國立臺灣大學醫學院醫學檢驗暨生物技術學系 博士論文
Department of Clinical Laboratory Sciences and Medical Biotechnology College of Medicine
National Taiwan University Doctoral Dissertation
探討兒童急性淋巴性白血病相關之微核醣核酸-181A 及微核醣核酸-151 的功能
Investigating the function of childhood acute lymphoblastic leukemia associated microRNAs: miR-181A and miR-151
顏靜慈 Ching-Tzu Yen
指導教授：林淑華 博士 Advisor: Shu-Wha Lin, Ph.D.
中華民國一百零四年十一月
November 2015
I
前 言
微核醣核酸為演化上高度保留的內生性非編碼小片段 RNA，具影響各種生理功能 的能力。除了調控個體發育及細胞功能，亦涉及癌症的病理機制。急性淋巴性白 血病為最常見的兒童癌症，先前與臺大基因體中心基因微陣列及晶片核心實驗室 合作分析兒童急性淋巴性白血病患者的微核醣核酸表現，找到了數個與急性淋巴 性白血病次分群相關的微核醣核酸。本論文主要探討兩個急性淋巴性白血病相關 之微核醣核酸基因：MIR181A1 及 MIR151A。第一部分著重於探討和 t(12;21)染色 體 轉 位 相 關 RNA― MIR181A1 。 t(12;21) 為 最 常 見 的 染 色 體 異 常 且 會 形 成 ETV6/RUNX1 融合致癌基因。本研究發現 MIR181A1 及 MIR151A 可相互調控，構
成 一 個 特 殊 的 雙 向 負 調 控 機 制 ， 並 發 現 過 量 表 現 微 核 醣 核 酸 181a 可 促 進 ETV6/RUNX1 陽性白血病細胞的分化。第二部分則著重於探討 MIR151A1 基因的功
能，此微核醣核酸在前 B 細胞急性淋巴性白血病中表現量遠高於 T 細胞急性淋巴 性白血病。在本研究中，我們藉由基因重組工程產製 Mir151 基因剔除小鼠，並鑑 定其表現型。
II
Preface
MicroRNAs (miRNAs) are endogenous noncoding small RNAs, which are highly
conserved during biological evolution and implicated in virtually all aspects of biology.
In addition to normal development and cellular function, they also involved in the
pathogenesis of many cancers. Acute lymphoblastic leukemia (ALL) is a special type of
cancer developed mostly in children. In collaboration with the NTU Microarray Core
facility we have applied miRNA profiling to childhood ALL patients and identified
several miRNAs correlated with various kinds of ALL subtypes. The thesis describes
two ALL-associated miRNA genes, MIR181A1 and MIR151A. The first part of the
thesis focuses on t(12;21)-positive ALL associated miRNA―MIR181A1. t(12;21) is the
most common chromosomal alteration which results in expression of ETV6/RUNX1
fusion oncogene. We demonstrated novel regulatory network comprising ETV6/RUNX1
and MIR181A1 in which ETV6/RUNX1 and MIR181A1 can regulate each other. We
further demonstrate that ectopic expression of miR-181a partially reversed the blockade
of B cell differentiation in ETV6/RUNX1-expressing leukemic cells. The second part of
the thesis focuses on MIR151A, which was identified to be differentially expressed in
B-ALL. We generated the Mir151 conventional knockout mice using recombineering
technique and characterized their phenotypes.
III
誌謝
每篇被完成的論文，不光是作者的心血結晶，還隱含著許多人的貢獻，今日 有此小小成果，心中由衷感激所有曾幫助我的人。
感謝給予我契機，讓我有幸踏入小兒血液腫瘤領域的林淑華老師及林東燦醫 師，並且感謝林淑華老師及林淑容老師的細心教導，充分滋養了我的學術基礎及 關於基因剔除小鼠的諸般知識與經驗，在撰寫期刊論文的同時，也習得了諸多寶 貴的經驗與技巧，讓我得以茁壯成長。謝謝所有擔任口試委員的老師們，俞松良 老師的精闢見解、楊性芳老師及曾慶平老師循循善導我的思路、以及陳淑惠醫師 提供的臨床意見，使得我的論文更臻完整。還要感謝 ALL 團隊的所有成員，特別 是建立了整個團隊研究基礎的楊永立醫師，耐心地教我 flow 分析方法及技巧的英 卉學姐，還有不吝分享實驗經驗的勝凱，都給予我莫大的幫助。另外還要感謝中 原大學基因功能研究室的諸位，在 mir-181a-1 研究上擔任開路先鋒的瑋柔、雖然 照顧老鼠並不輕鬆，但有認真又窩心的曉頴和瑀絜分擔雜務，著實幫了不少忙；
謝謝台大林淑華老師實驗室的同仁們，一直以來很慶幸有玉真學姐、軍宇學長及 明憲在實驗及待人處事方面的經驗分享，另外要特別感謝夢倪體貼我趕論文時無 暇顧及老鼠與實驗進度的龐大壓力，以及士鋒成功救援我那份差點毀損的論文電 子檔。
我要將此份小小成就歸功於我的家人，我最親愛的爸爸媽媽和姊姊們，無條 件支持任性的我念博士班的決定，即使我當了這麼多年的學生無法為家裡付出甚 麼，在家時間也不多，卻從不給我壓力，總是微笑地鼓勵我，用充滿愛的家常料 理撫慰我的身心。也謝謝我的婆家，對於這個總是忙於實驗的媳婦並未有太多要 求，甚至把我當成自家女兒般地愛護與支持。
最後，最感謝的是我的先生 白振學，我們互相扶持走過念博士班最艱辛的歲 月，也一起共享最美好的時光，就像彼此的左右手般，擁有不需言說的默契，是 我的最佳實驗夥伴，也是我最重要的人生伴侶。今年年底，我們的第一個孩子即 將出世，他在我腹中一起參與了 paper 及論文的撰寫和爸爸媽媽的畢業口試，感謝 他在我最忙碌的時刻始終乖巧相伴，謝謝你，我的寶貝。
IV
Abbreviation
7-AAD 7-aminoactinomycin
ABL1 ABL proto-oncogene 1
ALL Acute lymphoblastic leukemia
AML Acute myeloid leukemia
ANOVA Analysis of variance
APC Allophycocyanin
BAC Bacterial artificial clone B-ALL B-cell precursor ALL BCR Breakpoint Cluster Region
bp Base pair
BrdU 5'-bromo-2'-deoxyuridine
BSA Bovine serum albumin
CBC Complete blood counts
CD Cluster of differentiation
CDS Coding domain sequence
ChIP Chromatin Immunoprecipitation CLL Chronic lymphocytic leukemia CML chronic myeloid leukemia CRLF2 Cytokine receptor-like factor 2
DC Differential count
DMEM Dulbecco's Modified Eagle Medium DPBS Dulbecco's phosphate buffered saline EDTA Ethylenediaminetetraacetic acid
Epo Erythropoietin
ERG ETS-related gene
ETO Eight-Twenty-One
ETV6 Ets variant 6
FBS Fetal bovine serum
FITC Fluorescein isothiocyanate
G0 Gap 0
G1 Gap 1
GAPDH glyceraldehyde-3-phosphate dehydrogenase HCC Hepatocellular carcinoma
HCl Hydrochloric acid
HDAC Histone deacetylase
V
HEK Human embryonic kidney
HEPES 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid
HLH Helix-loop-helix
HRP horseradish peroxidase HSC Hematopoietic stem cell
iAMP21 Intrachromosomal amplification of chromosome 21 IGH Immunoglobulin Heavy chain locus
IKZF1 IKAROS family zinc finger
kb kilobase
kg kilogram
mg miligram
MIR151 microRNA 151
MIR181A1 microRNA 181A1
miRNA MicroRNA
MTT 3-(4,5- dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide
NC Negative control
N-CoR Nuclear receptor co-repressor 1 PAX5 Paired box homeotic gene 5 PBS Phosphate buffered saline PCR Polymerase chain reaction
PE R-Phycoreythin
PerCP Peridinin chlorophyll protein complex
PI propidium iodide
PLAG1 Pleomorphic adenomas gene 1 pre-B cell precursor B cell
pre-miRNA precursor microRNA pri-miRNA primary microRNA pro-B cell progenitor B cell
PVDF Polyvinylidene difluoride
PVDF Immobilon polyvinylidene difluoride
qRT-PCR Quantitative reverse transcription-polymerase chain reaction
RBC Red blood cell
RIPA Radioimmunoprecipitation assay RISC RNA-induced silencing complex ROS Reactive oxygen species
RPMI Roswell Park Memorial Institute
RT-PCR Reverse transcription-polymerase chain reaction
VI
RUNX1 Runt-related transcription factor 1 SDS Sodium dodecyl sulfate
SDS-PAGE Sodium dodecyl sulfate polyacrylamide gel electrophoresis SERCA2 Sarcoplasmic/endoplasmic reticulum Ca22+-ATPase 2 siRNA short interfering RNA
SMRT Silencing mediator of retinoic acid and thyroid hormone receptor TSS Transcriptional start site
UTR untranslated region
VPA Valproic acid
VII
Table of Contents
前言 I
Preface II
誌謝 III
Abbreviation IV
Table of content VII
The First Part of Childhood Acute Lymphoblastic Leukemia Associated MicroRNAs:
I. A Double Negative Loop Comprising of ETV6/RUNX1 and MIR181A1 1
摘要 2
Abstract 4
List of Figures 7
List of Tables 9
List of Appendixes 10
Chapter 1. Introduction 11
1.1. Childhood B-cell precursor acute lymphoblastic leukemia 11
1.2. Genetic aberrations in B-ALL 11
1.3. ETV6/RUNX1-positive B-ALL 13
1.3.1. ETV6/RUNX1 fusion gene 13
1.3.2. Structure and function of ETV6/RUNX1 fusion protein 14
1.3.3. Role of ETV6/RUNX1 in leukemogenesis of B-ALL 15
1.4. MicroRNAs 16
1.4.1. Overview 16
1.4.2. MicroRNAs in hematopoiesis and leukemogenesis 17
1.4.3. MIR181A1 gene 18
1.4.4. MicroRNAs associated with ETV6/RUNX1 19
1.5. Research motive and the aim 20
Chapter 2. Materials and Methods 21
2.1. Materials 21
2.1.1. Reagents 21
2.1.2. Kits 22
2.1.3. Antibodies 23
2.1.4. Vectors 24
2.1.5. Instruments 24
2.2. Methods 25
VIII
2.2.1. Patients 25
2.2.2. RNA preparation 26
2.2.3. Quantitative real-time PCR 26
2.2.4. MicroRNA expression profile 27
2.2.5. Cell culture 27
2.2.6. Cell viability 28
2.2.7. Proliferation and cell cycle 28
2.2.8. Apoptosis assay 29
2.2.9. Flow cytometric analysis of lineage markers 30
2.2.10. Chromatin immunoprecipitation 30
2.2.11. Western blotting 31
2.2.12. siRNA transfection 31
2.2.13. miRNA precursor transfection 32
2.2.14. ETV6/RUNX1 and RUNX1 protein overexpression 32
2.2.15. Lentiviral construct and infection 33
2.2.16. Luciferase reporter assay 34
2.2.17. Statistical analyses 34
Chapter 3. Results 36
3.1. ETV6/RUNX1 directly downregulates MIR181A1 36
3.1.1 ETV6/RUNX1-associated miRNA expressions in clinical samples 36 3.1.2 siRNA knockdown of ETV6/RUNX1 up-regulates miR-181a-1 expression 37 3.1.3 Overexpression of ETV6/RUNX1 down-regulates miR-181a-1 expression 38 3.1.4 ETV6/RUNX1 binds the regulatory region of MIR181A1 39 3.1.5 Transcriptional co-repressor HDAC3 is recruited to the regulatory region of
MIR181A1
39
3.2. MIR181A1 targets PLAG1 oncogene in B-ALL 40
3.2.1. Upregulation of PLAG1 mRNA in clinical samples 40 3.2.2. Overexpression of miR-181a inhibits PLAG1 expression in REH cells 41
3.3. MIR181A1 negatively regulates ETV6/RUNX1 42
3.3.1. Overexpression of miR-181a downregulates ETV6/RUNX1 in REH cells 42 3.3.2. miR-181a targets the miR-181a recognition sequence located in RUNX1-3' UTR 42
3.4. The cellular effects of MIR181A1 on B-ALL cells 43
3.4.1. Ectopic expression of MIR181A1 impedes REH cell growth 43 3.4.2. Apoptotic cells increases in MIR181A1-lentivirus transduced REH cells 44 3.4.3. The percentage of G0/G1 phase population increases in MIR181A1-lentivirus
transduced REH cells
44
3.4.4. Ectopic expression of MIR181A1 enhances REH cell differentiation 45
IX
3.4.5. MIR181A1 expression enhances apoptosis of differentiated cells 45 3.4.6. Ectopic expression of MIR181A1 induces partial differentiation of
ETV6/RUNX1-positive primary ALL blasts
46
Chapter 4. Discussion 48
4.1 Selection of Mir181A1 to be investigated 48
4.2 The relationship between ETV6/RUNX1 and MIR181A1 49
4.3 A double negative loop comprising ETV6/RUNX1 and MIR181A1 51
4.4 The effects of MIR181A1 on B-ALL cells 52
Chapter 5. Conclusion and Prospective 55
Bibliography 57
Figures 68
Tables 104
Appendixes 110
The Second Part of Childhood Acute Lymphoblastic Leukemia Associated
MicroRNAs: II. Establishment of Mir151 conventional knockout mice 116
摘要 117
Abstract 118
List of Figures 120
List of Tables 121
List of Appendixes 122
Chapter 6. Introduction 123
6.1. MicroRNAs 123
6.2. microRNA 151a 124
6.2.1. Human MIR151A gene 124
6.2.2. Clinical association and target mRNAs of MIR151A 125
6.2.3. Mouse Mir151 gene 126
6.3. Research motive and the strategy 126
Chapter 7. Materials and Methods 128
7.1. Materials 128
7.1.1. Reagents 128
7.1.2. Kits 129
7.1.3. Vectors 129
7.1.4. Equipment 129
7.2. Methods 130
7.2.1. Targeting vector construction 130
7.2.2. Gene targeting of ES cells and generation of Mir151 conventional knockout 131
X
(KO) mice
7.2.3. Animals 132
7.2.4. RNA preparation and reverse transcription 133
7.2.5. Quantitative real-time PCR 133
7.2.6. Complete blood counts and differential counts 134
7.2.7. Clinical chemistry 134
7.2.8. Chronic hypoxia 135
7.2.9. CoCl2 treatment 135
7.2.10. Tumor analysis 136
7.2.11. Urethane-induced lung cancer model 136
7.2.12. Histological analysis 137
7.2.13. Statistical analyses 137
Chapter 8. Results 138
8.1. Generation and identification of Mir151 conventional knockout mice 138 8.1.1. Generation of Mir151 conventional knockout mice 138 8.1.2. Identification of gene status in DNA and RNA level 138 8.2. Characterization of Mir151 conventional knockout mice 139
8.2.1. Mir151 is not essential to survival 139
8.2.2. A kinetic change of erythropoiesis 140
8.2.3. Elevated renal Epo mRNA level in young MiR151-/- mice 140
8.2.4. Induction of renal Epo by chronic hypoxia 141
8.2.5. No increase in renal Epo-producing cells in MiR151-/- mice 142 8.2.6. Increase of Hif-α target gene expressions in young MiR151-/- mice 143 8.2.7. Induction of Epo expression by CoCl2 treatment 144 8.3. Long-term observation of Mir151 conventional knockout mice 144
8.3.1. No difference in survival 144
8.3.2. Spontaneous developed lung cancer in elder Mir151 knockout mice 145 8.3.3. Increase number of urethane-induced lung tumors in elder Mir151 knockout
mice
145
8.4. Depletion of Mir151 protected young mice from urethane-induced lung cancer 146
Chapter 9. Discussion 147
9.1. Increased erythropoiesis in young mice 147
9.2. Spontaneous developed and urethane-induced lung cancer in elder mice v.s. the protective effect of Mir151 loss in young mice.
148
9.3. The potential application of Mir151 knockout mice 149
Chapter 10. Conclusion and Prospective 150
Bibliography 151
XI
Figures 155
Tables 175
Appendixes 180
1
The First Part
Childhood Acute Lymphoblastic Leukemia Associated MicroRNAs: I. A Double Negative Loop
Comprising of ETV6/RUNX1 and MIR181A1
2
摘要
急性淋巴性白血病為最常見的兒童癌症，而造成 ETV6/RUNX1 融合基因表現 的 t(12;21) 染色體轉位則是在兒童急性前 B 細胞淋巴性白血病中最常見的染色 體異常。微核醣核酸是一非編碼的小核醣核酸，長度僅 18–23 個核苷酸，是由 70-100 個核苷酸的微核醣核酸前驅物切割而來，其作用主要是在後基因轉錄階段 抑制基因表達。幾乎所有的生理機制都受到微核醣核酸的影響，包括造血細胞分 化，甚至已知有部分微核醣核酸參與在白血病癌化過程中。為探討與 ETV6/RUNX1 相關之微核醣核酸，本研究分析 50 個兒童前 B 細胞淋巴性白血病檢體中微核醣 核酸的表達情形，其中包含 10 個 ETV6/RUNX1 陽性病例。透過與 ETV6/RUNX1 陰性病人細胞相比較，本研究找到 17 個在 ETV6/RUNX1 陽性病人細胞表達量顯 著下降的微核醣核酸。在這些具顯著差異的微核醣核酸之中，由微核醣核酸 181a-1 前驅物 3ʹ 端衍生而來的成熟產物微核醣核酸 181a-1（miR-181a-1），其表現量的 改變最具統計意義（下降近百分之七十五，P值 < 0.001）。此外，MIR181A1 基 因中具有 RUNX1 蛋白的 DNA 結合位（TGTGGT），因此本研究選擇針對 MIR181A1 進行更深入的探討。
REH 細胞為 ETV6/RUNX1 陽性的前B細胞急性淋巴性白血病細胞株，本研究 利用小片段干擾 RNA 抑制 REH 細胞的 ETV6/RUNX1 表達可使 miR-181a-1 表現量上升，此外在人類胚胎腎臟 293FT 細胞中過量表達 ETV6/RUNX1 融合蛋 白 則可 抑制 miR-181a-1 表現量。本 研究 並以免疫染色質 沉澱 法 （chromatin immunoprecipitation ） 在 REH 細 胞 株 驗 證 被 預 測 的 RUNX1 結 合 位 ， 證 明 MIR181A1 直接受到 ETV6/RUNX1 融合蛋白調控。相較於 miR-181a-1，被報導
具有功能的是另一股微核醣核酸 181a（miR-181a），為尋找其下游標的，本研究 將 REH 細胞轉染 miR-181a 並檢測 PLAG1 表達，已知在慢性淋巴性白血病
（chronic lymphoblastic leukemia）中 PLAG1 為 miR-181a 標的。在過量表達 miR-181a 的 REH 細胞中，PLAG1 蛋白表達量下降；而在 ETV6/RUNX1 陽性臨
3
床檢體中 PLAG1 mRNA 表達明顯增加。上述證據皆指出，如同慢性淋巴性白血 病研究結果，PLAG1 基因在兒童急性淋巴性白血病亦為 miR-181a 標的。此外，
在 miR-181a 轉染之 REH 細胞中 ETV6/RUNX1 蛋白表達量亦顯著下降，且與 RUNX1 3ʹ 端 未 轉 譯 區 域 （ 3ʹ -untranslated region ） 上 的 miR-181a 辨 識 序 列
（UGAAUGU）相關。本研究並利用 REH 細胞證明過量表達 miR-181a 會促進 表現 ETV6/RUNX1 之前 B 細胞急性淋巴性白血病細胞由前 BI 階段（pre-BI stage ） 分 化為未成熟 B 細胞 （ immature B cells ） 。且 miR-181a 亦可 誘導 ETV6/RUNX1 陽性臨床病人檢體 CD10 抗原表達量減少，意即細胞有部分分化的
現象。
統整上述研究成果，本研究顯示 MIR181A1 及 ETV6/RUNX1 可相互調控，
並推論一涉及 MIR181A1 與 ETV6/RUNX1 的雙向負調控迴圈機制可能參與在由 ETV6/RUNX1 驅動之前B細胞急性淋巴性白血病分化停滯。
關鍵字：前 B 細胞急性淋巴性白血病，t(12;21)轉位，ETV6/RUNX1，微核醣核酸 181a-1，微核醣核酸 181a
4
Abstract
Acute lymphoblastic leukemia is the most common pediatric cancer, and the
chromosomal translocation t(12;21), which resulting in expression of ETV6/RUNX1
fusion gene, is the most frequent chromosomal lesion in childhood B-cell precursor
(pre-B) ALL. MicroRNAs (miRNAs) are small noncoding RNAs with 18–23
nucleotides arisen from cleavage of 70-100 nucleotide precursors and mostly down
regulate gene expression at post-transcriptional level. They have been implicated in
virtually all aspects of biology including hematopoietic cell differentiation and some of
them are also known to participate in leukemogenesis. To investigate the miRNAs that
are associated with regulation of ETV6/RUNX1 expression, we performed miRNA
expression profiling on fifty leukemic samples from children with pre-BALL, including
10 cases positive for ETV6/RUNX1. We identified 17 miRNAs that were
down-regulated in ETV6/RUNX1-positive, compared with ETV6/RUNX1-negative
B-ALL. Of these miRNAs, miR-181a-1, one of the mature form derived from the 3ʹ arm
of precursor hsa-mir-181a-1, gives the most significant fold-change (reduced by ~75%,
P<0.001). In addition, MIR181A1 contains a potential RUNX1 binding site (TGTGGT),
thus we selected MIR181A1 for further investigation.
In REH cells, an ETV6/RUNX1-positive B-ALL line, siRNA knockdown of
ETV6/RUNX1 resulted in increased miR-181a-1 expression, while overexpression of
5
ETV6/RUNX1 fusion protein in HEK-293FT cells resulted in reduction of miR-181a-1
level. The predicted RUNX1 binding site was also confirmed in REH cells by chromatin
immunoprecipitation analysis, indicating MIR181A1 is a direct target of the
ETV6/RUNX1 fusion protein. To search for downstream targets of miR-181a, the
functional counterpart of miR-181a-1, REH cells were transfected with miR-181a and
the expression of PLAG1, shown to be a target of miR-181a-1 in chronic lymphoblastic
leukemia (CLL) was examined. The PLAG1 protein level was decreased in miR-181a
over-expressed REH cells. In addition, the PLAG1 mRNA level was increased in
ETV6/RUNX1-positive clinical samples. This indicated that PLAG1 gene might be the
down-stream target of miR-181a in childhood ALL as in CLL. Furthermore, we found
ETV6/RUNX1 protein was also decreased in miR-181a-transfected REH cells,
correlating with the existence of a miR-181a recognition sequence (UGAAUGU) at the 3’-untranslated region of RUNX1. Using REH cells, we showed ectopic expression of
miR-181a could enhance ETV6/RUNX1-expressing B-ALL cells differentiate from
pre-BI stage to immature B cells. In addition, miR-181a could induce partial
differentiation of ETV6/RUNX1-positive clinical patient samples by diminishing CD10
expression.
Taken together, our results demonstrate that MIR181A1 and ETV6/RUNX1 regulate
each other, and we propose that a double negative loop involving MIR181A1 and
6
ETV6/RUNX1 may contribute to ETV6/RUNX1-driven differentiation arrest in B-ALL.
Key words：pre-B ALL，t(12;21) translocation，ETV6/RUNX1，miR-181a-1，miR-181a
7
List of Figures
Figure 1: hsa-mir-181a-1 and 5p sequence of miR-181 family 69 Figure 2: MicroRNA expression profile in childhood B-ALL patients 70 Figure 3: Validation of individual miRNA expression 71 Figure 4: Expression of ETV6/RUNX1 fusion protein and wild type RUNX1
protein in B-ALL cell lines
72
Figure 5: siRNA-mediated knockdown of ETV6/RUNX1 fusion gene 73 Figure 6: Effect of siRNA knockdown of ETV6/RUNX1 on miR-181a-1
expression
74
Figure 7: Overexpression of RUNX1 or ETV6/RUNX1 in HEK-293FT cells 75 Figure 8: Effects of RUNX1 and ETV6/RUNX1 overexpression on
miR-181a-1 expression
76
Figure 9: Schematic representation of the genomic structure of human MIR181A1 gene
77
Figure 10: ETV6/RUNX1 binds to regulatory region of MIR181A1 78 Figure 11: HDC3 binds to regulatory region of MIR181A1 79 Figure 12: Expression of PLAG1 mRNA in B-ALL clinical samples 80 Figure 13: Overexpression of miR-181a in REH cells by transfection with
precursor miRNA
81
Figure 14: PLAG1 expression in miRNA precursor transfected REH cells 82 Figure 15: ETV6/RUNX1 expression in miRNA mimics transfected REH cells 83 Figure 16: miR-181a targets the 3ʹ-UTR of ETV6/RUNX1 84 Figure 17: Stable expression of MIR181A1 in REH cells via lentiviral
transduction
86
Figure 18: Assessment of cell growth of lentivirus transduced REH cell 87 Figure 19: Apoptosis analysis of lentivirus transduced REH cell 88 Figure 20: Analysis of cell cycle and proliferation of lentivirus transduced
REH cells
89
Figure 21: Analysis of CD10 expression on lentivirus-infected REH cells 90 Figure 22: Analysis of CD20 expression on lentivirus-infected REH cells 91 Figure 23: Analysis of surface IgM expression on lentivirus-infected REH cells 92 Figure 24: Analysis of К-chain expression on lentivirus-infected REH cells 93 Figure 25: Analysis of λ-chain expression on lentivirus-infected REH cells 94 Figure 26: Analysis of CD10 expression on apoptotic REH cells 95 Figure 27: Ectopic expression of MIR181A1 in primary ALL cells via
lentiviral transduction
96
8
Figure 28: Detection of ETV6/RUNX1 mRNA in cultured primary ALL cells 97 Figure 29: Surface marker analysis of lentivirus-infected ETV6/RUNX1-positive
primary ALL cells derived from patient #747
98
Figure 30: Surface marker analysis of lentivirus-infected ETV6/RUNX1-positive primary ALL cells derived from patient #752
99
Figure 31: Surface marker analysis of lentivirus-infected ETV6/RUNX1-positive primary ALL cells derived from patient #745
100
Figure 32: Surface marker analysis of lentivirus-infected ETV6/RUNX1-negative primary ALL cells derived from patient #754
101
Figure 33: Viability of lentivirus-infected primary ALL cells derived from patient #745 and #752
102
Figure 34: Schematic representation of the double negative loop comprising ETV6/RUNX1 and MIR181A1
103
9
List of Tables
Table 1: Primer sequences 105
Table 2: siRNA sequences 106
Table 3: Clinical features of the ALL patients included in miRNA expression profiling study
107
Table 4: The statistic signature of 17 miRNAs 108
Table 5: The signature of 13 miRNAs/miRNA clusters and the locations of RUNX1 binding sites
109
10
List of Appendix
Appendix I: Frequency of cytogenetic subtypes of childhood ALL 111 Appendix II: Schematic diagram of the exon/intron structure of the
ETV6 and RUNX1 genes involved in t(12;21)(p13;q22)
112
Appendix III: A schematic representation of the full-length ETV6, RUNX1 and ETV6/RUNX1 proteins
113
Appendix IV: A hypothetical model for the molecular mechanism of ETV6/RUNX1 action
114
Appendix V: The canonical pathway of microRNA biosynthesis 115
11
Chapter 1. Introduction
1.1. Childhood B-cell precursor acute lymphoblastic leukemia
Acute lymphoblastic leukemia (ALL), characterized by excessive proliferation and
differentiation block of abnormal lymphoblast, is an a clinically and biologically
heterogeneous hematologic malignancy originating from a multipotent hematopoietic
stem cell (HSC). It is the most common childhood cancer, accounting for 31% of all
tumors (1), and the leading cause of cancer-related death in children and young adult (2).
According to the type of lymphocyte the leukemia cells come from, ALL has been
classified into B-cell and T-cell lineage ALL. At approximately 80% (1), the majority of
children with ALL have B-cell precursor ALL (B-ALL), which results from the
accumulation of genetic alterations in pre-B cells and demonstrates a pre-B cell-like
phenotype such as exhibiting cell surface markers of normal pre-B cells (CD19+,
CD10+), and appears to be clonal expansion of normal pre-B cells stalled at a particular
stage of the differentiation process (3, 4).
1.2. Genetic aberrations in B-ALL
Recurrent genetic aberrations that block pre-B cell differentiation and direct
aberrant proliferation and cell survival is the hallmark of B-ALL and the driving force
12
of leukemogenesis. Using standard chromosomal and molecular genetic analysis, 75%
of B-ALL cases can be detected harboring a recurring chromosomal abnormalities (2),
which can be subdivide into numerical aberrations, such as hyperdiploidy (>50
chromosomes) and hypodiploidy (<44 chromosomes), and structural alteration like
chromosomal rearrangement. Chromosomal rearrangements resulting from translocation,
inversion, deletion, and duplication are common in B-ALL. These rearrangements often
affect genes encoding regulators of hematopoiesis, tumor suppressors, oncogenes or
tyrosine kinases. Four common translocations, t(12;21)(p13;q22), t(1;19)(q23;p13),
t(9;22)(q34;q11.2), and t(4;11)(q21;q23), are most well-known and compart 30-35% of
childhood ALL (Appendix I) (1, 5). These translocations disrupts the normal genes
involved in regulating hematopoiesis and cause the formation of new fusion genes
which are critical to leukemogenesis. Some of them, such as t(9;22) which results in a
kinase fusion gene―BCR/ABL1, are powerful enough to induce leukemia individually
(6), but some others may require additional genetic hits to induce overt leukemia (7).
Standard methods can detect genetic abnormalities in most of the patients but still
have limitation. Recently, the aid of newly developed approaches including gene
expression microarray profiling, DNA copy number analysis, and next-generation
sequencing (NGS) technologies, have improved the number of lesions found in B-ALL
and virtually all cases can be detected harboring a genetic alteration (8). Numerous
13
mutations that targeted genes in key cellular pathways in B-ALL are identified and are
applied to characterize new subtypes or increase the new insights into known ALL
subtypes, such as BCR/ABL1-like ALL (~15%) (9, 10), intrachromosomal amplification
of chromosome 21 (iAMP21, ~2%) (11, 12), IGH- and CRLF2-rearrangement (5-7%)
(13), ERG-deregulated ALL (~7%) (14), PAX5-deletion (3%) /mutation (5-7%) /
rearrangement (2-3%) (5), and IKZF1-deregulated ALL (~15%) (9, 15).
1.3. ETV6/RUNX1-positive B-ALL
1.3.1. ETV6/RUNX1 fusion gene
The t(12;21) (p13;q22) translocation, which results in ETV6/RUNX1 fusion gene,
was first reported by two different group in 1995 (16, 17). It is the most common
chromosomal rearrangement in childhood B-ALL cases but less prevalent in adult
patients (1, 18). The incidence of t(12;21) in B-ALL is approximately 15-25%, and
patients with this translocation usually have excellent outcome (19). This rearrangement
is not able to be detected by conventional cytogenetic analysis but is readily detected by
fluorescent in situ hybridization and molecular techniques, such as RT-PCR and
quantitative RT-PCR (20-22).
The ETV6 gene, which is very large (240 kb) and consists of eight exons coding for
an ETS-like putative transcription factor containing a helix-loop-helix (HLH) and a
14
DNA-binding domain is on human chromosome 12p13 (23). The RUNX1 gene, which
belongs to the runt domain gene family of transcription factors, is on human
chromosome 21q22 and spans 260 kb consisting of 12 exons (24). Both ETV6 and
RUNX1 genes demonstrated critical roles in hematopoiesis in knockout mice studies
(25-27), and they are also frequently targeted by rearrangements and mutations in
leukemia (28, 29).
The breakpoints on ETV6 gene are clustered in a 15 kb region between exon 5 and
6 (23), and the breakpoints on RUNX1 gene may occur either in the ~100 kb intron 1 or
intron 2 (Appendix IIA) (18, 30). Most of all, the ETV6/RUNX1 fusion transcript shows
a joining of exon 5 of ETV6 to the second exon of RUNX1, while the junction occurred
at the third exon of RUNX1 is less frequently seen (Appendix IIB) (30). Wherever the
breakpoints occurred, these all result in fusion of the 5ʹ portion of ETV6 and almost the
entire coding region of RUNX1 gene.
1.3.2. Structure and function of ETV6/RUNX1 fusion protein
The ETV6/RUNX1 fusion protein is composed of the N-terminal
non-DNA-binding region of ETV6 and nearly complete RUNX1 protein, including the
DNA-binding and activation region, which is responsible for the essential function of
the fusion protein (Appendix III) (16, 18, 31). The ETV6 protein acts as a DNA-binding
15
transcriptional repressor, while RUNX1 can be a DNA-binding transcriptional activator
or repressor depending on the promoter specificity or cell context (18). In contrast to
RUNX1, transiently expressed ETV6/RUNX1 fusion protein generally represses the activities of reporter constructs driven by regulatory regions derived from
hematopoietic-specific genes (18). The ETV6/RUNX1 fusion protein acts as an aberrant
transcription factor that can interfere with the normal functions of wild-type ETV6 and
RUNX1 through multiple mechanisms. ETV6/RUNX1 can dimerize with wild-type
ETV6 through the HLH domain and disrupt the activity of ETV6 (31, 32).
ETV6/RUNX1 may also bind to RUNX1 target DNA sequences and recruit
transcriptional corepressors including mSinA, N-coR, and histone deacetylase-3
(HDAC3) via the fusion part of ETV6, resulting in dysregulated RUNX1-dependent
transcription (Appendix IV) (31, 33, 34).
1.3.3. Role of ETV6/RUNX1 in leukemogenesis of B-ALL
Analysis of monozygotic twins with concordant leukemia and retrospective
screening of neonatal blood spots of patients with leukemia indicate that chromosomal
translocations characteristic of pediatric leukemia often arise prenatally (35). However,
in normal cord blood ETV6/RUNX1 is detected at a frequency that is 100-fold greater
than the risk of the corresponding leukemia (36). Mouse models demonstrate that
16
expression of ETV6/RUNX1 in murine bone morrow stem cells impedes B cell
differentiation from the earliest stages of B cell development with a particular marked impact at the transition from pro-B to pre-B cell stages. Although the accumulation of
both multipotent and B-cell progenitors in vivo, ETV6/RUNX1 is insufficient to induce
leukemia by itself (37, 38). Collectively, these data suggests that ETV6/RUNX1 is a
frequent prenatal first hit in childhood leukemia which can initiate a preleukemic
phenotype remaining covert for up to 15 years but is insufficient for clinical leukemia.
1.4. MicroRNAs
1.4.1. Overview
MicroRNAs (miRNAs) are a group of single-stranded, endogenously initiated
non-coding RNAs which are first discovered in the early 1990s (39). They are
transcribed by RNA polymerase II as part of capped and polyadenylated primary
transcripts (pri-miRNAs) that can be either protein-coding or non-coding. The primary
transcript is cleaved by the Drosha ribonuclease III enzyme to produce an
approximately 70 to 100- nucleotide stem-loop precursor miRNA (pre-miRNA), which
is exported to cytoplasm and is further cleaved by the Dicer ribonuclease to generate the 20-23 nucleotides mature miRNA products from 3ʹ and/or 5ʹ arms (40). When two
mature miRNAs originate from opposite arms of the same pre-miRNA and express
17
similar amounts, they are represented as -3p and -5p (41). In human cells, the mature
miRNAs incorporate into a RNA-induced silencing complex (RISC) and then target
mRNAs for degradation or translational repression via partial or perfect
complementarity to the mRNA 3' untranslated region (3' UTR) through specific seed
sequences (Appendix V) (40, 42). By this way, miRNAs can downregulate gene
expression at the post-transcriptional level. In the recent years, there has been increasing
evidence that miRNAs also bind in the coding region to inhibit translation (43),
implying that miRNAs may flexibly tune the time-scale and magnitude of
post-transcriptional regulation through combination of multiple ways (44). Thus,
miRNAs have the ability to control fundamental cellular functions such as survival,
differentiation, apoptosis, and cell cycle.
1.4.2. MicroRNAs in hematopoiesis and leukemogenesis
A variety of miRNAs have been identified as important regulators of
hematopoiesis (45). For instance, miR-221/222 inhibits the erythropoiesis (46),
miR-223 is essential to modulate myeloid differentiation (47), miR-181a and miR-150
are dynamically regulated during T-cell and B-cell development, respectively, and
premature expression of certain miRNAs in hematopoietic progenitors may impair T-
and B-cell development at the stage transition during maturation (48-50).
18
MiRNA expression in hematological malignancies has also been extensively
studied. Some information about factors that modulate miRNA expression is now
available. Dysregulation of miRNA expression is frequently associated with cytogenetic
abnormalities, and indeed certain abnormalities have direct impacts on aberrant
miRNAs expressions (45). RUNX1/ETO, the most common acute myeloid leukemia–
associated fusion protein resulting from t(8;21), was first reported to directly repress
miR-223 expression by triggering chromatin remodeling and epigenetic silencing,
which in turn blocks the differentiation of myeloid precursor cells (51).
1.4.3. MIR181A1 gene
MIR181A1 gene is located on human chromosome 1q32.1 and is only 62 bp distant
from MIR181B1 gene, they are considered sharing the same primary transcript. There is
only one exon in MIR181A1 gene and expressed a pre-miRNA with 110 bp in length
referred to hsa-mir-181a-1. Hsa-mir-181a-1 will further generate two mature miRNAs
including miR-181a-3p and -5p, which are referred to miR-181a-1 and miR-181a,
respectively (Figure 1A). MIR181A1 belongs to the miR-181 family which consists of
mir-181a/b1, mir-181a/b2 and mir-181c/d, producing four highly similar mature 5p
miRNAs (miR-181a, miR-181b, miR-181c and miR-181d, respectively) with identical
seed sequence and a slight difference in 1~4 bp from three polycistronic transcripts
19
(Figure 1B).
Although both 3p and 5p of hsa-mir-181a can be detected in tissues, previous
reports only demonstrated that miR-181a targets various mRNAs and has physiological
roles and pathological meanings (52-54), while the function of miR-181a-1 remains
unclear. In healthy cells, miR-181a regulates B-cell development, influences T-cell
sensitivity to antigens by modulating T-cell receptor signaling, and is involved in early
steps of hematopoiesis (55). A tumor suppressor activity of miR-181a is reported in
chronic lymphocytic leukemia (CLL) (56, 57), glioma (53), and astrocytoma (54). In
addition, ectopic expression of miR-181a has been shown to sensitizes acute myeloid
leukemia (AML) cell lines to chemotherapy (58), and enhance the effect of radiation
treatment on malignant glioma cells via down-regulation of the Bcl-2 protein (59).
1.4.4. MicroRNAs associated with ETV6/RUNX1
Recent studies have shown that aberrant miRNA expression also plays an
important role in malignant transformation of ETV6/RUNX1 ALL. A highly expressed
miR-125b-2 cluster was found in ETV6/RUNX1 ALL and may provide leukemic cells
with a survival advantage against growth inhibitory signals in a p53-independent
mechanism (60). In addition, ETV6/RUNX1 was shown to regulate the cellular level of
the Survivin protein and apoptosis via suppression of miR-494 and miR-320a
20
expression by direct binding to the promoter regions of their encoding genes (61).
1.5. Research motive and the aim
It is believed that ETV6/RUNX1 expression may allow quiescent, preleukemic
cells to exist in the bone marrow via transcriptional deregulation of downstream genes.
There is ample evidence that leukemogenesis driven by ETV6/RUNX1 is mediated in
part by conferring survival signals through direct modulation of multiple targets such as
EPOR, MDM2, and miRNA genes (61-63). However, miRNAs involved in the
ETV6/RUNX1-mediated B-cell differentiation arrest is not well understood. For this
purpose, we designed and performed the experiments to identify the miRNAs which are
regulated by ETV6/RUNX1 and elucidated the underlying regulatory mechanism.
21
Chapter 2. Materials and Methods
2.1. Materials
2.1.1. Reagents
Product name Company
1 kb DNA Ladder Bertec
100 bp DNA Ladder Bertec
20x Human GAPDH VIC/MGB Applied Biosystems
2-Mercaptoethanol Sigma
2x TaqMan Universal PCR Master Mix Applied Biosystems
Acrylamide Sigma
Agar, Bacteriological ALPHA biosciences
Agarose Invitrogen
Ammonium persulfate Sigma
Ampicillin Sigma
bis-Acrylamide Sigma
Boric acid J.T.Baker
Bovine serum albumin (BSA) Sigma
BPB Sigma
Calf Intestine Alkaline Phosphatase (CIP) Fermentus
Chlorofrom J.T.Baker
Dithiothreitol (DTT) Invitrogen
DMEM Hyclone
dNTPs GeneDirex
Dulbecco's phosphate buffered saline (DPBS) Gibco
EDTA J.T.Baker
EDTA•Na2 J.T.Baker
Ethanol Sigma
Ethidium Bromide (EtBr) Amersco
Fetal bovine serum (FBS) Biological Industries
Formaldehyde Sigma
Glucose Sigma
Glycerol Mallinckrodt
22
Glycine J.T.Baker
HEPES Gibco
Isopropanol Fluka
L-glutamine Gibco
Methnol Sigma
OPTI-MEM I Invitrogen
PageRugular Pre-Stained Protein Ladder Fermentus Penicillin/Streptomycin (P/S) Gibco Phosphate buffered saline (PBS) Biowest
Polybrene Sigma
Protease inhibitor cocktails Roche
Puromycin Sigma
RIPA buffer Thermo
RPMI-1640 Gibco
Skimmed milk 安佳
Sodium bicarbonate Gibco
sodium dodecyl sulfate (SDS) Pierce
Sodium pyruvate Gibco
StemSpamTM CC100 StemCell Technologies
StemSpamTM SFEMII StemCell Technologies
TEMED Sigma
Tris base J.T.Baker
Trypan blue Biowest
Trypsin Gibco
Tryptone ALPHA biosciences
Tween 20 Sigma
Valproic acid sodium salt (VPA) Sigma
Xylene Sigma
Yeast extract ALPHA biosciences
2.1.2. Kits
Product name Company
Bradford protein assay Bio-Rad
BrdU-FITC flow kit BD
Cell Proliferation Kit I (MTT) Roche
Chromatin Immunoprecipitation (ChIP) Assay Kit Upstate
23
Dual-Luciferase Reporter Assay Promega
FastStart universal SYBR green master (ROX) Roche
FavorPrep plasmid extraction midi kit FAVORGEN
FavorPrep plasmid extraction mini kit FAVORGEN
FITC Annexin V apoptosis detection kit BD Immobiion Western Chemilum HRP substrate Millipore
Lipofectatime 2000 Invitrogen
NeonTM transfection system 10 μL kit Invitrogen
PEG-it kit SBI
Pre-miR™ miRNA Precursor Molecule (hsa-mir-181a) Ambion
Pre-miR™ Negative Control #1 Ambion
Qiagen Plasmid Maxi kit Qiagen
QuikChange Lightning Site-Directed Mutagenesis kit Agilent
RnaseOut Invitrogen
siPORT™ NeoFX™ Transfection Agent Ambion
siRNA Invitrogen
SuperScript III Reverse Transcriptase Invitrogen
T4 DNA Ligase NEB
Taq DNA polymerase Geneaid
TaqMan miRNA expression assay Applied Biosystems
TaqMan® MicroRNA Cells-to-CT™ Kit Ambion
TaqMan® miRNA RT kit Ambion
TransIT-LT1 Mirus Bio.
Trizol reagent Invitrogen
2.1.3. Antibodies
Product name Company
mouse monoclonal anti-PLAG1 (H00005324-M02) Abnova mouse polyclonal anti-β-actin (NB600-501) Novus rabbit polyclonal anti-RUNX1 (ab50541) Abcam HRP-conjugated Goat anti-mouse IgG (AP124P) Millipore HRP-conjugated Goat anti-rabbit IgG (AP132P) Millipore
24
2.1.4. Vectors
Product name Source
pGL3-SV40 Promoter Vector Promega
pRL-SV40 Vector Promega
pLKO.1.Null-T vector RNAi core facility, Acadamia Sinia
pMD.G vector RNAi core facility, Acadamia Sinia
pCMVΔR8.91 vector RNAi core facility, Acadamia Sinia pCMV-XL4-ETV6 (NM_001987.3) Origene
pCMV-XL4-RUNX1 (NM_001754.3) Origene
2.1.5. Instruments
Instrument name Company
7300 Real-Time PCR machine Appied Biosystems
GeneAmp PCR machine Appied Biosystems
AllegraTM 21R centriguge Beckman Coulter
Avanti® J-E high speed centrifuge Beckman Coulter
KUBOTA 2420 centriguge KUBOTA
Eppendorf microcentriguge (F45-24-11) Eppendorf Orbital shacking incubator (OSI500R) TKS
Analytical balance (TE124s) Sartorius
Ultrasound sonicator (UP200H) Hielscher
LAB ROTATOR Digisystem
Microporator MP-100 Digital Bio Tech.
Cell cuture incubator (MCO-15AC) SANYO
SynergyHT ELISA reader BioTek
SpectraMaxR M5 multi-detection reader Molecular Devices FUSION-SOLO Chemiluminescence imaging system Vilber Lourmat
FACSCalibur BD
Nanodrop Thermo
25
2.2. Methods
2.2.1. Patients
All of the B-cell precursor ALL patient samples were obtained at the time of diagnosis
and prior to treatment. Viable diagnostic bone marrow (BM) or peripheral blood (PB)
was obtained from 50 children who were diagnosed with B-ALL between July 1996 and
July 2014 at National Taiwan University Hospital (NTUH) and National Cheng Kung
University Hospital (NCKUH). The diagnosis of ALL was made based on the
morphologic findings of BM aspirates, as well as on immunophenotype analyses of
leukemic cells by flow cytometry. Conventional cytogenetics analyses were performed
as part of the routine workup. Patients were prospectively assigned to one of three risk
groups (standard, high, and very high) based on their presenting clinical features and the
biological features of their leukemic cells. Patients were considered to have
standard-risk (SR) ALL if they were between 1 and 9 years old with a presenting
leukocyte count less than 10×109 cells/L or were between 2 and 7 years old with a
presenting leukocyte count between 10×109 and 50×109 cells/L. Patients were
considered to have high-risk (HR) ALL if they were between 1 and 9 years old with a
presenting leukocyte count between 50×109 and 100×109 cells/L, or between 1 and 2 or
7 and 10 years old with a presenting leukocyte count between 10×109 and 50×109
26
cells/L. Patients with at least one of the following were assigned to the very-high-risk
(VHR) group: age younger than 1 year, initial leukocyte count greater than 100×109
cells/L, or presence of BCR-ABL, MLL-AF4 or other MLL rearrangements in B-ALL.
The Institutional Review Board of National Taiwan University Hospital approved the
study. In accordance with the Declaration of Helsinki, we obtained written informed
consent from the parents of each patient before collection.
2.2.2. RNA preparation
Mononuclear cells from bone marrow or peripheral blood were Ficoll purified and
immediately stored in liquid nitrogen. Cryopreserved samples were thawed and washed
in 2% FBS-supplemented 1X PBS prior to RNA extraction. Total RNA was extracted
using Trizol reagent according to the manufacturer’s instructions.
2.2.3. Quantitative real-time PCR
Transcripts of human ETV6/RUNX1 were quantified by TaqMan real-time PCR using
published primer probe combinations (22), and the TaqMan endogenous control assay
for GAPDH was combined in the same reaction. Expression of PLAG1 and the
reference gene GAPDH was determined by SYBR Green real-time PCR and measured
in two independent assays. The primer and probe sequences used in this study were
27
shown in Table 1. All of the assays were run in duplicate.
2.2.4. MicroRNA expression profile
MiRNA expression profiling was performed using the ABI PRISM 7900 Real
Time PCR System and stem-loop reverse transcription-quantitative PCR miRNA arrays
containing 397 mature human miRNAs. 365 miRNAs were assayed using TaqMan
miRNA arrays with 100 ng of RNA as the input for each reverse transcriptase reaction according to the manufacturer’s protocol. Each individual miRNA in primary ALL
blasts and cell line experiments was measured using TaqMan miRNA assays. All
miRNA assays were run concurrently with a calibration control, U6 snRNA and were
run in triplicate.
2.2.5. Cell culture
The REH cell line (human B-cell precursor leukemia, ETV6/RUNX1-positive) from
American Type Culture Collection was grown in 6-well plates at 105 to 106 cells/mL,
depending on experimental conditions. REH cells were cultured in RPMI medium
supplemented with 10 mM HEPES, 1 mM sodium pyruvate, 4.5 g/L glucose, 1.5 g/L
sodium bicarbonate, and 10% FBS. HEK-293FT cells were cultured in 24-well plates
and 60-mm dishes for the luciferase reporter assay and lentivirus packaging,
28
respectively. HEK-293FT cells were grown in DMEM supplemented with 6 mM
L-glutamine, 1 mM sodium pyruvate, and 10% FBS. Primary leukemic cells were obtained from patients with active precursor B-ALL. Primary ALL blasts were isolated
from freshly harvested bone marrow aspirates by density centrifugation using
Ficoll-Paque followed by two washes with RPMI medium. After freezing-thawing,
primary ALL blasts were cultured in SFEMII supplemented with a cytokine cocktail
containing recombinant human Flt3 ligand, stem cell factor, and thrombopoietin to
support the proliferation of hematopoietic progenitors.
2.2.6. Cell viability
The viability of cultured cells was determined by assaying the reduction of MTT to
formazan using the Cell Proliferation Kit I. Briefly, 100 L REH cells or primary ALL blasts were plated in 96-well plates, and 50 g MTT per 100 L 1X DPBS was added to each well at different times. Cells were then incubated at 37°C for 4 hours, and 100 L 10% SDS in 0.01 M HCl was added to dissolve the formazan crystals. Absorbance was
measured at 550 nm and 690 nm with a Synergy HT multi-detection microplate reader.
2.2.7. Proliferation and cell cycle
A BrdU flow kit was used to determine cell cycle kinetics and to measure BrdU
29
incorporation into DNA of proliferating cells. Briefly, cells (1.5 × 105 cells/mL) were
seeded in 6-well tissue culture plates and cultured for 48 hours followed by the addition
of 10 M BrdU, and the incubation was continued for an additional 30 minutes. Cells
were fixed in a solution containing paraformaldehyde and the detergent saponin, and
then they were incubated for 1 hour at 37°C with DNase (30 g per sample).
FITC-conjugated anti-BrdU (1:50 dilution in wash buffer) was added, and incubation
was continued for 20 minutes at room temperature. Cells were washed, and DNA was
stained using 7-AAD (20 L per sample), followed by flow cytometric analysis. The
BrdU content (FITC) and total DNA content (7-AAD) were determined using FCS
Express software. All experiments were carried out three times.
2.2.8. Apoptosis assay
Apoptosis was evaluated by staining with annexin V/PI and flow cytometric
analysis. Briefly, REH cells were harvested, washed, and resuspended in annexin V
binding buffer. Then, the cells were stained with annexin V-FITC in the dark at room
temperature for 10 minutes, centrifuged, and gently resuspended in annexin V binding
buffer. Finally, 10 μL PI staining solution was added and gently mixed, and cells were
kept on ice in the dark and immediately subjected to flow cytometry. All experiments
were carried out three times.
30
2.2.9. Flow cytometric analysis of lineage markers
To assess cell-surface markers, cells were suspended in 1% BSA/PBS and stained
with the appropriate dilution of the antibodies for 15 minutes at room temperature.
Before detection, cells were washed with 1% BSA/PBS and resuspended in 1%
paraformaldehyde/PBS. Monoclonal antibodies recognizing the following cell-surface
markers were used for flow cytometry: CD19, CD10, CD20, CD45, IgM, κ-chain, and λ-chain. Marker analyses were performed by using flow cytometry.
2.2.10. Chromatin immunoprecipitation
We used the chromatin immunoprecipitation (ChIP) kit to perform the assays. Briefly,
cells were harvested, and chromatin was cross-linked with formaldehyde at a final
concentration of 1%. After lysis of the cells, samples were sonicated to an average DNA
length of 300 to 500 bp. The chromatin was immunoprecipitated overnight with
antibodies against RUNX1 and HDAC3. The HDAC inhibitor valproic acid (VPA) was
used to release the binding of HDAC3; REH cells were treated with 2 mM VPA for 24
hours before harvesting. Chromatin was also purified from cross-linked DNA that had
not been immunoprecipitated to serve as an input DNA control. A genomic region close
to the putative RUNX1-binding site (P1), which is 3 kb upstream of the MIR181A1
transcription start site predicted by CoreBoost_HM
31
(http://rulai.cshl.edu/tools/CoreBoost_HM/) (64), and another MIR181A1 upstream
region (P2) was amplified by PCR. As a positive control, the primer set PC was used to
amplify the promoter region of MIR223 as previously described (51). All of the primers
used for PCR were listed in Table 1. The entire experiment was carried out three times
with similar results.
2.2.11. Western blotting
Cells were pelleted, washed with cold PBS, and lysed for 30 minutes on ice in RIPA
buffe with protease inhibitor cocktail. Lysates were cleared by centrifugation at 14,000
× g at 4°C for 15 minutes, and 35 μg total protein was separated by SDS-PAGE and
transferred to an PVDF membrane. The membrane was blocked and incubated
overnight with primary antibodies. After a final incubation with secondary antibodies
conjugated with HRP (1:5000 dilution), immune complexes were detected with HRP
chemiluminescent substrate. Antibodies and dilutions used were: anti-RUNX1 (1:1000),
anti-PLAG1 (1:500), and anti-β-actin (1:5000).
2.2.12. siRNA transfection
For ETV6/RUNX1 silencing with a siRNA, REH cells were transfected with a
mixture of siRNAs targeting the fusion region of ETV6/RUNX1 or a nonfunctional
32
control, siRNA-S (65, 66). The siRNA sequences were shown in Table 2. The siRNAs
were transfected into REH cells via electroporation with a MP-100 microporator in a
100-μL gold tip under the following conditions: 1 × 106 cells/mL antibiotic-free culture
medium, 230 nM siRNA, one pulse of 1,150 V for 30 milliseconds. After 48 hours of
transfection, cells were harvested to assess target gene expression.
2.2.13. miRNA precursor transfection
The miRNA precursors hsa-mir-181a and negative control 1 are partially
double-stranded RNAs that mimic endogenous precursor miRNAs. Each miRNA
precursor was transfected into cells at a final concentration of 50 nM using
siPORT NeoFx transfection agent. Two rounds of transfection were performed with a
48-hour interval between the first and second round. The effects manifested by the
introduction of the precursor miRNAs into the cells were assayed after the second round
of transfection.
2.2.14. ETV6/RUNX1 and RUNX1 protein overexpression
The pCMV6-XL4 vector expressing either ETV6/RUNX1 (pCMV-XL4-E/R) or
RUNX1 (pCMV-XL4-RUNX1) protein was transfected into HEK-293FT cells using
the transfection reagent TransIT-LT1 according to the manufacturer’s instructions. An
33
empty pCMV6-XL4 vector without insert was used as a transfection control. Cells were
harvested after 72 hours of transfection and further analyzed the protein and
miR-181a-1 expression by Western blot and qRT-PCR, respectively. The entire
experiment was carried out three times with similar results.
2.2.15. Lentiviral construct and infection
The sequence of MIR181A1 was PCR amplified from human bone marrow mononuclear
cells and then cloned into vector pLKO_TRC001, which contains a PGK-puromycin
acetyltransferase insert, and labeled as pLKO.1.181A1. An empty TRC1 vector,
pLKO.1.Null-T, was used as a negative control. Production and infection of lentivirus
followed the protocol from the National RNAi Core facility. Briefly, lentivirus was
generated by transfection of HEK-293FT cells using the transfection reagent
TransIT-LT1. The vectors used were pLKO.1.null-T or pLKO.1.181A1, and the packaging vectors were pMD.G and pCMVΔR8.91. Single infection of REH cells and
two sequential infections of primary ALL cells with concentrated lentiviral particles
were carried out in 24-well plates. Lentivirus-infected cells were selected by adding
puromycin (2 µg/mL) to the culture medium and collected after screening for a week.
34
2.2.16. Luciferase reporter assay
The luciferase activity assay was performed using the Dual-Luciferase Reporter Assay
System (Promega). A 678-bp fragment of the RUNX1 3′ UTR containing a binding site
for miR-181a (UGAAUGU) was cloned into the XbaI site at the distal end of the
luciferase reporter gene of pGL3-promoter vector. This construct was used to transiently
transfect HEK-293FT cells with Lipofectamine 2000 together with pRL-TK Renilla, a
transfection control used to calibrate the luciferase activity, and pLKO.1.181A1
(miR-181a-expressing vector) or pLKO.1.Null-T (negative control for
miR-181a-expressing vector). A mutated version of the binding sequence (AGAUCUG)
containing a BglⅡsite was obtained by site directed mutagenesis and was used as the
target site control. Cells were lysed, and the luciferase activity was measured 48 hours
after transfection.
2.2.17. Statistical analyses
In miRNA profiling analysis, to avoid low abundant expression issue, miRNA with
coefficient of variation (CV) < 0.2 was removed in the first step. In the second step, the student’s t test was used to evaluate different miRNA expression between
ETV6/RUNX1-positive (n=10) and ETV6/RUNX1-positive (n=40) groups. Finally, in
order to control multiple testing issue, false discovery rate method was performed to
35
adjust p value obtained from student’s t test (67). Data are represent the means ± SE or
± SD as indicated in the figure legends. The two-tailed unpaired Student’s t test or
ANOVA were used to test the difference between groups for continuous variables. For categorical data, Fisher’s exact was performed to test the difference between groups.
Calculation methods of P values were denoted in the figure legends or bottom of tables.
All tests were two-tailed and P values <0.05 were considered significant.
36
Chapter 3. Results
3.1 ETV6/RUNX1 directly downregulates MIR181A1
3.1.1 ETV6/RUNX1-associated miRNA expressions in clinical samples
Extensive miRNA profiling was carried out on the diagnostic samples of a cohort
of 50 childhood B-ALL patients in the cooperation with National Taiwan University
microarray core facility (Figure 2). Ten ETV6/RUNX1-positive and forty
ETV6/RUNX1-negative cases were included in this cohort. The clinical features
including gender, onset age, WBC count at first diagnosis, and distribution of risk
groups showed no statistical difference between the ETV6/RUNX1-positive and
ETV6/RUNX1-negative samples (Table 3). Because ETV6/RUNX1 retains the
DNA-binding ability of RUNX1, the fusion protein acts as a dominant-negative
repressor to downregulate RUNX1 target genes. Therefore, a reduction of specific
miRNAs in ETV6/RUNX1-positive samples compared with ETV6/RUNX1-negative
samples was evaluated. Seventeen miRNAs (let-7a, let-7b, miR-19a, miR-130b,
miR-155, miR-181a-1, miR-181c, miR-181d, miR-195, miR-221, miR-222,
miR-30e-3p, miR-342, miR-423, miR-425, miR-660, miR-92) were significantly
downregulated in ETV6/RUNX1-positive ALL samples (Table 4). According to
miRBase database (a searchable database of published miRNA sequences and
37
annotation, http://mirbase.org/) these miRNAs can be classified into 13 miRNA clusters.
In the use of CoreBoost_HM we predicted the transcriptional start site (TSS) of these
miRNA clusters, moreover, we identified that 92% of the ETV6/RUNX1-associated
miRNA clusters (12/13) possess the potential RUNX1 binding sites (TGT/cGGT) in the
region between upstream 4 kb and downstream 1 kb of their TSS (Table 5).
Of these miR-181a-1, which is derived from the 3′ arm of its precursor miRNA,
hsa-mir-181a-1 (Figure 1), had the most significant P-value and showed a remarkable
4-fold reduction (Table 4). The decreased expression of miR-181a-1 in
ETV6/RUNX1-positive leukemias was validated in another cohort of B-ALL primary
blasts analyzed by real-time quantitative RT-PCR. The relative miR-181a-1 levels in
ETV6/RUNX1-positive and ETV6/RUNX1-negative samples of validation set were 0.14
± 0.08 and 0.06 ± 0.03, respectively (Figure 3A). We also measure miR-181a level,
which is derived from the 5′ arm of hsa-mir-181a-1 and was not included in the miRNA
expression profile. Expression level of miR-181a was not associated with ETV6/RUNX1
status (Figure 3B), however, was positive correlated with miR-181a-1 expression level
in patient samples (Figure 3C).
3.1.2 siRNA knockdown of ETV6/RUNX1 up-regulates miR-181a-1 expression
Whether ETV6/RUNX1 regulates miR-181a-1 expression was further assessed by
38
knockdown of ETV6/RUNX1 and overexpression of ETV6/RUNX1 in cell lines. We
conducted siRNA-mediated knockdown of ETV6/RUNX1 in t(12;21)-positive REH cells,
which express the ETV6/RUNX1 fusion protein (Figure 4). A mixture of two
ETV6/RUNX1-specific siRNAs (siE/R), which target the fusion region of ETV6/RUNX1,
was used to suppress ETV6/RUNX1 expression (65). As a transfection control, we used
a nonfunctional siRNA (siRNA-S) that had no effect on ETV6/RUNX1 expression (66).
Compared with siRNA-S, both mRNA and protein of ETV6/RUNX1 were significantly
decreased by ~40% and ~35% after knockdown with siE/R without interfering the
RUNX1 protein expression (Figure 5). Further examination showed that miR-181a-1
levels increased significantly in REH cells that were treated with siE/R (221 ± 50.8%)
but not in those treated with siRNA-S (117 ± 13%) (Figure 6).
3.1.3 Overexpression of ETV6/RUNX1 down-regulates miR-181a-1 expression
Overexpression of ETV6/RUNX1 or RUNX1 protein was carried out by
transfection of pCMV6-vectors expressing ETV6/RUNX1 or RUNX1 into HEK-293FT
cells in the use of an empty vector as a transfection control (Figure 7). Compared with
the empty vector, while expression of RUNX1 protein increased the miR-181a-1 level
(114 ± 8.1%), expression of ETV6/RUNX1 protein significantly resulted in mR-181a-1
reduction (75 ± 14.6%) (Figure 8). This overexpression experiment confirmed that
![TraditionalMLCalgorithmsmainlytacklethebatchMLCproblem,wheretheinputdataarepresentedinabatch[24,28].Nevertheless,inmanyMLCapplicationssuchase-mailcategorization[22],multi-labelexamplesarriveasastream.Onlineanalysisistherefore dimensionreducermotivatedbyma](data:image/gif;base64,R0lGODlhAQABAIAAAP///wAAACH5BAEAAAAALAAAAAABAAEAAAICRAEAOw==)