell, primitive hematopoietic cell and endothelial cell-related genes compared with normal bone marrow cells. All these data suggested that RUNX1-EVI1 caused dysplastic hematopoiesis or leukemia of the megakaryocytic lineage and endowed gene expression profiles distinctive of immature hematopoietic cells.

P1-255Non-Hodgkin's lymphoma of the breast: a review of 6 cases including 3 patients with a history of autoimmune diseases

Case presentation: A 56-year-old female was received the resection of retroperitoneal leiomyosarcoma in December 2011. In July 2015, CT showed multiple liver metastases, and she received the treatment with ADR as first-line chemotherapy. After two cycles of ADR, her performance status was deteriorative, so she was treated with pazopanib as second-line chemotherapy. The PFS of pazopanib was 5 months and eribulin was 2.5 months. She was then treated with trabectedin as 4th-line chemotherapy from July 2016. Liver metastases was reduced and the disease has been controlled for 7 months from administration of trabectedin. We have still treated with trabectedin.

Expressional changes of genes and miRNA in common megakaryocyte-erythroid progenitors from lower-risk myelodysplastic syndrome

Myelodysplastic syndrome (MDS) is a stem cell tumor characterized by dysplastic features and ineffective hematopoiesis in the early phase and leukemic progression in the late phase. Speculating that differences in the expression of genes and microRNA (miRNA) in control and MDS-derived erythroid progenitors may cause ineffective erythropoiesis, we sorted common megakaryocyte-erythroid progenitors (MEPs) in bone marrow cells from three lower-risk MDS patients, and compared expression levels of genes and miRNA with those from controls. In apoptosis-related pathways, the expression of some pro-apoptotic genes, such as cell death-inducing DFFA-like effector A, caspase 5, and Fas ligand, was elevated in MDS-derived MEPs, while those of anti-apoptotic CD40 and tumor necrosis factor were lower. In hematopoiesis-regulating pathways, RUNX1 and ETV6 genes showed reduced expression. Expression profiling revealed that three and 35 miRNAs were significantly up- and down-regulated in MDS-derived MEPs. MIR9 exhibited robust expression in MEPs and CD71+GlyA+ erythroid cells derived from one of the three patients. Interestingly, overexpression of MIR9 inhibited the accumulation of hemoglobin in UT-7/GM cells. Some of these alterations in gene and miRNA expression may contribute to the pathogenesis of ineffective hematopoiesis in lower-risk MDS and provide molecular markers for sub-classification and making a prognosis.

Fadd and Skp2 are possible downstream targets of RUNX1-EVI1

RUNX1-EVI1 generated by t(3;21) is a causative gene in the leukemic transformation of chronic hematopoietic stem cell tumors. Recruitment of histone deacetylase via carboxyl terminal-binding protein by RUNX1-EVI1 results in transcriptional dysregulation in target genes of wild-type RUNX1, leading to differentiation block and apoptotic prevention of myeloid precursor cells. In the present study using mouse primary hematopoietic cells, we confirmed that RUNX1-EVI1 enhances replating activity of hematopoietic colonies and represses differentiation along the myeloid lineage under treatment with granulocyte colony-stimulating factor. We then observed that these biological effects of RUX1-EVI1 are canceled by the treatment of histone deacetylase inhibitors, trichostatin A and valproic acid. To identify target genes whose expression is suppressed by RUNX1-EVI1, we compared the expression profiles of apoptosis and cell-cycle-related genes in control and RUNX1-EVI1-expressing cells, and in RUNX1-EVI1-expressing cells with and without treatment with histone deacetylase inhibitors. Notably, the expression of three genes, Fadd, Skp2 and CD40lg, was found to be suppressed in RUNX1-EVI1-expressing cells and to be recovered on treatment with histone deacetylase inhibitors. Considering that these genes have some RUNX1-binding sites, they may be direct or indirect targets of RUNX1-EVI1, and changes in their expression may play some role in leukemogenesis by RUNX1-EVI1.