Up-regulation of MicroRNA 146b is Associated with Myelofibrosis in Myeloproliferative Neoplasms

Transcription

1 308 Available online at Up-regulation of MicroRNA 146b is Associated with Myelofibrosis in Myeloproliferative Neoplasms Jung-Sook Ha 1 and Hye-Ra Jung 2 1 Department of Laboratory Medicine, Keimyung University School of Medicine, Daegu and 2 Department of Pathology, Keimyung University School of Medicine, Daegu, Korea Abstract. In this study, our goal was to evaluate whether the expressions of microrna (mir)-150, mir- 146b, mir-31 and mir-95 demonstrate primary myelofibrosis (PMF) specificity, associations with fibrosis grade, hematologic phenotypes, or myeloproliferative neoplasm (MPN)-associated mutations. A total of 51 formalin-fixed and paraffin-embedded bone marrow MPN samples, including 15 polycythemia vera (PV), 26 essential thrombocythemia (ET), and 10 PMF, and 24 normal controls were included. The expression of microrna (mirna) was detected by quantitative real-time polymerase chain reaction using mirna specific primers. RNU6-2 was analyzed for all samples as endogenous control for relative quantification. Information for fibrosis, hematologic parameters, Janus kinase 2 (JAK2) V617F, and calreticulin (CALR) mutations was obtained from medical records. Significant increment of mir-146b was detected in PMF compared to normal controls (P=0.008). Moreover, expression of mir-146b tended to increase according to increment of fibrosis grade, and patients with myelofibrosis (MF) grade 3 showed significantly higher expression than patients with MF 0 to 2 (P=0.022, and 0.013, respectively) or normal controls (P<0.001). The expression of mir-31 also showed tendency to increase following fibrosis and mir-150 showed up-regulated expression in ET (P=0.015) compared to normal control. There was no relationship between mirna expression and hematologic indices except mir-95 showed negative correlation with platelet count (P=0.024). There was no significant correlation between mirna expression and JAK2 V617F or CALR mutation. Up-regulation of mir-146b could be used as a fibrosis-indicating marker and might be helpful in the study of fibrotic mechanism in MPN, as well as other fibrotic diseases. Key words: micrornas, mir-146b, primary myelofibrosis, myeloproliferative neoplasm Introduction Chronic myeloproliferative neoplasms (MPNs), including polycythemia vera (PV), essential thrombocythemia (ET), and primary myelofibrosis (PMF) are clonal stem cell disorders characterized by dysregulated stem cell expansion and over-production of red cells, white cells, and platelets alone or in combination. It also has a tendency to elicit extrameduallary hematopoiesis, complications of thrombosis and/or hemorrhage, and transformation to acute leukemia or myelofibrosis at variable rates [1]. Although somatic mutations, such as Janus kinase 2 (JAK2) V617F, thrombopoietin receptor (MPL, myelorproliferative leukemia virus oncogene) W515, and calreticulin (CALR) mutations have been discovered as major molecular Address correspondence to Jung-Sook Ha, Department of Laboratory Medicine, Keimyung University School of Medicine, 56 Dalseongro, Jung-gu, Daegu, Republic of Korea, ; phone: ; fax: ; e mail: ksksmom@naver.com and ksksmom@dsmc.or.kr markers, they are insufficient to describe significant heterogeneity in disease phenotype [2]. Therefore, the existence of other unknown genetic or epigenetic factors remains under investigation. Recently, microrna (mirna) has been identified as a novel epigenetic regulator of gene expression [3]. MiRNAs are small nucleotide, singlestranded, non-protein-coding RNAs which are phylogenetically conserved during evolution. Through sequence-specific base pairing with target messenger RNAs (mrna), mirnas regulate specific gene expression, contributing to a vast number of biological processes, including cell growth, differentiation, apoptosis, and potentially, the pathogenesis of malignant neoplasia [4]. Until now, many hematologic malignancies including MPN demonstrated specific mirna expression profiles and some of them obscured the aberrant expression of specific mirnas in primary patient /15/ by the Association of Clinical Scientists, Inc.

2 upregulation of mir-146b in myelofibrosis 309 Figure 1. (A-E). The melting cures of real-time polymerase chain reaction of RNU6-2 (A), mir-31 (B), mir-95 (C), mir-146b (D), and mir-150 (E). Melting curves for each microrna are showing clear one peak without non-specific amplification. Table 1. Clinicohematologic data of MPN patients PV (n=15) ET (n=26) PMF (n=10) Total (n=51) Age 66.6 ± ± ± ± 12.2 Men : women 7 : 8 16 : 10 8 : 2 31 : 20 WBC (x10 3 /µl) 14.0 ± ± ± ± 13.9 Hb (g/dl) 19.0 ± ± ± ± 4.5 Platelet (x10 3 /µl ) ± ± ± ± JAK2 V617F, n(%) 15(100.0) 20(76.9) 3(30.0) 38(74.5) CALR, n(%) 0(0.0) 4(15.4) 1(10.0) 5(9.8) Abbreviations: MPN, myeloproliferative neoplasm; PV, polycythemia vera; ET, essential thrombocythemia; PMF, primary myeloproliferative neoplasm; WBC, white blood cell; Hb, hemoglobin; JAK2, Janus kinase 2; CALR, calreticulin. Age and hematologic data is described as mean ± standard deviation. samples or mouse models [3,5-7]. Some of those mirnas showed distinct associations with specific subtype of MPNs, especially PMF [8-10]. PMF is characterized by extensive proliferation of abnormal megakaryocytes, which accumulate in the bone marrow (BM), resulting in the development of BM fibrosis and eventually, osteosclerosis. In this study, we analyzed the expression of selected mirnas (mir-150, mir-146b, mir-31 and mir- 95-2), whose deregulation was reported in PMF in previous studies, to validate whether their expressions demonstrate PMF specificity and show associations with fibrosis grade. Additionally, the possible associations with hematologic phenotypes or MPN-associated mutations were analyzed. Materials and Methods Patient samples. Formalin-fixed and paraffin-embedded (FFPE) bone marrow trephine biopsy samples obtained from MPN patients from were used. All samples were taken from posterior iliac crest for routine clinical care at first diagnosis before treatment and those with sufficient starting material for investigations were included in this study. For this study, morphologic review and diagnostic revision according to 2008 World Health Organization (WHO) classification criteria was performed and finally 15 PV, 26 ET, and 10 PMF patients were included. The patients consisted of 31 men and 20 women (average 65.7±12.2 years). For normal controls, 24 FFPE bone marrow samples (men:women 14:10, average age 62.2±14.5 years), which were collected for staging but diagnosed as normal without any abnormal findings, were selected. This study was approved by Institutional Review Board of Dongsan Medical Center (No. IRB ), and informed consent was obtained from subjects at the time of sample collection. Additional information for JAK2 V617F, CALR mutations, hematologic parameters including hemoglobin (Hb), white blood cell (WBC) count and platelet count, and fibrosis (grade MF 0 to 3) [11] were obtained from medical records (Table 1). RNA extraction, cdna synthesis and relative quantitative RT-PCR. Total RNA extraction from FFPE BM samples was performed using RNeasy FFPE kits (Qiagen, Valencia, CA, USA) according to the manufacturer s instructions. A total of 12 µl of extracted RNA was reverse-transcribed to synthesize complementary DNA

3 310 Figure 2. Expression level of mir-31(a), mir-95(b), mir-146b(c), and mir-150(d) according to MPN subtypes and control. MiR-150 showed slightly up-regulated expression in ET patients (P=0.015) compared to normal control. MiR- 146b displayed significantly up-regulated expression in PMF (P=0.008) compared to normal controls, but not among MPN subtype. The expression of micrornas were described as ΔCT values (ΔCT: Ct of each mirna Ct of endogenous control). Abbreviations: CT, cycle threshold; PV, polycythemia vera; ET, essential thrombocythemia; PMF, primary myelofibrosis; mir, microrna. using miscript II RT kit (Qiagen, Valencia, CA, USA). Quantitative real-time PCR was performed using miscript SYBR Green PCR kit (Qiagen, Valencia, CA, USA) in ABI real-time PCR machine to detect expression of each mirnas. At the end of the PCR cycles, melting curve analyses were performed in order to validate the specific amplification of the PCR product (Figure 1). MiRNA specific primers used for real-time PCR were purchased from Qiagen (miscript primer assay for mir- 150, mir-146b, mir-31 and mir-95, Qiagen, Valencia, CA, USA). RNU6-2 was analyzed in all samples as endogenous control. All reactions were performed in duplicate. Statistical analysis. Expression data was normalized for the expression of RNU6-2 and analyzed for the relative expression. In order to determine the differences of mir- NA expression between MPN subtypes and control, Mann-Whitney (for nonparametric analysis) or independent-t tests (for parametric analysis) were used. The expression differences of each mirna according to fibrosis grade or the harboring of JAK2 V617F or CALR mutation were also evaluated by either Mann-Whitney or independent-t tests. The relations of expression of mirnas and hematologic indices were analyzed using Pearson s correlations. All analyses were two-sided and P-values less than 0.05 were considered significant. SPSS 20.0 (SPSS Inc., Chicago, IL, USA) statistical software was used for all calculations. Results MiR-31, mir-95, mir-146b and mir-150 were constitutively expressed in FFPE samples of both MPN patients and normal controls (Figure 1). When expressions of mirnas were compared between MPN subtypes and normal controls,

4 upregulation of mir-146b in myelofibrosis 311 Figure 3. Expression of mir-31(a), mir-95(b), mir-146b(c), and mir-150(d) according to fibrosis grade (MF 0 to MF 3) and control. MiR-146b showed significantly higher expression in patients with MF 4 grade compared to normal controls (P<0.001), or patients with MF 0 to MF 2 (P=0.022, P=0.001 and P=0.013, respectively). Expression of mir-31 tended to increase according to fibrosis level, but statistical significance was detected only between moderate fibrosis and normal control (P=0.044). Abbreviations: CT, cycle threshold; mir, microrna; MF, myelofibrosis. mir-150 showed slightly up-regulated expression in ET patients (P=0.015), but not in PV or PMF specimens (P=0.386 and 0.371, respectively). MiR- 146b displayed significantly up-regulated expression in PMF (P=0.008) compared to controls, but not among MPN subtypes. MiR-31 and mir-95 expression in MPN subtypes did not show significant differences compared to controls (Figure 2). When compared mirnas expression according to fibrosis, expression of mir-146b was likely to increase according to the increment of fibrosis grade and showed significantly higher expression in patients with MF 3 than normal controls (P<0.001), or than cases with milder fibrosis MF 0 to 2 (P=0.022, and 0.013, respectively) (Figure 3). When analyzed for each PV and ET MPN subtypes, mir-146b also tend to increase according to increment of fibrosis grade and patients with MF 3 showed statistically significant difference compared to milder fibrosis patients (Figure 4). Expression of mir-31 tended to increase according to the increment of fibrosis, but statistical significance was detected only between MF 2 and normal control (P=0.044). MiR-95 and mir-150 did not show aberrant expression patterns compared to normal controls (Figure 3). In analyzing the correlations between the expression of mirnas and the hematologic indices (WBC count, Hb and platelet count), no significant correlations were demonstrated in MPN patients. However, only mir-95 showed slightly negative correlation with platelet count (P=0.024) (Figure 5).

5 312 Figure 4. Expression of mir-146b in PV(A) and ET(B) patients. The expression of mir-146b showed increased pattern according to the increment of myelofibrosis grade and patients with MF 3 showed significantly higher expression in both PV and ET. Abbreviations: CT, cycle threshold; mir, microrna; MF, myelofibrosis. deregulations were due to fibrosis (the most prominent finding of PMF). We found mir-146b to be up-regulated in PMF in additional association with fibrosis grade regardless of MPN subtypes. Figure 5. The expression of mir-95 showed negative relationship with platelet count in MPN patients (P=0.024). Abbreviations: CT, cycle threshold; mir, microrna. When the expression of mirnas was compared between patients with JAK2 V617F and those without JAK2 V617F, or between patients with CALR mutation and those without CALR, there were no significant differences (Figure 6). Discussion Although a number of studies have now reported deregulated mirnas in MPNs [3,7-10], most of those mirnas were not extensively validated. We focused on a few mirnas that had been reported to show PMF-specific deregulation to assess if those Previously, Hussein et al. demonstrated increased accumulation of mir-146b in megakaryocytes laser-dissected from BM trephine biopsies of PMF patients [10]. They reported that expression of mir-146b tended to increase according to increased megakaryopoiesis. However, they did not observe distinct relationships of mir146b with BM fibrotic changes. In this study, although we analyzed for mixed cell lineages rather than separated megakaryocytes, we also observed significantly higher expression of mir-146b in PMF, compared to normal controls or ET/PV, and found a trend of increasing mir-146b expression in association with advanced fibrosis regardless MPN subtype. Recently, other studies also reported the relationship of mir-146b with fibrosis based on up-regulated mir-146b in PS-induced hepatic fibrosis or angiotensin II-treated cardiac fibroblasts [12,13]. On the other hand, malignancy-related changes of mir-146b have also been reported, such as up-regulation in papillary thyroid cancer and down-regulation in chronic lymphocytic leukemia [14,15]. Because most of the PMF in this study had severe fibrotic changes, it is not clear whether the higher mir-146b expression in PMF was a malignancyrelated finding or fibrosis-associated changes. However, distinctive up-regulation of mir146b in

6 upregulation of mir-146b in myelofibrosis 313 Figure 6. The expression of mir-31, mir-95, mir-146b and mir-150 did not show significant differences according to harboring of JAK2 V617F or CALR mutation. MF 3 grade regardless of MPN subtypes in this study would suggest that the expression of mir- 146 might be fibrosis-related changes rather than malignancy-associated changes, and suggest that mir-146b might play a role on BM fibrosis development or at least secondary phenomenon following BM fibrosis in MPN. MiR-150 is known to be expressed in megakaryocyte-erythrocyte progenitor (MEP) cells and its overexpression commits MEPs toward megakaryocyte in normal hematopoiesis [16]. However, mir- 150 has been reported wit as variable functions in pathologic conditions. In MPNs, significant downregulation of mir-150 in PV or PMF patients, compared to normal controls, was reported in previous investigations [8,9]. Indeed, mir-150 expression positively correlated with the allelic burden of JAK2 617F in PV and PMF, and negatively correlated with platelet count, and positively correlated with peripheral WBC count [8]. MiR-150 was also suggested as a tumor suppressor in lymphoma patients [17]. In lupus nephritis, mir-150 over-expression was observed according to increasing chronicity index, including fibrosis, and it was demonstrated that mir-150 promotes renal fibrosis through down-regulation of SOCS1 [18]. In this study, we observed up-regulation of mir-150 in ET, compared to normal controls, which differ from previously published studies. Furthermore, we could not observe the relationship of mir-150 with JAK2 V617F, Hb, WBC, and platelet counts, or fibrosis grade. These controversial results for mir-150 would need to be studied for more MPN cases. Guglielmelli et al. reported down-regulated mir- 31 and mir-95 in PMF compared to controls or PV/ET [9]. Unfortunately, there had been no more significant reports for MPN to the best of knowledge. However, there have been variable reports for other diseases such as up-regulation of mir-31 in colorectal cancer [19], down-regulation of mir-31 in pulmonary fibrosis [20], or up-regulated patterns of mir-95 in lung cancer and pancreatic cancer [21,22]. In this study, we could not observe any significant differences of mir-95 expression according to MPN subtypes, fibrosis level, hematologic indices, or JAK2/CALR mutations except negative relation with platelet count. For mir-31, there was a light tendency to increase expression according to increment of fibrosis grade, but it was not proved statistically. These variable or conflicting results for expression of mir-150, 31, and 95 might be due to a small number of studies or differing sources of mirna. In previous studies, there had been variable cellular sources of mirnas such as granulocytes, CD34+ cells, platelets, or megakaryocytes from FFPE samples. Herein, we used FFPE for mirna extraction, which are rarely used for mirna studies due to unstable RNA stability. However, a previous study

7 314 demonstrated that mirnas from FFPE samples had similar expression profiles with those from fresh tissues, and they demonstrated that mirnas from FFPE tissues are suitable for mirna expression analyses [23]. However, because mirna expression could differ according to cell type, cellspecific expression profiles might be more informative in investigation than mixed cell lineages. Despite the potential limitations from this study, we found a distinct relationship between mir-146b expression and fibrosis, showing a tendency to increase according to fibrosis grade, regardless of MPN subtype. Thus, up-regulation of mir-146b could be used as a fibrotic indicator and be helpful in study of fibrotic mechanism in PMF, as well as other fibrotic diseases. Moreover, studies with larger sample sizes are needed to identify the specific functions of other mirnas in MPNs. Acknowledgment We thank KS Park, KS Lee and TY Park Ph.D. for good assistance with sample collection, DNA extraction and real-time analysis. This research was supported by the Bisa Research Grant of Keimyung University in References 1. Tefferi A, Thiele J, Vardiman JW. The 2008 World Health Organization classification system for myeloproliferative neoplasms: order out of chaos. Cancer 2009;115: Klampfl T, Gisslinger H, Harutyunyan AS, Nivarthi H, Rumi E, Milosevic JD, Them NC, Berg T, Gisslinger B, Pietra D, Chen D, Vladimer GI, Bagienski K, Milanesi C, Casetti IC, Sant'Antonio E, Ferretti V, Elena C, Schischlik F, Cleary C, Six M, Schalling M, Schonegger A, Bock C, Malcovati L, Pascutto C, Superti-Furga G, Cazzola M, Kralovics R. Somatic mutations of calreticulin in myeloproliferative neoplasms. N Engl J Med 2013;369: Lawrie CH. MicroRNAs in hematological malignancies. Blood Rev 2013;27: Ambros V. The evolution of our thinking about micrornas. Nat Med 2008;14: Omer A, Yadav NK, Singh P, Singh RK. Hematological malignancies: role of mirnas and their in silico aspects. Expert Rev Anticancer Ther 2013;13: Joshi D, Gosh K, Vundinti BR. MicroRNAs in hematological malignancies: a novel approach to targeted therapy. Hematology 2012;17: Zhan H, Cardozo C, Raza A. MicroRNAs in myeloproliferative neoplasms. Br J Haematol 2013;161: Gebauer N, Bernard V, Gebauer W, Feller AC, Merz H. MicroRNA expression and JAK2 allele burden in bone marrow trephine biopsies of polycythemia vera, essential thrombocythemia and early primary myelofibrosis. Acta Haematol 2013;129: Guglielmelli P, Tozzi L, Pancrazzi A, Bogani C, Antonioli E, Ponziani V, Poli G, Zini R, Ferrari S, Manfredini R, Bosi A, Vannucchi AM, Consortium MPDR. MicroRNA expression profile in granulocytes from primary myelofibrosis patients. Exp Hematol 2007;35: Hussein K, Theophile K, Dralle W, Wiese B, Kreipe H, Bock O. MicroRNA expression profiling of megakaryocytes in primary myelofibrosis and essential thrombocythemia. Platelets 2009;20: Barosi G, Bordessoule D, Briere J, Cervantes F, Demory JL, Dupriez B, Gisslinger H, Griesshammer M, Hasselbalch H, Kusec R, Le Bousse-Kerdiles MC, Liberato NL, Marchetti M, Reilly JT, Thiele J, European Myelofibrosis N. Response criteria for myelofibrosis with myeloid metaplasia: results of an initiative of the European Myelofibrosis Network (EUMNET). Blood 2005;106: Ge S, Wang X, Xie J, Yi X, Liu F. Deep sequencing analysis of microrna expression in porcine serum-induced hepatic fibrosis rats. Ann Hepatol 2014;13: Jiang X, Ning Q, Wang J. Angiotensin II induced differentially expressed micrornas in adult rat cardiac fibroblasts. J Physiol Sci 2013;63: Peng Y, Li C, Luo DC, Ding JW, Zhang W, Pan G. Expression profile and clinical significance of micrornas in papillary thyroid carcinoma. Molecules 2014;19: Wu PY, Zhang XD, Zhu J, Guo XY, Wang JF. Low expression of microrna-146b-5p and microrna-320d predicts poor outcome of large B-cell lymphoma treated with cyclophosphamide, doxorubicin, vincristine, and prednisone. Hum Pathol 2014;45: Lu J, Guo S, Ebert BL, Zhang H, Peng X, Bosco J, Pretz J, Schlanger R, Wang JY, Mak RH, Dombkowski DM, Preffer FI, Scadden DT, Golub TR. MicroRNA-mediated control of cell fate in megakaryocyte-erythrocyte progenitors. Dev Cell 2008;14: Watanabe A, Tagawa H, Yamashita J, Teshima K, Nara M, Iwamoto K, Kume M, Kameoka Y, Takahashi N, Nakagawa T, Shimizu N, Sawada K. The role of microrna-150 as a tumor suppressor in malignant lymphoma. Leukemia 2011;25: Zhou H, Hasni SA, Perez P, Tandon M, Jang SI, Zheng C, Kopp JB, Austin H, 3rd, Balow JE, Alevizos I, Illei GG. mir- 150 promotes renal fibrosis in lupus nephritis by downregulating SOCS1. J Am Soc Nephrol 2013;24: Chen T, Yao LQ, Shi Q, Ren Z, Ye LC, Xu JM, Zhou PH, Zhong YS. MicroRNA-31 contributes to colorectal cancer development by targeting factor inhibiting HIF-1alpha (FIH-1). Cancer Biol Ther 2014;15: Yang S, Xie N, Cui H, Banerjee S, Abraham E, Thannickal VJ, Liu G. mir-31 is a negative regulator of fibrogenesis and pulmonary fibrosis. FASEB J 2012;26: Chen X, Chen S, Hang W, Huang H, Ma H. MiR-95 induces proliferation and chemo- or radioresistance through directly targeting sorting nexin1 (SNX1) in non-small cell lung cancer. Biomed Pharmacother 2014;68: Zhang Y, Li M, Wang H, Fisher WE, Lin PH, Yao Q, Chen C. Profiling of 95 micrornas in pancreatic cancer cell lines and surgical specimens by real-time PCR analysis. World J Surg 2009;33: Liu A, Xu X. MicroRNA isolation from formalin-fixed, paraffin-embedded tissues. Methods Mol Biol 2011;724: