Functional Characterization of Mirnas in Tumour Biology

Functional characterization of miRNAs in tumour biology

Keerthana Krishnan

M. Biotech

A thesis submitted for the degree of Doctor of Philosophy at

The University of Queensland in 2014

Institute for Molecular Bioscience Abstract

MicroRNAs are non-coding, negative regulators of gene expression which act either by repressing protein translation or via mRNA degradation. They have been shown to play biologically significant roles in various processes like cell differentiation, proliferation, apoptosis and development in humans as well as other model organisms. miRNAs direct the wide repertoire of normal biological processes by down-regulating the expression of their target genes. Deregulation of miRNAs has been shown to be associated with various human diseases such as cancer. Although several oncomirs have been identified to date, functional studies to understand their mode of action have been hampered by the lack of tools to accurately predict and validate the gene networks repressed. This thesis aims to address this issue by using a combination of high- throughput technologies and subsequent experimental validation using cell based assays.

Chapter two takes a closer look at miR-182, which at the time this study was initiated was shown to be deregulated in several cancers, but its biological role and target networks in the context of breast cancer was relatively unknown. We used biotinylated synthetic miRNA to pull-down its endogenous mRNA targets to reveal that it disrupts key pathways underlying tumorigenesis, and subsequently confirmed its clinical relevance in human breast cancers. Chapter three takes a similar approach to identify the biologically relevant targets of miR-139, a novel breast cancer oncomir. The role of miR- 139 is more akin to being a potential tumour suppressor supporting our data and published datasets where its expression is frequently downregulated in human breast cancers. In Chapter four, we take a more global approach where using next-generation sequencing technology we identify miRNAs which show dynamic expression across various phases of the cell cycle, another biological process typically disrupted during tumorigenesis. Using online datasets, we try to identify if there is a significant correlation between these oscillating miRNAs and cancer, and also possible regulators of their expression.

These approaches facilitate the identification of novel oncomirs and subsequent characterization of their biologically relevant targets using context-dependent cell

i models. In addition these studies show that these miRNAs achieve their functional output by targeting multiple genes, which belong to the same pathway, adding to the existing notion of concomitant suppression. Together, this leads to a better understanding of the miRNA-mediated disruption to specific molecular processes underlying tumorigenesis. Such studies are imperative to explore the potential of oncomirs as possible prognostic, diagnostic or therapeutic tools.

ii Declaration by author

This thesis is composed of my original work, and contains no material previously published or written by another person except where due reference has been made in the text. I have clearly stated the contribution by others to jointly-authored works that I have included in my thesis.

I have clearly stated the contribution of others to my thesis as a whole, including statistical assistance, survey design, data analysis, significant technical procedures, professional editorial advice, and any other original research work used or reported in my thesis. The content of my thesis is the result of work I have carried out since the commencement of my research higher degree candidature and does not include a substantial part of work that has been submitted to qualify for the award of any other degree or diploma in any university or other tertiary institution. I have clearly stated which parts of my thesis, if any, have been submitted to qualify for another award.

I acknowledge that an electronic copy of my thesis must be lodged with the University Library and, subject to the General Award Rules of The University of Queensland, immediately made available for research and study in accordance with the Copyright Act 1968.

I acknowledge that copyright of all material contained in my thesis resides with the copyright holder(s) of that material. Where appropriate I have obtained copyright permission from the copyright holder to reproduce material in this thesis.

iii Publications during candidature

Peer-reviewed papers

Thiagarajan RD, Cloonan N, Gardiner BB, Mercer TR, Kolle G, Nourbakhsh E, Wani S, Tang D, Krishnan K, Georgas KM, Rumballe BA, Chiu HS, Steen JA, Mattick JS, Little MH, Grimmond SM. Refining transcriptional programs in kidney development by integration of deep RNA-sequencing and array-based spatial profiling. BMC Genomics, 2011. 12: p. 441.

Cloonan N, Wani S, Xu Q, Gu J, Lea K, Heater S, Barbacioru C, Steptoe AL, Martin HC, Nourbakhsh E, Krishnan K, Gardiner BB, Wang X, Nones K, Steen JA, Matigan N, Wood DLA, Kassahn KS, Waddell N, Shepherd J, Lee C, Ichikawa J, McKernan K, Bramlett K, Kuersten S and Grimmond SM. MicroRNAs and their isomiRs function cooperatively to target common biological pathways. Genome Biol, 2011. 12(12): p. R126.

Krishnan K, Steptoe AL, Martin HC, Wani S, Nones K, Waddell N, Mariasegaram M, Simpson PT, Lakhani SR, Gabrielli B, Vlassov A, Cloonan N, Grimmond SM. MicroRNA-182-5p targets a network of genes involved in DNA repair. RNA, 2013. 19(2): p. 230-42

Krishnan K, Steptoe AL, Martin HC, Pattabiraman DR, Nones K, Waddell N, Mariasegaram M, Simpson PT, Lakhani SR, Vlassov A, Grimmond SM, Cloonan N. miR-139-5p is a regulator of metastatic pathways in breast cancer. RNA, 2013. 19(12): p. 1767-80

Pattabiraman DR, McGirr C, Shakhbazov K, Barbier V, Krishnan K, Mukhopadhyay P, Hawthorne P, Trezise AEO, Grimmond SM, Papathanasiou P, Alexander WS, Perkins AC, Levesque JP, Winkler IG, Gonda TJ. Interaction of c-Myb with p300 is required for the induction of acute myeloid leukemia (AML) by human AML oncogenes. Blood, 2014. 123 (17): p. 2682-2690

Martin HC, Wani S, Steptoe AL, Krishnan K, Nones K, Nourbakhsh E, Vlassov A, Grimmond SM and Cloonan N. Imperfect centered miRNA binding sites are common and can mediate functional repression of target mRNAs. Genome Biol, 2014. 15(3): p. R51

Book Chapters

Krishnan K, Wood DLA, Steen JA, Grimmond SM and Cloonan N. “Tag Sequencing”. Epigenetic Regulation and Epigenomics – Advances in Molecular Biology and Medicine. 2012 Wiley-Blackwell. p. 145-166

Publications included in this thesis

Contributor Statement of contribution Author Krishnan K (Candidate) Designed experiments (70%) Performed experiments (70%) Data Analysis (70%) Wrote the paper (60%) Author Steptoe AL Performed experiments (10%) (Stable cell line generation with candidate) Author Martin HC Data Analysis (15%) (Biotin pulldown microarray) Author Wani S Performed experiments (10%) (Biotin pulldown) Author Nones K Performed experiments (7%) (Microarrays) Author Waddell N Designed experiments (5%) Author Mariasegaram M Performed experiments (3%) (Provided RNA from patient samples) Author Simpson PT Wrote the paper (2%) Author Lakhani SR Wrote the paper (2%) Author Gabrielli B Designed experiments (5%) Wrote the paper (5%) Author Vlassov A Provided reagents Author Cloonan N Designed experiments (15%) Data Analysis (15%) Wrote paper (25%) Author Grimmond SM Designed experiments (5%) Wrote paper (6%)

vi Krishnan K, Steptoe AL, Martin HC, Pattabiraman DR, Nones K, Waddell N, Mariasegaram M, Simpson PT, Lakhani SR, Vlassov A, Grimmond SM, Cloonan N. miR-139-5p is a regulator of metastatic pathways in breast cancer. RNA, 2013. 19(12): p. 1767-80: Incorporated as Chapter three

Contributor Statement of contribution Author Krishnan K (Candidate) Designed experiments (85%) Performed experiments (70%) Data Analysis (75%) Wrote the paper (80%) Author Steptoe AL Performed experiments (10%) (Stable cell line generation with candidate) Author Martin HC Data Analysis (15%) (Biotin pulldown microarray) Author Pattabiraman DR Performed experiments (10%) (Western Blots) Author Nones K Performed experiments (7%) (Microarrays) Author Waddell N Designed experiments (2%) Author Mariasegaram M Performed experiments (3%)(Provided RNA from patient samples) Author Simpson PT Wrote the paper (2%) Author Lakhani SR Wrote the paper (5%) Author Vlassov A Provided reagents Author Grimmond SM Designed experiments (5%) Wrote paper (5%) Author Cloonan N Designed experiments (8%) Data Analysis (10%) Wrote paper (8%)

Krishnan K, Wood DLA, Steen JA, Grimmond SM and Cloonan N. “Tag Sequencing”. Epigenetic Regulation and Epigenomics – Advances in Molecular Biology and Medicine. 2012 Wiley-Blackwell. p. 145-166: Incorporated as a section within Chapter one

Contributor Statement of contribution Author Krishnan K (Candidate) Wrote the paper (15%) Author Wood DLA Wrote the paper (15%) Author Steen JA Wrote the paper (5%) Author Grimmond SM Wrote and edited paper (5%) Author Cloonan N Wrote and edited paper (60%)

vii Contributions by others to the thesis

All the work presented in this thesis was performed at the Queensland Centre for Medical Genomics (QCMG) and would not have been possible without the resources provided by the centre, headed by Professor Sean Grimmond. All the microarray and sequencing data generation followed by data analysis were performed in-house (by personnel mentioned where appropriate). Special thanks to our system administrators Darrin Taylor and Scott Wood for help with data retrieval and storage.

Statement of parts of the thesis submitted to qualify for the award of another degree

None.

Acknowledgements

If nothing, my PhD years have taught me that my life prior to them was too simple, easy and sometimes boring! All the pain, joy, failures and successes during this time have been valuable life experiences and made me more appreciative of efforts that go into each and every scientific publication.

I would like to firstly thank my supervisor Sean, who gave me this incredible opportunity and through it all was extremely supportive. I am very grateful for all the help he has extended towards this thesis and outside of it. Thanks Sean for your mentorship! Everyone in Group Grimmond has contributed to this thesis, be it by listening to my sad stories or running around the campus to keep healthy or walking across to get coffee, special thanks to Shiv, Katia, Anita, Senel, Deb, Rathi, Milena, Daz, Dave, Nic and everyone else for helping to maintain my sanity. Thanks to Amanda and everyone in the postgrad team for helping with the milestones and all the paperwork. Finally, big thanks to my other supervisor, Nicole! This thesis would not have been possible without her, her guidance when I was in Brisbane and constant correspondence after I moved to Boston, were significant contributions towards this completion and are much appreciated. You got me into this Nicole, but you have also seen me through the entire journey, thanks for being a great teacher!

Thanks to friends in Brisbane who are more like family: Ravi, Sushmi, Vijay uncle, Neela aunty, Madhan, Sudarsanan, Preethi, Sree & Anu. I miss you guys and hopefully will get to see all of you very soon! Friends in Boston who supported (or nagged) me all of last year about thesis: Pavan, Shivang, Nitin, Aravind, Veena, Karthik, Shivangi, Nandini, Mona, Somnath, Maha, Neela, Christine & Susan, thank you! Huge thanks to my in-laws for their constant support and words of encouragement.

Appa, Amma and Krithika, you guys gave me the best opportunity for my career by sending me to Brisbane and since then, every successful step I have taken is thanks to you. Amma, you were not given the best of opportunities to study or work, so very early on you had decided and have since strived to give them to me. My success is yours and I dedicate this thesis to you.

Finally, my husband Diwakar, who still naively believes and waits when I say ‘our lives will get better after this’…Challenges are always going to be part of our lives but your patience, support and love will carry us through. This would not have been possible without you; my sanity depends on your calmness. Thanks for being there, always!

Keywords miRNA, breast cancer, biotin pull-down, target gene, microarray, DNA repair, metastasis, cell cycle, sequencing

Australian and New Zealand Standard Research Classifications (ANZSRC)

ANZSRC code: 111201 Cancer Cell Biology, 40%

ANZSRC code: 060405 Gene Expression, 40%

ANZSRC code: 060114 Systems Biology, 20%

Fields of Research (FoR) Classification

FoR code: 1112 Oncology and Carcinogenesis, 40%

FoR code: 0601 Biochemistry and Cell Biology, 40%

FoR code: 0604 Genetics, 20%

Table of Contents Abstract ...... i Declaration by author ...... iii Publications during candidature ...... iv Publications included in this thesis ...... vi Contributions by others to this thesis ...... viii Statement of parts of thesis submitted to qualify for the award of another degree ...... viii Acknowledgements ...... ix Keywords ...... x Australia and New Zealand Standard Research Classification ...... x Fields of Research Classification ...... x Table of contents ...... xi List of figures and tables ...... xiii List of abbreviations ...... xiv Chapter 1: Introduction ...... 1 1.1 Non-coding RNAs ...... 2 1.2 miRNAs ...... 4 1.2.1 miRNA biogenesis ...... 4 1.2.2 Role of AGO proteins in gene silencing ...... 7 1.2.3 Mechanism of gene silencing ...... 7 1.2.4 Target Recognition principles ...... 10 1.2.5 miRNA expression profiling ...... 16 1.2.6 miRNA diversity: Isomirs ...... 18 1.2.7 Role of miRNAs in cancer ...... 18 1.2.8 Tag Sequencing ...... 29 1.3 Hypotheses and Aims ...... 63

Chapter 2: miR-182-5p targets genes underlying the DNA Damage Repair pathway in breast cancer ...... 64 2.1 Summary ...... 65 2.2 Journal article: miR-182-5p targets genes underlying the DNA Damage Repair pathway in breast cancer ...... 67 Chapter 3: miR-139-5p is a potential tumour suppressor of breast cancer by targeting genes underlying metastasis related pathways ...... 80 3.1 Summary ...... 81 3.2 Journal article: miR-139-5p is a potential tumour suppressor of breast cancer by targeting genes underlying metastasis related pathways ...... 83 Chapter 4: Exploration of cell cycle associated miRNAs ...... 97 4.1 Introduction ...... 98 4.2 Materials and Methods ...... 100

4.3 Results ...... 103 4.3.1 Synchronization of cancer cell lines and verification of cell populations in different phases ...... 103 4.3.2 Small RNA sequencing of synchronized cancer cell lines for identification of miRNAs with dynamic expression across cell cycle ...... 105 4.3.3 Cell cycle associated transcription factors are not significantly enriched in the promoter regions of miRNAs that are dynamic across the cell cycle phases ...... 110 4.3.4 Expression analysis of associated transcription factors in the matching microarray data ...... 111 4.3.5 Expression of cell cycle dynamic miRNAs in human cancer samples ...... 113 4.3.6 miR-3607-5p and its predicted target cohort with a potential role in cell cycle and cancer ...... 117 4.4 Discussion ...... 119 Chapter 5: General discussion and future directions ...... 124 5.1 Summary of findings ...... 125 5.2 miRNAs and mRNA target interactions ...... 126 5.2.1 miRNAs act through multiple mRNA targets to exert a phenotypic effect ....126 5.2.2 miRNAs target multiple pathways to exert a biological effect ...... 129 5.2.3 The functional output of a miRNA is determined by the expression of its target genes ...... 131 5.3 MicroRNAs in breast cancer ...... 134 5.3.1 Molecular heterogeneity of breast cancer ...... 134 5.3.2 miRNA expression signatures in breast cancers ...... 134 5.3.3 miRNA signatures could predict prognosis ...... 136 5.3.4 Using miRNAs to direct therapy choices ...... 137 5.4 Conclusions and Future Directions ...... 139 References ...... 142 Appendix ...... 165 1. Supplementary Material for Chapter 2 ...... 166 2. Supplementary Material for Chapter 3 ...... 206 3. Supplementary Material for Chapter 4 ...... 228

List of Figures Chapter One: Figure 1.1:miRNA biogenesis pathway ...... 6 Figure 1.2: mRNA Target Recognition principles ...... 11

Chapter Four: Figure 4.1: Synchronization and cell cycle analysis of HeLa and MCF-7 cells ...... 104 Figure 4.2: Expression of cell cycle markers in synchronized HeLa and MCF-7 cells ...106 Figure 4.3: Comparison of dynamically expressed miRNAs to previous study ...... 108 Figure 4.4: Correlation between transcription factors and miRNA expression ...... 112 Figure 4.5: Analysis of miR-3607 in TCGA tumour samples ...... 116 Figure 4.6: Ingenuity Pathway Analysis of predicted targets of phasic miRNAs ...... 118

List of Tables Chapter One: Table 1.1: Known sub-classes of ncRNAs and their characteristics ...... 2 Table 1.2: List of commonly used target prediction software and their properties ...... 12 Table 1.3: List of widely studied miRNAs associated with human cancer ...... 19 Table 1.4: Known oncomirs in breast cancer with their validated targets and functional relevance ...... 24

Chapter Four: Table 4.1: List of miRNA with dynamic expression between two cell cycle phases in MCF7 and HeLa cells ...... 109 Table 4.2: Dynamically expressed miRNAs in HeLa and MCF7 and their expression across human cancers in The Cancer Genome Atlas dataset ...... 114

List of Abbreviations

AGO Argonaute ALL Acute Lymphocytic Leukemia AML Acute Myeloid Leukemia ASCII American Standard Code for Information Interchange BER Base excision repair CAGE Cap Analysis of Gene Expression CDK Cyclin-dependent kinase CDKi Cyclin-dependent kinase inhibitor cDNA Complementary DNA CDS Coding DNA sequence Chip-seq Chromatin-immunoprecipitation Sequencing CLASH Cross linking, ligation, and sequencing of hybrids CLIP-seq Cross-Linking Immuno-Precipitation sequencing CLL Chronic Lymphocytic Leukemia CML Chronic Myeloid Leukemia CNV Copy-number variation DCIS Ductal carcinoma in situ DDR DNA damage response DLBCL Diffuse Large B-cell lymphoma Dm-ChP DNA mediated chromatin pull-down DMEM Dulbecco's modified eagle's medium DNA Deoxy-Ribo nucleic acid DSB Double strand break EDTA Ethylenediaminetetraacetic acid EGF Epidermal growth factor ENCODE Encyclopedia of DNA Elements ePCR Emulsion PCR ER Estrogen receptor EST Expressed sequence tags FACS Fluorescence-activated cell sorting FBS Fetal bovine serum FC Fold change FDR False discovery rate GSEA Gene set enrichment analysis HEK293T Human embryonic kidney 293 T-antigen Her2 Human epidermal growth factor receptor HITS-CLIP High-throughput sequencing of RNA isolated by crosslinking immunoprecipitation HR Homologous recombination HRP Horseradish peroxidase ICGC International Cancer Genome Consortium IDC-NST Invasive ductal carcinoma of no special type

ILC Invasive lobular carcinoma IMC Invasive mucinous carcinoma IPA Ingenuity pathway analysis IRES Internal ribosome entry site kDa Kilo-dalton lncRNA Long non-coding RNA MeDIP-seq Methylated DNA ImmunoPrecipitation sequencing miRISC miRNA-induced silencing complex miRNA micro RNA miRNA-seq miRNA sequencing MMR Mismatch repair MRE miRNA response elements mRNA messenger RNA ncRNA Non-coding RNA NHEJ Non-homologous end joining ORF Open reading frame PABP polyA-binding protein PAR-CLIP Photoactivatable-ribonucleoside-enhanced crosslinking and immunoprecipitation PARP Poly ADP-ribose polymerase PASR Promoter-associated small RNA PBS Phosphate buffered saline PCR Polymerase chain reaction piRNA PIWI-interacting RNA PR Progesterone receptor Pre-miRNA Precursor miRNA Pri-miRNA Primary miRNA PROMPT Promoter upstream transcripts PTP Picotiter plate qRT-PCR Quantitative Realtime Polymerase Chain Reaction RAD-seq Restriction site Associate DNA sequencing RNA Ribo nucleic acid RNAi RNA interference RNAPII RNA Polymerase II RNA-seq RNA sequencing rRNA Ribosomal RNA RT Reverse transcription SILAC Stable isotope labelling with amino acids in cell culture siRNA Small interfering RNA SMS Single molecule sequencing snoRNA Small nucleolar RNA SNV Single nucleotide variants SOLiD Sequencing by Oligonucleotide Ligation and Detection SSB Single strand break TBST Tris Buffered Saline with Tween 20 TCGA The Cancer Genome Atlas

TF Transcription factor TFBS Transcription factor binding site tiRNA Transcription initiation RNA TMM Trimmed Mean of M-values TNBC Triple negative breast cancer TPM Transcripts per million TRBP Transactivation-responsive RNA binding protein tRNA Transfer RNA TSS Transcription start site TSSa-RNA Transcription-start site associated RNA UTR Untranslated region

CHAPTER ONE

Introduction

1. Introduction

1.1 Non-coding RNAs

The central dogma of molecular biology is that genetic information, stored in DNA is translated through to protein via RNA molecules. Since the 1960s, when studies first described the genetic code, it has been assumed that all genes encode proteins and collectively they were sufficient to cover all functions related to the cell, organismal development and homeostasis [1]. More recently, this theory has been refined to accommodate the findings that some genes can encode non-coding functional RNAs as an alternative to proteins. [2]. Indeed it is becoming apparent that more complex genomes e.g. human and mouse have similar number of proteins as Caenorhabditis elegans (C.elegans) and biological complexity correlates more with the size of the non-protein coding fraction of the genome [3].

Global analysis of the mouse genome revealed over 35000 non-coding transcripts mapping to ~10000 distinct loci [4] and ~98% of the human genome could transcribe long and short non-protein coding RNAs (ncRNAs) [5]. Several studies have now shown the regulatory potential of ncRNAs in key cellular processes underlying normal development and diseases in complex organisms [6, 7]. ncRNAs can be classified based on their size and functions as outlined in Table 1

Table 1.1: Known sub-classes of ncRNAs and their characteristics Name Characteristics References Small interfering RNAs ~21-22nt long, Dicer dependant biogenesis, [8-10] (siRNAs) associates with Argonaute (AGO) proteins to bring about sequence specific silencing of genes. Involved in gene and transposon regulation and viral defense microRNAs (miRNAs) Small RNAs ~21nt long (discussed in detail later)

2 PIWI-interacting RNAs 24-30 nt long, Dicer independent, associate [9, 10] (piRNAs) with the PIWI clade of AGO proteins and known functions include silencing transposons and DNA methylation Transcription initiation ~18nt long, map within -60 to +120nt of [11] RNAs (tiRNAs) transcription start sites (TSSs), could be involved in chromatin modification and regulation of protein recruitment in transcription initiation Small nucleolar RNAs Variable length, involved in guiding rRNA [12-14] (snoRNAs) methylation and pseudouridylation. Targets besides rRNAs have recently been discovered, suggesting gene regulatory functions Promoter-associated ~20-200nt, capped and found at the 5’ regions [11, 15] small RNAs (PASRs) of protein/non-coding genes, strongly associated with highly expressed genes and regions of RNAPII binding. Transcription-start site ~20–90nt, found within -250 to +50 of TSSs, [11, 16] associated RNAs strongly associated with highly expressed (TSSa-RNAs) genes and regions of RNAPII binding. Promoter upstream <200nts, found within -500 to -5kb of TSSs, [17] transcripts (PROMPTs) transcription is bidirectional and is associated with gene activity. May have gene regulatory functions. Long non-coding RNAs 200nt – 100kb in length with functional roles in [18-20] (lncRNAs) development through genetic and molecular processes including epigenetic regulation, cellular differentiation, cell cycle control, transcription, splicing and translation

3 1.2 miRNAs

miRNAs are conserved, short non-coding RNAs, ~ 22 nucleotides long with regulatory functions across several organisms [21]. lin-4 was the first miRNA to be discovered in C. elegans and was thought to be specific to this species [22, 23] . The second miRNA let-7 was discovered in C. elegans but was found to be conserved amongst worms, flies and humans [24-26]. This discovery led to an extensive search for miRNAs across different organisms and in the last decade several studies have addressed the questions of miRNA biogenesis, target recognition principles, identification of miRNA targets, miRNA expression profiling techniques and their roles in development and diseases. An overview of each of these topics will be presented in this introduction chapter. Since the focus of the thesis is to study the role of miRNAs in human diseases, all subsequent topics are discussed in the context of mammalian miRNAs unless otherwise specified.

1.2.1 miRNA biogenesis

The canonical biogenesis pathway for most animal miRNAs is summarized in Figure 1.1. Precursor molecules (pri-miRNAs) can originate either from full-length transcripts or spliced by-products deriving from intronic regions of protein coding genes. Multiple pri-miRNAs can be encoded in an individual transcript and are transcribed by RNA Polymerase II [8, 27]. miRNA biogenesis begins with the cleavage of the hairpin pri-miRNAs at the base by the DiGeorge syndrome critical region gene 8 (DGCR8) /Drosha (family of RNAse III) complex. DGCR8 recognizes the proximal 10bp of the stem and anchors Drosha, resulting in cleavage of ~55-70nt long pre-miRNA molecules [28]. miRNAs of intronic origin bypass this step and undergo normal splicing processes or a Drosha-only dependant pathway to produce the pre-miRNA molecules [29, 30].

Next, Exportin-5 and Ran-GTP recognizes the pre-miRNA and export them to the cytoplasm. Dicer then partners with the Transactivation-responsive (TAR) RNA binding protein (TRBP) to cleave the pre-miRNA to ~22nts long small RNA duplexes. Post- processing, one strand (the mature sequence) of the miRNA duplex gets incorporated into miRNA-induced silencing complex (miRISC) and the other (passenger) strand gets

4 degraded. Recent studies have shown that the passenger strands are not always degraded and can get loaded into miRISC to function as miRNAs [31-34]. Certain exceptions to the this ‘canonical biogenesis pathway’ occur in mammals where Argonaute 2 (AGO2), which has endonuclease activity, cleaves the 3’ arm of some pre- miRNAs to generate AGO2-cleaved precursor miRNA (ac-pre-miRNA) [35]. These intermediate molecules subsequently get processed by Dicer to form miRNA duplexes (Figure 1.1). However in the case of pre-miR-451, the ac-pre-miR-451 has been shown to be processed in a Dicer-independent manner to form miR-451 [36]. A more recent study has also shown supporting evidence for another 11 miRNAs undergoing this type of biogenesis [37]. The mature miRNAs in the miRISC interact with AGO proteins and glycine- tryptophan protein of 182 kDa (GW182) proteins, the key functional and effector molecules to mediate target repression [38].

Nucleus

RNA Pol II Transcription

pre-miRNA pri-miRNA Cytoplasm 5’ 3’ 5’ 3’ Drosha Exportin 5 Dicer Processing Transport Processing 3’ 5’

5’ 3’ 3’ 5’ 3’ 5’ 5’ AAAAAAAA 33’’ mRNA RNA-Induced Silencing Asymmetrical miRNA-mRNA interactions Complex (RISC) Unwinding miRNA duplex

Figure 1.1: miRNA biogenesis pathway: Illustration of miRNA biogenesis pathway showing the primary precursor molecules (pri-miRNA) which are processed by Drosha to ~70 nt long precursor hairpins (pre- miRNA) and get exported to the cytoplasm by Exportin 5. Dicer then cleaves the pre-miRNA hairpins to generate ~20 bp long miRNA duplexes, one strand of which gets incorporated into RNAinduced silencing complex (RISC) to bind to target mRNAs, leading to gene silencing.

6 1.2.2 Role of AGO proteins in gene silencing Mammals contain four Ago proteins (AGO1-AGO4), which have three conserved domains capable of interacting with miRNAs: PAZ, MID and PIWI domains [39, 40]. All four AGO proteins are involved in miRNA-mediated gene silencing, however, only AGO2 with an enzymatically active RNAse-H like PIWI domain also functions in RNAi [41]. The differential expression of individual AGO proteins in a tissue/cell specific manner and their ability to repress targets at different efficacy suggests differential roles for these proteins [42]. This is strengthened by studies showing a much stronger effect on miRNA mediated gene silencing in HEK293 cells with knockdown of AGO2 compared to the other AGO proteins [43] and embryonic lethality in AGO2 knockout mice [41]. The GW182 proteins act downstream of AGO proteins and are important for miRNA- mediated gene silencing [44]; there are three known mammalian GW182 proteins including TNRC6A, B, and C. The interaction between GW182 and AGO proteins in the miRISC is essential for target repression by miRNAs [45, 46].

1.2.3 Mechanism of gene silencing

miRNAs in association with the AGO proteins in the miRISC, direct target repression by either mRNA degradation, deadenylation or translational repression, due to imperfect base-pair complementarity between miRNA and target mRNA. In few cases perfect base-pair complementarity has been shown to lead to mRNA cleavage by AGO proteins [47-49]. An overview of studies attempting to decode the mechanism employed by miRNAs to efficiently silence their targets is given below.

1.2.3.1 Translational repression

Translation from mRNA to proteins, in eukaryotes, requires several processes to work harmoniously for efficient translation: (i) initiation (at the correct codon); (ii) elongation of the polypeptide chain, and (iii) termination and release from the ribosome. These processes require a 5’ capped and 3’ poly-A tailed mRNA; the eukaryotic translation initiation factor 4G (eIF4G) which circularizes the mRNA by interacting with the cap binding protein eIF4E (associated with the cap); and the cytoplasmic poly(A)-

7 binding protein (PABPC, associated with the poly(A) tail). This circularization activates the translation initiation process and protects the mRNA from degradation [50, 51]. Studies thus far suggest interference of miRNAs on the roles of the eIF4F complex and PABPC as possible mechanisms for translational repression.

The first studies to investigate miRNA mediated target repression in C.elegans showed lin4 (miRNA) to repress translation of its targets lin-14 and lin-28 with no apparent change in their mRNA abundance [52, 53]. The mRNA was found to be localized in polysomes suggesting translation repression to have occurred post-initiation. Later studies in HeLa cells supported this finding and suggested degradation of nascent polypeptide chain to be the causal mechanism for translational repression [54]. Petersen et. al. suggested repression happened before synthesis of the polypeptide and was due to ‘ribosome drop-off’ during the elongation process [55]. Translational repression at initiation has also been shown to be the case, where the mRNA targets do not sediment in larger polysomal fractions of sucrose gradients and are instead found in lighter fractions with fewer ribosomes [56]. Replacing the cap structure with Internal Ribosome Entry Site (IRES) showed abrogation of translational repression, suggesting that the cap and its binding protein eIF4E are key targets of this mechanism [56, 57]. Further studies in human and mouse cell extracts support this model where using IRES led to efficient translation of target genes [58, 59]. More recently this has gained support in model organisms like drosophila [60] and zebrafish [61] where using different approaches, the authors concluded that miRNA targeting occurs at initiation rather than post-elongation or as an early termination. These studies have reasonably shown that the primary mechanism used by miRNAs to silence their targets is translational repression during the initiation process.

1.2.3.2 mRNA deadenylation and degradation

Evidence for this type of target repression comes from several transcriptome-wide studies showing inverse correlation in the expression of miRNAs and their target mRNAs [62-65]. Subsequent studies with depletion of crucial factors, part of the miRNA machinery, like Dicer or Ago [43] and models which over-express [62-66] or knock-

8 down miRNAs [62, 63, 67] of interest have suggested mRNA degradation as one possible mechanism for miRNA mediated gene silencing.

Direct cleavage of mRNA targets has been reported [49], however it is a rare phenomenon in animal cells where the miRNAs can function with only partial complementarity to their mRNA targets. In such instances, they usually employ shortening of the poly-A tail followed by a 3’ to 5’ decay activated by the exosome or a 5’ to 3’ degradation preceded by decapping of the mRNA catalysed by XRN1 [68-71]. In Drosophila depletion of key factors from the cellular mRNA decay pathway led to increased expression of miRNA targets, emphasizing its widespread role in target repression by miRNAs [70, 72-74]. mRNA deadenylation alone can lead to repression by interfering with PABPC association with the tail, hence disabling the circularization of the mRNA and subsequent translation [59]. Translationally repressed mRNAs can localize in cytoplasmic foci called P-bodies, which can act as the destination for the final steps involving mRNA degradation, aided by the accumulation of proteins involved in deadenylation, decapping and mRNA degradation [75, 76]. A study involving miR-223 knockout mice showed much greater correlation between mRNA and protein changes, and suggested mRNA degradation to be the principle cause of protein reduction [63]. Recently Guo et. al. showed that about 84% of target repression at the protein level can be attributed to decrease in mRNA abundance at both 12 and 32 hrs post transfection of miRNAs, suggesting a close-link between mRNA deadenylation and translational repression [65]. In contrast, a study utilizing both microarrays and a proteomic approach to identify targets of miR-143 showed more targets regulated through translational repression compared to mRNA decay [77]. It has also been shown that depending on the time point chosen for the analysis many targets could be regulated only at the protein level [62]. Another possible explanation for the widespread reduction of target mRNA in response to miRNA expression is the over representation of transcription factors in its target cohort [78]. This could lead to a secondary effect of mRNA suppression by the loss of activating transcription factors. Another factor confounding these studies is the presence of mRNA isoforms with UTRs of different length which may or may not be able to undergo miRNA mediated silencing [79]. This is more relevant in the case of

9 cancer cells which show widespread shortening of 3’ UTRS compared to normal cells [80]. Recent studies in drosophila [60] and zebrafish [61] have shown reduction in (a) ribosomes bound to miRNA targets or (b) protein expression respectively to occur 2 hours before any change in the mRNA abundance is seen, again suggesting a miRNA mediated translational repression. Taken together, translational control of targets genes is the dominant direct-mechanism underlying miRNA mediated gene silencing. However, the effects of miRNAs are made far-reaching and efficient by directly targeting expression of primary targets e.g. transcription factors which then subsequently leads to changes in widespread mRNA abundance, resulting in the desired phenotype.

1.2.4 Target Recognition principles

In animals, the majority of miRNAs are partially complementary to their mRNA targets. While this increases their ability to target multiple genes, it also makes target prediction and identification more complicated. Studies to date have elucidated several ‘target recognition principles’, which follow certain rules of base-pairing depending on locations both in miRNAs and the target genes. The most common type of site is the ‘seed’ region in the 5’ end of the miRNA involving perfect base pairing between nucleotides 2 and 7 (6mer) [21]. This type of site recognition can be sub divided into: 8mer, 7mer-m8, 7mer- A1 and 6mer sites, where 8mer sites have an adenine at position 1 of the target site and base pairing at position 8. 7mer-A1 and 7mer-m8 sites either have an adenine at position 1 or base-pairing at position 8 respectively [81]. It is speculated that a component of RISC recognizes and enables interaction between the adenine on the target site and the first base of the miRNA, which in general has been shown to have a strong bias towards being a Uracil (U) [82]. This type of site recognition was initially thought to be both necessary and sufficient for effective miRNA-mediated silencing [83].

Subsequent studies have since shown other types of site recognition models including ‘3’- supplementary’ or ‘3’- compensatory sites’ to either enhance target recognition or to overcome imperfect base-pairing at the seed region [21]. The 3’- supplementary sites require at least 3-4 perfect base-pairing from nucleotides 13 to 16 in the miRNA and this type of site conservation predict targets with greater specificity [81]. Mismatches or bulges in the seed region can also be ‘compensated’ by the 3’-

10 complementary sites spanning nucleotides 13 to 17 with at least nine perfect base-pairing [21]. However these types of site recognition seem relatively rare and experimental evidence to support their functionality in the miRNA-pathway are relatively few [21, 48]. Most recently, ‘centered sites’ beginning at position 3,4 or 5 of the miRNA and spanning 11 nucleotides of perfect base-pairing to the target have been reported [48]. The different models of miRNA-mRNA base-pairing are outlined in Figure 1.2.

Figure 1.2: mRNA Target Recognition principles: (A) Seed site involving perfect base-pairing at the 5’ end of the miRNA (B) 3’ Supplementary and (C) 3’ Compensatory sites involving base-pairing at the 3’ end of miRNA to enhance target recognition (D) Centered sites starting at position 3,4 or 5 spanning 11 nucleotides of perfect base-pairing. Figure adapted from Brennecke et al., (2005) PLoS Biology.

1.2.4.1 mRNA Target site location

Most studies to date report target sites in the 3’ UTR of the target gene, leading to the hypothesis that target genes have longer 3’ UTRs as opposed to house- keeping genes which avoid miRNA-mediated repression by having shorter 3’ UTRs [84]. Target sites are found about 15-20 nts away from the stop codons in short 3’ UTRs [81] and are found at both ends in longer (>2000nts) 3’ UTRs [85]. Alternative polyadenylation and alternative splicing are classic mechanisms employed by the cell in specific contexts (e.g. cancer) to disrupt miRNA-mediated target repression [80, 86, 87].

11 Although functional miRNA sites are more prevalent in the 3’ UTR, evidence suggesting their presence in 5’ UTRs and coding regions leading target repression have been found [88-91]. This number would no doubt increase when computational tools evolve to account for CDS and 5’ UTRs into their prediction algorithms.

1.2.4.2 Target prediction tools

Initial target prediction tools generally yielded non-overlapping targets [21], which limited subsequent analysis and biological validations in specific cellular contexts. The computational tools have since evolved, predominantly due to the general consensus of importance attributed to ‘seed-sites’. Although each tool employs its own method to reduce false-positive rates and increase accuracy, the most commonly used tools follow certain general guidelines: complementary bases to the miRNA seed sequence, evolutionary conservation of the target region, and the thermal stability of the miRNA- mRNA duplex [63, 92-94]. Some algorithms like Target Scan [82, 95, 96] assign ‘context’ scores depending on the location of the predicted target site (proximity to stop codon) and other features flanking in the mRNA. A table summarizing commonly used miRNA target prediction software is shown (Table 2).

Table 1.2: List of commonly used target prediction software and their properties No. Name Clades Criteria for prediction Refere nce 1 MiRanda Human, Sequence match, conservation and [97] Drosophila, thermal stability. zebrafish 2 RNAhybrid Mammals, Fly Analyzing secondary structure of [98] miRNA/mRNA duplex 3 TargetScan Mammals, Thermodynamic model of RNA [82, and nematodes, fish, interaction and sequence alignment 95, 96] TargetscanS fly analysis 4 DIANA- Mammals, fly Single miRNA-recognition elements [99]

12 microT 5 PicTar Mammals, fly, Stringent seed pairing for at least one [100] nematodes of the sites for the miRNA, site number, overall predicted pairing stability 6 Target gene Drosophila Sequence search with evaluation of [101] prediction the predicted miRNA–target at EMBL heteroduplex structures and energies 7 PITA Top Mammals, Sequence search with site [102] nematodes, fly accessibility and thermodynamic model of interaction 8 EIMMo Mammals, fish, Seed pairing, likely hood of [85] fly, nematodes conservation. 9 mirWIP C. elegans Site accessibility, pairing stability, [103] total free energy of miRNA-target hybridization. 10 RNA22 Mammals, fly, Matches to sequence patterns, [104] nematode sequence stability, does not take conservation 11 FASTH Human, mouse Free energy score thresholds, perfect [105] complementarity to the 5’ end of miRNA, minimum base pairing at the 3’ end, and other optional criteria

1.2.4.3 miRNAs target genetic networks

It is now becoming increasingly clear that each miRNA has a large number of target mRNAs that it can regulate. It has been proposed that a large portion of miRNA targets are mRNAs that fall under similar pathways, ablation of which ultimately contribute to the outcome [94]. One such miRNA that has been shown to regulate different mRNA targets that function along the same pathway is miR-24. This antiproliferative miRNA was shown to target a network of genes whose function was highly

13 enriched for DNA repair and cell cycle progression such as MYC, E2F2 and VHL [106]. Subsequent studies showed that the miRNA regulated H2AX, a protein that plays a key role in the double-strand break response [106]. Hence miR-24 plays a role in the regulation of DNA repair and cell cycle progression through the regulation of various mRNAs that are key players in these cellular processes.

Another such example is miR-34, a p53-regulated miRNA that has been shown to be a suppressor of Wnt signaling through the repression of WNT1, WNT3, LRP6, AXIN2, β-catenin and LEF1, all of which are components of the canonical Wnt signaling cascade [107]. p53 has been shown to inhibit the epithelial-to-mesenchymal transition program and subsequent metastasis of cancers [108] whereas Wnt signaling is known to promote the process [109]. Hence, p53 presumably exerts its anti-metastatic functions through the transactivation of miR-34, which then acts through the inhibition of various components of the Wnt signaling cascade, inhibiting their translation. Hence to fully understand the role of a particular miRNA, a thorough study of the target gene network that it regulates through a robust assay is an essential first step. The mechanism by which a particular miRNA is able to regulate the expression of numerous target mRNAs is still poorly understood, perhaps resulting from our inability to accurately identify a major portion of targets of a particular miRNA. Lal et al. observe that almost all target genes regulated by miR-24 lack the presence of a seed sequence in their 3’-UTRs although they do contain sequences that are highly complementary to the miRNA, thus resembling non-canonical miRNA response elements (MREs) [106]. Moreover, independent studies have demonstrated the presence of complementary sequences within the ORFs of genes to be as effective in miRNA repression, with many well-known tumor suppressors such as RB1 harbouring such sequences in the coding region complementary to miR-181 [110]. Hence, the search for binding sequences in the 3’UTRs of genes as the sole predictor of miRNA targets could be misleading and provide an incomplete picture of miRNA biology. As better algorithms and lab-based protocols are developed to identify miRNA targets, a better picture of how miRNAs are able to regulate large networks of genes will evolve, enabling better understanding of the properties of these molecules. This will also be greatly aided by the techniques used to both biologically identify and validate miRNA targets.

1.2.4.4 Identification and validation of putative targets

Forward and reverse genetics were the first techniques used to identify miRNAs and their target genes [111]. Since then, microarray analysis of samples after overexpression or knock down of a miRNA has widely been used to identify targets whose expression are inversely correlated to the miRNA and was first used to show the effect of miR-124 in the brain [66]. These studies are typically hampered by limitations due to probe-design and are not able to distinguish between direct targets or indirect targets (genes down regulated by transcription factors which are direct targets of the miRNA). It will also overlook translationally repressed targets, which show no apparent change at the transcript level. Immunoprecipitation of AGO followed by assaying the ‘co-precipitated’ mRNA targets by microarrays or next-generation sequencing technology has helped identify targets potentially regulated at the level of mRNA degradation or translational repression [112-116]. This approach is typically specific to one AGO protein and can potentially lead to false negatives when other AGO proteins are present in miRISC [117]. Another limitation is the non-specific binding of AGO proteins to cognate mRNAs in the cell lysis leading to false positives [114].

Proteomic approaches including Stable isotope labelling with amino acids in cell culture (SILAC) in combination with mass spectrometry have been used to gauge the global changes in protein expression after manipulation of miRNA expression [62, 63], which supported the importance of seed sites. Target validations subsequent to these large-scale analyses are typically carried out using reporter assays where the predicted binding site is cloned downstream of a reporter gene and assayed for the effect of miRNA regulation to that site. This method can be labour intensive, expensive and most importantly loses context-specificity. To overcome some of these limitations, new approaches are being employed to identify direct targets of miRNAs. One such approach is the high-throughput sequencing of RNA isolated by crosslinking immunoprecipitation (HITS-CLIP), where to recapitulate endogenous interactions, ultraviolet radiation is used to crosslink RNA and associated RNA binding proteins prior to immunoprecipitation, followed by sequencing to assay the targets [116, 118]. Photoactivatable-ribonucleoside-enhanced crosslinking

15 and immunoprecipitation (PAR-CLIP) has optimized the UV crosslinking by incubating cultured cells with a photoactivatable nucleoside such as 4-thiouridine and is capable of indicating the site of targeting [112]. Both studies found comparable number of sites in the 3’ UTR, CDS and the 5’ UTR [119]. Biotinylation, the process of attaching biotin covalently to a protein or nucleic acid, has [94] been used extensively in molecular biology owing to its high affinity to streptavidin, which can be used to specifically isolate the labeled molecules. Pull-downs using biotinylated proteins has been successfully used to identify protein-protein interactions [120] and genome-wide binding sites for transcription factors using bioChIP- seq [121] or to perform DNA mediated chromatin pull-down (Dm-ChP) [122]. Pull- downs using biotinylated miRNAs followed by microarrays or high-throughput sequencing has also been used to identify miRNA targets [37, 123]. Recent studies have demonstrated the utility of this technique to specifically identify miRNA targets with higher accuracy than routinely used bioinformatics algorithms. The major advantage of this system is the highly specific nature of the biotin-streptavidin binding which ensures minimal non-specificity and noise [124, 125]. This robustness of the interaction ensures higher fidelity compared to other techniques such as HITS-CLIP, which call for the availability of a ChIP grade antibody against a miRISC component and a challenging multistep protocol [94]. Additionally, such protocols are carried out by pull-down of a generic miRISC complex protein in cells that overexpress a particular miRNA with the assumption that majority of the miRISC component will be co-localized specifically to the miRNA in question. This provides room for non-specificity especially in the case of isomirs where distinguishing targets of a miRNA and its isomiR is difficult using these approaches. The use of biotinylated miRNA on the other hand ensures the identification of targets specifically associated to the miRNA or the isomiR in question [37].

1.2.5 miRNA expression profiling

miRNA expression profiling across sample cohorts are essential and have greatly aided in elucidating their role in normal developmental processes and disease states. Their small size and ability to generate isomiRs (discussed later) pose certain challenges in accurate quantification of their abundance; however certain approaches are robust,

16 reproducible and routinely used in global miRNA expression profiling studies. miRBase [126, 127] is both the repository and valuable resource for such studies, constantly updated when novel miRNAs are identified. Version 18 of miRBase (current at time of writing) has a total of 18,226 entries of hairpin precursor miRNAs expressing 21,643 mature miRNA products derived from a broad range of plant and animal species.

Reverse transcription of miRNA to cDNA followed by real time PCR provides relative quantification of the specific miRNA interrogated; is highly sensitive and requires very small amounts of input RNA. This method cannot be used to identify novel miRNAs and scalability is limited [128]. Two most commonly used commercial products include ABI’s specific stem-loop primers, used to generate cDNA, subsequently amplified by primers specific to mature miRNA, and Exiqon’s LNA based primers which adds a polyA tail to the mature miRNA before it is converted to cDNA and amplified using miRNA specific primers. Microarrays provide a cost-effective, high-throughput approach to simultaneously profile hundreds of miRNAs. Typically this approach involves fluorescent labeling of miRNAs in samples, followed by hybridization to DNA probes attached to glass slides or in some cases beads [129]. Estimating the fluorescence associated with specific probes gives an indication of relative abundance of specific miRNAs. Limitations include inability to identify novel miRNAs, less specificity and lower dynamic range compared to qRT-PCR or sequencing [128]. Sequence similarity of more than 75% results in cross hybridization leading to noise [130]. The probe design depends on known miRNA sequences and generally lags behind new releases from miRBase thereby excluding several families of miRNAs. Another popular approach widely used now is next generation sequencing which is sensitive, and becoming increasingly very cost-effective. This topic is discussed in detail in the book chapter titled “Tag Sequencing” (attached below). Briefly, the advantages of this method include its non-reliance on previously known sequences of miRNAs, which can be used to identify novel miRNAs. Sequencing based studies have shown it to be highly sensitive, reproducible and quantitative [131]. Most importantly it can very accurately distinguish between miRNAs of similar sequence or even isomiRs [37].

1.2.6 miRNA diversity: Isomirs

Next-generation sequencing technology has enabled the identification of miRNA variants called ‘isomiRs’ [132] or non-canonical miRNAs, which are generated from a common hairpin pre-miRNA molecule. When first observed, isomiRs were thought to be artifacts of library generation, sequencing or poor alignment [133-135], degraded RNA [136] or by products of inefficient processing by Drosha/Dicer [137, 138]. In more recent studies however, a more functional role for the variants are emerging; suggesting a model where their expression allows a more fine-tuned modulation of target repression [37].

IsomiRs differ from their canonical counterparts in a wide variety of ways including: substitutions, insertions or deletions, 3' end non-templated additions, and 5' and/or 3' cleavage variations [37]. 3’ isomiRs can result from a variation in Drosha/Dicer processing [31, 137-140] and have been shown to be developmentally regulated suggesting a functional role [141]. 5’ isomiRs potentially represent a new ‘seed region’ and hence a new set of targets. Although this could explain the reduced expression of 5’ isomiRs relative to 3’ isomiRs in some sequencing data [31, 142] there are contradictory reports where 5′ isomiR expression was comparable to the canonical miRNAs [31, 138, 142]. Research into isomiRs is still at its infancy and requires a more targeted approach to unravel their specific biological functions in a context dependent manner.

1.2.7 Role of miRNAs in cancer

miRNAs are gene regulators which play key roles in development and homeostasis as well as in human diseases. For the purpose of this thesis, an overview of miRNAs frequently deregulated in human cancers and their contribution to the pathogenesis of the disease is outlined below.

Cancer is an accumulation of mutations in the genome and epigenome leading to activation of oncogenes or silencing of tumour suppressors. The disease is characterized by a multi-step process involving the: (1) Uncontrolled proliferation of aberrant cells developing into a primary tumour, (2) Escape of tumour cells into the systemic circulation, and (3) Establishment in a distant tissue followed by outgrowth of secondary

18 tumours (metastasis). High-throughput analyses have reported altered miRNA expression in almost all tumours investigated till date, signifying the major role they play in oncogenesis. The term “oncomir” has been commonly used in reference to miRNAs that play a role in tumourigenesis. The following sections will discuss oncogenic and tumour suppressive miRNAs and the role they play in the pathogenesis of various cancers, a list of well characterized oncomirs is shown in Table 3.

Table 1.3: List of widely studied miRNAs associated with human cancer miRNA Activity Cancer Type Reference 17-92 cluster Oncogene B cell lymphoma; lung cancer; [143-147] colon cancer; multiple myeloma, medulloblastoma 155 Oncogene CLL, non-Hodgkin’s lymphoma, [148-153] Hodgkin’s disease, Burkitt’s lymphoma, breast, lung, colon, pancreatic cancers. 21 Oncogene Breast, lung, gastric, cervical, [154-162] prostate, colorectal, head and neck cancers, glioblastoma 221 Oncogene Hepatocarcinoma, CLL, melanoma, [163-171] glioblastoma, breast, thyroid, lung and prostate cancer 372/373 Oncogene Testicular tumours [172] 106b-93-25 Oncogene Colon, pancreatic, gastric and [173] cluster prostate cancer, multiple myeloma LIN28 Oncogene CML, hepatocellular carcinoma, [174] Wilms tumour 151 Oncogene Hepatocellular carcinoma [175] 10b Oncogene Breast, pancreatic cancer, glioma [58, 176, 177] 9 Oncogene Breast cancer, renal cell carcinoma, [178-180] colon cancer

19 27a Oncogene Breast cancer [181] 517c, 520g Oncogene Neuro-ectodermal brain tumours [169] 15a, 16-1 Tumour CLL, DLBCL, prostate, pancreatic [169, 182-187] Suppressor cancer, multiple myeloma 34 Tumour Colon and pancreatic cancers, [188-191] Suppressor Burkitt’s lymphoma Let-7 Tumour Lung, breast, colon, gastric, prostate [192-195] Suppressor cancer, CLL 29 Tumour Colon, breast, lung cancer, AML, [196-198] Suppressor cholangiocarcinoma 145 Tumour Breast cancer [199] Suppressor 221, 222 Tumour Erythroblastic leukemia [200] Suppressor 200 Tumour Colon, breast, lung, bladder cancers [201-204] Suppressor 148a Tumour Pancreatic and gastric cancer [205, 206] Suppressor 124a Tumour Glioblastoma, ALL, gastric cancer [207-209] Suppressor 31 Tumour Breast cancer, T-cell leukemia, [210-212] Suppressor colon carcinoma 137 Tumour Glioblastoma, squamous cell [213, 214] Suppressor carcinoma 129-2 Tumour Gastric, endometrial cancer [215, 216] Suppressor

1.2.7.1 Oncogenic miRNAs

One of the main indications of a potential oncogenic role for a miRNA is overexpression in tumours. miR-155 and the miR-17-92 cluster were the first set of

20 miRNAs to be causatively associated with human cancer, the coding regions of both being amplified in B-cell lymphomas [217, 218]. Using Eμ-miR-155 transgenic mice, overexpression of this miRNA was demonstrated to be causative of high-grade lymphoma, the first evidence that dysregulation of a single miRNA can lead to cancer [219]. Follow up studies have shown that miR-155 is overexpressed in a range of haematopoietic malignancies as well as lung and breast cancer [148, 150, 152, 220]. More recent studies into its oncogenic nature have revealed that it is a target for epigenetic silencing by BRCA1 through the recruitment of HDAC2, which deacetylates its promoter [221].

The miR-17-92 cluster consists of six miRNA hairpins – miR-17, miR-18a, miR- 19a, miR-20a, miR-19b and miR-92a-1. The cluster has been found to be overexpressed in various other solid tumours including those of the breast, colon, lung and prostate [152]. Importantly, a causative role for this cluster was established when overexpression in B cells was shown to induce a lymphoproliferative disorder in mice. Studying the mechanism of tumourigenicity revealed that the cluster can be transactivated by Myc and promotes cell cycle progression and cell proliferation through the regulation of the E2F transcription factors [222, 223]. A paralog of miR-17-92 is miR-106b-25 cluster, which has also been shown to act as an oncogene in gastric, prostate and pancreatic neuroendocrine tumours by cooperating with MCM7 in the inhibition of PTEN [173, 224, 225]. Expression profiling of both tumour and normal tissues has evolved as one of the most effective ways of identifying novel miRNAs that could play a role in tumourigenesis. miR-31 was found to be overexpressed in malignant lung cancers, its knockdown leading to repressed growth of the tumour [226]. Its tumourigenicity was attributed to its ability to repress the LATS2 and PPP2R2A tumour suppressors. Expression of another miRNA, miR-27a, was found to be elevated in a panel of breast cancer cell lines, contributing to their oncogenicity through the regulation of its target genes ZBTB10 and Myt-1 [181]. These target genes repress the Sp family of transcription factors that contribute to the proliferative and angiogenic phenotype of cancer cells. The miR-106b family was also identified through microarray profiling, its expression levels being highly correlated with the expression of cell cycle genes [227]. It was shown to

21 directly target CDKN1A (p21), promoting exit from the G1 phase and entry into the S- phase. A member of the miR-106b family, miR-17-5p, was shown to promote cell cycle progression through the modulation of over 20 oncogenes and tumour-suppressor genes, including MAPK9/JNK2 [228]. Recent studies have also focused on the role of miRNAs in tumour progression and metastasis. miR-10b was identified as an oncomir highly expressed in metastatic breast cancer cells, conferring properties of invasiveness and metastasis [58]. It was shown to inhibit translation of HOXD10 upon induction by Twist, leading to increased expression of RHOC. Further study showed that administration of miR-10b antagomirs to mice bearing highly metastatic tumours effectively suppressed lung metastasis, proving to be a promising candidate for therapeutic intervention [229]. Similarly miR-9, which is upregulated in breast cancer cells, was shown to directly inhibit expression of E-cadherin, leading to activation of beta-catenin signaling which results in increased cell motility and invasiveness [179]. Activation of beta-catenin also contributed to upregulation of VEGF signaling, promoting tumour angiogenesis.

1.2.7.2 Tumour-suppressor miRNAs

miR-15a and miR-16-1 were the first miRNAs shown to be lost in human cancer, the 13q14 locus being frequently lost in CLL [184]. Further study showed that these two miRNAs target the BCL2 oncogene known to be causative of follicular lymphoma [184, 185]. Let-7, a founding member of the miRNA family identified in C. elegans was also the first miRNA shown to target an oncogene, with all three RAS genes harbouring let-7 complementary sites in their 3’UTRs [192]. Accordingly, expression levels of let-7 were reduced in lung tumour tissue, reciprocal to that of RAS expression. Through the above two studies it became clear that the loss of miRNAs could contribute to tumourigenesis through deregulation of cellular oncogenes. Further study has shown that miR-15a and 16-1 are also lost in prostate cancer and multiple myeloma [185, 230, 231] whereas the let-7 family is lost in lung and breast cancers [193, 232].

The miR-29 family consists of three members: miR-29a, b and c, which are lost in various haematological malignancies such as chronic lymphocytic leukemia and acute myeloid leukemia, as well as solid tumours such as breast cancers [196, 197, 233]. They

22 were shown to inhibit apoptosis through repression of the anti-apoptotic gene Mcl-1 [197]. miR-129 is also involved in the activation of p53 (TP53) through the repression of p85a, the regulatory subunit of PI3 kinase and CDC42, both of which are negative regulators of p53 (TP53) [198]. Through the targeting of DNMT3A and B, miR-29 relieves epigenetic silencing of tumour suppressor genes such as FHIT and WWOX, which are repressed by promoter methylation during the pathogenesis of non-small cell lung cancer [234]. Although most studies have revealed a tumour suppressive role for this miRNA, it has been shown to act as a context-dependent oncogenic miRNA. Through the repression of TTP (tristetraprolin), miR-29a has been shown to promote the epithelial-to- mesenchymal transition and metastasis in EpRas cells [235]. Several miRNAs have been implicated in the inhibition of cancer progression and metastasis. Expression of miR-31 was inversely correlated with metastatic potential and its overexpression in aggressive breast cancer cells sufficient to suppress metastasis [210]. Its effects were shown to be mediated by several targets, including RHOA, integrin alpha5 (ITGAV) and radixin (RDX) [236]. The suppression of just these three genes could mimic suppression of miR-31. Moreover, activation of miR-31 function in established metastasis, even in the form of a brief induction, leads to its regression [237]. The previous section addressed the role of miR-31 as an oncogenic miRNA whereas other studies discussed here establish it as a tumour suppressive miRNA. This points out to a dual, context-dependent role for miR-31 in tumourigenesis. The miR-200 family also plays tumour suppressive roles in various cancers. The family consists of miR-200a, miR-200b, miR-200c, miR-141 and miR-429, which were shown to inhibit the epithelial-to-mesenchymal transition (EMT) and maintain the epithelial state by targeting the transcriptional repressors Zeb1 and Zeb2 [201]. p53 has been shown to bind to the miR-200c promoter and transactivate it, preventing the EMT program and metastasis [238]. As described above, most studies suggest correlation between miRNAs and cancer. This is based on expression profiling in tumour samples but does not always imply a causative role in the disease initiation or progression. Such large-scale profiling analyses also possess certain inherent limitations including: inter-study variability in terms of sample size and variable validation approaches [239]. However it is increasingly

23 becoming evident that miRNAs do have a fundamental role to play in tumorigenesis [240, 241], suggesting it is important to tease out the molecular mechanism underlying the role of miRNAs in cancer. This can be achieved by identifying deregulation of clusters of miRNAs and by identifying the functional network of genes they modulate in a given context. Using these approaches we can then decipher the specific mechanism underlying the role of the deregulated miRNA in a given cancer type.

1.2.8 Role of miRNAs in breast cancer

Breast cancer, a major health burden around the world, is one of the leading causes of death amongst women. It has often been referred to as a collection of breast diseases with diverse histopathologies, genetic and genomic variations, distinct expression profiles and clinical outcomes. This diversity impedes our understanding of its biology and is a major barrier to both identifying novel therapies and improving existing therapies for treatment and prevention [242].

Contributing to this molecular heterogeneity is the differential expression of miRNAs across the various clinical subgroups [199]. The expression signatures of miRNAs were also shown to correlate with histopathological features including tumour size, nodal involvement, proliferative, and invasive abilities of the cancer cells [199]. Shown in Table 4 are known oncomirs of breast cancer and the functional relevance of their validated targets.

Table 1.4: Known oncomirs in breast cancer with their validated targets and functional relevance miRNA_name Function Targets Related cellular event Reference miR-21 Oncogenic RHOB, PTEN, Cell migration, elongation, [158, 243- TPM1, invasion and apoptosis 246] miR-10b* Oncogenic HOXD10, Metastasis, cell motility, [247, 248] TIAM1 invasiveness

24 miR-155 Oncogenic RhoA, Cell survival, [249-251] FOXO3a, chemoresistance, cell miR-373 Oncogenic CD44 Invasion, metastasis [252] miR-520c Oncogenic CD44 Invasion, metastasis [252] miR-27a Oncogenic ZBTB10, Cell cycle dysregulation, [181, 253] Myt1, proliferation, angiogenesis, miR-206 TS ER-alpha† Estrogen-receptor mediated [254, 255] proliferation miR17-5p TS HBP-1, AIB1, Cell migration, invasion, [256-258] cyclinD1 proliferation, miR-125a,b TS HuR, ERBB2, Anchorage dependant [259, 260] ERBB3 growth, migration, invasion miR-200c TS TCF-8 (ZEB1) EMT, invasion [261] let-7 family TS ER-alpha, H- Proliferation, apoptosis, [262, 263] Ras, HMGA2 self-renewal and miR-31 TS Rho-A, Fzd3, Metastasis [210] ITGA5, RDX miR-335 TS SOX4, Metastasis [264] PTPRN2, miR-27b TS CYP1B1 Modulation of anticancer [265] drugs miR-126 TS IRS-1 Cell growth [266] miR-185 TS SIX-1 Proliferation, apoptosis [267] miR-145 TS Mucin1, Cell invasion, metastasis, [268-270] RTKN, ER- cell growth, apoptosis

25 miR-205 TS ERBB3, Proliferation, anchorage- [271, 272] VEGF-A, independent growth, miR101 TS EZH2 Proliferation, invasion, [273] H3K27-trimethylation

Several clinical studies reported on the possibility of using miRNAs as prognostic markers by highlighting the differential expression of certain miRNAs in large patient sample cohorts [274-276]; however given the inherent heterogeneity of the disease this has proved to be difficult. This highlights the scope for identification of novel oncomirs in breast cancer and the subsequent validation of specific biological processes they disrupt. The following section will focus on three processes that are commonly deregulated during tumorigenesis, which are also the focus of the results chapters forming part of this thesis.

1.2.8.1 DNA Damage and Repair

Maintenance of genome integrity is one of the major cellular functions that ensures proper replication of DNA and its transmission across generations, deregulation of which can lead to the occurrence of mutations and aberrations that ultimately result in tumorigenesis. DNA damage can result from physical agents such as ionizing radiation (IR) and ultraviolet light (UV), or by chemical means from alkylating agents, DNA crosslinking agents and cigarette smoke, all of which can induce both single-strand (SSBs) and double-strand DNA breaks (DSBs). DNA repair mechanisms counteract such damage through processes such as homologous recombination (HR) or nonhomologous end joining (NHEJ) in the case of DSBs [277] or mismatch repair (MMR) and base excision repair (BER) to repair mismatched or damaged bases [278, 279]. This repair is carried out by the action of several enzymes including nucleases, ligases, kinases and phosphatases, amongst others that are recruited upon sensing of damage, a process known as the DNA damage response (DDR). DDR is a signal transduction pathway that senses DNA damage and initiates the cascade of signals that sets out to counter the threat to the cell, primarily mediated by three kinases, ATM, ATR and DNA-PK and poly (ADP-

26 ribose) polymerase (PARP) family members [280]. More recently, miRNAs have also been identified to play a major role in this process, as will be described in detail in the first results chapter of this thesis.

1.2.8.2 Metastasis

Over 90% of cancer-related deaths are a result of metastasis and not the primary tumor itself [281], with the ability to form metastasis being an indicator of the prognosis of the cancer. The invasion-metastasis cascade is the process by which tumor cells are able to escape from the primary tumor site by the acquisition of invasive and migratory properties, intravasate into the bloodstream and lodge in distal sites where they would ultimately colonize and lead to the outgrowth of metastases [282]. In the case of carcinomas, individual cancer cells acquire invasive properties through a change in cell state termed the epithelial-to-mesenchymal transition characterized by a loss of epithelial properties including cell junctions, accompanied by the gain of mesenchymal properties that enables their translocation from the primary tumor [283]. Several signaling pathways have been implicated in the ability of an epithelial cell to acquire mesenchymal properties that enable it to metastasize including TGFb, Wnt, Hedgehog, NOTCH and NFkB [284- 287]. More recently, several miRNAs have been implicated in processes underlying metastasis as will be discussed in the second results chapter.

1.2.8.3 Cell Cycle The cell cycle is a fundamental cellular process that ensures DNA replication and cell division. Given its high relevance to cell division and proliferation, its deregulation has severe implications for normal cell functions and is one of the hallmarks of cancer [288]. It is composed of the initial G1 resting phase where the cell synthesizes proteins required for DNA synthesis, regulated by cyclinD1-CDK4/6 complexes [289]. The transition between these two phases, known as the G1-S checkpoint ensures that cells with damaged DNA do not progress further and is tightly controlled by cyclinE-CDK2 complexes, which upon entry into S-phase changes to a cyclinA-CDk2 complex. The S- phase where DNA is replicated is followed by the G2 phase where the cell, through

27 protein synthesis, prepares for mitosis. The transition to the M phase, known as G2-M transition is initially regulated by cyclinA-CDK1 and later cyclinB1-CDK1 complexes and is a major checkpoint for genotoxic stress ensuring damaged DNA is not passed onto the next generation [289]. Cyclin-dependent kinase inhibitors (CKIs; the INK4 and CIP family) bind to cyclins or cyclin-CDK complexes and negatively regulate progression of the cell cycle and incidentally, happen to be important tumor suppressors that curb the process of tumorigenesis that results from uncontrolled cell cycle progression. The complexity of this process requires the involvement of a large number of regulators, which also includes miRNAs as will be expanded upon in the third results chapter of this thesis.

28 Tag Sequencing

Keerthana Krishnan, David L. A. Wood, Jason A. Steen, Sean M. Grimmond, and Nicole Cloonan*

Queensland Centre for Medical Genomics, Institute for Molecular Bioscience, The University of Queensland, Brisbane, QLD, Australia

*Email: [email protected]

Keywords

Tag sequencing The process of generating short fragments of nucleic acid sequence derived from longer nucleic acids. Originally performed using standard capillary sequencing, most tag sequencing is now performed by massively parallel sequencing.

PCR: Polymerase chain reaction A molecular biology technique used to replicate and amplify DNA fragments using DNA polymerase, nucleotides, primers, and the modulation of reaction temperature.

Library In this context, library refers to a collection of DNA fragments, each flanked by a pair of defined nucleic acid sequences (adaptors) from which molecular biology protocols can be anchored.

Genome The full complement of heritable nucleic acids from an organism, and includes all genes (protein coding and non-coding genes). In most cases the genome is DNA, however some viral genomes are RNA based. Unless otherwise specified, the genome contains mitochondrial DNA too.

- - 29 Transcriptome The full complement of RNA transcripts expressed in a particular cell population or a particular tissue, in a particular environment at a particular time. While an organism usually has only one genome, it may have many transcriptomes. The transcriptome is considered to be the active portion of the genome.

Epigenome The full complement of heritable material that are is not nucleic acid. This includes modifications to DNA (methylation), and the modifications and positioning of proteins that can affect the stability and accessibility of the genome to transcription.

- - 30 Abstract

Technological advances in sequencing are revolutionizing the way genome scale biological research is performed – leading to an explosion of applications made possible by the massive scale of data generated and a similarly massive decrease in cost. Specific mention of throughputs, lengths, or costs will become out of date very quick- ly as the technology is still rapidly evolving, however the concepts and protocols surrounding the technology have largely stabilized. This article will explore the technology, analysis, and common applications of tag sequencing, and how these are impacting the major facets of nucleic acid research: genomics, epigenomics, and transcriptomics.

1 Library construction

The process of creating libraries for massive scale tag sequencing will determine the kinds of information that can be extracted at the time of analysis. Although each strategy differs in the specifics, there are two very general steps for every library construction (i) fragment the nucleic acids to an appropriate size; (ii) capture and amplify the fragments between adaptors of a defined sequence.

1.1 Fragment libraries

By far the simplest library preparation protocol is a "fragment" library, where DNA is sheared down to fragments that are slightly larger than the desired tag length. This strategy provides the best use of DNA, as almost the entire DNA used for library generation is available for sequencing (Figure 1A). Data from this style of library is typically used in applications where an appropriate reference genome has already been assembled. De novo assembly applications using fragment data alone are confounded by repetitive regions

- - 31 longer than the read length, computationally intensive, and are often hampered by the high error rates in massive-scale sequencing [1]. There are three main approaches to adding adaptors to genomic fragments: (i) adaptors are added via PCR (Figure 1A); (ii) two unique double stranded molecules (designated A and B) are ligated to the fragment simultaneously, and only molecules both an A and B adaptor are selected by PCR amplification (Figure 1B); or (iii) the use of Y shaped adaptors which allow only the generation of AB molecules (Figure 1C). There are obvious advantages to the use of Y-adaptors, which result in significantly more molecules suitable for sequencing (if we assume that both A and B adaptors ligate equally, half of the molecules in an AB ligation protocol will be either AA or BB) and allow libraries to be generated from smaller amounts of material. Most recently, the use of Y shaped adaptors that include Taqman probe binding sites have been described for use in pyrosequencing (see section 2.2.1), which allows for the specific titration of library molecule concentrations required for bead generation [2]. Generating libraries via PCR is generally considered less favorable than ligation due to the difficulty in ensuring even coverage across the genome, the difficulty in generating tight insert size distributions, and the creation of PCR artifacts which require more extensive purification [3].

1.2 Paired-end libraries

"Paired-end" libraries are fragment-style libraries where short tags are read from both ends of a single fragment (typically 200- 300nt long). A critical step in this protocol is size selecting the DNA fragments, so the physical distance between the paired tags is known. Once paired-reads are aligned to a reference genome, the distance between the tags can be compared with the expected distance, and discordant mapping distances identify variations in genomic structure (see section 4.2 and Figure 7). The vast majority of genome re-sequencing data comes from paired-end libraries, due to their relatively low requirement for input DNA and the simplicity of the library preparation.

- - 32

- - 33 Figure 1. Sequencing library construction. (A) Chromosomal DNA can be left intact, or sheared. PCR primers containing known adaptor sequence and random hexamers (depicted as 6 N’s) are used to PCR amplify genomic fragments ready for sequencing. Optionally, sequence specific primers can be used, depending on the application, although this may require substantial optimization for multiplex PCR, or many individual PCR reactions. (B) Standard adaptor libraries are prepared by fragmenting the DNA into small fragments slightly larger than the desired tag length. Adaptors of a known sequence are ligated to each end, which are used for subsequent PCR amplification and priming for the sequencing reactions. For a simple fragment run, tags are usually sequenced from the 5’ end of the library (black arrowhead). For a paired-end run, tags are sequenced from the 5’ (black arrowhead) and the 3’ (white arrowhead) end of the library. (C) Y-adaptor libraries are prepared as in (B), except a single Y adaptor is used rather than two distinct double-stranded adaptors. This eliminates the situation where two of the same adaptor ligate to the same DNA fragment and not amplified during the PCR stage. Additionally, as each strand now contains a forward primer (black arrowhead), the number of DNA fragments present in the library that can be sequenced directly are again increased. (D) Mate-pair libraries are usually constructed from much larger fragments than fragment or paired-end libraries (1-10kb vs 50-500nt). Adaptors are added, and the longer molecules are then circularized by ligation. The ends of the DNA fragments (dotted lines) are then removed from the remaining DNA (by restriction digest or by nick translation) producing a linear molecule of each end separated by the first adaptors. Finally, the sequencing adaptors are ligated to each end to allow for amplification. To generate mate-pair data, tags are generated from the 5’ end (black arrowhead), and the 3’ end (white arrowheads), although the direction of the second read will depend on both the protocol and sequencing platform employed.

1.3 Mate-pair libraries

The "mate-pair" strategy is less commonly used for large scale genomic sequencing, but offers greater sensitivity to detect structural variations such as translocations and inversions. The protocol involves breaking the target molecule into large fragments (often between 1-10kb, allowing tag pairs to straddle repetitive regions) and rather than sequence the entire fragment, each end is captured and sequenced as a pair of short tags (Figure 1D). Although sensitive, this protocol is more complex, and requires substantially more DNA - typically 10-50X more than paired-end libraries depending on the size of the fragmentation desired.

- - 34 1.4 Transcriptome libraries

Sequencing can also be used to capture information about gene expression levels and RNA editing events. Termed RNA-seq, these libraries are conceptually similar to DNA libraries, with the addition of a cDNA synthesis step to convert the RNA to DNA. Where the RNA from a sample is limited, an effective way to prepare RNA libraries for sequencing is to perform cDNA synthesis using random-hexamer priming prior to fragmentation and adaptor ligation [4]. This step amplifies the amount of starting material available for subsequent steps in the protocol, however as it creates double stranded molecules, information regarding the originating strand is lost. Subsequent analysis of the data is problematic in cases of overlapping genes, overlapping anti-sense transcripts, or in experiments where the goal is to identify novel transcriptional events. By capturing RNA fragments prior to cDNA synthesis, fragment or paired-end libraries can be created which retain information regarding the strand of transcription. Strategies used so far include: adaptor tagged random hexamers (Cloonan et al. 2008), serial adaptor ligation [6], or simultaneous adaptor ligation [7]. However all these approaches require microgram scale reactions to produce high quality (high complexity) libraries.

2 Advanced Sequencing Technologies

Conventional “Sanger” sequencing [8] requires the isolation of individual DNA fragments, which is typically achieved either through PCR or molecular cloning into bacteria. The rapid advance in sequencing that has been observed in recent years was kick- started by the miniaturization of the PCR step – a protocol often referred to as “emulsion PCR” or ePCR [9].

- - 35 2.1 Massively parallel PCR

Emulsion PCR requires a population of DNA fragments that have been captured individually between adaptors of known sequence (see Section 1.1). This population is diluted in PCR reagents, and microscopic beads coated with primers complementary to the adaptor sequence are added to the mix. This aqueous mix of PCR reactants is then vortexed with oil to produce micro-reactors or ‘micelles’ (water in oil emulsion) which ideally contain a single DNA fragment, a single bead, and PCR reagents. This mixture is then thermally cycled, and each individual DNA fragment is clonally amplified onto a bead (Figure 2). Billions of micelles present in the emulsion allow the massive parallelization of the PCR step of sequencing, enabling in just hours what would take years to achieve using 384 well plates. For clonal amplification of individual library molecules, it is crucial to ensure that any micelle contains only a single DNA fragment. The probability of any micelle containing exactly N DNA fragments given a ratio of DNA fragments to micelles can be mod- eled by the Poisson distribution: (RN.e-R/N!). For example, using a DNA to micelle ratio of 1, the chance of a micelle containing exactly 1 DNA fragment is 36.8%, and there is 26.4% of containing 2 or more DNA fragments. As multiple templates on a single bead will lead to mixed signals when sequencing, the optimal ratio of DNA to micelle numbers is actually much less than this, usually about 0.1. At this ratio, the chance of a micelle containing exactly 1 DNA fragment is only 9.1%, however there is only a 0.47% chance of 2 or more fragments in the same micelle. Also important, but less crucial is ensuring that each micelle only contains a single bead. Multiple beads in a micelle can deplete the PCR reagents available and lead to lower intensity signals when sequencing. However using low ratios of beads can lead to insufficient yields of amplified material: the probability of a micelle containing exactly 1 DNA fragment and exactly 1 bead is the product of the individual probabilities. Using a DNA ratio of 0.1 and a bead ratio of 1 we can calculate an overall 3.3% chance of single bead and single template micelles. This means that the optimization of ratios is a trade-off between specificity and yield, and in practice this

- - 36 leaves the majority of micelles empty, and the majority of beaded micelles without DNA.

Figure 2. Emulsion PCR. Oil and PCR reagents (including the reverse primer) are mixed to form bubbles of aqueous reactors in an emulsion (micelles). The DNA template is amplified on beads coated with primers complimentary to the forward adaptor. Only a small fraction of all micelles will contain exactly one bead and one DNA fragment.

- - 37 2.2 Advanced sequencing technologies

2.2.1 Pyrosequencing

The first commercial second-generation sequencer (Roche’s 454 GS FLX) utilized pyrosequencing [10,11], a protocol where incorporation of a single nucleotide results ultimately in the emis- sion of light via the firefly enzyme luciferase [12]. Emulsion PCR (see section 2.1) is used to generate beads with amplified template DNA. These beads are arrayed onto a picotiter plate (PTP; a fused silica capillary structure), providing a fixed location at which the subsequent sequencing reaction can be recorded. The reaction is initiated by the addition of a primer and DNA polymerase to the PTP followed by a solution of each nucleotide in a stepwise fashion. The incorporation of a complementary base in the growing DNA strand releases pyrophosphate, and leads to the emis- sion of light captured by camera. The amount of light emitted is proportional to the number of nucleotides present in the template (Fig- ure 3). Sequencing across a homopolymer fragment (stretches of the same nucleotide within a sequence) can sometimes be troublesome for this technology due to limitations in the dynamic range of the cameras used, or insufficient nucleotides to complete the extension of the homopolymer, which can result in insertions or deletions when base-calling. By contrast, because each incorporation step is nucleotide-speciﬁc, substitution errors are rarely encountered in py- roseqeuncing reads [13].

2.2.2 Sequencing by synthesis

Illumina’s Genome Analyzer (previously Solexa) is based on the ‘sequence by synthesis’ concept to generate millions of reads per run. Unlike other massive-scale sequencing platforms, the Genome Analyzer doesn’t use emulsion PCR to amplify the signal from DNA fragments. Instead, this platform uses solid-phase amplification of DNA clusters [14]. Template DNA (captured between two different

- - 38

Figure 3. Schematic diagram of Pyrosequencing. Emulsion PCR is used to generate beads covered in sequencing template. These are added to individual wells along with a cocktail of enzymes and reactants, including DNA polymerase, ATP sulfurylase, and Luciferase. The sequential addition of either dATP, dCTP, dGTP, or dTTP results in an enzymatic cascade which converts nucleotide incorporation into the growing DNA strand to light. This conversion is semi- quantitative– the light generated is proportional to the number of nucleotides incorporated. The signal is then captured by a high performance camera, and digital- ly converted to a DNA sequence.

adaptors of known sequence) is added to a glass flowcell coated with covalently attached oligos complementary to the two adaptors. Hy- bridization of the library molecules to the flowcell is followed by its

- - 39 amplification via bridge PCR to produce randomly distributed, clonally amplified clusters (Figure 4A). Each sequencing cycle involves flushing the flowcell with reaction mixture, including: primers, a nucleotide mix, and DNA polymerase. The nucleotide mix contains four differently labeled fluorescent nucleotides that are reversibly blocked at the 3′-OH to ensure that only a single base is incorporated per cycle by the DNA polymerase. The fluorescence at each cluster and its position on theglass surface is captured by camera, followed by a chemical step to remove the fluorescent group and unblock the 3’ end. The read length in each run is dependent on the number of cycles performed (Figure 4B). Since this method reads each base individually, errors are not introduced while sequencing homopolymer sequences. The use of a modified polymerase and reversible dye terminator nucleotides, however, results in base-substitution errors and shorter read lengths than compared to pyrosequencing.

2.2.3 Sequencing by ligation

The SOLiD system from Applied Biosystems (now Life Tech- nologies) is based on the hybridization-ligation chemistry, also referred to as sequencing by ligation [15]. Also utilizing emulsion PCR, the beads generated are covalently attached to a glass slide and placed into a fluidics cassette within the sequencer (Figure 5A). The sequencing process starts with the annealing of a universal sequencing primer complementary to the forward adapter on the library molecules. Fluorescently labeled nucleic acid probes are introduced to hybridize to the template, and are ligated to the primer by DNA ligase. The unbound probes are washed away and fluorescent signals are recorded. A cleavage and wash step is performed to remove the final 3 bases along with the fluorescent group of the li gated probe to enable subsequent rounds of ligation and imaging (Figure 5B).

- - 40

Figure 4. Sequencing by synthesis on the Illumina Genome Analyzer. (A) Instead of emulsion PCR amplified beads, single library molecules are clonally amplified by bridge PCR. Library molecules are hybridized to the flow cell which is covered with primers complementary to the library adaptors. PCR occurs through the bridging of single-stranded DNA molecules to nearby adaptors, causing replication to occur in clonal clusters. (B) Sequencing by synthesis is achieved through reversible terminator chemistry. The simultaneous addition of fluorescently labeled dNTPs (colored circles) that cannot extend ensures incorporation of a single nucleotide per cycle. The flow cell is washed, imaged, and then the terminator (black triangle) is cleaved from the nucleotide allowing extension for the next cycle.

Rather than directly probing individual nucleotides, SOLiD sequencers detect relationship between nucleotides due to the degenerate design of the probes (di-base encoding). Each probe is an octamer consisting of 2 probe-specific bases and 6 universal bases with one of 4 fluorescent labels attached to the 5' end. In the first

- - 41

Figure 5. Sequencing by ligation. (A) Emulsion PCR is used to generate beads which are subsequently attached to the surface of the slide. (B) DNA ligase is used to extend the DNA strand by incorporating fluorescently labeled octamer probes (inset). Each probe contains one of 16 possible di-base combinations, and an additional 5 random nucleotides (for a total of 1024 unique probes). A phosphorothio- ate cleavage point allows the fluorescent label to be cleaved, and further probes can be ligated. (C) Once the full length of the tag has been extended, the DNA strands are denatured, and new primers are annealed. These new primers differ in their starting positions within the adaptor, which allows the di-base probes to interrogate different nucleotides. For primers n-2 onwards, unlabelled “bridge probes” are used to push the interrogated area out so that the adaptor sequence is not interrogated. After five ligation cycles, every nucleotide would have been

- - 42 interrogated twice (dashed line). The overall length of the tag can be extended by performing extra ligation cycles per primer. Additional primer sets using different probes can be used to increase the accuracy of the reads. For example, the “Exact Call Chemistry” uses a single extension from the n-4 primer with tri-base probes containing known nucleotides at nucleotides 1, 2, and 4, which can increase the error correcting capabilities according to the manufacturer. The sizes of primers, probes, and DNA fragments are not to scale.

ligation step, probes consisting one of 16 possible 2-base combinations (eg. TT, AT etc) compete to anneal to the template sequence adjacent to the universal primer followed by its ligation. The synthesized strand is then denatured and washed away, and the process is restarted with a different primer, offset from the original. Five rounds of primer reset are completed for every sequenced tag – each primer is offset from the previous primer by 1nt to ensure that every nucleotide in the DNA template gets probed by the di-base twice. The number of ligation cycles determines the eventual read length. Although slower than the previously discussed technologies, di-base encoding results in sequencing every nucleotide twice, leading to higher system accuracy than single-pass sequencing according to the manufacturer (McKernan et al. 2009). However, the output of di-base encoding is a string of digits (0, 1, 2, or 3) representing the di-nucletotide relationships rather than the familiar strings of A, C, G, or T, and presents specific challenges for analysis. This representation, known as “color-space”, has not been widely adopted by the sequencing community, and the development of analysis tools lags behind those available for other platforms. The most common error reported from SOLiD sequenced tags include base-substitutions and sequence bias due to the under- representation of sequences at the extremes of G/C % content [17].

2.2.4 Semiconductor sequencing

A recent development in massive scale sequencing has been the release of the Ion Torrent by Life Technologies. Most closely analogous to pyrosequencing, the Ion Torrent also arrays beads generated by emulsion PCR, and sequentially washes the array with individual nucleotides and DNA polymerase. However two major

- - 43 advances differentiate these technologies. Firstly, the array is a semiconductor microchip with wells positioned above nano-scale pH detectors. Secondly, native (unlabelled) nucleotides are used in the sequencing reactions. Nucleotide incorporation is detected as a change in pH when protons are released as a byproduct of DNA polymerization (Figure 6). This change in process reduces the cost of the machines and the reagents considerably, as no specialized optics are required. The sequencing runs are also considerably faster using native reactants, a typical run (generating up to 200nt) will last only two hours. The technology is in its infancy, but early reports suggest that the homopolymer tract issue is less of a problem than with pyrosequencing [18]. However this may not be due to any fundamental superiority of the platform, but to the sophistication of the correction algorithms used in the primary analysis (see section 3.1).

Figure 6. Semiconductor sequencing. Emulsion PCR is used to generate beads covered in sequencing template. These are added to individual wells along with DNA polymerase. The sequential addition of either dATP, dCTP, dGTP, or dTTP results in the release of a hydrogen ion (H+), which is detected by the sensitive pH meters built into the semiconductor chip. The signal is converted to a sequencing read analogous to pyrosequencing, however as the optics are not required, the runs are faster, and both the reagents and machines are substantially less expensive than other currently available technologies.

- - 44 2.2.5 Massive scale of data generation

Regardless of the platform used, parallelizing sequencing dramatically increases the volume of data produced versus non-parallel sequencing. These new sequencers can analyze hundreds of millions of DNA fragments per run, compared with 96 to 384 DNA fragments using the traditional method. Although the read-lengths obtained are substantially shorter than with capillary sequencing (50 to 400 nucleotides versus > 1000 nucleotides), the volume of sequence produced is substantially higher. For example, a SOLiD run generating 500 million mappable 50-mer tags will produce 25 Gb of usable sequence in 7 days – almost 1500-fold more than previously possible using a 384-well capillary sequencing instrument in the same time frame.

3 Data analysis and bioinformatics

Experiments using deep sequencing platforms require extensive data analysis, normally consisting of three loosely coupled phases.

3.1 Primary analysis

In the first, primary phase, tag sequence and quality values are extracted from original high resolution images of each sequencing cycle (or in the case of Ion Torrent, digital signals of the change in pH). This analysis (termed base-calling) is normally performed by vendor maintained instrument embedded software and hardware. When raw intensity signals, normally of optical origin, are transformed into base calls, each base call is ascribed a quality (generally a phred score), representing the likelihood that base is correct, and are closely linked to the molecular biology featured by that platform (see sections 2.3.1 – 2.3.4). Additionally, all platforms suffer from deterioration of signal as the read extends, leading to noise towards

- - 45 the end of sequences – high throughput is traded for lower accuracy of reads. Although software has been written to improve the base- calling of 454, Illumina, and SOLiD [19,20], they are often made at great computational expense in terms of both disk space for raw data storage and CPU time for re-processing it, hence not widely adopted. The extracted data has historically been text (ASCII) files which are easy to view, parse, and manipulate, however as the size of the output data increases, binary file formats are being adopted to better manage data storage issue. The use of binary formats for sequencing data will be advantageous as their file size is smaller than plain ASCII text, and also allows for more rapid access to individual sequencing reads when required.

3.2 Secondary analysis

Secondary analysis typically involves aligning the tags to some reference genome, if it exists, a process both computationally intensive and often strategically challenging [21]. This is due in part to systematic error in deep sequencing experiments, the size of the data sets being utilised, and the approach of aligning to reference sequences derived from different samples to the source data. Deep sequencing alignment software packages employ a variety of algorithms and approaches, often forcing a trade-off between accuracy and speed, as well as providing functions to facilitate certain tertiary analysis requirements. For example, aligners producing ‘gapped’ alignments tend to perform slower than ‘ungapped’ aligners [22], but can detect insertion and deletion events, which may be a requirement for experimental analysis. Some aligners adopt a heu- ristic approach, often with very fast performance, but at the expense of potentially missing some alignments. To increase performance, a common algorithmic approach is to ‘seed and extend’ an alignment. In this approach, small sub-sequences (seeds) of a specified length are used to rapidly determine candidate alignments, which can be extended to the full length of the read, or partially to include a pa- rameterised minimum amount (variable length alignments). For additional speed and sensitivity, ‘spaced seeds’ are most often used.

- - 46 For highly accurate short read alignments, initial seed hits can be extended using computationally expensive algorithms such as Smith-Waterman to determine the most accurate local alignment. In some applications of deep sequencing methodologies, such as mammalian transcriptomics (RNA-seq), the mapped data can be subjected to further multiple sequence alignment among the tags to produce a reference-based assembly of transcripts. The adoption of such assembly approaches can assist in identification of novel sequence events. In all cases, the goal of secondary analysis is to identify biological signal from the data, reduce the data size and complexity, and transform data to a format suitable with the continued analysis requirements.

3.3 Tertiary analysis

Unlike the previous two phases, the third phase is application specific, and the analysis will depend on the experimental design, the library protocol employed, and the precise biological question to be addressed. Many qualitative and quantitative aspects can be extracted from the sequencing data: sequence content (eg. mutations or RNA editing events), gene expression estimates, structural variation, copy number analysis, DNA or RNA footprinting, are but some of the applications (see section 4). Crucially, in order to obtain biologically meaningful results, and reduce experimental noise, this ‘bottom up’ analysis approach imparts heavy reliance on the appropriate completion of previous stages for effective tertiary analysis [23].

4 Applications of tag sequencing

4.1 Historical applications

Although tag sequencing has often been thought of as a recent invention made possible by the rapid evolution of sequencing technology, it has been used since the early 1990s when “Sanger” se-

- - 47 quencing became broadly accessible to many laboratories. The de- sire to increase the throughput of gene discovery and gene expression studies while lowering the experimental costs led to the adoption of tag sequencing and spawned the field that would later become known as transcriptomics. At the heart of this drive were the “Expressed Sequence Tags” (ESTs), generated by single stranded sequencing of the extreme ends of cloned cDNAs [24]. Large scale EST screens were slow to be adopted as the results were seen as incomplete and inaccurate [25]. However these tags soon proved to be invaluable, enabling rapid gene discovery (both novel genes in the same species [26], or the same genes in novel species [27]), exon identification [28], and genetic mapping [29]. By 1995 more than half of the records in Gen- bank were ESTs, and were being heavily accessed by scientific community [25]. The development of SAGE (Serial Analysis of Gene Expres- sion) - where transcripts are cleaved by one or more restriction enzymes, ligated together, and then sequenced - allowed for the first time broad detailed surveying of the mRNA population of cells [30]. SAGE had the advantage of not relying on prior knowledge of transcript sequences for their detection, and that quantification of transcript abundance was possible using the counts of sequenced tags. The primary disadvantage was the very short length of tag sequenced, typically only 10 – 14 nucleotides long. This short length, combined with the redundancy of complex genomes confounded the unambiguous detection of transcripts with multiple tag-to-gene mappings. This meant that statistically significant detection of differentially expressed transcripts was difficult, particularly for those transcripts that were very poorly expressed (Lu et al. 2004). While improvements to the SAGE protocol extended tag lengths to 21 bases using LongSAGE [32], and then to 26 bases using SuperSAGE [33], these protocols (like EST sequencing) were still limited by cost, where it was prohibitively expensive to thoroughly survey multiple transcriptomes. When lower capillary sequencing costs failed to materialize and DNA microarray technology became the predomi- nant genomic workhorse [34], interest in tag sequencing waned.

4.2 Whole genome sequencing

- - 48 The first, and arguably the most common application of the newer massive-scale sequencing was to perform whole genome sequencing or re-sequencing [12]. Full prokaryote de-novo genomic sequencing is now possible in single runs, although typically longer sequence reads (from capillary or 454 sequencing) are used as scaf- folds and paired with the shorter massive scale technologies to generate high levels of sequence coverage for complex eukaryotic genomes. There has also been a widespread adoption of massive scale sequencing to detect mutations in cancer genomes. Many different kinds of mutations can be discovered by tag sequencing, including: simple nucleotide variations (SNVs; substitutions and small insertions or deletions); copy number variations (CNVs; amplification or loss); and structural variations (SVs; insertions, deletions, inversions, and translocations) (Figure 7).

4.3 Targeted genome re-sequencing

Another popular application of massive-scale tag sequencing is the re-sequencing of specific sub-regions of genomes. This includes “exome” re-sequencing, where genomic libraries are enriched by hybridization to arrays of exon sequences, providing a more cost effective approach when sequence variations of interest are likely to impact the transcriptionally active genome. The DNA binding locations of specific proteins can be interrogated using Chromatin ImmunoPrecipitation followed by deep sequencing (ChIP-seq) experiments [35]. In such experiments, transcription factor binding sites, RNA polymerase II binding sites and other proteins, as well as nucleosome locations can be determined. These experiments provide insight into transcriptional regulation and the functional organization of the genome. A more general application of this idea, DNAse footprinting, can be used to assess genome- wide positions of all DNA binding proteins simultaneously [36].

- - 49

Figure 7. Detecting structural variations using paired sequencing. Tags from “paired-end” or “mate-pair” libraries are aligned to a reference genome, and the distance between the two tags is used to detect specific structural variants. Where the alignment distance is concordant with the size of the DNA fragments, no structural variation can be inferred. Discordant mapping distances and orientations between the two tags can identify specific molecular events. Deletions: are detected when the observed alignment distance for paired end reads is much larger than what would be expected from the library. Insertions: can be inferred when the observed alignment difference is much smaller than the expected alignment distance. Translocations: Inter-chromosomal translocations can be detected when the paired ends of a DNA fragment map to different chromosomes. Intra- chromosomal events are often seen as pairs of insertion and deletion events. Inver- sions: are identified when the orientation of the observed tags is different to the expected orientation, and depending on the size of the inversion, may also be coupled with discordant pairs detecting insertion and deletion events.

There are a number of strategies for capturing epigenomic signatures. Antibody based capture protocols, including ChIP-seq and MeDIP-seq (Methylated DNA ImmunoPrecipitation sequencing) [37], are easily combined with fragment library making strategies to profile DNA methylation and histone modification signatures [38].

- - 50 Similarly, capture using methyl-CpG binding domains (MBDs) [39] is also well suited for traditional fragment library sequencing however neither MeDIP nor MBD strategies are able to precisely determine the methylation status of each cytosine within the capture fragment. Bisulfite sequencing (a protocol that converts unmethylat- ed cytosines to uracils) [40] provides this resolution, but is difficult to combine with short-tag sequencing due to the ambiguity in aligning bisulfite converted short tags. However, recent improvements to both the experimental protocol and to the computational analysis may make this approach more widely adopted [41]. Another less commonly applied targeted resequencing protocol is Restriction site Associate DNA sequencing (RAD-seq) [42]. RAD tags are generated through restriction enzyme digest and ligation of a biotinylated adaptor, followed by DNA shearing and isolation of biotinylated tags using streptavidin beads. Such libraries are used for genetic mapping [43], genotyping studies [42], and can also be used for targeted enrichment of CpG islands for DNA methylation studies [44,45]

4.4 Whole transcriptome sequencing

Whole transcriptome RNAseq can provide accurate and sensitive gene expression data, but also information on novel exons, expressed mutations, alternative splicing, and fusion gene identification (Figure 8) [46]. RNAseq data can be prepared as fragment or paired-end libraries, analogous to those described above - and the identification of exons and introns parallels the identification of deletions in genomic DNA. Fragment library sequencing can easily detect known exon-exon junction sequences, but the sensitivity for novel events is low. Paired-end libraries are far more sensitive, as they do not rely on a tag to cross an exon-exon boundary, and can also detect the relationship between novel exonic sequence and the transcriptional framework. Once the framework is assembled, the presence of individual transcripts can be modelled and quantified [47,48]. The single nucleotide resolution of RNA-seq data also allows for determination of allelic specific expression if the parental genotype is known [49].

- - 51 4.5 Targeted transcriptome sequencing

Strictly speaking, most “whole transcriptome” RNA-seq experiments could already be considered as targeted transcriptome tag sequencing. The majority include a step to deplete the very large proportion of cellular RNA composed of ribosomal RNA (rRNA) or transfer RNA (tRNA), together accounting for >99% of RNA in an average eukaryotic cell. Semantics aside though, targeted transcriptome sequencing has revealed a great deal about the structure and function of complex transcriptomes. Short-tag sequencing is ideally suited to detecting microRNAs (~22nt) and other small RNA populations (siRNAs, piRNAs, tiR- NAs, snoRNAs, etc), as these molecules do not need to be fragment- ed prior to library preparation [50]. Such libraries are generally prepared from size selected RNA, either prior to or after the ligation of adaptors (see section 1.1). SAGE (see section 4.1) and CAGE (Cap Analysis of Gene Expression) [51] are tag sequencing strategies that both pre-date massive-scale sequencing – but these have been given a new lease of life with the dramatically reduced cost of obtaining data. CAGE, used to measure the transcriptional start sites of RNAs with a methyl-G cap, has been particularly boosted by the new technology, spawning new protocols for comprehensive profiling (deep-CAGE) [52], and for applications where only tiny amounts of sample are available (nano-CAGE) [53]. Cross-Linking Immuno-Precipitation sequencing (CLIP-seq) [54] or High-throughput Sequencing CLIP (HITS-CLIP) [55], are two names to describe a protocol to profile the specific binding sites of RNA-binding proteins. RNA is cross-linked to bound proteins by UV irradiation. Unbound RNA is degraded, and the protected RNA is isolated and sequenced. It can be applied to all endogenous RNA- binding proteins (RNA foot-printing), or to specific RNA-binding proteins by immuno-precipitaion or other protein tagging approaches. This technique is rapidly gaining popularity, as the immense regulatory potential of RNA-protein interactions becomes apparent. Recently, HITS-CLIP has been used to identify miRNA binding sites in mRNA [56].

- - 52

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

5 Clinical applications of tag sequencing

Massively-parallel sequencing technology has progressed to the point where entire human genomes can be sequenced rapidly and comprehensively – and it is only a matter of time until sequencing costs fall to the magic $1000 mark, widely thought to be the threshold price for consumer adoption of genome sequencing as a clinical test [57]. This is causing much excitement among medical research-

- - 53 ers, as it heralds an era of personalized medical genomics where genomic analysis is used to take the guesswork out of treatment. Such tests could be used to determine what mutations are present within an individual’s cancer, and guide chemotherapy choice to achieve the best possible outcomes for that patient. Despite the hyperbole, there are various non-trivial challenges that must be overcome before whole genome sequencing can be adopted as a mainstream clinical test. These include issues of starting material availability, availability of 'normal' control samples, and the time involved for sample gathering and substantial computational analysis [58]. While not insurmountable, these challenges make it far more likely that whole genome sequencing for mutation or pol- ymorphism detection will not be routinely used as a clinical test in the short to medium term, remaining predominantly a research tool until these challenges are met. Of course the potential of tag sequencing in the clinic goes beyond whole tumor interrogation, and there are smaller scale tests with less ambitious outcomes that are currently under development. For example, Sequenom is developing a test known as MaterniT21 to detect the aneuploidy that causes Down syndrome. This approach uses tag sequencing to detect an over-representation of tags deriving from chromosome 21 (indicative of a trisomy) in the maternal bloodstream [59]. As single nucleotide resolution of chromosome 21 derived tags is not required, the analysis and interpretation of this test is far simpler and more amenable to routine clinical use. Research projects conducted in clinically accredited laboratories to discover the causative mutations of rare disease are also becoming more common: Miller syndrome [60], Dopa-responsive dys- tonia [61], and a rare manifestation of inflammatory bowel disease [62] have all had causative mutations identified through the use of massive-scale sequencing. While these success stories emphasize the potential benefits of tag sequencing in the clinic (although the majority of efforts to locate causal mutations are not as successful [63]), they are not routine genetic tests, and there is still significant ethical debate around using whole genome sequencing in the clinic [64].

- - 54 6 Future perspectives

The prospect of technology improvements is very good with all current sequencing platforms continuing to drive higher throughputs, longer reads, and improved accuracy from their current approaches. It is predicted that rapidly increasing data outputs from these technologies will make all tag sequencing projects more af- fordable. However, even with current throughputs though, sequencing is already coming up against its next major challenge – the computational requirements of storage, processing and analysis. What is becoming a headache of major genome sequencing centres world- wide will only become worse as throughputs increase, and larger experiments become more common. Major breakthroughs in both hardware and software will be required to cope with the onslaught of genomic-scale data in the coming years. The commercial release of single molecule sequencing technologies (SMS) will be an area to watch. Early access users of the PacBio sequencing platform have demonstrated that with SMS, it is possible to generate sequence reads in excess of 1,000 nt, in real time, from minute amounts of template [65]. However error rates for this technology are substantially higher than the other platforms, and this creates substantial analytical challenges and pragmatically limits the application of this technology [66]. It is hoped that the likes of SMS will complement current sequencing approaches in the short to intermediate term but significant improvements in both data generation and error rates will be required for SMS to overtake current sequencing platforms as the technology of choice.

Acknowledgements

This work was supported by the National Health and Medical Re- search Council (455857, 456140, 631701); Australian Research Council (DP1093164, DP0988754). KK and DLW receive an Aus- tralian Postgraduate Award from the Australian Federal Govern-

- - 55 ment. SMG is supported by an NHMRC Senior Research Fellow- ship, and NC is supported by an ARC Postdoctoral Fellowship.

References

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

- - 61 20 2)!&+#)**2,07,5!) &1!1!> %$ =ILL@MI> MJ> %!0BIGHHC 3*+$!+!2& 1=!+,*!1,+.0!1 0&.2&,+> %&#!$%=II@IK> MK> 7 !+BIGHHC! 0!21,#2%!%3*+$!+,*! &1 ),1! > %&#!$%=HN> ML> %&+@<,0!+1,+ < 00&1 <,&+1<%0)!1 BIGHHC%!,0&$&+,#2%! &2&+ %,)!0,320!(120&+> ' !&# ! #!=JJ@KI> MM> %,*.1,+ <20& !BIGHHC%!.0,.!02&!1+ ..)& @ 2&,+1,#1&+$)!@*,)! 3)!1!/3!+ &+$> ! !!( =IHN>

- - 62 1.3 Hypotheses and Aims

The extensive studies on miRNAs described above have revealed that these molecules are master regulators of genetic networks, frequently shown to be involved in human cancers. To truly understand their biological role in tumour initiation and progression it is imperative to characterize their biologically relevant targets using context-dependent cell models. This will lead to a better understanding of the miRNA- mediated disruption to specific molecular processes underlying tumorigenesis, and will maximize their potential for medical applications. There are three specific hypotheses that will be tested in this thesis: 1. The role of miRNAs in breast cancer is far from resolved because of the molecular heterogeneity of the disease. Identifying miRNAs whose expression is recurrently deregulated in human cancer samples will help discover novel oncomirs, whose role in the pathogenesis of the human cancer can then be elucidated. 2. The oncogenic potential of miRNAs is driven by the coordinated targeting of transcripts in key pathways. Accurately surveying miRNA-mRNA binding events by biotinylated miRNA pull-downs in human cells will reveal the biological processes and pathways underlying their cancer promoting potential. 3. Cell cycle is a key process underlying cell survival, proliferation and replication, and is frequently disrupted in disease states. Integrating leads from cell cycle related miRNA profiling and miRNAs deregulated in human cancers will enrich for novel oncomiRs, with either tumor suppressive or an oncogenic potential.

This thesis specifically aims to address these hypotheses by functionally characterizing miR-182 and miR-139, both of which are deregulated in human breast cancers. To study the miRNA-mRNA interactions we use a biotinylated pull-down approach, and perform experimental validations using cell based assays. We also perform next generation sequencing technology to identify miRNAs, which show phasic expression across the cell cycle and perform an overlap analysis with miRNAs deregulated in human cancers to identify novel oncomirs.

CHAPTER TWO miR-182-5p targets genes underlying the DNA Damage Repair pathway in breast

cancer

2 miR-182-5p targets genes underlying the DNA Damage Repair pathway in breast cancer

2.1 Summary

The role of microRNAs (miRNAs) in cancer were first discovered in 2002 when the miRs-15 and -16 were implicated in the causation of chronic lymphocytic leukemia [184]. Since then, hundreds of miRNAs have been identified and a large number of them have been linked to tumorigenesis [290]. Acting through the degradation of mRNA or inhibition of translation, they are believed to regulate the expression levels of genes exerting both oncogenic as well as tumour-suppressive properties, diving them into two groups based on their targets. Despite numerous studies reporting a causative role in different cancers, very few of these have provided deeper insight by defining the myriad of target genes that the miRNA could be acting upon. Identification of miRNA targets has been pursued, initially by carrying our miRNA microarrays [291], next-generation RNA-sequencing [292] and SILAC to quantitate changes in protein expression [62], high-throughput sequencing of RNA isolated by crosslinking immunoprecipitation (HITS-CLIP) [116] and biotinylated synthetic tagging [123] have been used more recently to directly identify sites on mRNA that the miRNA associates with. Despite these technological advances, very few studies on cancer have exploited these tools to venture deeper into miRNA targets and their integrative function in tumorigenesis. Chapter two of this thesis, presented in the form of a published manuscript titled ‘MicroRNA-182-5p targets a network of genes involved in DNA repair’, characterizes the role of miR-182-5p in mammary tumorigenesis by identifying its targets using a biotinylated pulldown approach. Based on this technique, we were able to identify over a thousand putative targets for this miRNA, a significant proportion of which have been previously documented to play a role in cancer biology. Targets of this miRNA were shown to be involved in the DNA damage repair pathway and to play a role in cell cycle. While characterizing the role of miR-182 in tumorigenesis, another study reported its role in DNA damage through its target gene BRCA1 and its ability to sensitize cells to PARP inhibition [293]. From our study, we identify a significant number of miR-182-5p target genes, including BRCA1, which play a role in DNA damage repair such as CHEK2 and

65 TP53BP1. We find that miR-182-5p targets a network of genes involved in DNA damage repair and carry out functional assays to show that re-expression of CHEK2 (and BRCA1 as previously reported) in miR-182 over-expressing cells can overcome their PARP sensitivity. Supplementary tables associated with the journal article are included as Appendix 1 of this thesis. The details of contributions to the following manuscript can be found on page vi under the heading ‘Publications included in this thesis’ This study, while identifying novel target genes for miR-182, also fosters the hypothesis that miRNAs act by coordinated repression of a large number of target genes that lie in the same pathways to achieve functional outcomes. Not only do these hypotheses apply to the case of miR-182-5p, but they will be an overriding theme through the following chapters of this thesis.

66 Downloaded from rnajournal.cshlp.org on April 13, 2014 - Published by Cold Spring Harbor Laboratory Press

MicroRNA-182-5p targets a network of genes involved in DNA repair

KEERTHANA KRISHNAN,1 ANITA L. STEPTOE,1 HILARY C. MARTIN,1,5 SHIVANGI WANI,1 KATIA NONES,1 NIC WADDELL,1 MYTHILY MARIASEGARAM,2 PETER T. SIMPSON,2 SUNIL R. LAKHANI,2 BRIAN GABRIELLI,3 ALEXANDER VLASSOV,4 NICOLE CLOONAN,1,6 and SEAN M. GRIMMOND1,6 1Queensland Centre for Medical Genomics, Institute for Molecular Bioscience, The University of Queensland, St Lucia, QLD, Australia 4072 2The University of Queensland, UQ Centre for Clinical Research (UQCCR), Herston, QLD, Australia 4029 3Diamantina Institute, Princess Alexandra Hospital, The University of Queensland, Woolloongabba, QLD, Australia 4102 4Life Technologies, Austin, Texas 78744, USA

ABSTRACT MicroRNAs are noncoding regulators of gene expression, which act by repressing protein translation and/or degrading mRNA. Many have been shown to drive tumorigenesis in cancer, but functional studies to understand their mode of action are typically limited to single-target genes. In this study, we use synthetic biotinylated miRNA to pull down endogenous targets of miR-182-5p. We identified more than 1000 genes as potential targets of miR-182-5p, most of which have a known function in pathways underlying tumor biology. Specifically, functional enrichment analysis identified components of both the DNA damage response pathway and cell cycle to be highly represented in this target cohort. Experimental validation confirmed that miR-182-5p-mediated disruption of the homologous recombination (HR) pathway is a consequence of its ability to target multiple components in that pathway. Although there is a strong enrichment for the cell cycle ontology, we do not see primary proliferative defects as a consequence of miR-182-5p overexpression. We highlight targets that could be responsible for miR-182-5p-mediated disruption of other biological processes attributed in the literature so far. Finally, we show that miR- 182-5p is highly expressed in a panel of human breast cancer samples, highlighting its role as a potential oncomir in breast cancer. Keywords: PARP inhibition; biotin pull-down; miRNA; target identification

INTRODUCTION the role of miRNAs as oncogenes or tumor suppressors has been reported in almost every type of cancer. In the case of MicroRNAs (miRNAs) are short (∼22 nt), single-stranded, breast cancer, miRNA expression profiling of more than 70 noncoding, negative regulators of gene expression in eukary- primary tumors and cell lines identified five miRNAs otes. An individual miRNA is capable of targeting hundreds (miR-10b, miR125b, miR-145, miR-21, and miR-155) that of distinct mRNAs (Thomas et al. 2010), and together the 1150+ human miRNAs are believed to modulate more than were consistently deregulated (Iorio et al. 2005). miR-10b a third of the mRNA species encoded in the genome (Bartel has subsequently been shown to play a role in the metastatic 2009). miRNAs have been shown to be biologically significant ability of breast cancer, positively regulating cell migration in various cellular processes such as cell differentiation, pro- and invasion (Ma et al. 2007). miR-21, initially shown to liferation, apoptosis, and development in humans as well as be overexpressed in several human breast cancers (Iorio other model organisms (for review, see Kloosterman and et al. 2005; Volinia et al. 2006), was later established to func- Plasterk 2006). Thus, deregulation of miRNAs can result in tion as an oncogene (Si et al. 2007) by targeting anti-meta- abnormal growth and development leading to several disor- static genes TPM1 (Zhu et al. 2007), PDCD4 (Frankel et al. ders, including cancer. 2008), and Maspin (Zhu et al. 2008). These and other studies Since the first study showing direct evidence of miR-17-92 highlighted the importance of identifying miRNAs driving acting as an oncogene in B-cell lymphomas (He et al. 2005), tumorigenesis and of accurate prediction followed by characterization of their biologically relevant targets. Intriguingly, some miRNAs (such as miR-17-5p) have been shown to 5Present address: Wellcome Trust Centre for Human Genetics, University of Oxford, Oxford OX3 7BN, UK have a dual role as both oncogenes and tumor suppressors, 6Corresponding authors depending on the cellular model under investigation (He E-mail [email protected] et al. 2005; Hossain et al. 2006; Volinia et al. 2006; Zhang E-mail [email protected] Article published online ahead of print. Article and publication date are at et al. 2006), and this has been due to a combination of the http://www.rnajournal.org/cgi/doi/10.1261/rna.034926.112. gene networks targeted by the miRNA and the expression

230 RNA (2013), 19:230–242. Published by Cold Spring Harbor Laboratory Press. Copyright © 2013 RNA Society. Downloaded from rnajournal.cshlp.org on April 13, 2014 - Published by Cold Spring Harbor Laboratory Press

miR-182-5p targets the DNA repair pathway levels of target transcripts (Cloonan et al. 2008). These dual- the BRCA1-dependent DNA damage response pathway and functioning miRNAs are of particular importance because the downstream G2 cell cycle checkpoint. In the context of they are key biological hubs that could be used to leverage a breast cancer, we demonstrate that miR-182-5p’s primary therapeutic outcome. effect is to mediate the double-stranded DNA damage re- Another miRNA with this potential dual function is miR- sponse. Finally, we observe overexpression of miR-182-5p 182-5p. This miRNA was first identified and cloned from the in a panel of human breast cancer patient samples, estab- mouse eye (Lagos-Quintana et al. 2003) and has shown to be lishing its role as a potential oncomir in human breast tumorigenic in melanoma (Segura et al. 2009) and endome- cancer. trial cancer (Myatt et al. 2010) and is overexpressed in lung (Cho et al. 2009), prostate (Schaefer et al. 2010), and colon RESULTS cancers (Sarver et al. 2009). It is also up-regulated in primary gliomas and is associated with poor prognosis for patients Identification of putative miR-182-5p targets using with metastasis (Jiang et al. 2010). In contrast, a tumor-sup- biotinylated pull-down method pressive role for miR-182-5p has been established in lung cancer (Sun et al. 2010; Zhang et al. 2011), and gastric adeno- Endogenous targets of miR-182-5p in HEK293T cells were carcinoma (Kong et al. 2012), where overexpression of this captured using biotinylated synthetic miRNAs as described miRNA leads to suppression of cell growth. Despite the wide- in Cloonan et al. (2011). Transient transfections of biotiny- spread association of altered miR-182-5p expression across a lated miR-182-5p molecules and mock transfections, to be range of human cancers, only a few targets have been identi- used as negative controls in the subsequent analysis, were car- fied so far; including FOXO3, MITF (Segura et al. 2009), ried out simultaneously in the same cell line. Expression pro- FOXO1 (Guttilla and White 2009), BRCA1 (Moskwa et al. filing was then performed on the pull-down fraction versus 2011), CTTN (Sun et al. 2010), RGS17 (Sun et al. 2010), the mock-transfected samples using microarrays (Fig. 1A). and more recently CREB1 (Kong et al. 2012) and MTSS1 The false discovery rate (FDR) was calculated to account (Liu et al. 2012). Recently, miR-182-5p, along with other for multiple testing (Benjamini and Hochberg 1995). miRNAs in its cluster, has been shown to affect apoptosis, Probes that met the 5% FDR threshold (for one-sided tests) senescence, proliferation, and migration/motility in medul- and with a fold-change >1.25 were considered significantly loblastoma (Weeraratne et al. 2012). We believe that a enriched in the pull-down (Fig. 1B). This differential expres- more thorough and systematic screen of miR-182-5p targets sion analysis (see Materials and Methods) revealed 1235 will help elucidate its biological role in tumor biology. probes (1091 genes) to be significantly enriched in the pull- Accurate miRNA target gene prediction has been challeng- down fractions (Supplemental Table 1). ing due to several reasons, mostly the constantly evolving hy- Previously validated mRNA targets of miR-182-5p, potheses surrounding the miRNA–mRNA target recognition BRCA1, RGS17, and FOXO3 were significantly enriched in principles. Initial studies showed miRNAs binding to the 3′ our pull-down (P-values ∼0.005) (Fig. 1B), confirming the UTRs of their target mRNAs by partial complementation validity of our approach. Although there is no large-scale ex- spanning 6–8 nt of the seed sequence in their 5′ end (Lim perimental validation of miR-182-5p targets with which to et al. 2005), and most computational target-prediction pro- compare, we would expect to see an enrichment of predicted grams rely on evolutionary conservation of these seed sites targets in our biotin pull-down data. To assess this, we com- to limit the high false-positive rates for such small sequence pared the TargetScan (Lewis et al. 2005) predicted targets of motifs (Brennecke et al. 2005; Krek et al. 2005; Lewis et al. miR-182-5p with our significantly enriched genes from the 2005). Even with this limitation, there are typically hundreds biotin pull-down (Fig. 1B,C). We observed an overlap of to thousands of targets predicted for a single miRNA, and 113 genes, which was significantly more than we would pre- − many are shown to be false positives when interrogated by lu- dict by chance (χ2 test; P ≈ 8×10 24; degrees of freedom = ciferase assays (Bentwich 2005; Rajewsky 2006; Baek et al. 1), indicating that the biotin pull-down is enriched for 2008; Cloonan et al. 2008). Along with the lack of overlap predicted targets of miR-182-5p. Finally, we selected four between targets predicted by different programs (Saito previously uncharacterized targets (CDKN1B, CHEK2, and Saetrom 2010), this limits streamlined analysis of biolog- SMARCD3, and NFKBIB), which also fall within our strin- ically relevant targets and their subsequent experimental gent significance threshold, and asked whether miR-182-5p validation. could interact with these mRNAs. We cloned the predicted Our study overcomes these obstacles by using a combina- binding sites into the 3′ UTR of a luciferase reporter gene, tion of biotinylated synthetic miRNA pull-downs (Cloonan and cotransfected either miR-182-5p mimic or a negative et al. 2011) to capture endogenous mRNA targets of miR- mimic control, and measured the luciferase activity 48 h 182-5p, and microarray expression profiling to identify post-transfection. Using this assay, we validated all four bind- them. We show that miR-182-5p targets more than 1000 ing sites tested (Fig. 1B,D). Taken together, these results con- genes including many well-characterized oncogenes and tu- firm that our pull-downs are enriching for genuine targets of mor suppressors. Its targets include multiple components of miR-182-5p.

www.rnajournal.org 231 Downloaded from rnajournal.cshlp.org on April 13, 2014 - Published by Cold Spring Harbor Laboratory Press

Krishnan et al.

miR-182-5p targets genes involved A Sample relations based on 9599 genes with sd/mean >0.1 in the DNA damage response 50 pathway 45 40 35 To infer the broad biological process- 30 es regulated by miR-182-5p, we per- 25 formed gene set enrichment analysis 20

mock transfection 3 (GSEA) on the full set of experimentally mock transfection 1 mock transfection 2 determined targets miR-182-5p (biotin pull-down) using Ingenuity Pathway miR-182-5p pulldown 1 miR-182-5p pulldown 2 Analysis (IPA, Ingenuity Systems). The enrichment of ontologies found in this 7 set was compared with 10 randomly B TargetScan predicted target Previously identified target generated gene lists of equivalent size. 6 Luciferase assay validated target Supplemental Table 2 lists all “Molec- ular and Cellular Functions” (P-value 5 < 0.002), and Supplemental Table 3 lists all “Canonical Pathways” (P-value 4 fold-change threshold < 0.02) associated with the miR-182- BRCA1 SMARCD3 − (p-value) CDKN1B 5p-predicted targets where the log(P- 10 3 value) of the enrichment was at least FOXO3 -log one standard deviation away from the 2 mean −log(P-value) from random gene RGS17 NFKBIB lists. For all subsequent analysis and 1 significance threshold validations, we considered an ontology CHEK2 FOXO3 FOXO1 as significantly enriched if its −log(P- 0 value) was at least four standard devia- –4 –3 –2 –1 0 1 2 3 4 tions away from the mean of the −log higher in pull-down higher in mock log2(fold-change) (P-value) from the random gene sets (Cloonan et al. 2008). Only two molecular processes were C p ≈ 8.4 X 10-24 D identified above our stringent thresh- -ve control miR-182-5p old: “Gene Expression” (160 genes; P- 1.1 ≈ −11 “ ” 1.0 value 1.15 × 10 ), and Cell Cycle −10 0.9 **** (114 genes; P-value ≈ 1.46 × 10 ). 0.8 Like many miRNAs, transcription fac- 479 113 978 0.7 0.6 tors are over-represented in miR-182- 0.5 0.4 5p targets (Cui et al. 2006). The cell 0.3 cycle ontology is similarly broad, en- 0.2 TargetScan predictions Relative Luciferase Activity 0.1 compassing a variety of proliferative 0.0 and regulatory functions from the direct miR-182-5p pull-down targets pmirGlo CDKN1B CHEK2 SMARCD3 NFKBIB control of proliferation to checkpoint FIGURE 1. Identifying targets of miR-182-5p via biotin pull-down. (A) Hierarchical clustering control. Like others (Moskwa et al. of microarray data was performed using the plotSampleRelations function in the lumi package. 2011), we could not identify a prolifera- “ ” Total vertical distance between samples indicates similarity. (B)A volcano plot showing the tive defect in MDA-MB-231 cells upon log2-transformed fold-change (mock/pull-down) versus the log10-transformed P-value for that fold-change for every gene detected above background in the microarray. (Blue) Genes that are inducing overexpression of miR-182- targets validated by previous studies; (orange) genes predicted by TargetScan to be targets of 5p (Fig. 2A,B). We constructed three miR-182-5p; (green) genes selected for validation using luciferase assays (D). There is an apparent independent cell lines that could over- enrichment of the targets in the pull-downs compared with the controls. (C) Venn diagram express miR-182-5p in response to showing the overlap of genes between TargetScan-predicted targets of miR-182-5p (also expressed above background in HEK293Ts) and biotinylated miR-182-5p pull-down-predicted tar- doxycycline (Fig. 2A). Using all three gets. This overlap is significantly more than expected by chance. (D) Dual luciferase assay used to stable cell lines, we examined the rate validate CHEK2, SMARCD3, CDKN1B, and NFKBIB as targets of miR-182-5p. HEK293T cells of proliferation (Fig. 2B) and the dis- were transiently cotransfected with 20 nM miR-182-5p or control mimic with a pmirGlo-lucif- tribution of cells in each phase of the erase construct containing the predicted binding site from the indicated target gene. Luciferase activity was normalized to Renilla activity; (∗) P < 0.05 in a Student’s t-test. The data plotted cell cycle (Fig. 2C), but were unable to are the mean and SEM of three independent biological replicates. detect any substantial differences. We

232 RNA, Vol. 19, No. 2 Downloaded from rnajournal.cshlp.org on April 13, 2014 - Published by Cold Spring Harbor Laboratory Press

miR-182-5p targets the DNA repair pathway

A B progression in the cell lines tested here. 1000 0ng/mL doxycycline However, this does not exclude a role 1.00 1000ng/mL doxycycline 0ng/mL doxycycline for this miRNA in proliferation in other 1000ng/mL doxycycline 100 biological contexts or other aspects of 0.75 cell cycle biology, such as checkpoint 10 0.50 regulation. We next sought to identify the canon- MTT activity 1 0.25 ical pathways enriched with predicted miR-182-5p targets. The most signifi- 0.1 Relative miR-182-5p expression 0.00 cantly enriched pathways were all facets #6 #9 #29 0 10 20 30 40 50 60 70 80 90 100 110 of the DNA damage response (DDR) Stable Cell line ID (MDA-MB-231) Hours Post Doxycycline (Supplemental Table 3): “Cell Cycle: G2/M DNA Damage Checkpoint Regu- C D ” ≈ −05 “ 15 lation (P-value 2.90 × 10 ), Role 50 0ng/mL doxycycline of BRCA1 in DNA Damage Response” 1000ng/mL doxycycline − 40 (P-value ≈ 8.50 × 10 05), “Hereditary 10 30 Breast Cancer Signaling” (P-value ≈ − 1.1 × 10 04), “Cyclins and Cell Cycle 20 − 5 Regulation” (P-value ≈ 1.49 × 10 04), 10 (relative expression) Proportion of cells 10 “Role of CHK Proteins in Cell Cycle

0 Log 0 Checkpoint Control” (P-value ≈ 5.6 × G1/G0 S G2/M miR-182-5p mimic Control mimic − 10 04), and “ATM Signaling” (P-val- Cell Cycle Phase Transfection Molecule − ue ≈ 8.20 × 10 04). To confirm that E this enrichment is not occurring in 70 * response to a foreign molecule being in- 60 miR-182-5p mimic troduced into these cells, we transiently Control mimic 50 transfected 20 nM miR-182-5p mimic 40 (∼3× the concentration of the biotiny- 30 lated molecule) into MDA-MB-231 20 cells and evaluated expression levels of

Proportion of cells 10 0 known DNA repair genes. Figure 3A G1/G0 S G2/M shows that only ATF1 changed signifi- Cell Cycle Phase cantly between cells transfected with FIGURE 2. Overexpression of miR-182-5p does not induce a proliferative defect. (A) Expression miR-182-5p or negative mimic control, of miR-182-5p as assessed by qRT-PCR in MDA-MB-231 cells (with low endogenous expression but only to a small extent (1.16×). This of miR-182-5p) stably transfected with miR-182-5p whose expression is induced in response to confirms that there is no substantial doxycycline. Shown here are three independent cell lines grown in the presence of 0 or 1000 ng/ mL of doxycycline for 48 h. RNU6B was used as an endogenous control for normalization of ex- and widespread DNA-damage response pression. (B) MTT cell proliferation assays of MDA-MB-231 cells stably expressing miR-182-5p. as a result of introducing an exogenous The graph plots the mean and SEM of the previously studied stable cell lines grown with either 0 molecule into these cells. To further or 1000 ng/mL doxycycline. The induction of miR-182-5p does not affect the proliferation rates verify if this response is miR-182-5p of MDA-MB-231 cells. (C) DNA profile analysis of MDA-MB-231 cells stably expressing miR- 182-5p. The graph shows the mean and SEM of the percentage of cells in different cell cycle phas- specific, we performed functional en- es, as assessed by FACS. There was no significant difference between MDA-MB-231 cells express- richment analysis on biotin pull-down ing or not expressing miR-182-5p. (D) Expression of miR-182-5p in HeLa cells transiently targets of other miRNAs (data generated transfected with miR-182-5p mimic or negative mimic control as assessed by qRT-PCR. in-house). Figure 3B shows the canon- RNU6B was used as an endogenous control for normalization of expression. (E) A graph showing “ the DNA profile analysis of HeLa cells transiently transfected with either miR-182-5p mimic or a ical pathway Role of BRCA1 in DNA control mimic. Shown is the mean and SEM of three independent biological replicates, each per- Damage Response” to be highly signifi- ∗ formed in technical triplicates. ( ) P < 0.05 in a Student’s t-test (n = 3). cant only in the miR-182-5p pull-down analysis. As shown in the figure, the level of enrichment in the miR-182-5p were also unable to find a significant change in the proportion pull-down is at least four standard deviations away from of cells in each phase of the cell cycle upon miR-182-5p over- the mean of −log(significance) of other miRNA pull-downs, expression in this system (Fig. 2C), and only a small (5%) ac- suggesting that this enrichment is not occurring by chance cumulation in the G1 phase of transiently transfected HeLa and is a specific response to miR-182-5p overexpression. cells (Fig. 2D,E). Taken together, we conclude that miR- DDR is a key pathway in cancer, and its disruption leads to 182-5p is not sufficient to modulate proliferation or cell cycle genetic instability and promotes tumorigenesis (Deng 2006).

www.rnajournal.org 233 Downloaded from rnajournal.cshlp.org on April 13, 2014 - Published by Cold Spring Harbor Laboratory Press

Krishnan et al.

A 1.6 ulation of the response to DNA damage, so their repression 1.4 by miR-182-5p could lead to a disruption of the DDR path- 1.2 * way. There are, however, two classes of proteins that are an 1 exception to this coherent model. First is the Cyclin-depen- 0.8 dent Kinase 6 (CDK6), which, along with CDK4, is known 0.6 to phosphorylate and inactivate RB (RB1), leading to the 0.4 cell cycle progression (Meyerson and Harlow 1994). How- Relative Expression Relative 0.2 ever, in response to DNA damage, this would typically lead 0 to an accumulation of genomic instability and is therefore CHEK2 BRCA1 ATF1 TP53 FOXO1 ATMIN ACTIN HPRT repressed by p53 (TP53), through p21 (CDKN1A) in vitro (see Fig. 4; Meyerson and Harlow 1994; Harper et al. B Role of BRCA1 in DNA Damage Response 5 1995). Hence, repression of the CDKs by miR-182-5p would lead to activation of RB and a cell cycle arrest, enabling effi- 4 cient DNA repair. The second class is the components of the SCF complex that are responsible for the ubiquitylation of 3 several proteins that are essential for the DDR, e.g., SKP1 and 2, components of the SCF complex, which are known KIP1 2 to degrade p27 (CDKN1B), an essential mediator of cell cycle arrest in response to DNA damage (Cuadrado et al. -log (significance) 2009). Another component of the SCF complex, BTRC, is 1 also known to ubiquitylate ATF4, which plays a role in main- taining genomic integrity (Lassot et al. 2001). The repres- 0 miR-182-5p miR-10a miR-10b miR-17-5p miR-23b miR-27a sion of these two classes of proteins by miR-182-5p could lead to a normal functioning DDR, which may explain the FIGURE 3. Enrichment of targets in the DNA damage response is spe- tumor-suppressive effects of this miRNA observed in lung cific to miR-182-5p overexpression. (A) Real-time PCR analysis to eval- uate the relative mRNA levels of known DNA repair genes was cancer (Sun et al. 2010; Zhang et al. 2011), and human gastric performed in MDA-MB-231 cells transiently transfected with 20 nM adenocarcinoma (Kong et al. 2012), in a manner analogous miR-182-5p mimic or negative mimic. (∗) Significant changes (P < to miR-17-5p’s dual functionality (He et al. 2005; Hossain 0.05). Data were normalized using HPRT as the internal control. The re- et al. 2006; Volinia et al. 2006; Zhang et al. 2006). sults shown are from three independent biological replicates, each performed in technical quadruplicates. (B) Functional enrichment analysis was performed using biotin-enriched targets of several miRNAs including miR-182-5p. As shown, the canonical pathway “Role of BRCA1 in miR-182-5p modulates PARP inhibitor sensitivity DNA Damage Response” is highly significant in the miR-182-5p pull- by targeting multiple components of the DNA repair down with its −log(significance) at least four standard deviations pathway away (top arrow) from the mean −log(significance) (lower arrow) of the other miRNAs. This confirms that the enrichment seen for targets miR-182-5p has recently been reported to target BRCA1 in the DNA damage response is miR-182-5p specific and not occurring (Moskwa et al. 2011), a critical component of the homolo- by chance. gous recombination (HR) pathway, in breast cancer cell lines. Since it has been shown that some miRNAs concurrently target functionally related genes to drive a specific biological sig- Typically, this pathway is activated in response to double- nal (Cloonan et al. 2008; Tsang et al. 2010; Ulitsky et al. 2010; stranded breaks (DSBs) leading to cell cycle arrest at either Su et al. 2011) and miR-182-5p targets are enriched for genes the p53-dependent G1/S-phase checkpoint or the p53-inde- involved in DNA damage repair, we hypothesized that miR- pendent G2/M-phase DNA damage checkpoint (Moynahan 182-5p’s action on the HR pathway extends beyond the tar- et al. 1999; Moynahan and Jasin 2010), and regulates a specific geting of BRCA1. miR-182-5p targeting of CHEK2, an up- set of gene products involved in DNA repair, such as BRCA1 stream regulator of BRCA1, should also alter sensitivity to (Khanna and Jackson 2001). Disruption to this pathway has the HR-mediated repair. Direct targeting of the predicted been associated with a wide variety of malignancies including miR-182-5p-binding site in CHEK2 was confirmed by dual breast, ovarian, and pancreatic cancer (Lord and Ashworth luciferase assay (P < 0.05) (Fig. 1D). We then used the 2012). Figure 4 shows an overview of this pathway and its BRCA1 wild-type MDA-MB-231 cells transiently transfected downstream effects, with the pull-down-identified targets of with miR-182-5p mimics in PARP inhibition assays, using miR-182-5p highlighted in dark gray. Of the 54 genes in different concentrations of ANI (PARP1 inhibitor 4-amino- this pathway, we have now identified 36 (66.66%) as likely tar- 1,8-naphthalimide) ranging from 0 to 10 μM. These cells gets of miR-182-5p (Fig. 4; Supplemental Table 4). were also shown to have an ∼1000-fold increase in the ex- A majority (32 of 36) of the miR-182-5p targets in this pression of miR-182-5p relative to the negative control mim- pathway are proteins that play a major role in the positive reg- ic in the MDA-MB-231 cells (Fig. 5B). PARP inhibitors are

234 RNA, Vol. 19, No. 2 Downloaded from rnajournal.cshlp.org on April 13, 2014 - Published by Cold Spring Harbor Laboratory Press

miR-182-5p targets the DNA repair pathway

Resected DSB/ssDNA

Cell Growth Nuclear Export & Survival YWHAZ RAD1/9A RAD17 RFC YWHAE TP53BP1 HUS1 CDC25B/C CREB5 Homologous RFC MSH2 Recombination CHEK1 ATM/ATR RBBP8 MRN Complex MRE11A RAD50 TLK2 NBN H2AFX CHEK2 BRCA1 BARD1 Chromatin Assembly DNA Repair CDC25A RNA PolII Regulation of RB1 HDAC3 TP53 SMARCD3 Transcription CDK2 CDKN1A ATF1 TUBG1

G1S Arrest Chromatin Remodelling/ Transcriptional Regulation TOP2A GADD45A CCND1 CCNA

CDK4/6 CDK2 CCNH Transcriptional response to DNA damage CDK7

TFDP1 E2F RB1 Ub Proteasomal CDKN1B CDKN1B Degradation

P E2F RB1 SKP1 CUL1 S Phase genes SCF Complex (CDC2, E2F, CCNA) F-box/ SKP2 G1S progression Cell cycle BTRC progression

FIGURE 4. Overview of biotinylated miR-182-5p pull-down-predicted targets involved in the DNA damage response pathway genes involved in different canonical pathways underlying the DNA damage response, including the G2M cell cycle checkpoint, BRCA1-dependent HR-mediated pathway, cyclins and role of cell cycle regulators, role of CHK proteins, and ATM signaling. (Dark gray) miR-182-5p targets identified by biotin pull-down. These canonical pathways were found to be significantly enriched compared with 10 random gene lists of similar size using IPA (P-value < 0.02). cytotoxic in cells deficient in HR-mediated repair, because Validation of predicted binding sites the inhibitors suppress base excision repair (BER), which in MDA-MB-231 cells would normally compensate an HR deficiency. Increased sensitivity to PARP inhibitors would therefore indicate a re- Since the original target identification analysis was performed duction in HR function, and reintroduction of a miR-182-5p in the nonmalignant HEK293T cell lines, we sought to vali- target ORF should rescue the phenotype (Fig. 5A). date the interaction between miR-182-5p and some of the As previously reported (Moskwa et al. 2011), overexpres- genes enriched in the biotin pull-down using MDA-MB- sion of miR-182-5p sensitized the cells to PARP inhibition, 231 breast cancer cells. We selected genes for which there as measured by clonogenic survival assays. We confirmed was a single predicted binding site, for validation in luciferase that reintroduction of the BRCA1 ORF could rescue this phe- assays. The predicted binding sites (and ∼60 nt of surround- notype (Fig. 5C) with a P-value < 0.05 at 0.01 μM and 0.1 μM ing sequence) were cloned into the 3′ UTR of the Dual showing 10%–20% increase in rescue of the cells. Our results Luciferase pmiRGlo vector and transiently transfected into also demonstrate that the phenotype can be rescued by cells. Luciferase activity, indicative of translation from the CHEK2, where similar results were obtained with 20%– plasmid, was measured in the presence of miR-182-5p mimic 30% increase in survival at the 0.1 and 1 μM concentrations of or negative control mimic and normalized using Renilla ac- the PARP inhibitor (P-value < 0.05), indicating that CHEK2 tivity. Using this approach, we were able to validate seven is another target of miR-182-5p that contributes to this out of eight of the targets picked (87.5%) (Fig. 6), including phenotype. ATF1, RAD17, CHEK2, SMARD3, CREB5, TP53BP1, and

www.rnajournal.org 235 Downloaded from rnajournal.cshlp.org on April 13, 2014 - Published by Cold Spring Harbor Laboratory Press

Krishnan et al.

A PARPi PARP miR-182-5p 5’ 3’ 5’ 3’ mimic 5’ 3’

3’ 5’ 3’ 5’ 3’ 5’ Single Stranded Breaks Double Stranded Breaks DSBs persist cell death PARP miR-182-5p mimic target ORF

5’ 3’ 5’ 3’

3’ 5’ 3’ 5’ Base Excision Repair Homologous Recombination cell survival B C 100000 1.2 miR-182-5p mimic

10000 1 miR-182-5p mimic + BRCA1 ORF miR-182-5p mimic + 0.8 ** CHEK2 ORF 1000 *** Control mimic

Control mimic + 0.6 BRCA1 ORF 100 ** Control mimic + CHEK2 ORF 0.4

Fraction of Surviving Cells 0.2 Relative miR-18-5p expression miR-18-5p Relative

1 0 miR-182-5p negative 0 0.01 0.1 1 10 mimic mimic Concentration of PARP inhibitor (μM)

FIGURE 5. The effect of miR-182-5p on sensitivity to PARP inhibition in MDA-MB-231cells. (A) A schematic diagram depicting the principle of the PARP inhibitor assays performed here. Typically, cells with functional PARP repair single-strand breaks via the base excision repair pathway. In PARP inhibition assays, the inhibited PARP leads single-stranded breaks to decay to double-stranded breaks, which are then repaired via the homologous recombination (HR) pathway. However, in cells with a dysfunctional HR pathway (i.e., overexpressing miR-182-5p mimic leading to down-regulation of the DNA damage response [DDR] components), the DNA damage persists, leading to cell death. If a DDR target of an miR-182-5p target (e.g., BRCA1 or CHEK2 open reading frame [ORF]) is reintroduced, this would rescue any effect of miR-182-5p leading to increased cell survival. (B)To assess the level of miR-182-5p overexpression, we performed real-time analysis in MDA-MB-231 cells transiently transfected with 10 nM miR-182-5p or negative control mimic and find an ∼1000-fold increase from the base level expression in these breast cancer cells. (C) BRCA1 and CHEK2 mediate sensitivity to PARP1 inhibitor, induced by overexpression of miR-182-5p. Cells were transiently transfected with miR-182-5p mimic or control mimic (±ORF). Cell viability was assessed using the clonogenic survival assay in the presence of 4-amino-1,8-naphthalimide (ANI; PARP1 inhibitor) at the indicated concentrations on the x-axis. The data plotted are the mean and SEM of at least three independent biological replicates. (∗) P < 0.05 in a Student’s t-test.

CDKN1B. These data further strengthen our hypothesis that Queensland Centre for Clinical Research, UQCCR). The sam- miR-182-5p targets the DNA repair pathway through a net- ple cohort (n = 40) included “Invasive Ductal Carcinomas– work of functionally related genes. Additionally, the high No Special Type” (IDC-NST) of different molecular subtypes rate of validation seen here contrasts favorably with valida- including triple negative (n = 18); Her2+ (n = 4), ER+/PR+ (n tion rates based on TargetScan predictions alone (40%) = 9); Invasive Lobular Carcinomas (ILC) (n = 3); and normal (Cloonan et al. 2008), and confirms the ability of the biotin breast tissue (n = 6) (Supplemental Table 5). We included pull-down approach to enrich for genuine biological targets. several molecular subtypes of breast cancer and assayed the expression levels of miR-182-5p relative to an endogenous control RNU6B, using qRT-PCR. As shown in Figure 7 (upper miR-182-5p is frequently up-regulated in human panel), miR-182-5p is highly expressed (log fold-change breast cancer 10 [tumor/normal] >1.5) in 32/40 tumor samples (∼80%) as- Although miR-182-5p has been shown to be important for sayed relative to the normal controls, and all but one sample development of breast cancer in cell lines and mice, no screens had higher expression in the tumors than any of the normal have been performed to determine its relevance to human controls. The average expression of miR-182-5p in every tu- breast cancer subtypes. We extended the analysis of miR- mor subtype is significantly higher (P < 0.0001) than the av- 182-5p expression to a cohort of human breast cancer patient erage expression across the normal breast tissue (Fig. 7, samples (from the Brisbane Breast Bank, The University of upper panel). Expression of miR-182-5p is highest across

236 RNA, Vol. 19, No. 2 Downloaded from rnajournal.cshlp.org on April 13, 2014 - Published by Cold Spring Harbor Laboratory Press

miR-182-5p targets the DNA repair pathway

miR-182-5p -ve control p<0.05 1.15 ** et al. 2011). Gene set enrichment analysis (GSEA) of miR- p<0.1 1.10 * 182-5p targets revealed that the DNA damage response was 1.05 a core molecular pathway regulated by this miRNA. This pathway is critical in the normal functioning and replication 1.00 ** ** ** ** **** ** 0.95 of cells, and disruption can cause cellular transformation. 0.90 PARP plays a key role in DNA repair via the base excision re- 0.85 pair (BER) pathway. When PARP is inhibited, single-strand 0.80 breaks (SSBs) degenerate to lethal DSBs, which, in the case 0.75 of BRCA-negative cells or cells deficient in HR, leads to 0.70 cell death (Bryant et al. 2005; McCabe et al. 2006). 0.65 Importantly, we were able to partially rescue miR-182-5p- 0.60 0.55

0.50 10000 **** 0.45 Relative Luciferase Activity 0.40 1000 0.35 100 0.30

0.25 10 0.20 0.15 1 0.10 0.1 0.05 miR-182-5p expression relative to RNU6B miR-182-5p expression Triple Negative HER2+ ER+ PR+ ILC Normal 0.00 pmirGlo ATF1 BARD1 CREB5 RAD17 TP53BP1 CHEK2 CDKN1B SMARCD3 1000000 FIGURE 6. Dual luciferase assay used to validate targets of miR-182-5p. * MDA-MB-231 cells were transiently cotransfected with 20 nM miR- 182-5p or control mimic with a pmirGlo-luciferase construct containing the predicted binding site from the indicated target gene. Luciferase activity was normalized to Renilla activity; (∗∗) P < 0.05 as indicated in a Student’s t-test. The data plotted are the mean and SEM of three independent biological replicates with three technical 100000 replicates. the ER+/PR+ luminal subtype; however, the expression of miR-182-5p is highly variable across the triple negative tumors, perhaps indicative of the heterogeneity within this class. 10000 To confirm the relevance of miR-182-5p in an indepen-

dent cohort of human breast cancers, we downloaded (Reads per million) miR-182-5p expression miRNA-seq data from the Cancer Genome Atlas (TCGA) data portal (https://tcga-data.nci.nih.gov/tcga/), which included both tumor (n = 741) and normal tissues (n = 99). 1000 The relative abundance of miRNAs was counted, and shown Tumour Normal in Figure 7 (lower panel) is the reads per million of miR-182- FIGURE 7. Expression analysis of miR-182-5p across tumor subtypes 5p across both the tumor and normal cohort, where we find a and normal tissue from human breast cancer patient samples. (Upper significant difference (P < 0.0001) in its level of expression. panel) Real-time PCR was performed in breast cancer patient samples These results confirm that miR-182-5p is an important of various subtypes to assess miR-182-5p expression, where lines and error bars represent the mean and SEM of miR-182-5p expression (nor- miRNA relevant to human cancer and that further study is malized to RNU6B) across sample subtypes: Invasive Ductal warranted to clearly delineate its role in the pathogenesis Carcinoma of Triple Negative (n − 19); Her2+ (n =4) or ER+/PR+ and progression in specific breast cancer subtypes. (n = 9) phenotype; Invasive Lobular Carcinomas (n = 3); and normal breast tissue (n = 6). (▪) Expression levels for individual patients; (∗) indicate where the difference in the means subtype was (P < 0.0001) when compared with the normal samples. (Lower panel) Expression of miR- DISCUSSION 182-5p across several breast cancer patient samples as assessed by In this study, we have characterized transcriptome-wide miRNA-seq. Shown in the x-axis is reads per million of miR-182-5p in data downloaded from the TCGA web portal, which included tumors mRNA targets of miR-182-5p using the affinity copurifica- (n = 741) and normal tissue (n = 99). (∗) Where the difference in the tion of mRNA bound to biotin-labeled miRNAs (Cloonan means was (P < 0.0001) when compared with the normal samples.

www.rnajournal.org 237 Downloaded from rnajournal.cshlp.org on April 13, 2014 - Published by Cold Spring Harbor Laboratory Press

Krishnan et al. induced sensitivity to PARP inhibitors by restoring either of this pathway. This would increase the burden of mutation BRCA1 or CHEK2 expression. Moskwa et al. (2011) previ- in these cells, possibly quite slowly, allowing the cells to ac- ously concluded that BRCA1 was the key gene responsible commodate and adapt to a modestly increasing mutation for the miR-182-5p-induced sensitivity for PARP inhibitors, load. Such a hypothesis, while plausible, has not been directly because 100% reversal of the sensitivity was observed upon tested in this study. The second (but not mutually exclusive) reintroduction of BRCA1. However, the overexpression of hypothesis is that miR-182-5p also directly suppresses the BRCA1 under a strong promoter could easily compensate apoptotic pathway triggered by DNA damage. CDKN1B for the relatively weaker disruption to the pathway achieved (p27KIP1) was a target of miR-182-5p identified by the biotin by induction of an miRNA, and hence mask the contribution pull-down and validated using luciferase assay. This gene reg- of other pathway members. Our results directly demonstrate ulates G0-to-S phase transitions by interacting with cyclin- that CHEK2 contributes to the miR-182-5p-induced sensi- dependent kinases (CDKs) (Hengst and Reed 1998; Sherr tivity to PARP inhibition, but also suggest that other mem- and Roberts 1999), and overexpression of this gene has bers of the pathway could contribute as well, confirmed by been shown to trigger apoptosis in the MDA-MB-231 cell the luciferase assay validation in the breast cancer cells (Fig. line (Katayose et al. 1997) with pro-apoptotic properties in 6). Together, repression of DNA damage repair could lead other cancers (Fujieda et al. 1999; Wu et al. 1999; Tenjo to genomic instability followed by cellular transformation et al. 2000). Apoptosis has recently been identified as one (Fig. 8). of the deregulated pathways by the miR-182-5p cluster in One of the well-studied consequences of suppressing the medulloblastoma (Weeraratne et al. 2012), and suppression repair of DNA damage is the activation of apoptotic path- of CDKN1B by miR-182-5p, along with other pro-apoptotic ways, and this seems counterintuitive for a potential driver genes like BAK and BAX (Supplemental Table 1), suggests of tumorigenesis. There are two hypotheses that could ex- that apoptosis triggered by DNA damage could be deregulat- plain this apparent contradiction. The first is that the com- ed by this miRNA. bined effect of miR-182-5p targeting across a broad range Not all targets of miR-182-5p easily fit into the model pro- of genes involved in DNA damage repair would manifest as posed here (Fig. 8). Some targets—like BTRC, SKP1, SKP2, a reduction in the fidelity of DNA repair, rather than a failure and the components of the SCF complex—are suppressors of DNA damage repair, and miR-182-5p suppression of these proteins would act to ensure genome fidelity. Relevant to this DSBs point are the conflicting reports regarding miR-182-5p’smo- lecular role in tumorigenesis. Whereas there appears to be an oncogenic role for this miRNA in some cancers including Sensors melanoma (Segura et al. 2009), endometrioid endometrial cancer (Myatt et al. 2010), and glioma (Jiang et al. 2010);

BRCA1 in others, such as lung adenocarcinoma (Sun et al. 2010; Transducers CHEK2 Zhang et al. 2011) and human gastric adenocarcinoma TP53BP1 miR-182-5p (Kong et al. 2012), its role is more akin to that of a tumor sup- ATF1 pressor. Such dual-function miRNAs have been previously RAD17 Effectors SMARCD3 reported: miR-26a (Sander et al. 2008; Huse et al. 2009; CDKN1B Kota et al. 2009; Kim et al. 2010), miR-205 (Iorio et al. DNA repair 2007, 2009; Gandellini et al. 2009; Wu et al. 2009), and miR-17-5p (Hossain et al. 2006; Mraz et al. 2009; Yu et al. 2010; Li et al. 2011) are all characterized examples. In the Oncogene Growth Genetic Apoptosis activation/ Arrest Instability latter case, the molecular mechanism underlying the dual TSG mutation phenotype was uncovered through systematic screening of ? predicted targets through luciferase assays (Cloonan et al. 2008). Targeting both inhibitors and activators of DNA dam- Cancer age repair could either promote or inhibit genomic stability depending on the relative expression levels of those targets FIGURE 8. Model showing the effect of miR-182-5p on the DNA re- —and this could explain the conflicting reports of miR- “ pair pathway. When a cell undergoes double-strand breaks (DSB), sen- 182-5p’s role in tumorigenesis. For a firm conclusion to be sor” genes activate a signaling cascade including transducer and effector genes that leads to an efficient repair of the DNA damage. Under this made, independent validation of each target would be re- model, miR-182-5p-mediated deregulation of the DNA damage required. While the biotin pull-down has been optimized for sponse pathway (orange) typically results in impaired DNA repair high specificity, the presence of false-positive targets in the with subsequent effects on the cell cycle, apoptosis, or genetic stability enrichment analysis should be considered while interpreting leading to tumorigenesis. We have also highlighted previously validated and novel targets that we have mechanistically shown to interact with these results. Ongoing work suggests that the rate of false pos- miR-182-5p, belonging to the DNA damage response pathway. itives is ≤5% (data not shown), which would not be high

238 RNA, Vol. 19, No. 2 Downloaded from rnajournal.cshlp.org on April 13, 2014 - Published by Cold Spring Harbor Laboratory Press

miR-182-5p targets the DNA repair pathway enough to interfere with GSEA but may become critical when mediated DNA repair pathway, which is the major determi- considering individual targets. nant of a cell’s sensitivity to PARP inhibition. Further studies Many of the key genes involved in DNA repair also have are required to address if the status of BRCA mutation and annotated roles in the cell cycle, and GSEA correctly identi- overexpression of miR-182-5p are mutually exclusive and if fied the cell cycle as an important molecular function of either one is sufficient to make cells PARP-inhibition sensi- miR-182-5p targets. Although we found that ectopic expres- tive. This could include nonbreast cancer tumors, such as sion of miR-182-5p was not sufficient to drive proliferation ovarian cancer, where several sporadic cancer types display (confirming earlier reports) (Moskwa et al. 2011), we did a “BRCA-like” phenotype (Turner et al. 2004). Intriguingly, observe minor cell cycle effects consistent with a DNA dam- overexpression of miR-182-5p has recently been shown to age response. The HR-mediated repair is restricted to the target BRCA1 in ovarian papillary serous carcinoma (Liu S and G2 phases of the cell cycle (Moynahan et al. 1999; et al. 2012). An important subject of future studies would Moynahan and Jasin 2010) by several factors and detection be to determine how well miR-182-5p performs as a prog- of double-strand breaks would typically result in a G2 arrest. nostic or therapeutic biomarker. In our analysis with overexpression of miR-182-5p in HeLa cells (which would suppress this checkpoint), we see fewer MATERIALS AND METHODS cells in both the S and G2 phases with increased accumulation in the G1 phase. This could be a result of a disrupted G2 arrest Cell culture as a result of the deregulated HR-mediated repair pathway (Fig. 8). Other studies have shown the miR-182-5p in con- HEK293T, MDA-MB-231, and HeLa cells were maintained in junction with miR-96 and miR-27a (Guttilla and White DMEM (Life Technologies) with 10% FBS and 1% Penicillin– 2009) can alter proliferation rates in MCF-7 breast cancer Streptomycin (Life Technologies) and grown in a 5% CO2 atmo- cells; hence, it could be that concomitant targeting of genes sphere at 37°C. MDA-MB-231 and HeLa cell lines were purchased from Cell Bank; the HEK293T cell line was purchased from ATCC. by all members of the miR-182-5p cluster and other mature miRNAs is essential to drive this phenotype. We confirm the relevance of miR-182-5p dysregulation to Biotin pull-downs, microarray hybridizations, human breast cancer by showing that this miRNA is overex- and analysis pressed across a panel of human breast cancer samples, which Pull-downs of miR-182-5p targets were carried out as previously de- belong to several molecular and pathological subtypes. Al- scribed (Cloonan et al. 2011), using biotin-labeled oligonucleotides though overexpressed in the triple negative subtype, the var- specific for miR-182-5p (Supplemental Table 6). Briefly, 50 pmol of iability of miR-182-5p expression within this subtype was biotin-labeled oligos (IDT) was transiently transfected into substantially more than seen in any other classification. Tri- HEK293T cells and cultured for 24 h. This was followed by cell lysis ple negative tumors are typically characterized by BRCA defi- and binding of 50 μL (Streptavidin), myOne C1 Dynabeads ciency and a disrupted HR pathway. Given miR-182-5p’s role (Invitrogen) to the RNA fraction for enrichment. Fifty nanograms in HR-mediated DNA repair, it is possible that transcription of captured mRNA fractions (three independent biological repli- factors that typically dysregulate protein components of this cates) was amplified and labeled using the Illumina Total Prep ’ pathway are also dysregulating miR-182-5p transcription. In RNA amplification kit (Ambion) as per the manufacturer s instructions. Samples were profiled on Illumina Human HT-12 chips along melanoma, miR-182-5p has been associated with the inva- with control RNA from mock-transfected cells. sion/metastatic signaling cascade (Segura et al. 2009). We Microarray data were normalized using the lumi package (Du did not find a strong association between miR-182-5p and the et al. 2008) by applying background adjustment, variance-stabilizing metastatic lymph node status of samples, suggesting that its transformation (Lin et al. 2008), and robust spline normalization role in breast cancer is more likely tumorigenesis than tumor (Workman et al. 2002) successively. The lmFit and eBayes functions progression. However, further studies using matched prima- in the limma package (Smyth 2004) were used to test differential ex- ry and distant metastatic samples followed by functional val- pression between the pull-down samples and the controls (Cloonan idations are required to elucidate miR-182-5p’s molecular et al. 2011). The false discovery rate (FDR) was calculated to account role in each cancer subtype. for multiple testing (Benjamini and Hochberg 1995). Probes that Of particular relevance to therapeutic biomarkers is miR- met the 5% FDR threshold (for one-sided tests) were considered sig- 182-5p’s ability to induce sensitivity to PARP inhibitors nificantly enriched in the pull-down. The transcripts (ENSEMBL V62) to which they matched exactly were considered putative targets through multiple effectors. This result could be of immense of that miRNA. The targets enriched using the biotin pull-downs therapeutic value, potentially widening the opportunity to were analyzed using Ingenuity Pathway Analysis as previously de- expand treatment from breast cancers with mutations in scribed (Cloonan et al. 2008). BRCA1/2 to tumors with miR-182-5p disrupted HR-mediated pathways. It is currently not clear what percentage of the sporadic breast tumor patients with a functional HR pathway Stable cell line generation would respond to treatment with PARP inhibitors. We have MDA-MB-231 cells stably expressing miR-182-5p were generated now shown miR-182-5p as a potential regulator of the HR- using the Mir-X Inducible miRNA Systems (Clontech). Briefly,

www.rnajournal.org 239 Downloaded from rnajournal.cshlp.org on April 13, 2014 - Published by Cold Spring Harbor Laboratory Press

Krishnan et al.

MDA-MB-231 cells were transfected with the pTet-on Advanced luminometer (PerkinElmer). Luciferase activity was normalized to Vector using Lipofectamine 2000 (Life Technologies Australia, Invi- the internal control, Renilla activity in each well. Assays were con- trogen Division), and cells stably expressing the plasmid were select- ducted in triplicate and independently repeated three times. ed using 800 μg/mL G418 (Life Technologies); maintenance concentration: 400 μg/mL G418. Primers (Supplemental Table 6) were used to amplify the miR-182-5p hairpin from human genomic MTT proliferation assays DNA and cloned into the pmRI-ZsGreen I vector plasmid Stable pmRi-MDA-MB-231 cell lines overexpressing miR-182-5p supplied and subsequently transfected into the 231-pTet-on parent (1000 ng/mL doxycycline) and parent stables (with no doxycycline) line. Cells stably expressing pmRi-Zsgreen-miR-182-5p were select- were plated at 1 × 104 cells per well. MTT (3-[4,5-dimethylthiazol- ed using 1 μg/mL puromycin and further maintained in 0.5 μg/mL 2-yl]-2,5-diphenyl tetrazolium bromide) activity was assayed using puromycin. Stable expression of miR-182-5p was confirmed using a Cell Growth Determination Kit (Sigma-Aldrich) according to the the miRNA Taqman Assay (Applied Biosystems) specific for miR- manufacturer’s instructions and detected on a PowerWave XS spec- 182-5p (Fig. 2A). trophotometer (BioTek).

Clonogenic cell survival assay Clinical samples, RNA purification, and qRT-PCR MDA-MB-231 cells were seeded overnight (2 × 105 cells/well in a analyses 12-well plate) and transfected with 10 nM miRNA mimics Human breast tumors were derived from the Brisbane Breast Bank, (Ambion). In rescue experiments, miR-182-5p or control mimics collected from consenting patients and with ethical approval from were cotransfected with 0.5 μg of BRCA1 or CHEK2 cDNA clones. the research ethics committees of the Royal Brisbane & Women’s After 48 h, 500 cells in 2 mL of DMEM media (10% FBS, v/v) Hospital and the University of Queensland. Histological type, tumor were seeded on six-well plates in triplicate and incubated overnight grade, tumor size, lymph node status, and ER, PR, and HER2 status before treatment. PARP inhibitors (4-amino-1, 8-naphthalimide were obtained from the pathology reports. ER, PR, and HER2 bio- [Sigma-Aldrich] in DMSO) were added to the growth media at 0 markers were used to infer molecular subtype as luminal, HER2, or μM, 0.01 μM, 0.1 μM, 1 μM, and 10 μM concentrations. Cells in triple negative. Total RNA from human tumor samples was extract- the presence of PARP inhibitor were allowed to form colonies for ed using tumor homogenization followed by TRIzol extraction 14 d. For evaluation, formed colonies were stained with Crystal Vio- (Invitrogen). let and surviving colonies containing more than 50 cells were count- Total RNA was purified from cell lines using the miRNeasy Mini – ed. The plating efficiency was 20% 35%. Kit (QIAGEN), and RNA integrity was assessed using an Agilent Bioanalyzer 2100. For mature miRNA, cDNA (5–10 ng of total RNA) was synthesized using a Taqman MicroRNA RT Kit Flow cytometry for cell cycle analysis (Applied Biosystems), and qRT-PCR was performed using a miR- 182-5p MicroRNA Taqman assay (Applied Biosystems). For HeLa cells were transiently transfected using 50 nM miR-182-5p mRNA expression analysis, 500 ng of total RNA was reverse-tran- mirVana mimic (Life Technologies) using Lipofectamine 2000 scribed using SuperScript III (Invitrogen), and a 1:50 dilution of (Invitrogen) as per the manufacturer’s instructions. All cells were the cDNA was used in the real-time PCR reaction. All RT-PCR harvested and fixed in 70% ethanol overnight at −20°C. DNA was was performed on an Applied Biosystems 7000 Sequence stained using 10 μg/mL propidium iodide (Sigma-Aldrich), and Detection System. For small RNA expression analysis, RNU6B was RNA was removed using 200 μg/mL RNase A (Sigma-Aldrich). used as an endogenous control to normalize the data. Cells were filtered through a 35-μm cell strainer mesh (BectonDickinson) and analyzed on Becton Dickinson LSR II flow cytometer fitted with a 488-nm laser. Cell data were gated and ana- DATA DEPOSITION lyzed using FlowJo 7.2.2 (Tree Star, Inc.). The same fixing and FACS protocol was applied to MDA-MB-231 cells stably overex- The raw microarray data used in this study are available from pressing miR-182-5p. the Gene Expression Omnibus (GEO) under accession number GSE38593.

Dual luciferase assay to validate predicted binding sites SUPPLEMENTAL MATERIAL Predicted target sites of miR-182-5p were cloned into the Nhe1 and SalI sites of pmirGLO Dual-Luciferase miRNA Target Expression Supplemental material is available for this article. Vector (Promega). Synthetic oligos (Supplemental Table 6) corresponding to 60 nucleotides surrounding the target sequence were ACKNOWLEDGMENTS annealed before ligation into the pmirGlo plasmid. All constructs were verified by sequencing. HEK293T or MDA-MB-231 cells This work was partially supported by Australian Research Council were cotransfected with 50 ng of a pmirGlo construct and miR- (ARC) Discovery Project Grant DP1093164. K.K. is supported by 182-5p or negative mimic (Ambion) to a final concentration of an Australian Post-graduate Award (APA); P.T.S. is supported by 20 nM. Post-transfection, cells were incubated for 48 h prior to as- a fellowship from the National Breast Cancer Foundation, saying. Luciferase activity was assayed using the Dual Luciferase Australia; N.C. is supported by an ARC Postdoctoral Fellowship; Reporter Assay System (Promega) and detected on a Wallac 1420 and S.M.G. is supported by a National Health and Medical

240 RNA, Vol. 19, No. 2 Downloaded from rnajournal.cshlp.org on April 13, 2014 - Published by Cold Spring Harbor Laboratory Press

miR-182-5p targets the DNA repair pathway

Research Council (NHMRC) Principal Research Fellowship. The Harper JW, Elledge SJ, Keyomarsi K, Dynlacht B, Tsai LH, Zhang P, Arrayed Retroviral Expression Cloning (ARVEC) facility kindly sup- Dobrowolski S, Bai C, Connell-Crowley L, Swindell E, et al. 1995. plied the BRCA1 and CHEK2 ORFs. We are grateful for the helpful Inhibition of cyclin-dependent kinases by p21. Mol Biol Cell 6: 387–400. discussions with all members of QCMG, and we are particularly He L, Thomson JM, Hemann MT, Hernando-Monge E, Mu D, thankful to John Pearson, Darrin Taylor, and Scott Wood for Goodson S, Powers S, Cordon-Cardo C, Lowe SW, Hannon GJ, HPC infrastructure and support. et al. 2005. A microRNA polycistron as a potential human oncogene. Nature 435: 828–833. Received June 13, 2012; accepted November 14, 2012. Hengst L, Reed SI. 1998. Inhibitors of the Cip/Kip family. Curr Top Microbiol Immunol 227: 25–41. Hossain A, Kuo MT, Saunders GF. 2006. Mir-17-5p regulates breast REFERENCES cancer cell proliferation by inhibiting translation of AIB1 mRNA. Mol Cell Biol 26: 8191–8201. Baek D, Villen J, Shin C, Camargo FD, Gygi SP, Bartel DP. 2008. The Huse JT, Brennan C, Hambardzumyan D, Wee B, Pena J, impact of microRNAs on protein output. Nature 455: 64–71. Rouhanifard SH, Sohn-Lee C, le Sage C, Agami R, Tuschl T, et al. Bartel DP. 2009. MicroRNAs: Target recognition and regulatory func- 2009. The PTEN-regulating microRNA miR-26a is amplified in tions. Cell 136: 215–233. high-grade glioma and facilitates gliomagenesis in vivo. Genes Dev Benjamini Y, Hochberg Y. 1995. Controlling the false discovery rate: 23: 1327–1337. A practical and powerful approach to multiple testing. J R Stat Soc Iorio MV, Ferracin M, Liu CG, Veronese A, Spizzo R, Sabbioni S, Series B Stat Methodol 57: 289–300. Magri E, Pedriali M, Fabbri M, Campiglio M, et al. 2005. Bentwich I. 2005. Prediction and validation of microRNAs and their MicroRNA gene expression deregulation in human breast cancer. targets. FEBS Lett 579: 5904–5910. Cancer Res 65: 7065–7070. Brennecke J, Stark A, Russell RB, Cohen SM. 2005. Principles of Iorio MV, Visone R, Di Leva G, Donati V, Petrocca F, Casalini P, microRNA–target recognition. PLoS Biol 3: e85. doi: 10.1371/ Taccioli C, Volinia S, Liu CG, Alder H, et al. 2007. MicroRNA signa- journal.pbio.0030085. tures in human ovarian cancer. Cancer Res 67: 8699–8707. Bryant HE, Schultz N, Thomas HD, Parker KM, Flower D, Lopez E, Iorio MV, Casalini P, Piovan C, Di Leva G, Merlo A, Triulzi T, Ménard S, Kyle S, Meuth M, Curtin NJ, Helleday T. 2005. Specific killing of Croce CM, Tagliabue E. 2009. microRNA-205 regulates HER3 in BRCA2-deficient tumours with inhibitors of poly(ADP-ribose) po- human breast cancer. Cancer Res 69: 2195–2200. lymerase. Nature 434: 913–917. Jiang L, Mao P, Song L, Wu J, Huang J, Lin C, Yuan J, Qu L, Cheng SY, Cho WC, Chow AS, Au JS. 2009. Restoration of tumour suppressor Li J. 2010. miR-182 as a prognostic marker for glioma progression hsa-miR-145 inhibits cancer cell growth in lung adenocarcinoma and patient survival. Am J Pathol 177: 29–38. patients with epidermal growth factor receptor mutation. Eur J Katayose Y, Kim M, Rakkar AN, Li Z, Cowan KH, Seth P. 1997. Cancer 45: 2197–2206. Promoting apoptosis: A novel activity associated with the cyclin-de- Cloonan N, Brown MK, Steptoe AL, Wani S, Chan WL, Forrest AR, pendent kinase inhibitor p27. Cancer Res 57: 5441–5445. Kolle G, Gabrielli B, Grimmond SM. 2008. The miR-17-5p Khanna KK, Jackson SP. 2001. DNA double-strand breaks: Signaling, microRNA is a key regulator of the G1/S phase cell cycle transition. repair and the cancer connection. Nat Genet 27: 247–254. Genome Biol 9: R127. doi: 10.1186/gb-2008-9-8-r127. Kim H, Huang W, Jiang X, Pennicooke B, Park PJ, Johnson MD. 2010. Cloonan N, Wani S, Xu Q, Gu J, Lea K, Heater S, Barbacioru C, Integrative genome analysis reveals an oncomir/oncogene clus- Steptoe AL, Martin HC, Nourbakhsh E, et al. 2011. MicroRNAs ter regulating glioblastoma survivorship. Proc Natl Acad Sci 107: and their isomiRs function cooperatively to target common biolog- 2183–2188. ical pathways. Genome Biol 12: R126. doi: 10.1186/gb-2011-12-12- Kloosterman WP, Plasterk RH. 2006. The diverse functions of micro- r126. RNAs in animal development and disease. Dev Cell 11: 441–450. Cuadrado M, Gutierrez-Martinez P, Swat A, Nebreda AR, Fernandez- Kong WQ, Bai R, Liu T, Cai CL, Liu M, Li X, Tang H. 2012. MicroRNA- Capetillo O. 2009. p27Kip1 stabilization is essential for the mainte- 182 targets cyclic adenosine monophosphate responsive element nance of cell cycle arrest in response to DNA damage. Cancer Res binding protein 1 (CREB1) and suppresses cell growth in human 69: 8726–8732. gastric adenocarcinoma. FEBS J 279: 1252–1260. Cui Q, Yu Z, Purisima EO, Wang E. 2006. Principles of microRNA reg- Kota J, Chivukula RR, O’Donnell KA, Wentzel EA, Montgomery CL, ulation of a human cellular signaling network. Mol Syst Biol 2: 46. Hwang HW, Chang TC, Vivekanandan P, Torbenson M, Clark KR, doi: 10.1038/msb4100089. et al. 2009. Therapeutic microRNA delivery suppresses tumorigen- Deng CX. 2006. BRCA1: Cell cycle checkpoint, genetic instability, DNA esis in a murine liver cancer model. Cell 137: 1005–1017. damage response and cancer evolution. Nucleic Acids Res 34: Krek A, Grün D, Poy MN, Wolf R, Rosenberg L, Epstein EJ, 1416–1426. MacMenamin P, da Piedade I, Gunsalus KC, Stoffel M, et al. 2005. Du P, Kibbe WA, Lin SM. 2008. lumi: A pipeline for processing Illumina Combinatorial microRNA target predictions. Nat Genet 37: 495–500. microarray. Bioinformatics 24: 1547–1548. Lagos-Quintana M, Rauhut R, Meyer J, Borkhardt A, Tuschl T. 2003. Frankel LB, Christoffersen NR, Jacobsen A, Lindow M, Krogh A, New microRNAs from mouse and human. RNA 9: 175–179. Lund AH. 2008. Programmed cell death 4 (PDCD4) is an important Lassot I, Ségéral E, Berlioz-Torrent C, Durand H, Groussin L, Hai T, functional target of the microRNA miR-21 in breast cancer cells. J Benarous R, Margottin-Goguet F. 2001. ATF4 degradation relies β Biol Chem 283: 1026–1033. on a phosphorylation-dependent interaction with the SCF TrCP Fujieda S, Inuzuka M, Tanaka N, Sunaga H, Fan GK, Ito T, Sugimoto C, ubiquitin ligase. Mol Cell Biol 21: 2192–2202. Tsuzuki H, Saito H. 1999. Expression of p27 is associated with Bax Lewis BP, Burge CB, Bartel DP. 2005. Conserved seed pairing, often expression and spontaneous apoptosis in oral and oropharyngeal flanked by adenosines, indicates that thousands of human genes carcinoma. Int J Cancer 84: 315–320. are microRNA targets. Cell 120: 15–20. Gandellini P, Folini M, Longoni N, Pennati M, Binda M, Colecchia M, Li H, Bian C, Liao L, Li J, Zhao RC. 2011. miR-17-5p promotes human Salvioni R, Supino R, Moretti R, Limonta P, et al. 2009. miR-205 ex- breast cancer cell migration and invasion through suppression of erts tumor-suppressive functions in human prostate through down- HBP1. Breast Cancer Res Treat 126: 565–575. regulation of protein kinase Cε. Cancer Res 69: 2287–2295. Lim LP, Lau NC, Garrett-Engele P, Grimson A, Schelter JM, Castle J, Guttilla IK, White BA. 2009. Coordinate regulation of FOXO1 by miR- Bartel DP, Linsley PS, Johnson JM. 2005. Microarray analysis shows 27a, miR-96, and miR-182 in breast cancer cells. J Biol Chem 284: that some microRNAs downregulate large numbers of target 23204–23216. mRNAs. Nature 433: 769–773.

www.rnajournal.org 241 Downloaded from rnajournal.cshlp.org on April 13, 2014 - Published by Cold Spring Harbor Laboratory Press

Krishnan et al.

Lin SM, Du P, Huber W, Kibbe WA. 2008. Model-based variance-stabi- Si ML, Zhu S, Wu H, Lu Z, Wu F, Mo YY. 2007. miR-21-mediated tu- lizing transformation for Illumina microarray data. Nucleic Acids Res mor growth. Oncogene 26: 2799–2803. 36: e11. doi: 10.1093/nar/gkm1075. Smyth GK. 2004. Linear models and empirical Bayes methods for assess- Liu Z, Liu J, Segura MF, Shao C, Lee P, Gong Y, Hernando E, Wei JJ. ing differential expression in microarray experiments. Stat Appl 2012. MiR182 overexpression in tumorigenesis of high-grade ovari- Genet Mol Biol 3: Article3. doi: 10.2202/1544-6115.1027. an papillary serous carcinoma. J Pathol 228: 204–215. Su WL, Kleinhanz RR, Schadt EE. 2011. Characterizing the role of Lord CJ, Ashworth A. 2012. The DNA damage response and cancer miRNAs within gene regulatory networks using integrative geno- therapy. Nature 481: 287–294. mics techniques. Mol Syst Biol 7: 490. doi: 10.1038/msb.2011.23. Ma L, Teruya-Feldstein J, Weinberg RA. 2007. Tumour invasion and Sun Y, Fang R, Li C, Li L, Li F, Ye X, Chen H. 2010. Hsa-mir-182 sup- metastasis initiated by microRNA-10b in breast cancer. Nature presses lung tumorigenesis through down regulation of RGS17 ex- 449: 682–688. pression in vitro. Biochem Biophys Res Commun 396: 501–507. McCabe N, Turner NC, Lord CJ, Kluzek K, Bialkowska A, Swift S, Tenjo T, Toyoda M, Okuda J, Watanabe I, Yamamoto T, Tanaka K, Giavara S, O’Connor MJ, Tutt AN, Zdzienicka MZ, et al. 2006. Ohtani M, Nohara T, Kawasaki H, Tanigawa N. 2000. Prognostic Deficiency in the repair of DNA damage by homologous recombina- significance of p27kip1 protein expression and spontaneous apoptosis tion and sensitivity to poly(ADP-ribose) polymerase inhibition. in patients with colorectal adenocarcinomas. Oncology 58: 45–51. Cancer Res 66: 8109–8115. Thomas M, Lieberman J, Lal A. 2010. Desperately seeking microRNA Meyerson M, Harlow E. 1994. Identification of G1 kinase activity for targets. Nat Struct Mol Biol 17: 1169–1174. cdk6, a novel cyclin D partner. Mol Cell Biol 14: 2077–2086. Tsang JS, Ebert MS, van Oudenaarden A. 2010. Genome-wide dissec- Moskwa P, Buffa FM, Pan Y, Panchakshari R, Gottipati P, Muschel RJ, tion of microRNA functions and cotargeting networks using gene Beech J, Kulshrestha R, Abdelmohsen K, Weinstock DM, et al. 2011. set signatures. Mol Cell 38: 140–153. miR-182-mediated downregulation of BRCA1 impacts DNA repair Turner N, Tutt A, Ashworth A. 2004. Hallmarks of ‘BRCAness’ in spora- and sensitivity to PARP inhibitors. Mol Cell 41: 210–220. dic cancers. Nat Rev Cancer 4: 814–819. Moynahan ME, Jasin M. 2010. Mitotic homologous recombination Ulitsky I, Laurent LC, Shamir R. 2010. Towards computational predic- maintains genomic stability and suppresses tumorigenesis. Nat Rev tion of microRNA function and activity. Nucleic Acids Res 38: e160. Mol Cell Biol 11: 196–207. doi: 10.1093/nar/gkq570. Moynahan ME, Chiu JW, Koller BH, Jasin M. 1999. Brca1 controls ho- Volinia S, Calin GA, Liu CG, Ambs S, Cimmino A, Petrocca F, Visone R, mology-directed DNA repair. Mol Cell 4: 511–518. Iorio M, Roldo C, Ferracin M, et al. 2006. A microRNA expression Mraz M, Malinova K, Kotaskova J, Pavlova S, Tichy B, Malcikova signature of human solid tumors defines cancer gene targets. Proc J, Stano Kozubik K, Smardova J, Brychtova Y, Doubek M, Natl Acad Sci 103: 2257–2261. et al. 2009. miR-34a, miR-29c and miR-17-5p are down- Weeraratne SD, Amani V, Teider N, Pierre-Francois J, Winter D, regulated in CLL patients with TP53 abnormalities. Leukemia 23: Kye MJ, Sengupta S, Archer T, Remke M, Bai AH, et al. 2012. 1159–1163. Pleiotropic effects of miR-183∼96∼182 converge to regulate cell sur- Myatt SS, Wang J, Monteiro LJ, Christian M, Ho KK, Fusi L, Dina RE, vival, proliferation and migration in medulloblastoma. Acta Brosens JJ, Ghaem-Maghami S, Lam EW. 2010. Definition of Neuropathol 123: 539–552. microRNAs that repress expression of the tumor suppressor gene Workman C, Jensen LJ, Jarmer H, Berka R, Gautier L, Nielser HB, FOXO1 in endometrial cancer. Cancer Res 70: 367–377. Saxild HH, Nielsen C, Brunak S, Knudsen S. 2002. A new non-linear Rajewsky N. 2006. microRNA target predictions in animals. Nat Genet normalization method for reducing variability in DNA microarray 38: S8–S13. experiments. Genome Biol 3: presearch0048–research0048.16. Saito T, Saetrom P. 2010. MicroRNAs—targeting and target prediction. Wu J, Shen ZZ, Lu JS, Jiang M, Han QX, Fontana JA, Barsky SH, N Biotechnol 27: 243–249. Shao ZM. 1999. Prognostic role of p27Kip1 and apoptosis in human Sander S, Bullinger L, Klapproth K, Fiedler K, Kestler HA, Barth TF, breast cancer. Br J Cancer 79: 1572–1578. Möller P, Stilgenbauer S, Pollack JR, Wirth T. 2008. MYC stimulates Wu H, Zhu S, Mo YY. 2009. Suppression of cell growth and invasion by EZH2 expression by repression of its negative regulator miR-26a. miR-205 in breast cancer. Cell Res 19: 439–448. Blood 112: 4202–4212. Yu J, Ohuchida K, Mizumoto K, Fujita H, Nakata K, Tanaka M. 2010. Sarver AL, French AJ, Borralho PM, Thayanithy V, Oberg AL, MicroRNA miR-17-5p is overexpressed in pancreatic cancer, associ- Silverstein KA, Morlan BW, Riska SM, Boardman LA, ated with a poor prognosis, and involved in cancer cell proliferation Cunningham JM, et al. 2009. Human colon cancer profiles show dif- and invasion. Cancer Biol Ther 10: 748–757. ferential microRNA expression depending on mismatch repair status Zhang L, Huang J, Yang N, Greshock J, Megraw MS, Giannakakis A, and are characteristic of undifferentiated proliferative states. BMC Liang S, Naylor TL, Barchetti A, Ward MR, et al. 2006. Cancer 9: 401. doi: 10.1186/1471-2407-9-401. microRNAs exhibit high frequency genomic alterations in human Schaefer A, Jung M, Mollenkopf HJ, Wagner I, Stephan C, Jentzmik F, cancer. Proc Natl Acad Sci 103: 9136–9141. Miller K, Lein M, Kristiansen G, Jung K. 2010. Diagnostic and prog- Zhang L, Liu T, Huang Y, Liu J. 2011. microRNA-182 inhibits the pro- nostic implications of microRNA profiling in prostate carcinoma. liferation and invasion of human lung adenocarcinoma cells through Int J Cancer 126: 1166–1176. its effect on human cortical actin-associated protein. Int J Mol Med Segura MF, Hanniford D, Menendez S, Reavie L, Zou X, Alvarez-Diaz S, 28: 381–388. Zakrzewski J, Blochin E, Rose A, Bogunovic D, et al. 2009. Aberrant Zhu S, Si ML, Wu H, Mo YY. 2007. MicroRNA-21 targets the tumor miR-182 expression promotes melanoma metastasis by repressing suppressor gene tropomyosin 1 (TPM1). J Biol Chem 282: 14328– FOXO3 and microphthalmia-associated transcription factor. Proc 14336. Natl Acad Sci 106: 1814–1819. Zhu S, Wu H, Wu F, Nie D, Sheng S, Mo YY. 2008. MicroRNA-21 tar- Sherr CJ, Roberts JM. 1999. CDK inhibitors: Positive and negative reg- gets tumor suppressor genes in invasion and metastasis. Cell Res 18: ulators of G1-phase progression. Genes Dev 13: 1501–1512. 350–359.

242 RNA, Vol. 19, No. 2

CHAPTER THREE miR-139-5p is a potential tumour suppressor of breast cancer by targeting genes underlying metastasis related pathways

3 miR-139-5p is a potential tumour suppressor of breast cancer by targeting genes underlying metastasis related pathways

3.1 Summary

As discussed in the previous chapter, miRNAs exert their functions through the regulation of a multitude of targets, many of which are components of the same pathway. Through the coordinate repression of these targets genes, they are able to induce global changes to gene expression resulting in the alteration of cellular and organismal phenotypes. In the case of miR-182, identification of its target genes led to uncovering its function in DNA damage repair, which, although alluded to in previous publications from its target BRCA1 [293], was not a complete representation of its function. The following chapter will discuss the role of another miRNA, miR-139-5p, whose function was previously poorly defined based on a few identified targets.

Carcinomas that result from malignant transformation of epithelial cells are the most common types of cancers in humans. 90% of mortalities from this form of cancer are a result of the spreading of the disease to other organs and tissue sites of the body through the process of metastasis [281]. Although the mechanisms involved in this process are poorly understood, there are a series of steps that cancer cells undergo before being able to metastasize to distant tissue, commonly known as the invasion-metastasis cascade [294]. Briefly, the carcinoma cells acquire migratory and invasive properties that enable them to renounce their cell-cell adhesion junctions as well as adherence to the basement membrane at their orthotopic site. This enables them to invade local tissues and intravasate into the bloodstream where they can travel to distant sites. Following this, they extravasate and colonize foreign tissue where they form macrometastases. The mechanisms that govern the initial acquisition of migratory and invasive properties have been under intense investigation, with many potential players identified [295].

The following chapter will discuss the role of miR-139-5p in the suppression of migratory and invasive properties of human breast cancer cells. From data presented in this chapter, miR-139-5p is shown to be downregulated in the most invasive subtypes of human breast cancers that fall under the triple-negative subgroup. Moreover, mining of two other studies uncover its downregulation in aggressive forms of breast cancer. To better understand the role of

81 miR-139-5p in breast cancer progression and metastasis, we carried out biotin pull-downs to identify its target genes. We found its targets to be closely associated with pathways involved in metastases, specifically the TGFbeta and Wnt signaling cascades. Other pathways that were enriched were the PRKC, PI3Kinase and MAP kinase signaling cascades, all of which have been shown to play a role in cancer cell invasion and migration. To functionally test its role in the acquisition of properties required for breast cancer metastasis, we tested the ability of the breast cancer cell line MDA-MB-231 to migrate through a transwell membrane and to invade through a matrigel extracellular matrix layer. We found a significant reduction in the ability of these cells to both migrate and invade, showing the ability of this miRNA to play a major role in the metastatic ability of breast cancer cells. Supplementary tables associated with the journal article are included as Appendix 2 of this thesis. The details of contributions to the following manuscript can be found on page vii under the heading ‘Publications included in this thesis’ This chapter recapitulates the previously known functions of miR-139-5p (from other cancer types) in breast cancer and extends its network of target cohort. Also, miR-139-5p is not only shown to regulate multiple targets but can alter multiple pathways underlying the invasive phenotype, providing another framework as part of the miRNA targeting mechanisms.

82 Downloaded from rnajournal.cshlp.org on April 13, 2014 - Published by Cold Spring Harbor Laboratory Press

miR-139-5p is a regulator of metastatic pathways in breast cancer

KEERTHANA KRISHNAN,1 ANITA L. STEPTOE,1 HILARY C. MARTIN,1,9 DIWAKAR R. PATTABIRAMAN,2 KATIA NONES,1 NIC WADDELL,1 MYTHILY MARIASEGARAM,3 PETER T. SIMPSON,3 SUNIL R. LAKHANI,3,4,5 ALEXANDER VLASSOV,6 SEAN M. GRIMMOND,1,8,10 and NICOLE CLOONAN1,7,10 1Queensland Centre for Medical Genomics, Institute for Molecular Bioscience, University of Queensland, St Lucia, QLD, Australia 4072 2Whitehead Institute for Biomedical Research, Cambridge, Massachusetts 02142, USA 3The University of Queensland, UQ Centre for Clinical Research (UQCCR), Herston, QLD, Australia 4029 4The University of Queensland, School of Medicine, Herston, QLD, Australia 4029 5Pathology Queensland, The Royal Brisbane and Women’s Hospital, Herston, QLD, Australia 4029 6Life Technologies, Austin, Texas 78744, USA 7QIMR Berghofer Medical Research Institute, Genomic Biology Laboratory, Herston, Australia 4006 8Wolfson Wohl Cancer Research Centre, Institute of Cancer Sciences, University of Glasgow, Glasgow, G61 1BD, United Kingdom

ABSTRACT Metastasis is a complex, multistep process involved in the progression of cancer from a localized primary tissue to distant sites, often characteristic of the more aggressive forms of this disease. Despite being studied in great detail in recent years, the mechanisms that govern this process remain poorly understood. In this study, we identify a novel role for miR-139-5p in the inhibition of breast cancer progression. We highlight its clinical relevance by reviewing miR-139-5p expression across a wide variety of breast cancer subtypes using in-house generated and online data sets to show that it is most frequently lost in invasive tumors. A biotin pull-down approach was then used to identify the mRNA targets of miR-139-5p in the breast cancer cell line MCF7. Functional enrichment analysis of the pulled-down targets showed significant enrichment of genes in pathways previously implicated in breast cancer metastasis (P < 0.05). Further bioinformatic analysis revealed a predicted disruption to the TGFβ, Wnt, Rho, and MAPK/PI3K signaling cascades, implying a potential role for miR-139-5p in regulating the ability of cells to invade and migrate. To corroborate this finding, using the MDA-MB-231 breast cancer cell line, we show that overexpression of miR-139-5p results in suppression of these cellular phenotypes. Furthermore, we validate the interaction between miR-139-5p and predicted targets involved in these pathways. Collectively, these results suggest a significant functional role for miR-139-5p in breast cancer cell motility and invasion and its potential to be used as a prognostic marker for the aggressive forms of breast cancer. Keywords: biomarker; breast cancer; miRNA

INTRODUCTION gan, (3) extravasation into the parenchyma of distant tissues, and (4) colonization and outgrowth of tumors in the distant Breast cancer is the most commonly diagnosed cancer in site (for review, see Fidler 2003). The molecular mechanisms women and a leading cause of cancer mortality. One of the underlying metastatic migration and invasion are only par- major determinants of breast cancer mortality is the stage tially understood, despite several signaling pathways being of disease at diagnosis; patients who present with metastatic implicated (Blanco and Kang 2011). Interaction between car- disease have a 5-year survival rate of 21% (Cardoso and cinoma cells and their neighboring stroma has also been Castiglione 2009). The progression of primary tumors to shown to play a critical role (Bhowmick et al. 2004). metastatic disease is known to involve (1) invasion of extra- Recently, miRNAs have also been found to play a key role in cellular matrix and stromal layers by the tumor cells, (2) metastases (Ma et al. 2007; Valastyan et al. 2009). miRNAs are intravasation into the bloodstream to travel to a distant or- short noncoding RNAs that suppress target gene activity predominantly through binding to target mRNAs and inhibiting 9Present address: Wellcome Trust Centre for Human Genetics, University their translation. miRNAs have also been shown to promote of Oxford, Oxford OX3 7BN, UK target gene degradation (for review, see Fabian et al. 2010; 10Corresponding authors Huntzinger and Izaurralde 2011). Several miRNAs have E-mail [email protected] E-mail [email protected] Article published online ahead of print. Article and publication date are at © 2013 Krishnan et al. This article, published in RNA, is available under a http://www.rnajournal.org/cgi/doi/10.1261/rna.042143.113. Freely available Creative Commons License (Attribution-NonCommercial 3.0 Unported), as online through the RNA Open Access option. described at http://creativecommons.org/licenses/by-nc/3.0/.

RNA 19:1767–1780; Published by Cold Spring Harbor Laboratory Press for the RNA Society 1767 Downloaded from rnajournal.cshlp.org on April 13, 2014 - Published by Cold Spring Harbor Laboratory Press

Krishnan et al. causative or correlative links with metastasis. In the case of molecular subtypes of invasive ductal carcinomas–no special breast cancer, miR-21 overexpression has been shown to pro- type (IDC-NST): triple negative (n = 18), Her2+ (n = 4), ER+/ mote metastasis (Asangani et al. 2008; Zhu et al. 2008). miR- PR+ (n = 9); invasive lobular carcinomas (ILC) (n = 3); 31 regulates the expression of the key metastatic genes integ- and normal breast tissue (n = 6) (Supplemental Table 1). rin-α5, radixin, and RhoA (Valastyan et al. 2009). miR-200 The expression levels of miR-139-5p were assayed by qRT- has also been shown to increase metastatic potential of mam- PCR relative to an endogenous control RNU6B (Fig. 1A). mary carcinoma cell lines (Korpal et al. 2011), while other We observe an increase in the levels of miR-139-5p in normal studies have observed an inverse correlation between miR- mammary tissue and several subtypes, but the triple negative 200 expression and tumor invasion and metastasis (Gregory subtype showed a marked variable pattern where 38% of the et al. 2008; Korpal et al. 2008). Overexpression of miR-139- samples had lower expression compared to the normal con- 5p was recently shown to correlate with reduced metastatic ac- trols. Since this subtype is heterogeneous at clinical, morpho- tivity in hepatocellular carcinoma and gastric cancer cells logical, and molecular levels, it is possible that the low miR- (Bao et al. 2011; Wong et al. 2011; Li et al. 2013) and down- 139-5p expressing subgroup is one with a very different prog- regulated in glioblastoma (Li et al. 2013). Furthermore, in pa- nosis (Cheang et al. 2008), and further studies are warranted tients with invasive squamous cell carcinoma, loss of miR- to try to validate this. Although the difference in the popula- 139-5p expression is associated with increased metastatic dis- tion average did not reach statistical significance, the loss of ease (Mascaux et al. 2009). miR-139-5p expression may help to identify a new molecular Although a strong association between miR-139-5p and subtype important for the biological understanding of disease metastasis exists, there is little knowledge of the mechanisms and for clinical management within this invasive subgroup by which it contributes to this process or of the gene networks of breast cancer. it regulates; the specific pathways that are disrupted are still poorly understood. Only five targets of miR-139-5p have miR-139-5p is frequently down-regulated in invasive been identified and validated so far: FoxO1 (Hasseine et al. breast carcinoma 2009), Rho-kinase2 (Wong et al. 2011), CXCR4 (Bao et al. 2011), RAP1B (Guo et al. 2012), and Type I Insulin-like GF Next, we reviewed miR-139-5p expression in previously pub- (Shen et al. 2012). However, its role in breast cancer has not lished data using TaqMan Low-Density Arrays to analyze 29 been studied so far. breast tumors and 21 normal adjacent controls (Romero- miRNAs have the potential to be oncogenes or tumor sup- Cordoba et al. 2012). This sample cohort included inva- pressors in a given cellular context. Depending on the specific sive ductal carcinomas (n = 26), invasive lobular carcinomas tissues or cancer type they are expressed in, miRNAs achieve (n = 1), invasive mucinous carcinomas (IMC) (n = 1), and functional specificity by targeting a core network of genes ductal carcinoma in situ (DCIS) (n = 1). Of the IDCs, only that belong to the same pathway. This interaction is highly five samples were triple negative. As shown in Figure 1B, dependent on the relative abundance of multiple mRNA tar- miR-139-5p is significantly (P value < 0.0001) down-regulat- gets. For instance, miR-17-5p is oncogenic in hepatocellular ed in the tumor cohort compared to normal controls. To and colorectal carcinomas (Ma et al. 2012; Shan et al. 2013) strengthen the validity of this expression profile, we also and, in contrast, has been shown to have tumor-suppressive looked for changes in expression of known metastasis-associ- properties in cervical cancer cells (Wei et al. 2012). Similarly, ated miRNAs in breast cancer. Importantly, miR-139-5p ex- miR-182-5p was shown to have oncogenic properties in blad- pression positively correlates with miR-31 (r = 0.44) and der, ovarian, and breast cancers (Hirata et al. 2012; Liu et al. miR-200b (r = 0.36), which are well-characterized anti-meta- 2012; Krishnan et al. 2013), whereas it acts as a tumor sup- static miRNAs in breast cancer (Korpal et al. 2008; Valastyan pressor in lung cancer (Sun et al. 2010). Given this molecular et al. 2009). This result suggested that miR-139-5p could be and context specificity of miRNAs, we wished to explore another marker for metastatic breast cancer besides the asso- whether miR-139-5p was a potential oncomir of breast can- ciation with triple negative tumors. cer and what the output of its functional repression was, and To further investigate the expression of miR-139-5p across identify the network of genes possibly regulating its functions a larger cohort of patient samples, we chose to analyze a in the context of breast cancer. miRNA-seq data set (Farazi et al. 2011) consisting of normal breast tissue (n = 16) and various types of breast cancer including: adenoid cystic carcinoma ((n = 2), apocrine carcino- RESULTS ma (n = 4), atypical medullary carcinoma (n = 9), metaplastic carcinoma (n = 11), mucinous carcinoma (n = 1), ductal car- miR-139-5p is de-regulated in human triple negative cinoma in situ (n = 21), and invasive ductal carcinoma (n = breast cancer samples 174). Although the adenoid cystic carcinoma, a proportion To assess the clinical relevance of miR-139-5p in human of apocrine carcinomas, atypical medullary, and metaplastic breast cancer, we measured its expression in a cohort of breast carcinomas can be classified as basal-like molecular subtypes, cancer patient samples (n = 40) that included the following they differ in their morphology, aggressiveness, and prognosis

1768 RNA, Vol. 19, No. 12 Downloaded from rnajournal.cshlp.org on April 13, 2014 - Published by Cold Spring Harbor Laboratory Press

miR-139-5p and breast cancer

A (Yerushalmi et al. 2009; Marchio et al. 2010; Park et al. 2010). Metaplastic carcinomas are a heterogeneous group of tumors 10 characterized by the presence of epithelial cells (spindle, squamous) with or without metaplastic elements such as 1 bone, cartilage, and muscle. They are aggressive tumors with a poor prognosis (Park et al. 2010). Medullary carcinoma is a 0.1 controversial entity defined by Ridolfi and, despite the high grade, is said to have a good prognosis. In contrast, tumors 0.01 that do not fulfill the criteria for typical medullary carcinomas

miR-139-5p expression (normalised to RNU6B) have been designated as atypical medullary carcinomas 0.001 Triple Negative Her2+ve ER/PR+ve ILC Normal (Ridolfi et al. 1977). These are more aggressive and have mor- IDC phological features and prognosis that overlap with high grade

B 1 * IDC (Ridolfi et al. 1977). Adenoid cystic carcinoma, despite falling into the basal-like category, is an indolent tumor in

0.1 the breast compared to its salivary gland counterpart. Carcinomas with apocrine differentiation are heterogeneous,

0.01 and the behavior is dependent on grade and stage (Yerushalmi et al. 2009; Marchio et al. 2010). Interestingly, as shown in

0.001 Figure 1C, the subtypes that have the worst prognosis and the highest propensity to form distant metastases (metaplastic miR expression (normalized to RNU48) to (normalized expression miR carcinomas, atypical medullary, and the high grade TN IDCs) 0.0001 TTN N T N miR-139-5p miR-31 miR-200b exhibit the lowest levels of miR-139-5p expression. Within the IDCs, those patients with a triple negative phenotype show C ** * a more significant (P < 0.001) down-regulation compared 1000 to those that express either ER/PR or Her2 (P ∼ 0.007). However, not all triple negative patient samples show a 100 down-regulation of miR-139-5p expression as seen in Figure

10 1A and in some subtypes in Figure 1C, but the common fea- ture among these results is that miR-139-5p is frequently

1 lost in the most aggressive subtypes of breast cancer, suggest-

miR-139-5p expression (tpm) expression miR-139-5p ing that it may play a key role in the metastatic cascade of breast

0.1 cancer. Normal DCIS IDC IDC TN Metaplastic Atypical ApocrineAdenoid TN Medullary TN TN TN Identification of direct targets of miR-139-5p using a FIGURE 1. Expression analysis of miR-139-5p across tumor subtypes biotinylated miRNA duplex and normal tissue from human breast cancer patient samples. (A) Lines represent the mean of miR-139-5p expression (normalized to It is well established that the function of microRNAs depends RNU6B) across sample subtypes: triple negative (n = 18), Her2+ (n =4), ER+/PR+ (n = 9), invasive lobular carcinomas (n = 3), normal breast on the expression of their targets (Sood et al. 2006) and that tissue (n = 6). Triangles represent the expression levels for individual pa- the differing expression of these targets in different cellular tients. (B) Expression of miRNAs in breast cancer as assessed by states can lead to opposing phenotypic outcomes—for exam- TaqMan Low Density Arrays (Romero-Cordoba et al. 2012). Lines rep- ple, overexpression of miR-17-5p acts as either an oncogene resent the mean expression of miR-139-5p, alongside well-characterized anti-metastatic miRNAs like miR-31 and miR-200b (normalized to or a tumor suppressor depending on what tissue it is ex- RNU6B) across the sample cohort (comparing tumor versus normal pressed in (Cloonan et al. 2008). Therefore, it is critical to ro- along the x-axis). Triangles represent the expression levels for individual bustly identify the direct targets of miR-139-5p in breast patients. Asterisks indicate significant difference between the expression cancer cells. Given that target prediction programs are noisy of miR-139-5p in the tumour compared to the adjacent normal tissue ([∗] P < 0.0001, Student’s t-test). (C) miR-139-5p expression as assessed and have high false-positive rates (Bentwich 2005; Sood et al. using miRNA-sequencing from Farazi et al. (2011) across normal breast 2006), we resorted to an experimental identification of tar- tissue (n = 16) and a panel of breast cancer patient samples of various gets. We used the recently reported miRNA pull-down ap- subtypes, including adenoid cystic carcinoma (n = 2), apocrine carcino- proach (Cloonan et al. 2011) that involves transfecting ma (n = 4), atypical medullary carcinoma (n = 9), metaplastic carcinoma (n = 11), mucinous carcinoma (n = 1), ductal carcinoma in situ synthetic biotinylated miR-139-5p duplexes (Fig. 2A) into (DCIS) (n = 21), and invasive ductal carcinoma (IDC) (n = 174). cells (in our case, MCF7 cells) and surveying the captured Lines and scatter plot represent the mean across a subtype and the tran- target mRNAs using microarrays. Figure 2B shows sample scripts per million (tpm) of miR-139-5p in each sequenced sample, re- clustering of cells transfected with biotinylated miR-139-5p spectively. Asterisks indicate significant difference between the expression of miR-139-5p in the specific tumor subtype compared to and the total lysates where total distance between samples in- the normal tissue. (∗∗) P < 0.0001, (∗) P ∼ 0.007, Student’s t-test. dicates similarity, highlighting the reproducibility of the pull-

www.rnajournal.org 1769 Downloaded from rnajournal.cshlp.org on April 13, 2014 - Published by Cold Spring Harbor Laboratory Press

Krishnan et al.

6 C A TargetScan predicted target 5 Previously identified target Functionally enriched target

4 fold-change threshold

3 (p-value) 10 -log 2

ROCK2 significance threshold

1 B Sample relations based on 16181 genes with sd/mean > 0.1 IGF1R FFOXO1OXO1 0 -6 -5 -4 -3 -2 -10123 4

higher in pull-down log2(fold-change) higher in cell lysate

D p ≈ 5.5 X 10-7

30 40 50 60 300 46 1310 20

TargetScan Predictions mir−139-5p_pulldown_rep1 Total CellLysate_rep2 miR-139-5p pull-down targets Total CellLysate_rep3 Total CellLysate_rep1 mir−139-5p_pulldown_rep3 mir−139-5p_pulldown_rep2

FIGURE 2. Identifying targets of miR-139-5p via biotin pull-down. (A) Sequence and position of biotin in the molecule used to transfect MCF7 cells for the pull-down approach used to identify biologically relevant targets of miR-139-5p. (B) Hierarchical clustering of microarray data was performed using the plotSampleRelations function in the lumi package. Total vertical distance between samples indicates similarity. (C)A“volcano plot” showing the log2-transformed fold-change versus the log10-transformed P-value of that fold-change for every gene detected above background in the microarray. Genes highlighted in blue are targets validated by previous studies. Genes highlighted in orange are predicted by TargetScan to be targets of miR- 139-5p, showing an enrichment of the targets in the pull-downs compared to the controls. Genes highlighted in green have known association with pathways frequently involved in metastasis. (D) Venn diagram showing the overlap of genes between TargetScan predicted targets of miR-139-5p (also expressed above background in HEK293Ts) and biotinylated miR-139-5p pull-down predicted targets. This overlap is significantly more than expected by chance. down approach between biological replicates. We identified are associated with the metastatic cascade, specifically in the mRNAs significantly enriched in the biotin fraction com- context of breast cancer. miRNAs regulate cellular processes pared to the input RNA at a 5% false discovery rate (FDR) by concomitant suppression of a network of genes, so we in- threshold, with a fold-change > 2 (Fig. 2C). Using this ap- cluded genes which do not pass our stringent threshold cutoff proach, 884 probes (targeting 879 genes) were found to be sig- for target identification, since they belong to closely connect- nificantly enriched in the pull-down fractions (Supplemental ed pathways associated with the invasive ability of cancer cells. Table 2). Together, these results demonstrate that the pull-down has Next, we compared the miR-139-5p pull-down targets to enriched for miR-139-5p’s biological targets. TargetScan predicted targets and saw an overlap between the experimentally determined and the predicted targets miR-139-5p targets genes involved in pathways which was significantly more than expected by chance (P ≈ − associated with metastasis 5.5 × 10 7) (Fig. 2D), supporting the notion that we are enriching for biologically relevant targets of miR-139-5p. As ex- A gene set enrichment analysis (GSEA) was performed using pected, some targets identified in other pathologies are either ingenuity pathway analysis (IPA) on the miR-139-5p targets not expressed in MCF7 cells (2/5) or do not meet our strin- significantly enriched in the biotin pull-down. The list of gent threshold for target identification (Fig. 2C). Figure 2C pull-down-enriched 879 genes (Supplemental Table 2) was highlights (in green) the significantly enriched genes that compared to 10 other random gene lists of the same size to

1770 RNA, Vol. 19, No. 12 Downloaded from rnajournal.cshlp.org on April 13, 2014 - Published by Cold Spring Harbor Laboratory Press

miR-139-5p and breast cancer confirm specificity. A functional ontology was only consid- to mesenchymal transition (EMT) (Zavadil and Bottinger ered significant if its −log(P-value) was at least four standard 2005), and breast cancer metastasis (Padua et al. 2008). deviations away from the mean −log(P-value) of the 10 ran- Interestingly, TGFB also cross-talks with several other path- dom gene lists. Supplemental Table 3 lists all canonical pathways, including estrogen receptor signaling (Matsuda et al. ways (IPA) showing significant enrichment (P-value < 0.05) 2001), which have also been identified as miR-139-5p targets for miR-139-5p pull-down targets. As predicted, several sig- from our analyses. Also interestingly, proteins underlying this nificantly enriched pathways have been previously implicated pathway were identified to be potentially regulated by miR- in metastatic biology (Table 1), with confirmed roles for all 139-5p using target prediction software (Lee et al. 2013). but one in breast cancer. This analysis outlines the complex set of targets that ap- Functional enrichment analysis identified several canonical pear to be regulated by miR-139-5p, a major proportion of pathways previously implicated in the metastatic cascade, but them being involved in tumor progression and metastasis. the specific mechanism by which miR-139-5p modulates key It suggests possible mechanistic targets of miR-139-5p in signaling pathways underlying these processes is still un- the regulation of metastasis, warranting further functional known. To explore this, we analyzed the list of significantly characterization of its role in suppressing these functional enriched targets from the pull-down approach and explored phenotypes. the literature for their known role specifically in cellular processes underlying metastasis, including cell proliferation, mi- miR-139-5p does not alter proliferation or DNA profile gration, and invasion. Shown in Figure 3 are key signaling in MDA-MB-231 cells cascades underlying these processes, with targets of miR- 139-5p highlighted in dark gray (log2 FC > 1) and light gray The pathways analysis suggested a role for miR-139-5p in pro- (log2 FC > 0.5) and targets previously validated in other stud- moting breast cancer metastasis. To confirm its relevance in ies. The signaling molecules identified from this analysis were processes underlying malignancy of tumor cells, we under- found to be part of three major pathways: (1) Wnt signaling took a series of cell-based assays. Previous studies have report- which consists of canonical and noncanonical arms, depend- ed a prognostic role for cell proliferation in the metastatic ing on the ligand; canonical Wnt signaling occurs upon re- ability of tumors (Maeda et al. 1996; Panizo-Santos et al. cruitment of β-catenin to the nucleus (Clevers and Nusse 2000). Moreover, cell proliferation has been reported to be al- 2012), whereas the noncanonical streams work through tered by miR-139-5p in human colorectal carcinoma through downstream activation of molecules such as PRKC and regulating the expression of RAP1B (Guo et al. 2012). RHO (Schlessinger et al. 2009), both of which are known To test whether this phenotype was relevant in breast players in breast cancer progression (Scheel et al. 2011); (2) cancer, we generated four stable cell lines with inducible receptor tyrosine kinases (RTK) signal through several path- expression of miR-139-5p in MDA-MB-231 cells, a well-char- ways including RAS–MAPK and PI3 kinase, activating several acterized invasive breast cancer cell line. Doxycycline-induced downstream signal transducers, such as RAF1 and RAC1, and expression of miR-139-5p for each cell line was confirmed us- transcription factors such as MAPK3/1 (ERK1/2), NFKB, ing TaqMan real time PCR (Fig. 4A). The four stable cell lines CEBPB, and TWIST1, all of which play important roles in show between20- and 200-fold higherexpression of miR-139- breast cancer progression (Chaudhary et al. 2000; Bundy 5p in the presence of 1000 ng/mL doxycycline, which is also and Sealy 2003; Buchholz et al. 2005; Kim et al. 2008; Hong comparable to the difference in expression seen between nor- et al. 2011); (3) TGFB signaling which plays an important mal breast tissues relative to tumor patient samples (Fig. 1C). role in breast cancer (Buck and Knabbe 2006), epithelial Sequencing the small RNA population of the MDA-MB-231

TABLE 1. miR-139-5p target enriched canonical pathways and their association with tumor progression

Associated with Pathway P-value metastasis? Cancer type References

− Protein ubiquitination pathway 1.00 × 10 5 Yes Breast Kim et al. (2011) − ERK5 signaling 2.95 × 10 5 Yes Breast, prostate Cronan et al. (2011); Ramsay et al. (2011) − Aminoacyl-tRNA biosynthesis 1.95 × 10 4 Maybe (tumorigenesis) Lung, colon Kim et al. (2011) − N-Glycan biosynthesis 2.90 × 10 4 Yes Breast Lau and Dennis (2008) − Glucocorticoid receptor 3.55 × 10 4 Maybe (tumorigenesis) Breast Moutsatsou and Papavassiliou (2008) signaling − NGF signaling 3.90 × 10 4 Yes Breast Adriaenssens et al. (2008) − Xenobiotic metabolism 5.80 × 10 4 Maybe (tumorigenesis) Breast Aust et al. (2005); Naushad et al. (2011) signaling − HGF signaling 6.50 × 10 4 Yes Breast, melanoma Maroni et al. (2007); Previdi et al. (2010) − PI3K/AKT signaling 1.15 × 10 3 Yes Breast, lung, thyroid Smirnova et al. (2012); Xue et al. (2012)

www.rnajournal.org 1771 Downloaded from rnajournal.cshlp.org on April 13, 2014 - Published by Cold Spring Harbor Laboratory Press

Krishnan et al.

Ligand

RTK WNT7B Insulin PRKCE P Extracellular FZD LRP5/6 IGF1R space DAGL HRAS PIK3CA TGFB1 IRS1 P

TGFBR2 PLC P RAC1 AKT1 TNFRSF P RAF1 Dvl Dvl DAAM1 P TRAF6 SMAD2/3 WASF1 P GSK3 APC MAP3K1 IKBKB MAP3K7 P CTNNB1 AXIN1 CSNK1A1 RHOT1 Actin P IkB P polymerization ROCK1 ROCK2 MDM2 SMAD2/3 SMAD4 P NFKBIB MAPK3/1 P CTNNB1 Cytoplasm Actin Cell Motility FOXO1 microfilaments NFKB1 E2 TP53 ESR1 P CTNNB1 ETV5 MYC EP300 CEBPB TWIST1 TCF/LEF P FOXO1 SMAD2/3 SMAD4 CDKN1A BCL2 Cell Proliferation ESR1 Nucleus BMI1 Cell cycle Cell survival Cell Migration/Invasion regulation Degradation of ECM MYB Epithelial-to-mesenchymal Transition

Non target Ligand miR-139 target with log2 FC > 1 Previously identified miR-139 target P miR-139 target with log2 FC > 0.5 Phosphorylated protein Transcription factor Receptor

FIGURE 3. Overview of biotinylated miR-139-5p pull-down predicted targets involved in breast cancer invasion and metastasis. Illustration of major pathways being altered by miR-139-5p and the specific components that are targeted. These major pathways, TGFβ, Wnt, and RTK-induced MAPK and PI3K all play a major role in the process of metastasis. Several components within each of these pathways are targeted by miR-139-5p in our pull- downs with log2 FC > 1 (dark gray) or log2 FC > 0.5 (light gray), or in previously reported studies leading to distinct downstream phenotypes that contribute to the metastatic properties of the cells. cells (Supplemental Table 4) shows miR-139-5p to be ex- significant difference in the proliferation rates of the parental pressed at ∼300 transcripts per million (tpm), where the top (uninduced) cell lines versus the induced cell lines overex- 10% of miRNAs are expressed (miRNAs with <100 tpm expressing miR-139-5p. We also performed FACS analysis of cluded) with >1000 tpm. The highest expressed miRNA in the PI-stained cells to measure any change in the DNA profile the cell line had ∼74,000 tpm (a difference of ∼240-fold be- of MDA-MB-231 cells in response to miR-139-5p expres- tween miR-139-5p and the highest expressed miRNA). This sion. As shown in Figure 4C, there was no statistically signifi- suggested our stable overexpression of miR-139-5p waswithin cant change in DNA profile. Together, these results suggest physiologically relevant levels. Additionally, an analysis of the that overexpression of miR-139-5p has no effect on cell pro- data downloaded from The Cancer Genome Atlas (TCGA) liferation or progression through the cell cycle in MDA-MB- that contains data from 893 breast invasive carcinomas and 231 cells. This result also suggests that miR-139-5p may tar- normal patient samples show expression of miRNAs in nor- get different gene networks in other cancer types where alter- mal breast tissue to be between 100 tpm and 300,000 tpm ations in proliferation rates have been shown to result from with the top 10% of miRNAs with >10,000 tpm. These find- miR-139-5p overexpression. ings suggest our levels of expression are within the realms of normal copy numbers per cell. At this low level of overexpres- miR-139-5p suppresses both invasion and migration sion, we did not observe any gross morphological changes or in MDA-MB-231 cells changes in cellular integrity. All four cell lines were used in the subsequent cell-based Cellular migration and invasion are key processes under- assays. We performed MTT assays to measure the rate of pro- lying metastasis. To test the hypothesis that miR-139-5p liferation over a time course of four days. Figure 4B shows no overexpression can suppress these phenotypes in breast

1772 RNA, Vol. 19, No. 12 Downloaded from rnajournal.cshlp.org on April 13, 2014 - Published by Cold Spring Harbor Laboratory Press

miR-139-5p and breast cancer

0ng/mL doxycycline μ A 1000 on the top of a transwell insert containing 8- M pores in basal 1000ng/mL doxycycline medium and placed in a well containing medium enriched 100 with EGF. After 6 h, cells that migrated to the bottom of the insert were stained with crystal violet and counted. 10 Induction of miR-139-5p expression, upon addition of doxycycline, significantly decreased the migratory ability of the 1 MDA-MD-231 cells to between 40% and 70% (P < 0.05) (Fig. 5A,B) of their original levels in the four stable cell lines Relative miR-139-5p expression 0.1 tested. #26 #87 #88 #107 Stable Cell line (MDA-MB-231) The invasive ability of the cells was tested using a similar procedure as outlined above using transwells that contained B 0.5 a layer of matrigel to mimic the basement membrane. Cells 0.4 0ng/mL doxycycline were plated on top of the matrigel and invading cells were 1000ng/mL doxycycline stained and counted 24 h after plating. Overexpression of 0.3 miR-139-5p decreased the ability of the MBA-MB-231 cells to invade to 40%–70% (Fig. 5C,D) of the original levels in 0.2 MTT Activity all four stable cell lines (reached significance in two of them

0.1 [P < 0.05]). These assays reveal a novel role for miR-139-5p in the inhibition of properties that account for the metastatic 0.0 potential of breast cancer cells. Similar effects of anti-invasive 6 24 48 72 96 and anti-migratory roles for miR-139-5p have also been Hours post doxycycline shown in other cancer types, like human hepatocellular carcinoma (Wong et al. 2011; Fan et al. 2012) and colorectal cancer C 50 0ng/mL doxycycline 1000ng/mL doxycycline cells (Shen et al. 2012). 40

30 Validation of miR-139-5p target-binding sites and

20 change in protein expression in MDA-MB-231 cells

Proportion of cells The initial pull-down analysis was performed in MCF7 cells, 10 which is a breast cancer cell line, albeit less invasive than the 0 MDA-MB-231 cells. Since these cells wereused in all function- G1 S G2 al assays, we sought to validate the interaction between miR- FIGURE 4. Overexpression of miR-139-5p does not induce a prolifer- 139-5p and some of the genes enriched in the biotin pull- ative defect. (A) Expression of miR-139-5p as assessed by qRT-PCR in down using MDA-MB-231 breast cancer cells. The predicted MDA-MB-231 cells (with low endogenous expression of miR-139-5p) binding sites (and ∼60 nt of surrounding sequence) were stably transfected with miR-139-5p whose expression is induced in re- cloned into the 3′ UTR of the pMIR-REPORT Luciferase con- sponse to doxycycline. Four independent cell lines grown in the presence of 0 or 1000 ng/mL of doxycycline for 48 h are shown. RNU6B struct and transiently transfected into cells. Luciferase activity, was used as an endogenous control for normalization of expression. indicative of translation from the plasmid, was measured in (B) MTT cell proliferation assays of MDA-MB-231 cells stably express- the presence of a miR-139-5p mimic or negative control mim- ing miR-139-5p. The graph plots the mean and SEM of the stable cell ic and normalized using β-galactosidase activity. Using this lines grown with either 0 or 1000 ng/mL doxycycline. The induction of miR-139-5p does not affect the proliferation rates of MDA-MB-231 approach, we were able to validate five of seven genes selected cells. (C) DNA profile analysis of MDA-MB-231 cells stably expressing (Fig. 6A), including HRAS, NFKB1, PIK3CA, RAF, and miR-139-5p. Graph shows the mean and SEM of the percentage of cells RHOT1. These genes are key modulators of the pathways pre- in different cell cycle phases, as assessed by FACS. There was no significant difference between MDA-MB-231 cells expressing or not express- viously discussed (Fig. 3). These data further strengthen our ing miR-139-5p. hypothesis that miR-139-5p targets a network of genes underlying cellular processes involved in metastasis. The validation of target binding in the MDA-MB-231 cells also suggests that cancer cells, we used the stable cell lines with inducible expres- these targets are possibly the mediators of miR-139-5p’s influ- sion of miR-139-5p in MDA-MB-231 cells. The migratory ence on invasion and migration (Fig. 5). The high rate of val- ability of the stable cell lines was tested by quantifying their idation is further evidence of the ability of the biotin pull- ability to traverse a transwell membrane in the presence down approach to enrich for biologically relevant targets of (1000 ng/mL) or absence of doxycycline. The parent cell the miRNA. line stably expressing an empty vector without the miR- We further tested if regulation of these target genes by 139-5p construct was included in the assay to identify poten- miR-139-5p could result in observable changes in their pro- tial side effects of the doxycycline treatment. Cells were plated tein expression. Western blotting carried out with lysates

www.rnajournal.org 1773 Downloaded from rnajournal.cshlp.org on April 13, 2014 - Published by Cold Spring Harbor Laboratory Press

Krishnan et al.

A 0ng/mL doxycycline C 0ng/mL doxycycline Rho-kinase2 (Wong et al. 2011), and 1000ng/mL doxycycline 1000ng/mL doxycycline c-Fos (Fan et al. 2012) in hepatocellular

d carcinoma, CXCR4 (Bao et al. 2011) in

a r

d gastric cancer cells, RAP1B (Guo et al.

2012), and Type I insulin-like growth fac-

l e

* e

c c

tor (Shen et al. 2012) in colorectal cancer.

g g

a *

t In this study, we show miR-139-5p to be t

n *

c r

r associated with human breast cancer,

e P P where its expression is frequently down- regulated in the more aggressive subtypes. Using the biotin pull-down method, B Migration D Invasion followed by GSEA, we experimentally miR-139-5p miR-139-5p miR-139-5p miR-139-5p OFF ON OFF ON identified a large cohort of miR-139-5p targets in breast cancer cells, with known functions in pathways underlying the cellular migration and invasion processes. Overexpressing miR-139-5p in MDA-MB-231 cells reduced their invasive and migratory abilities. Together, FIGURE 5. The effect of miR-139-5p overexpression on migration and invasion in MDA-MB- these findings support the hypothesis – 231 cells. (A D) MDA-MB-231 cells stably expressing miR-139-5p and empty vector constructs that miR-139-5p is a potential anti-meta- were subjected to Boyden chamber transwell migration and invasion assays as described. Upon doxycycline-inducible (1000 ng/mL) expression of miR-139-5p. (A) The ability of the cells to mi- static oncomir of solid tumors. grate decreased to 40%–70% of the original levels in the stable cell lines. (B) Representative images There are certain caveats to be consid- of migrated cells not expressing miR-139-5p (left) and overexpressing miR-139-5p (right). (C) ered when interpreting our results. For The ability of the cells to invade through a matrigel basement membrane-like matrix decreased instance, although we did not observe to 40%–70% of original levels in the stable cell lines. (D) Representative images of invaded cells not expressing miR-139-5p (left) and overexpressing miR-139-5p (right). The cell lines stably ex- gross morphological changes or changes pressing vector controls did not show significant changes in the invasion or migration assays. in the cell cycle (Fig. 4B,C), it is possi- These experiments were performed with the vector and four different stable cells lines in tripli- ble that the introduction of an exogenous cate. Data plotted are mean and SEM of three independent biological replicates with at least two technical replicates each. Asterisks indicate P < 0.05 in a Student’s t-test. molecule could lead to changes in mRNA expression affecting our ability to detect miRNA targets. Reduced or absence of from stably transfected MDA-MB-231 cells in the presence or target expression in the cell line being studied could lead to absence of doxycycline showed reduction in the protein levels false negatives, and thus the targets identified in this study of NFKB1/p50 and PIK3CA and a dramatic loss in the levels should not be considered to be a definitive list of all possible of HRAS when miR-139-5p is overexpressed (Fig. 6B,C). targets in all possible cell types. Similarly, overexpression of Quantification of the levels of protein expression showed an exogenous miRNA (even at the low levels used in this statistical significance in the reduction of HRAS and study) could also lead to competition for gene targets with PIK3CA levels across all four stable cell lines, whereas the av- endogenous miRNAs, leading to false positives in our assay. erage 65% reduction of NFKB1/p50 was not statistically sig- Although we do not observe major disruption to the tran- nificant when including the increase in protein observed in scriptional landscape either here or in previous studies one cell line. This difference could be attributed to the inher- (Cloonan et al. 2011; Krishnan et al. 2013), this potential ent differences in expression of miR-139-5p across the differ- requires that studies of individual miRNA targets will need ent stable cell lines and the rate of transcription, which might to be individually validated (Fig. 6). However, the inclusion alter protein levels at different time points. These observa- of a small proportion of false positives is unlikely to affect tions further substantiate our initial findings, showing that the functional enrichment analysis performed here (Supple- miR-139-5p does have a substantial effect on the levels of tar- mental Table 3). This style of analyses has shown reliable re- get genes that are relevant to tumorigenesis, cell migration, sults with a 0.25 FDR (Subramanian et al. 2005), and for and invasion. strong biological signals, even a false-positive rate of 50% can yield accurate results (Cloonan et al. 2008). The accuracy of the analyses in this study is highlighted by our recapitula- DISCUSSION tion of known biological functions for miR-139-5p (Wong In prior studies, miR-139-5p has been shown to have anti- et al. 2011; Fan et al. 2012; Shen et al. 2012). invasive, anti-migratory, and in some cases, anti-proliferative miRNAs achieve specific regulation of cellular processes by effects on cancer cells. However, studies focusing on its target concomitant suppression of a network of genes underlying cohort have been limited to FoxO1 (Hasseine et al. 2009), the same function and/or pathways (Cloonan et al. 2008,

1774 RNA, Vol. 19, No. 12 Downloaded from rnajournal.cshlp.org on April 13, 2014 - Published by Cold Spring Harbor Laboratory Press

miR-139-5p and breast cancer

A miR-139-5p -ve control

1.00 * * * * * 0.90

0.80

0.70

0.60

0.50

0.40

0.30

0.20 Relative Luciferase Activity Relative Luciferase

0.10

0.00 pmirLuc BMI1 HRAS NFKB1-A NFKB1-B PIK3CA PRKCE RAF-A RAF-B RHOT1

BC#26 #87 #88 #107

- Dox+ Dox - Dox + Dox - Dox + Dox - Dox + Dox 1.0

NFkB1/p50 * * HRAS 0.5

PIK3CA Relative Intensity

0.0 Histone H3 -Dox +Dox -Dox +Dox -Dox +Dox NFkB1/p50 HRAS PIK3CA

FIGURE 6. Validation of miR-139-5p targets using a luciferase assay and Western blots. (A) MDA-MB-231 cells were transiently cotransfected with 20 nM miR-139-5p or a control mimic with a pMIR-REPORT Luciferase construct containing the predicted binding site from the indicated target gene. Luciferase activity was normalized to β-galactosidase activity (asterisk indicates P < 0.05 as indicated in a Student’s t-test). Data plotted are mean and SEM of at least two independent biological replicates with three technical replicates (n ≥ 2). (B) Immunoblotting showing the effects on protein expression of NFKB1/p50, HRAS, and PIK3CA upon induction of miR-139-5p in stably transfected MDA-MB-231 cells. (C) Quantitation of the blots was carried out after normalization to loading control, showing a statistically significant reduction in levels of HRAS and PIK3CA across all stable cell lines. A substantial reduction in NFKB1/p50 was observed in three of four cell lines.

2011; Shirdel et al. 2011; Gennarino et al. 2012). In this study, 1999) or through RTK-induced downstream Ras and PI3K we have shown miR-139-5p to be able to target several path- signaling (Huang et al. 2004; Du et al. 2010). ways underlying the invasive and migratory phenotypes of Through luciferase assays and Western blotting, we were cancer cells. The Wnt signaling pathway activation has been able to confirm Ras and PI3K members to be directly regulat- shown in several cancers and more specifically in breast can- ed by miR-139-5p, suggesting that the miRNA is likely inhib- cer (Schlange et al. 2007; Khalil et al. 2012). Schepeler et al. iting the invasive phenotypes through regulation of RTK- have shown that disruption of Wnt signaling leads to an in- mediated downstream signaling. Additionally, NFkB, which crease in the levels of miR-139-5p, among other miRNAs, can also be regulated by PI3K (Romashkova and Makarov in colorectal carcinoma cells (Schepeler et al. 2012). In our 1999), has been shown to be a direct target of miR-139-5p, study, we observe components of the Wnt pathway being tar- which is capable of conferring anti-apoptotic properties on gets of miR-139-5p. If disruption of Wnt in breast cancer cells metastatic cancer cells (Buchholz et al. 2005). Ras activates also leads to increase in levels of miR-139-5p, it is possible that the canonical MAPK pathway (RAF → MEK → ERK), this miRNA could be part of a regulatory feedback loop that through which they regulate Rho GTPases, which are key could keep the Wnt pathway in check in normal cells. The in- players in cell migration and invasion (Vega and Ridley vasive and migratory capacity of cells can also be attributed to 2008; Makrodouli et al. 2011). Interaction of PI3K with Rho MAPK signaling, which can occur downstream from PRKC GTPase members Rac1 and Cdc42 has also been shown to signaling through the Wnt-calcium pathway (Sheldahl et al. regulate downstream actin reorganization (Tolias et al. 1995)

www.rnajournal.org 1775 Downloaded from rnajournal.cshlp.org on April 13, 2014 - Published by Cold Spring Harbor Laboratory Press

Krishnan et al. andcancercellmigration(BarberandWelch2006),suggesting other breast cancer cell lines and subtypes would help to that both validated targets of miR-139-5p could be mediating identify its potential as a prognostic marker for the invasive anti-migratory and anti-invasive properties through the same property. effectors, the Rho GTPases. These properties are classical hall- Being widely implicated in the progression of various can- marks of aggressive breast tumors, suggesting miR-139-5p to cers would indicate a common mode of action by miR-139- be anti-metastatic. 5p, presumably through the inhibition of signaling pathways In this study, we show loss of miR-139-5p across the inva- underlying the metastatic traits of a wide range of carcinomas sive subtypes in three different data sets. However, this loss (Fig. 7). Metastatic properties of different carcinoma cells does not always correlate with other molecular markers usu- have been attributed to the pathways described in this study, ally employed to classify breast cancers, e.g., the triple nega- among others, suggesting that inhibition of these pathways tive phenotype has been routinely used to classify aggressive would be the initial step in reversing the aggressive nature of breast cancers. More recently, the heterogeneity underlying late stage tumors. Our study now shows experimental evi- this subgroup has been noted, warranting more detailed dence for regulation of these pathways by a single miRNA. prognostic tools (Metzger-Filho et al. 2012). Consistent Future studies in the characterization of these downstream with this view, we observe a wide distribution in the expres- signaling pathways and the mode of miR-139-5p silencing sion of miR-139-5p in the triple negative breast cancers ana- in human cancers would provide further information for its lyzed (Fig. 1A). Moreover, analysis of this and the Farazi et al. potential use as a therapeutic in the reversal of the metastatic data sets (Fig. 1A,C) suggests that loss of miR-139-5p could phenotype of breast and other cancers. help identify an aggressive subgroup of triple negative cancers. However, a comprehensive screen of a large number MATERIALS AND METHODS of patient samples, including those from relatively rare subtypes, is required to further understand the potential of Cell culture miR-139-5p as a biomarker for aggressive breast cancer. Experimental validations as performed in this study across MDA-MB-231 cells were maintained in DMEM (Life Technologies Australia) with 10% FBS and 1% Pen-strep and grown in a 5% CO2 atmosphere at 37°C. MCF7 cells were maintained in similar condi- Stromal Cells Autocrine Signals tions with the exception of 10 μg/mL bovine insulin (SigmaAldrich)

Paracrine Signals added to the growth media. MDA-MB-231 and MCF7 cell lines were purchased from Cell Bank Australia. Primary tumour

Clinical samples, RNA purification, and qRT-PCR analyses PI3K MAPK Wnt TGFb miR-139-5p Human breast tumors were derived from the Brisbane Breast Bank, collected from consenting patients and with ethical approval from the research ethics committees of The Royal Brisbane & Women’s Rho family Hospital and The University of Queensland. Histological type, tumor grade, tumor size, lymph node status, and ER, PR, and HER2 status were obtained from the pathology reports. ER, PR, and HER2 biomarkers were used to infer molecular subtype as luminal, HER2, Migration/Invasion/Proliferation/Survival Loss of motility/cell survival or triple negative. Total RNA from human tumor samples was extracted using tumor homogenization followed by TRIzol extraction (Invitrogen). Total RNA was purified from cell lines using the miRNeasy Mini Kit (Qiagen), and RNA integrity was assessed using an Agilent Bioanalyzer 2100. For mature miRNA, cDNA (5–10 ng total RNA) Inhibition of metastasis? was synthesized using a TaqMan MicroRNA RT Kit (Applied Metastatic cancer cells Biosystems), and qRT-PCR was performed using a miR-139-5p MicroRNA TaqMan Assay (Applied Biosystems). All RT-PCR was FIGURE 7. Model showing effect of miR-139-5p on tumor progres- performed on an Applied Biosystems 7000 SequenceDetection sion. Primary tumors, upon receiving autocrine and paracrine signals, respond by downstream activation of key signaling pathways such System. For small RNA expression analysis, RNU6B was used as an as TGF-β, Wnt, MAPK, and PI3K that are responsible for promoting endogenous control to normalize the data. their migratory, invasive, proliferative, and anti-apoptotic properties. Combined acquisition of these phenotypes enables the cells to become metastatic and seed distant organs. Shown in the figure is a potential Biotin pull-downs, microarray hybridizations model where miR-139-5p-mediated inhibition of these pathways can abrogate the onset of metastatic traits, potentially making them more Pull-downs of miR-139-5p targets were carried out as previously de- susceptible for currently employed chemotherapeutic regimes. scribed (Cloonan et al. 2011), using biotin-labeled oligonucleotides

1776 RNA, Vol. 19, No. 12 Downloaded from rnajournal.cshlp.org on April 13, 2014 - Published by Cold Spring Harbor Laboratory Press

miR-139-5p and breast cancer specific for miR-139-5p (Fig. 2A). Briefly, the 50 pmol of biotin- Flow cytometry for cell-cycle analysis labeled oligos (IDT) were transiently transfected into MCF7 cells, cultured for 24 h, followed by cell lysis and binding of 50 μL MDA-MB-231 cells stably expressing miR-139-5p were harvested − myOne C1 Streptavidin Dynabeads (Invitrogen) to the RNA frac- and fixed in 70% ethanol at 20°C overnight. DNA was stained μ tion for enrichment. Fifty nanograms of captured mRNA fractions using 10 g/mL propidium iodide (SigmaAldrich), and RNA was μ (three independent biological replicates) were amplified and labeled removed using 200 g/mL RNase A (SigmaAldrich). Cells were fil- μ using the Illumina Total Prep RNA Amplification Kit (Ambion) as tered through 35- m cell strainer mesh (BectonDickinson) and an- per the manufacturer’s instructions. Samples were profiled on alyzed on a Becton Dickinson LSR II flow cytometer fitted with a Illumina Human HT-12 chips along with input total RNA from 488-nm laser. Cell data were gated and analyzed using FlowJo the same cells as negative controls. Microarray data have been 7.2.2 (Tree Star). deposited to the Gene Expression Omnibus and can be accessed with accession number GSE40411. Cell invasion assay MDA-MB-231 cells stably overexpressing miR-139-5p and parent Bioinformatic analysis of pulldown stables (in the presence or absence of doxycycline) were counted microarray data and resuspended in serum-free media. Cell were plated at a density of 2 × 105 cells/well in the upper chamber of a matrigel (200 μg)- Microarray data were normalized using the lumi package (Du et al. coated transwell filter (8.0-μm pore) from Corning. To the reservoir, 2008) by applying background adjustment, variance-stabilizing 650 μL of serum-free media with 0.1% BSA, 1% tet-FCS, and 10 ng/ transformation (Lin et al. 2008), and robust spline normalization mL EGF was added. At 24 h, noninvaded cells on the upper side of the (Workman et al. 2002) successively. The lmFit and eBayes functions matrigel were removed carefully with a cotton swab. The cells bound in the limma package (Smyth 2004) were used to test differential ex- to the lower side of the filter were washed twice with PBS and fixed pression between the pull-down samples and the controls (Cloonan with 5% gluteraldehyde at room temperature for 10 min. Fixed cells et al. 2011). The false discovery rate was calculated to account for were washed twice with PBS and stained using 1% crystal violet. multiple testing (Benjamini and Hochberg 1995). Probes that met Excess stain was removed by washing with water, and the filters the 5% FDR threshold (for one-sided tests) with a fold-change > 2 were dried overnight. Migrated cells were counted in six random were considered significantly enriched in the pull-down. The tran- fields and images obtained using light microscopy and camera. scripts (ENSEMBL V62) to which they matched exactly were considered putative targets of that miRNA. The targets enriched using the biotin pull-downs were analyzed using ingenuity pathway analysis as Cell migration assay previously described (Cloonan et al. 2008). MDA-MB-231 cells stably overexpressing miR-139-5p and parent stables (in the presence or absence of doxycycline) were counted Stable cell line generation and resuspended in serum-free media. Cells were plated at a density of 2 × 105 cells/well in the upper chamber of a transwell filter (8.0- MDA-MB-231 cells stably expressing miR-139-5p were generated μm pore) from Corning. To the reservoir, 650 μL of serum-free me- using the Mir-X Inducible miRNA Systems (Clontech). Briefly, dia with 0.1% BSA, 1% tet-FCS, and 10 ng/mL EGF was added. At MDA-MB-231 cells were transfected with the pTet-on Advanced 6 h, nonmigrated cells on the upper side of the filter were removed Vector using Lipofectamine 2000 (Life Technologies Australia, carefully with a cotton swab. The cells bound to the lower side of the Invitrogen Division), and cells stably expressing the plasmid were filter were washed twice with PBS and fixed with 5% gluteraldehyde selected using 800 μg/mL G418 (Life Technologies Australia) (main- at room temperature for 10 min. Fixed cells were washed twice with tenance concentration: 400 μg/mL G418). Primers (Supplemental PBS and stained using 1% crystal violet. Excess stain was removed by Table 5) were used to amplify the miR-139-5p hairpin from human washing with water, and the filters were dried overnight. Migrated genomic DNA and cloned into the pmRI-Zsgreen vector plasmid cells were counted in six random fields and images obtained using supplied and subsequently transfected into the 231-pTet-on parent light microscopy and camera. line. Cells stably expressing the pmRi-Zsgreen-miR-139-5p were selected using 1 μg/mL puromycin and further maintained in 0.5 μg/ mL puromycin. Stable expression of miR-139-5p was confirmed us- Luciferase assay to validate predicted binding sites ing TaqMan MicroRNA Assay (Applied Biosystems) specific for miR-139-5p (Fig. 4A). Predicted target sites of miR-139-5p were cloned into the HindIII and SpeI sites of the pMIR-REPORT Luciferase vector. Synthetic oligos (Supplemental Table 5) corresponding to 60 nt surround- MTT proliferation assays ing the target sequence were annealed before ligation into the pMIR-REPORT Luciferase vector. All constructs were verified by Stable pmRi-MDA-MB-231 cell lines overexpressing miR-139-5p sequencing. MDA-MB-231 cells were cotransfected with 50 ng of a (1000 ng/mL dox) and parent stables (with no doxycycline) were pMIR-REPORT Luciferase construct and 50 ng of pMIR-REPORT plated at 1 × 104 cells per well. MTT (3-[4,5-dimethylthiazol-2-yl]- β-galactosidase Reporter Control Vector (Ambion) along with 2,5-diphenyl tetrazolium bromide) activity was assayed using a Cell miR-139-5p or a negative mimic (Ambion) to a final concentration Growth Determination Kit (SigmaAldrich) according to the manu- of 20 nM. Post-transfection, cells were incubated for 48 h prior to as- facturer’s instructions and detected on a PowerWave XS spectropho- saying. Luciferase activity was assayed using the Luciferase Assay tometer (BioTek). System (Promega Corporation) and detected on a Wallac

www.rnajournal.org 1777 Downloaded from rnajournal.cshlp.org on April 13, 2014 - Published by Cold Spring Harbor Laboratory Press

Krishnan et al.

1420 luminometer (PerkinElmer). β-galactosidase activity was de- and S.M.G. is supported by a National Health and Medical termined using the β-Galactosidase Enzyme Assay System (Prom- Research Council (NHMRC) Principal Research Fellowship. We ega) and detected on a PowerWave XS spectrophotometer also thank all the members of QCMG for the helpful discussions (BioTek). Luciferase activity was normalized to β-galactosidase activ- and John Pearson, Darrin Taylor, and Scott Wood for HPC infra- ity in each well. Assays were conducted in triplicate and independent- structure and support. The authors would also like to acknowledge ly repeated at least twice. the contribution of tissue donors and staff for the generation of the Brisbane Breast Bank.

Western blotting Received August 26, 2013; accepted September 12, 2013. MDA-MB-231 cells stably overexpressing miR-139-5p (in the presence or absence of 1000 ng/mL doxycycline) grown in 10-cm dishes REFERENCES for 24 h were washed gently in 5 mL of ice cold PBS and then lysed using 300 μL of ice-cold aqueous lysis buffer (50 mM Tris, pH 7.5, Adriaenssens E, Vanhecke E, Saule P, Mougel A, Page A, Romon R, 150 mM NaCl, 10 mM EDTA, pH 8.0, 0.2% sodium azide, 50 mM Nurcombe V, Le Bourhis X, Hondermarck H. 2008. Nerve growth factor is a potential therapeutic target in breast cancer. Cancer Res NaF, 0.5% NP40) containing protease (Cat.# P8340, SigmaAldrich) 68: 346–351. and phosphatase inhibitors (Cat.# P5726 and P0044, SigmaAldrich). Asangani IA, Rasheed SA, Nikolova DA, Leupold JH, Colburn NH, Lysates were spun at 12,000g for 30 min and the supernatants col- Post S, Allgayer H. 2008. MicroRNA-21 (miR-21) post-transcrip- lected and stored at −80°C. Estimation of protein concentration tionally downregulates tumor suppressor Pdcd4 and stimulates inva- was performed using Bradford reagent (Cat.# 500-0001, Biorad) us- sion, intravasation and metastasis in colorectal cancer. Oncogene 27: – ing a standard curve created with known concentrations of BSA. 2128 2136. Optical density measurements were carried out on an Ultrospec Aust S, Obrist P, Klimpfinger M, Tucek G, Jager W, Thalhammer T. 2005. Altered expression of the hormone- and xenobiotic-metabo- 6300 pro (Amersham Biosciences). Thirty micrograms of protein lizing sulfotransferase enzymes 1A2 and 1C1 in malignant breast tis- containing lysate were loaded onto each well of a NuPAGE SDS- sue. Int J Oncol 26: 1079–1085. PAGE gel (Invitrogen) and run for 1.5 h at 130V. Protein from Bao W, Fu HJ, Xie QS, Wang L, Zhang R, Guo ZY, Zhao J, Meng YL, the gel was transferred onto PVDF membrane (Millipore) using Ren XL, Wang T, et al. 2011. HER2 interacts with CD44 to up-reg- NuPAGE transfer buffer at 20V for 2 h. Following transfer, mem- ulate CXCR4 via epigenetic silencing of microRNA-139 in gastric – branes were blocked in TBS-T containing 5% skim milk powder cancer cells. Gastroenterology 141: 2076 2087 e2076. Barber MA, Welch HC. 2006. PI3K and RAC signalling in leukocyte and for 1 h at room temperature, after which they were incubated cancer cell migration. Bull Cancer 93: E44–E52. with primary antibody (1:500)-containing solution (in 5% BSA) Benjamini Y, Hochberg Y. 1995. Controlling the false discovery rate: A overnight rocking at 4°C. Primary antibodies used were: rabbit practical and powerful approach to multiple testing. J R Stat Soc anti-p50/p105 antibody (Cat.# 3035P, Cell Signaling), rabbit anti- Series B (Methodological) 57: 289–300. HRAS (Cat.# SC520, Santa Cruz Biotechnology), rabbit anti- Bentwich I. 2005. Prediction and validation of microRNAs and their tar- – PIK3CA p110 antibody (Cat.# 4249S, Cell Signaling), and rabbit gets. FEBS Lett 579: 5904 5910. Bhowmick NA, Neilson EG, Moses HL. 2004. Stromal fibroblasts in anti-histone H3 (Cat.# 9715, Cell Signaling). Membranes were cancer initiation and progression. Nature 432: 332–337. then washed in TBS-T three times (30 min each), followed by incu- Blanco MA, Kang Y. 2011. Signaling pathways in breast cancer metasta- bation with anti-rabbit HRP secondary antibody (Cat.# 7074, Cell sis—novel insights from functional genomics. Breast Cancer Res 13: Signaling) at 1:5000 in TBS-T containing 5% skim milk powder 206. for 1 h at room temperature. Membranes were then washed in Buchholz TA, Garg AK, Chakravarti N, Aggarwal BB, Esteva FJ, Kuerer TBS-T and developed using the SuperSignal West Dura HM, Singletary SE, Hortobagyi GN, Pusztai L, Cristofanilli M, et al. κ Chemiluminescent Substrate (Cat.# 34076, ThermoScientific) on 2005. The nuclear transcription factor B/bcl-2 pathway correlates with pathologic complete response to doxorubicin-based neoadju- a Konica Minolta film processor (SRX 201A, Konica Minolta). vant chemotherapy in human breast cancer. Clin Cancer Res 11: 8398–8402. Buck MB, Knabbe C. 2006. TGF-β signaling in breast cancer. Ann N Y DATA DEPOSITION Acad Sci 1089: 119–126. Bundy LM, Sealy L. 2003. CCAAT/enhancer binding protein beta Microarray data have been deposited to the Gene Expression (C/EBPβ)-2 transforms normal mammary epithelial cells and induces Omnibus under accession number GSE40411. epithelial to mesenchymal transition in culture. Oncogene 22: 869–883. Cardoso F, Castiglione M. 2009. Locally recurrent or metastatic breast cancer: ESMO clinical recommendations for diagnosis, treatment – SUPPLEMENTAL MATERIAL and follow-up. Ann Oncol 20: 15 18. Chaudhary A, King WG, Mattaliano MD, Frost JA, Diaz B, Morrison DK, Supplemental material is available for this article. Cobb MH, Marshall MS, Brugge JS. 2000. Phosphatidylinositol 3-kinase regulates Raf1 through Pak phosphorylation of serine 338. Curr Biol 10: 551–554. Cheang MC, Voduc D, Bajdik C, Leung S, McKinney S, Chia SK, ACKNOWLEDGMENTS Perou CM, Nielsen TO. 2008. Basal-like breast cancer defined by This work was partially supported by an Australian Research Council five biomarkers has superior prognostic value than triple-negative phenotype. Clin Cancer Res 14: 1368–1376. (ARC) Discovery Project Grant DP1093164. K.K. is supported by an Clevers H, Nusse R. 2012. Wnt/β-catenin signaling and disease. Cell 149: Australian Post Graduate Award (APA), P.T.S. is supported by a fel- 1192–1205. lowship from the National Breast Cancer Foundation, Australia, Cloonan N, Brown MK, Steptoe AL, Wani S, Chan WL, Forrest AR, N.C. is supported by an ARC Future Fellowship (FT 120100453), Kolle G, Gabrielli B, Grimmond SM. 2008. The miR-17-5p

1778 RNA, Vol. 19, No. 12 Downloaded from rnajournal.cshlp.org on April 13, 2014 - Published by Cold Spring Harbor Laboratory Press

miR-139-5p and breast cancer

microRNA is a key regulator of the G1/S phase cell cycle transition. Korpal M, Lee ES, Hu G, Kang Y. 2008. The miR-200 family inhibits ep- Genome Biol 9: R127. ithelial-mesenchymal transition and cancer cell migration by direct Cloonan N, Wani S, Xu Q, Gu J, Lea K, Heater S, Barbacioru C, targeting of E-cadherin transcriptional repressors ZEB1 and ZEB2. Steptoe AL, Martin HC, Nourbakhsh E, et al. 2011. MicroRNAs J Biol Chem 283: 14910–14914. and their isomiRs function cooperatively to target common biolog- Korpal M, Ell BJ, Buffa FM, Ibrahim T, Blanco MA, Celia-Terrassa T, ical pathways. Genome Biol 12: R126. Mercatali L, Khan Z, Goodarzi H, Hua Y, et al. 2011. Direct targeting Cronan MR, Nakamura K, Johnson NL, Granger DA, Cuevas BD, of Sec23a by miR-200s influences cancer cell secretome and pro- Wang JG, Mackman N, Scott JE, Dohlman HG, Johnson GL. motes metastatic colonization. Nat Med 17: 1101–1108. 2011. Defining MAP3 kinases required for MDA-MB-231 cell tumor Krishnan K, Steptoe AL, Martin HC, Wani S, Nones K, Waddell N, growth and metastasis. Oncogene 31: 3889–3900. Mariasegaram M, Simpson PT, Lakhani SR, Gabrielli B, et al. Du P, Kibbe WA, Lin SM. 2008. lumi: A pipeline for processing Illumina 2013. MicroRNA-182-5p targets a network of genes involved in microarray. Bioinformatics 24: 1547–1548. DNA repair. RNA 19: 230–242. Du J, Sun C, Hu Z, Yang Y, Zhu Y, Zheng D, Gu L, Lu X. 2010. Lau KS, Dennis JW. 2008. N-Glycans in cancer progression. Lysophosphatidic acid induces MDA-MB-231 breast cancer cells Glycobiology 18: 750–760. migration through activation of PI3K/PAK1/ERK signaling. PLoS Lee CH, Kuo WH, Lin CC, Oyang YJ, Huang HC, Juan HF. 2013. One 5: e15940. MicroRNA-regulated protein-protein interaction networks and Fabian MR, Sonenberg N, Filipowicz W. 2010. Regulation of mRNA their functions in breast cancer. Int J Mol Sci 14: 11560–11606. translation and stability by microRNAs. Annu Rev Biochem 79: Li RY, Chen LC, Zhang HY, Du WZ, Feng Y, Wang HB, Wen JQ, Liu X, 351–379. Li XF, Sun Y, et al. 2013. MiR-139 inhibits Mcl-1 expression and po- Fan Q, He M, Deng X, Wu WK, Zhao L, Tang J, Wen G, Sun X, Liu Y. tentiates TMZ-induced apoptosis in glioma. CNS Neurosci Ther 19: 2012. Derepression of c-Fos caused by microRNA-139 down-regu- 477–483. lation contributes to the metastasis of human hepatocellular carci- Lin SM, Du P, Huber W, Kibbe WA. 2008. Model-based variance-stabi- noma. Cell Biochem Funct 31: 319–324. lizing transformation for Illumina microarray data. Nucleic Acids Res Farazi TA, Horlings HM, Ten Hoeve JJ, Mihailovic A, Halfwerk H, 36: e11. Morozov P, Brown M, Hafner M, Reyal F, van Kouwenhove M, Liu Z, Liu J, Segura MF, Shao C, Lee P, Gong Y, Hernando E, Wei JJ. et al. 2011. MicroRNA sequence and expression analysis in breast 2012. MiR-182 overexpression in tumourigenesis of high-grade tumors by deep sequencing. Cancer Res 71: 4443–4453. serous ovarian carcinoma. J Pathol 228: 204–215. Fidler IJ. 2003. The pathogenesis of cancer metastasis: The ‘seed and soil’ Ma L, Teruya-Feldstein J, Weinberg RA. 2007. Tumour invasion and hypothesis revisited. Nat Rev Cancer 3: 453–458. metastasis initiated by microRNA-10b in breast cancer. Nature Gennarino VA, D’Angelo G, Dharmalingam G, Fernandez S, 449: 682–688. Russolillo G, Sanges R, Mutarelli M, Belcastro V, Ballabio A, Ma Y, Zhang P, Wang F, Zhang H, Yang Y, Shi C, Xia Y, Peng J, Liu W, Verde P, et al. 2012. Identification of microRNA-regulated gene net- Yang Z, et al. 2012. Elevated oncofoetal miR-17-5p expression reg- works by expression analysis of target genes. Genome Res 22: ulates colorectal cancer progression by repressing its target gene 1163–1172. P130. Nat Commun 3: 1291. Gregory PA, Bert AG, Paterson EL, Barry SC, Tsykin A, Farshid G, Maeda K, Chung YS, Onoda N, Ogawa M, Kato Y, Nitta A, Arimoto Y, Vadas MA, Khew-Goodall Y, Goodall GJ. 2008. The miR-200 family Kondo Y, Arakawa T, Sowa M. 1996. Association of tumor cell pro- and miR-205 regulate epithelial to mesenchymal transition by tar- liferation with lymph node metastasis in early gastric cancer. geting ZEB1 and SIP1. Nat Cell Biol 10: 593–601. Oncology 53: 1–5. Guo H, Hu X, Ge S, Qian G, Zhang J. 2012. Regulation of RAP1B by Makrodouli E, Oikonomou E, Koc M, Andera L, Sasazuki T, Shirasawa miR-139 suppresses human colorectal carcinoma cell proliferation. S, Pintzas A. 2011. BRAF and RAS oncogenes regulate Rho Int J Biochem Cell Biol 44: 1465–1472. GTPase pathways to mediate migration and invasion properties Hasseine LK, Hinault C, Lebrun P, Gautier N, Paul-Bellon R, Van in human colon cancer cells: A comparative study. Mol Cancer Obberghen E. 2009. miR-139 impacts FoxO1 action by decreasing 10: 118. FoxO1 protein in mouse hepatocytes. Biochem Biophys Res Commun Marchio C, Weigelt B, Reis-Filho JS. 2010. Adenoid cystic carcinomas of 390: 1278–1282. the breast and salivary glands (or ‘The strange case of Dr Jekyll and Hirata H, Ueno K, Shahryari V, Tanaka Y, Tabatabai ZL, Hinoda Y, Mr Hyde’ of exocrine gland carcinomas). J Clin Pathol 63: 220–228. Dahiya R. 2012. Oncogenic miRNA-182-5p targets Smad4 and Maroni P, Bendinelli P, Matteucci E, Desiderio MA. 2007. HGF induces RECK in human bladder cancer. PLoS One 7: e51056. CXCR4 and CXCL12-mediated tumor invasion through Ets1 and Hong J, Zhou J, Fu J, He T, Qin J, Wang L, Liao L, Xu J. 2011. NF-κB. Carcinogenesis 28: 267–279. Phosphorylation of serine 68 of Twist1 by MAPKs stabilizes Twist1 Mascaux C, Laes JF, Anthoine G, Haller A, Ninane V, Burny A, protein and promotes breast cancer cell invasiveness. Cancer Res 71: Sculier JP. 2009. Evolution of microRNA expression during human 3980–3990. bronchial squamous carcinogenesis. Eur Respir J 33: 352–359. Huang C, Jacobson K, Schaller MD. 2004. MAP kinases and cell migra- Matsuda T, Yamamoto T, Muraguchi A, Saatcioglu F. 2001. Cross-talk tion. J Cell Sci 117: 4619–4628. between transforming growth factor-β and estrogen receptor signal- Huntzinger E, Izaurralde E. 2011. Gene silencing by microRNAs: ing through Smad3. J Biol Chem 276: 42908–42914. Contributions of translational repression and mRNA decay. Nat Metzger-Filho O, Tutt A, de Azambuja E, Saini KS, Viale G, Loi S, Rev Genet 12: 99–110. Bradbury I, Bliss JM, Azim HA Jr, Ellis P, et al. 2012. Dissecting Khalil S, Tan GA, Giri DD, Zhou XK, Howe LR. 2012. Activation status the heterogeneity of triple-negative breast cancer. J Clin Oncol 30: of Wnt/ss-catenin signaling in normal and neoplastic breast tissues: 1879–1887. Relationship to HER2/neu expression in human and mouse. PLoS Moutsatsou P, Papavassiliou AG. 2008. The glucocorticoid receptor sig- One 7: e33421. nalling in breast cancer. J Cell Mol Med 12: 145–163. Kim J, Shao Y, Kim SY, Kim S, Song HK, Jeon JH, Suh HW, Chung JW, Naushad SM, Reddy CA, Rupasree Y, Pavani A, Digumarti RR, Yoon SR, Kim YS, et al. 2008. Hypoxia-induced IL-18 increases hyp- Gottumukkala SR, Kuppusamy P, Kutala VK. 2011. Cross-talk oxia-inducible factor-1α expression through a Rac1-dependent NF- between one-carbon metabolism and xenobiotic metabolism: κB pathway. Mol Biol Cell 19: 433–444. Implications on oxidative DNA damage and susceptibility to breast Kim B, Nam HJ, Pyo KE, Jang MJ, Kim IS, Kim D, Boo K, Lee SH, cancer. Cell Biochem Biophys 61: 715–723. Yoon JB, Baek SH, et al. 2011. Breast cancer metastasis suppressor Padua D, Zhang XH, Wang Q, Nadal C, Gerald WL, Gomis RR, 1 (BRMS1) is destabilized by the Cul3-SPOP E3 ubiquitin ligase Massague J. 2008. TGFβ primes breast tumors for lung metastasis complex. Biochem Biophys Res Commun 415: 720–726. seeding through angiopoietin-like 4. Cell 133: 66–77.

www.rnajournal.org 1779 Downloaded from rnajournal.cshlp.org on April 13, 2014 - Published by Cold Spring Harbor Laboratory Press

Krishnan et al.

Panizo-Santos A, Sola I, Vega F, de Alava E, Lozano MD, Idoate MA, Shirdel EA, Xie W, Mak TW, Jurisica I. 2011. NAViGaTing the micro- Pardo-Mindan J. 2000. Predicting metastatic risk of gastrointestinal nome—using multiple microRNA prediction databases to identify stromal tumors: Role of cell proliferation and cell cycle regulatory signalling pathway-associated microRNAs. PLoS One 6: e17429. proteins. Int J Surg Pathol 8: 133–144. Smirnova T, Zhou ZN, Flinn RJ, Wyckoff J, Boimel PJ, Pozzuto Park HS, Park S, Kim JH, Lee JH, Choi SY, Park BW, Lee KS. 2010. M, Coniglio SJ, Backer JM, Bresnick AR, Condeelis JS, et al. Clinicopathologic features and outcomes of metaplastic breast carci- 2012. Phosphoinositide 3-kinase signaling is critical for ErbB3- noma: Comparison with invasive ductal carcinoma of the breast. driven breast cancer cell motility and metastasis. Oncogene 31: Yonsei Med J 51: 864–869. 706–715. Previdi S, Maroni P, Matteucci E, Broggini M, Bendinelli P, Desiderio Smyth GK. 2004. Linear models and empirical Bayes methods for assess- MA. 2010. Interaction between human-breast cancer metastasis ing differential expression in microarray experiments. Stat Appl and bone microenvironment through activated hepatocyte growth Genet Mol Biol 3: doi: 10.2202/1544-6115.1027. factor/Met and β-catenin/Wnt pathways. Eur J Cancer 46: Sood P, Krek A, Zavolan M, Macino G, Rajewsky N. 2006. Cell-type- 1679–1691. specific signatures of microRNAs on target mRNA expression. Ramsay AK, McCracken SR, Soofi M, Fleming J, Yu AX, Ahmad I, Proc Natl Acad Sci 103: 2746–2751. Morland R, Machesky L, Nixon C, Edwards DR, et al. 2011. ERK5 Subramanian A, Tamayo P, Mootha VK, Mukherjee S, Ebert BL, signalling in prostate cancer promotes an invasive phenotype. Br J Gillette MA, Paulovich A, Pomeroy SL, Golub TR, Lander ES, Cancer 104: 664–672. et al. 2005. Gene set enrichment analysis: A knowledge-based ap- Ridolfi RL, Rosen PP, Port A, Kinne D, Mike V. 1977. Medullary carci- proach for interpreting genome-wide expression profiles. Proc Natl noma of the breast: A clinicopathologic study with 10 year follow- Acad Sci 102: 15545–15550. up. Cancer 40: 1365–1385. Sun Y, Fang R, Li C, Li L, Li F, Ye X, Chen H. 2010. Hsa-mir-182 Romashkova JA, Makarov SS. 1999. NF-κB is a target of AKT in anti- suppresses lung tumorigenesis through down regulation of apoptotic PDGF signalling. Nature 401: 86–90. RGS17 expression in vitro. Biochem Biophys Res Commun 396: 501– Romero-Cordoba S, Rodriguez-Cuevas S, Rebollar-Vega R, Quintanar- 507. Jurado V, Maffuz-Aziz A, Jimenez-Sanchez G, Bautista-Pina V, Tolias KF, Cantley LC, Carpenter CL. 1995. Rho family GTPases bind Arellano-Llamas R, Hidalgo-Miranda A. 2012. Identification and to phosphoinositide kinases. J Biol Chem 270: 17656–17659. pathway analysis of microRNAs with no previous involvement in Valastyan S, Reinhardt F, Benaich N, Calogrias D, Szasz AM, Wang breast cancer. PLoS One 7: e31904. ZC, Brock JE, Richardson AL, Weinberg RA. 2009. A pleiotropically Scheel C, Eaton EN, Li SH, Chaffer CL, Reinhardt F, Kah KJ, Bell G, acting microRNA, miR-31, inhibits breast cancer metastasis. Cell Guo W, Rubin J, Richardson AL, et al. 2011. Paracrine and autocrine 137: 1032–1046. signals induce and maintain mesenchymal and stem cell states in the Vega FM, Ridley AJ. 2008. Rho GTPases in cancer cell biology. FEBS Lett breast. Cell 145: 926–940. 582: 2093–2101. Schepeler T, Holm A, Halvey P, Nordentoft I, Lamy P, Riising EM, Wei Q, Li YX, Liu M, Li X, Tang H. 2012. MiR-17-5p targets TP53INP1 Christensen LL, Thorsen K, Liebler DC, Helin K, et al. 2012. and regulates cell proliferation and apoptosis of cervical cancer cells. Attenuation of the β-catenin/TCF4 complex in colorectal cancer IUBMB Life 64: 697–704. cells induces several growth-suppressive microRNAs that target can- Wong CC, Wong CM, Tung EK, Au SL, Lee JM, Poon RT, Man K, cer promoting genes. Oncogene 31: 2750–2760. Ng IO. 2011. The microRNA miR-139 suppresses metastasis and Schlange T, Matsuda Y, Lienhard S, Huber A, Hynes NE. 2007. progression of hepatocellular carcinoma by down-regulating Rho- Autocrine WNT signaling contributes to breast cancer cell prolifer- kinase 2. Gastroenterology 140: 322–331. ation via the canonical WNT pathway and EGFR transactivation. Workman C, Jensen LJ, Jarmer H, Berka R, Gautier L, Nielser HB, Breast Cancer Res 9: R63. Saxild HH, Nielsen C, Brunak S, Knudsen S. 2002. A new non- Schlessinger K, Hall A, Tolwinski N. 2009. Wnt signaling pathways meet linear normalization method for reducing variability in DNA Rho GTPases. Genes Dev 23: 265–277. microarray experiments. Genome Biol 3: doi: 10.1186/gb-2002-3-9- Shan SW, Fang L, Shatseva T, Rutnam ZJ, Yang X, Du W, Lu WY, research0048. Xuan JW, Deng Z, Yang BB. 2013. Mature miR-17-5p and passenger Xue G, Restuccia DF, Lan Q, Hynx D, Dirnhofer S, Hess D, Ruegg C, miR-17-3p induce hepatocellular carcinoma by targeting PTEN, Hemmings BA. 2012. Akt/PKB-mediated phosphorylation of GalNT7 and vimentin in different signal pathways. J Cell Sci 126: Twist1 promotes tumor metastasis via mediating cross-talk between 1517–1530. PI3K/Akt and TGF-β signaling axes. Cancer Discov 2: 248–259. Sheldahl LC, Park M, Malbon CC, Moon RT. 1999. Protein kinase C Yerushalmi R, Hayes MM, Gelmon KA. 2009. Breast carcinoma—rare is differentially stimulated by Wnt and Frizzled homologs in a G- types: Review of the literature. Ann Oncol 20: 1763–1770. protein-dependent manner. Curr Biol 9: 695–698. Zavadil J, Bottinger EP. 2005. TGF-β and epithelial-to-mesenchymal Shen K, Liang Q, Xu K, Cui D, Jiang L, Yin P, Lu Y, Li Q, Liu J. 2012. transitions. Oncogene 24: 5764–5774. MiR-139 inhibits invasion and metastasis of colorectal cancer by Zhu S, Wu H, Wu F, Nie D, Sheng S, Mo YY. 2008. MicroRNA-21 tar- targeting the type I insulin-like growth factor receptor. Biochem gets tumor suppressor genes in invasion and metastasis. Cell Res 18: Pharmacol 84: 320–330. 350–359.

1780 RNA, Vol. 19, No. 12 CHAPTER FOUR

Exploration of cell cycle associated miRNAs 4 Exploration of cell cycle associated miRNAs

4.1 Introduction MicroRNAs (miRNAs) are non-coding RNAs, predominantly 22 nucleotides, long that act to regulate protein production through either repression of translation [296] or (more rarely) mRNA degradation [297]. An individual miRNA is capable of directly targeting hundreds to thousands of distinct mRNAs [94, 123, 298-302], but achieve their functional specificity by coordinate suppression of a network of genes involved in specific cellular processes [101, 298, 299, 303, 304], and by the cooperative targeting of isomiRs [305]. As the biological role of a miRNA depends on the presence of specific mRNA targets, the pathways targeted by individual miRNAs may differ in different cell types. This results in some miRNAs behaving as either oncogenes or tumor suppressors depending on the cell type they are expressed in [183, 298, 299].

One of the most crucial processes underlying cell biology is the cell cycle, which goes through three phases followed by mitosis. The G1 phase is where, in response to signals from the environment, the cell synthesizes proteins required for its division [306]. This is followed by the S (synthesis)-phase where the DNA is replicated, and enters the second gap phase (the G2 pre-mitotic phase). During this phase rapid increase in cell mass occurs owing to increased protein synthesis [289]. Finally, mitosis occurs in several stages that include chromatin condensation and segregation, ultimately resulting in cell division [307]. This process is tightly controlled by a host of proteins including cyclins, cyclin-dependent kinases (CDK) and cyclin-dependent kinase inhibitors (CDKi) that regulate the transition through each phase, ensuring that only cells with undamaged DNA proceed to division [289]. In addition to regulation by these proteins, miRNAs and proteins that belong to the miRNA machinery such as Dicer and Argonaute also regulate the cell cycle.

Dicer, an endoribonuclease which catalyzes the excision of double-stranded RNAs into short 21-22bp fragments during miRNA biogenesis [308], has been shown through knockout studies to induce a G1 arrest in response to nitrogen limiting conditions in fission yeast [309, 310]. In a chicken-human hybrid cell line, loss of Dicer leads to accumulation of cells in the G2-M phase [311]. Knockout of Dicer 1 leads to

98 accumulation in the G1-S phase in germ line stem cells of Drosophila melanogaster [312], also recapitulated in mouse embryonic stem cells where depletion of DGCR8 leads to an extended G1-S phase [313]. Knockdown of AGO2, which constitute catalytic units of the RNA-induced silencing complex (RISC) [40], leads to inhibition of cell proliferation by inducing apoptosis in myeloma cell lines [314]. Several miRNAs have also been shown to play an important role in the different stages of the cell cycle, including the G1-S transition and entry and progression through mitosis by controlling the actions of several key regulators. For example, the miR-15a-16 cluster regulates the cell cycle through its targets that include CDK1, CDK2 and CDK6 and cyclins D1, D3 and E1 [315-318]. A summary of miRNAs that regulate various phases of the cell cycle is shown in Appendix 3.1, and the dysregulation of these miRNAs in cancer has been recently reviewed [319]. Together, these studies provide strong evidence that miRNAs play a significant role in regulation of the cell cycle.

miRNAs with dynamic expression across the cell cycle phases and their potential functions have been difficult to study. Zhou et. al [320] previously used synchronized HeLa cells to identify 25 miRNAs as differentially expressed across the cell cycle phases, however the study was limited to ~700 human miRNAs, and suffered from a lower level of synchronization. In this report we expand this study using miRNA-seq to characterize cell cycle expression patterns from synchronized HeLa and MCF7 cells, with data mapped back to miRBase V19 with ~2000 precursor human miRNAs. We integrate this data with ENCODE regulatory factor profiling, and TCGA tumor expression data to determine the relationship between miRNAs with differential expression in the cell cycle and their expression across human cancers.

99 4.2 Materials and Methods

4.2.1 Cell Culture HeLa and MCF7 cells were maintained in DMEM (Life Technologies) with 10% FBS and 1% Penicillin–Streptomycin (Life Technologies) and grown in a 5% CO2 atmosphere at 37°C. HeLa cell lines were purchased from Cell Bank.

4.2.2 Synchronization of cells HeLa cells were synchronized by incubation for 18 h with 2.5 mM thymidine (Sigma- Aldrich), released into fresh media for 8 h, and treated again with 2.5 mM thymidine for another 18 h. To obtain synchronized populations, these cells were then released for 0 h

(S phase), 8 h (G2-M), and 14 h (G0-G1). Chemically synchronized populations were verified by flow cytometry. MCF7 cells were synchronized by culture in the absence of beta-estradiol for 48 hours followed by its re-addition when cells were in G-0-G1 at 6h, S phase at 22h and G2-M at 26h. Brian Gabrielli provided synchronized HeLa cells and Elizabeth Musgrove provided synchronized MCF7 cells along with data for the Figures 4.1b and c.

4.2.3 Cell cycle analysis by flow cytometry All cells were harvested and fixed in 70% ethanol at -20°C overnight, then resuspended in buffer (5 mM EDTA, PBS, pH 7.4) approximately 1 h prior to analysis. DNA was stained using 40 μg/ml propidium iodide (Sigma-Aldrich), and RNA was removed using 400 μg/ml RNase A (Sigma-Aldrich). Cells were filtered through 35 μm cell strainer mesh (Becton Dickinson, North Ryde, NSW, Australia) and analyzed on Becton Dickinson LSR II flow cytometer fitted with 488 nm laser. Cell data were gated using WinList v6.0 and analyzed in Modfit LT v3.0, both programs from Verity Software House (Topsham, ME, USA). Melissa Brown performed FACS analysis for Figure 4.1a.

4.2.4 RNA Purification and microarray analysis Total RNA was purified from cell pellets using either an RNeasy Mini Kit (Qiagen), or a miRNeasy Mini Kit (Qiagen), and in both cases RNA integrity was assessed using an Agilent Bioanalyzer 2100. Total RNA samples were profiled on Illumina Human HT-12 chips along with control RNA from mock-transfected cells. Microarray data were

100 normalized using the lumi package (Du et al. 2008) by applying background adjustment, variance-stabilizing transformation (Lin et al. 2008), and robust spline normalization (Workman et al. 2002) successively. Shivangi Wani and Katia Nones performed RNA purification and microarray data generation respectively.

4.2.5 Library preparation and sequencing Construction of small RNA libraries was performed using the Small RNA Expression Kit (SREK, Ambion, Austin, TX, USA), according to the manufacturer's instructions. Briefly, small RNA purified using the miRNeasy kit (Qiagen) was combined with a mixture of both 5' and 3' adaptors and ligated in a single reaction. cDNA was synthesized, and PCR amplified. During the PCR, sequencing adaptors were incorporated along with a unique six-nucleotide barcode sequence. Following library construction and purification, barcoded libraries were pooled and sequenced together using the Applied Biosystems (Beverly, MA, USA) SOLiD system. Shivangi Wani and Keerthana Krishnan generated small RNA sequencing libraries. Ehsan Nourbaksh sequenced all libraries.

4.2.6 Bioinformatic Analysis of small RNA sequencing data Sequenced data was de-barcoded and small RNA reads for individual samples were mapped back to human hairpin sequences from miRBase V19, allowing up to 2 mismatches. Mature names for mapped miRNAs were derived from miRBase if there was a seed length of at least 20 with up to 2 mismatches, with another stipulation being at least 8 consecutive bases should map before a mismatch is found. The expression of the mature miRNAs were quantified using the sum of all reads mapping to the isoforms arising from the same arm of the hairpin. The mapped data was scaled using a limma based Trimmed Mean of M-values (TMM) and quantile normalized. The miRNAs were called ‘dynamically expressed’ if the expression values were >2 fold different between any two phases of the cell cycle. To remove potentially cross-hybridizing miRNAs to the probes used on the Ncode miRNA microarray from Zhou et.al’s study, we used the probe sequences to map against mature miRNA sequences from miRBase v19. Microarray probes can potentially cross-hybridize with targets with ~80% similarity in sequence. To account for this in a ~22 nucleotide long miRNA, we allowed up to 4 mismatches to generate the list of potential cross-hybridizing miRNAs (Appendix 3.6).

101 4.2.7 Bioinformatic Analysis of transcription factor binding Analysis was performed using the UCSC genome browser, where the promoter regions of the specific miRNAs were scanned for binding regions of transcription factors that play a role in the cell cycle. These binding regions were obtained from ChIP-seq data generated by ENCODE and can be viewed on the genome browser. To ensure that we cover both promoters and enhancer regions, we scanned upto 10kb upstream of the transcription start site of each miRNA in the presence of the H3K27Ac and H3K4me3 marks. The analysis was performed for well-characterized cell cycle associated transcription factors such as c- Myc, FOXM1, NRF1, GATA3, STAT1, E2F4 and E2F6 amongst others.

102 4.3 RESULTS

4.3.1 Synchronization of cancer cell lines and verification of cell populations in different phases To identify miRNAs with dynamic expression across cell cycle phases, we used HeLa and MCF7 cells. We chose these two cell lines, representing two cancer types and the two cell lines were synchronized using different methods. This would overcome biases based on cell specificity and the methodologies used for synchronization to identify a broad set of potential cell cycle associated miRNAs. HeLa cells were synchronized using a double-thymidine block protocol, following which synchrony was assessed by flow cytometry of propidium iodide (PI) stained cells, these cells have been previously used and described (Cloonan, Brown et al. 2008). As shown in Figure 4.1A, at 0hrs all the cells were cycling in S-phase (>97% synchrony), 8 hrs after addition of thymidine majority of cells are in the G2-M phase (>93% synchrony), whereas after 14 hrs cells were in the G0-G1 phase (>74% synchrony). MCF7 cells are estrogen-dependent and undergo growth arrest in its absence. These cells were arrested in the G0 phase by withdrawal of estrogen for 48hrs. A modified version of previously used synchronization protocols was employed [321], resulting in >80% of cells accumulating in the G0-G1 phase after 6hrs of estrogen re-addition, >60% of cells residing in S-phase at 22hrs and

40%, of cells passing through the G2-M phase at 27hrs (Fig 4.1B). Validation of these phases was carried out by immunoblotting for the expression of representative markers i.e. expression of cyclin E2 and cyclin A, whose expression is at its peak during the G1-S transition [322], showed high levels of expression at 18-24hrs following estrogen addition, with cyclin E2 undergoing a time-dependent decrease after 24hrs (Fig 4.1C). From 24-30 hrs following estrogen addition, there was a concomitant increase in the expression of Cyclin B1, known for its role in the G2-M transition and mitosis [323] (Fig 4.1C). Biological replicates were generated of HeLa and MCF7 cells, which were harvested at the above mentioned time points for extraction of total RNA for sequencing (miRNA) analysis. One replicate of the cell line at each time point was then used in gene expression profiling to estimate the relative expression of known cell cycle markers across the three cell cycle phases.

103 A S phase G2/M phase G1/G0 phase

T = 8 hrs T = 0 hrs T = 14 hrs

Hela Cell number Hela Cell number Hela Cell number

DNA content DNA content DNA content

B S phase G2/M phase G1/G0 phase

100 90 80 70 60 50 40 30

% Cells - MCF7 20 10 0 0 18202224262830 Time (hrs) C Estrogen

Ti m e (h ) 0 18202224262830 cycE2

cycA

cycB1

Figure 4.1: Synchronization and cell cycle analysis of HeLa and MCF-7 cells: (A) HeLa cells were synchronized by a double-thymidine block protocol and cells harvested at 0 hrs to represent S-phase, 8 hrs following thymidine addition to represent G2-M phase and 14 hrs post-addition for the G0-G1 time point. This figure panel is reproduced from a previously published study (Cloonan, Brown et al. 2008 (Ref. 228)). (B) MCF-7 cells were synchronized by estrogen withdrawal for 48 hrs followed by its re-addition, with >80% of cells in the G0-G1 phase after 6 hrs of re-addition, >60% of cells in S-phase at 22hrs and >20% of cells in G2-M at 27hrs. (C) Cell cycle phase of synchronized MCF-7 cells was validated by immunoblotting of protein extracts for Cyclin E2 and Cyclin B1 that show maximal expression in the G1-S and G2-M phases respectively.

104

Expression levels of cell cycle markers were analyzed from the microarray data as a means of synchronization validation. Typically, genes such as CDKN2A (p16) decrease at the late G1 stage to allow progression into the S-phase whereas CCND1, CCNE1 and

CCNE2 (Cyclins D1, E1 and E2) are expressed maximally during the G1-S transition, with levels decreasing after entry into S-phase [322]. As shown in Figure 4.2, the expression of CDKN2A decreased in S phase in HeLa cells. Markers of G1-S transition such as CCNE1, CCNE2, CDK2 and CDC25A [322] all showed a decrease in expression after S phase in both cell lines, except CCNE1 and CDK2 whose fold-change in MCF7 and HeLa, respectively, was <2-fold. Markers of G2-M transition such as CCNB1, CDC2, CDC25B and CDC25C [323] showed maximal expression at this stage in both cell lines (Fig 4.2). These results demonstrate efficient synchronization of HeLa and MCF7 cells and are an accurate representation of cell populations in the specific phases of the cell cycle.

4.3.2 Small RNA sequencing of synchronized cancer cell lines for identification of miRNAs with dynamic expression across cell cycle

miRNA sequencing was performed on small RNA populations from G1, S and

G2/M phases of the two synchronized cancer cell lines (Appendix 3.2). Data was mapped to miRBase V19, scaled using a limma based Trimmed Mean of M-Values (TMM) method and quantile normalized. We first filtered out potential cross-mappers within our sequencing dataset by aligning the mature miRNA sequences from miRBase v19 to the hairpin sequences, with the same parameters as the data (see M&M). This list (Appendix 3.3) was then cross-referenced with the sequencing data and potential cross-mapping miRNAs were excluded from further analysis. A fold change of > 2 between any two phases of the cell cycle was used to identify miRNAs with dynamic expression across the cell cycle phases (Appendix 3.4).

105 A

Figure 4.2: Expression of cell cycle markers in synchronized HeLa and MCF-7 cells: Microarray data obtained from the synchronized cells was analyzed for expression of markers of cell cycle progression in (A) HeLa and (B) MCF-7 cells. Expression of markers shown is relative to their levels in the G1 phase. Yellow bars indicate expression levels in the G1, light blue bars indicate relative levels in the S, and dark blue bars in the G2M phase.

106 We looked for miRNAs with known association to cell cycle (Appendix 3.1) in our dataset and found 7/37 miRNAs (Appendix 3.5) to have dynamic expression; five across HeLa (hsa-miR-145-5p, hsa-miR-15a-5p, hsa-miR-195-5p, hsa-miR-331-3p and hsa- miR-424-5p) and two across the MCF7 data (hsa-miR-200b-5p and hsa-miR-503-5p). Surprisingly, although we had previously identified miR-17-5p as dynamic by qRT-PCR of the mature and hairpin sequences [298], we did not identify it to be dynamic across our sequenced dataset. One possible reason for this anomaly is the presence of isomiRs [37]; indeed, separating isomiR reads from the mature miRNA highlighted a different pattern of expression (Appendix 3.5). The mature hsa-miR-17-5p did not show differential expression across the phases whereas the sum of reads mapping back to isomiRs were more representative of results from Cloonan et. al [228]. They showed highest expression in the G2M phase followed by G1G0, whereas the S phase had almost half the number of reads. The TaqMan based qRT-PCR assay relies on a target specific stem-loop primer which then extends the 3’ end for quantitation of the target, suggesting a higher sensitivity to differences in that region. As expected, when we split the reads mapping back to the shorter versions of the miRNA (stop position <37) and the longer versions (stop positions >37), there was a larger variation between the cell cycle phases. This was not found to be the case in MCF7 cells, where neither the mature nor the isomiRs forms had any phase specific distribution, suggesting it was specific to the HeLa cells. Analysis of isomiRs was able to reproduce the expression pattern of miR-17-5p in the same sample set as Cloonan et. al [228], highlighting the importance to understand the measurement bias across technologies and its application to accurate interpretation of the data. These suggest the low overlap between dynamically expressed miRNAs in our dataset and known cell cycle associated miRNAs can be attributed to tissue specificity where expression patterns are highly specific to the cell type, and limitation and differences due the technology used. More importantly, not all cell cycle associated miRNAs are dynamically expressed across the phases; they could exert their influence across the cell cycle phases by modulating specific targets’ expression.

As an additional quality control we examined the expression of the 25 HeLa cyclic miRNAs identified by Zhou et al using array-based miR profiling and our dataset

107 (Appendix 3.5). Three miRNAs have expression below our threshold (>100 transcripts per million (tpm) in at least one phase). Eleven miRNAs were discounted after it was deduced that they are likely false positives arising from cross-hybrization of miRNAs with >80% homology to target probes on the array (Appendix 3.6). Four of the remaining 11 cell cycle related miRNAs were common to both studies (hsa-miR-126, hsa-miR-582, hsa-miR-34a and hsa-miR-221) with concordance in pattern of fold change (Fig 4.3). The disparity arising due to cross-hybridization and lower sensitivity suggests that the major differences between the two studies are likely due to technical variations in protocol.

A miR-34a B miR-126

C miR-221D miR-582

Figure 4.3: Comparison of dynamically expressed miRNAs to previous study: Comparison of miRNAs that show dynamic expression across the cell cycle in HeLa cells between our current study (red line) and Zhou et al., (blue line) where after filtering for cross-mappers and cross-hybridizing miRNAs, an overlap of 4 dynamic miRNAs was observed. (A) miR-34a (B) miR-126 (C) miR-221 (D) miR-582 shows the concordance in pattern of relative fold change Data is represented as fold change ratios between phases since this was the only available data format from Zhou et al.

108 Finally, to shortlist miRNAs for further analysis, we applied a stringent threshold of expression greater than 100 tpm in at least one phase of the cell cycle to ensure that the miRNA was robustly expressed in at least one sample. Using this criterion we identified 9 miRNAs to be dynamically expressed across both cell lines, with an additional 12 miRNAs only in HeLa cells, and 8 miRNAs only in MCF7 cells (Table 4.1), which were used in all subsequent analyses.

Table 4.1: List of miRNA with >100tpm in at least one sample and >2 fold difference between two cell cycle phases in MCF7 and HeLa cells

MCF7 HeLa >100 tpm* miRNA G1G0 S G2M G1G0 S G2M MCF7 Hela mir-451a-5p 464.4 1160.4 776.9 539.6 435.4 8463.4 Yes Yes mir-421-3p 1902.9 792.5 1160.4 1084.4 1204.5 504.4 Yes Yes mir-3975-3p 7.4 55.1 133.7 29.3 3364.7 1587.1 Yes Yes mir-3607-5p 455.0 455.0 2238.3 776.9 3443.9 1033.9 Yes Yes mir-185-5p 1778.7 2387.2 629.5 109.4 49.3 78.3 Yes Yes mir-15b-3p 88.9 427.0 341.4 365.9 85.7 326.1 Yes Yes mir-1234-5p 6576.0 2699.7 5160.1 1033.9 1186.6 497.5 Yes Yes mir-106b-3p 273.2 140.5 301.7 381.1 589.4 275.7 Yes Yes let-7d-5p 1445.1 2323.8 575.4 192.0 85.7 154.4 Yes Yes mir-5095-3p 5.6 12.5 12.4 161.7 479.3 175.0 Yes mir-4485 12.1 17.0 28.7 501.0 2959.1 722.6 Yes mir-3940-3p 1.4 10.6 19.4 222.9 182.5 90.6 Yes mir-3609-3p 21.6 12.5 26.9 149.3 683.6 144.5 Yes mir-338-3p 35.0 22.8 15.7 163.0 78.5 208.2 Yes mir-1468-3p 12.1 5.2 7.2 35.8 121.8 23.0 Yes mir-3159-3p 7.4 21.9 26.9 423.5 890.6 316.8 Yes mir-143-3p 3.5 10.6 9.2 45.6 60.7 1058.4 Yes mir-1248-5p 21.6 49.7 49.1 15.0 186.4 8.9 Yes mir-1247-5p 45.0 86.1 96.7 21.9 116.4 29.1 Yes mir-1246-5p 13.9 19.3 32.4 35.8 226.4 15.1 Yes mir-1180-3p 5.6 22.8 22.9 57.8 153.4 51.5 Yes mir-503-5p 144.5 346.4 232.3 21.9 27.8 61.9 Yes mir-4263-3p 331.2 86.1 166.4 25.3 12.8 3.5 Yes mir-3908-5p 672.1 742.1 1397.8 15.0 32.3 9.5 Yes mir-3196-5p 121.8 36.3 216.9 45.6 2.4 5.4 Yes mir-3195-5p 184.6 42.9 101.2 12.8 9.4 3.5 Yes

109 mir-3195-3p 306.5 156.4 358.3 21.9 0.0 3.5 Yes mir-15a-3p 52.8 173.9 195.8 48.9 23.8 57.9 Yes let-7d-3p 205.8 216.9 82.5 48.9 70.9 12.2 Yes

*>100 transcripts per million in atleast one cell cycle phase

4.3.3 Cell cycle associated transcription factors are not significantly enriched in the promoter regions of miRNAs that are dynamic across the cell cycle phases To explore the possible upstream transcription factors that could regulate the dynamic expression of miRNAs, we used data from the ENCODE project [324]. We asked if there was evidence for binding of transcription factors previously reported to regulate the cell cycle in the promoter regions of our shortlisted miRNAs. The ENCODE Chip-Seq analysis was also performed in HeLa and MCF7 cells, which helps to accurately capture the biological relevance of the presence of such binding sites, since the miRNA-seq was performed in the same cell lines. Defining the transcriptional start sites of the transcripts giving rise to miRNAs is challenging. Studies have reported transcription start sites for miRNAs can be between 2- 10kb upstream, within their promoter regions [325, 326]. To strengthen the promoter analysis, we used the presence of the H3K27Ac and H3K4me3 marks to support the location of proximal promoters and enhancers. H3K27Ac is found near active regulatory elements and active enhancers [327], whereas H3K4me3 marks active promoters [328]. The majority of transcription factor binding sites (TFBS) in the promoter regions of the miRNAs were in regions that contained both these marks (Appendix 3.7), reinforcing the validity of the binding sites. As summarized in Appendix 3.7, we observe that 8 out of these 9 miRNAs (except hsa-let-7d-5p) have binding of transcription factors with a known role in cell cycle, as demonstrated by ChIP-seq such as E2F4, ELK4, FOXM1, NRF1, c-Myc amongst others. To determine if the presence of these TF binding sites occur by chance or are actual effectors of the miRNAs’ dynamic expression across the phases, we looked upstream of an equal number of matched stationary miRNAs, whose expression levels had the same distribution as the miRNAs with dynamic expression across the cell cycle

110 phases (Appendix 3.7). There was no statistically significant difference (p=1.57E-01; degrees of freedom=1), when we used a chi-squared test between the two sets, suggesting the presence of the TFBS could be occurring by chance. Although a significant association was not found between cell cycle related TFs and dynamic miRNAs, we wondered if any cell cycle related TF was bound to only dynamic miRNAs and not the stationary miRNAs. On visual inspection and comparing the TFs (Appendix 3.7), we identified a subset of 5 TFs (FOXM1 in MCF7 and NRF1, E2F4, E2F6 and ELK4 in HeLa) exclusively bound to promoter regions of the cell cycle associated miRNAs. We restricted further analysis to this subset of TFs, to explore if they could potentially regulate the expression of the miRNAs.

4.3.4 Expression analysis of associated transcription factors in the matching microarray data If the transcription factors, uniquely bound to dynamically regulated miRNAs, are indeed regulators of miRNA expression, then the expression of the TFs could also be dynamic across the cell cycle. To determine whether this is the case, the expression levels of the TFs with the potential to regulate our miRNAs of interest (Appendix 3.7) were analysed using microarray data. The expression of NRF1 was dynamic across the cell cycle in HeLa cells, showing an increase in the G2-M phase, whereas FOXM1 was elevated in the S-phase of MCF7 cells (Fig 4.4A). E2F4, E2F6 and ELK4 were not differentially expressed across the cell cycle in HeLa. This suggests that the expression of certain TFs may be different across the different phases of the cell cycle, possibly resulting in differences in their target miRNA expression and inducing phenotypic effects such as changes in the cell cycle.

111 A MCF7 HeLa

B HeLa

C MCF7

Figure 4.4: Correlation between transcription factors and miRNA expression: (A) Expression of NRF1 and FOXM1 were dynamic in HeLa and MCF7 cells, respectively. NRF1 is maximally expressed in the G2-M phase whereas FOXM1 is maximally expressed in the S-phase. E2F4, E2F6 and ELK4 do not show dynamic expression across the HeLa cell cycle (B) Expression pattern of NRF1 correlates with increased expression of miR-15b-3p in G2M but a relative downregulation in S phase, with an inverse pattern of expression for miR-1234-5p across the phase of HeLa cells (C) Expression pattern of FOXM1 correlates with increased expression of miR-15b-3p in the S and G2M phases with reduced expression of miR-1234-5p and miR-106b-3p in the S-phase of MCF7 cells.

112 Next, we compared the expression patterns of the transcription factors (Fig 4.4A, Appendix 3.7) to those of the miRNAs (Appendix 3.7) that they possibly regulate. NRF1 is bound to the promoters of miR-1234-5p and miR-15b-3p and its expression is maximal in the G2-M phase in HeLa cells. The expression of miR-1234-5p decreases whereas that of miR-15b-3p increases in the G2-M phase (Fig 4.4B). Similarly, FOXM1 is expressed maximally in the S-phase of MCF7 cells and is bound to the promoters of miR-1234-5p, miR-106b-3p and miR-15b-3p. The expression of miR-1234-5p and miR-106b-3p decrease in the S-phase in MCF7 cells, whereas that of miR-15b-3p increases during this phase (Fig 4.4C). This brings about an interesting possibility that transcription factors such as NRF1 and FOXM1 could not only activate miRNAs but also repress their transcription. NRF1 has previously been shown to repress the DSPP gene in odontoblasts [329] whereas FOXM1 has been shown to act as a repressor of GATA3 in luminal mammary epithelial cells [330]. Although these findings make it tempting to infer a direct relationship between the expression of the miRNA and TF and their functional output, there are confounding factors like the lag time between transcription and expression of active mature miRNAs. This duration represents about a fourth of the time for the entire cell cycle process (HeLa has doubling time of ~24 hrs and MCF7 has doubling time of ~29hrs), during which changes in level of TF may not be able to alter miRNA transcription levels. Moreover, other aspects such as phosphorylation state can have a major effect on the activity of transcription factors as has been demonstrated for both FOXM1 and NRF1 [331, 332]. Given the small number of samples and many variables not addressed using experimental evidence, this data must be interpreted with caution. Large-scale analysis combined with further downstream validations is warranted, for instance TF-binding data through ChIP may provide a clearer picture of the dynamics of miRNA regulation.

4.3.5 Expression of cell cycle dynamic miRNAs in human cancer samples

As regulation of the cell cycle is one of the 10 hallmarks for cancer initiation and progression [288], we hypothesized that the cell cycle-associated miRNAs identified in this study may also be involved in cancer. Six of the nine miRNAs have previously known associations with at least one cancer type (Table 4.2). To determine if these

113 miRNAs are dysregulated in multiple cancer types, we examined their expression across 12 cancers from The Cancer Genome Atlas (TCGA) database (https://tcga- data.nci.nih.gov/tcga/). We used stringent criteria to define differential expression: (i) miRNAs must be expressed more than 100tpm on average in either tumour or normal; (ii) at least 2 fold difference in expression between normal and tumour samples; and (iii) a significance value <0.05 with bonferonni FDR correction. Supporting the known association of cell-cycle with cancer, 7 out of 9 cell cycle-associated miRNAs were differentially expressed in ≥ 50% of tumor types analyzed, when compared to normal samples of the same tissue type (Table 4.2 and Fig 4.5). For the remaining two microRNAs, (miR-3975-3p and miR-1234-5p), data was not present. In most cases, their over or under expression was concordant with their predicted role as oncogenic or tumour suppressive in the relevant cancer types (Table 4.2). The contradictory finding for some miRNAs (miR-185-5p in colon and breast cancer; let-7d-5p in prostate and breast cancer) and their previously shown role may be attributed to the presence of molecular subtypes within the cancer where the miRNAs could have distinct roles to play. Of the 7 miRNAs with data in TCGA, five miRNAs (miR-421-3p, miR-15b-3p, mir-451a-5p, let-7d-tp and miR-185-5p) were significantly overexpressed in both breast and cervical carcinomas that were represented by the two cell lines used in our study, suggesting their biological relevance.

Table 4.2: Dynamically expressed miRNAs in HeLa and MCF7 and their expression across human cancers in The Cancer Genome Atlas dataset

miRNA Literature TCGA Cancer Role Cancer* Expression Up Down miR-421 Gastric Onc [333] STAD Up 6 0 Hepatocellular Onc [334] LIHC Up miR-185 Prostate TS [335, PRAD n.s. 6 0 336] Colon TS [337] COAD Up Breast TS [338] BRCA Up Let-7d Head and Neck TS [339] HNSC Up 7 1 Single Cell Breast TS [340] BRCA Up Prostate TS [341] PRAD Up miR-106b Hepatocellular Onc [342] LIHC Up 3 0

114 Breast Onc [343] BRCA n.s. Prostate Onc [344] PRAD Up Head and Neck Onc [345] HNSC Up Single Cell miR-451a Lung TS [346] LUAD Down 2 6 Hepatocellular TS [347] LIHC n.s. Colorectal Onc [348] COAD Up Breast TS [349] BRCA Down Lung TS [350] LUAD Down miR-15b Cervical Onc [351] CESC Up 8 1 Hepatocellular TS [352] LIHC Up mir-3607-5p None 5 1 miR-1234-5p None n/a n/a miR-3975 None n/a n/a

*BRCA – Breast adenocarcinoma; CESC – Cervical squamous cell carcinoma and endocervical adenocarcinoma; COAD – Colon adenocarcinoma; HNSC – Head and Neck squamous cell carcinoma; LIHC – Liver hepatocellular carcinoma; LUAD – Lung adenocarcinoma; PRAD – Prostate adenocarcinoma; STAD – Stomach adenocarcinoma; n.s. – not significant; n/a – not available.

We also identified a novel potential oncomiR, e.g. miR-3607-5p with dynamic expression across both cell lines, which does not have a previously reported role in human cancer, but is significantly over-expressed in 5 cancer types and down-regulated in one (Fig 4.5). This apparently contradictory role is seen with at least 3 other miRNAs, miR-451a-5p, miR-15b-3p and lte-7d-5p (Table 4.2), which show both up and down regulation across the different cancer types. Previously, miRNAs like miR-17-5p [218, 353] and miR-182-5p [299, 302, 354-359] have been shown to act as potential tumour suppressors or oncogenes depending on cancer cell type and the target mRNAs expressed in those cells. We have now identified four other miRNAs, which have the potential to be dual functioning in cancer. Further functional characterization of these miRNAs is essential to elucidate their specific role in each of the cancer types.

115 hsa-mir-3607-5p

Breast Adenocarcinoma Cervical Squamous Carcinoma Bladder Urothelial Carcinoma Head & Neck Squamous Carcinoma 104 3 10 103 103 * * *

3 10 2 10 102 102 2 10 1 1 10 10

101 100 100 101

100 10-1 10-1

miRNA expression (transcripts per million) per (transcripts expression miRNA -1 million) per (transcripts expression miRNA 0 10 expressionmiRNA (transcripts per million) 10 -2 million) per (transcripts expression miRNA -2 Tumour Normal 10 10 Tumour Normal Tumour Normal Tumour Normal

Kidney Renal Papillary Carcinoma Lung Adenocarcinoma Liver Hepatocellular Carcinoma Lung Squamous Cell Carcinoma 4 10 3 3 * 3 10 * 10 10 *

llion)

103 102

102 102

102 101

101 101 ession (transcripts per million) per ession (transcripts 101 100

miRNA exprmiRNA miRNA expression (transcripts per million) per (transcripts expression miRNA

0 -1 expressionmiRNA (transcripts per million) 10 10 100 expressionmiRNA (transcripts10 per mi 0 Tumour Normal Tumour Normal Tumour Normal Tumour Normal

Stomach Adenocarcinoma Thyroid Carcinoma Uti er ne Corpus Endometri oi dCarcinoma Prostate Adenocarcinoma 103 104 103 103

102 103 102 102 1 10 2 101 10 100 101 101 100 10-1

miRNA expression (transcripts per million) per (transcripts expression miRNA miRNA expression (transcripts per million) per (transcripts expression miRNA miRNA expression (transcripts per million) per (transcripts expression miRNA 10-2 million) per (transcripts expression miRNA 100 100 10-1 Tumour Normal Tumour Normal Tumour Normal Tumour Normal

Figure 4.5: Analysis of miR-3607 in TCGA tumour samples: Expression of miR-3607 across a range of tumours from publicly available TCGA data showing significant overexpression (as marked by asterisk) in 5 tumour types and down-regulation in one. *represents p-value < 0.05 (unpaired t-test).

116 4.3.6 miR-3607-5p and its predicted target cohort with a potential role in cell cycle and cancer Very little is known about miR-3607-5p, apart from its expression upon exposure to ionizing radiation in lymphoblast cells [360]. To further assess its role in cell cycle regulation and disease, we analyzed TargetScan predicted targets of miR-3607-5p [82] for functional enrichment using Ingenuity Pathway Analysis. We also included predicted targets of miR-3975-3p and miR-1234-5p, about which there is little published evidence supporting biological functions. To identify cellular and molecular functions over represented within these targets, we performed gene set enrichment analysis with these predicted targets and 10 random gene lists of the same size. Appendix 3.8 lists the molecular and cellular functions associated with targets of miR-3607-5p. A category was considered significant if its –log (p-value) was at least 3 standard deviations away from the significance of the mean of the random gene lists. Figure 4.6 shows ontological categories passing these criteria. We find cellular growth and proliferation, DNA Replication, Recombination & Repair, and Tumour Morphology as significantly (p-value < 0.05) over represented within these predicted targets of miR-3607-5p. The enriched targets include well-characterized genes underlying tumorigenesis such as FGF2, FOXO1, FZD7, MMP24, BMP2 amongst others (Appendix 3.8). However we didn’t find such significant associations with either miR3975-3p or miR-1234-5p. Although the predicted targets need experimental validation, it is plausible for miR-3607-5p to have a role in tumorigenesis.

117 Cellular Growth & Proliferation

5 4.5 4 3.5 3 2.5 2 - log (p-value) 1.5 1 0.5 0

miR miR miR 1 2345678910 3607-5p 3975-3p 1234-5p

DNA Replication, Recombination & Repair

5 4.5 4 3.5 3 2.5 2 - log (p-value) 1.5 1 0.5 0

miR miR miR 1 2345678910 3607-5p 3975-3p 1234-5p

Tumor Morphology

3.5

2.5

- log (p-value) 1.5

0.5

0 miR miR miR 1 234567892 3 4 5 6 7 8 9 10 3607-5p 3975-3p 1234-5p

Figure 4.6: Ingenuity Pathway Analysis of predicted targets of phasic miRNAs: Graph displaying the significance of functional enrichment for TargetScan predicted targets of miR-3607-5p, miR-3975-3p and miR-1234-5p cyclic miRNAs. Black arrows indicate the mean significance of randomly selected gene sets of equivalent size, and the grey boxes show +3 standard deviations (grey arrows) away from the mean.

118 4.4 DISCUSSION Understanding the cell cycle is important to gain insights into its dysregulation in diseases such as cancer. In this study we used synchronized HeLa and MCF7 cells at each phase of the cell cycle and profiled their miRNA expression using deep sequencing, to globally survey miRNAs whose expression is dynamic across the phases, and to identify potential regulators of cell cycle. We carried out FACS analysis in HeLa cells and immunoblotting for classic cell cycle markers in MCF7 cells to ensure proper synchronization of cells. We also examined the expression of well-known regulators of the cell cycle to determine their dynamics in our cells as a validation of the individual cell cycle phases in both cell lines. From these data, we identify 9 miRNAs whose expression is dynamic in both HeLa and MCF7 cells. Analysis of upstream regulators using data from the ENCODE consortium on the same cancer cell lines, showed the presence of binding sites for cell cycle regulating transcription factors, some of which are also dynamic in matching mRNA data. The differential expression of about 80% of the oscillating miRNAs in the TCGA cancer dataset potentially suggests that this property might impart a tendency to be either driving or being a passenger in tumorigenesis. For the first time we now show dynamic expression of miR-3607-5p in cell cycle phases and dysregulation in cancer using online datasets, with gene set enrichment analysis of its predicted targets supporting a role for it in cell cycle and tumorigenesis.

Only one previous study has looked at expression of miRNAs across different phases of the cell cycle in HeLa cells using microarrays [320]. Recognizing that identifying miRNAs with dynamic changes across the cell cycle of HeLa cells might not be representative of all epithelial cells, we chose to perform synchronization of another epithelial cancer line, MCF7, an estrogen receptor-positive breast cancer cell line. Although this might still not eliminate cell type-specific idiosyncrasies, miRNAs which show differential expression in both cells lines would be a better representation of those that play a broad role in the cell cycle. To improve on the previous study, we have also performed miRNA-seq. This technology is not bound by known miRNAs, and can identify miRNAs more accurately and with higher sensitivity than microarrays [128, 361], providing a good resource for the community studying miRNAs in cell cycle. We have also expanded our studies to look for an association between these miRNAs and

119 their expression levels across various tumor types and also begun to explore their upstream regulation.

A caveat of the study is that the method of synchronization might have introduced bias in the data, especially in MCF7 cells where estrogen withdrawal and re-addition was used. The ability of estrogen to lead to c-Myc-induced proliferation of MCF7 breast cancer cells has been well documented [362, 363]. Hence, activation of such a transcription factor that has a wide array of targets could potentially lead to the changes in expression of miRNAs that are not cell cycle-relevant, but rather, targets of Myc. One could argue that a double thymidine block would abolish such biases, however, the addition of thymidine to cancer cell lines also has side effects depending on whether mismatch repair pathways are functional or not [364]. Hence we have used a combination of both synchronization techniques to obtain a list of miRNAs that are altered during the cell cycle irrespective of the method of synchronization across two different carcinoma cell lines. The use of different synchronization techniques could also explain the differences in dynamic expression of these miRNAs e.g. miR-3607-5p is expressed at high levels in G2-M phase of MCF7, whereas its expression is maximal at S-phase in HeLa cells. However, the expression of the miRNA is dynamic across the cell cycle and follow up analysis using real-time PCR-based validation could resolve such differences. Future studies on specific cancer cell types would benefit from a combination of synchronization techniques for the same cell line, eg. A comparison of estrogen and thymidine induced synchronization on MCF7 cells will help in delineating any methodology based bias. Another major limitation of the study is the absence of technical replicates for the miRNA-seq and microarray analysis. We attempt to overcome this through the use of two different biologically distinct cell lines and appropriate bioinformatics controls for the analysis, but further downstream validations are required to strengthen these findings. Given the tissues specific nature of a miRNA’s functional repression, the overlap of nine of the 21 (HeLa) and 17 (MCF7) dynamically expressed miRNAs between these two distinct cancer types suggests the use of cancer cell lines from different tissues will help to eliminate miRNAs that play tissue-specific roles and identify regulators of common processes such as the cell cycle

120 Analysis of TCGA data revealed that all the cell cycle-associated miRNAs, excepting mir-1234-5p and mir-3975-3p (where data was not available), are differentially expressed in most of the cancer types analysed. This is perhaps not surprising given that miRNAs that regulate the cell cycle are well known to be deregulated in cancers [173, 365]. We also identified 4 miRNAs with a possible dual nature of either being oncogenic or tumour suppressive in different cell types. This result suggests that further research on multiple cancer types and patient cohorts is necessary to begin exploring miRNAs as therapeutic options and cannot be relied on their extensive characterization in one cell/cancer type.

miRNAs could change with the different phases of the cell cycle if they are required to drive a major role in regulation of those particular phases or transitions. Alternatively, this could be a passenger effect of a cell cycle-dependent transcription factor, which regulates the expression of these miRNAs as a means of fine-tuning its function. In this scenario, the expression of the transcription factor would change during the different phases resulting in a concomitant change in the expression of the miRNA, which in turn could regulate key target mRNAs that are critical for a certain checkpoint. The miRNA may serve as a homeostatic monitor or feedback mechanism to ensure fine regulation of the changes in cell cycle phase. We explored these possibilities by the analysis of promoter regions of miRNAs that were cell cycle phase-dependent in both HeLa and MCF7 cells. This analysis revealed a list of potential regulating transcription factors that could play a role in their transactivation or repression. Interestingly, the transcription factor data (Appendix 3.7) shows binding of the dynamic miRNAs by a unique set of factors that are absent in the promoter regions of the stationary miRNAs including FOXM1, NRF1, E2F4 and E2F6, each of which plays an important role in cell cycle regulation [366-369]. In fact, E2F4 has a known role in the regulation of various miRNAs to regulate the cell cycle, including miR-17-92, mir-22 and Let-7a [370]. Regulation of target genes such as MYC through these miRNAs, enables E2F4 to play a major role in the repression of G0-G1 progression [370]. Hence, these miRNAs that the TFs are bound to could be intermediate factors that they control to regulate the expression of a large number of downstream proteins that ultimately result in changes in the cell cycle. Our analysis did not reveal a significant correlation when comparing stationary and

121 dynamic miRNAs potentially due to the small sample set. A similar analysis done at the genome wide level, may be able to identify a correlation in the presence of cell cycle regulated TFs upstream of dynamically expressed miRNAs, which might provide a platform to study the role of these transcription factors in miRNA-mediated cell cycle control.

In order to test whether these cell cycle-associated miRNAs actually play a role in cell cycle regulation or cancer, further functional assays have to be performed. Cell proliferation assays in response to overexpression and down regulation of these miRNAs will allow us to estimate their effect on a cell’s ability to survive and grow, followed by FACS analysis of the cell cycle to understand its significance on the cell cycle transitions. These results can be expanded further using studies to identify upstream regulators of these functional miRNAs, this could be done by knockdown or overexpression of TFs and observing its effects on the expression of the miRNA simultaneously monitoring the cells’ progression across the cell cycle phases using FACS. This would provide evidence if it is the target repression by the TF or miRNA or both that is causing downstream functional regulation. Direct interaction studies between transcription factors and the promoter regions of miRNAs fused to luciferase-expressing constructs will provide evidence for interaction and mode of regulation. Additional ChIP-seq or ChIP-qPCR experiments may also be performed across different phases of the cell cycle to confirm binding of specific TFs to the promoter regions of miRNAs in a phase-dependent manner.

Although these above mentioned miRNAs were singled out based on their dynamics across the cell cycle, this does not necessarily indicate a functional role in cell cycle progression. Hence the short-listed miRNAs would need to be studied further in the context of their target mRNAs using direct experimental approaches such as biotin pull- down [37, 123, 299, 301, 371-377]. Target identification can be followed up with target- specific characterization studies. These can be based on the targets, their specific functions or the pathways they belong to. Experiments to determine their function could involve target overexpression or inhibition followed by a measure of their effect on the biological pathway. Furthermore studies focusing on regulators of miRNA transcription, which are also responsible for the downstream signaling outcomes need to be performed.

122 What signals do these factors respond to and which of these are responsible for the activation and repression of specific miRNAs? Databases such as TransmiR attempt to consolidate all published TF-miRNA associations [378], however, computational tools that predict TF binding to miRNA promoter regions based on their binding consensus sequence would facilitate prediction of miRNA regulating factors and enable more comprehensive understanding and streamlined validations of signaling, both upstream and downstream of miRNAs.

4.5 ACKNOWLEDGEMENTS Brian Gabrielli1 provided synchronized HeLa cells and Elizabeth Musgrove2 provided synchronized MCF7 cells along with data for the Figures 4.1b and c. Melissa Brown3 performed FACS analysis for Figure 4.1a. Shivangi Wani3 and Keerthana Krishnan3 generated small RNA sequencing libraries. Ehsan Nourbaksh3 sequenced all libraries. Keerthana Krishnan3 performed all data analysis under the supervision of Nicole Cloonan4 and Sean Grimmond5. This work was partially supported by Australian Research Council (ARC) Discovery Project Grant DP1093164. KK is supported by an Australian Post-graduate Award (APA), NC is supported by an ARC Postdoctoral Fellowship.

1Diamantina Institute, Princess Alexandra Hospital, The University of Queensland, Woolloongabba, QLD, Australia 4102 2Cancer Genomics & Biochemistry Laboratory, Peter MacCallum Cancer Centre, East Melbourne, VIC, Australia, 3002 3Queensland Centre for Medical Genomics, Institute for Molecular Bioscience, The University of Queensland, St Lucia, QLD, Australia, 4072 4QIMR Berghofer Medical Research Institute, Genomic Biology Laboratory, 300 Herston Road Herston, QLD, Australia, 4006 5Wolfson Wohl Cancer Research Centre, Institute of Cancer Sciences, University of Glasgow, Glasgow, G61 1BD, United Kingdom.

123

CHAPTER FIVE

General Discussion and Future Directions

5 General Discussion and Future Directions

5.1 Summary of findings: Since their discovery in 1993, miRNAs have been recognized as crucial in diverse cellular and organismal processes, from development to cellular proliferation, and differentiation. There is still considerable effort invested in discovering new miRNAs, and the number of known miRNAs has grown substantially. In 2003, miRBase Release 2 had 506 miRNA entries [379]. Now, miRBase V20 in 2013 contains 24521 miRNA loci [379, 380]. This indicates a relatively new and evolving field of research with the role of miRNAs still being explored in various cellular processes and disease states.

This thesis aimed to identify and characterize the functional role of miRNAs with a cancer association, particularly breast cancer. We have done this by identifying targets of dysregulated miRNAs and elucidating their effect on key cellular processes. The introductory chapter discusses what we know about the mechanisms by which miRNAs are generated, their mode of action and known associations in cancer. Chapter two focuses on miR-182-5p and its role in the DNA Damage Repair pathway of breast cancer and Chapter three elucidates the role of miR-139-5p in cellular processes underlying breast tumor progression. Chapter four takes a more exploratory look at expression profiles of miRNAs across the different phases of cell cycle and their potential association with cancer. In the following sections, the work performed as part of this thesis and its contribution to major themes of miRNA research are discussed.

125 5.2 miRNAs and mRNA target interactions

5.2.1 miRNAs act through multiple mRNA targets to exert a phenotypic effect Since the discovery of miR-15 and miR-16 in the frequently lost regions of chronic lymphocytic leukemia patient samples [184], several miRNAs have been implicated in the initiation and progression of cancer. Initially the studies focused on finding ‘the target’ gene of the miRNA or the cluster, eg. let-7 was shown to downregulate RAS [192] and miR-17-5p down-regulated the expression of AIB1 in breast cancer [257] and, along with miR-20a, was able to regulate expression of E2F1 [381]. These studies typically relied on expensive loss/gain of function studies, and therefore assessing multiple targets was impractical. This was followed by development of bioinformatics tools, like TargetScan [82] and Pictar [382] which could predict targets of miRNAs based on known principles of miRNA targeting mechanisms. However, validation was still typically restricted to single genes or gene families, as was the case with the miR-182’s repression of FOXO3 and MITF in melanoma [302] and its cluster’s regulation of FOXO1 in breast cancer [253] were both considered sufficient to explain its role in tumorigenesis.

siRNAs are another class of small RNAs which unlike miRNAs, can be exogenous (viral or transgene induced) [8] or endogenous (dsRNA, typically from repeats, pseudogenes) [383] in origin. However, siRNAs share the same silencing machinery as miRNAs, Dicer for cleavage [384] and Ago proteins within the silencing complex to regulate target genes [384, 385]. Transfection of different siRNAs resulted in unique gene expression profiles [386] specific to the siRNA. Twelve siRNAs against three genes generated 347 off-targeted genes [387] and >969 transcripts were dysregulated in response to six siRNAs against six genes HeLa cells [388]. This regulation extended to changes in protein levels of unintended targets [389]. Analysis of the off-target transcripts revealed sequence complementarity in their 3’UTR to the 5’end of the siRNAs [388]. Despite strong reliance on single target validations, several studies began to show differential expression of hundreds of genes in response to overexpression of a single miRNA, using gene expression arrays [20, 390]. These findings

126 and similarities of siRNAs to miRNA targeting mechanisms suggested that miRNAs could target multiple genes to achieve their functional specificity and output.

With the increase in the number of bioinformatics software that consistently showed large numbers of predicted targets [82, 102, 104, 382, 391], more studies explored multiple candidates per miRNA. miR-181a could regulate T-cell sensitivity by targeting multiple phosphatases [392]. miR-31 could suppress metastasis by coordinated regulation of RhoA, integrin-alpha5 and radixin [236]. Using large numbers of luciferase assays, miR-17-5p was shown to target 26 binding sites across 21 genes [228]. A study [104] performing large scale validation of 226 target sites (44 predicted targets of miR- 375, 24 for miR-296 and 158 for miR-134) found 168 target sites with at least 30% repression in luciferase activity [104]. High-throughput sequencing of RNA isolated by crosslinking immunoprecipitation (HITS-CLIP) was applied to mouse brain, where the top twenty most abundant miRNAs had an average of 655 targets [302]. A more efficient cross-linking using photoactivatable nucleoside such as 4-thiouridine in the Photoactivatable-ribonucleoside-enhanced crosslinking and immunoprecipitation (PAR- CLIP) was used to compare Ago2-IP on mock transfected and miR-124 and miR-7 transfected Hek293 cells, identifying ~3000 and ~920 transcripts containing PAR-CLIP sequence read clusters [112]. Stable isotope labeling with amino acids in cell culture (SILAC), a high throughput analysis of changes in protein levels, showed 12 of 504 quantified proteins to be repressed (2.4%) by over-expressing miR-1 in HeLa cells [393]. Although methods for high-throughput analysis of proteins are still limited, this study suggested that a single miRNA could regulate a substantial portion of the proteome.

In this thesis, the biotin pull-down method described previously [123], but significantly modified to increase the specificity of the enrichment [394] was employed. This technique has been successfully used in several studies from this laboratory [37, 299, 377] and others [123, 301, 371-375] to experimentally identify hundreds of target genes for miRNAs of interest. Furthermore several targets belonging to the functionally relevant pathways were validated using luciferase assays (validation rate of 87.5% for miR-182-5p and 83.3% for miR-139-5p), strengthening the validity of this approach. This finding supports the principle of concomitant suppression of targets underlying the same

127 biological process by a miRNA to achieve a functional output, as shown before for miR- 31 (Valastyan, Benaich et al. 2009) and miR-17-5p [228, 236].

One of the biggest criticisms of all the high-throughput methods of target identification is the potentially high rate of false positives, clouding our understanding of the true number of miRNA targets. The CLIP based analysis by employing the AGO footprint to determine ‘genuine’ miRNA interactions result in <10% false positive rate [395]. Using analyses from biotin pull-down of several miRNAs, sufficient complementarity to account for ~80% of the biological targets can be identified [394]. Using RNAseq on the pulled down transcripts, at least a further 40% where the transcript structure is different to what is defined in RefSeq can be explained, suggesting a false positive rate of <10% [394] for this technique. Such low false positive rates with a high rate of ‘true-positive’ biologically relevant target identification make these techniques highly suited for miRNA characterization.

Through the identification of multiple target mRNAs, this thesis has substantially added to what was previously known about the underlying biology of the miRNAs examined in this study. Upon response to double-strand DNA breaks, ATM phosphorylates CHEK2, which in turn phosphorylates several substrates including BRCA1, CDC25A, p53 and E2F1 (Stolz, Ertych et al. 2011). If BRCA1 were the only target of miR-182, as had been shown by Moskwa et. al. then its inhibition alone would be insufficient to interfere with homologous recombination (HR) pathway [293]. In fact, being one of the early responders to damage, CHEK2 which also phosphorylates BRCA1 to regulate the double stranded repair mechanism [396] might play a crucial role, as it can activate other downstream regulators. Hence, by inhibiting a range of different proteins that play major roles in the response to DNA damage, including CHEK2, RAD17, ATF1 and CREB5, TP53BP1, SMARCD3 and CDKN1B, miR-182 is able to effectively knock down the ability of the cell to respond to damage and undergo apoptosis. Similarly with miR-139-5p, although literature supported a perceived role underlying metastasis in other cancers [397, 398], this study first showed concomitant suppression of several targets, HRAS, NFKB1, PIK3CA, RAF, and RHOT1 (luciferase validation of binding sites) and protein level repression for HRAS, NFKB1, PIK3CA which belong to key signaling

128 networks underlying metastasis. This in fact has been shown to be the case in other metastasis related miRNAs like miR-31 and miR10b, which target gene regulatory networks to control progression of metastasis [399], making it abundantly clear that miRNAs target multiple genes to exert a particular phenotype.

5.2.2 miRNAs target multiple pathways to exert a biological effect

Tumorigenesis is a multi-step process, which requires disruption of multiple processes and their underlying pathways for progression. This process is typically initiated by the acquisition of certain cellular traits or ‘hallmarks’ [400], such as sustaining proliferative signaling, avoiding immune destruction, and activating invasion and metastasis. Several miRNAs have now been linked to individual hallmarks [401], consistent with the hypothesis above that miRNAs regulate multiple members of the same pathway. Intriguingly, it is now becoming evident that miRNAs can be associated with more than one pathway (and more than one hallmark) to achieve the same biological outcome. Several in-silico approaches first suggested that miRNAs co-regulate multiple pathways to exert a biological affect in a given cancer type [402-404], followed by functional studies which supported this notion. miR-155 was shown to have oncogenic properties by promoting the proliferation of breast cancer cells through the targeting of TP53INP1 [405]. It also has a well-defined role in the promotion of tumor cell invasion and metastasis by the activation of an epithelial-to-mesenchymal transition (EMT) in breast cancer cells in cooperation with TGF-β [249]. Additionally, it also targets SOCS1 leading to activation of the JAK-STAT pathway promoting inflammation [251]. Hence miR-155 is able to promote tumorigenesis through the modulation of three hallmarks and regulating targets with varied functions. Similarly, miR-21 is able to promote proliferation in breast cancers through the targeting of PCDC4 [406]; it is also able to induce their migratory properties [407] and enable anti-apoptotic functions [407]. Prior to the publication on miR-139-5p (Chapter three), only six targets had been identified in six different studies. FoxO1 (Hasseine et al. 2009), Rho-kinase2 (Wong et al. 2011), and c-Fos (Fan et al. 2012) in hepatocellular carcinoma, CXCR4 (Bao et al. 2011) in gastric cancer cells, RAP1B (Guo et al. 2012), and Type I insulin-like growth

129 factor (Shen et al. 2012) in colorectal cancer. Chapter three dramatically expanded this cohort, and new targets include genes belonging to key signaling pathways underlying metastasis. Multiple separate pathways, including Wnt signalling, TGF-β pathway, PI3K lipid signaling and MAP kinase cascade were targeted by regulation of multiple genes. While this finding strengthened its association as an anti-invasive and anti-migratory miRNA, it also highlighted that an assumption that a miRNA only targets one pathway may be overly simplistic. miR-139-5p targeting of mRNAs in the Wnt pathway also suggests a potential role in tumour initiation and stemness [408], which is distinct from its function in inhibiting migration and invasion identified in our study. Subsequent to the publication on miR-182 (Chapter two), its association with breast cancer was expanded to genes and other pathways besides the DNA Damage and Repair pathway. Activation of miR-182 by β-catenin signaling, leads to increased invasiveness and metastasis by targeting RECK [409]. Another study showed miR-182’s repression of MIM (Missing in Metastasis) leading to increased motility and invasiveness by activating RhoA [410]. Knockdown of miR-182 led to activation of CBX7 a positive regulator of E-cadherin, which is involved in normal epithelial cell morphology and frequently lost in neoplasia [411]. In combination to the findings in Chapter two that miR-182 targets genes in the apoptosis pathway such as CDKN1B, BAX and BAK, these studies suggest that this miRNA can contribute to breast tumorigenesis by regulating target genes underlying distinct signaling pathways, which culminate in altering different phenotypic changes. Given that a single miRNA can target hundreds of genes in a given cancer type, this raises the question of whether repression of all targets contribute to a phenotypic change, or if only a subset do. GSEA performed on the targets of miR-182-5p revealed an enrichment of genes that play a role in the cell cycle, prompting us to carry out proliferation assays to test its role in cell cycle progression. However, these assays showed no changes in the cell proliferation levels in the presence of miR-182-5p expression. This highlights a limitation in GSEA based analysis, where the same set of genes are involved in several cellular processes leading to a ‘pseudo-enrichment’ of that function. Another contributing factor is the genuine ability of the miRNA to interact with these target genes based on sequence complementarity but does not always result in a

130 functional repression with a phenotypic output in every cell type. In such cases multiple miRNAs targeting the same genes or isomiRs that also target the same cohort along with the canonical form, together enrich for the ‘functional network’ of genes belonging to core pathways whose repression leads to a specific biological signal [37].

5.2.3 The functional output of a miRNA is determined by the expression of its target genes miRNAs can exert oncogenic phenotypes across different cancer types by targeting the same mRNAs. An example of this is miR-21 targeting PDCD4 in breast [412], cervical [413], and colorectal cancers [414]. However, miRNAs can also exert the same phenotype by targeting a different cohort of genes depending on the cell model. For instance, miR-139 targets NR5A2 in esophageal squamous cell carcinomas to effectively impose anti-invasive properties [415]. It also inhibits the metastasis of laryngeal squamous carcinoma cells through the targeting of CXCR4 [397], and of hepatocellular carcinoma cells by the targeting of Wnt-TCF4 [416]. In another example, miR-155, a well studied oncogenic miRNA, promotes the proliferation of MCF-7 breast cancer cells through the targeting of TP53INP1 [405]. It also promotes the proliferation of MDA-MB- 231 breast cancer cells through the targeting of SOCS1 [251] and BT-474 breast cancer cells through its downstream targeting of FOXO3a [250]. It is possible that common target mRNAs targeted by the miRNAs in different cell types were missed due to technical limitations and differences in the modes of target identification. In the case of miR-139, all studies except from this thesis, employed computational target predicting softwares, typically reliant on conservative miRNA- target binding principles, leading to high number of false negatives. Another possible explanation is the differences in target mRNAs expressed between each cancer/cell type, such that a miRNA target expressed in one cell type might be absent in another. In Chapter three, we performed an analysis for overlap of previously validated target genes and found that 2 out of 5 genes (CXCR4 and RAP1B) were not expressed in MCF7 cells. Hence, miRNAs could regulate distinct mRNAs, depending on their levels of expression, to achieve similar outcomes across different cell types and conditions. An apparently paradoxical phenomenon is the ability of a single miRNA to potentially have both oncogenic and tumor suppressive functions across different tumor

131 types. Analysis of TCGA miRNA sequencing data in the Chapter four highlights several examples of miRNAs that are up-regulated in some cancer types, whilst down-regulated in others. These have the potential to be oncogenic or tumour suppressive, respectively, although this still needs to be experimentally tested. Of the nine miRNAs that were dynamic across the phases of the cell cycle, four showed this pattern. This suggests that apparently contradictory roles for miRNAs may be more common than currently appreciated. There are several examples already in the literature: miR-17-5p, miR-182 and miR-21, amongst others, have been studied extensively and shown to have dual roles in different cancer types [157, 199, 218, 257, 299, 417]. If the function of a miRNA was largely dependent on the mRNAs that it targets, then it is conceivable that its phenotypic effects could vary significantly based on whether or not its targets are expressed. When expressed, the level of mRNA abundance and its fluctuation alters the ability of miRNA to regulate its target genes. The cell can employ various mechanisms, for instance, a transcription factor can increase the expression of mRNA, which is also a target of the miRNA, to levels at which translational repression is no longer effective, thereby altering its functional output. Expression patterns of a specific set of miRNA targets across different tumor types where the miRNA has contradictory effects may help understand the role of these dual functioning miRNAs. miR-182 has oncogenic potential in melanoma [302], endometrial cancer [356] and colon cancer [357], amongst other tumor types. However, its role is more akin to a tumour suppressor in lung [359] and gastric carcinomas [418]. Chapter two identified targets of miR-182, uncovering a broad regulation of the DNA damage response as discussed before. As a potential oncogene, it also targets inhibitors of the cell cycle such as such as CDKN1A, CDKN1B and RB [322] that will presumably prevent cell cycle progression in the event that errors in DNA replication are encountered. Contradictory to the above function, is that miR-182 also targets genes that are important for cell cycle progression such as CDK4/6, CDK2 and CDK7 [322]. Also opposing its role in the repression of DNA repair, is its ability to target BTRC, SKP1, SKP2, and the components of the SCF complex, which are suppressors of the DNA repair pathway. Although further analyses and experiments need to be directed to understanding how the targeting of such counter-intuitive mRNAs could still lead to the observed outcome, prior work on miR-

132 17-5p suggests that up-regulation of mRNAs via transcription factors overcome the functional effects of miRNA targeting [228], and it seems reasonable to hypothesize that a similar mechanism is at play here. There are also instances where the same target mRNAs regulated by the same miRNAs exhibit different effects in multiple cancers. One example is NCOA3, a nuclear receptor co-activator known to promote cell proliferation, which is a target of miR-17-5p in HEK293T cells [228]. miR-17-5p also targets NCOA3 in breast cancer cells where it inhibits cell proliferation [257]. Perhaps the presence of functional estrogen receptor also a nuclear hormone receptor, in the breast cancer cells could alter the function of the co- activator. Moreover, NCOA3 is a co-activator that is actively recruited to transcription factors to alter gene expression. Regulation of different target genes by these transcription factors could also result in differing phenotypes indirectly. Another possibility is the preferential ability of certain signaling pathways to confer oncogenic traits in certain cell types, while performing opposing functions in others. A good example of this scenario is the TGFβ pathway, which has a well-documented role in tumor progression and metastasis in breast cancers. However, TGFβ also plays a tumor- suppressive role in colorectal and pancreatic cancers where deletions and loss-of-function mutations can be found in key proteins of the pathway [419, 420]. Hence a miRNA targeting key components of this pathway would bestow oncogenicity in colorectal and pancreatic tumors while displaying a tumor-suppressive role in breast cancer. Thus, like the individual phenotypic effects of a miRNA, the overall oncogenic or tumor-suppressor potential of a miRNA is a function of the mRNAs it targets. The involvement of large networks of genes ensures that phenotypic effects observed are a result of combinatorial outcomes of multiple signals and pathways making it evident that the function of a miRNA cannot be attributed to one target or one phenotype associated with a specific target. Hence, a rigorous experimental approach to identify and validate miRNA targets should constitute the first step towards understanding its function in a given cell type.

133 5.3 MicroRNAs in breast cancer

5.3.1 Molecular heterogeneity of breast cancer There is no single dominant pathway which characterizes breast tumours, unlike colon cancers which are primarily driven by alterations in the Wnt signaling pathway [421] or pancreatic cancer which are driven by mutations in KRAS [422]. Breast cancer has been classified into approximately 20 major and 18 minor subtypes primarily by histology and morphological properties of the tumor at time of diagnosis [423]. However, around 70% of all breast tumours ultimately fall under the invasive ductal carcinoma not otherwise specified (IDC NOS) category or invasive lobular carcinoma (ILC) subtype [423], indicating that the classification system is not able to account for the wider heterogeneity of the disease and is thus of minimal prognostic value and clinical utility [423]. More recently, gene expression profiling enabled classification of breast cancers into five molecular subtypes: luminal A, luminal B, ERBB2, basal and normal-like [424, 425]. Triple negative breast cancers (TNBCs), a subclass of the basal subtype, lack expression of estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER2). TNBCs account for around 15% of all invasive breast cancers and are associated with a higher probability of relapse and poorer overall survival in the first few years after diagnosis [426]. Stratification of TNBC patients using more reliable markers, based on dysregulated pathways would help to personalize treatment, possibly increasing efficacy. miRNAs being regulators of pathways, have the potential to be valuable therapeutic candidates, providing an avenue to be explored.

5.3.2 miRNA expression signatures in breast cancers Expression profiling has enabled the identification of miRNAs that are differentially expressed between the five molecular subtypes of breast cancer with the same miRNAs able to distinguish luminal and basal subtypes in independent datasets [427]. Recent studies have identified a nine miRNAs signature that could differentiate ductal carcinoma in-situ (DCIS) from invasive breast cancers, this includes let-7d, miR- 210, and -221 which are up-regulated in the invasive transition, possibly regulating signaling pathways that are responsible for the progression of the cancers [428]. However, studies that stratify different subtypes of breast cancer using miRNA signatures

134 do so solely based on expression profiling, with very little or no insight into the targets and function of the specific miRNAs in that context. Perhaps this information is not necessary for the purposes of classification, however the same set of miRNAs may not be useful to classify a different cancer type due to their highly target dependent functionality. Hence, the interpretation of a miRNA signature may change based on which signaling pathways are active in a particular tumor. With the realization that breast cancers are very heterogeneous, it may be a futile exercise identifying a signature that works for a small set of tumors without understanding the biological functions of its composite miRNAs. One way to do this is to develop an understanding of the miRNA function based on accurate determination of its targets and what signaling pathways they regulate.

The first two results chapters of this thesis explored the roles of miR-182-5p and miR-139-5p in breast cancer. The first study analyzed expression of miR-182-5p in different subtypes of breast cancer showing a consistent upregulation in all tumor subtypes, except the TNBC, which had more variability. From identification of targets, miR-182-5p was shown to regulate the homologous recombination (HR) pathway of DNA damage repair (DDR), targeting several key transcripts such as BRCA1 and CHEK2. Hence high expression of miR-182-5p, aside from BRCA mutations, could also be an indicator of a perturbation in the HR pathway. Similarly, expression of miR-139-5p was variable across the TNBC subtype, highlighting the heterogeneity present within this subtype. On a larger cohort of patient samples, it was frequently down-regulated in the most invasive subtypes and subsequent functional analysis showed its ability to regulate pathways that are important for cell migration and invasion that initiate the process of metastasis. Hence the loss of expression of this miRNA may be predictive of an activated MAPK, PI3K, TGF-beta or Wnt signaling pathway. This predictive power based on expression pattern of a miRNA could be used as a tool to decode intra- and inter-tumoral heterogeneity and understand the signaling pathways that are active in different subsets of carcinoma cells.

135 5.3.3 miRNA signatures could predict prognosis As discussed above, the widely used histological classification of breast cancers does not take into account the heterogeneity of the disease. The classification of breast cancers into the molecular subtypes requires gene expression profiling followed by data analysis, protocols that are usually not part of a diagnostic lab that analyses clinical tumor samples to provide a histological report to the physician. Such a setup would benefit from the use of a simple prognostic marker that would substitute for having to perform extensive transcriptome analyses but still provide information on pathways that are active in a particular tumor. This is now possible with miRNA signatures, which have been identified to segregate molecular subtypes of breast cancer as well as differentiate between the DCIS and IDC stages of tumor progression [427, 428]. miRNA expression has been associated with clinicopathological features such as ER positivity [429]. Such signatures have been reliably used to specifically identify ER, PR and HER2 status of breast cancers [430] and have also been used to stratify TNBCs based on predictability of overall survival and metastasis-free survival [431].

Presumed advantages for the use of miRNAs as diagnostics include their stability and ease of detection in serum [432], plasma [433] and even whole blood [434] of patients. Studies have compared expression of miRNA in the tumor to those in the serum, where only 7 out of the 20 oncomiRs highly expressed in the tumor are also present in the serum [429]. This represents about a third of overexpressed miRNAs, a level of non- specificity, which may not be applicable for accurate diagnostic purposes. Circulating levels of miR-155 have shown promise in separating breast cancer patients from normal with expression levels in the serum dropping following chemotherapy or surgery [435]. However, the role of miR-155 in inflammation and immunity [436, 437] maybe a confounding factor in determining its prognostic value in this context. It might just be a ‘side-effect’ of chemotherapy as opposed to a sign of tumour regression. Levels of circulating miR-182 are higher in breast cancer patients than in healthy controls [438]. Moreover, circulating levels of this miRNA also have utility as a prognostic marker as its levels in ER negative and PR negative patients are higher than those in ER positive and PR positive patients, respectively [438]. As observed in Chapter three, expression of miR-139-5p is known to be heterogeneous across the TNBC subtype of breast cancers.

136 Lower expression levels of miR-139-5p may be able to predict higher invasive potential and high propensity for metastasis, whose applicability could be extended to other cancer types such as squamous cell, cervical and hepatocellular carcinoma, where it has been shown to have similar anti-invasive and anti-migratory properties. Such techniques might provide easier and more feasible options to segregate between cancer subtypes in addition to histological procedures, however their utility in a clinical setting requires more studies to establish specificity and reliability before being considered as genuine biomarkers.

5.3.4 Using miRNAs to direct therapy choices Therapy choice based on expression markers began with the uncovering of amplifications in the HER2 gene in 15% of breast cancers, tumors which were classified as having poor clinical prognosis are now treated with a monoclonal antibody against the HER2 receptor as neoadjuvant therapy, significantly increasing the overall survival and disease-free survival of patients that even harbor metastatic disease [439, 440]. The use of miRNAs as clinical tools (discussed above) could also enable a better selection of appropriate treatments based on the specific signaling pathways that are active in a particular cancer type.

A directly translateable case would be miR-182-5p, whose expression impairs functioning of the HR pathway, and directly modulates the sensitivity of cells to PARP inhibitor therapy. miR-182-5p over-expression, to a certain extent, phenocopies the presence of a BRCA mutation. The incidence of BRCA mutation induced clinical breast cancer is around 5% of all breast cancers [441]. This subset of patients would be classified under the TNBC and could possibly receive neo-adjuvant chemotherapy followed by surgery. However, a similar proportion of patients, may not harbor a defect in the BRCA genes but have aberrant expression of miR-182. When gene expression data of tumors containing high levels of miR-182 expression from the TCGA are compared to those with BRCA1/2 mutations, there is almost no overlap: of the 98 patients with significant over-expression of miR-182, only a single patient also had a BRCA mutation, indicating that these two lesions rarely occur simultaneously in a tumor. Although diagnostic tools would have to encompass both these complementary scenarios when

137 being used to direct therapy, testing this miRNA would open up the treatment to a new group of patients previously thought not to respond.

Similarly, expression levels of miR-139-5p could also be used to direct therapy. Current diagnostic tools monitor for expression of HER2 and EGFR as an indication of active MAPK and/or PI3K signaling which will then direct therapy towards the use of Trastuzumab or lapatinib [442]. However, the loss of miR-139-5p may also lead to activation of the MAPK and P13K pathway, which may not manifest as aberrant HER2 or EGFR, but rather affect Ras, which is a downstream protein. Additionally, loss of miR-139 expression could potentially direct the use of anti-TGF-beta [443] and anti-Wnt therapy [444] using agents that are currently in development.

This thesis has thus far explored how understanding the miRNA target repertoire could be useful when deciphering its functional role and to derive prognostic information for clinical utility. Another possible, but longer-term goal would be to use knowledge on miRNA function towards developing novel therapies. The outcomes observed from manipulating miRNAs experimentally make it highly appealing to design miRNA-based therapeutics. It is known that miRNAs are highly stable molecules which resist degradation by RNAses by being enclosed in exosomes and microvesicles [445]. miRNA mimics in the form of synthetic RNA duplexes or anti-miRs that reduce the levels of endogenous miRNAs have both been attempted as therapeutic tools [445, 446]. With the cost of high-throughput genomics decreasing significantly, personalized cancer medicine is now becoming a possible reality, where every cancer patient will receive tailored therapy that is specifically targeted to their disease. To match the speed of diagnostics, new avenues for drug development are also highly sought after and the pursuit of any viable avenue is warranted, especially miRNAs, which have a widespread effect on a whole signaling pathway as opposed to other inhibitors, which would only target nodes. Similar to siRNA and gene therapy, several key considerations need to be taken into account for miRNA-based therapy: (1) delivery of the mimic or anti-miR to the appropriate tissue, (2) vehicle of delivery, (3) minimal or no side-effects and (4) appropriate dosage levels without toxic effects are crucial. Administration of short hairpin RNAs in adult mice have resulted in long term adverse side-effects due to

138 saturation of their cellular processing pathways, also shared by miRNAs [447], such studies need to be performed with potential therapeutic miRNA mimics or anti-miRs for their potential as therapy to be inferred.

5.4 Conclusions and Future Directions

Findings from this thesis and other studies have shown that miRNAs act through multiple targets to exert a phenotypic effect, by targeting multiple genes of a single pathway and multiple pathways in a given cellular state. The functional output of a miRNA is determined by the expression of its target cohort in the given context, which can largely explain the multitude of roles a single miRNA is capable of. Results from this thesis have also been discussed in the context of breast cancer heterogeneity and how miRNA expression signatures and functional analysis can be applied to leverage a therapeutic benefit. Given the importance of miRNAs and subsequent characterization of their target genes, the need for accurate and high-throughput technologies to directly identify endogenous miRNA-target interactions still exists. A new technique to address this issue is crosslinking, ligation, and sequencing of hybrids (CLASH) [448], which combines immunoprecipitation of miRNA machinery (AGO protein) with interacting RNAs followed by partial hydrolysis, ligation, reverse transcription and sequencing. This technique has the advantage of identifying targets and their specific binding sites. Similar to other crosslinking techniques, this method relies on pulling down a component of the miRNA machinery to analyse miRNA-mRNA interactions, and currently requires tagged protein-RNA complexes, which limits its use to cell cultures. Delivery using viral vectors for primary cells or developing mouse models expressing tagged proteins will allow for more cell and animal models to be studied under different conditions. Another caveat is the low efficiency of RNA-RNA ligation which leads the chimeric reads to be less than 2% of the total reads obtained from a library [449]. Optimization steps are required to stabilize and recover more RNA-RNA hybrids. This would enable identification of higher number of miRNA interactions. This protocol provides an avenue to identify novel target binding principles, novel targets and in real time analyse changes to miRNA targeting mechanisms in response to different conditions.

139 Another direct future goal arising out of this thesis would be to explore the regulators of miRNAs. It is conceivable that miR-182-5p or other potential oncogenic miRNAs, could be induced to high levels as a consequence of a genetic aberration in its promoter or enhancer regions or even as a result of up-regulation of a transcription factor that could activate its transcription during the pathogenesis of breast cancer. This could be carried out by surveying its promoter region for binding of transcription factors that are elevated in breast cancer. Alternatively sequence data from TCGA or ICGC could be queried for alterations in the genomic regions, which could identify copy number changes to explain the dysregulation in the miRNA expression.

Similarly, genomic or epigenomic alterations in a regulatory region that controls the transcription of miR-139-5p could result in its silencing across breast tumours. In fact, the promoter region of miR-139-5p has been reported to be silenced by EZH2- induced H3K27me3 in hepatocellular carcinomas [450]. Similarly, CD44 and HER2 can induce deacetylation of the H3K9 mark leading to silencing of miR-139 in gastric cancer cells [451]. Similar mechanisms could be observed in breast cancer cells by carrying out methylation-specific PCR to detect DNA methylation of the promoter or ChIP-qPCR for silencing histone marks such as H3K27me3 or H3K9me3 at the promoter regions of miR- 139-5p. Alternatively, publicly available data from the NIH epigenomics roadmap that catalogues the epigenetic status of various cell types could be mined and analyzed to specifically identify regions of silencing in breast cancer cells. Independent of the role of upstream regulators in the context of miR-139-5p, it would be interesting to test whether expression of mir-139-5p induces a transition from a more mesenchymal morphology to an epithelial phenotype, a reverse EMT (or mesenchymal-to-epithelial transition; MET) in MDA-MB-231 cells. This would provide a mechanism by which miR-139-5p is able to alter the migratory, invasive and metastatic properties of these cells. Understanding the signals to which miRNAs and/or their upstream regulators respond, would enable a top to bottom delineation of signaling pathway that miRNAs are part of.

The level of dysregulation of individual miRNAs in cancer raises the question of whether they are drivers of cancer, or just bystanders. As has been shown in this thesis and several other studies in the last few years, dysregulation of miRNA expression and

140 their target repression has been implicated in tumorigenesis. miRNAs are also found in ‘fragile loci’, which are frequently associated with cancer [449] and promote several phenotypes of cancer. However, high expression of miRNA does not always lead to biological significance. miR-21 is one of the most highly expressed miRNAs in normal liver tissue but has a subdued effect on target repression [376]. Very few miRNAs have been shown to be able to induce transformation: miR-17 [143] and miR-21 [157] are two examples. Mechanistic network analysis of general miRNA functions show they could either switch off target expression or in many cases try to maintain target expression within an optimal window suggesting a more modulatory effect than a driver effect for miRNAs [452, 453]. So is the regulation at the miRNA machinery level more relevant in the analysis of therapeutic benefits for cancer? Are individual subsets of miRNAs dysregulated only to fine tune processes in a cell? Regardless of the answer to these questions, their potential clinical utility in breast and other cancers as diagnostic, prognostic, and therapeutic biomarkers is not impacted, as an oncogenic function is not required for their application in this field. It is likely that miRNAs will continue to attract a lot of attention for their clinical potential in cancer and other diseases.

141

REFERENCES

6 References

1. Mattick, J.S., Challenging the dogma: the hidden layer of non-protein-coding RNAs in complex organisms. Bioessays, 2003. 25(10): p. 930-9. 2. Alexander, R.P., et al., Annotating non-coding regions of the genome. Nat Rev Genet, 2010. 11(8): p. 559-71. 3. Taft, R.J., et al., Non-coding RNAs: regulators of disease. J Pathol, 2010. 220(2): p. 126- 39. 4. Carninci, P., et al., The transcriptional landscape of the mammalian genome. Science, 2005. 309(5740): p. 1559-63. 5. Mattick, J.S., The functional genomics of noncoding RNA. Science, 2005. 309(5740): p. 1527-8. 6. Mattick, J.S., Non-coding RNAs: the architects of eukaryotic complexity. EMBO Rep, 2001. 2(11): p. 986-91. 7. Mattick, J.S., A new paradigm for developmental biology. J Exp Biol, 2007. 210(Pt 9): p. 1526-47. 8. Carthew, R.W. and E.J. Sontheimer, Origins and Mechanisms of miRNAs and siRNAs. Cell, 2009. 136(4): p. 642-55. 9. Ghildiyal, M. and P.D. Zamore, Small silencing RNAs: an expanding universe. Nat Rev Genet, 2009. 10(2): p. 94-108. 10. Malone, C.D. and G.J. Hannon, Small RNAs as guardians of the genome. Cell, 2009. 136(4): p. 656-68. 11. Taft, R.J., et al., Tiny RNAs associated with transcription start sites in animals. Nat Genet, 2009. 41(5): p. 572-8. 12. Dieci, G., M. Preti, and B. Montanini, Eukaryotic snoRNAs: a paradigm for gene expression flexibility. Genomics, 2009. 94(2): p. 83-8. 13. Filipowicz, W. and V. Pogacic, Biogenesis of small nucleolar ribonucleoproteins. Curr Opin Cell Biol, 2002. 14(3): p. 319-27. 14. Matera, A.G., R.M. Terns, and M.P. Terns, Non-coding RNAs: lessons from the small nuclear and small nucleolar RNAs. Nat Rev Mol Cell Biol, 2007. 8(3): p. 209-20. 15. Kapranov, P., et al., RNA maps reveal new RNA classes and a possible function for pervasive transcription. Science, 2007. 316(5830): p. 1484-8. 16. Seila, A.C., et al., Divergent transcription from active promoters. Science, 2008. 322(5909): p. 1849-51. 17. Preker, P., et al., RNA exosome depletion reveals transcription upstream of active human promoters. Science, 2008. 322(5909): p. 1851-4. 18. Clark, M.B. and J.S. Mattick, Long noncoding RNAs in cell biology. Semin Cell Dev Biol, 2011. 22(4): p. 366-76. 19. Amaral, P.P. and J.S. Mattick, Noncoding RNA in development. Mamm Genome, 2008. 19(7-8): p. 454-92. 20. Wang, K.C. and H.Y. Chang, Molecular mechanisms of long noncoding RNAs. Mol Cell, 2011. 43(6): p. 904-14. 21. Bartel, D.P., MicroRNAs: target recognition and regulatory functions. Cell, 2009. 136(2): p. 215-33. 22. Lee, R.C., R.L. Feinbaum, and V. Ambros, The C. elegans heterochronic gene lin-4 encodes small RNAs with antisense complementarity to lin-14. Cell, 1993. 75(5): p. 843- 54.

143 23. Wightman, B., I. Ha, and G. Ruvkun, Posttranscriptional regulation of the heterochronic gene lin-14 by lin-4 mediates temporal pattern formation in C. elegans. Cell, 1993. 75(5): p. 855-62. 24. Lagos-Quintana, M., et al., Identification of novel genes coding for small expressed RNAs. Science, 2001. 294(5543): p. 853-8. 25. Lau, N.C., et al., An abundant class of tiny RNAs with probable regulatory roles in Caenorhabditis elegans. Science, 2001. 294(5543): p. 858-62. 26. Lee, R.C. and V. Ambros, An extensive class of small RNAs in Caenorhabditis elegans. Science, 2001. 294(5543): p. 862-4. 27. Kim, V.N., J. Han, and M.C. Siomi, Biogenesis of small RNAs in animals. Nat Rev Mol Cell Biol, 2009. 10(2): p. 126-39. 28. Han, J., et al., Molecular basis for the recognition of primary microRNAs by the Drosha- DGCR8 complex. Cell, 2006. 125(5): p. 887-901. 29. Havens, M.A., et al., Biogenesis of mammalian microRNAs by a non-canonical processing pathway. Nucleic Acids Res, 2012. 30. Sibley, C.R., et al., The biogenesis and characterization of mammalian microRNAs of mirtron origin. Nucleic Acids Res, 2012. 40(1): p. 438-48. 31. Chiang, H.R., et al., Mammalian microRNAs: experimental evaluation of novel and previously annotated genes. Genes Dev, 2010. 24(10): p. 992-1009. 32. Ghildiyal, M., et al., Sorting of Drosophila small silencing RNAs partitions microRNA* strands into the RNA interference pathway. RNA, 2010. 16(1): p. 43-56. 33. Czech, B., et al., Hierarchical rules for Argonaute loading in Drosophila. Mol Cell, 2009. 36(3): p. 445-56. 34. Okamura, K., N. Liu, and E.C. Lai, Distinct mechanisms for microRNA strand selection by Drosophila Argonautes. Mol Cell, 2009. 36(3): p. 431-44. 35. Diederichs, S. and D.A. Haber, Dual role for argonautes in microRNA processing and posttranscriptional regulation of microRNA expression. Cell, 2007. 131(6): p. 1097-108. 36. Cifuentes, D., et al., A novel miRNA processing pathway independent of Dicer requires Argonaute2 catalytic activity. Science, 2010. 328(5986): p. 1694-8. 37. Cloonan, N., et al., MicroRNAs and their isomiRs function cooperatively to target common biological pathways. Genome Biol, 2011. 12(12): p. R126. 38. Krol, J., I. Loedige, and W. Filipowicz, The widespread regulation of microRNA biogenesis, function and decay. Nat Rev Genet, 2010. 11(9): p. 597-610. 39. Jinek, M. and J.A. Doudna, A three-dimensional view of the molecular machinery of RNA interference. Nature, 2009. 457(7228): p. 405-12. 40. Peters, L. and G. Meister, Argonaute proteins: mediators of RNA silencing. Mol Cell, 2007. 26(5): p. 611-23. 41. Liu, J., et al., Argonaute2 is the catalytic engine of mammalian RNAi. Science, 2004. 305(5689): p. 1437-41. 42. Wu, L., J. Fan, and J.G. Belasco, Importance of translation and nonnucleolytic ago proteins for on-target RNA interference. Curr Biol, 2008. 18(17): p. 1327-32. 43. Schmitter, D., et al., Effects of Dicer and Argonaute down-regulation on mRNA levels in human HEK293 cells. Nucleic Acids Res, 2006. 34(17): p. 4801-15. 44. Eulalio, A., F. Tritschler, and E. Izaurralde, The GW182 protein family in animal cells: new insights into domains required for miRNA-mediated gene silencing. RNA, 2009. 15(8): p. 1433-42. 45. Till, S., et al., A conserved motif in Argonaute-interacting proteins mediates functional interactions through the Argonaute PIWI domain. Nat Struct Mol Biol, 2007. 14(10): p. 897-903.

144 46. Eulalio, A., E. Huntzinger, and E. Izaurralde, GW182 interaction with Argonaute is essential for miRNA-mediated translational repression and mRNA decay. Nat Struct Mol Biol, 2008. 15(4): p. 346-53. 47. Karginov, F.V., et al., Diverse endonucleolytic cleavage sites in the mammalian transcriptome depend upon microRNAs, Drosha, and additional nucleases. Mol Cell, 2010. 38(6): p. 781-8. 48. Shin, C., et al., Expanding the microRNA targeting code: functional sites with centered pairing. Mol Cell, 2010. 38(6): p. 789-802. 49. Yekta, S., I.H. Shih, and D.P. Bartel, MicroRNA-directed cleavage of HOXB8 mRNA. Science, 2004. 304(5670): p. 594-6. 50. Wells, S.E., et al., Circularization of mRNA by eukaryotic translation initiation factors. Mol Cell, 1998. 2(1): p. 135-40. 51. Derry, M.C., et al., Regulation of poly(A)-binding protein through PABP-interacting proteins. Cold Spring Harb Symp Quant Biol, 2006. 71: p. 537-43. 52. Olsen, P.H. and V. Ambros, The lin-4 regulatory RNA controls developmental timing in Caenorhabditis elegans by blocking LIN-14 protein synthesis after the initiation of translation. Dev Biol, 1999. 216(2): p. 671-80. 53. Seggerson, K., L. Tang, and E.G. Moss, Two genetic circuits repress the Caenorhabditis elegans heterochronic gene lin-28 after translation initiation. Dev Biol, 2002. 243(2): p. 215-25. 54. Nottrott, S., M.J. Simard, and J.D. Richter, Human let-7a miRNA blocks protein production on actively translating polyribosomes. Nat Struct Mol Biol, 2006. 13(12): p. 1108-14. 55. Petersen, C.P., et al., Short RNAs repress translation after initiation in mammalian cells. Mol Cell, 2006. 21(4): p. 533-42. 56. Pillai, R.S., et al., Inhibition of translational initiation by Let-7 MicroRNA in human cells. Science, 2005. 309(5740): p. 1573-6. 57. Humphreys, D.T., et al., MicroRNAs control translation initiation by inhibiting eukaryotic initiation factor 4E/cap and poly(A) tail function. Proc Natl Acad Sci U S A, 2005. 102(47): p. 16961-6. 58. Mathonnet, G., et al., MicroRNA inhibition of translation initiation in vitro by targeting the cap-binding complex eIF4F. Science, 2007. 317(5845): p. 1764-7. 59. Wakiyama, M., et al., Let-7 microRNA-mediated mRNA deadenylation and translational repression in a mammalian cell-free system. Genes Dev, 2007. 21(15): p. 1857-62. 60. Djuranovic, S., A. Nahvi, and R. Green, miRNA-mediated gene silencing by translational repression followed by mRNA deadenylation and decay. Science, 2012. 336(6078): p. 237-40. 61. Bazzini, A.A., M.T. Lee, and A.J. Giraldez, Ribosome profiling shows that miR-430 reduces translation before causing mRNA decay in zebrafish. Science, 2012. 336(6078): p. 233-7. 62. Selbach, M., et al., Widespread changes in protein synthesis induced by microRNAs. Nature, 2008. 455(7209): p. 58-63. 63. Baek, D., et al., The impact of microRNAs on protein output. Nature, 2008. 455(7209): p. 64-71. 64. Hendrickson, D.G., et al., Concordant regulation of translation and mRNA abundance for hundreds of targets of a human microRNA. PLoS Biol, 2009. 7(11): p. e1000238. 65. Guo, H., et al., Mammalian microRNAs predominantly act to decrease target mRNA levels. Nature, 2010. 466(7308): p. 835-40. 66. Lim, L.P., et al., Microarray analysis shows that some microRNAs downregulate large numbers of target mRNAs. Nature, 2005. 433(7027): p. 769-73.

145 67. Krutzfeldt, J., et al., Silencing of microRNAs in vivo with 'antagomirs'. Nature, 2005. 438(7068): p. 685-9. 68. Parker, R. and H. Song, The enzymes and control of eukaryotic mRNA turnover. Nat Struct Mol Biol, 2004. 11(2): p. 121-7. 69. Wu, L. and J.G. Belasco, Micro-RNA regulation of the mammalian lin-28 gene during neuronal differentiation of embryonal carcinoma cells. Mol Cell Biol, 2005. 25(21): p. 9198-208. 70. Rehwinkel, J., et al., A crucial role for GW182 and the DCP1:DCP2 decapping complex in miRNA-mediated gene silencing. RNA, 2005. 11(11): p. 1640-7. 71. Wu, L., J. Fan, and J.G. Belasco, MicroRNAs direct rapid deadenylation of mRNA. Proc Natl Acad Sci U S A, 2006. 103(11): p. 4034-9. 72. Eulalio, A., et al., Target-specific requirements for enhancers of decapping in miRNA- mediated gene silencing. Genes Dev, 2007. 21(20): p. 2558-70. 73. Eulalio, A., et al., Deadenylation is a widespread effect of miRNA regulation. RNA, 2009. 15(1): p. 21-32. 74. Behm-Ansmant, I., et al., mRNA degradation by miRNAs and GW182 requires both CCR4:NOT deadenylase and DCP1:DCP2 decapping complexes. Genes Dev, 2006. 20(14): p. 1885-98. 75. Parker, R. and U. Sheth, P bodies and the control of mRNA translation and degradation. Mol Cell, 2007. 25(5): p. 635-46. 76. Eulalio, A., I. Behm-Ansmant, and E. Izaurralde, P bodies: at the crossroads of posttranscriptional pathways. Nat Rev Mol Cell Biol, 2007. 8(1): p. 9-22. 77. Yang, Y., et al., Identifying targets of miR-143 using a SILAC-based proteomic approach. Mol Biosyst, 2010. 6(10): p. 1873-82. 78. Cui, Q., et al., Principles of microRNA regulation of a human cellular signaling network. Mol Syst Biol, 2006. 2: p. 46. 79. Clancy, J.L., et al., mRNA isoform diversity can obscure detection of miRNA-mediated control of translation. RNA, 2011. 17(6): p. 1025-31. 80. Mayr, C. and D.P. Bartel, Widespread shortening of 3'UTRs by alternative cleavage and polyadenylation activates oncogenes in cancer cells. Cell, 2009. 138(4): p. 673-84. 81. Grimson, A., et al., MicroRNA targeting specificity in mammals: determinants beyond seed pairing. Mol Cell, 2007. 27(1): p. 91-105. 82. Lewis, B.P., C.B. Burge, and D.P. Bartel, Conserved seed pairing, often flanked by adenosines, indicates that thousands of human genes are microRNA targets. Cell, 2005. 120(1): p. 15-20. 83. Doench, J.G. and P.A. Sharp, Specificity of microRNA target selection in translational repression. Genes Dev, 2004. 18(5): p. 504-11. 84. Stark, A., et al., Animal MicroRNAs confer robustness to gene expression and have a significant impact on 3'UTR evolution. Cell, 2005. 123(6): p. 1133-46. 85. Gaidatzis, D., et al., Inference of miRNA targets using evolutionary conservation and pathway analysis. BMC Bioinformatics, 2007. 8: p. 69. 86. Majoros, W.H. and U. Ohler, Spatial preferences of microRNA targets in 3' untranslated regions. BMC Genomics, 2007. 8: p. 152. 87. Sandberg, R., et al., Proliferating cells express mRNAs with shortened 3' untranslated regions and fewer microRNA target sites. Science, 2008. 320(5883): p. 1643-7. 88. Kloosterman, W.P., et al., Substrate requirements for let-7 function in the developing zebrafish embryo. Nucleic Acids Res, 2004. 32(21): p. 6284-91. 89. Lytle, J.R., T.A. Yario, and J.A. Steitz, Target mRNAs are repressed as efficiently by microRNA-binding sites in the 5' UTR as in the 3' UTR. Proc Natl Acad Sci U S A, 2007. 104(23): p. 9667-72.

146 90. Tay, Y., et al., MicroRNAs to Nanog, Oct4 and Sox2 coding regions modulate embryonic stem cell differentiation. Nature, 2008. 455(7216): p. 1124-8. 91. Zhang, J., et al., MiR-145, a new regulator of the DNA fragmentation factor-45 (DFF45)- mediated apoptotic network. Mol Cancer, 2010. 9: p. 211. 92. Rajewsky, N., microRNA target predictions in animals. Nat Genet, 2006. 38 Suppl: p. S8-13. 93. Bentwich, I., Prediction and validation of microRNAs and their targets. FEBS Lett, 2005. 579(26): p. 5904-10. 94. Thomas, M., J. Lieberman, and A. Lal, Desperately seeking microRNA targets. Nat Struct Mol Biol, 2010. 17(10): p. 1169-74. 95. Lewis, B.P., et al., Prediction of mammalian microRNA targets. Cell, 2003. 115(7): p. 787-98. 96. Friedman, R.C., et al., Most mammalian mRNAs are conserved targets of microRNAs. Genome Res, 2009. 19(1): p. 92-105. 97. Betel, D., et al., The microRNA.org resource: targets and expression. Nucleic Acids Res, 2008. 36(Database issue): p. D149-53. 98. Rehmsmeier, M., et al., Fast and effective prediction of microRNA/target duplexes. RNA, 2004. 10(10): p. 1507-17. 99. Kiriakidou, M., et al., A combined computational-experimental approach predicts human microRNA targets. Genes Dev, 2004. 18(10): p. 1165-78. 100. Lall, S., et al., A genome-wide map of conserved microRNA targets in C. elegans. Curr Biol, 2006. 16(5): p. 460-71. 101. Stark, A., et al., Identification of Drosophila MicroRNA targets. PLoS Biol, 2003. 1(3): p. E60. 102. Kertesz, M., et al., The role of site accessibility in microRNA target recognition. Nat Genet, 2007. 39(10): p. 1278-84. 103. Hammell, M., et al., mirWIP: microRNA target prediction based on microRNA- containing ribonucleoprotein-enriched transcripts. Nat Methods, 2008. 5(9): p. 813-9. 104. Miranda, K.C., et al., A pattern-based method for the identification of MicroRNA binding sites and their corresponding heteroduplexes. Cell, 2006. 126(6): p. 1203-17. 105. Ragan, C., et al., Transcriptome-wide prediction of miRNA targets in human and mouse using FASTH. PLoS One, 2009. 4(5): p. e5745. 106. Lal, A., et al., miR-24-mediated downregulation of H2AX suppresses DNA repair in terminally differentiated blood cells. Nat Struct Mol Biol, 2009. 16(5): p. 492-8. 107. Cha, Y.H., et al., MiRNA-34 intrinsically links p53 tumor suppressor and Wnt signaling. Cell Cycle, 2012. 11(7): p. 1273-81. 108. Kim, T., et al., p53 regulates epithelial-mesenchymal transition through microRNAs targeting ZEB1 and ZEB2. J Exp Med, 2011. 208(5): p. 875-83. 109. Yook, J.I., et al., A Wnt-Axin2-GSK3beta cascade regulates Snail1 activity in breast cancer cells. Nat Cell Biol, 2006. 8(12): p. 1398-406. 110. Schnall-Levin, M., et al., Unusually effective microRNA targeting within repeat-rich coding regions of mammalian mRNAs. Genome Res, 2011. 21(9): p. 1395-403. 111. Ambros, V., The functions of animal microRNAs. Nature, 2004. 431(7006): p. 350-5. 112. Hafner, M., et al., Transcriptome-wide identification of RNA-binding protein and microRNA target sites by PAR-CLIP. Cell, 2010. 141(1): p. 129-41. 113. Beitzinger, M., et al., Identification of human microRNA targets from isolated argonaute protein complexes. RNA Biol, 2007. 4(2): p. 76-84. 114. Karginov, F.V., et al., A biochemical approach to identifying microRNA targets. Proc Natl Acad Sci U S A, 2007. 104(49): p. 19291-6.

147 115. Hendrickson, D.G., et al., Systematic identification of mRNAs recruited to argonaute 2 by specific microRNAs and corresponding changes in transcript abundance. PLoS One, 2008. 3(5): p. e2126. 116. Chi, S.W., et al., Argonaute HITS-CLIP decodes microRNA-mRNA interaction maps. Nature, 2009. 460(7254): p. 479-86. 117. Su, H., et al., Essential and overlapping functions for mammalian Argonautes in microRNA silencing. Genes Dev, 2009. 23(3): p. 304-17. 118. Zisoulis, D.G., et al., Comprehensive discovery of endogenous Argonaute binding sites in Caenorhabditis elegans. Nat Struct Mol Biol, 2010. 17(2): p. 173-9. 119. Thomson, D.W., C.P. Bracken, and G.J. Goodall, Experimental strategies for microRNA target identification. Nucleic Acids Res, 2011. 39(16): p. 6845-53. 120. de Boer, E., et al., Efficient biotinylation and single-step purification of tagged transcription factors in mammalian cells and transgenic mice. Proc Natl Acad Sci U S A, 2003. 100(13): p. 7480-5. 121. He, A. and W.T. Pu, Genome-wide location analysis by pull down of in vivo biotinylated transcription factors. Curr Protoc Mol Biol, 2010. Chapter 21: p. Unit 21 20. 122. Kliszczak, A.E., et al., DNA mediated chromatin pull-down for the study of chromatin replication. Sci Rep, 2011. 1: p. 95. 123. Orom, U.A. and A.H. Lund, Isolation of microRNA targets using biotinylated synthetic microRNAs. Methods, 2007. 43(2): p. 162-5. 124. Green, N.M., Avidin. 3. The Nature of the Biotin-Binding Site. Biochem J, 1963. 89: p. 599-609. 125. Gonzalez, M., et al., Interaction of biotin with streptavidin. Thermostability and conformational changes upon binding. J Biol Chem, 1997. 272(17): p. 11288-94. 126. Griffiths-Jones, S., et al., miRBase: microRNA sequences, targets and gene nomenclature. Nucleic Acids Res, 2006. 34(Database issue): p. D140-4. 127. Griffiths-Jones, S., et al., miRBase: tools for microRNA genomics. Nucleic Acids Res, 2008. 36(Database issue): p. D154-8. 128. Pritchard, C.C., H.H. Cheng, and M. Tewari, MicroRNA profiling: approaches and considerations. Nat Rev Genet, 2012. 13(5): p. 358-69. 129. Lu, J., et al., MicroRNA expression profiles classify human cancers. Nature, 2005. 435(7043): p. 834-8. 130. Kane, M.D., et al., Assessment of the sensitivity and specificity of oligonucleotide (50mer) microarrays. Nucleic Acids Res, 2000. 28(22): p. 4552-7. 131. Cloonan, N. and S.M. Grimmond, Transcriptome content and dynamics at single- nucleotide resolution. Genome Biol, 2008. 9(9): p. 234. 132. Morin, R.D., et al., Application of massively parallel sequencing to microRNA profiling and discovery in human embryonic stem cells. Genome Res, 2008. 18(4): p. 610-21. 133. Kim, J., et al., Identification and characterization of new microRNAs from pig. Mamm Genome, 2008. 19(7-8): p. 570-80. 134. Reese, T.A., et al., Identification of novel microRNA-like molecules generated from herpesvirus and host tRNA transcripts. J Virol, 2010. 84(19): p. 10344-53. 135. de Hoon, M.J., et al., Cross-mapping and the identification of editing sites in mature microRNAs in high-throughput sequencing libraries. Genome Res, 2010. 20(2): p. 257- 64. 136. Sdassi, N., et al., Identification and characterization of new miRNAs cloned from normal mouse mammary gland. BMC Genomics, 2009. 10: p. 149. 137. Ruby, J.G., et al., Large-scale sequencing reveals 21U-RNAs and additional microRNAs and endogenous siRNAs in C. elegans. Cell, 2006. 127(6): p. 1193-207.

148 138. Ruby, J.G., et al., Evolution, biogenesis, expression, and target predictions of a substantially expanded set of Drosophila microRNAs. Genome Res, 2007. 17(12): p. 1850-64. 139. Landgraf, P., et al., A mammalian microRNA expression atlas based on small RNA library sequencing. Cell, 2007. 129(7): p. 1401-14. 140. Burroughs, A.M., et al., A comprehensive survey of 3' animal miRNA modification events and a possible role for 3' adenylation in modulating miRNA targeting effectiveness. Genome Res, 2010. 20(10): p. 1398-410. 141. Fernandez-Valverde, S.L., R.J. Taft, and J.S. Mattick, Dynamic isomiR regulation in Drosophila development. RNA, 2010. 16(10): p. 1881-8. 142. Berezikov, E., et al., Deep annotation of Drosophila melanogaster microRNAs yields insights into their processing, modification, and emergence. Genome Res, 2011. 21(2): p. 203-15. 143. Hayashita, Y., et al., A polycistronic microRNA cluster, miR-17-92, is overexpressed in human lung cancers and enhances cell proliferation. Cancer Res, 2005. 65(21): p. 9628- 32. 144. Ota, A., et al., Identification and characterization of a novel gene, C13orf25, as a target for 13q31-q32 amplification in malignant lymphoma. Cancer Res, 2004. 64(9): p. 3087- 95. 145. Inomata, M., et al., MicroRNA-17-92 down-regulates expression of distinct targets in different B-cell lymphoma subtypes. Blood, 2009. 113(2): p. 396-402. 146. Manni, I., et al., The microRNA miR-92 increases proliferation of myeloid cells and by targeting p63 modulates the abundance of its isoforms. FASEB J, 2009. 23(11): p. 3957- 66. 147. Wang, P.Y., et al., Regulating A549 cells growth by ASO inhibiting miRNA expression. Mol Cell Biochem, 2010. 339(1-2): p. 163-71. 148. Garzon, R., et al., MicroRNA signatures associated with cytogenetics and prognosis in acute myeloid leukemia. Blood, 2008. 111(6): p. 3183-9. 149. Greither, T., et al., Elevated expression of microRNAs 155, 203, 210 and 222 in pancreatic tumors is associated with poorer survival. Int J Cancer, 2010. 126(1): p. 73- 80. 150. Kluiver, J., et al., BIC and miR-155 are highly expressed in Hodgkin, primary mediastinal and diffuse large B cell lymphomas. J Pathol, 2005. 207(2): p. 243-9. 151. Metzler, M., et al., High expression of precursor microRNA-155/BIC RNA in children with Burkitt lymphoma. Genes Chromosomes Cancer, 2004. 39(2): p. 167-9. 152. Volinia, S., et al., A microRNA expression signature of human solid tumors defines cancer gene targets. Proc Natl Acad Sci U S A, 2006. 103(7): p. 2257-61. 153. Calin, G.A., et al., A MicroRNA signature associated with prognosis and progression in chronic lymphocytic leukemia. N Engl J Med, 2005. 353(17): p. 1793-801. 154. Cai, X., C.H. Hagedorn, and B.R. Cullen, Human microRNAs are processed from capped, polyadenylated transcripts that can also function as mRNAs. RNA, 2004. 10(12): p. 1957-66. 155. Fujita, S., et al., miR-21 Gene expression triggered by AP-1 is sustained through a double-negative feedback mechanism. J Mol Biol, 2008. 378(3): p. 492-504. 156. Chan, J.A., A.M. Krichevsky, and K.S. Kosik, MicroRNA-21 is an antiapoptotic factor in human glioblastoma cells. Cancer Res, 2005. 65(14): p. 6029-33. 157. Si, M.L., et al., miR-21-mediated tumor growth. Oncogene, 2007. 26(19): p. 2799-803. 158. Zhu, S., et al., MicroRNA-21 targets tumor suppressor genes in invasion and metastasis. Cell Res, 2008. 18(3): p. 350-9.

149 159. Asangani, I.A., et al., MicroRNA-21 (miR-21) post-transcriptionally downregulates tumor suppressor Pdcd4 and stimulates invasion, intravasation and metastasis in colorectal cancer. Oncogene, 2008. 27(15): p. 2128-36. 160. Sathyan, P., H.B. Golden, and R.C. Miranda, Competing interactions between micro- RNAs determine neural progenitor survival and proliferation after ethanol exposure: evidence from an ex vivo model of the fetal cerebral cortical neuroepithelium. J Neurosci, 2007. 27(32): p. 8546-57. 161. Yamanaka, Y., et al., Aberrant overexpression of microRNAs activate AKT signaling via down-regulation of tumor suppressors in natural killer-cell lymphoma/leukemia. Blood, 2009. 114(15): p. 3265-75. 162. Park, J.K., et al., Antisense inhibition of microRNA-21 or -221 arrests cell cycle, induces apoptosis, and sensitizes the effects of gemcitabine in pancreatic adenocarcinoma. Pancreas, 2009. 38(7): p. e190-9. 163. Pallante, P., et al., MicroRNA deregulation in human thyroid papillary carcinomas. Endocr Relat Cancer, 2006. 13(2): p. 497-508. 164. Fornari, F., et al., MiR-221 controls CDKN1C/p57 and CDKN1B/p27 expression in human hepatocellular carcinoma. Oncogene, 2008. 27(43): p. 5651-61. 165. Di Leva, G., et al., MicroRNA cluster 221-222 and estrogen receptor alpha interactions in breast cancer. J Natl Cancer Inst, 2010. 102(10): p. 706-21. 166. Felicetti, F., et al., The promyelocytic leukemia zinc finger-microRNA-221/-222 pathway controls melanoma progression through multiple oncogenic mechanisms. Cancer Res, 2008. 68(8): p. 2745-54. 167. le Sage, C., et al., Regulation of the p27(Kip1) tumor suppressor by miR-221 and miR- 222 promotes cancer cell proliferation. EMBO J, 2007. 26(15): p. 3699-708. 168. Garofalo, M., et al., MicroRNA signatures of TRAIL resistance in human non-small cell lung cancer. Oncogene, 2008. 27(27): p. 3845-55. 169. Garofalo, M., et al., miR-221&222 regulate TRAIL resistance and enhance tumorigenicity through PTEN and TIMP3 downregulation. Cancer Cell, 2009. 16(6): p. 498-509. 170. Pineau, P., et al., miR-221 overexpression contributes to liver tumorigenesis. Proc Natl Acad Sci U S A, 2010. 107(1): p. 264-9. 171. Galardi, S., et al., miR-221 and miR-222 expression affects the proliferation potential of human prostate carcinoma cell lines by targeting p27Kip1. J Biol Chem, 2007. 282(32): p. 23716-24. 172. Voorhoeve, P.M., et al., A genetic screen implicates miRNA-372 and miRNA-373 as oncogenes in testicular germ cell tumors. Cell, 2006. 124(6): p. 1169-81. 173. Petrocca, F., et al., E2F1-regulated microRNAs impair TGFbeta-dependent cell-cycle arrest and apoptosis in gastric cancer. Cancer Cell, 2008. 13(3): p. 272-86. 174. Viswanathan, S.R., et al., Lin28 promotes transformation and is associated with advanced human malignancies. Nat Genet, 2009. 41(7): p. 843-8. 175. Ding, J., et al., Gain of miR-151 on chromosome 8q24.3 facilitates tumour cell migration and spreading through downregulating RhoGDIA. Nat Cell Biol, 2010. 12(4): p. 390-9. 176. Nakata, K., et al., MicroRNA-10b is overexpressed in pancreatic cancer, promotes its invasiveness, and correlates with a poor prognosis. Surgery, 2011. 150(5): p. 916-22. 177. Gabriely, G., et al., Human glioma growth is controlled by microRNA-10b. Cancer Res, 2011. 71(10): p. 3563-72. 178. Hildebrandt, M.A., et al., Hsa-miR-9 methylation status is associated with cancer development and metastatic recurrence in patients with clear cell renal cell carcinoma. Oncogene, 2010. 29(42): p. 5724-8. 179. Ma, L., et al., miR-9, a MYC/MYCN-activated microRNA, regulates E-cadherin and cancer metastasis. Nat Cell Biol, 2010. 12(3): p. 247-56.

150 180. Zhu, L., et al., MicroRNA-9 up-regulation is involved in colorectal cancer metastasis via promoting cell motility. Med Oncol, 2011. 181. Mertens-Talcott, S.U., et al., The oncogenic microRNA-27a targets genes that regulate specificity protein transcription factors and the G2-M checkpoint in MDA-MB-231 breast cancer cells. Cancer Res, 2007. 67(22): p. 11001-11. 182. Bandi, N., et al., miR-15a and miR-16 are implicated in cell cycle regulation in a Rb- dependent manner and are frequently deleted or down-regulated in non-small cell lung cancer. Cancer Res, 2009. 69(13): p. 5553-9. 183. Bonci, D., et al., The miR-15a-miR-16-1 cluster controls prostate cancer by targeting multiple oncogenic activities. Nat Med, 2008. 14(11): p. 1271-7. 184. Calin, G.A., et al., Frequent deletions and down-regulation of micro- RNA genes miR15 and miR16 at 13q14 in chronic lymphocytic leukemia. Proc Natl Acad Sci U S A, 2002. 99(24): p. 15524-9. 185. Cimmino, A., et al., miR-15 and miR-16 induce apoptosis by targeting BCL2. Proc Natl Acad Sci U S A, 2005. 102(39): p. 13944-9. 186. Klein, U., et al., The DLEU2/miR-15a/16-1 cluster controls B cell proliferation and its deletion leads to chronic lymphocytic leukemia. Cancer Cell, 2010. 17(1): p. 28-40. 187. Tolcher, A.W., et al., A phase II, pharmacokinetic, and biological correlative study of oblimersen sodium and docetaxel in patients with hormone-refractory prostate cancer. Clin Cancer Res, 2005. 11(10): p. 3854-61. 188. He, L., et al., A microRNA component of the p53 tumour suppressor network. Nature, 2007. 447(7148): p. 1130-4. 189. Cole, K.A., et al., A functional screen identifies miR-34a as a candidate neuroblastoma tumor suppressor gene. Mol Cancer Res, 2008. 6(5): p. 735-42. 190. Li, Y., et al., MicroRNA-34a inhibits glioblastoma growth by targeting multiple oncogenes. Cancer Res, 2009. 69(19): p. 7569-76. 191. Gallardo, E., et al., miR-34a as a prognostic marker of relapse in surgically resected non- small-cell lung cancer. Carcinogenesis, 2009. 30(11): p. 1903-9. 192. Johnson, S.M., et al., RAS is regulated by the let-7 microRNA family. Cell, 2005. 120(5): p. 635-47. 193. Lee, Y.S. and A. Dutta, The tumor suppressor microRNA let-7 represses the HMGA2 oncogene. Genes Dev, 2007. 21(9): p. 1025-30. 194. Sampson, V.B., et al., MicroRNA let-7a down-regulates MYC and reverts MYC-induced growth in Burkitt lymphoma cells. Cancer Res, 2007. 67(20): p. 9762-70. 195. Takamizawa, J., et al., Reduced expression of the let-7 microRNAs in human lung cancers in association with shortened postoperative survival. Cancer Res, 2004. 64(11): p. 3753- 6. 196. Garzon, R., et al., MicroRNA 29b functions in acute myeloid leukemia. Blood, 2009. 114(26): p. 5331-41. 197. Mott, J.L., et al., mir-29 regulates Mcl-1 protein expression and apoptosis. Oncogene, 2007. 26(42): p. 6133-40. 198. Park, S.Y., et al., miR-29 miRNAs activate p53 by targeting p85 alpha and CDC42. Nat Struct Mol Biol, 2009. 16(1): p. 23-9. 199. Iorio, M.V., et al., MicroRNA gene expression deregulation in human breast cancer. Cancer Res, 2005. 65(16): p. 7065-70. 200. Felli, N., et al., MicroRNAs 221 and 222 inhibit normal erythropoiesis and erythroleukemic cell growth via kit receptor down-modulation. Proc Natl Acad Sci U S A, 2005. 102(50): p. 18081-6. 201. Gregory, P.A., et al., The miR-200 family and miR-205 regulate epithelial to mesenchymal transition by targeting ZEB1 and SIP1. Nat Cell Biol, 2008. 10(5): p. 593- 601.

151 202. Wiklund, E.D., et al., Coordinated epigenetic repression of the miR-200 family and miR- 205 in invasive bladder cancer. Int J Cancer, 2011. 128(6): p. 1327-34. 203. Davalos, V., et al., Dynamic epigenetic regulation of the microRNA-200 family mediates epithelial and mesenchymal transitions in human tumorigenesis. Oncogene, 2012. 31(16): p. 2062-74. 204. Roybal, J.D., et al., miR-200 Inhibits lung adenocarcinoma cell invasion and metastasis by targeting Flt1/VEGFR1. Mol Cancer Res, 2011. 9(1): p. 25-35. 205. Zheng, B., et al., MicroRNA-148a suppresses tumor cell invasion and metastasis by downregulating ROCK1 in gastric cancer. Clin Cancer Res, 2011. 17(24): p. 7574-83. 206. Liffers, S.T., et al., MicroRNA-148a is down-regulated in human pancreatic ductal adenocarcinomas and regulates cell survival by targeting CDC25B. Lab Invest, 2011. 91(10): p. 1472-9. 207. Fowler, A., et al., miR-124a is frequently down-regulated in glioblastoma and is involved in migration and invasion. Eur J Cancer, 2011. 47(6): p. 953-63. 208. Agirre, X., et al., Epigenetic silencing of the tumor suppressor microRNA Hsa-miR-124a regulates CDK6 expression and confers a poor prognosis in acute lymphoblastic leukemia. Cancer Res, 2009. 69(10): p. 4443-53. 209. Ando, T., et al., DNA methylation of microRNA genes in gastric mucosae of gastric cancer patients: its possible involvement in the formation of epigenetic field defect. Int J Cancer, 2009. 124(10): p. 2367-74. 210. Valastyan, S., et al., A pleiotropically acting microRNA, miR-31, inhibits breast cancer metastasis. Cell, 2009. 137(6): p. 1032-46. 211. Yamagishi, M., et al., Polycomb-mediated loss of miR-31 activates NIK-dependent NF- kappaB pathway in adult T cell leukemia and other cancers. Cancer Cell, 2012. 21(1): p. 121-35. 212. Cottonham, C.L., S. Kaneko, and L. Xu, miR-21 and miR-31 converge on TIAM1 to regulate migration and invasion of colon carcinoma cells. J Biol Chem, 2010. 285(46): p. 35293-302. 213. Chen, L., et al., miR-137 is frequently down-regulated in glioblastoma and is a negative regulator of Cox-2. Eur J Cancer, 2012. 214. Langevin, S.M., et al., MicroRNA-137 promoter methylation is associated with poorer overall survival in patients with squamous cell carcinoma of the head and neck. Cancer, 2011. 117(7): p. 1454-62. 215. Huang, Y.W., et al., Epigenetic repression of microRNA-129-2 leads to overexpression of SOX4 oncogene in endometrial cancer. Cancer Res, 2009. 69(23): p. 9038-46. 216. Shen, R., et al., Epigenetic repression of microRNA-129-2 leads to overexpression of SOX4 in gastric cancer. Biochem Biophys Res Commun, 2010. 394(4): p. 1047-52. 217. Eis, P.S., et al., Accumulation of miR-155 and BIC RNA in human B cell lymphomas. Proc Natl Acad Sci U S A, 2005. 102(10): p. 3627-32. 218. He, L., et al., A microRNA polycistron as a potential human oncogene. Nature, 2005. 435(7043): p. 828-33. 219. Costinean, S., et al., Pre-B cell proliferation and lymphoblastic leukemia/high-grade lymphoma in E(mu)-miR155 transgenic mice. Proc Natl Acad Sci U S A, 2006. 103(18): p. 7024-9. 220. Raveche, E.S., et al., Abnormal microRNA-16 locus with synteny to human 13q14 linked to CLL in NZB mice. Blood, 2007. 109(12): p. 5079-86. 221. Chang, S., et al., Tumor suppressor BRCA1 epigenetically controls oncogenic microRNA- 155. Nat Med, 2011. 17(10): p. 1275-82. 222. Ventura, A., et al., Targeted deletion reveals essential and overlapping functions of the miR-17 through 92 family of miRNA clusters. Cell, 2008. 132(5): p. 875-86.

152 223. Xiao, C., et al., Lymphoproliferative disease and autoimmunity in mice with increased miR-17-92 expression in lymphocytes. Nat Immunol, 2008. 9(4): p. 405-14. 224. Petrocca, F., A. Vecchione, and C.M. Croce, Emerging role of miR-106b-25/miR-17-92 clusters in the control of transforming growth factor beta signaling. Cancer Res, 2008. 68(20): p. 8191-4. 225. Poliseno, L., et al., Identification of the miR-106b~25 microRNA cluster as a proto- oncogenic PTEN-targeting intron that cooperates with its host gene MCM7 in transformation. Sci Signal, 2010. 3(117): p. ra29. 226. Liu, X., et al., MicroRNA-31 functions as an oncogenic microRNA in mouse and human lung cancer cells by repressing specific tumor suppressors. J Clin Invest, 2010. 120(4): p. 1298-309. 227. Ivanovska, I., et al., MicroRNAs in the miR-106b family regulate p21/CDKN1A and promote cell cycle progression. Mol Cell Biol, 2008. 28(7): p. 2167-74. 228. Cloonan, N., et al., The miR-17-5p microRNA is a key regulator of the G1/S phase cell cycle transition. Genome Biol, 2008. 9(8): p. R127. 229. Ma, L., et al., Therapeutic silencing of miR-10b inhibits metastasis in a mouse mammary tumor model. Nat Biotechnol, 2010. 28(4): p. 341-7. 230. Ambs, S., et al., Genomic profiling of microRNA and messenger RNA reveals deregulated microRNA expression in prostate cancer. Cancer Res, 2008. 68(15): p. 6162- 70. 231. Roccaro, A.M., et al., MicroRNAs 15a and 16 regulate tumor proliferation in multiple myeloma. Blood, 2009. 113(26): p. 6669-80. 232. Yanaihara, N., et al., Unique microRNA molecular profiles in lung cancer diagnosis and prognosis. Cancer Cell, 2006. 9(3): p. 189-98. 233. Pekarsky, Y., et al., Tcl1 expression in chronic lymphocytic leukemia is regulated by miR-29 and miR-181. Cancer Res, 2006. 66(24): p. 11590-3. 234. Fabbri, M., et al., MicroRNA-29 family reverts aberrant methylation in lung cancer by targeting DNA methyltransferases 3A and 3B. Proc Natl Acad Sci U S A, 2007. 104(40): p. 15805-10. 235. Gebeshuber, C.A., K. Zatloukal, and J. Martinez, miR-29a suppresses tristetraprolin, which is a regulator of epithelial polarity and metastasis. EMBO Rep, 2009. 10(4): p. 400-5. 236. Valastyan, S., et al., Concomitant suppression of three target genes can explain the impact of a microRNA on metastasis. Genes Dev, 2009. 23(22): p. 2592-7. 237. Valastyan, S., et al., Activation of miR-31 function in already-established metastases elicits metastatic regression. Genes Dev, 2011. 25(6): p. 646-59. 238. Chang, C.J., et al., p53 regulates epithelial-mesenchymal transition and stem cell properties through modulating miRNAs. Nat Cell Biol, 2011. 13(3): p. 317-23. 239. Nair, V.S., L.S. Maeda, and J.P. Ioannidis, Clinical outcome prediction by microRNAs in human cancer: a systematic review. J Natl Cancer Inst, 2012. 104(7): p. 528-40. 240. Volinia, S., et al., Reprogramming of miRNA networks in cancer and leukemia. Genome Res, 2010. 20(5): p. 589-99. 241. Ferracin, M., et al., MicroRNA profiling for the identification of cancers with unknown primary tissue-of-origin. J Pathol, 2011. 225(1): p. 43-53. 242. Vargo-Gogola, T. and J.M. Rosen, Modelling breast cancer: one size does not fit all. Nat Rev Cancer, 2007. 7(9): p. 659-72. 243. Connolly, E.C., et al., Overexpression of miR-21 Promotes an In vitro Metastatic Phenotype by Targeting the Tumor Suppressor RHOB. Molecular Cancer Research, 2010. 8(5): p. 691-700.

153 244. Qi, L., et al., Expression of miR-21 and its targets (PTEN, PDCD4, TM1) in flat epithelial atypia of the breast in relation to ductal carcinoma in situ and invasive carcinoma. BMC Cancer, 2009. 9(1): p. 163. 245. Zhu, S., et al., MicroRNA-21 Targets the Tumor Suppressor Gene Tropomyosin 1 (TPM1). Journal of Biological Chemistry, 2007. 282(19): p. 14328-14336. 246. Frankel, L.B., et al., Programmed Cell Death 4 (PDCD4) Is an Important Functional Target of the MicroRNA miR-21 in Breast Cancer Cells. Journal of Biological Chemistry, 2008. 283(2): p. 1026-1033. 247. Ma, L., J. Teruya-Feldstein, and R.A. Weinberg, Tumour invasion and metastasis initiated by microRNA-10b in breast cancer. Nature, 2007. 449(7163): p. 682-8. 248. Moriarty, C.H., B. Pursell, and A.M. Mercurio, miR-10b targets Tiam1: implications for Rac activation and carcinoma migration. J Biol Chem, 2010. 285(27): p. 20541-6. 249. Kong, W., et al., MicroRNA-155 is regulated by the transforming growth factor beta/Smad pathway and contributes to epithelial cell plasticity by targeting RhoA. Mol Cell Biol, 2008. 28(22): p. 6773-84. 250. Kong, W., et al., MicroRNA-155 regulates cell survival, growth, and chemosensitivity by targeting FOXO3a in breast cancer. J Biol Chem, 2010. 285(23): p. 17869-79. 251. Jiang, S., et al., MicroRNA-155 functions as an OncomiR in breast cancer by targeting the suppressor of cytokine signaling 1 gene. Cancer Res, 2010. 70(8): p. 3119-27. 252. Huang, Q., et al., The microRNAs miR-373 and miR-520c promote tumour invasion and metastasis. Nat Cell Biol, 2008. 10(2): p. 202-10. 253. Guttilla, I.K. and B.A. White, Coordinate regulation of FOXO1 by miR-27a, miR-96, and miR-182 in breast cancer cells. J Biol Chem, 2009. 284(35): p. 23204-16. 254. Adams, B.D., D.M. Cowee, and B.A. White, The role of miR-206 in the epidermal growth factor (EGF) induced repression of estrogen receptor-alpha (ERalpha) signaling and a luminal phenotype in MCF-7 breast cancer cells. Mol Endocrinol, 2009. 23(8): p. 1215-30. 255. Adams, B.D., H. Furneaux, and B.A. White, The micro-ribonucleic acid (miRNA) miR- 206 targets the human estrogen receptor-alpha (ERalpha) and represses ERalpha messenger RNA and protein expression in breast cancer cell lines. Mol Endocrinol, 2007. 21(5): p. 1132-47. 256. Li, H., et al., miR-17-5p promotes human breast cancer cell migration and invasion through suppression of HBP1. Breast Cancer Res Treat, 2010. 257. Hossain, A., M.T. Kuo, and G.F. Saunders, Mir-17-5p regulates breast cancer cell proliferation by inhibiting translation of AIB1 mRNA. Mol Cell Biol, 2006. 26(21): p. 8191-201. 258. Yu, Z., et al., A cyclin D1/microRNA 17/20 regulatory feedback loop in control of breast cancer cell proliferation. J Cell Biol, 2008. 182(3): p. 509-17. 259. Guo, X., Y. Wu, and R.S. Hartley, MicroRNA-125a represses cell growth by targeting HuR in breast cancer. RNA Biol, 2009. 6(5): p. 575-83. 260. Scott, G.K., et al., Coordinate suppression of ERBB2 and ERBB3 by enforced expression of micro-RNA miR-125a or miR-125b. J Biol Chem, 2007. 282(2): p. 1479-86. 261. Hurteau, G.J., et al., Overexpression of the microRNA hsa-miR-200c leads to reduced expression of transcription factor 8 and increased expression of E-cadherin. Cancer Res, 2007. 67(17): p. 7972-6. 262. Zhao, Y., et al., Let-7 family miRNAs regulate estrogen receptor alpha signaling in estrogen receptor positive breast cancer. Breast Cancer Res Treat, 2010. 263. Yu, F., et al., let-7 regulates self renewal and tumorigenicity of breast cancer cells. Cell, 2007. 131(6): p. 1109-23. 264. Tavazoie, S.F., et al., Endogenous human microRNAs that suppress breast cancer metastasis. Nature, 2008. 451(7175): p. 147-52.

154 265. Tsuchiya, Y., et al., MicroRNA regulates the expression of human cytochrome P450 1B1. Cancer Res, 2006. 66(18): p. 9090-8. 266. Zhang, J., et al., The cell growth suppressor, mir-126, targets IRS-1. Biochem Biophys Res Commun, 2008. 377(1): p. 136-40. 267. Imam, J.S., et al., MicroRNA-185 suppresses tumor growth and progression by targeting the Six1 oncogene in human cancers. Oncogene, 2010. 268. Sachdeva, M. and Y.Y. Mo, MicroRNA-145 suppresses cell invasion and metastasis by directly targeting mucin 1. Cancer Res, 2010. 70(1): p. 378-87. 269. Wang, S., et al., miR-145 inhibits breast cancer cell growth through RTKN. Int J Oncol, 2009. 34(5): p. 1461-6. 270. Spizzo, R., et al., miR-145 participates with TP53 in a death-promoting regulatory loop and targets estrogen receptor-alpha in human breast cancer cells. Cell Death Differ, 2010. 17(2): p. 246-54. 271. Iorio, M.V., et al., microRNA-205 regulates HER3 in human breast cancer. Cancer Res, 2009. 69(6): p. 2195-200. 272. Wu, H., S. Zhu, and Y.Y. Mo, Suppression of cell growth and invasion by miR-205 in breast cancer. Cell Res, 2009. 19(4): p. 439-48. 273. Varambally, S., et al., Genomic loss of microRNA-101 leads to overexpression of histone methyltransferase EZH2 in cancer. Science, 2008. 322(5908): p. 1695-9. 274. Camps, C., et al., hsa-miR-210 Is induced by hypoxia and is an independent prognostic factor in breast cancer. Clin Cancer Res, 2008. 14(5): p. 1340-8. 275. Foekens, J.A., et al., Four miRNAs associated with aggressiveness of lymph node- negative, estrogen receptor-positive human breast cancer. Proc Natl Acad Sci U S A, 2008. 105(35): p. 13021-6. 276. Rothe, F., et al., Global microRNA expression profiling identifies MiR-210 associated with tumor proliferation, invasion and poor clinical outcome in breast cancer. PLoS One, 2011. 6(6): p. e20980. 277. West, S.C., Molecular views of recombination proteins and their control. Nat Rev Mol Cell Biol, 2003. 4(6): p. 435-45. 278. Jiricny, J., The multifaceted mismatch-repair system. Nat Rev Mol Cell Biol, 2006. 7(5): p. 335-46. 279. Lindahl, T. and D.E. Barnes, Repair of endogenous DNA damage. Cold Spring Harb Symp Quant Biol, 2000. 65: p. 127-33. 280. Ciccia, A. and S.J. Elledge, The DNA damage response: making it safe to play with knives. Mol Cell, 2010. 40(2): p. 179-204. 281. Gupta, G.P. and J. Massague, Cancer metastasis: building a framework. Cell, 2006. 127(4): p. 679-95. 282. Fidler, I.J., The pathogenesis of cancer metastasis: the 'seed and soil' hypothesis revisited. Nat Rev Cancer, 2003. 3(6): p. 453-8. 283. Thiery, J.P., Epithelial-mesenchymal transitions in tumour progression. Nat Rev Cancer, 2002. 2(6): p. 442-54. 284. Bailey, J.M., P.K. Singh, and M.A. Hollingsworth, Cancer metastasis facilitated by developmental pathways: Sonic hedgehog, Notch, and bone morphogenic proteins. J Cell Biochem, 2007. 102(4): p. 829-39. 285. Wang, Z., et al., Down-regulation of notch-1 inhibits invasion by inactivation of nuclear factor-kappaB, vascular endothelial growth factor, and matrix metalloproteinase-9 in pancreatic cancer cells. Cancer Res, 2006. 66(5): p. 2778-84. 286. Scheel, C., et al., Paracrine and autocrine signals induce and maintain mesenchymal and stem cell states in the breast. Cell, 2011. 145(6): p. 926-40. 287. Zavadil, J. and E.P. Bottinger, TGF-beta and epithelial-to-mesenchymal transitions. Oncogene, 2005. 24(37): p. 5764-74.

155 288. Hanahan, D. and R.A. Weinberg, The hallmarks of cancer. Cell, 2000. 100(1): p. 57-70. 289. Vermeulen, K., D.R. Van Bockstaele, and Z.N. Berneman, The cell cycle: a review of regulation, deregulation and therapeutic targets in cancer. Cell Prolif, 2003. 36(3): p. 131-49. 290. Visone, R. and C.M. Croce, MiRNAs and cancer. Am J Pathol, 2009. 174(4): p. 1131-8. 291. Liu, C.G., et al., An oligonucleotide microchip for genome-wide microRNA profiling in human and mouse tissues. Proc Natl Acad Sci U S A, 2004. 101(26): p. 9740-4. 292. Xu, G., et al., Transcriptome and targetome analysis in MIR155 expressing cells using RNA-seq. RNA, 2010. 16(8): p. 1610-22. 293. Moskwa, P., et al., miR-182-mediated downregulation of BRCA1 impacts DNA repair and sensitivity to PARP inhibitors. Mol Cell, 2011. 41(2): p. 210-20. 294. Valastyan, S. and R.A. Weinberg, Tumor metastasis: molecular insights and evolving paradigms. Cell, 2011. 147(2): p. 275-92. 295. Friedl, P. and K. Wolf, Tumour-cell invasion and migration: diversity and escape mechanisms. Nat Rev Cancer, 2003. 3(5): p. 362-74. 296. Brodersen, P., et al., Widespread translational inhibition by plant miRNAs and siRNAs. Science, 2008. 320(5880): p. 1185-90. 297. Bagga, S., et al., Regulation by let-7 and lin-4 miRNAs results in target mRNA degradation. Cell, 2005. 122(4): p. 553-63. 298. Cloonan, N., et al., The miR-17-5p microRNA is a key regulator of the G1/S phase cell cycle transition. Genome biology, 2008. 9(8): p. R127. 299. Krishnan, K., et al., MicroRNA-182-5p targets a network of genes involved in DNA repair. RNA, 2013. 19(2): p. 230-42. 300. Lal, A., et al., Capture of microRNA-bound mRNAs identifies the tumor suppressor miR- 34a as a regulator of growth factor signaling. PLoS Genet, 2011. 7(11): p. e1002363. 301. Orom, U.A., F.C. Nielsen, and A.H. Lund, MicroRNA-10a binds the 5'UTR of ribosomal protein mRNAs and enhances their translation. Mol Cell, 2008. 30(4): p. 460-71. 302. Segura, M.F., et al., Aberrant miR-182 expression promotes melanoma metastasis by repressing FOXO3 and microphthalmia-associated transcription factor. Proc Natl Acad Sci U S A, 2009. 106(6): p. 1814-9. 303. Hilgers, V., N. Bushati, and S.M. Cohen, Drosophila microRNAs 263a/b confer robustness during development by protecting nascent sense organs from apoptosis. PLoS Biol, 2010. 8(6): p. e1000396. 304. Krishnan, K., et al., miR-139-5p is a regulator of metastatic pathways in breast cancer. RNA, 2013. 305. Cloonan, N., et al., MicroRNAs and their isomiRs function cooperatively to target common biological pathways. Genome biology, 2011. 12(12): p. R126. 306. Blomen, V.A. and J. Boonstra, Cell fate determination during G1 phase progression. Cell Mol Life Sci, 2007. 64(23): p. 3084-104. 307. Mitchison, T.J. and E.D. Salmon, Mitosis: a history of division. Nat Cell Biol, 2001. 3(1): p. E17-21. 308. Sontheimer, E.J., Assembly and function of RNA silencing complexes. Nat Rev Mol Cell Biol, 2005. 6(2): p. 127-38. 309. Carmichael, J.B., et al., ago1 and dcr1, two core components of the RNA interference pathway, functionally diverge from rdp1 in regulating cell cycle events in Schizosaccharomyces pombe. Mol Biol Cell, 2004. 15(3): p. 1425-35. 310. Provost, P., et al., Dicer is required for chromosome segregation and gene silencing in fission yeast cells. Proc Natl Acad Sci U S A, 2002. 99(26): p. 16648-53. 311. Fukagawa, T., et al., Dicer is essential for formation of the heterochromatin structure in vertebrate cells. Nat Cell Biol, 2004. 6(8): p. 784-91.

156 312. Hatfield, S.D., et al., Stem cell division is regulated by the microRNA pathway. Nature, 2005. 435(7044): p. 974-8. 313. Wang, Y., et al., DGCR8 is essential for microRNA biogenesis and silencing of embryonic stem cell self-renewal. Nat Genet, 2007. 39(3): p. 380-5. 314. Zhou, Y., et al., High-risk myeloma is associated with global elevation of miRNAs and overexpression of EIF2C2/AGO2. Proc Natl Acad Sci U S A, 2010. 107(17): p. 7904-9. 315. Takeshita, F., et al., Systemic delivery of synthetic microRNA-16 inhibits the growth of metastatic prostate tumors via downregulation of multiple cell-cycle genes. Mol Ther, 2010. 18(1): p. 181-7. 316. Liu, Q., et al., miR-16 family induces cell cycle arrest by regulating multiple cell cycle genes. Nucleic Acids Res, 2008. 36(16): p. 5391-404. 317. Linsley, P.S., et al., Transcripts targeted by the microRNA-16 family cooperatively regulate cell cycle progression. Mol Cell Biol, 2007. 27(6): p. 2240-52. 318. Wang, F., et al., Down-regulation of the cyclin E1 oncogene expression by microRNA-16- 1 induces cell cycle arrest in human cancer cells. BMB Rep, 2009. 42(11): p. 725-30. 319. Bueno, M.J. and M. Malumbres, MicroRNAs and the cell cycle. Biochim Biophys Acta, 2011. 1812(5): p. 592-601. 320. Zhou, J.Y., et al., Analysis of microRNA expression profiles during the cell cycle in synchronized HeLa cells. BMB Rep, 2009. 42(9): p. 593-8. 321. D'Assoro, A.B., et al., Impaired p53 function leads to centrosome amplification, acquired ERalpha phenotypic heterogeneity and distant metastases in breast cancer MCF-7 xenografts. Oncogene, 2008. 27(28): p. 3901-11. 322. Sherr, C.J., Cancer cell cycles. Science, 1996. 274(5293): p. 1672-7. 323. Stark, G.R. and W.R. Taylor, Control of the G2/M transition. Molecular biotechnology, 2006. 32(3): p. 227-48. 324. Wang, J., et al., Sequence features and chromatin structure around the genomic regions bound by 119 human transcription factors. Genome Res, 2012. 22(9): p. 1798-812. 325. Saini, H.K., S. Griffiths-Jones, and A.J. Enright, Genomic analysis of human microRNA transcripts. Proceedings of the National Academy of Sciences of the United States of America, 2007. 104(45): p. 17719-24. 326. Marsico, A., et al., PROmiRNA: a new miRNA promoter recognition method uncovers the complex regulation of intronic miRNAs. Genome biology, 2013. 14(8): p. R84. 327. Creyghton, M.P., et al., Histone H3K27ac separates active from poised enhancers and predicts developmental state. Proceedings of the National Academy of Sciences of the United States of America, 2010. 107(50): p. 21931-6. 328. Hon, G.C., R.D. Hawkins, and B. Ren, Predictive chromatin signatures in the mammalian genome. Hum Mol Genet, 2009. 18(R2): p. R195-201. 329. Narayanan, K., et al., The CCAAT enhancer-binding protein (C/EBP)beta and Nrf1 interact to regulate dentin sialophosphoprotein (DSPP) gene expression during odontoblast differentiation. J Biol Chem, 2004. 279(44): p. 45423-32. 330. Carr, J.R., et al., FoxM1 regulates mammary luminal cell fate. Cell Rep, 2012. 1(6): p. 715-29. 331. Joshi, K., et al., MELK-dependent FOXM1 phosphorylation is essential for proliferation of glioma stem cells. Stem Cells, 2013. 31(6): p. 1051-63. 332. Wang, C., et al., Cyclin D1 repression of nuclear respiratory factor 1 integrates nuclear DNA synthesis and mitochondrial function. Proc Natl Acad Sci U S A, 2006. 103(31): p. 11567-72. 333. Jiang, Z., et al., Increased expression of miR-421 in human gastric carcinoma and its clinical association. Journal of gastroenterology, 2010. 45(1): p. 17-23.

157 334. Zhang, Y., et al., Downregulation of human farnesoid X receptor by miR-421 promotes proliferation and migration of hepatocellular carcinoma cells. Molecular cancer research : MCR, 2012. 10(4): p. 516-22. 335. Li, X., et al., MicroRNA-185 and 342 inhibit tumorigenicity and induce apoptosis through blockade of the SREBP metabolic pathway in prostate cancer cells. PLoS One, 2013. 8(8): p. e70987. 336. Qu, F., et al., MicroRNA-185 suppresses proliferation, invasion, migration, and tumorigenicity of human prostate cancer cells through targeting androgen receptor. Mol Cell Biochem, 2013. 377(1-2): p. 121-30. 337. Akcakaya, P., et al., miR-185 and miR-133b deregulation is associated with overall survival and metastasis in colorectal cancer. Int J Oncol, 2011. 39(2): p. 311-8. 338. Imam, J.S., et al., MicroRNA-185 suppresses tumor growth and progression by targeting the Six1 oncogene in human cancers. Oncogene, 2010. 29(35): p. 4971-9. 339. Childs, G., et al., Low-level expression of microRNAs let-7d and miR-205 are prognostic markers of head and neck squamous cell carcinoma. The American journal of pathology, 2009. 174(3): p. 736-45. 340. Volinia, S., et al., Breast cancer signatures for invasiveness and prognosis defined by deep sequencing of microRNA. Proceedings of the National Academy of Sciences of the United States of America, 2012. 109(8): p. 3024-9. 341. Ramberg, H., et al., Regulation of PBX3 expression by androgen and Let-7d in prostate cancer. Mol Cancer, 2011. 10: p. 50. 342. Yau, W.L., et al., Over-expression of miR-106b promotes cell migration and metastasis in hepatocellular carcinoma by activating epithelial-mesenchymal transition process. PLoS One, 2013. 8(3): p. e57882. 343. Smith, A.L., et al., The miR-106b-25 cluster targets Smad7, activates TGF-beta signaling, and induces EMT and tumor initiating cell characteristics downstream of Six1 in human breast cancer. Oncogene, 2012. 31(50): p. 5162-71. 344. Hudson, R.S., et al., MicroRNA-106b-25 cluster expression is associated with early disease recurrence and targets caspase-7 and focal adhesion in human prostate cancer. Oncogene, 2013. 32(35): p. 4139-47. 345. Hui, A.B., et al., Comprehensive MicroRNA profiling for head and neck squamous cell carcinomas. Clin Cancer Res, 2010. 16(4): p. 1129-39. 346. Vosa, U., et al., Meta-analysis of microRNA expression in lung cancer. Int J Cancer, 2013. 132(12): p. 2884-93. 347. Li, H.P., et al., miR-451 inhibits cell proliferation in human hepatocellular carcinoma through direct suppression of IKK-beta. Carcinogenesis, 2013. 34(11): p. 2443-51. 348. Bitarte, N., et al., MicroRNA-451 is involved in the self-renewal, tumorigenicity, and chemoresistance of colorectal cancer stem cells. Stem Cells, 2011. 29(11): p. 1661-71. 349. Bergamaschi, A. and B.S. Katzenellenbogen, Tamoxifen downregulation of miR-451 increases 14-3-3zeta and promotes breast cancer cell survival and endocrine resistance. Oncogene, 2012. 31(1): p. 39-47. 350. Wang, R., et al., MicroRNA-451 functions as a tumor suppressor in human non-small cell lung cancer by targeting ras-related protein 14 (RAB14). Oncogene, 2011. 30(23): p. 2644-58. 351. Myklebust, M.P., et al., MicroRNA-15b is induced with E2F-controlled genes in HPV- related cancer. Br J Cancer, 2011. 105(11): p. 1719-25. 352. Wu, C.S., et al., Downregulation of microRNA-15b by hepatitis B virus X enhances hepatocellular carcinoma proliferation via fucosyltransferase 2-induced Globo H expression. Int J Cancer, 2014. 134(7): p. 1638-47.

158 353. Hossain, A., M.T. Kuo, and G.F. Saunders, Mir-17-5p regulates breast cancer cell proliferation by inhibiting translation of AIB1 mRNA. Molecular and cellular biology, 2006. 26(21): p. 8191-201. 354. Cho, W.C., A.S. Chow, and J.S. Au, Restoration of tumour suppressor hsa-miR-145 inhibits cancer cell growth in lung adenocarcinoma patients with epidermal growth factor receptor mutation. Eur J Cancer, 2009. 45(12): p. 2197-206. 355. Kong, W.Q., et al., MicroRNA-182 targets cyclic adenosine monophosphate responsive element binding protein 1 (CREB1) and suppresses cell growth in human gastric adenocarcinoma. FEBS J, 2012. 356. Myatt, S.S., et al., Definition of microRNAs that repress expression of the tumor suppressor gene FOXO1 in endometrial cancer. Cancer Res, 2010. 70(1): p. 367-77. 357. Sarver, A.L., et al., Human colon cancer profiles show differential microRNA expression depending on mismatch repair status and are characteristic of undifferentiated proliferative states. BMC Cancer, 2009. 9: p. 401. 358. Schaefer, A., et al., Diagnostic and prognostic implications of microRNA profiling in prostate carcinoma. Int J Cancer, 2010. 126(5): p. 1166-76. 359. Sun, Y., et al., Hsa-mir-182 suppresses lung tumorigenesis through down regulation of RGS17 expression in vitro. Biochem Biophys Res Commun, 2010. 396(2): p. 501-7. 360. Chaudhry, M.A., et al., Identification of radiation-induced microRNA transcriptome by next-generation massively parallel sequencing. J Radiat Res, 2013. 54(5): p. 808-22. 361. Git, A., et al., Systematic comparison of microarray profiling, real-time PCR, and next- generation sequencing technologies for measuring differential microRNA expression. RNA, 2010. 16(5): p. 991-1006. 362. Dubik, D., T.C. Dembinski, and R.P. Shiu, Stimulation of c-myc oncogene expression associated with estrogen-induced proliferation of human breast cancer cells. Cancer Res, 1987. 47(24 Pt 1): p. 6517-21. 363. Dubik, D. and R.P. Shiu, Transcriptional regulation of c-myc oncogene expression by estrogen in hormone-responsive human breast cancer cells. J Biol Chem, 1988. 263(25): p. 12705-8. 364. Mohindra, A., et al., Defects in homologous recombination repair in mismatch-repair- deficient tumour cell lines. Hum Mol Genet, 2002. 11(18): p. 2189-200. 365. Lal, A., et al., miR-24 Inhibits cell proliferation by targeting E2F2, MYC, and other cell- cycle genes via binding to "seedless" 3'UTR microRNA recognition elements. Mol Cell, 2009. 35(5): p. 610-25. 366. Leung, T.W., et al., Over-expression of FoxM1 stimulates cyclin B1 expression. FEBS Lett, 2001. 507(1): p. 59-66. 367. Jacquot, C., et al., Effect of four genes (ALDH1, NRF1, JAM and KBL) on proliferation arrest in a non-small cell bronchopulmonary cancer line. Anticancer Res, 2002. 22(4): p. 2229-35. 368. Beshiri, M.L., et al., Coordinated repression of cell cycle genes by KDM5A and E2F4 during differentiation. Proc Natl Acad Sci U S A, 2012. 109(45): p. 18499-504. 369. Bertoli, C., et al., Chk1 Inhibits E2F6 Repressor Function in Response to Replication Stress to Maintain Cell-Cycle Transcription. Curr Biol, 2013. 23(17): p. 1629-37. 370. Lee, B.K., A.A. Bhinge, and V.R. Iyer, Wide-ranging functions of E2F4 in transcriptional activation and repression revealed by genome-wide analysis. Nucleic Acids Res, 2011. 39(9): p. 3558-73. 371. Christoffersen, N.R., et al., p53-independent upregulation of miR-34a during oncogene- induced senescence represses MYC. Cell Death Differ, 2010. 17(2): p. 236-45. 372. Hutchison, E.R., et al., Evidence for miR-181 involvement in neuroinflammatory responses of astrocytes. Glia, 2013. 61(7): p. 1018-28.

159 373. Le, M.T., et al., Conserved regulation of p53 network dosage by microRNA-125b occurs through evolving miRNA-target gene pairs. PLoS Genet, 2011. 7(9): p. e1002242. 374. Li, S., et al., Hepato-specific microRNA-122 facilitates accumulation of newly synthesized miRNA through regulating PRKRA. Nucleic Acids Res, 2012. 40(2): p. 884- 91. 375. Zhang, Z., et al., Negative regulation of lncRNA GAS5 by miR-21. Cell Death Differ, 2013. 20(11): p. 1558-68. 376. Androsavich, J.R., et al., Disease-linked microRNA-21 exhibits drastically reduced mRNA binding and silencing activity in healthy mouse liver. RNA, 2012. 18(8): p. 1510- 26. 377. Krishnan, K., et al., miR-139-5p is a regulator of metastatic pathways in breast cancer. RNA, 2013. 19(12): p. 1767-80. 378. Wang, J., et al., TransmiR: a transcription factor-microRNA regulation database. Nucleic Acids Res, 2010. 38(Database issue): p. D119-22. 379. Griffiths-Jones, S., The microRNA Registry. Nucleic Acids Res, 2004. 32(Database issue): p. D109-11. 380. Kozomara, A. and S. Griffiths-Jones, miRBase: annotating high confidence microRNAs using deep sequencing data. Nucleic Acids Res, 2014. 42(Database issue): p. D68-73. 381. O'Donnell, K.A., et al., c-Myc-regulated microRNAs modulate E2F1 expression. Nature, 2005. 435(7043): p. 839-43. 382. Krek, A., et al., Combinatorial microRNA target predictions. Nat Genet, 2005. 37(5): p. 495-500. 383. Golden, D.E., V.R. Gerbasi, and E.J. Sontheimer, An inside job for siRNAs. Mol Cell, 2008. 31(3): p. 309-12. 384. Tomari, Y. and P.D. Zamore, Perspective: machines for RNAi. Genes Dev, 2005. 19(5): p. 517-29. 385. Meister, G. and T. Tuschl, Mechanisms of gene silencing by double-stranded RNA. Nature, 2004. 431(7006): p. 343-9. 386. Jackson, A.L., et al., Expression profiling reveals off-target gene regulation by RNAi. Nat Biotechnol, 2003. 21(6): p. 635-7. 387. Birmingham, A., et al., 3' UTR seed matches, but not overall identity, are associated with RNAi off-targets. Nat Methods, 2006. 3(3): p. 199-204. 388. Jackson, A.L., et al., Widespread siRNA "off-target" transcript silencing mediated by seed region sequence complementarity. RNA, 2006. 12(7): p. 1179-87. 389. Scacheri, P.C., et al., Short interfering RNAs can induce unexpected and divergent changes in the levels of untargeted proteins in mammalian cells. Proc Natl Acad Sci U S A, 2004. 101(7): p. 1892-7. 390. Dahiya, N., et al., MicroRNA expression and identification of putative miRNA targets in ovarian cancer. PLoS One, 2008. 3(6): p. e2436. 391. John, B., et al., Human MicroRNA targets. PLoS Biol, 2004. 2(11): p. e363. 392. Li, Q.J., et al., miR-181a is an intrinsic modulator of T cell sensitivity and selection. Cell, 2007. 129(1): p. 147-61. 393. Vinther, J., et al., Identification of miRNA targets with stable isotope labeling by amino acids in cell culture. Nucleic Acids Res, 2006. 34(16): p. e107. 394. Martin, H.C., et al., Imperfect centered miRNA binding sites are common and can mediate repression of target mRNAs. Genome Biol, 2014. 15(3): p. R51. 395. Lebedeva, S., et al., Transcriptome-wide analysis of regulatory interactions of the RNA- binding protein HuR. Mol Cell, 2011. 43(3): p. 340-52. 396. Zhang, J., et al., Chk2 phosphorylation of BRCA1 regulates DNA double-strand break repair. Mol Cell Biol, 2004. 24(2): p. 708-18.

160 397. Luo, H.N., et al., MiR-139 targets CXCR4 and inhibits the proliferation and metastasis of laryngeal squamous carcinoma cells. Med Oncol, 2014. 31(1): p. 789. 398. Shen, K., et al., MiR-139 inhibits invasion and metastasis of colorectal cancer by targeting the type I insulin-like growth factor receptor. Biochem Pharmacol, 2012. 84(3): p. 320-30. 399. Pencheva, N. and S.F. Tavazoie, Control of metastatic progression by microRNA regulatory networks. Nat Cell Biol, 2013. 15(6): p. 546-54. 400. Hanahan, D. and R.A. Weinberg, Hallmarks of cancer: the next generation. Cell, 2011. 144(5): p. 646-74. 401. Ruan, K., X. Fang, and G. Ouyang, MicroRNAs: novel regulators in the hallmarks of human cancer. Cancer Lett, 2009. 285(2): p. 116-26. 402. Chen, S.J. and H.C. Chen, Analysis of targets and functions coregulated by microRNAs. Methods Mol Biol, 2011. 676: p. 225-41. 403. Gusev, Y., Computational methods for analysis of cellular functions and pathways collectively targeted by differentially expressed microRNA. Methods, 2008. 44(1): p. 61- 72. 404. Zhang, H., et al., In-depth bioinformatic analysis of lung cancer-associated microRNA targets. Oncol Rep, 2013. 30(6): p. 2945-56. 405. Zhang, C.M., J. Zhao, and H.Y. Deng, MiR-155 promotes proliferation of human breast cancer MCF-7 cells through targeting tumor protein 53-induced nuclear protein 1. J Biomed Sci, 2013. 20: p. 79. 406. Frankel, L.B., et al., Programmed cell death 4 (PDCD4) is an important functional target of the microRNA miR-21 in breast cancer cells. J Biol Chem, 2008. 283(2): p. 1026-33. 407. Yan, L.X., et al., Knockdown of miR-21 in human breast cancer cell lines inhibits proliferation, in vitro migration and in vivo tumor growth. Breast Cancer Res, 2011. 13(1): p. R2. 408. Lindvall, C., et al., Wnt signaling, stem cells, and the cellular origin of breast cancer. Stem Cell Rev, 2007. 3(2): p. 157-68. 409. Chiang, C.H., M.F. Hou, and W.C. Hung, Up-regulation of miR-182 by beta-catenin in breast cancer increases tumorigenicity and invasiveness by targeting the matrix metalloproteinase inhibitor RECK. Biochim Biophys Acta, 2013. 1830(4): p. 3067-76. 410. Lei, R., et al., Suppression of MIM by microRNA-182 activates RhoA and promotes breast cancer metastasis. Oncogene, 2014. 33(10): p. 1287-96. 411. Hannafon, B.N., et al., Expression of microRNA and their gene targets are dysregulated in preinvasive breast cancer. Breast Cancer Res, 2011. 13(2): p. R24. 412. Lu, Z., et al., MicroRNA-21 promotes cell transformation by targeting the programmed cell death 4 gene. Oncogene, 2008. 27(31): p. 4373-9. 413. Yao, Q., et al., MicroRNA-21 promotes cell proliferation and down-regulates the expression of programmed cell death 4 (PDCD4) in HeLa cervical carcinoma cells. Biochem Biophys Res Commun, 2009. 388(3): p. 539-42. 414. Pan, X., Z.X. Wang, and R. Wang, MicroRNA-21: a novel therapeutic target in human cancer. Cancer Biol Ther, 2010. 10(12): p. 1224-32. 415. Liu, R., et al., Tumor-suppressive function of miR-139-5p in esophageal squamous cell carcinoma. PLoS One, 2013. 8(10): p. e77068. 416. Gu, W., X. Li, and J. Wang, miR-139 regulates the proliferation and invasion of hepatocellular carcinoma through the WNT/TCF-4 pathway. Oncol Rep, 2014. 31(1): p. 397-404. 417. Zhang, L., et al., microRNAs exhibit high frequency genomic alterations in human cancer. Proc Natl Acad Sci U S A, 2006. 103(24): p. 9136-41.

161 418. Kong, W.Q., et al., MicroRNA-182 targets cAMP-responsive element-binding protein 1 and suppresses cell growth in human gastric adenocarcinoma. FEBS J, 2012. 279(7): p. 1252-60. 419. Villanueva, A., et al., Disruption of the antiproliferative TGF-beta signaling pathways in human pancreatic cancer cells. Oncogene, 1998. 17(15): p. 1969-78. 420. Grady, W.M., et al., Mutational inactivation of transforming growth factor beta receptor type II in microsatellite stable colon cancers. Cancer research, 1999. 59(2): p. 320-4. 421. Kinzler, K.W. and B. Vogelstein, Lessons from hereditary colorectal cancer. Cell, 1996. 87(2): p. 159-70. 422. Deramaudt, T. and A.K. Rustgi, Mutant KRAS in the initiation of pancreatic cancer. Biochim Biophys Acta, 2005. 1756(2): p. 97-101. 423. Viale, G., The current state of breast cancer classification. Ann Oncol, 2012. 23 Suppl 10: p. x207-10. 424. Perou, C.M., et al., Molecular portraits of human breast tumours. Nature, 2000. 406(6797): p. 747-52. 425. Sorlie, T., et al., Gene expression patterns of breast carcinomas distinguish tumor subclasses with clinical implications. Proc Natl Acad Sci U S A, 2001. 98(19): p. 10869- 74. 426. Metzger-Filho, O., et al., Dissecting the heterogeneity of triple-negative breast cancer. J Clin Oncol, 2012. 30(15): p. 1879-87. 427. Blenkiron, C., et al., MicroRNA expression profiling of human breast cancer identifies new markers of tumor subtype. Genome Biol, 2007. 8(10): p. R214. 428. Volinia, S., et al., Breast cancer signatures for invasiveness and prognosis defined by deep sequencing of microRNA. Proc Natl Acad Sci U S A, 2012. 109(8): p. 3024-9. 429. Chan, M., et al., Identification of circulating microRNA signatures for breast cancer detection. Clin Cancer Res, 2013. 19(16): p. 4477-87. 430. Lowery, A.J., et al., MicroRNA signatures predict oestrogen receptor, progesterone receptor and HER2/neu receptor status in breast cancer. Breast Cancer Res, 2009. 11(3): p. R27. 431. Cascione, L., et al., Integrated microRNA and mRNA signatures associated with survival in triple negative breast cancer. PLoS One, 2013. 8(2): p. e55910. 432. Roth, C., et al., Circulating microRNAs as blood-based markers for patients with primary and metastatic breast cancer. Breast Cancer Res, 2010. 12(6): p. R90. 433. Zhao, H., et al., A pilot study of circulating miRNAs as potential biomarkers of early stage breast cancer. PLoS One, 2010. 5(10): p. e13735. 434. Heneghan, H.M., et al., Circulating microRNAs as novel minimally invasive biomarkers for breast cancer. Ann Surg, 2010. 251(3): p. 499-505. 435. Sun, Y., et al., Serum microRNA-155 as a potential biomarker to track disease in breast cancer. PLoS One, 2012. 7(10): p. e47003. 436. Dudda, J.C., et al., MicroRNA-155 is required for effector CD8+ T cell responses to virus infection and cancer. Immunity, 2013. 38(4): p. 742-53. 437. Tili, E., C.M. Croce, and J.J. Michaille, miR-155: on the crosstalk between inflammation and cancer. Int Rev Immunol, 2009. 28(5): p. 264-84. 438. Wang, P.Y., et al., Higher expression of circulating miR-182 as a novel biomarker for breast cancer. Oncol Lett, 2013. 6(6): p. 1681-1686. 439. Slamon, D.J., et al., Human breast cancer: correlation of relapse and survival with amplification of the HER-2/neu oncogene. Science, 1987. 235(4785): p. 177-82. 440. Slamon, D.J., et al., Use of chemotherapy plus a monoclonal antibody against HER2 for metastatic breast cancer that overexpresses HER2. N Engl J Med, 2001. 344(11): p. 783- 92.

162 441. Balmana, J., et al., BRCA in breast cancer: ESMO clinical recommendations. Ann Oncol, 2009. 20 Suppl 4: p. 19-20. 442. Milanezi, F., S. Carvalho, and F.C. Schmitt, EGFR/HER2 in breast cancer: a biological approach for molecular diagnosis and therapy. Expert Rev Mol Diagn, 2008. 8(4): p. 417-34. 443. Bhola, N.E., et al., TGF-beta inhibition enhances chemotherapy action against triple- negative breast cancer. J Clin Invest, 2013. 123(3): p. 1348-58. 444. Bilir, B., O. Kucuk, and C.S. Moreno, Wnt signaling blockage inhibits cell proliferation and migration, and induces apoptosis in triple-negative breast cancer cells. J Transl Med, 2013. 11: p. 280. 445. van Rooij, E., A.L. Purcell, and A.A. Levin, Developing microRNA therapeutics. Circ Res, 2012. 110(3): p. 496-507. 446. Prakash, T.P. and B. Bhat, 2'-Modified oligonucleotides for antisense therapeutics. Curr Top Med Chem, 2007. 7(7): p. 641-9. 447. Grimm, D., et al., Fatality in mice due to oversaturation of cellular microRNA/short hairpin RNA pathways. Nature, 2006. 441(7092): p. 537-41. 448. Helwak, A., et al., Mapping the human miRNA interactome by CLASH reveals frequent noncanonical binding. Cell, 2013. 153(3): p. 654-65. 449. Helwak, A. and D. Tollervey, Mapping the miRNA interactome by cross-linking ligation and sequencing of hybrids (CLASH). Nat Protoc, 2014. 9(3): p. 711-28. 450. Au, S.L., et al., Enhancer of zeste homolog 2 epigenetically silences multiple tumor suppressor microRNAs to promote liver cancer metastasis. Hepatology, 2012. 56(2): p. 622-31. 451. Bao, W., et al., HER2 interacts with CD44 to up-regulate CXCR4 via epigenetic silencing of microRNA-139 in gastric cancer cells. Gastroenterology, 2011. 141(6): p. 2076-2087 e6. 452. Bushati, N. and S.M. Cohen, microRNA functions. Annu Rev Cell Dev Biol, 2007. 23: p. 175-205. 453. Herranz, H. and S.M. Cohen, MicroRNAs and gene regulatory networks: managing the impact of noise in biological systems. Genes Dev, 2010. 24(13): p. 1339-44. 454. Schultz, J., et al., MicroRNA let-7b targets important cell cycle molecules in malignant melanoma cells and interferes with anchorage-independent growth. Cell Res, 2008. 18(5): p. 549-57. 455. Qin, X., et al., MicroRNA-19a mediates the suppressive effect of laminar flow on cyclin D1 expression in human umbilical vein endothelial cells. Proc Natl Acad Sci U S A, 2010. 107(7): p. 3240-4. 456. Wang, P., et al., microRNA-21 negatively regulates Cdc25A and cell cycle progression in colon cancer cells. Cancer research, 2009. 69(20): p. 8157-65. 457. Lal, A., et al., p16(INK4a) translation suppressed by miR-24. PLoS One, 2008. 3(3): p. e1864. 458. Kim, Y.K., et al., Functional links between clustered microRNAs: suppression of cell- cycle inhibitors by microRNA clusters in gastric cancer. Nucleic Acids Res, 2009. 37(5): p. 1672-81. 459. Kota, J., et al., Therapeutic microRNA delivery suppresses tumorigenesis in a murine liver cancer model. Cell, 2009. 137(6): p. 1005-17. 460. Malhas, A., N.J. Saunders, and D.J. Vaux, The nuclear envelope can control gene expression and cell cycle progression via miRNA regulation. Cell Cycle, 2010. 9(3): p. 531-9. 461. Sun, F., et al., Downregulation of CCND1 and CDK6 by miR-34a induces cell cycle arrest. FEBS Lett, 2008. 582(10): p. 1564-8.

163 462. Shi, W., et al., Significance of Plk1 regulation by miR-100 in human nasopharyngeal cancer. Int J Cancer, 2010. 126(9): p. 2036-48. 463. Pierson, J., et al., Regulation of cyclin dependent kinase 6 by microRNA 124 in medulloblastoma. J Neurooncol, 2008. 90(1): p. 1-7. 464. Shi, L., et al., MiR-125b is critical for the suppression of human U251 glioma stem cell proliferation. Brain Res, 2010. 1312: p. 120-6. 465. Huang, L., et al., MicroRNA-125b suppresses the development of bladder cancer by targeting E2F3. Int J Cancer, 2011. 128(8): p. 1758-69. 466. Butz, H., et al., Down-regulation of Wee1 kinase by a specific subset of microRNA in human sporadic pituitary adenomas. J Clin Endocrinol Metab, 2010. 95(10): p. E181-91. 467. Wu, J., et al., miR-129 regulates cell proliferation by downregulating Cdk6 expression. Cell Cycle, 2010. 9(9): p. 1809-18. 468. Wu, S., et al., Multiple microRNAs modulate p21Cip1/Waf1 expression by directly targeting its 3' untranslated region. Oncogene, 2010. 29(15): p. 2302-8. 469. Sachdeva, M. and Y.Y. Mo, p53 and c-myc: how does the cell balance "yin" and "yang"? Cell Cycle, 2009. 8(9): p. 1303. 470. Wang, X., et al., MicroRNAs181 regulate the expression of p27Kip1 in human myeloid leukemia cells induced to differentiate by 1,25-dihydroxyvitamin D3. Cell Cycle, 2009. 8(5): p. 736-41. 471. Giannakakis, A., et al., miR-210 links hypoxia with cell cycle regulation and is deleted in human epithelial ovarian cancer. Cancer Biol Ther, 2008. 7(2): p. 255-64. 472. Guo, X., et al., miRNA-331-3p directly targets E2F1 and induces growth arrest in human gastric cancer. Biochem Biophys Res Commun, 2010. 398(1): p. 1-6. 473. Sarkar, S., B.K. Dey, and A. Dutta, MiR-322/424 and -503 are induced during muscle differentiation and promote cell cycle quiescence and differentiation by down-regulation of Cdc25A. Mol Biol Cell, 2010. 21(13): p. 2138-49. 474. Yang, X., et al., miR-449a and miR-449b are direct transcriptional targets of E2F1 and negatively regulate pRb-E2F1 activity through a feedback loop by targeting CDK6 and CDC25A. Genes Dev, 2009. 23(20): p. 2388-93. 475. Xia, W., et al., MicroRNA-200b regulates cyclin D1 expression and promotes S-phase entry by targeting RND3 in HeLa cells. Molecular and cellular biochemistry, 2010. 344(1-2): p. 261-6. 476. Majid, S., et al., Regulation of minichromosome maintenance gene family by microRNA- 1296 and genistein in prostate cancer. Cancer Res, 2010. 70(7): p. 2809-18. 477. Wang, J., et al., MiR-365b-3p, down-regulated in retinoblastoma, regulates cell cycle progression and apoptosis of human retinoblastoma cells by targeting PAX6. FEBS Lett, 2013. 478. Shen, W.W., et al., MiR-142-3p functions as a tumor suppressor by targeting CD133, ABCG2, and Lgr5 in colon cancer cells. J Mol Med (Berl), 2013. 479. Yao, C.X., et al., miR-200b targets GATA-4 during cell growth and differentiation. RNA Biol, 2013. 10(4). 480. Furuta, M., et al., The tumor-suppressive miR-497-195 cluster targets multiple cell-cycle regulators in hepatocellular carcinoma. PLoS One, 2013. 8(3): p. e60155. 481. Zhu, G., et al., Downregulated microRNA-32 expression induced by high glucose inhibits cell cycle progression via PTEN upregulation and Akt inactivation in bone marrow- derived mesenchymal stem cells. Biochemical and biophysical research communications, 2013. 433(4): p. 526-31.

164

APPENDIX

1. APPENDIX

Supplementary Material for Chapter 2

Supplementary Table 1: Significantly enriched probes, predicted to be miR-182 targets

Probe_ID Gene logFC Avg. adj.P. ILMN_1783985 COQ6 -1.93 9.88 0.00 Symbol Exp Value ILMN_1711766 SKP1A -1.93 11.33 0.00 ILMN_1714083 KLHL8 -3.25 9.66 0.00 ILMN_1679520 AGPAT1 -1.91 9.26 0.00 ILMN_1711853 MED24 -2.81 10.59 0.00 ILMN_1669940 TMEM38B -1.89 9.89 0.00 ILMN_1772455 HDAC3 -2.81 11.60 0.00 ILMN_1766435 WBP11 -1.87 12.39 0.00 ILMN_2252136 YWHAE -2.78 9.95 0.01 ILMN_2091792 ENTPD6 -1.87 9.52 0.00 ILMN_2358457 ATF4 -2.72 10.49 0.00 ILMN_1738424 CDC42 -1.87 10.44 0.00 ILMN_2189222 KLHL8 -2.71 10.47 0.00 ILMN_2396813 C19orf62 -1.86 9.51 0.00 ILMN_1805800 RAB5A -2.61 10.03 0.00 ILMN_1812441 C17orf63 -1.86 9.70 0.00 ILMN_1746171 H2AFY -2.60 11.25 0.00 ILMN_1655165 RNF138 -1.85 10.44 0.00 ILMN_1686254 FAM127B -2.55 10.69 0.00 ILMN_1744442 TTPAL -1.85 8.93 0.00 ILMN_2388272 MED24 -2.48 10.32 0.00 ILMN_1700546 ELOVL6 -1.85 8.79 0.00 ILMN_1809496 COPG2 -2.43 9.93 0.00 ILMN_1719471 MSH3 -1.85 10.07 0.00 ILMN_1723871 OTUB1 -2.31 9.63 0.00 ILMN_1813246 LOC728554 -1.85 9.15 0.00 ILMN_1685625 UCP2 -2.30 9.20 0.00 ILMN_2399896 SEC31A -1.84 10.69 0.00 ILMN_1700690 VAT1 -2.30 9.80 0.00 ILMN_2370208 CMTM3 -1.84 8.66 0.00 ILMN_2349006 USP21 -2.26 9.02 0.00 ILMN_1742981 TUBA1A -1.83 12.61 0.00 ILMN_1658911 LOC647349 -2.25 9.88 0.00 ILMN_2352563 CLDND1 -1.83 10.80 0.00 ILMN_1718354 INTS7 -2.25 9.15 0.00 ILMN_1772302 MTHFS -1.83 10.78 0.00 ILMN_1747460 TMEM184B -2.24 9.63 0.00 ILMN_1791332 ATP5O -1.83 13.53 0.00 ILMN_1696021 KPNA6 -2.23 10.40 0.00 ILMN_2222008 KIFC1 -1.83 10.98 0.00 ILMN_1743939 C5orf24 -2.22 8.36 0.00 ILMN_1793616 RNF38 -1.82 9.44 0.00 ILMN_2396672 ABLIM1 -2.20 8.85 0.00 ILMN_1753002 RAB2B -1.81 10.87 0.00 ILMN_1752947 C17orf79 -2.19 10.19 0.00 ILMN_1785424 ABLIM1 -1.81 10.92 0.00 ILMN_1670931 PDS5A -2.18 9.76 0.00 ILMN_2283388 C20orf24 -1.79 11.96 0.00 ILMN_1783806 DTNBP1 -2.16 8.57 0.00 ILMN_1791002 SKP2 -1.79 9.70 0.00 ILMN_2314140 PAX6 -2.14 8.67 0.00 ILMN_1704793 LOC339344 -1.79 10.85 0.00 ILMN_1859946 n/a -2.13 10.44 0.00 ILMN_1654289 ELK1 -1.79 10.37 0.00 ILMN_1707336 ARPC4 -2.10 9.42 0.00 ILMN_1678919 YOD1 -1.79 10.17 0.00 ILMN_1758087 TAOK1 -2.09 8.85 0.00 ILMN_1790951 C19orf50 -1.78 10.31 0.00 ILMN_1807873 SNX6 -2.08 9.71 0.00 ILMN_2394250 PLEKHA1 -1.78 9.82 0.00 ILMN_2364131 TTPAL -2.08 9.74 0.00 ILMN_1690963 DDEF1 -1.78 10.33 0.00 ILMN_1699496 PHF21A -2.08 10.14 0.00 ILMN_1793360 APITD1 -1.78 10.55 0.00 ILMN_1751444 NCAPG -2.06 11.25 0.00 ILMN_1808587 ZFHX3 -1.78 8.73 0.00 ILMN_1743034 KIF1B -2.05 8.65 0.00 ILMN_1785635 BRD3 -1.77 10.19 0.00 ILMN_1675612 BLCAP -2.05 9.35 0.00 ILMN_1719039 UBE2G1 -1.77 9.81 0.00 ILMN_1792672 POLR2D -2.05 10.33 0.00 ILMN_2373779 COPS8 -1.76 9.03 0.00 ILMN_2415926 THOC3 -2.04 10.75 0.00 ILMN_1667670 SLC25A15 -1.76 9.54 0.00 ILMN_2351638 BEX4 -2.03 11.30 0.00 ILMN_2181125 NAPB -1.76 9.20 0.00 ILMN_1780887 USP21 -2.02 9.18 0.00 ILMN_1680782 PATL1 -1.75 10.32 0.00 ILMN_1683658 FKBP1A -2.01 11.28 0.00 ILMN_2100689 MAP2K4 -1.75 9.48 0.00 ILMN_2311537 HMGA1 -2.01 11.44 0.00 ILMN_1911042 n/a -1.74 11.29 0.00 ILMN_1728083 EIF4EBP2 -2.01 9.74 0.00 ILMN_1706553 SMG7 -1.74 9.97 0.00 ILMN_1694799 PIAS2 -2.01 9.03 0.00 ILMN_1720526 CENPN -1.74 11.18 0.00 ILMN_1782403 PRR11 -2.00 8.52 0.00 ILMN_2207393 CNOT3 -1.73 9.04 0.00 ILMN_2109197 EPB41L3 -1.99 10.24 0.00 ILMN_2371685 UBE2E1 -1.73 8.35 0.00 ILMN_2074258 BARD1 -1.98 10.01 0.00 ILMN_1691942 CCNI -1.73 13.30 0.00 ILMN_1728845 SMARCD1 -1.97 11.82 0.00 ILMN_1722066 ARMC1 -1.73 9.84 0.00 ILMN_1651819 GALNT11 -1.97 9.99 0.00 ILMN_1711189 EXOSC10 -1.73 10.13 0.00 ILMN_1723007 ZCCHC9 -1.97 11.17 0.00 ILMN_1688322 ADIPOR1 -1.73 10.23 0.00 ILMN_1706990 ZNF271 -1.96 10.13 0.00 ILMN_1792712 LOC201725 -1.72 9.52 0.00 ILMN_2311761 AP3S1 -1.96 9.84 0.00 ILMN_1798496 HOXB8 -1.72 10.97 0.00 ILMN_2328433 NOL1 -1.95 9.01 0.00 ILMN_2409793 MAZ -1.72 8.24 0.00 ILMN_1678235 KIAA1267 -1.95 9.43 0.00 ILMN_1673185 CPSF2 -1.72 9.41 0.00 ILMN_1762835 HELZ -1.95 9.23 0.00 ILMN_1710326 CLDND1 -1.72 10.71 0.00 ILMN_1688158 CYB5R4 -1.94 9.09 0.00 ILMN_1758311 NET1 -1.72 11.41 0.00 ILMN_1760360 ZNF294 -1.94 9.58 0.00 ILMN_1771019 MTMR4 -1.72 10.54 0.00 ILMN_2385097 NDRG3 -1.94 8.95 0.00 ILMN_1734833 NBN -1.71 9.94 0.00 ILMN_1722774 VPS72 -1.93 9.75 0.00 ILMN_1656682 AZIN1 -1.71 11.12 0.00

167 ILMN_1807243 PRPF18 -1.71 9.01 0.00 ILMN_1813834 PRMT6 -1.58 9.51 0.00 ILMN_1714622 TNRC6A -1.70 8.84 0.00 ILMN_1757467 H1F0 -1.58 11.45 0.01 ILMN_1700024 UST -1.70 8.63 0.00 ILMN_2415235 CSNK1E -1.57 12.82 0.00 ILMN_2374244 DYRK2 -1.70 9.42 0.00 ILMN_1666004 WASL -1.57 10.11 0.00 ILMN_1732772 PPME1 -1.70 9.42 0.00 ILMN_1682180 VCPIP1 -1.57 8.52 0.00 ILMN_1727809 STK35 -1.69 8.46 0.00 ILMN_2374865 ATF3 -1.57 9.49 0.00 ILMN_1721116 USP10 -1.69 9.58 0.00 ILMN_1763260 HIF1A -1.57 8.28 0.00 ILMN_1720965 TULP4 -1.69 8.97 0.00 ILMN_2072603 MRPL14 -1.57 12.25 0.00 ILMN_1724148 ORAI1 -1.69 8.22 0.00 ILMN_1806266 RAP1GDS1 -1.57 8.95 0.00 ILMN_1760143 ADRM1 -1.69 11.55 0.00 ILMN_2106902 CHES1 -1.57 9.48 0.00 ILMN_1781360 MPHOSPH6 -1.69 9.36 0.00 ILMN_2070896 BMPR2 -1.56 9.03 0.00 ILMN_1697614 NHP2L1 -1.68 10.12 0.00 ILMN_2379130 IRAK1 -1.56 11.72 0.00 ILMN_2333367 FKBP1A -1.67 9.33 0.00 ILMN_1667561 IFRD1 -1.56 9.10 0.00 ILMN_1744308 DHX33 -1.66 9.71 0.00 ILMN_1682264 WDR68 -1.56 10.87 0.00 ILMN_2090397 ISG20L2 -1.66 10.36 0.00 ILMN_1697597 KIAA0494 -1.55 9.29 0.00 ILMN_1724293 KDELR2 -1.66 9.98 0.00 ILMN_1671257 DKC1 -1.55 12.31 0.00 ILMN_2155172 BXDC2 -1.66 11.05 0.00 ILMN_1679800 BXDC2 -1.55 12.43 0.00 ILMN_1764415 ZNF585A -1.66 8.17 0.00 ILMN_2320964 ADAR -1.55 10.19 0.00 ILMN_2077886 C1orf109 -1.66 9.62 0.00 ILMN_1743397 PIGW -1.55 9.42 0.00 ILMN_1683609 UBE1 -1.65 11.97 0.00 ILMN_1745415 BBX -1.55 10.03 0.00 ILMN_1745813 KIAA1279 -1.65 10.80 0.00 ILMN_1680987 HAND1 -1.55 12.98 0.00 ILMN_1654262 ZMAT3 -1.65 8.29 0.00 ILMN_1789999 SLC30A7 -1.55 9.61 0.00 ILMN_2124951 RBMX -1.65 11.50 0.00 ILMN_1753885 YTHDF1 -1.55 10.65 0.00 ILMN_1770425 CDIPT -1.65 8.37 0.00 ILMN_2055760 KIAA1715 -1.55 9.20 0.00 ILMN_1655608 KLHL18 -1.64 8.39 0.00 ILMN_2393763 ARPC4 -1.54 9.23 0.00 ILMN_1810488 NFYC -1.64 9.95 0.00 ILMN_2391551 C13orf23 -1.54 10.03 0.00 ILMN_1727996 BAG4 -1.64 8.24 0.01 ILMN_1669722 LRRC61 -1.54 8.07 0.00 ILMN_1755077 HEBP2 -1.64 11.78 0.00 ILMN_1788203 HEY1 -1.54 11.59 0.00 ILMN_1778917 CDK7 -1.64 9.27 0.00 ILMN_1800975 PSME3 -1.54 11.27 0.00 ILMN_1715416 NUP188 -1.64 9.96 0.00 ILMN_1688639 FBXL2 -1.54 8.55 0.00 ILMN_1699695 TNFRSF21 -1.64 10.79 0.00 ILMN_1743208 NEDD1 -1.54 7.89 0.00 ILMN_1700044 SAP130 -1.63 9.85 0.00 ILMN_1685763 WDR45L -1.54 9.47 0.00 ILMN_1917341 n/a -1.63 9.12 0.00 ILMN_2403458 SMARCB1 -1.53 9.62 0.00 ILMN_1752582 RAB5B -1.63 10.86 0.00 ILMN_2322552 NCKAP1 -1.53 9.25 0.00 ILMN_1790953 TBCB -1.63 11.01 0.00 ILMN_2321485 PPP1R8 -1.53 8.73 0.00 ILMN_1673369 SEPHS1 -1.63 9.74 0.00 ILMN_1745686 MFHAS1 -1.53 8.10 0.00 ILMN_1724497 ABI2 -1.63 9.30 0.00 ILMN_1786328 WDR40A -1.53 9.80 0.00 ILMN_1730433 CD2AP -1.62 8.40 0.00 ILMN_1782579 IMMT -1.53 11.09 0.00 ILMN_1801928 YWHAZ -1.62 13.13 0.00 ILMN_1754304 C6orf151 -1.53 8.83 0.00 ILMN_1810514 SLC25A44 -1.62 10.06 0.00 ILMN_1682938 ARF3 -1.53 9.65 0.00 ILMN_1728024 TUBG1 -1.62 10.10 0.00 ILMN_1657754 RPAP2 -1.53 9.21 0.00 ILMN_1712687 PAK2 -1.62 9.60 0.00 ILMN_1806328 PDIK1L -1.53 9.17 0.00 ILMN_1804798 BEXL1 -1.61 12.52 0.00 ILMN_1694259 TINP1 -1.52 11.97 0.00 ILMN_1704342 UBE3C -1.61 10.16 0.00 ILMN_1763694 RSPRY1 -1.52 10.30 0.00 ILMN_1752591 LEPROTL1 -1.61 10.53 0.00 ILMN_1658847 MGC61598 -1.51 9.83 0.00 ILMN_1668507 DDAH1 -1.61 9.63 0.00 ILMN_1783676 CCDC15 -1.51 8.28 0.00 ILMN_1693108 RUVBL1 -1.61 10.74 0.00 ILMN_1685574 TSC22D2 -1.51 8.81 0.00 ILMN_1689817 LCOR -1.61 9.24 0.00 ILMN_1757636 C5orf35 -1.51 8.98 0.00 ILMN_1738229 NDRG3 -1.61 10.41 0.00 ILMN_1656540 RUVBL1 -1.51 9.88 0.00 ILMN_1775744 MRPS16 -1.61 10.47 0.00 ILMN_1669046 FOXQ1 -1.51 8.67 0.00 ILMN_1723185 ELOF1 -1.61 8.92 0.00 ILMN_1703487 LMO4 -1.50 10.30 0.00 ILMN_1737426 PCMTD1 -1.61 8.51 0.00 ILMN_1781580 BRI3 -1.50 10.66 0.00 ILMN_1706502 EIF2AK2 -1.61 10.83 0.00 ILMN_1657139 ADAT1 -1.50 8.61 0.00 ILMN_1665423 ZFP91 -1.60 11.38 0.00 ILMN_1673069 DPP9 -1.50 9.17 0.00 ILMN_1671742 UPF3A -1.60 9.35 0.00 ILMN_1738491 SNX30 -1.50 8.76 0.00 ILMN_2173451 GPI -1.60 10.95 0.00 ILMN_1659895 MSN -1.50 10.14 0.00 ILMN_1783023 LOC285636 -1.60 10.88 0.00 ILMN_1779374 AMMECR1 -1.50 10.00 0.00 ILMN_1912737 n/a -1.60 8.40 0.00 ILMN_1761531 SGPL1 -1.50 8.39 0.00 ILMN_1708858 CSNK1E -1.60 11.10 0.00 ILMN_1708414 GNL3L -1.50 10.63 0.00 ILMN_2089340 FAM127B -1.60 8.04 0.00 ILMN_1789510 STIP1 -1.49 11.82 0.00 ILMN_1815733 EIF5 -1.59 10.42 0.00 ILMN_1774028 MTFR1 -1.49 10.14 0.00 ILMN_1815158 GPS2 -1.59 9.95 0.00 ILMN_2394242 AMMECR1 -1.49 8.60 0.00 ILMN_1790987 HUWE1 -1.59 8.96 0.00 ILMN_1722718 BMP2 -1.49 8.92 0.00 ILMN_1793671 TFDP2 -1.58 7.94 0.00 ILMN_1687782 RAD17 -1.49 8.06 0.00 ILMN_2359345 NET1 -1.58 8.01 0.00 ILMN_2399627 AP1G1 -1.49 8.59 0.00

168 ILMN_1725612 NUP50 -1.49 9.40 0.00 ILMN_1736510 FOXN2 -1.41 9.47 0.00 ILMN_2187487 HEATR5B -1.49 9.38 0.00 ILMN_1795435 ZNF264 -1.40 8.31 0.00 ILMN_2325008 DHX40 -1.48 9.05 0.00 ILMN_1663486 TLK2 -1.40 9.67 0.00 ILMN_1729976 ZNF828 -1.48 10.78 0.00 ILMN_1661491 SH3GL2 -1.40 9.18 0.00 ILMN_1911605 n/a -1.48 8.90 0.00 ILMN_2367020 SEC61G -1.40 13.36 0.00 ILMN_1722811 CDKN1B -1.48 9.15 0.00 ILMN_1773066 CDKN2AIP -1.40 8.71 0.00 ILMN_1701131 C2orf49 -1.48 9.32 0.00 ILMN_1684051 WASF2 -1.39 8.84 0.00 ILMN_1669394 EI24 -1.48 11.93 0.00 ILMN_2312719 EXOSC9 -1.39 9.47 0.00 ILMN_2330371 TATDN3 -1.48 9.10 0.00 ILMN_1837935 TNPO1 -1.39 9.38 0.00 ILMN_1771884 ZNF45 -1.48 8.44 0.00 ILMN_1794588 DYRK2 -1.39 8.13 0.00 ILMN_1673024 RBM15B -1.47 8.07 0.00 ILMN_1769665 RAB5C -1.38 10.11 0.00 ILMN_1713129 TTF1 -1.47 8.96 0.00 ILMN_2374687 PTPN13 -1.38 8.60 0.00 ILMN_1720270 CDR2 -1.47 9.87 0.00 ILMN_1800837 CFDP1 -1.38 9.71 0.00 ILMN_1654441 OAT -1.47 11.12 0.01 ILMN_1801215 PDDC1 -1.38 8.36 0.00 ILMN_1727444 C16orf53 -1.47 9.84 0.00 ILMN_1679655 WDR82 -1.37 10.57 0.00 ILMN_1651705 CAT -1.47 8.94 0.00 ILMN_1784602 CDKN1A -1.37 9.04 0.00 ILMN_2129388 KIAA1128 -1.47 8.59 0.00 ILMN_2374683 PTPN13 -1.37 8.71 0.00 ILMN_1726359 NECAP1 -1.46 10.55 0.00 ILMN_1765574 TFAP2A -1.37 11.06 0.01 ILMN_2374293 DYRK1A -1.46 9.15 0.00 ILMN_1753440 C4orf30 -1.37 9.33 0.00 ILMN_1753345 SCAMP5 -1.46 8.56 0.00 ILMN_1802894 VKORC1L1 -1.37 9.34 0.00 ILMN_1761797 CSTB -1.46 12.74 0.00 ILMN_1682316 TRIM33 -1.37 10.79 0.00 ILMN_2321634 RAD17 -1.46 8.04 0.00 ILMN_1738173 METTL4 -1.36 8.06 0.00 ILMN_1750167 PRR3 -1.46 8.45 0.00 ILMN_2366177 IFT122 -1.35 7.89 0.00 ILMN_2181968 CBL -1.45 8.57 0.00 ILMN_1701269 ATP5C1 -1.35 12.44 0.00 ILMN_1674302 PPAT -1.45 10.50 0.00 ILMN_2311089 BRCA1 -1.35 9.57 0.00 ILMN_1688404 ZMYM4 -1.45 10.26 0.00 ILMN_1723117 IPO9 -1.35 9.32 0.00 ILMN_1655497 EIF4B -1.45 12.26 0.00 ILMN_1732343 KIAA0999 -1.35 9.00 0.00 ILMN_1667432 HYAL3 -1.45 8.23 0.00 ILMN_2106167 RAP1GDS1 -1.35 8.75 0.00 ILMN_1710676 FBXO5 -1.45 10.49 0.00 ILMN_1665117 C6orf89 -1.35 7.83 0.00 ILMN_1776260 LOC653505 -1.45 12.29 0.00 ILMN_1680955 AURKA -1.35 11.64 0.00 ILMN_1771962 GLI3 -1.45 7.97 0.00 ILMN_1815169 MCM5 -1.35 10.40 0.00 ILMN_1767662 LASS6 -1.44 11.11 0.00 ILMN_1703946 ADORA2B -1.35 10.27 0.00 ILMN_1812552 PHCA -1.44 8.63 0.00 ILMN_1656427 C17orf39 -1.35 8.71 0.00 ILMN_1757272 THAP1 -1.44 8.48 0.00 ILMN_1803256 STOX2 -1.35 9.78 0.00 ILMN_1807042 MARCKS -1.44 12.60 0.00 ILMN_2404539 C20orf30 -1.35 11.97 0.00 ILMN_1790100 C11orf82 -1.44 9.70 0.00 ILMN_2403006 TJP1 -1.34 8.88 0.00 ILMN_1759097 MLLT11 -1.44 11.49 0.00 ILMN_1672940 ZNF562 -1.34 8.90 0.00 ILMN_1658472 APH1A -1.44 10.97 0.00 ILMN_2346573 PSME3 -1.34 11.30 0.00 ILMN_1741422 FUT8 -1.44 8.69 0.00 ILMN_1813236 C6orf136 -1.34 9.59 0.00 ILMN_1675387 LIMS1 -1.44 8.54 0.00 ILMN_1662166 PTK7 -1.34 8.43 0.00 ILMN_1716552 ENAH -1.43 9.62 0.00 ILMN_2330267 ABCE1 -1.34 11.60 0.00 ILMN_1659953 3-Sep -1.43 8.61 0.00 ILMN_1692684 SMARCD2 -1.34 9.06 0.00 ILMN_1738632 PRKAR1A -1.43 11.35 0.00 ILMN_1676846 ABCE1 -1.34 11.54 0.00 ILMN_1759008 ZNF689 -1.43 11.09 0.00 ILMN_1696383 POP4 -1.34 9.13 0.00 ILMN_1751898 C12orf4 -1.43 9.16 0.00 ILMN_1770623 FAM58A -1.33 10.07 0.00 ILMN_2414399 NME1 -1.43 10.64 0.01 ILMN_1701558 MAP1A -1.33 8.54 0.00 ILMN_1668270 ZDHHC18 -1.43 8.68 0.00 ILMN_2096322 ADIPOR1 -1.33 9.78 0.00 ILMN_1742167 TUBA1C -1.42 14.30 0.00 ILMN_1656676 ZYG11B -1.33 11.17 0.00 ILMN_1671326 EIF2C1 -1.42 8.95 0.00 ILMN_2324421 TXNRD1 -1.33 8.50 0.01 ILMN_2337941 COPS8 -1.42 11.01 0.00 ILMN_1902658 n/a -1.32 8.86 0.00 ILMN_1771964 GSTA4 -1.42 9.77 0.00 ILMN_1775901 PHF5A -1.32 9.45 0.00 ILMN_2243553 ZNF275 -1.42 9.40 0.00 ILMN_2057220 HRSP12 -1.32 8.78 0.01 ILMN_1674282 PPARD -1.42 7.94 0.00 ILMN_1731184 MELK -1.32 9.62 0.00 ILMN_1705442 CMTM3 -1.42 10.51 0.00 ILMN_1705871 DDHD2 -1.32 9.59 0.00 ILMN_1729175 FBXO3 -1.42 8.72 0.00 ILMN_1687440 HIPK2 -1.32 9.09 0.00 ILMN_1796962 PPP3R1 -1.42 10.47 0.00 ILMN_1740216 ERCC3 -1.32 9.79 0.00 ILMN_1795128 C13orf23 -1.42 9.74 0.00 ILMN_2365544 NOLA2 -1.32 11.39 0.00 ILMN_1697503 DHX29 -1.42 9.24 0.00 ILMN_1658624 UBXN6 -1.32 9.01 0.00 ILMN_1651936 SETD8 -1.41 9.24 0.00 ILMN_1689123 CCNK -1.32 10.44 0.00 ILMN_1758906 GNA13 -1.41 10.39 0.00 ILMN_2156267 EIF2AK1 -1.32 11.98 0.00 ILMN_1705515 UPF3A -1.41 9.48 0.00 ILMN_1809285 DCP1A -1.32 8.69 0.00 ILMN_1670383 PCDH7 -1.41 8.31 0.00 ILMN_1652580 POLD1 -1.32 9.43 0.00 ILMN_1695509 PTPN12 -1.41 8.46 0.00 ILMN_1724811 PARN -1.31 9.26 0.00 ILMN_1659415 MAP2K1IP1 -1.41 10.20 0.00 ILMN_1755235 XPO6 -1.31 10.27 0.00 ILMN_1672504 PDXK -1.41 10.78 0.00 ILMN_1709439 CHMP1A -1.31 9.83 0.00

169 ILMN_1695334 PYGO2 -1.31 8.03 0.00 ILMN_1790577 SLC35F2 -1.25 9.23 0.00 ILMN_2331636 ACACA -1.31 11.07 0.00 ILMN_1780598 PIAS1 -1.25 9.43 0.00 ILMN_1732705 HCFC1 -1.31 12.60 0.00 ILMN_1700231 IHPK1 -1.25 10.15 0.00 ILMN_1762615 FAM175B -1.31 10.36 0.00 ILMN_2400500 LASS2 -1.25 9.44 0.00 ILMN_2062381 LCOR -1.31 9.86 0.00 ILMN_1716080 CBL -1.25 8.24 0.00 ILMN_1812080 PARP16 -1.31 8.21 0.00 ILMN_1806312 C20orf30 -1.25 12.31 0.00 ILMN_1782504 MTERFD1 -1.31 10.11 0.00 ILMN_1808196 GSTO1 -1.25 12.04 0.00 ILMN_1762764 SH3BGRL2 -1.31 9.39 0.00 ILMN_1754912 GLE1 -1.25 10.64 0.00 ILMN_1803254 KIAA2010 -1.31 9.72 0.00 ILMN_1724493 LYSMD2 -1.25 8.88 0.00 ILMN_1754121 CSK -1.31 12.03 0.00 ILMN_1680703 MRPS15 -1.25 12.03 0.00 ILMN_2361695 BAG5 -1.31 10.74 0.00 ILMN_1741585 SFMBT1 -1.24 8.30 0.00 ILMN_1705985 PIGA -1.31 10.54 0.00 ILMN_1799104 SPAG9 -1.24 9.69 0.00 ILMN_1665538 SKP2 -1.30 12.43 0.00 ILMN_1761560 PHF13 -1.24 10.49 0.00 ILMN_2389013 ADRM1 -1.30 9.86 0.00 ILMN_1802669 PPP3CB -1.24 10.03 0.00 ILMN_1800889 FIG4 -1.30 9.06 0.00 ILMN_2196097 PPP2CA -1.24 11.87 0.00 ILMN_1708619 SEH1L -1.30 8.84 0.00 ILMN_1765204 ST13 -1.24 10.95 0.00 ILMN_1651254 LPP -1.30 8.97 0.00 ILMN_1782459 OSBPL8 -1.24 9.39 0.01 ILMN_1728517 FNTB -1.30 9.42 0.01 ILMN_1720422 G3BP2 -1.24 9.68 0.00 ILMN_1758750 EARS2 -1.29 9.14 0.00 ILMN_1786852 ZCCHC3 -1.24 8.78 0.00 ILMN_1741356 PRICKLE1 -1.29 9.01 0.00 ILMN_1806937 C19orf62 -1.24 11.45 0.00 ILMN_1786601 PLAGL2 -1.29 10.09 0.00 ILMN_1712320 DDX50 -1.24 10.39 0.00 ILMN_1717524 LIPT1 -1.29 8.60 0.00 ILMN_1742238 SET -1.24 11.72 0.00 ILMN_2357438 AURKA -1.29 11.45 0.00 ILMN_1724825 PCBP2 -1.23 12.43 0.00 ILMN_1659206 RARA -1.29 9.47 0.00 ILMN_1718265 ATG5 -1.23 8.79 0.00 ILMN_1672004 TOB1 -1.29 9.21 0.00 ILMN_1680104 SLC35C1 -1.23 8.09 0.00 ILMN_1654118 BCL2L1 -1.29 9.69 0.00 ILMN_1652244 POPDC3 -1.23 8.00 0.00 ILMN_1676075 KLHL12 -1.29 8.57 0.00 ILMN_2112402 PHF5A -1.23 11.82 0.00 ILMN_1683300 BAX -1.29 8.05 0.00 ILMN_2362122 AP3M1 -1.23 9.81 0.01 ILMN_1725485 RGS17 -1.29 8.75 0.00 ILMN_2249288 TIPRL -1.23 8.75 0.01 ILMN_1728984 PA2G4 -1.29 11.41 0.00 ILMN_1703408 FZD3 -1.23 8.62 0.00 ILMN_1676010 SP1 -1.29 9.38 0.00 ILMN_2093674 GMFB -1.23 11.75 0.00 ILMN_2164081 KLHL12 -1.28 10.34 0.00 ILMN_1813400 CBR4 -1.23 9.82 0.00 ILMN_1806818 MCM3 -1.28 11.15 0.01 ILMN_2144116 CPSF2 -1.22 9.46 0.01 ILMN_1702835 SH3BGRL -1.28 10.02 0.01 ILMN_1713491 VAMP2 -1.22 8.67 0.00 ILMN_2061950 RABGAP1 -1.28 10.09 0.00 ILMN_1736939 UGCG -1.22 8.68 0.00 ILMN_1779486 FAM126B -1.28 8.27 0.00 ILMN_2149766 APPBP2 -1.22 10.21 0.00 ILMN_1741003 ANXA5 -1.28 11.83 0.00 ILMN_2129015 AFF1 -1.22 8.01 0.00 ILMN_1762725 EIF3EIP -1.28 12.70 0.00 ILMN_1670353 RAD51AP1 -1.22 11.15 0.00 ILMN_2370296 ENAH -1.28 8.46 0.00 ILMN_1812856 ZSWIM1 -1.22 9.86 0.00 ILMN_1721713 EXOSC9 -1.28 10.22 0.00 ILMN_2362457 ZDHHC16 -1.22 9.99 0.00 ILMN_1771149 MRPL19 -1.28 9.21 0.00 ILMN_1666156 MORF4L2 -1.22 12.34 0.01 ILMN_1801923 ATF1 -1.28 8.08 0.00 ILMN_1708841 GOLPH3 -1.21 10.52 0.01 ILMN_2399622 AP1G1 -1.27 9.17 0.00 ILMN_2387599 C20orf24 -1.21 13.24 0.00 ILMN_2175075 SFRS4 -1.27 12.21 0.00 ILMN_2393254 CAPNS1 -1.21 9.53 0.00 ILMN_1815723 NUP35 -1.27 8.46 0.00 ILMN_1657153 ACTR3 -1.21 10.30 0.00 ILMN_1705032 SEH1L -1.27 8.21 0.00 ILMN_1809889 CCDC117 -1.21 11.00 0.00 ILMN_1718672 NOLA2 -1.27 11.69 0.00 ILMN_1771627 ZMIZ1 -1.21 9.93 0.00 ILMN_1708059 USP13 -1.27 8.84 0.00 ILMN_1674034 H2AFY -1.21 8.28 0.00 ILMN_1785926 ZNF621 -1.27 9.43 0.00 ILMN_2128293 APIP -1.21 9.39 0.00 ILMN_1704972 TRIM5 -1.27 8.14 0.00 ILMN_1683811 TNPO3 -1.20 9.68 0.00 ILMN_1766951 MSX2 -1.27 8.31 0.00 ILMN_1801348 GOT2 -1.20 12.61 0.00 ILMN_1718071 AGTPBP1 -1.27 9.33 0.00 ILMN_1710873 ZNF330 -1.20 10.72 0.00 ILMN_1679438 MLF1IP -1.27 8.90 0.00 ILMN_1736816 C13orf3 -1.20 8.90 0.00 ILMN_1662161 TBC1D13 -1.26 9.22 0.00 ILMN_1723158 NOL1 -1.20 12.03 0.00 ILMN_1772690 OGFOD1 -1.26 9.66 0.00 ILMN_2323979 WARS2 -1.20 8.82 0.00 ILMN_1811195 ZNF211 -1.26 8.29 0.00 ILMN_1696870 TGFBRAP1 -1.20 8.54 0.00 ILMN_2047112 RP11-529I10.4 -1.26 10.00 0.00 ILMN_1728514 BAG5 -1.20 8.70 0.00 ILMN_2343105 LIPT1 -1.26 8.66 0.00 ILMN_2261784 CCNY -1.20 9.63 0.00 ILMN_1788347 KIAA1737 -1.26 9.41 0.00 ILMN_1708502 AFF4 -1.20 8.62 0.00 ILMN_1691949 LOC728554 -1.26 13.13 0.00 ILMN_1772540 ATMIN -1.19 8.32 0.00 ILMN_1657857 TMEM14C -1.26 11.39 0.00 ILMN_1685824 B4GALT5 -1.19 10.50 0.00 ILMN_1775762 GNAI2 -1.26 9.51 0.00 ILMN_1773369 MRPL48 -1.19 10.04 0.00 ILMN_1761844 ZCCHC17 -1.25 10.71 0.00 ILMN_1734742 ARHGDIA -1.19 11.44 0.00 ILMN_1731714 CREB5 -1.25 9.12 0.00 ILMN_1810805 HEATR5B -1.19 8.66 0.00 ILMN_2170515 METTL11A -1.25 10.16 0.00 ILMN_1715905 DSN1 -1.19 9.61 0.00

170 ILMN_1737380 BCAP29 -1.19 8.35 0.00 ILMN_1738677 PRPF8 -1.15 12.10 0.00 ILMN_1651872 UBIAD1 -1.19 9.22 0.00 ILMN_1733863 FAM100A -1.15 8.69 0.00 ILMN_1686555 FYN -1.19 9.38 0.00 ILMN_2103295 TINP1 -1.15 12.98 0.00 ILMN_1701991 SYNJ1 -1.18 8.68 0.00 ILMN_1732725 SAPS3 -1.14 9.73 0.00 ILMN_1651385 MFN2 -1.18 9.75 0.00 ILMN_1678922 HERC4 -1.14 9.00 0.00 ILMN_1785284 ALDH6A1 -1.18 9.37 0.00 ILMN_1790533 PHACTR2 -1.14 9.14 0.00 ILMN_1791097 RSBN1 -1.18 9.48 0.00 ILMN_1775304 DNAJB1 -1.14 9.67 0.00 ILMN_1798659 CCDC28A -1.18 10.21 0.00 ILMN_2307450 ZNF302 -1.14 10.08 0.00 ILMN_1718285 HOXC8 -1.18 8.77 0.00 ILMN_1686235 GNPNAT1 -1.14 7.96 0.01 ILMN_1701374 NUP35 -1.18 8.39 0.00 ILMN_2102693 NUFIP2 -1.14 9.86 0.00 ILMN_1660691 RAB31 -1.18 11.51 0.01 ILMN_1717765 NUDT11 -1.14 9.74 0.00 ILMN_1666192 DCTN5 -1.18 10.56 0.00 ILMN_2399264 6-Sep -1.14 8.39 0.00 ILMN_1843198 n/a -1.18 10.30 0.00 ILMN_1709101 AKAP13 -1.14 7.92 0.00 ILMN_1734254 ZNF770 -1.18 9.53 0.00 ILMN_2052208 GADD45A -1.14 9.77 0.00 ILMN_1807600 NPLOC4 -1.18 9.26 0.00 ILMN_2172969 STXBP6 -1.14 8.01 0.01 ILMN_1746673 3-Sep -1.18 10.01 0.00 ILMN_2095840 MYST3 -1.14 10.08 0.00 ILMN_1727692 TRIT1 -1.18 9.88 0.00 ILMN_2166384 IPO5 -1.13 9.40 0.01 ILMN_1803005 MMACHC -1.18 9.54 0.00 ILMN_1763634 PEX14 -1.13 8.82 0.00 ILMN_1806017 PSME1 -1.18 11.58 0.00 ILMN_1812191 C12orf57 -1.13 11.77 0.00 ILMN_1774844 MAPKAPK2 -1.18 7.83 0.00 ILMN_1689318 NUAK1 -1.13 9.51 0.00 ILMN_2223350 C13orf1 -1.18 8.16 0.00 ILMN_1812776 FBXO28 -1.13 9.44 0.00 ILMN_1708611 RDX -1.18 10.05 0.00 ILMN_2139100 SHISA5 -1.13 11.01 0.00 ILMN_2402766 AFTPH -1.18 8.11 0.00 ILMN_1742069 ZSWIM5 -1.13 8.42 0.00 ILMN_1768050 SCOC -1.17 10.91 0.00 ILMN_1658464 GTF3A -1.13 11.86 0.01 ILMN_1742379 IFT122 -1.17 8.41 0.00 ILMN_1682197 NFXL1 -1.13 9.19 0.00 ILMN_1694323 GLE1 -1.17 7.86 0.00 ILMN_2111187 ELOVL6 -1.13 10.69 0.00 ILMN_1692056 HS3ST3A1 -1.17 10.64 0.00 ILMN_1865764 n/a -1.13 8.50 0.01 ILMN_1666050 TMUB1 -1.17 11.07 0.00 ILMN_1773906 NCOA4 -1.13 12.66 0.00 ILMN_2398587 ZNRD1 -1.17 9.53 0.00 ILMN_1754660 ZCCHC24 -1.13 8.67 0.00 ILMN_1775759 NRAS -1.17 9.18 0.00 ILMN_2377430 AGPAT2 -1.13 9.55 0.00 ILMN_1775008 NCAPD2 -1.17 10.57 0.00 ILMN_1742456 OSTF1 -1.13 9.08 0.00 ILMN_1809818 PRCC -1.17 8.13 0.00 ILMN_1768197 ROD1 -1.13 10.92 0.00 ILMN_1735548 HIVEP1 -1.17 7.87 0.00 ILMN_1744138 CHCHD7 -1.12 9.28 0.00 ILMN_1789095 BMPR2 -1.17 8.78 0.00 ILMN_1736340 ANGEL2 -1.12 10.40 0.00 ILMN_1695276 MAPRE2 -1.17 10.98 0.00 ILMN_1679092 RAB9B -1.12 8.32 0.00 ILMN_1784785 COPS7B -1.17 10.45 0.00 ILMN_1682799 STAMBPL1 -1.12 9.31 0.00 ILMN_1676899 YEATS2 -1.17 9.36 0.00 ILMN_1800225 PPARG -1.12 9.75 0.00 ILMN_1738712 GPR180 -1.17 9.86 0.00 ILMN_1737611 VAMP1 -1.12 10.14 0.01 ILMN_1753279 HNRNPA0 -1.17 11.55 0.01 ILMN_1679838 WBP5 -1.12 11.84 0.00 ILMN_1711703 C16orf70 -1.17 8.21 0.00 ILMN_2097858 KIAA1737 -1.12 9.57 0.00 ILMN_1796210 PPRC1 -1.17 10.59 0.00 ILMN_1703053 ZFP91 -1.12 7.98 0.00 ILMN_1661636 ZMYM2 -1.17 8.95 0.00 ILMN_1814165 SSBP3 -1.12 7.94 0.00 ILMN_1738657 SATB2 -1.17 9.71 0.00 ILMN_1657771 CRTC2 -1.12 7.91 0.00 ILMN_1762972 CHD9 -1.16 9.56 0.00 ILMN_1716766 CEBPG -1.12 9.74 0.00 ILMN_1714335 RDH10 -1.16 9.40 0.00 ILMN_1677432 SRGAP1 -1.11 8.19 0.00 ILMN_1785107 NXT2 -1.16 8.77 0.01 ILMN_1652521 MTMR9 -1.11 9.03 0.00 ILMN_1761808 MCFD2 -1.16 9.37 0.01 ILMN_2301083 UBE2C -1.11 12.58 0.00 ILMN_2404906 SGOL1 -1.16 8.58 0.00 ILMN_2373495 H2AFY -1.11 14.22 0.00 ILMN_2395236 CHEK2 -1.16 9.00 0.00 ILMN_1803476 KCTD20 -1.11 9.96 0.00 ILMN_2404112 TRDMT1 -1.16 8.00 0.00 ILMN_1741148 ALDOA -1.11 12.85 0.00 ILMN_1811997 ZNF364 -1.16 10.17 0.01 ILMN_1691432 PRDM4 -1.11 9.59 0.00 ILMN_1795228 ZFAND5 -1.16 11.05 0.00 ILMN_2318685 ABHD12 -1.11 8.00 0.00 ILMN_1697559 G6PD -1.15 8.44 0.00 ILMN_1676745 ZNF142 -1.11 8.95 0.00 ILMN_1656501 DUSP5 -1.15 8.60 0.00 ILMN_1764186 LOC146517 -1.11 9.12 0.00 ILMN_1783709 RRAGA -1.15 11.56 0.00 ILMN_1764871 PIGP -1.11 11.23 0.00 ILMN_1739876 RAB3GAP1 -1.15 10.88 0.00 ILMN_1661622 TBC1D7 -1.11 10.41 0.00 ILMN_1741054 SLC5A6 -1.15 10.85 0.00 ILMN_2380801 FYN -1.11 8.13 0.00 ILMN_1722894 ZNRD1 -1.15 9.46 0.00 ILMN_2225348 ZNF805 -1.11 8.13 0.00 ILMN_1806106 GNL3 -1.15 10.86 0.00 ILMN_1713143 MRPL3 -1.10 12.60 0.00 ILMN_1729980 RNF216 -1.15 9.45 0.00 ILMN_1656826 SH3RF1 -1.10 8.51 0.00 ILMN_1661337 SRM -1.15 11.51 0.00 ILMN_2109526 CSNK2A1P -1.10 8.84 0.00 ILMN_1782385 POLR2A -1.15 11.46 0.00 ILMN_2097793 ZBTB4 -1.10 9.21 0.00 ILMN_1808501 SH3KBP1 -1.15 9.49 0.00 ILMN_1758673 SLC44A1 -1.10 8.43 0.01 ILMN_1652082 ELF4 -1.15 8.58 0.00 ILMN_2135709 C8orf47 -1.10 8.37 0.00 ILMN_1659364 RFC5 -1.15 11.49 0.00 ILMN_1658143 RFC3 -1.10 9.38 0.01

171 ILMN_1692962 CTDSP2 -1.10 10.20 0.00 ILMN_2222074 PTPN12 -1.04 9.17 0.00 ILMN_2189605 FAM122B -1.10 9.52 0.00 ILMN_2397024 SPOP -1.04 11.57 0.00 ILMN_1765060 FBXO34 -1.10 9.34 0.00 ILMN_2313926 CDC42SE2 -1.04 9.23 0.00 ILMN_1805225 LPCAT3 -1.10 8.86 0.00 ILMN_2392674 PRR3 -1.04 8.26 0.00 ILMN_1766222 LARP5 -1.09 10.24 0.00 ILMN_1748591 ODC1 -1.04 12.28 0.00 ILMN_1744949 RHOBTB3 -1.09 11.23 0.00 ILMN_2100693 MAP2K4 -1.04 7.90 0.00 ILMN_1739032 TMEM70 -1.09 8.09 0.00 ILMN_1657683 C1orf198 -1.04 9.34 0.00 ILMN_1672128 ATF4 -1.09 14.17 0.00 ILMN_1695585 RPS26L -1.04 13.82 0.01 ILMN_2386355 CSNK2A1 -1.09 9.08 0.01 ILMN_1758412 COPS7A -1.04 11.54 0.00 ILMN_1777526 MED20 -1.09 11.46 0.00 ILMN_1764945 AP3D1 -1.04 10.13 0.00 ILMN_1656145 GOT1 -1.09 11.24 0.00 ILMN_1804812 ANAPC1 -1.04 10.69 0.00 ILMN_1800512 HMOX1 -1.09 10.09 0.00 ILMN_2323526 WAC -1.03 10.20 0.00 ILMN_1669635 NUP85 -1.09 11.09 0.00 ILMN_2390114 AP3D1 -1.03 10.27 0.00 ILMN_1776216 TMEM32 -1.09 11.77 0.00 ILMN_1794470 ANKFY1 -1.03 9.18 0.00 ILMN_1772692 DICER1 -1.09 9.39 0.00 ILMN_1666761 PPP2R5E -1.03 10.39 0.01 ILMN_1788961 PPP2R2A -1.09 10.73 0.00 ILMN_1746856 RAB21 -1.03 8.94 0.00 ILMN_1738027 BRCA1 -1.08 9.76 0.00 ILMN_1766359 GATAD2B -1.03 8.46 0.00 ILMN_1657993 ADNP -1.08 9.45 0.01 ILMN_1787410 EIF6 -1.03 11.34 0.00 ILMN_1691499 TJP1 -1.08 10.25 0.00 ILMN_1711381 RFC3 -1.03 7.98 0.00 ILMN_1778255 FARSA -1.08 9.59 0.00 ILMN_1673215 PCBP1 -1.03 12.82 0.00 ILMN_1679195 C20orf24 -1.08 13.25 0.00 ILMN_1876924 WNK1 -1.03 10.10 0.00 ILMN_2115696 USP42 -1.08 9.99 0.00 ILMN_1769118 9-Sep -1.03 12.69 0.00 ILMN_1706957 BMPR1A -1.08 9.05 0.00 ILMN_1761260 COBLL1 -1.03 9.15 0.01 ILMN_2227573 GSTO1 -1.08 12.63 0.00 ILMN_1758250 TRAFD1 -1.03 8.94 0.00 ILMN_1777449 IFT74 -1.08 8.68 0.00 ILMN_1681081 AGPAT2 -1.03 8.27 0.01 ILMN_1719103 UBE2CBP -1.08 7.89 0.00 ILMN_2181445 BCL2L13 -1.03 10.65 0.00 ILMN_1856861 n/a -1.08 8.07 0.00 ILMN_1779264 PSMG1 -1.03 11.04 0.00 ILMN_1671885 MLF2 -1.08 10.46 0.00 ILMN_1673682 GATAD2A -1.03 11.42 0.00 ILMN_2075927 STK40 -1.08 10.53 0.01 ILMN_2108938 FNBP4 -1.02 10.76 0.00 ILMN_1699859 GNPAT -1.08 10.15 0.00 ILMN_1692121 USO1 -1.02 10.57 0.00 ILMN_2403906 ARFIP1 -1.08 9.97 0.00 ILMN_1795905 ZBTB4 -1.02 8.25 0.00 ILMN_1713380 EIF2B2 -1.08 11.05 0.00 ILMN_1744628 FDX1L -1.02 9.54 0.01 ILMN_1736548 PHACTR4 -1.07 10.82 0.00 ILMN_1714383 TPD52L1 -1.02 9.04 0.00 ILMN_1726108 LASS2 -1.07 10.09 0.00 ILMN_1827736 n/a -1.02 10.73 0.00 ILMN_2128428 DAB2 -1.07 8.79 0.00 ILMN_1682783 TUG1 -1.02 11.89 0.00 ILMN_1723020 MAP3K1 -1.07 9.27 0.00 ILMN_1669599 DENND4C -1.02 8.25 0.00 ILMN_1804329 TUSC2 -1.07 8.41 0.00 ILMN_1756898 COQ9 -1.02 10.37 0.00 ILMN_1807106 LDHA -1.07 13.93 0.00 ILMN_1765860 DOCK11 -1.02 8.70 0.01 ILMN_2356890 MRPL42 -1.07 9.02 0.00 ILMN_2200636 KIAA1267 -1.02 10.30 0.00 ILMN_1700695 SLC44A1 -1.07 10.99 0.00 ILMN_1813938 CHCHD4 -1.02 10.79 0.00 ILMN_1713756 GLUD1 -1.07 11.35 0.01 ILMN_1787705 ATP6V1B2 -1.02 11.61 0.00 ILMN_1811648 DCAKD -1.07 10.80 0.00 ILMN_1802615 CDK6 -1.02 10.18 0.00 ILMN_1654697 ZNF280B -1.06 8.28 0.00 ILMN_1726554 IREB2 -1.02 9.44 0.00 ILMN_1694075 GADD45A -1.06 8.70 0.00 ILMN_1708203 OTUD4 -1.02 8.83 0.00 ILMN_1794349 XYLB -1.06 8.20 0.00 ILMN_2234229 PRMT6 -1.02 10.57 0.00 ILMN_2147306 PNRC2 -1.06 11.06 0.00 ILMN_2053921 CAPZB -1.02 8.48 0.00 ILMN_1742578 MKLN1 -1.06 9.87 0.00 ILMN_1669931 TM9SF3 -1.02 9.55 0.00 ILMN_1685774 LOC647340 -1.06 13.41 0.00 ILMN_1761031 PTPDC1 -1.02 8.11 0.00 ILMN_1862018 ATXN7L3 -1.06 9.48 0.00 ILMN_2380946 EIF4G2 -1.02 13.45 0.00 ILMN_1684594 USP24 -1.06 9.88 0.00 ILMN_1798108 C6orf211 -1.01 8.06 0.00 ILMN_1686152 GGA2 -1.05 9.95 0.00 ILMN_1664909 LOC729351 -1.01 7.99 0.00 ILMN_1691575 SNX2 -1.05 10.96 0.00 ILMN_1705908 RPL7L1 -1.01 11.74 0.01 ILMN_2263718 SPAG9 -1.05 10.60 0.00 ILMN_2413141 C6orf162 -1.01 8.19 0.00 ILMN_1763265 CHMP1B -1.05 9.20 0.00 ILMN_1671809 DUSP22 -1.01 9.68 0.00 ILMN_1663605 RNF123 -1.05 8.84 0.01 ILMN_1659099 ROCK2 -1.01 8.62 0.01 ILMN_1732300 POLR2C -1.05 8.81 0.00 ILMN_1743711 LOC650215 -1.01 10.19 0.00 ILMN_1748697 LIN28B -1.05 11.27 0.01 ILMN_2312732 DPP8 -1.01 8.26 0.00 ILMN_1653404 NKIRAS2 -1.05 9.15 0.01 ILMN_2045911 FBXO28 -1.01 9.74 0.00 ILMN_1655117 WDR19 -1.05 9.56 0.00 ILMN_2337655 WARS -1.01 10.43 0.00 ILMN_2335669 ZC3H14 -1.05 9.19 0.00 ILMN_1748651 PSMB3 -1.01 13.68 0.00 ILMN_1727271 WARS -1.05 9.66 0.00 ILMN_1730645 TMEFF2 -1.01 7.89 0.00 ILMN_1813275 DUSP22 -1.05 11.11 0.00 ILMN_1681703 FOXO3 -1.01 11.50 0.01 ILMN_2209027 RPS26 -1.05 12.61 0.00 ILMN_1721729 PPARBP -1.01 9.50 0.01 ILMN_1841334 n/a -1.04 8.54 0.00 ILMN_1709474 ALMS1 -1.01 7.97 0.00 ILMN_1730294 INO80C -1.04 10.52 0.00 ILMN_1800993 CLUAP1 -1.00 8.13 0.00

172 ILMN_1685978 ATPIF1 -1.00 8.67 0.00 ILMN_1695432 TPST2 -0.96 8.57 0.00 ILMN_1799289 MRPL55 -1.00 9.86 0.00 ILMN_1691425 LDOC1L -0.96 9.54 0.00 ILMN_2090105 TAGLN2 -1.00 11.23 0.00 ILMN_1750029 GABPA -0.96 8.35 0.01 ILMN_1803060 MAML1 -1.00 7.90 0.00 ILMN_1805990 BAK1 -0.96 8.52 0.00 ILMN_1804479 MRPL18 -1.00 11.29 0.01 ILMN_1811489 OXSR1 -0.96 10.24 0.01 ILMN_1724052 LOC730051 -1.00 7.90 0.00 ILMN_1659285 PSMG1 -0.96 11.37 0.00 ILMN_2391355 STAMBP -1.00 9.26 0.00 ILMN_1677396 NDFIP2 -0.96 11.74 0.01 ILMN_1769869 PIP5K1A -1.00 7.83 0.00 ILMN_1778444 FKBP5 -0.96 9.49 0.00 ILMN_1799642 TRIM24 -1.00 9.78 0.00 ILMN_2336109 L3MBTL2 -0.96 10.22 0.00 ILMN_1675577 C14orf172 -1.00 8.39 0.00 ILMN_1773741 GOLGA5 -0.96 9.56 0.00 ILMN_1784227 MCRS1 -1.00 11.75 0.00 ILMN_2312296 PCBP2 -0.96 13.41 0.00 ILMN_1793729 C15orf39 -1.00 9.47 0.00 ILMN_2311518 TROVE2 -0.96 11.23 0.01 ILMN_2337923 TPD52L1 -1.00 9.05 0.00 ILMN_1666122 HEG1 -0.96 8.02 0.00 ILMN_1745116 ABHD12 -1.00 7.86 0.00 ILMN_1663042 SDC4 -0.95 7.77 0.00 ILMN_2346727 MTUS1 -1.00 7.79 0.00 ILMN_2124769 YBX1 -0.95 12.70 0.00 ILMN_1659544 STX3 -1.00 9.22 0.01 ILMN_2245305 ABHD12 -0.95 8.49 0.00 ILMN_1800871 RAB6A -1.00 8.31 0.00 ILMN_2088612 XPO4 -0.95 9.24 0.00 ILMN_1658884 ATP11B -1.00 7.78 0.00 ILMN_1666049 NUP214 -0.95 9.01 0.00 ILMN_2132161 KIF18A -1.00 8.51 0.00 ILMN_1788783 TRAM2 -0.95 8.90 0.01 ILMN_1774949 PIGP -1.00 9.05 0.00 ILMN_1767848 PCMTD2 -0.95 8.84 0.00 ILMN_2038776 TXN -1.00 14.39 0.00 ILMN_1784037 ZBTB40 -0.95 8.60 0.01 ILMN_1802642 TOM1L1 -0.99 7.99 0.00 ILMN_2297710 PLEKHB2 -0.95 8.71 0.00 ILMN_1799604 OCIAD1 -0.99 11.54 0.01 ILMN_1752810 LARP6 -0.95 9.42 0.00 ILMN_1736481 SECISBP2 -0.99 9.14 0.00 ILMN_1736575 TRIM28 -0.95 10.48 0.00 ILMN_2189668 NUDT11 -0.99 10.05 0.00 ILMN_1786036 GPATCH2 -0.95 7.93 0.00 ILMN_2171461 ZNF322A -0.99 8.11 0.00 ILMN_1660840 FAM152A -0.95 10.27 0.00 ILMN_1743456 ZCCHC14 -0.99 9.48 0.00 ILMN_2327795 RERE -0.95 8.44 0.00 ILMN_2398388 APH1A -0.99 9.17 0.01 ILMN_1734096 DCLRE1A -0.95 9.33 0.00 ILMN_1664440 TP53BP1 -0.99 9.52 0.01 ILMN_1698968 ASXL2 -0.95 9.37 0.00 ILMN_1797342 FNBP1 -0.99 10.27 0.00 ILMN_1771224 C18orf24 -0.94 8.63 0.00 ILMN_1653871 NAMPT -0.99 8.00 0.00 ILMN_2358842 PCBP4 -0.94 8.86 0.00 ILMN_1691119 RNF122 -0.99 7.87 0.00 ILMN_1806778 UBE2E1 -0.94 12.46 0.00 ILMN_1728676 KIAA0196 -0.99 10.45 0.01 ILMN_2106227 KIAA2026 -0.94 7.86 0.00 ILMN_2214098 BIVM -0.99 9.61 0.00 ILMN_2158242 SHOC2 -0.94 8.37 0.00 ILMN_1708906 C2orf29 -0.99 12.03 0.01 ILMN_1676005 KPNA1 -0.94 9.72 0.00 ILMN_1763436 SETX -0.99 7.83 0.00 ILMN_2169856 C12orf43 -0.94 9.16 0.01 ILMN_2124471 SLC36A1 -0.99 8.25 0.00 ILMN_1745962 FBXO7 -0.94 10.14 0.00 ILMN_1799128 SLC30A9 -0.99 10.42 0.00 ILMN_1791478 MTPN -0.94 10.75 0.00 ILMN_1732176 AGPAT2 -0.99 8.62 0.00 ILMN_1665357 EPS15 -0.94 8.21 0.00 ILMN_1758049 NFIA -0.99 7.90 0.00 ILMN_1738420 TMEM201 -0.94 8.29 0.00 ILMN_1727309 FAM82A2 -0.99 9.81 0.01 ILMN_2208495 LASS5 -0.94 8.76 0.01 ILMN_1733421 PRKCQ -0.98 9.89 0.00 ILMN_1803180 PRDX6 -0.94 12.71 0.00 ILMN_1657873 XPO4 -0.98 10.09 0.00 ILMN_1722533 KATNAL1 -0.94 8.14 0.00 ILMN_2396956 AKAP13 -0.98 9.28 0.00 ILMN_2317348 APTX -0.94 9.76 0.00 ILMN_1681590 LARP1 -0.98 9.87 0.00 ILMN_1807540 CBARA1 -0.94 9.09 0.00 ILMN_1779886 TBC1D14 -0.98 10.02 0.00 ILMN_1661039 MRPL30 -0.94 7.96 0.00 ILMN_2357577 PRKAA1 -0.98 9.54 0.00 ILMN_2230683 CDCA7L -0.94 8.52 0.01 ILMN_2083567 PHLPPL -0.98 8.13 0.01 ILMN_1657624 NSFL1C -0.94 9.27 0.00 ILMN_1721842 RYBP -0.98 9.72 0.01 ILMN_1759801 DPP8 -0.93 8.71 0.00 ILMN_1669070 MIPEP -0.98 8.69 0.00 ILMN_1726752 APTX -0.93 8.64 0.00 ILMN_1652735 RFXAP -0.98 8.54 0.00 ILMN_1761519 EIF4G2 -0.93 13.02 0.00 ILMN_1753467 SAMD4B -0.98 8.19 0.00 ILMN_1704446 SLC6A10P -0.93 8.78 0.00 ILMN_2176768 SEPHS1 -0.98 10.93 0.00 ILMN_1662839 PLEKHA1 -0.93 10.21 0.00 ILMN_1793203 SMCR7L -0.98 10.71 0.00 ILMN_1763828 MTF1 -0.93 8.68 0.00 ILMN_1739749 B3GALT6 -0.98 9.13 0.00 ILMN_2248589 DHX40 -0.93 10.98 0.00 ILMN_2081883 IQCK -0.97 8.87 0.00 ILMN_1724540 CART1 -0.93 8.60 0.01 ILMN_1666306 SRRD -0.97 8.86 0.00 ILMN_1774974 CLUAP1 -0.93 7.80 0.01 ILMN_1704665 GPM6B -0.97 9.78 0.00 ILMN_2328972 DNMT3B -0.93 8.78 0.00 ILMN_1666372 ATP5H -0.97 12.99 0.00 ILMN_2117987 TFDP1 -0.93 11.89 0.01 ILMN_1664560 DYRK1A -0.97 8.96 0.00 ILMN_2328378 OSBPL3 -0.93 8.46 0.00 ILMN_1693323 C6orf162 -0.97 8.08 0.00 ILMN_1711919 SCYL2 -0.92 9.02 0.00 ILMN_1785406 NAP1L3 -0.97 8.41 0.00 ILMN_1672135 ZNF615 -0.92 8.58 0.00 ILMN_1725300 MLL3 -0.97 8.51 0.01 ILMN_1754235 SLC35C2 -0.92 8.16 0.00 ILMN_2353697 MIZF -0.97 8.49 0.00 ILMN_1674506 MED23 -0.92 8.71 0.01 ILMN_1705310 VEZF1 -0.97 10.52 0.00 ILMN_2405031 TRIM24 -0.92 7.99 0.00

173 ILMN_1747775 STX2 -0.92 9.72 0.01 ILMN_2288232 PHKB -0.88 8.54 0.01 ILMN_1699208 NAP1L1 -0.92 12.18 0.01 ILMN_2224031 CETN3 -0.88 9.41 0.00 ILMN_1769473 SETD2 -0.92 9.83 0.01 ILMN_1667213 DFFA -0.88 10.27 0.01 ILMN_1736796 RB1CC1 -0.92 9.96 0.00 ILMN_2294274 S100PBP -0.88 8.18 0.00 ILMN_2290204 ARHGAP28 -0.92 7.87 0.00 ILMN_1679891 NAF1 -0.88 8.52 0.00 ILMN_2327860 MAL -0.92 7.79 0.00 ILMN_1667319 LPPR2 -0.87 8.19 0.01 ILMN_1698487 SDHD -0.92 10.62 0.01 ILMN_1808374 SNTB2 -0.87 9.44 0.00 ILMN_1679880 THOC6 -0.92 8.86 0.01 ILMN_1737833 ATN1 -0.87 8.58 0.01 ILMN_1775243 LOC646766 -0.92 14.02 0.01 ILMN_1757347 C22orf9 -0.87 7.77 0.00 ILMN_1714278 C9orf30 -0.92 8.95 0.00 ILMN_1779381 SEC61A2 -0.87 8.61 0.00 ILMN_1692486 ZNRD1 -0.91 10.03 0.01 ILMN_1781099 ISY1 -0.87 9.85 0.00 ILMN_1677953 OGFOD1 -0.91 9.81 0.00 ILMN_1762787 RNF26 -0.87 9.28 0.00 ILMN_1802831 MTMR12 -0.91 7.91 0.01 ILMN_2215631 OTUD6B -0.87 8.93 0.00 ILMN_1665510 ERRFI1 -0.91 9.98 0.00 ILMN_1684321 CYB5B -0.87 12.37 0.00 ILMN_1777584 KARS -0.91 9.75 0.01 ILMN_1669424 LOC646531 -0.87 13.58 0.01 ILMN_1714990 DBT -0.91 9.68 0.00 ILMN_1778991 NFIB -0.87 8.60 0.01 ILMN_1716400 FOXM1 -0.91 9.22 0.00 ILMN_2313158 MBNL1 -0.87 8.40 0.00 ILMN_1788738 ZNRF3 -0.91 7.89 0.00 ILMN_1749081 AUTS2 -0.87 10.32 0.00 ILMN_2117223 ROD1 -0.91 10.63 0.00 ILMN_1663195 MCM7 -0.87 13.44 0.01 ILMN_2367743 TUBG1 -0.91 11.98 0.00 ILMN_1798886 NUDT21 -0.87 10.10 0.01 ILMN_1745811 TDRD3 -0.91 8.01 0.00 ILMN_1799516 DNAJC9 -0.87 11.84 0.01 ILMN_2192620 LOC90379 -0.91 8.80 0.00 ILMN_2403889 PRMT5 -0.87 10.78 0.01 ILMN_1712774 IRS4 -0.91 8.22 0.01 ILMN_2358041 NBN -0.87 7.72 0.01 ILMN_1742147 UBL4A -0.91 9.63 0.01 ILMN_1716272 KBTBD8 -0.87 8.84 0.00 ILMN_1733627 NEDD4L -0.91 8.20 0.00 ILMN_1666305 CDKN3 -0.86 9.88 0.01 ILMN_1814573 FTSJD1 -0.91 8.49 0.00 ILMN_1739586 FEZ2 -0.86 10.58 0.00 ILMN_1710738 RC3H2 -0.91 8.71 0.00 ILMN_1725043 ADAL -0.86 8.10 0.01 ILMN_2089656 C1orf107 -0.91 8.24 0.00 ILMN_1772527 C12orf44 -0.86 9.89 0.01 ILMN_2144088 FDFT1 -0.91 12.48 0.01 ILMN_1765578 TIPARP -0.86 8.69 0.00 ILMN_1686097 TOP2A -0.90 12.26 0.01 ILMN_1686136 NSD1 -0.86 7.95 0.00 ILMN_1688526 ARL5A -0.90 11.83 0.01 ILMN_1746276 EPC1 -0.86 7.72 0.00 ILMN_1726783 RNASEH1 -0.90 10.00 0.00 ILMN_1762224 HPS3 -0.86 7.77 0.00 ILMN_1814966 PIAS3 -0.90 7.83 0.00 ILMN_2223903 PPIC -0.86 10.49 0.00 ILMN_2404049 RBM38 -0.90 9.01 0.01 ILMN_1656316 ZMYM3 -0.85 8.69 0.00 ILMN_1682336 MASTL -0.90 8.96 0.00 ILMN_1705876 NAP1L1 -0.85 9.80 0.00 ILMN_1844692 FOXO3 -0.90 9.97 0.01 ILMN_1767542 THAP10 -0.85 9.29 0.00 ILMN_1747016 CEP55 -0.90 10.20 0.01 ILMN_1679797 ADARB1 -0.85 9.20 0.01 ILMN_1669905 DCP2 -0.90 11.42 0.01 ILMN_1699476 RPE -0.85 9.13 0.01 ILMN_1687743 BTBD7 -0.90 8.93 0.01 ILMN_2403946 FEZ2 -0.85 9.71 0.01 ILMN_1802380 RERE -0.90 9.43 0.00 ILMN_1697529 RNF10 -0.85 8.35 0.01 ILMN_1662852 IQCK -0.90 7.85 0.00 ILMN_1719097 C18orf8 -0.85 9.54 0.00 ILMN_2364384 PPARG -0.90 7.80 0.00 ILMN_1763409 LRRC8D -0.85 8.09 0.00 ILMN_1748916 C18orf55 -0.89 9.61 0.01 ILMN_1727134 KLHDC5 -0.85 9.07 0.00 ILMN_1760849 NETO2 -0.89 10.52 0.00 ILMN_1772876 ZNF395 -0.85 9.12 0.00 ILMN_1671895 ZNF613 -0.89 8.23 0.01 ILMN_2335718 HNRNPAB -0.85 12.41 0.01 ILMN_1743104 RBM4B -0.89 9.69 0.00 ILMN_1749641 FBXO3 -0.85 8.41 0.01 ILMN_1755352 TUB -0.89 9.12 0.00 ILMN_1799814 WDR57 -0.85 10.77 0.01 ILMN_1764803 SMU1 -0.89 8.14 0.00 ILMN_1683950 SNX11 -0.85 9.54 0.01 ILMN_1769566 ATG3 -0.89 10.03 0.00 ILMN_2387799 PDPK1 -0.85 9.79 0.01 ILMN_1711089 DNAJC21 -0.89 7.97 0.00 ILMN_1810797 WASF3 -0.85 8.93 0.01 ILMN_1722858 PPP2CA -0.89 13.39 0.01 ILMN_1805826 BIVM -0.85 9.45 0.00 ILMN_1745798 GTF2F2 -0.89 10.92 0.00 ILMN_2225974 GCLM -0.85 8.01 0.00 ILMN_1785179 UBE2G2 -0.89 10.28 0.00 ILMN_1714444 KLF12 -0.85 8.36 0.00 ILMN_1775170 MT1X -0.89 11.17 0.01 ILMN_1797813 SUZ12 -0.85 12.50 0.00 ILMN_2381753 G3BP2 -0.89 11.28 0.00 ILMN_1740395 RAVER1 -0.84 9.40 0.00 ILMN_1741264 MRPS33 -0.88 11.59 0.00 ILMN_1781526 PPP1R8 -0.84 8.20 0.01 ILMN_1692651 PHB -0.88 11.54 0.00 ILMN_1732489 SLC10A7 -0.84 8.45 0.01 ILMN_1736002 COPS5 -0.88 11.98 0.01 ILMN_1682232 MIER1 -0.84 9.19 0.01 ILMN_1773459 SOX11 -0.88 8.47 0.01 ILMN_1692260 MAFG -0.84 7.97 0.01 ILMN_1730048 C7orf26 -0.88 9.68 0.00 ILMN_1737211 ZNF585A -0.84 7.88 0.00 ILMN_1774083 TRIAP1 -0.88 10.02 0.01 ILMN_1754839 DHX15 -0.84 11.66 0.01 ILMN_2335319 KCNG3 -0.88 7.75 0.00 ILMN_1674662 C15orf42 -0.84 8.74 0.00 ILMN_1714216 TSC2 -0.88 8.83 0.00 ILMN_1763104 TRAF4 -0.84 7.65 0.00 ILMN_1760890 SEPN1 -0.88 9.79 0.01 ILMN_2127624 UBE2CBP -0.84 8.23 0.00 ILMN_1770412 AHCYL1 -0.88 11.21 0.01 ILMN_2096405 WDR37 -0.84 9.16 0.01

174 ILMN_1691567 GNPDA2 -0.84 8.23 0.00 ILMN_1784238 SEC22B -0.79 9.09 0.01 ILMN_2192281 CARD8 -0.84 8.38 0.00 ILMN_2187533 FAM130A1 -0.79 8.09 0.01 ILMN_1656186 SLC41A1 -0.84 8.41 0.00 ILMN_1797585 MYO1B -0.79 7.95 0.00 ILMN_1748438 POLR2G -0.84 12.09 0.01 ILMN_2182531 C18orf55 -0.79 10.72 0.01 ILMN_1763091 C14orf43 -0.84 8.80 0.00 ILMN_1808860 STX5 -0.79 8.51 0.01 ILMN_1683450 CDCA5 -0.84 11.96 0.00 ILMN_1665554 BRF2 -0.79 9.10 0.01 ILMN_1743829 ATXN2 -0.84 10.71 0.00 ILMN_1736650 JMJD2A -0.79 7.71 0.01 ILMN_1696485 HNRNPAB -0.84 11.24 0.01 ILMN_1739297 GALNT4 -0.79 7.86 0.01 ILMN_1699015 H3F3A -0.84 13.94 0.01 ILMN_1686194 SDCCAG10 -0.79 10.44 0.01 ILMN_2080637 ZBTB44 -0.84 7.95 0.00 ILMN_1807016 LHX2 -0.79 9.11 0.01 ILMN_1707156 LRRFIP2 -0.84 9.13 0.01 ILMN_1709549 PLEKHM1 -0.79 8.11 0.00 ILMN_2346460 NARG2 -0.83 8.00 0.00 ILMN_2058468 BACH2 -0.79 7.95 0.00 ILMN_1699703 ARCN1 -0.83 11.13 0.01 ILMN_2061310 ZNF280C -0.79 8.88 0.00 ILMN_1710962 TMEM97 -0.83 12.24 0.01 ILMN_1788988 THAP1 -0.79 7.62 0.00 ILMN_1815121 PLAGL1 -0.83 7.80 0.00 ILMN_1682727 JAZF1 -0.79 8.11 0.00 ILMN_1706376 OSBP -0.83 11.02 0.01 ILMN_1694041 PDCL -0.79 7.81 0.00 ILMN_1734288 DUSP18 -0.83 8.16 0.00 ILMN_1706901 HOXD13 -0.79 8.95 0.00 ILMN_1668283 HYAL2 -0.83 8.48 0.01 ILMN_1757644 UBE2H -0.78 8.16 0.00 ILMN_1690223 CNTNAP2 -0.83 10.10 0.01 ILMN_2220739 TMCO3 -0.78 8.74 0.01 ILMN_1658677 DTX3 -0.83 7.60 0.00 ILMN_1797367 TSC1 -0.78 8.44 0.01 ILMN_1693259 PDCD6IP -0.83 9.20 0.01 ILMN_1808590 GUCY1A3 -0.78 8.87 0.01 ILMN_1726421 METTL9 -0.83 8.42 0.00 ILMN_1803398 SRF -0.78 10.25 0.00 ILMN_1664776 EFR3A -0.83 9.78 0.01 ILMN_2133638 DULLARD -0.78 8.47 0.00 ILMN_1720844 SSX2IP -0.83 8.28 0.01 ILMN_1709623 MAPK7 -0.78 8.92 0.00 ILMN_1775937 DDB1 -0.83 13.48 0.01 ILMN_2401618 MLX -0.78 8.06 0.01 ILMN_1659189 C9orf89 -0.83 8.58 0.00 ILMN_1702279 KIF3B -0.78 8.51 0.00 ILMN_1794956 BBS9 -0.82 8.10 0.00 ILMN_1691892 TAGLN2 -0.78 9.09 0.01 ILMN_1767015 BCORL1 -0.82 7.97 0.01 ILMN_2163306 FAM120A -0.78 11.60 0.00 ILMN_2157951 STX6 -0.82 9.06 0.00 ILMN_1813389 MRPS7 -0.78 11.22 0.01 ILMN_2142284 SLC25A43 -0.82 8.67 0.00 ILMN_1733511 GOLGA3 -0.78 10.44 0.01 ILMN_1807177 KIAA1797 -0.82 8.50 0.00 ILMN_1726704 RSC1A1 -0.77 7.97 0.01 ILMN_1699598 AP2M1 -0.82 10.09 0.00 ILMN_1760575 PTP4A1 -0.77 8.56 0.01 ILMN_1667453 MIZF -0.82 8.46 0.00 ILMN_1741133 NME1 -0.77 14.52 0.01 ILMN_1741331 C14orf108 -0.82 9.03 0.01 ILMN_1789233 VPS37C -0.77 10.88 0.01 ILMN_1677765 LRP8 -0.82 8.37 0.00 ILMN_2374249 DYRK2 -0.77 8.95 0.00 ILMN_1662719 GPBP1L1 -0.82 8.34 0.00 ILMN_1681135 SPATA2 -0.77 7.75 0.00 ILMN_1749014 ACLY -0.82 11.75 0.01 ILMN_1708805 NCOA3 -0.77 8.95 0.01 ILMN_1744046 DIAPH2 -0.82 8.48 0.01 ILMN_1777915 STX6 -0.77 8.67 0.01 ILMN_1653797 C6orf62 -0.82 10.41 0.01 ILMN_1788878 MORN4 -0.77 8.03 0.01 ILMN_2174805 CD300LG -0.82 7.84 0.01 ILMN_1744649 PSMB5 -0.77 12.34 0.00 ILMN_1745946 CCDC5 -0.82 8.82 0.01 ILMN_1697118 ARMC6 -0.77 8.63 0.01 ILMN_2175474 MTRF1L -0.82 10.06 0.00 ILMN_1709114 MAP3K7IP1 -0.77 8.55 0.01 ILMN_1699676 C14orf147 -0.81 8.76 0.01 ILMN_1800976 NFATC3 -0.77 7.90 0.00 ILMN_1802397 GNA11 -0.81 9.54 0.01 ILMN_1720311 SLC25A46 -0.77 9.40 0.00 ILMN_1669252 CUL2 -0.81 10.19 0.01 ILMN_1739454 USP34 -0.76 8.17 0.01 ILMN_1792314 ACTR1A -0.81 10.85 0.01 ILMN_1741005 RG9MTD2 -0.76 8.02 0.01 ILMN_1691736 ST6GALNAC6 -0.81 8.80 0.00 ILMN_1670096 NRBP1 -0.76 7.80 0.00 ILMN_1678546 PEX11B -0.81 9.61 0.00 ILMN_1696532 RBBP5 -0.76 8.59 0.01 ILMN_1754547 R3HDM1 -0.81 8.16 0.00 ILMN_1719064 KCTD10 -0.76 8.22 0.00 ILMN_1717477 PSD3 -0.81 8.13 0.00 ILMN_1759117 XK -0.76 7.81 0.01 ILMN_1746375 CSNK2A1P -0.81 8.51 0.01 ILMN_1657679 VAV3 -0.76 7.76 0.01 ILMN_2180866 RPS26L1 -0.81 13.94 0.01 ILMN_1811972 MYCBP2 -0.76 10.02 0.01 ILMN_1700660 RNF135 -0.81 8.65 0.01 ILMN_1664168 SLC25A11 -0.76 9.08 0.01 ILMN_1696591 RB1 -0.81 7.75 0.01 ILMN_1666208 C14orf106 -0.76 9.88 0.00 ILMN_2324375 CHCHD7 -0.80 7.89 0.01 ILMN_2225144 EIF4E3 -0.76 8.50 0.01 ILMN_1675674 UBE4B -0.80 9.53 0.01 ILMN_2057981 FAM164A -0.75 7.85 0.01 ILMN_1726025 ASXL1 -0.80 9.28 0.01 ILMN_1738276 TMEM185A -0.75 9.02 0.01 ILMN_1726289 C12orf35 -0.80 7.80 0.00 ILMN_1813423 NAT15 -0.75 7.88 0.01 ILMN_1731518 LOC201164 -0.80 8.49 0.00 ILMN_1697117 TBP -0.75 9.77 0.01 ILMN_2355168 MGST1 -0.80 10.26 0.01 ILMN_1665205 ZNF260 -0.75 8.64 0.00 ILMN_1703617 AHSA1 -0.80 11.34 0.00 ILMN_1698770 C5orf33 -0.75 8.02 0.00 ILMN_1762407 CABLES2 -0.80 8.04 0.01 ILMN_1762932 CHMP2A -0.75 7.74 0.00 ILMN_1687538 ETS1 -0.80 8.37 0.01 ILMN_2167922 TRMT5 -0.75 13.30 0.01 ILMN_1720233 CCDC49 -0.80 8.63 0.01 ILMN_1772722 MRPS33 -0.75 8.78 0.01 ILMN_1751645 OR3A4 -0.80 7.74 0.01 ILMN_1800058 NKX2-5 -0.75 8.94 0.00

175 ILMN_1742230 BAZ1A -0.75 10.36 0.01 ILMN_1663618 STAT3 -0.67 7.90 0.01 ILMN_2354478 CYFIP2 -0.75 8.22 0.01 ILMN_1703791 ANXA7 -0.67 10.81 0.01 ILMN_1806349 SLC6A8 -0.75 8.02 0.01 ILMN_1717809 RNF24 -0.67 7.68 0.01 ILMN_1787280 C1orf135 -0.75 9.78 0.01 ILMN_1658494 C13orf15 -0.67 8.61 0.01 ILMN_1780188 B3GALNT2 -0.74 7.77 0.00 ILMN_2278636 CUTL1 -0.66 8.79 0.01 ILMN_1751051 C7orf25 -0.74 8.43 0.00 ILMN_2120210 RCAN2 -0.66 7.92 0.01 ILMN_2314007 TCF12 -0.74 8.53 0.01 ILMN_1789136 SERF2 -0.66 12.37 0.01 ILMN_1675472 LOC644799 -0.74 9.01 0.01 ILMN_1666364 COQ10A -0.66 9.21 0.01 ILMN_1758852 ENTPD7 -0.74 7.81 0.00 ILMN_1680378 RBM45 -0.65 8.78 0.01 ILMN_1679071 MTX3 -0.74 9.21 0.01 ILMN_1727761 GMEB1 -0.65 8.48 0.01 ILMN_2069593 SFRS2IP -0.74 8.85 0.01 ILMN_1666739 RBM15 -0.64 10.08 0.01 ILMN_1750563 CERCAM -0.74 8.07 0.01 ILMN_1815718 BTRC -0.64 7.87 0.01 ILMN_1793302 WDR4 -0.74 7.78 0.01 ILMN_1658083 ABT1 -0.64 7.72 0.01 ILMN_1789909 TBC1D9B -0.74 8.56 0.01 ILMN_1793241 SRD5A1 -0.63 7.72 0.01 ILMN_1759549 SRGAP2 -0.74 8.50 0.01 ILMN_1709044 TGIF2 -0.74 8.59 0.01 ILMN_2394571 FBXW11 -0.74 10.44 0.01 ILMN_1717393 PTCHD1 -0.73 7.77 0.01 ILMN_1761322 FHOD3 -0.73 8.89 0.00 ILMN_1672843 FBXO8 -0.73 9.02 0.01 ILMN_1666482 SP2 -0.73 8.60 0.01 ILMN_1742450 TAPBP -0.73 8.60 0.00 ILMN_1722726 RICS -0.73 8.25 0.00 ILMN_1815705 LZTFL1 -0.72 8.96 0.00 ILMN_1776522 RAG1AP1 -0.72 10.26 0.01 ILMN_1722738 ROGDI -0.72 7.76 0.01 ILMN_1801600 CCDC97 -0.72 9.50 0.01 ILMN_1759513 RND3 -0.72 8.84 0.01 ILMN_1781691 TRAK2 -0.72 8.92 0.01 ILMN_1657436 FGFR1OP2 -0.72 7.89 0.01 ILMN_1788135 APITD1 -0.72 7.94 0.01 ILMN_1769601 MGC16169 -0.72 8.27 0.01 ILMN_1670000 IQWD1 -0.71 10.24 0.01 ILMN_2199022 SAP30BP -0.71 8.00 0.01 ILMN_1792092 ZCCHC8 -0.71 8.60 0.01 ILMN_2323418 KRIT1 -0.71 8.39 0.01 ILMN_2063925 CTNNBL1 -0.71 9.72 0.01 ILMN_2360028 NFATC3 -0.71 8.10 0.01 ILMN_1689007 SFRS14 -0.71 7.98 0.00 ILMN_1684034 STAT5B -0.71 8.27 0.01 ILMN_1681591 PTPN1 -0.71 10.18 0.01 ILMN_2387505 AP1B1 -0.71 7.65 0.01 ILMN_1796772 ARHGAP28 -0.71 7.85 0.01 ILMN_1671583 MKRN1 -0.71 11.70 0.01 ILMN_2140700 CRIPAK -0.70 8.01 0.01 ILMN_1672331 MAP3K7IP2 -0.70 7.91 0.01 ILMN_1764166 BCKDHB -0.70 8.99 0.01 ILMN_1733155 GIT1 -0.70 7.83 0.01 ILMN_2065783 EXOC2 -0.70 8.51 0.01 ILMN_1771620 SNRPB2 -0.70 10.20 0.01 ILMN_1746031 RIMS4 -0.70 8.29 0.01 ILMN_1658182 MEX3C -0.70 8.29 0.01 ILMN_1709882 ICK -0.69 8.40 0.01 ILMN_1727938 ZNF764 -0.69 7.62 0.00 ILMN_2376833 ZNF200 -0.69 7.67 0.01 ILMN_1761147 GABPB2 -0.68 8.47 0.01 ILMN_1703132 LYRM2 -0.68 10.22 0.01 ILMN_1671902 THUMPD3 -0.68 7.99 0.01 ILMN_1809344 BTBD10 -0.68 8.61 0.01 ILMN_1692272 TEX261 -0.68 8.80 0.01 ILMN_2195703 PPARGC1B -0.68 7.72 0.01 ILMN_1769478 LOC202051 -0.67 8.01 0.01 ILMN_2363621 RBBP8 -0.67 10.85 0.01 ILMN_1749848 SLC35F1 -0.67 9.23 0.01 ILMN_1775939 SF3B2 -0.67 10.98 0.01

176 Supplementary Table 2: Functional Enrichment Analysis of genes predicted to be miR-182 targets

Category Function Function p-value Molecules Annotation Gene transcription transcription 1.15E-11 ACTR3, AFF1, AFF4, AKAP13, ALX1, ATF1, ATF3, ATF4, ATN1, Expression ATXN7L3, BACH2 (includes EG:60468), BAZ1A, BCL2L1, BMP2, BMPR2, BRCA1, BTRC, CBL, CCNK, CDK7, CDKN1A, CEBPG, CHMP1A, COPS5, CREB5, CRTC2, CSNK1E, CSRNP2, CTDSP2, CUX1, DAB2, DCAF6, DDB1, DHX15, ELF4, ELK1, EPC1, ERCC3, ETS1, FOXM1, FOXN2, FOXN3, FOXO3, GABPA, GABPB1, GATAD2A, GLI3, GPS2, GTF3A, HAND1, HCFC1, HEY1, HIF1A, HINFP, HIPK2, HMGA1, HOXD13, JAZF1, KAT6A, KLF12, LMO4, MAML1, MAP3K1, MAZ, MED1, MED23, MED24, MLX, MTERFD1, MTF1 (includes EG:17764), NCKAP1, NCOA3, NCOA4, NFATC3, NFIA, NFIB, NFYC, NKX2-5, NME1, NSD1, OTUB1, PA2G4, PAX6, PHF21A, PIAS1, PIAS2, PIAS3, PLAGL1, PLAGL2, POLR2A, POLR2C, POLR2D, POLR2G, PPARD, PPARG, PPARGC1B, PPP1R8, PRDM4, PRKAR1A, PRKCQ, PRMT6, PRPF8, PYGO2, R3HDM1, RARA, RB1, RBBP8, RNF10, RNF216, RUVBL1, SMARCB1, SMARCD1, SP1, SP2, SRF, STAT3, STAT5B, TAB1, TAB2, TAGLN2, TBP, TCF12, TFAP2A, TFDP1, TFDP2, TGIF2, THAP1, TRIM24, TRIM28, TROVE2, VEZF1, VPS72, WNK1, YBX1, ZFHX3, ZNF45 Gene transcription transcription of DNA 3.32E-11 AFF4, ALX1, ATF4, BMPR2, BRCA1, CCNK, CDK7, CEBPG, COPS5, Expression endogenous promoter CREB5, CSRNP2, CUX1, DCAF6, ELF4, ELK1, EPC1, ERCC3, ETS1, FOXM1, FOXO3, GABPA, GABPB1, GLI3, HAND1, HCFC1, HIF1A, HOXD13, JAZF1, KLF12, MAML1, MAZ, MED1, MED23, MTF1 (includes EG:17764), NCOA3, NFATC3, NFYC, NKX2-5, PHF21A, PLAGL1, POLR2A, POLR2C, POLR2D, POLR2G, PPARD, PPARG, PPARGC1B, PRDM4, PRKAR1A, RARA, RBBP8, RUVBL1, SMARCB1, SMARCD1, SP1, SP2, SRF, STAT3, STAT5B, TBP, TCF12, TFDP1, TRIM24, TRIM28, TROVE2, VEZF1, VPS72, YBX1, ZFHX3 Gene transcription transcription of DNA 6.94E-07 ACTR3, AFF1, ATF1, ATF3, ATF4, ATXN7L3, CBL, CHMP1A, CREB5, Expression CSRNP2, DAB2, ELK1, ETS1, FOXM1, FOXN2, FOXN3, FOXO3, GABPB1, GATAD2A, GLI3, HEY1, HIF1A, HINFP, HIPK2, HMGA1, KLF12, LMO4, MAZ, MED1, MED23, MED24, MLX, MTERFD1, MTF1 (includes EG:17764), NFIB, NFYC, NKX2-5, NSD1, PA2G4, PAX6, PLAGL2, PPARD, PPARG, PRKCQ, PRMT6, RARA, RB1, SP1, SRF, STAT3, STAT5B, TBP, TCF12, TFDP1, THAP1, TRIM28, YBX1, ZFHX3, ZNF45 Gene transcription transcription of gene 1.37E-04 ACTR3, ATF1, ATN1, BAZ1A, CCNK, CDK7, DDB1, ERCC3, HCFC1, Expression HIF1A, HIPK2, KAT6A, MAP3K1, MED1, NCOA3, NFIA, NFIB, NME1, POLR2A, PYGO2, RARA, RB1, SMARCB1, SP1, SRF, STAT3, TBP, TFAP2A, TGIF2 Gene transcription transcription of HIF- 7.88E-04 HIF1A, PLAGL2, SP1 Expression 1 response element Gene transcription initiation of 1.78E-03 AKAP13, DHX15, ERCC3, NCKAP1, RNF10, TBP Expression transcription Gene transcription transcription of 6.03E-03 PIAS1, PIAS2, PIAS3 Expression androgen-responsive element Gene transcription transcription of 1.01E-02 ATF4, CRTC2 Expression cAMP response element Gene transcription transcription of HIV- 1.93E-02 PPP1R8, SP1 Expression 1 Gene transcription transcription of 3.05E-02 ATF1, ATF3, BRCA1, CDKN1A, CUX1, DAB2, MAML1, MED1, NCOA3, Expression synthetic promoter NCOA4, OTUB1, PPARG, PRKCQ, PYGO2, RARA, RB1, RBBP8, STAT3, TAB2, TBP, TFDP1 Gene transcription transcription of E2F 3.10E-02 TFDP1, TFDP2 Expression binding site Gene transcription transcription of RNA 3.78E-02 SMARCB1, SP1, TBP Expression Gene activation activation of 1.53E-06 AKAP13, ASXL1, ATF1, ATF4, BMP2, BMPR2, BRCA1, BTRC, CBL, Expression synthetic promoter CDK7, DUSP22, ELK1, EPC1, ETS1, FKBP5, GNA13, HDAC3, HIF1A, JAZF1, LRRFIP2, MED1, MLX, MTF1 (includes EG:17764), NCOA3, NCOA4, NME1, PDPK1, PIAS1, PIAS2, PPARD, PPARG, PPARGC1B, PRKAR1A, PRKCQ, RARA, RB1, RFXAP, RNF216, SEPT9, SP1,

177 STAT5B, TAB1, TBP, TFAP2A, TSC1, TSC2, TXN Gene activation activation of 4.22E-04 BRCA1, FKBP5, FOXO3, PA2G4, PIAS1, PIAS2 Expression androgen-responsive element Gene activation activation of AP2 9.22E-03 SP1, TFAP2A, YBX1 Expression binding site Gene activation activation of gene 2.79E-02 GABPA, NAP1L1, NKX2-5, PIAS1, PIAS3, SMARCB1, SP1, STAT3, TBP, Expression TP53BP1, TXNRD1 Gene activation activation of ATF 3.10E-02 ELK1, SRF Expression binding site Gene activation activation of p53 4.47E-02 SP1, YBX1 Expression consensus binding site Gene transactivatio transactivation 5.88E-04 AKAP13, ASXL1, BRCA1, BTRC, CDC42, CDK7, CRTC2, ELK1, ETS1, Expression n FOXM1, GABPA, GABPB1, GNA13, HDAC3, HIF1A, ISG20L2, JAZF1, LMO4, MAML1, MAZ, MCM5, MED1, MSX2, NAP1L1, NCOA3, NCOA4, NFATC3, NKX2-5, NME1, PIAS1, PIAS2, POLR2A, PPARD, PPARG, PRKCQ, RARA, SMARCB1, SP1, SRF, STAT3, STAT5B, TBP, TP53BP1, TXN Gene transactivatio transactivation of 1.01E-02 GABPA, GABPB1 Expression n PEA3/ets binding site Gene elongation elongation of RNA 1.09E-03 ADRM1, CDK7, ERCC3, GTF2F2, POLR2A, POLR2C, POLR2D, Expression POLR2G, TBP Gene repression repression of DNA 2.87E-03 ATF1, BRCA1, PIAS3, TFAP2A Expression Gene gene gene silencing 3.60E-03 CHMP1A, TRIM24, TRIM33 Expression silencing Gene gene gene silencing of 1.93E-02 TRIM24, TRIM33 Expression silencing gene Gene expression expression of cAMP 4.10E-03 ATF4, CRTC2, PRKAR1A, RB1 Expression response element Gene expression expression of 7.80E-03 ATF1, ATF3, BMPR2, BRCA1, CDKN1A, CUX1, DAB2, HIF1A, HMGA1, Expression synthetic promoter MAML1, MED1, NCOA3, NCOA4, OTUB1, PPARG, PRKCQ, PYGO2, RARA, RB1, RBBP8, SOX11, STAT3, TAB2, TBP, TFDP1, TXN Gene regulation regulation of Osf1 1.01E-02 SP1, SP2 Expression binding site Gene binding binding of p53 3.03E-02 ANXA7, BRCA1, MED1 Expression consensus binding site Cell Cycle cell division cell division process 1.46E-10 ACLY, ASXL1, ATF3, AURKA, BABAM1, BAX, BMP2, BRCA1, BTRC, process C13orf15, CAT, CBL, CD2AP, CDC42, CDCA5, CDK6, CDK7, CDKN1A, CDKN1B, CDKN3, CEP55, CETN3, CHEK2, CHMP1A, COPS5, CSNK1E, CSNK2A1, CUL2, DDB1, DIAPH2, DSN1, DYRK2, EIF4G2, ELF4, ERCC3, ETS1, FBXO5, FOXM1, FOXN3, FOXO3, FYN, GADD45A, GNA13, GNAI2, GNL3, GPS2, HAUS1, HCFC1, HDAC3, HIF1A, HINFP, HIPK2, HMGA1, HMOX1, KIF3B, KIFC1, MAPK7, MASTL, MCM7, MELK, MPHOSPH6, NAMPT, NBN, NCAPD2, NCAPG (includes EG:64151), NCOA3, NEDD1, NEDD4L, NME1, NRAS, PA2G4, PCBP4, PDK3, PEX11B, PHB (includes EG:5245), PHF13, PIAS1, PLAGL1, POLD1, PPARG, PPP2R2A, PRCC, PRDM4, PRMT5, PSMB5, RAD17, RB1, RB1CC1, RBBP8, SEPT6, SEPT9, SETD8, SGOL1, SKA1, SKA3, SKP2, SMARCB1, SP1, STAT3, TFDP1, THAP1, TOP2A, TP53BP1, TPD52L1, TRIM33, TSC2, TUSC2, TXN, UBE2C, UBIAD1, VAV3, WNK1, YWHAE Cell Cycle cell division cell division process 4.69E-06 ASXL1, AURKA, BAX, BMP2, BRCA1, BTRC, CD2AP, CDC42, CDK6, process of tumor cell lines CDKN1A, CDKN1B, CEP55, CHEK2, COPS5, CSNK2A1, FBXO5, FOXM1, FOXO3, FYN, GADD45A, GNAI2, GNL3, GPS2, HIF1A, HMGA1, HMOX1, KIFC1, MAPK7, MASTL, MCM7, MELK, NCAPD2, NCOA3, NEDD4L, PCBP4, PHB (includes EG:5245), PIAS1, PLAGL1, PPARG, PPP2R2A, RAD17, RB1, RB1CC1, RBBP8, SEPT6, SEPT9, SKP2, SMARCB1, SP1, STAT3, TOP2A, TP53BP1, TRIM33, TUSC2, VAV3, YWHAE Cell Cycle cell division cell division process 7.67E-06 AURKA, BRCA1, CDC42, CDK7, CDKN1A, CSNK1E, DSN1, DYRK2, process of chromosomes HIF1A, NCAPD2, NCAPG (includes EG:64151), NME1, NRAS, PDK3, PHF13, SGOL1, SKA1, SKA3, TOP2A, UBIAD1, WNK1 Cell Cycle cell division cell division process 1.17E-05 BRCA1, BTRC, CDKN1A, CDKN1B, CHEK2, FOXM1, HMOX1, MCM7, process of lung cancer cell PCBP4, PIAS1, RAD17, RB1, TUSC2, YWHAE lines Cell Cycle cell division cell division process 1.27E-05 ASXL1, AURKA, BAX, BMP2, BRCA1, BTRC, CD2AP, CDC42, CDK6,

178 process of cell lines CDKN1A, CDKN1B, CEP55, CHEK2, COPS5, CSNK2A1, DDB1, FBXO5, FOXM1, FOXO3, FYN, GADD45A, GNAI2, GNL3, GPS2, HIF1A, HIPK2, HMGA1, HMOX1, KIFC1, MAPK7, MASTL, MCM7, MELK, NCAPD2, NCOA3, NEDD4L, NME1, PCBP4, PHB (includes EG:5245), PIAS1, PLAGL1, PPARG, PPP2R2A, RAD17, RB1, RB1CC1, RBBP8, SEPT6, SEPT9, SKP2, SMARCB1, SP1, STAT3, TFDP1, TOP2A, TP53BP1, TRIM33, TUSC2, VAV3, YWHAE Cell Cycle cell division cell division process 5.31E-05 ASXL1, AURKA, BAX, BMP2, BRCA1, BTRC, C13orf15, CD2AP, process of eukaryotic cells CDC42, CDK6, CDKN1A, CDKN1B, CEP55, CHEK2, COPS5, CSNK2A1, DDB1, FBXO5, FOXM1, FOXO3, FYN, GADD45A, GNAI2, GNL3, GPS2, HIF1A, HIPK2, HMGA1, HMOX1, KIFC1, MAPK7, MASTL, MCM7, MELK, NBN, NCAPD2, NCOA3, NEDD4L, NME1, PCBP4, PHB (includes EG:5245), PIAS1, PLAGL1, PPARG, PPP2R2A, RAD17, RB1, RB1CC1, RBBP8, SEPT6, SEPT9, SKP2, SMARCB1, SP1, STAT3, TFDP1, TOP2A, TP53BP1, TRIM33, TUSC2, TXN, VAV3, YWHAE Cell Cycle cell division arrest in cell division 8.29E-05 BRCA1, BTRC, CDKN1A, CDKN1B, CHEK2, PCBP4, PIAS1, RB1, process process of lung TUSC2, YWHAE cancer cell lines Cell Cycle cell division cell division process 2.00E-04 ASXL1, AURKA, CD2AP, CDC42, CDKN1B, CEP55, CSNK2A1, FBXO5, process of cervical cancer FOXM1, GNAI2, KIFC1, MAPK7, MASTL, NCAPD2, NEDD4L, cell lines PPP2R2A, RB1, SEPT6, SEPT9, TOP2A, VAV3 Cell Cycle cell division arrest in cell division 9.60E-04 AURKA, BMP2, BRCA1, BTRC, CDC42, CDKN1A, CDKN1B, CDKN3, process process CHEK2, CUL2, DDB1, EIF4G2, FBXO5, FOXM1, FOXO3, GADD45A, GNL3, GPS2, HIPK2, MASTL, MELK, NBN, PA2G4, PCBP4, PIAS1, PLAGL1, PPARG, PPP2R2A, RB1, RB1CC1, SEPT9, SGOL1, SKP2, SMARCB1, SP1, TP53BP1, TUSC2, YWHAE Cell Cycle cell division entry into cell 1.06E-03 AURKA, CDKN1A, CDKN1B, FOXM1, HMOX1, MAPK7, NCAPD2, process division process of NCOA3, RB1, SKP2, SMARCB1 cell lines Cell Cycle cell division arrest in cell division 1.21E-03 CDC42, CDKN1A, FOXM1, GNL3, GPS2, PLAGL1, RB1, TP53BP1 process process of bone cancer cell lines Cell Cycle cell division entry into cell 1.24E-03 AURKA, C13orf15, CDKN1A, CDKN1B, CHEK2, FOXM1, HMOX1, process division process of MAPK7, NCAPD2, NCOA3, RB1, SKP2, SMARCB1 eukaryotic cells Cell Cycle cell division cell division process 1.77E-03 BRCA1, CDKN1A, CDKN1B, CHEK2, FOXM1, HMOX1, MCM7, RAD17, process of carcinoma cell RB1 lines Cell Cycle cell division arrest in cell division 2.20E-03 AURKA, BMP2, BRCA1, BTRC, CDC42, CDKN1A, CDKN1B, CHEK2, process process of tumor cell FBXO5, FOXM1, FOXO3, GADD45A, GNL3, GPS2, MASTL, MELK, lines PCBP4, PIAS1, PLAGL1, PPARG, PPP2R2A, RB1, RB1CC1, SMARCB1, SP1, TP53BP1, TUSC2, YWHAE Cell Cycle cell division delay in cell division 2.22E-03 AURKA, BAX, BRCA1, CDC42, CDKN1A, CHEK2, FOXM1, NCOA3, process process TFDP1, TOP2A Cell Cycle cell division entry into cell 2.43E-03 AURKA, CDKN1B, FOXM1, MAPK7, NCAPD2 process division process of cervical cancer cell lines Cell Cycle cell division delay in cell division 2.74E-03 AURKA, BAX, BRCA1, CDC42, CDKN1A, CHEK2, NCOA3, TFDP1, process process of cell lines TOP2A Cell Cycle cell division entry into cell 2.74E-03 AURKA, CDKN1B, FOXM1, HMOX1, MAPK7, NCAPD2, RB1, SKP2, process division process of SMARCB1 tumor cell lines Cell Cycle cell division arrest in cell division 2.87E-03 CDKN1A, GADD45A, NBN, SKP2 process process of fibroblasts Cell Cycle cell division arrest in cell division 3.08E-03 AURKA, BMP2, BRCA1, BTRC, CDC42, CDKN1A, CDKN1B, CHEK2, process process of eukaryotic DDB1, FBXO5, FOXM1, FOXO3, GADD45A, GNL3, GPS2, HIPK2, cells MASTL, MELK, NBN, PCBP4, PIAS1, PLAGL1, PPARG, PPP2R2A, RB1, RB1CC1, SKP2, SMARCB1, SP1, TP53BP1, TUSC2, YWHAE Cell Cycle cell division delay in cell division 3.49E-03 CHEK2, TFDP1 process process of epithelial cell lines Cell Cycle cell division arrest in cell division 4.75E-03 AURKA, BMP2, BRCA1, BTRC, CDC42, CDKN1A, CDKN1B, CHEK2, process process of cell lines DDB1, FBXO5, FOXM1, FOXO3, GADD45A, GNL3, GPS2, HIPK2, MASTL, MELK, PCBP4, PIAS1, PLAGL1, PPARG, PPP2R2A, RB1, RB1CC1, SMARCB1, SP1, TP53BP1, TUSC2, YWHAE Cell Cycle cell division cell division process 6.03E-03 BAX, CDKN1A, CDKN1B process of skin cancer cell lines

179 Cell Cycle cell division cell division process 6.58E-03 CDKN1A, CDKN1B, NCOA3, RB1, SMARCB1 process of fibroblast cell lines Cell Cycle cell division exit from cell 9.22E-03 CDKN1A, CDKN1B, FOXO3 process division process of colon cancer cell lines Cell Cycle cell division cell division process 9.23E-03 CDC42, CDK6, CDKN1A, FOXM1, GNL3, GPS2, PLAGL1, RB1, process of bone cancer cell TP53BP1 lines Cell Cycle cell division cell division process 1.23E-02 CDKN1B, CHEK2, DDB1, FBXO5, HIPK2, NME1, TFDP1 process of epithelial cell lines Cell Cycle cell division cell division process 1.32E-02 CDKN1B, FOXM1, SEPT9 process of endothelial cell lines Cell Cycle cell division cell division process 1.56E-02 COPS5, FOXM1, HMGA1, STAT3 process of pancreatic cancer cell lines Cell Cycle cell division exit from cell 1.56E-02 CDKN1A, CDKN1B, FOXO3, MASTL process division process of tumor cell lines Cell Cycle cell division arrest in cell division 2.03E-02 BRCA1, CDKN1A, CDKN1B, CHEK2, GADD45A, PPARG process process of colon cancer cell lines Cell Cycle cell division arrest in cell division 2.29E-02 CDKN1A, GADD45A, NBN, RB1, SKP2, SP1 process process of normal cells Cell Cycle cell division cell division process 2.29E-02 CDKN1A, CHEK2, GADD45A, NBN, SKP2, TXN process of connective tissue cells Cell Cycle cell division arrest in cell division 2.38E-02 CDKN1A, CDKN1B, SMARCB1 process process of fibroblast cell lines Cell Cycle cell division delay in cell division 2.42E-02 AURKA, BAX, BRCA1, CDC42, CDKN1A, NCOA3, TOP2A process process of tumor cell lines Cell Cycle cell division cell division process 2.55E-02 BRCA1, CDK6, CDKN1A, CDKN1B, CHEK2, FOXO3, GADD45A, process of colon cancer cell PPARG, RAD17 lines Cell Cycle cell division exit from cell 2.97E-02 CDKN1A, CDKN1B, FOXO3, MASTL, PPARG process division process Cell Cycle cell division entry into cell 3.10E-02 C13orf15, MAPK7 process division process of smooth muscle cells Cell Cycle cell division arrest in cell division 3.25E-02 CDKN1B, DDB1, FBXO5, HIPK2 process process of embryonic cell lines Cell Cycle cell division cell division process 4.29E-02 CDKN1B, DDB1, FBXO5, HIPK2, TFDP1 process of embryonic cell lines Cell Cycle cell division arrest in cell division 4.38E-02 CDKN1B, DDB1, FBXO5, HIPK2 process process of epithelial cell lines Cell Cycle cell division entry into cell 4.38E-02 C13orf15, CDKN1A, CHEK2, MAPK7 process division process of normal cells Cell Cycle cell stage cell stage 4.75E-09 AURKA, BABAM1, BAX, BMP2, BRCA1, BTRC, C13orf15, CBL, CD2AP, CDC42, CDCA5, CDK6, CDKN1A, CDKN1B, CDKN3, CEP55, CHEK2, CHMP1A, COPS5, CUL2, DDB1, DIAPH2, ELF4, FBXO5, FOXM1, FOXN3, FOXO3, FYN, GADD45A, GNA13, GNAI2, GNL3, GPS2, HAUS1, HDAC3, HIF1A, HINFP, HMGA1, HMOX1, KIFC1, MAPK7, MASTL, MPHOSPH6, NBN, NCAPD2, NCOA3, NEDD4L, NME1, PCBP4, PHF13, PIAS1, POLD1, PPARG, PPP2R2A, PRMT5, RAD17, RB1, RB1CC1, RBBP8, SEPT6, SEPT9, SETD8, SGOL1, SKA1, SKA3, SKP2, SMARCB1, SP1, TFDP1, THAP1, TIPRL, TOP2A, TP53BP1, TPD52L1, TRIM33, TUSC2, TXN, UBE2C, VAV3, YWHAE Cell Cycle cell stage cell stage of cell lines 2.29E-06 AURKA, BAX, BMP2, BRCA1, BTRC, CD2AP, CDC42, CDK6, CDKN1A, CDKN1B, CEP55, CHEK2, COPS5, DDB1, FBXO5, FOXM1, FOXO3, GADD45A, GNAI2, GNL3, GPS2, HIF1A, HMGA1, HMOX1, KIFC1, MAPK7, MASTL, NCAPD2, NCOA3, NEDD4L, NME1, PCBP4, PIAS1,

180 PPARG, PPP2R2A, RAD17, RB1, RB1CC1, RBBP8, SEPT6, SEPT9, SKP2, SMARCB1, SP1, TFDP1, TOP2A, TP53BP1, TUSC2, VAV3, YWHAE Cell Cycle cell stage cell stage of tumor 2.67E-06 AURKA, BAX, BMP2, BRCA1, BTRC, CD2AP, CDC42, CDK6, CDKN1A, cell lines CDKN1B, CEP55, CHEK2, COPS5, FBXO5, FOXM1, FOXO3, GADD45A, GNAI2, GNL3, GPS2, HIF1A, HMGA1, HMOX1, KIFC1, MAPK7, MASTL, NCAPD2, NCOA3, NEDD4L, PCBP4, PIAS1, PPP2R2A, RAD17, RB1, RB1CC1, RBBP8, SEPT6, SEPT9, SKP2, SMARCB1, SP1, TOP2A, TP53BP1, TUSC2, VAV3, YWHAE Cell Cycle cell stage cell stage of 5.33E-06 AURKA, BAX, BMP2, BRCA1, BTRC, C13orf15, CD2AP, CDC42, CDK6, eukaryotic cells CDKN1A, CDKN1B, CEP55, CHEK2, COPS5, DDB1, FBXO5, FOXM1, FOXO3, GADD45A, GNAI2, GNL3, GPS2, HIF1A, HMGA1, HMOX1, KIFC1, MAPK7, MASTL, NBN, NCAPD2, NCOA3, NEDD4L, NME1, PCBP4, PIAS1, PPARG, PPP2R2A, RAD17, RB1, RB1CC1, RBBP8, SEPT6, SEPT9, SKP2, SMARCB1, SP1, TFDP1, TOP2A, TP53BP1, TUSC2, VAV3, YWHAE Cell Cycle cell stage cell stage of lung 1.29E-05 BRCA1, BTRC, CDKN1A, CDKN1B, CHEK2, FOXM1, HMOX1, PCBP4, cancer cell lines PIAS1, RB1, TUSC2, YWHAE Cell Cycle cell stage entry into cell stage 2.71E-04 AURKA, C13orf15, CDKN1A, CDKN1B, CHEK2, FOXM1, HMOX1, of eukaryotic cells MAPK7, NCAPD2, NCOA3, RB1, SKP2, SMARCB1 Cell Cycle cell stage cell stage of cervical 3.93E-04 AURKA, CD2AP, CDC42, CDKN1B, CEP55, FBXO5, FOXM1, GNAI2, cancer cell lines KIFC1, MAPK7, MASTL, NCAPD2, NEDD4L, PPP2R2A, SEPT6, SEPT9, TOP2A, VAV3 Cell Cycle cell stage entry into cell stage 4.37E-04 AURKA, CDKN1A, CDKN1B, FOXM1, HMOX1, MAPK7, NCAPD2, of cell lines NCOA3, RB1, SKP2, SMARCB1 Cell Cycle cell stage arrest in cell stage 1.25E-03 BMP2, BRCA1, BTRC, CDC42, CDKN1A, CDKN1B, CHEK2, DDB1, FBXO5, FOXM1, FOXO3, GADD45A, GNL3, GPS2, MASTL, NBN, PCBP4, PIAS1, PPP2R2A, RB1, RB1CC1, SEPT9, SGOL1, SKP2, SMARCB1, SP1, TP53BP1, TUSC2 Cell Cycle cell stage entry into cell stage 1.51E-03 AURKA, CDKN1B, FOXM1, HMOX1, MAPK7, NCAPD2, RB1, SKP2, of tumor cell lines SMARCB1 Cell Cycle cell stage entry into cell stage 2.43E-03 AURKA, CDKN1B, FOXM1, MAPK7, NCAPD2 of cervical cancer cell lines Cell Cycle cell stage arrest in cell stage of 3.01E-03 BMP2, BRCA1, BTRC, CDC42, CDKN1A, CDKN1B, CHEK2, FBXO5, tumor cell lines FOXM1, FOXO3, GADD45A, GNL3, GPS2, MASTL, PCBP4, PIAS1, PPP2R2A, RB1, RB1CC1, SMARCB1, SP1, TP53BP1, TUSC2 Cell Cycle cell stage arrest in cell stage of 3.47E-03 BMP2, BRCA1, BTRC, CDC42, CDKN1A, CDKN1B, CHEK2, DDB1, eukaryotic cells FBXO5, FOXM1, FOXO3, GADD45A, GNL3, GPS2, MASTL, NBN, PCBP4, PIAS1, PPP2R2A, RB1, RB1CC1, SKP2, SMARCB1, SP1, TP53BP1, TUSC2 Cell Cycle cell stage cell stage of 6.03E-03 CDKN1B, FOXM1, SEPT9 endothelial cell lines Cell Cycle cell stage cell stage of normal 6.57E-03 BMP2, C13orf15, CDKN1A, CHEK2, GADD45A, MAPK7, NBN, NME1, cells RB1, SKP2, SP1 Cell Cycle cell stage arrest in cell stage of 7.45E-03 BMP2, BRCA1, BTRC, CDC42, CDKN1A, CDKN1B, CHEK2, DDB1, cell lines FBXO5, FOXM1, FOXO3, GADD45A, GNL3, GPS2, MASTL, PCBP4, PIAS1, PPP2R2A, RB1, RB1CC1, SMARCB1, SP1, TP53BP1, TUSC2 Cell Cycle cell stage cell stage of 1.01E-02 BRCA1, CDKN1B, CHEK2, FOXM1, HMOX1, RB1 carcinoma cell lines Cell Cycle cell stage cell stage of 1.18E-02 CDKN1B, CHEK2, DDB1, FBXO5, NME1, TFDP1 epithelial cell lines Cell Cycle cell stage entry into cell stage 1.56E-02 C13orf15, CDKN1A, CHEK2, MAPK7 of normal cells Cell Cycle cell stage delay in cell stage 1.76E-02 BRCA1, CDC42, CDKN1A, CHEK2, FOXM1, NCOA3, TOP2A Cell Cycle cell stage cell stage of 3.03E-02 NME1, RB1, SP1 epithelial cells Cell Cycle cell stage exit from cell stage 3.03E-02 CDKN1A, FOXO3, MASTL of tumor cell lines Cell Cycle cell stage entry into cell stage 3.10E-02 C13orf15, MAPK7 of smooth muscle cells Cell Cycle cell stage cell stage of smooth 3.78E-02 BMP2, C13orf15, MAPK7 muscle cells Cell Cycle cell stage delay in cell stage of 4.36E-02 BRCA1, CDC42, CDKN1A, CHEK2, NCOA3, TOP2A eukaryotic cells Cell Cycle cell cycle cell cycle progression 7.67E-06 ACLY, ATF3, AURKA, BAX, BMP2, BRCA1, BTRC, CAT, CBL, CDC42, progression CDCA5, CDK6, CDKN1A, CDKN1B, CDKN3, CHEK2, CSNK2A1, CUL2, EIF4G2, ETS1, FBXO5, FOXM1, FOXO3, FYN, GADD45A, GNA13, GPS2, HAUS1, HCFC1, HDAC3, HIPK2, HMOX1, MAPK7, MASTL,

181 MELK, NAMPT, NBN, NCAPD2, PA2G4, PHB (includes EG:5245), PHF13, PLAGL1, PPARG, PRDM4, PRMT5, PSMB5, RB1, SEPT9, SGOL1, SKA1, SKA3, SKP2, SMARCB1, STAT3, TFDP1, THAP1, TOP2A, TRIM33, TSC2, TXN, UBE2C, YWHAE Cell Cycle cell cycle arrest in cell cycle 1.70E-03 AURKA, BRCA1, BTRC, CDKN1A, CDKN1B, FOXM1, FOXO3, progression progression of GADD45A, HIPK2, MELK, PLAGL1, PPARG, RB1, SKP2, SMARCB1, eukaryotic cells YWHAE Cell Cycle cell cycle arrest in cell cycle 2.23E-03 AURKA, BRCA1, BTRC, CDKN1A, CDKN1B, CDKN3, CUL2, EIF4G2, progression progression FOXM1, FOXO3, GADD45A, HIPK2, MELK, NBN, PA2G4, PLAGL1, PPARG, RB1, SKP2, SMARCB1, YWHAE Cell Cycle cell cycle arrest in cell cycle 3.19E-03 AURKA, BRCA1, BTRC, CDKN1A, CDKN1B, FOXM1, FOXO3, HIPK2, progression progression of cell MELK, PLAGL1, PPARG, RB1, SMARCB1, YWHAE lines Cell Cycle cell cycle arrest in cell cycle 3.49E-03 AURKA, FOXO3 progression progression of squamous cell carcinoma cell lines Cell Cycle cell cycle arrest in cell cycle 5.01E-03 AURKA, BRCA1, CDKN1A, CDKN1B, FOXM1, FOXO3, MELK, progression progression of tumor PLAGL1, PPARG, RB1, SMARCB1, YWHAE cell lines Cell Cycle cell cycle cell cycle progression 5.09E-03 AURKA, BAX, BRCA1, BTRC, CDK6, CDKN1A, CDKN1B, CSNK2A1, progression of cell lines FOXM1, FOXO3, HIPK2, HMOX1, MELK, PHB (includes EG:5245), PLAGL1, PPARG, RB1, SMARCB1, STAT3, TFDP1, YWHAE Cell Cycle cell cycle delay in cell cycle 5.65E-03 AURKA, BAX, CDKN1A, TFDP1 progression progression of cell lines Cell Cycle cell cycle cell cycle progression 7.41E-03 AURKA, BAX, BRCA1, BTRC, CDK6, CDKN1A, CDKN1B, CSNK2A1, progression of eukaryotic cells FOXM1, FOXO3, GADD45A, HIPK2, HMOX1, MELK, PHB (includes EG:5245), PLAGL1, PPARG, RB1, SKP2, SMARCB1, STAT3, TFDP1, YWHAE Cell Cycle cell cycle cell cycle progression 8.84E-03 AURKA, BAX, BRCA1, CDK6, CDKN1A, CDKN1B, CSNK2A1, FOXM1, progression of tumor cell lines FOXO3, HMOX1, MELK, PHB (includes EG:5245), PLAGL1, PPARG, RB1, SMARCB1, STAT3, YWHAE Cell Cycle cell cycle arrest in cell cycle 1.01E-02 CDKN1A, SKP2 progression progression of fibroblasts Cell Cycle cell cycle arrest in cell cycle 1.01E-02 CDKN1A, CDKN1B progression progression of skin cancer cell lines Cell Cycle cell cycle arrest in cell cycle 1.32E-02 CDKN1A, RB1, YWHAE progression progression of lung cancer cell lines Cell Cycle cell cycle cell cycle progression 1.81E-02 CDKN1A, RB1, SMARCB1 progression of fibroblast cell lines Cell Cycle cell cycle cell cycle progression 2.31E-02 CDKN1A, HMOX1, RB1, YWHAE progression of lung cancer cell lines Cell Cycle cell cycle delay in cell cycle 3.03E-02 AURKA, BAX, CDKN1A progression progression of tumor cell lines Cell Cycle cell cycle arrest in cell cycle 3.10E-02 CDKN1A, PPARG progression progression of colon cancer cell lines Cell Cycle cell cycle cell cycle progression 3.78E-02 CDK6, CDKN1A, PPARG progression of colon cancer cell lines Cell Cycle cell cycle arrest in cell cycle 3.79E-02 BRCA1, FOXM1, FOXO3, RB1 progression progression of breast cancer cell lines Cell Cycle cell cycle arrest in cell cycle 4.47E-02 PLAGL1, RB1 progression progression of bone cancer cell lines Cell Cycle cell cycle arrest in cell cycle 4.47E-02 CDKN1A, SMARCB1 progression progression of fibroblast cell lines Cell Cycle cell cycle delay in cell cycle 4.47E-02 BAX, CDKN1A progression progression of lymphoblastoid cell lines

182 Cell Cycle cytokinesis cytokinesis of cell 1.05E-05 CD2AP, CEP55, GNAI2, KIFC1, MASTL, NEDD4L, NME1, SEPT6, lines SEPT9, TOP2A, VAV3 Cell Cycle cytokinesis cytokinesis of 2.90E-05 CD2AP, CEP55, GNAI2, KIFC1, MASTL, NEDD4L, SEPT6, SEPT9, cervical cancer cell VAV3 lines Cell Cycle cytokinesis cytokinesis of tumor 5.15E-05 CD2AP, CEP55, GNAI2, KIFC1, MASTL, NEDD4L, SEPT6, SEPT9, cell lines TOP2A, VAV3 Cell Cycle cytokinesis cytokinesis 5.98E-05 AURKA, CD2AP, CDKN1A, CEP55, DIAPH2, GNAI2, KIFC1, MASTL, NEDD4L, NME1, SEPT6, SEPT9, SETD8, TOP2A, VAV3 Cell Cycle M phase M phase of cervical 2.05E-05 CD2AP, CDC42, CEP55, GNAI2, KIFC1, MASTL, NCAPD2, NEDD4L, cancer cell lines SEPT6, SEPT9, TOP2A, VAV3 Cell Cycle M phase M phase 4.27E-05 AURKA, CD2AP, CDC42, CDKN1A, CEP55, DIAPH2, FBXO5, GNAI2, KIFC1, MASTL, MPHOSPH6, NCAPD2, NEDD4L, NME1, SEPT6, SEPT9, SETD8, SKA1, TOP2A, VAV3 Cell Cycle M phase M phase of cell lines 4.99E-05 CD2AP, CDC42, CEP55, FBXO5, GNAI2, KIFC1, MASTL, NCAPD2, NEDD4L, NME1, SEPT6, SEPT9, TOP2A, VAV3 Cell Cycle M phase M phase of tumor 1.28E-04 CD2AP, CDC42, CEP55, FBXO5, GNAI2, KIFC1, MASTL, NCAPD2, cell lines NEDD4L, SEPT6, SEPT9, TOP2A, VAV3 Cell Cycle M phase M phase of epithelial 1.01E-02 FBXO5, NME1 cell lines Cell Cycle interphase interphase 2.56E-05 BABAM1, BAX, BMP2, BRCA1, BTRC, C13orf15, CDC42, CDCA5, CDK6, CDKN1A, CDKN1B, CDKN3, CHEK2, CHMP1A, COPS5, CUL2, DDB1, ELF4, FOXM1, FOXN3, FOXO3, GADD45A, GNL3, GPS2, HINFP, HMGA1, HMOX1, MASTL, NBN, NCOA3, PCBP4, PIAS1, POLD1, PPARG, PPP2R2A, RAD17, RB1, RB1CC1, RBBP8, SKP2, SMARCB1, SP1, TFDP1, TP53BP1, TPD52L1, TUSC2 Cell Cycle interphase interphase of lung 6.56E-05 BRCA1, BTRC, CDKN1A, CDKN1B, CHEK2, HMOX1, PCBP4, PIAS1, cancer cell lines RB1, TUSC2 Cell Cycle interphase arrest in interphase of 3.37E-04 BRCA1, BTRC, CDKN1A, CDKN1B, CHEK2, PCBP4, PIAS1, TUSC2 lung cancer cell lines Cell Cycle interphase interphase of cell 4.76E-04 BAX, BMP2, BRCA1, BTRC, CDC42, CDK6, CDKN1A, CDKN1B, lines CHEK2, COPS5, DDB1, FOXM1, FOXO3, GADD45A, GNL3, GPS2, HMGA1, HMOX1, MASTL, NCOA3, PCBP4, PIAS1, PPARG, PPP2R2A, RAD17, RB1, RB1CC1, RBBP8, SKP2, SMARCB1, SP1, TFDP1, TP53BP1, TUSC2 Cell Cycle interphase interphase of tumor 5.06E-04 BAX, BMP2, BRCA1, BTRC, CDC42, CDK6, CDKN1A, CDKN1B, cell lines CHEK2, COPS5, FOXM1, FOXO3, GADD45A, GNL3, GPS2, HMGA1, HMOX1, MASTL, NCOA3, PCBP4, PIAS1, PPP2R2A, RAD17, RB1, RB1CC1, RBBP8, SKP2, SMARCB1, SP1, TP53BP1, TUSC2 Cell Cycle interphase interphase of 5.31E-04 BAX, BMP2, BRCA1, BTRC, C13orf15, CDC42, CDK6, CDKN1A, eukaryotic cells CDKN1B, CHEK2, COPS5, DDB1, FOXM1, FOXO3, GADD45A, GNL3, GPS2, HMGA1, HMOX1, MASTL, NBN, NCOA3, PCBP4, PIAS1, PPARG, PPP2R2A, RAD17, RB1, RB1CC1, RBBP8, SKP2, SMARCB1, SP1, TFDP1, TP53BP1, TUSC2 Cell Cycle interphase arrest in interphase of 1.48E-03 BMP2, BRCA1, BTRC, CDC42, CDKN1A, CDKN1B, CHEK2, DDB1, eukaryotic cells FOXM1, FOXO3, GADD45A, GNL3, GPS2, MASTL, NBN, PCBP4, PIAS1, PPP2R2A, RB1, RB1CC1, SKP2, SMARCB1, SP1, TP53BP1, TUSC2 Cell Cycle interphase arrest in interphase of 1.48E-03 BMP2, BRCA1, BTRC, CDC42, CDKN1A, CDKN1B, CHEK2, FOXM1, tumor cell lines FOXO3, GADD45A, GNL3, GPS2, MASTL, PCBP4, PIAS1, PPP2R2A, RB1, RB1CC1, SMARCB1, SP1, TP53BP1, TUSC2 Cell Cycle interphase arrest in interphase of 1.51E-03 CDC42, CDKN1A, FOXM1, GNL3, GPS2, RB1, TP53BP1 bone cancer cell lines Cell Cycle interphase interphase of bone 1.73E-03 CDC42, CDK6, CDKN1A, FOXM1, GNL3, GPS2, RB1, TP53BP1 cancer cell lines Cell Cycle interphase interphase of 1.91E-03 CDKN1A, CDKN1B, NCOA3, SMARCB1 fibroblast cell lines Cell Cycle interphase arrest in interphase of 3.40E-03 BMP2, BRCA1, BTRC, CDC42, CDKN1A, CDKN1B, CHEK2, DDB1, cell lines FOXM1, FOXO3, GADD45A, GNL3, GPS2, MASTL, PCBP4, PIAS1, PPP2R2A, RB1, RB1CC1, SMARCB1, SP1, TP53BP1, TUSC2 Cell Cycle interphase interphase of 5.26E-03 CDKN1A, CHEK2, GADD45A, NBN, SKP2 fibroblasts Cell Cycle interphase interphase of normal 5.83E-03 C13orf15, CDKN1A, CHEK2, GADD45A, NBN, RB1, SKP2, SP1 cells Cell Cycle interphase arrest in interphase of 9.86E-03 CDKN1A, GADD45A, NBN, RB1, SP1 normal cells Cell Cycle interphase arrest in interphase of 1.32E-02 CDKN1A, GADD45A, NBN fibroblasts

183 Cell Cycle interphase interphase of 1.67E-02 BRCA1, CDKN1B, CHEK2, HMOX1, RB1 carcinoma cell lines Cell Cycle interphase arrest in interphase of 1.93E-02 CDKN1B, FOXM1 endothelial cell lines Cell Cycle interphase entry into interphase 1.93E-02 CDKN1A, CHEK2 of fibroblasts Cell Cycle interphase interphase of 2.38E-02 COPS5, FOXM1, HMGA1 pancreatic cancer cell lines Cell Cycle interphase interphase of colon 2.95E-02 BRCA1, CDKN1A, CDKN1B, CHEK2, FOXO3, GADD45A, RAD17 cancer cell lines Cell Cycle interphase arrest in interphase of 2.97E-02 BRCA1, CDKN1A, CDKN1B, CHEK2, GADD45A colon cancer cell lines Cell Cycle interphase interphase of prostate 3.25E-02 CDKN1B, NCOA3, SKP2, SP1 cancer cell lines Cell Cycle interphase interphase of 3.79E-02 CDKN1B, CHEK2, DDB1, TFDP1 epithelial cell lines Cell Cycle G1 phase G1 phase 4.93E-05 BAX, BMP2, BRCA1, BTRC, CDC42, CDCA5, CDK6, CDKN1A, CDKN1B, CDKN3, CHEK2, COPS5, CUL2, ELF4, FOXM1, GADD45A, GNL3, GPS2, HINFP, HMGA1, NBN, NCOA3, PIAS1, RB1, RB1CC1, SKP2, SP1, TFDP1, TUSC2 Cell Cycle G1 phase G1 phase of tumor 7.53E-04 BAX, BMP2, BRCA1, BTRC, CDC42, CDK6, CDKN1A, CDKN1B, cell lines COPS5, FOXM1, GNL3, GPS2, HMGA1, NCOA3, PIAS1, RB1, RB1CC1, SKP2, SP1, TUSC2 Cell Cycle G1 phase G1 phase of cell lines 7.68E-04 BAX, BMP2, BRCA1, BTRC, CDC42, CDK6, CDKN1A, CDKN1B, CHEK2, COPS5, FOXM1, GNL3, GPS2, HMGA1, NCOA3, PIAS1, RB1, RB1CC1, SKP2, SP1, TFDP1, TUSC2 Cell Cycle G1 phase G1 phase of bone 1.86E-03 CDC42, CDK6, CDKN1A, FOXM1, GNL3, GPS2, RB1 cancer cell lines Cell Cycle G1 phase arrest in G1 phase of 3.49E-03 CDC42, CDKN1A, FOXM1, GNL3, GPS2, RB1 bone cancer cell lines Cell Cycle G1 phase G1 phase of skin 3.49E-03 BAX, CDKN1B cancer cell lines Cell Cycle G1 phase arrest in G1 phase of 3.81E-03 BMP2, BRCA1, BTRC, CDC42, CDKN1A, CDKN1B, FOXM1, GNL3, tumor cell lines GPS2, PIAS1, RB1, RB1CC1, SP1, TUSC2 Cell Cycle G1 phase arrest in G1 phase of 4.15E-03 BRCA1, BTRC, CDKN1A, PIAS1, TUSC2 lung cancer cell lines Cell Cycle G1 phase arrest in G1 phase of 6.92E-03 BMP2, BRCA1, BTRC, CDC42, CDKN1A, CDKN1B, FOXM1, GNL3, eukaryotic cells GPS2, PIAS1, RB1, RB1CC1, SKP2, SP1, TUSC2 Cell Cycle G1 phase arrest in G1 phase of 1.01E-02 RB1, SP1 epithelial cells Cell Cycle G1 phase arrest in G1 phase of 1.93E-02 CDKN1A, SKP2 tumor cells Cell Cycle G1 phase G1 phase of 2.38E-02 CDKN1B, CHEK2, TFDP1 epithelial cell lines Cell Cycle G1 phase G1 phase of prostate 3.78E-02 CDKN1B, NCOA3, SP1 cancer cell lines Cell Cycle G1/S phase G1/S phase transition 8.50E-05 CDC42, CDCA5, CDKN1A, CDKN1B, CDKN3, CHEK2, COPS5, CUL2, transition ELF4, FOXM1, GADD45A, HINFP, HMGA1, NBN, NCOA3, RB1, RB1CC1, TFDP1 Cell Cycle G1/S phase G1/S phase transition 3.49E-03 COPS5, HMGA1 transition of pancreatic cancer cell lines Cell Cycle G1/S phase G1/S phase transition 6.37E-03 CDC42, CDKN1A, CDKN1B, CHEK2, COPS5, FOXM1, HMGA1, transition of cell lines NCOA3, RB1CC1, TFDP1 Cell Cycle G1/S phase G1/S phase transition 2.42E-02 CDC42, CDKN1A, CDKN1B, COPS5, FOXM1, HMGA1, NCOA3, transition of tumor cell lines RB1CC1 Cell Cycle G1/S phase G1/S phase transition 3.10E-02 CHEK2, TFDP1 transition of epithelial cell lines Cell Cycle ploidy ploidy 4.37E-04 AURKA, CDK7, CDKN1A, CSNK1E, DYRK2, HIF1A, NME1, PDK3, TOP2A, UBIAD1, WNK1 Cell Cycle ploidy ploidy of cell lines 2.42E-03 CDK7, CDKN1A, CSNK1E, DYRK2, NME1, PDK3, UBIAD1, WNK1 Cell Cycle mitosis mitosis 4.81E-04 AURKA, BMP2, BRCA1, CBL, CDC42, CDCA5, CDKN1A, CDKN1B, FBXO5, FOXM1, FYN, GNA13, HAUS1, HDAC3, MAPK7, MASTL, NCAPD2, PHF13, PRMT5, RB1, SEPT9, SGOL1, SKA1, SKA3, THAP1, TOP2A, TRIM33, TXN, UBE2C, YWHAE Cell Cycle mitosis mitosis of cell lines 3.39E-03 AURKA, BMP2, BRCA1, CDKN1A, CDKN1B, FBXO5, FOXM1, MAPK7,

184 MASTL, SEPT9, TOP2A, YWHAE Cell Cycle mitosis mitosis of breast cell 3.60E-03 CDKN1A, CDKN1B, SEPT9 lines Cell Cycle mitosis entry into mitosis of 9.81E-03 AURKA, CDKN1A, CDKN1B, MAPK7 cell lines Cell Cycle mitosis entry into mitosis of 1.01E-02 CDKN1A, CDKN1B breast cell lines Cell Cycle mitosis mitosis of tumor cell 1.25E-02 AURKA, BMP2, BRCA1, CDKN1A, FBXO5, FOXM1, MAPK7, MASTL, lines TOP2A, YWHAE Cell Cycle mitosis mitosis of lung 1.93E-02 FOXM1, YWHAE cancer cell lines Cell Cycle mitosis exit from mitosis of 3.10E-02 CDKN1A, MASTL tumor cell lines Cell Cycle checkpoint checkpoint control 6.46E-04 CHEK2, DDB1, ERCC3, MCM7, PRCC, RAD17, RB1, RBBP8, TP53BP1 control Cell Cycle checkpoint checkpoint control of 2.87E-03 CHEK2, MCM7, RAD17, TP53BP1 control tumor cell lines Cell Cycle S phase entry into S phase of 1.28E-03 C13orf15, CDKN1A, CDKN1B, CHEK2, FOXM1, HMOX1, NCOA3, RB1, eukaryotic cells SKP2, SMARCB1 Cell Cycle S phase entry into S phase of 2.84E-03 CDKN1A, CDKN1B, FOXM1, HMOX1, NCOA3, RB1, SKP2, SMARCB1 cell lines Cell Cycle S phase S phase of fibroblasts 5.65E-03 CDKN1A, CHEK2, NBN, SKP2 Cell Cycle S phase initiation of S phase 1.01E-02 CDKN1A, CDKN1B of breast cancer cell lines Cell Cycle S phase entry into S phase of 1.01E-02 CDKN1B, FOXM1, HMOX1, RB1, SKP2, SMARCB1 tumor cell lines Cell Cycle S phase S phase of normal 1.19E-02 C13orf15, CDKN1A, CHEK2, NBN, SKP2 cells Cell Cycle S phase S phase 1.20E-02 C13orf15, CDKN1A, CDKN1B, CHEK2, CHMP1A, FOXM1, FOXO3, HMOX1, NBN, NCOA3, POLD1, PPARG, RB1, SKP2, SMARCB1 Cell Cycle S phase S phase of eukaryotic 1.29E-02 C13orf15, CDKN1A, CDKN1B, CHEK2, FOXM1, FOXO3, HMOX1, NBN, cells NCOA3, PPARG, RB1, SKP2, SMARCB1 Cell Cycle S phase entry into S phase of 1.93E-02 CDKN1A, CDKN1B breast cell lines Cell Cycle S phase entry into S phase of 1.93E-02 NCOA3, SMARCB1 fibroblast cell lines Cell Cycle S phase entry into S phase of 1.93E-02 CDKN1B, SKP2 prostate cancer cell lines Cell Cycle S phase initiation of S phase 2.38E-02 CDKN1A, CDKN1B, SKP2 of eukaryotic cells Cell Cycle S phase entry into S phase of 3.10E-02 HMOX1, RB1 lung cancer cell lines Cell Cycle S phase re-entry into S phase 3.78E-02 CDKN1A, PPARG, SKP2 of eukaryotic cells Cell Cycle S phase entry into S phase of 4.47E-02 HMOX1, RB1 carcinoma cell lines Cell Cycle S phase entry into S phase of 4.47E-02 CDKN1B, FOXM1 cervical cancer cell lines Cell Cycle S phase exit from S phase of 4.47E-02 CDKN1A, FOXO3 colon cancer cell lines Cell Cycle DNA DNA damage 1.78E-03 CHEK2, FOXN3, HINFP, NBN, RAD17, TIPRL damage checkpoint checkpoint Cell Cycle endoreduplic endoreduplication 1.88E-03 CDKN1A, NRAS, RB1 ation Cell Cycle endoreduplic endoreduplication of 1.01E-02 CDKN1A, RB1 ation cells Cell Cycle segregation segregation of 2.22E-03 BRCA1, CDC42, DSN1, NCAPD2, NCAPG (includes EG:64151), PHF13, chromosomes SGOL1, SKA1, SKA3, TOP2A Cell Cycle G2 phase G2 phase 2.97E-03 BABAM1, BAX, BRCA1, CDKN1A, CHEK2, DDB1, FOXM1, FOXN3, GADD45A, MASTL, NBN, PCBP4, PPP2R2A, RAD17, RBBP8, TP53BP1, TPD52L1 Cell Cycle G2 phase G2 phase of colon 3.21E-03 BRCA1, CDKN1A, CHEK2, GADD45A, RAD17 cancer cell lines

185 Cell Cycle G2 phase arrest in G2 phase of 4.10E-03 BRCA1, CDKN1A, CHEK2, GADD45A colon cancer cell lines Cell Cycle G2 phase G2 phase of tumor 4.14E-03 BAX, BRCA1, CDKN1A, CHEK2, FOXM1, GADD45A, MASTL, PCBP4, cell lines PPP2R2A, RAD17, RBBP8, TP53BP1 Cell Cycle G2 phase G2 phase of cell lines 4.19E-03 BAX, BRCA1, CDKN1A, CHEK2, DDB1, FOXM1, GADD45A, MASTL, PCBP4, PPP2R2A, RAD17, RBBP8, TP53BP1 Cell Cycle G2 phase arrest in G2 phase of 6.03E-03 CDKN1A, FOXM1, TP53BP1 bone cancer cell lines Cell Cycle G2 phase arrest in G2 phase of 1.63E-02 BRCA1, CDKN1A, CHEK2, DDB1, FOXM1, GADD45A, MASTL, PCBP4, cell lines PPP2R2A, TP53BP1 Cell Cycle G2 phase arrest in G2 phase of 1.82E-02 BRCA1, CDKN1A, CHEK2, FOXM1, GADD45A, MASTL, PCBP4, tumor cell lines PPP2R2A, TP53BP1 Cell Cycle G2 phase arrest in G2 phase of 2.38E-02 BRCA1, CHEK2, PCBP4 lung cancer cell lines Cell Cycle G2/M phase arrest in G2/M phase 3.49E-03 FOXM1, TP53BP1 of bone cancer cell lines Cell Cycle G2/M phase G2/M phase of tumor 2.03E-02 BRCA1, CHEK2, FOXM1, RAD17, RBBP8, TP53BP1 cell lines Cell Cycle G2/M phase G2/M phase 2.65E-02 BABAM1, BRCA1, CHEK2, FOXM1, NBN, RAD17, RBBP8, TP53BP1 Cell Cycle G2/M phase arrest in G2/M phase 4.38E-02 BRCA1, CHEK2, FOXM1, TP53BP1 of tumor cell lines Cell Cycle S phase S phase checkpoint 3.49E-03 MCM7, RAD17 checkpoint control of carcinoma control cell lines Cell Cycle S phase S phase checkpoint 3.49E-03 MCM7, RAD17 checkpoint control of lung control cancer cell lines Cell Cycle S phase S phase checkpoint 6.03E-03 MCM7, RAD17, TP53BP1 checkpoint control of tumor cell control lines Cell Cycle senescence senescence of cells 4.41E-03 ACLY, BRCA1, CAT, CDKN1A, CDKN1B, CHEK2, CSNK2A1, NAMPT, PSMB5, RB1, SMARCB1 Cell Cycle senescence senescence of 8.63E-03 ACLY, BRCA1, CAT, CDKN1A, CHEK2, CSNK2A1, NAMPT, PSMB5, eukaryotic cells RB1, SMARCB1 Cell Cycle senescence senescence of breast 3.10E-02 CDKN1A, RB1 cell lines Cell Cycle senescence senescence of normal 3.25E-02 BRCA1, CAT, CDKN1A, CSNK2A1 cells Cell Cycle G0/G1 phase arrest in G0/G1 phase 1.01E-02 CDKN1A, CDKN1B transition transition of fibroblast cell lines Cell Cycle G0/G1 phase arrest in G0/G1 phase 4.38E-02 CDKN1A, CDKN1B, FOXO3, SMARCB1 transition transition of tumor cell lines Cell Cycle prometaphas prometaphase 2.38E-02 CDC42, KIFC1, MASTL e Cell Cycle anaphase anaphase 3.03E-02 NCAPD2, SKA1, TOP2A Cell Cycle anaphase anaphase of cervical 3.10E-02 NCAPD2, TOP2A cancer cell lines Cell Cycle antiproliferat antiproliferative 4.47E-02 ASXL1, CDKN1B ive response response of tumor cell lines Post- ubiquitinatio ubiquitination 1.22E-07 ATG3 (includes EG:171415), BAG5, BRCA1, BTRC, CBL, CUL2, DDB1, Translational n FBXL2, FBXO3, FBXO7, FBXW11, FKBP1A, HUWE1, IRAK1, Modification LOC728622/SKP1, NDFIP2, OTUB1, PRICKLE1, RNF115, RNF135, TBC1D7, TRIM24, TRIM33, TSC1, UBE2C, UBE2E1, UBE2H, UBE3C, UBE4B, USP21 Post- ubiquitinatio ubiquitination of 2.84E-07 ATG3 (includes EG:171415), BAG5, BRCA1, BTRC, CBL, CUL2, DDB1, Translational n protein FBXL2, FBXO3, FBXO7, FBXW11, FKBP1A, HUWE1, IRAK1, Modification LOC728622/SKP1, NDFIP2, PRICKLE1, RNF115, RNF135, TBC1D7, TRIM24, TRIM33, TSC1, UBE2C, UBE2E1, UBE2H, UBE3C, UBE4B, USP21 Post- moiety moiety attachment of 3.18E-04 ATG3 (includes EG:171415), BAG5, BMP2, BMPR1A, BRCA1, BTRC, Translational attachment protein CBL, CDK6, CUL2, DAB2, DDB1, DYRK2, EIF2AK2, EPB41L3, FBXL2, Modification FBXO3, FBXO7, FBXW11, FKBP1A, FNTB, FUT8, FYN, GMFB, HUWE1, ICK, IRAK1, LOC728622/SKP1, MAPKAPK2, NDFIP2, OXSR1,

186 PAK2, PRICKLE1, PRKAA1, RARA, RNF115, RNF135, SIK3, SLC35C1, TBC1D7, TLK2, TPST2, TRIM24, TRIM33, TSC1, UBE2C, UBE2E1, UBE2H, UBE3C, UBE4B, USP21, UST, WNK1 Post- modification modification of 3.27E-04 APH1A, ATG3 (includes EG:171415), BAG4, BAG5, BMP2, BMPR1A, Translational protein BRCA1, BTRC, CBL, CDK6, CTDNEP1, CTDSP2, CUL2, DAB2, DDB1, Modification DNAJB1, DYRK2, EIF2AK2, EPB41L3, FBXL2, FBXO3, FBXO7, FBXW11, FKBP1A, FNTB, FUT8, FYN, GMFB, HDAC3, HUWE1, ICK, IRAK1, LOC728622/SKP1, MAPKAPK2, MCRS1, NDFIP2, NFYC, OXSR1, PAK2, PPME1, PPP2CA, PPP2R2A, PRICKLE1, PRKAA1, RARA, RNF115, RNF135, SEPHS1, SIK3, SLC35C1, TBC1D7, TLK2, TPST2, TRIM24, TRIM33, TSC1, UBE2C, UBE2E1, UBE2H, UBE3C, UBE4B, UBL4A, USP10, USP21, UST, WNK1 Post- activation activation of Protein 1.01E-02 CRIPAK, PAK2, PDPK1, RB1, TAB2, TRAF4 Translational kinase Modification Post- heterodimeri heterodimerization of 1.93E-02 BAK1, BAX Translational zation protein Modification Post- methylation methylation of basic 3.10E-02 PRMT5, PRMT6 Translational amino acid Modification Post- methylation methylation of 3.10E-02 PRMT5, PRMT6 Translational glutamine family Modification amino acid Post- sulfation sulfation of protein 3.10E-02 TPST2, UST Translational Modification Cell Death cell death cell death 4.88E-07 ABCE1, ACACA, ADNP, ADRM1, AGPAT2, AHSA1, AKAP13, ANXA7, APH1A, APIP, ARHGDIA, ATF1, ATF3, ATG5, ATMIN, ATN1, ATXN2, AURKA, B4GALT5, BACH2 (includes EG:60468), BAG4, BAG5, BAK1, BARD1, BAX, BCL2L1, BCL2L13, BLCAP, BMP2, BRCA1, BTRC, CABLES2, CAPNS1, CARD8, CAT, CDC42, CDK6, CDKN1A, CDKN1B, CERS5, CHEK2, COPS5, CSNK1E, CSNK2A1, CTNNBL1, CYFIP2 (includes EG:26999), DAB2, DFFA, DNAJB1, DNMT3B, DUSP22, EIF2AK2, EIF4G2, ELK1, ERCC3, ETS1, EXOC2, FDFT1, FKBP1A, FKBP5, FOXM1, FOXO3, GABPA, GABPB1, GADD45A, GMEB1, GPS2, GTF2F2, HAUS1, HDAC3, HEY1, HIF1A, HIPK2, HMGA1, HMOX1, HUWE1, IMMT, LIMS1, MAML1, MAP2K4, MAP3K1, MAPK7, MED1, MELK, MFN2, MSN, MT1X, MTMR9, NAMPT, NBN, NCKAP1, NCOA3, NET1, NKX2-5, NME1, NRAS, NUAK1, ODC1, OGFOD1, PAK2, PCBP2, PDCD6IP, PDPK1, PHB (includes EG:5245), PIAS1, PLAGL1, POLR2A, PPARD, PPARG, PPP2CA, PPP2R2A, PPP3R1, PRKAA1, PRKAR1A, PRKCQ, PSME3, PTPN13, RARA, RB1, RB1CC1, RBM4B, RDX, RND3, RNF216, RRAGA, RYBP, SAP30BP, SEC61G, SET, SGPL1, SH3RF1, SKP2, SMARCB1, SP1, STAT3, STAT5B, TCF12, TFAP2A, TFDP1, THAP1, TNFRSF21, TOP2A, TPD52L1, TRAF4, TRIAP1, TRIM24, TRIM28, TSC2, TTF1, TUBA1A, TXN, TXNRD1, UBE2C, UBE4B, UCP2, UGCG, VAV3, YBX1, YWHAE, YWHAZ, ZMYM2 Cell Death cell death cell death of tumor 2.38E-06 ABCE1, ACACA, ADNP, ADRM1, AGPAT2, AHSA1, ARHGDIA, ATF1, cell lines ATF3, ATG5, ATMIN, AURKA, BACH2 (includes EG:60468), BAK1, BARD1, BAX, BCL2L1, BCL2L13, BMP2, BRCA1, BTRC, CABLES2, CAPNS1, CARD8, CAT, CDC42, CDK6, CDKN1A, CDKN1B, CERS5, CHEK2, CSNK2A1, DAB2, DFFA, DNAJB1, EIF2AK2, EIF4G2, ELK1, ETS1, EXOC2, FDFT1, FKBP1A, FKBP5, FOXO3, GABPA, GADD45A, GPS2, HDAC3, HIF1A, HIPK2, HMGA1, HMOX1, HUWE1, IMMT, LIMS1, MAML1, MAP2K4, MAPK7, MELK, MSN, MT1X, MTMR9, NAMPT, NCOA3, NME1, NRAS, NUAK1, OGFOD1, PAK2, PCBP2, PDCD6IP, PDPK1, PHB (includes EG:5245), PIAS1, PLAGL1, PPARD, PPARG, PPP2CA, PPP2R2A, PRKAA1, PRKAR1A, PTPN13, RARA, RB1, RBM4B, RDX, SEC61G, SH3RF1, SKP2, SP1, STAT3, TFAP2A, TNFRSF21, TOP2A, TRAF4, TRIAP1, TRIM28, TSC2, TTF1, TUBA1A, TXN, UBE2C, UCP2, UGCG, YBX1, YWHAE, YWHAZ Cell Death cell death cell death of cell lines 7.73E-06 ABCE1, ACACA, ADNP, ADRM1, AGPAT2, AHSA1, ARHGDIA, ATF1, ATF3, ATG5, ATMIN, AURKA, BACH2 (includes EG:60468), BAG4, BAK1, BARD1, BAX, BCL2L1, BCL2L13, BMP2, BRCA1, BTRC, CABLES2, CAPNS1, CARD8, CAT, CDC42, CDK6, CDKN1A, CDKN1B, CERS5, CHEK2, CSNK2A1, DAB2, DFFA, DNAJB1, EIF2AK2, EIF4G2, ELK1, ETS1, EXOC2, FDFT1, FKBP1A, FKBP5, FOXO3, GABPA, GADD45A, GPS2, HDAC3, HIF1A, HIPK2, HMGA1, HMOX1, HUWE1,

187 IMMT, LIMS1, MAML1, MAP2K4, MAPK7, MELK, MSN, MT1X, MTMR9, NAMPT, NBN, NCOA3, NME1, NRAS, NUAK1, OGFOD1, PAK2, PCBP2, PDCD6IP, PDPK1, PHB (includes EG:5245), PIAS1, PLAGL1, PPARD, PPARG, PPP2CA, PPP2R2A, PPP3R1, PRKAA1, PRKAR1A, PRKCQ, PTPN13, RARA, RB1, RBM4B, RDX, RNF216, SEC61G, SGPL1, SH3RF1, SKP2, SP1, STAT3, STAT5B, TCF12, TFAP2A, TFDP1, TNFRSF21, TOP2A, TRAF4, TRIAP1, TRIM28, TSC2, TTF1, TUBA1A, TXN, UBE2C, UCP2, UGCG, YBX1, YWHAE, YWHAZ Cell Death cell death cell death of 3.42E-05 ABCE1, ACACA, ADNP, ADRM1, AGPAT2, AHSA1, APIP, ARHGDIA, eukaryotic cells ATF1, ATF3, ATG5, ATMIN, ATXN2, AURKA, B4GALT5, BACH2 (includes EG:60468), BAG4, BAG5, BAK1, BARD1, BAX, BCL2L1, BCL2L13, BMP2, BRCA1, BTRC, CABLES2, CAPNS1, CARD8, CAT, CDC42, CDK6, CDKN1A, CDKN1B, CERS5, CHEK2, COPS5, CSNK2A1, DAB2, DFFA, DNAJB1, DNMT3B, EIF2AK2, EIF4G2, ELK1, ETS1, EXOC2, FDFT1, FKBP1A, FKBP5, FOXO3, GABPA, GADD45A, GPS2, GTF2F2, HDAC3, HIF1A, HIPK2, HMGA1, HMOX1, HUWE1, IMMT, LIMS1, MAML1, MAP2K4, MAPK7, MELK, MFN2, MSN, MT1X, MTMR9, NAMPT, NBN, NCOA3, NKX2-5, NME1, NRAS, NUAK1, OGFOD1, PAK2, PCBP2, PDCD6IP, PDPK1, PHB (includes EG:5245), PIAS1, PLAGL1, PPARD, PPARG, PPP2CA, PPP2R2A, PPP3R1, PRKAA1, PRKAR1A, PRKCQ, PTPN13, RARA, RB1, RBM4B, RDX, RND3, RNF216, SEC61G, SET, SGPL1, SH3RF1, SKP2, SP1, STAT3, STAT5B, TCF12, TFAP2A, TFDP1, TNFRSF21, TOP2A, TRAF4, TRIAP1, TRIM28, TSC2, TTF1, TUBA1A, TXN, UBE2C, UCP2, UGCG, YBX1, YWHAE, YWHAZ Cell Death cell death cell death of cervical 1.68E-04 ADRM1, ATG5, BAK1, BAX, BCL2L1, CAT, CDKN1A, CDKN1B, cancer cell lines CSNK2A1, EIF2AK2, EIF4G2, EXOC2, GADD45A, IMMT, LIMS1, MSN, MTMR9, NME1, OGFOD1, PAK2, PCBP2, PDCD6IP, PDPK1, PPP2R2A, RB1, SEC61G, TFAP2A, TNFRSF21, TOP2A, TSC2, TXN, UCP2 Cell Death cell death cell death of bone 1.90E-04 ATG5, ATMIN, BAX, BMP2, BRCA1, CABLES2, CAPNS1, CDK6, cancer cell lines CDKN1A, ETS1, GPS2, HMGA1, HUWE1, PIAS1, PLAGL1, PRKAA1, RB1, TRAF4 Cell Death cell death cell death of colon 1.07E-03 AHSA1, ATF3, ATMIN, BAK1, BAX, BCL2L1, CDKN1A, CDKN1B, cancer cell lines CHEK2, ETS1, EXOC2, FOXO3, GABPA, GADD45A, HDAC3, HIPK2, HMGA1, LIMS1, MAML1, NUAK1, PPARD, PPARG, TRIAP1, UBE2C, YWHAE Cell Death cell death cell death of 1.91E-03 BAX, BCL2L1, HMOX1, PPP2CA unspecified cell lines Cell Death cell death cell death of 2.38E-03 BAX, BCL2L1, CDKN1A, LIMS1, NAMPT, PDPK1, TOP2A, TRIM28, fibrosarcoma cell YBX1 lines Cell Death cell death cell death of tumor 3.05E-03 ATXN2, B4GALT5, BAX, BCL2L1, BRCA1, CARD8, CAT, CDKN1A, cells CDKN1B, DNMT3B, HIF1A, HMOX1, NCOA3, PPP2CA, RARA, STAT3, TFAP2A, TXN Cell Death cell death cell death of cancer 1.21E-02 ATXN2, B4GALT5, BAX, BCL2L1, BRCA1, CAT, CDKN1A, CDKN1B, cells DNMT3B, HMOX1, NCOA3, RARA, TFAP2A, TXN Cell Death cell death cell death of 1.71E-02 ADNP, ATG5, BAK1, BCL2L1, CAT, CDC42, CDKN1A, DNAJB1, neuroblastoma cell EIF2AK2, FKBP1A, TUBA1A, TXN, UGCG lines Cell Death cell death cell death of muscle 1.76E-02 APIP, BCL2L1, BMP2, CAT, NKX2-5, RB1, STAT3 cells Cell Death cell death cell death of breast 1.76E-02 ARHGDIA, ATG5, BAK1, BARD1, BAX, BCL2L1, BCL2L13, BRCA1, cancer cell lines CDC42, CDKN1A, CERS5, DAB2, ELK1, EXOC2, FOXO3, GADD45A, HIF1A, NCOA3, PHB (includes EG:5245), PPARG, PTPN13, SH3RF1, SP1, TFAP2A, UGCG Cell Death cell death cell death of prostate 2.89E-02 ABCE1, ACACA, BAK1, BAX, BCL2L1, BMP2, BRCA1, CAT, FDFT1, cancer cell lines FOXO3, GADD45A, HIF1A, NCOA3, RB1, STAT3, TXN Cell Death cell death cell death of 3.06E-02 ATG5, BAG4, BAX, BCL2L1, CAT, CDC42, EIF2AK2, HIPK2, MAP2K4, epithelial cell lines NAMPT, PAK2, PIAS1, PPARD, PPARG, PPP3R1, PRKAA1, RNF216, SGPL1, TCF12 Cell Death cell death cell death of skin cell 3.10E-02 BAK1, CAT lines Cell Death apoptosis apoptosis 1.37E-06 ABCE1, ACACA, AGPAT2, AHSA1, AKAP13, ANXA7, APH1A, APIP, ARHGDIA, ATF1, ATF3, ATG5, ATMIN, ATN1, ATXN2, AURKA, B4GALT5, BACH2 (includes EG:60468), BAK1, BARD1, BAX, BCL2L1, BCL2L13, BLCAP, BMP2, BRCA1, BTRC, CABLES2, CAPNS1, CARD8, CAT, CDC42, CDK6, CDKN1A, CDKN1B, CHEK2, COPS5, CSNK1E, CSNK2A1, CTNNBL1, CYFIP2 (includes EG:26999), DAB2, DFFA, DNAJB1, DNMT3B, DUSP22, EIF2AK2, EIF4G2, ERCC3, ETS1, EXOC2,

188 FKBP1A, FKBP5, FOXO3, GABPA, GABPB1, GADD45A, GMEB1, GPS2, HAUS1, HDAC3, HEY1, HIF1A, HIPK2, HMGA1, HMOX1, HUWE1, IMMT, LIMS1, MAP2K4, MAP3K1, MAPK7, MED1, MELK, MFN2, MSN, MTMR9, NAMPT, NBN, NCKAP1, NCOA3, NET1, NKX2-5, NME1, NRAS, ODC1, OGFOD1, PAK2, PCBP2, PDCD6IP, PDPK1, PHB (includes EG:5245), PIAS1, PLAGL1, POLR2A, PPARD, PPARG, PPP2CA, PPP2R2A, PPP3R1, PRKAA1, PRKAR1A, PRKCQ, PSME3, PTPN13, RARA, RB1, RB1CC1, RBM4B, RND3, RNF216, RYBP, SAP30BP, SEC61G, SET, SGPL1, SH3RF1, SKP2, SMARCB1, STAT3, STAT5B, TCF12, TFAP2A, THAP1, TNFRSF21, TOP2A, TPD52L1, TRAF4, TRIAP1, TRIM24, TRIM28, TSC2, TTF1, TXN, UBE4B, UGCG, VAV3, YBX1, YWHAE, YWHAZ Cell Death apoptosis apoptosis of tumor 1.78E-05 ABCE1, ACACA, AGPAT2, AHSA1, ARHGDIA, ATF1, ATF3, ATG5, cell lines ATMIN, AURKA, BACH2 (includes EG:60468), BAK1, BARD1, BAX, BCL2L1, BCL2L13, BMP2, BRCA1, BTRC, CABLES2, CAPNS1, CARD8, CAT, CDC42, CDK6, CDKN1A, CDKN1B, CHEK2, CSNK2A1, DAB2, DFFA, DNAJB1, EIF2AK2, EIF4G2, ETS1, EXOC2, FKBP5, FOXO3, GABPA, GADD45A, GPS2, HDAC3, HIF1A, HIPK2, HMGA1, HMOX1, HUWE1, IMMT, LIMS1, MAP2K4, MAPK7, MELK, MSN, MTMR9, NCOA3, NME1, NRAS, OGFOD1, PAK2, PCBP2, PDCD6IP, PDPK1, PHB (includes EG:5245), PIAS1, PLAGL1, PPARD, PPARG, PPP2CA, PPP2R2A, PRKAA1, PRKAR1A, PTPN13, RARA, RB1, RBM4B, SEC61G, SH3RF1, SKP2, STAT3, TNFRSF21, TOP2A, TRAF4, TRIAP1, TRIM28, TSC2, TTF1, TXN, UGCG, YBX1, YWHAE, YWHAZ Cell Death apoptosis apoptosis of cell lines 3.69E-05 ABCE1, ACACA, AGPAT2, AHSA1, ARHGDIA, ATF1, ATF3, ATG5, ATMIN, AURKA, BACH2 (includes EG:60468), BAK1, BARD1, BAX, BCL2L1, BCL2L13, BMP2, BRCA1, BTRC, CABLES2, CAPNS1, CARD8, CAT, CDC42, CDK6, CDKN1A, CDKN1B, CHEK2, CSNK2A1, DAB2, DFFA, DNAJB1, EIF2AK2, EIF4G2, ETS1, EXOC2, FKBP5, FOXO3, GABPA, GADD45A, GPS2, HDAC3, HIF1A, HIPK2, HMGA1, HMOX1, HUWE1, IMMT, LIMS1, MAP2K4, MAPK7, MELK, MSN, MTMR9, NBN, NCOA3, NME1, NRAS, OGFOD1, PAK2, PCBP2, PDCD6IP, PDPK1, PHB (includes EG:5245), PIAS1, PLAGL1, PPARD, PPARG, PPP2CA, PPP2R2A, PPP3R1, PRKAA1, PRKAR1A, PRKCQ, PTPN13, RARA, RB1, RBM4B, RNF216, SEC61G, SGPL1, SH3RF1, SKP2, STAT3, STAT5B, TCF12, TNFRSF21, TOP2A, TRAF4, TRIAP1, TRIM28, TSC2, TTF1, TXN, UGCG, YBX1, YWHAE, YWHAZ Cell Death apoptosis apoptosis of cervical 5.43E-05 ATG5, BAK1, BAX, BCL2L1, CAT, CDKN1A, CDKN1B, CSNK2A1, cancer cell lines EIF2AK2, EIF4G2, EXOC2, GADD45A, IMMT, LIMS1, MSN, MTMR9, NME1, OGFOD1, PAK2, PCBP2, PDCD6IP, PDPK1, PPP2R2A, RB1, SEC61G, TNFRSF21, TSC2, TXN Cell Death apoptosis apoptosis of bone 7.25E-05 ATG5, ATMIN, BAX, BMP2, BRCA1, CABLES2, CAPNS1, CDK6, cancer cell lines CDKN1A, ETS1, GPS2, HUWE1, PIAS1, PLAGL1, PRKAA1, RB1, TRAF4 Cell Death apoptosis apoptosis of 1.89E-04 ABCE1, ACACA, AGPAT2, AHSA1, ARHGDIA, ATF1, ATF3, ATG5, eukaryotic cells ATMIN, ATXN2, AURKA, B4GALT5, BACH2 (includes EG:60468), BAK1, BARD1, BAX, BCL2L1, BCL2L13, BMP2, BRCA1, BTRC, CABLES2, CAPNS1, CARD8, CAT, CDC42, CDK6, CDKN1A, CDKN1B, CHEK2, COPS5, CSNK2A1, DAB2, DFFA, DNAJB1, DNMT3B, EIF2AK2, EIF4G2, ETS1, EXOC2, FKBP5, FOXO3, GABPA, GADD45A, GPS2, HDAC3, HIF1A, HIPK2, HMGA1, HMOX1, HUWE1, IMMT, LIMS1, MAP2K4, MAPK7, MELK, MSN, MTMR9, NAMPT, NBN, NCOA3, NKX2-5, NME1, NRAS, OGFOD1, PAK2, PCBP2, PDCD6IP, PDPK1, PHB (includes EG:5245), PIAS1, PLAGL1, PPARD, PPARG, PPP2CA, PPP2R2A, PPP3R1, PRKAA1, PRKAR1A, PRKCQ, PTPN13, RARA, RB1, RBM4B, RND3, RNF216, SEC61G, SET, SGPL1, SH3RF1, SKP2, STAT3, STAT5B, TCF12, TFAP2A, TNFRSF21, TOP2A, TRAF4, TRIAP1, TRIM28, TSC2, TTF1, TXN, UGCG, YBX1, YWHAE, YWHAZ Cell Death apoptosis apoptosis of 2.44E-03 ARHGDIA, BACH2 (includes EG:60468), BCL2L1, BMP2, CARD8, CAT, lymphoma cell lines CDKN1A, CHEK2, EIF2AK2, MSN, PPARG, RARA, STAT3, UGCG, YWHAZ Cell Death apoptosis apoptosis of colon 2.72E-03 AHSA1, ATF3, ATMIN, BAK1, BAX, BCL2L1, CDKN1A, CHEK2, ETS1, cancer cell lines EXOC2, FOXO3, GABPA, GADD45A, HDAC3, HIPK2, HMGA1, LIMS1, PPARD, PPARG, TRIAP1, YWHAE Cell Death apoptosis apoptosis of tumor 3.90E-03 ATXN2, B4GALT5, BAX, BCL2L1, BRCA1, CAT, CDKN1A, CDKN1B, cells DNMT3B, HIF1A, NCOA3, PPP2CA, RARA, STAT3, TFAP2A, TXN Cell Death apoptosis apoptosis of 6.03E-03 BAX, BCL2L1, PPP2CA unspecified cell lines Cell Death apoptosis apoptosis of cancer 8.24E-03 ATXN2, B4GALT5, BAX, BCL2L1, BRCA1, CAT, CDKN1A, CDKN1B,

189 cells DNMT3B, NCOA3, RARA, TFAP2A, TXN Cell Death apoptosis apoptosis of glioma 9.81E-03 B4GALT5, BCL2L1, CAT, CDKN1B cells Cell Death apoptosis apoptosis of 1.08E-02 BAX, BCL2L1, CDKN1A, LIMS1, PDPK1, TRIM28, YBX1 fibrosarcoma cell lines Cell Death apoptosis apoptosis of prostate 1.93E-02 NCOA3, RARA cancer cells Cell Death apoptosis apoptosis of 2.91E-02 BAK1, BAX, BCL2L1, CDKN1B, FOXO3, HMGA1, PRKAR1A, RB1, carcinoma cell lines RBM4B, SKP2, TTF1, YBX1 Cell Death apoptosis apoptosis of lung 3.03E-02 BAK1, BAX, BCL2L1, CDKN1A, CDKN1B, FOXO3, GADD45A, HIPK2, cancer cell lines PPP2CA, RB1, SKP2, TRIAP1, TTF1, YBX1 Cell Death apoptosis apoptosis of prostate 4.37E-02 ABCE1, ACACA, BAK1, BAX, BCL2L1, BMP2, BRCA1, CAT, FOXO3, cancer cell lines GADD45A, HIF1A, NCOA3, RB1, STAT3 Cell Death apoptosis apoptosis of 4.47E-02 BAX, BCL2L1 sympathetic neuron Cell Death survival survival of breast 7.52E-04 BCL2L1, BRCA1, CDKN1A, FKBP5, FOXM1, HMGA1, KIFC1, MED1, cancer cell lines PTPN13, RB1CC1, STAT3, UGCG Cell Death survival survival of 9.22E-03 BAX, CDKN1A, NBN lymphoblastoid cell lines Cell Death survival survival of tumor cell 1.32E-02 AKAP13, BAX, BCL2L1, BRCA1, CAT, CDK6, CDKN1A, CDKN1B, lines CHEK2, DUSP22, DUSP5, FKBP5, FOXM1, GLUD1, HDAC3, HMGA1, HMOX1, KIFC1, LAMTOR3, MAP3K1, MED1, MTMR12, NBN, NRAS, NUAK1, PPAT, PPP1R8, PPP2CA, PRKAA1, PRKCQ, PTPN13, RB1CC1, STAT3, TRIM28, TXN, UGCG Cell Death survival survival of cell lines 4.38E-02 AKAP13, BAX, BCL2L1, BRCA1, CAT, CDK6, CDKN1A, CDKN1B, CHEK2, DUSP22, DUSP5, FKBP5, FOXM1, GLUD1, HDAC3, HMGA1, HMOX1, KIFC1, LAMTOR3, MAP3K1, MED1, MGST1, MTMR12, NBN, NRAS, NUAK1, PPAT, PPP1R8, PPP2CA, PRKAA1, PRKCQ, PTPN13, RB1CC1, STAT3, TRIM28, TXN, UGCG Cell Death survival survival of 4.47E-02 NRAS, TRIM28 fibrosarcoma cell lines Cell Death cell viability cell viability of 1.32E-02 CAT, HMOX1, RARA lymphoma cell lines Cell Death cell viability cell viability of 1.93E-02 BMP2, HMOX1 cancer cells Cell Death colony colony survival of 1.56E-02 BCL2L1, CDKN1A, CHEK2, RB1CC1 survival tumor cell lines Cell Death colony colony survival of 1.93E-02 CDKN1A, CHEK2 survival colon cancer cell lines Cell Death necrosis necrosis of 3.10E-02 CAT, CDKN1A lymphoma cell lines Cancer colon cancer colon cancer 2.31E-06 ACLY, AKAP13, ARHGDIA, ATF3, AURKA, BAK1, BAX, BMP2, CAT, CDCA5, CDK7, CDKN1A, CDKN1B, CDKN3, CEP55, CHMP1A, CNTNAP2, DUSP5, EIF2AK2, EXOSC9, FKBP1A, GNA11, HAND1, HDAC3, HUWE1, KIFC1, LRP8, MCM3, MCM5, MELK, MLF1IP, MSH3, MT1X, NKX2-5, NRAS, ODC1, PDPK1, POLD1, PPARG, PPPDE1, RUVBL1, SET, SRD5A1, STIP1, STXBP6, TMEM97, TPD52L1, TRIAP1, TUBA1A, TUBA1C, TUBG1, UBA1, YWHAE Cancer central central nervous 2.64E-04 B4GALT5, CDCA7L, CDK6, CDK7, CDKN1B, ELK1, FDFT1, FKBP1A, nervous system tumor FNTB, GNA13, HDAC3, HIF1A, HIPK2, MARCKS, MLL3, POLD1, system PPARG, PSMB5, RARA, RB1, SEC61G, SOX11, TAOK1, TOP2A, tumor TUBA1A, TUBA1C, TUBG1, UBE2C, VAV3 Cancer colon tumor colon tumor 2.70E-04 ACLY, ARHGDIA, AURKA, CAT, HAND1, HDAC3, HUWE1, LRP8, PDPK1, POLD1, PPARG, PPPDE1, STIP1, TMEM97, TUBG1, UBA1, YWHAE Cancer lung lung adenocarcinoma 3.59E-04 ALDOA, CDCA5, DICER1, EIF2C1, FOXM1, FOXO3, GPI, LDHA, adenocarcino NRAS, POLR2A, SMEK1, SRM, UBE2C ma Cancer lung tumor lung tumor 3.71E-04 ALDOA, BAX, BCL2L1, CDCA5, CDKN1A, CDKN1B, CERS2, CERS5, DICER1, EIF2C1, FOXM1, FOXO3, GPI, HUWE1, LDHA, NRAS, PDPK1, POLR2A, PSMB5, RB1, SEC61G, SMEK1, SRM, TOP2A, TUBA1A, TUBA1C, TUBG1, UBE2C, YWHAE Cancer colon colon carcinoma 4.29E-04 ACLY, ARHGDIA, CAT, HAND1, HDAC3, HUWE1, LRP8, PDPK1, carcinoma POLD1, PPARG, PPPDE1, STIP1, TMEM97, TUBG1, UBA1, YWHAE

190 Cancer transformatio transformation of 4.78E-04 AURKA, BCL2L1, BMP2, BRCA1, CAT, CDKN1A, CDKN1B, HMGA1, n cells HYAL2, NCOA3, NRAS, ODC1, SEPT9, SKP2, STAT3, TBP Cancer tumor tumor 4.83E-04 ABI2, ACLY, ADARB1, ADORA2B, AGPAT2, AHCYL1, ALDOA, ANXA5, ARHGDIA, AURKA, B4GALT5, BAK1, BAX, BCL2L1, BMP2, BMPR1A, BMPR2, BRF2, C13orf15, C6orf211, CAT, CBL, CCDC28A, CDCA5, CDCA7L, CDK6, CDK7, CDKN1A, CDKN1B, CDKN3, CEBPG, CERS2, CERS5, CETN3, CHEK2, COBLL1, COPS5, CUX1, DAB2, DDHD2, DICER1, DKC1, DYRK2, EIF2AK1, EIF2AK2, EIF2C1, EIF4B, ELK1, EPS15, ETS1, FDFT1, FKBP1A, FNTB, FOXM1, FOXN2, FOXO3, FYN, FZD3, GADD45A, GLUD1, GNA13, GNAI2, GOLGA5, GPI, GPM6B, GSTO1, HAND1, HDAC3, HEY1, HIF1A, HIPK2, HMGA1, HPS3, HS3ST3A1, HUWE1, KIF1B, KRIT1, LDHA, LOC728622/SKP1, LRP8, LTN1, MARCKS, MBNL1, MCM5, MCM7, MGST1, MIPEP, MLF1IP, MLL3, MLX, MSH3, MSX2, MT1X, MTUS1, NCAPG (includes EG:64151), NCOA3, NCOA4, NETO2, NFYC, NKX2-5, NME1, NRAS, NUP50, ODC1, PA2G4, PDPK1, POLD1, POLR2A, PPARD, PPARG, PPP3CB, PPPDE1, PRCC, PRDX6, PRKAR1A, PRKCQ, PRPF8, PSMB5, RARA, RB1, RBM4B, RCAN2, ROCK2, RPE, SDHD, SEC22B, SEC31A, SEC61G, SKP2, SMARCB1, SMEK1, SOX11, SRD5A1, SRM, STAMBP, STAT3, STAT5B, STIP1, STK35, STX3, STXBP6, TAOK1, TFAP2A, TFDP2, TMEM97, TNPO3, TOM1L1, TOP2A, TRIM24, TRIM33, TSC1, TSC2, TUBA1A, TUBA1C, TUBG1, TXN, UBA1, UBE2C, UGCG, VAV3, WDR19, WNK1, YWHAE, YWHAZ, ZMYM2 Cancer tuberous tuberous sclerosis 7.02E-04 FKBP1A, HMOX1, TSC1, TSC2 sclerosis Cancer head and head and neck cancer 7.71E-04 AURKA, B4GALT5, BRCA1, CDC42, CDCA7L, CDK6, CDKN1A, neck cancer CDKN1B, ELK1, EPS15, ETS1, FDFT1, FKBP1A, FNTB, FYN, GNA13, GOLGA5, HDAC3, HIF1A, HIPK2, HUWE1, LDHA, LMO4, LRRC8D, MARCKS, MLL3, NCOA4, NRAS, PAK2, PPARD, PPARG, PRKAR1A, PSMB5, RB1, SEC61G, SET, SOX11, STAT5B, TAOK1, TMEM38B, TOP2A, TRIM24, TRIM33, TUBA1A, TUBA1C, TUBG1, UBE2C, VAV3 Cancer sarcoma sarcoma 8.78E-04 CDK6, CDK7, CDKN1A, CHEK2, FDFT1, FKBP1A, FNTB, FOXM1, FYN, HDAC3, HIPK2, HMGA1, POLD1, PPARG, PSMB5, RARA, RB1, SMARCB1, STK35, TAOK1, TOP2A, TUBA1A, TUBA1C, TUBG1, UBE2C, WNK1 Cancer soft tissue soft tissue sarcoma 8.86E-04 CDK6, CDKN1A, FKBP1A, FNTB, FOXM1, FYN, HDAC3, HMGA1, sarcoma PPARG, RB1, SMARCB1, TOP2A, TUBA1A, TUBA1C, TUBG1, UBE2C Cancer colorectal colorectal cancer 9.26E-04 ACLY, AKAP13, ARHGDIA, ATF3, AURKA, BAK1, BAX, BEX4, BMP2, cancer CAT, CDCA5, CDK7, CDKN1A, CDKN1B, CDKN3, CEP55, CHMP1A, CNTNAP2, COBLL1, DUSP5, EIF2AK1, EIF2AK2, ENAH, EPB41L3, ERRFI1, EXOSC9, FBXL2, FKBP1A, FNTB, GNA11, HAND1, HDAC3, HIF1A, HMOX1, HUWE1, KIFC1, LDHA, LRP8, MCM3, MCM5, MELK, MLF1IP, MSH3, MT1X, NCAPG (includes EG:64151), NFYC, NKX2-5, NRAS, NUAK1, ODC1, PDPK1, POLD1, PPARD, PPARG, PPP3CB, PPPDE1, PSMB5, RAB31, RUVBL1, SEPT9, SET, SRD5A1, SSBP3, STIP1, STK35, STXBP6, TAOK1, TFDP1, TMEFF2, TMEM97, TOP2A, TPD52L1, TRIAP1, TUBA1A, TUBA1C, TUBG1, UBA1, UBE2C, WARS, YWHAE, ZNF200 Cancer brain cancer brain cancer 1.07E-03 B4GALT5, BAX, CDCA7L, CDK6, CDKN1B, ELK1, FKBP1A, FNTB, GNA13, HIF1A, HIPK2, MARCKS, MLL3, PPARG, PSMB5, RB1, SEC61G, SMARCB1, SOX11, TAOK1, TOP2A, TUBA1A, TUBA1C, TUBG1, UBE2C, VAV3 Cancer digestive digestive organ 1.20E-03 ACLY, ARHGDIA, AURKA, BAX, CAT, CCDC28A, CDKN1A, CDKN1B, organ tumor tumor CDKN3, COBLL1, COPS5, EIF2AK2, EPS15, FKBP1A, FNTB, FYN, GLUD1, GSTO1, HAND1, HDAC3, HIF1A, HUWE1, LRP8, MLF1IP, MLL3, MSH3, MSX2, MTUS1, NCAPG (includes EG:64151), PA2G4, PDPK1, POLD1, PPARG, PPP3CB, PPPDE1, PRCC, PRKCQ, PSMB5, RARA, RB1, SEC31A, STIP1, TFDP2, TMEM97, TOP2A, TUBA1A, TUBA1C, TUBG1, TXN, UBA1, UBE2C, YWHAE Cancer colorectal colorectal carcinoma 1.32E-03 ACLY, ARHGDIA, CAT, COBLL1, FNTB, HAND1, HDAC3, HUWE1, carcinoma LRP8, PDPK1, POLD1, PPARG, PPP3CB, PPPDE1, STIP1, TMEM97, TUBG1, UBA1, UBE2C, YWHAE Cancer tumorigenesi tumorigenesis 1.32E-03 ABI2, ACLY, ACTR3, ADARB1, ADORA2B, AGPAT2, AHCYL1, s AKAP13, ALDOA, ANXA5, ARHGDIA, ASXL1, ATF3, ATP5C1, AURKA, B4GALT5, BAG4, BAK1, BARD1, BAX, BCL2L1, BEX4, BMP2, BMPR1A, BMPR2, BRCA1, BRF2, BTRC, C13orf15, C18orf8, C6orf211, CAT, CBL, CCDC28A, CDC42, CDCA5, CDCA7L, CDK6, CDK7, CDKN1A, CDKN1B, CDKN3, CEBPG, CEP55, CERS2, CERS5, CETN3, CHEK2, CHMP1A, CNTNAP2, COBLL1, COPS5, CSNK1E,

191 CUX1, DAB2, DDHD2, DIAPH2, DICER1, DKC1, DOCK11, DUSP5, DYRK2, EI24, EIF2AK1, EIF2AK2, EIF2C1, EIF4B, EIF5, ELK1, ELOVL6, ENAH, EPB41L3, EPS15, ERRFI1, ETS1, EXOSC9, FBXL2, FDFT1, FGFR1OP2, FHOD3, FKBP1A, FNTB, FOXM1, FOXN2, FOXO3, FYN, FZD3, G3BP2, GADD45A, GLUD1, GNA11, GNA13, GNAI2, GOLGA5, GOT1, GPI, GPM6B, GSTO1, GUCY1A3, HAND1, HDAC3, HEY1, HIF1A, HIPK2, HMGA1, HMOX1, HPS3, HS3ST3A1, HUWE1, KDELR2, KIF1B, KIFC1, KRIT1, LDHA, LMO4, LOC728622/SKP1, LPP, LRP8, LRRC8D, LTN1, MAL, MAP3K1, MAPK7, MARCKS, MBNL1, MCM3, MCM5, MCM7, MELK, MGST1, MIPEP, MLF1IP, MLL3, MLX, MSH3, MSX2, MT1X, MTUS1, NAMPT, NBN, NCAPG (includes EG:64151), NCOA3, NCOA4, NETO2, NFYC, NKX2-5, NME1, NRAS, NSD1, NUAK1, NUP214, NUP50, ODC1, PA2G4, PAK2, PDCD6IP, PDPK1, PHB (includes EG:5245), PLAGL2, POLD1, POLR2A, PPARD, PPARG, PPAT, PPP2CA, PPP3CB, PPP3R1, PPPDE1, PRCC, PRDX6, PRKAR1A, PRKCQ, PRPF8, PSD3, PSMB5, PTP4A1, PTPN1, RAB31, RARA, RB1, RB1CC1, RBBP8, RBM15, RBM4B, RCAN2, RND3, ROCK2, RPE, RUVBL1, SDHD, SEC22B, SEC31A, SEC61G, SEPT9, SET, SKP2, SLC35C1, SMARCB1, SMEK1, SOX11, SRD5A1, SRM, SSBP3, STAMBP, STAT3, STAT5B, STIP1, STK35, STX3, STXBP6, SUZ12, TAOK1, TFAP2A, TFDP1, TFDP2, TGIF2, TMEFF2, TMEM38B, TMEM97, TNPO3, TOM1L1, TOP2A, TPD52L1, TRIAP1, TRIM24, TRIM33, TSC1, TSC2, TUBA1A, TUBA1C, TUBG1, TXN, UBA1, UBE2C, UGCG, VAV3, WARS, WDR19, WNK1, YEATS2, YWHAE, YWHAZ, ZBTB4, ZBTB40, ZFHX3, ZMYM2, ZNF200 Cancer colorectal colorectal tumor 1.40E-03 ACLY, ARHGDIA, AURKA, CAT, COBLL1, FNTB, HAND1, HDAC3, tumor HUWE1, LRP8, PDPK1, POLD1, PPARG, PPP3CB, PPPDE1, STIP1, TMEM97, TUBA1A, TUBA1C, TUBG1, UBA1, UBE2C, YWHAE Cancer neuroepitheli neuroepithelial tumor 1.53E-03 B4GALT5, CDK6, CDK7, CDKN1B, ELK1, FKBP1A, GNA13, HIF1A, al tumor HIPK2, MARCKS, MLL3, PPARG, RARA, RB1, SEC61G, SOX11, TAOK1, TOP2A, UBE2C, VAV3 Cancer head and head and neck tumor 1.70E-03 AURKA, B4GALT5, CDCA7L, CDK6, CDKN1B, ELK1, EPS15, ETS1, neck tumor FKBP1A, FNTB, GNA13, GOLGA5, HDAC3, HIF1A, HIPK2, HUWE1, LDHA, MARCKS, MLL3, NCOA4, NRAS, PPARD, PPARG, PRKAR1A, PSMB5, RB1, SEC61G, SOX11, STAT5B, TAOK1, TOP2A, TRIM24, TRIM33, TUBA1A, TUBA1C, TUBG1, UBE2C, VAV3 Cancer acute acute myeloid 1.86E-03 ASXL1, CBL, FKBP1A, FNTB, FYN, HDAC3, LPP, NRAS, NSD1, myeloid leukemia NUP214, PDPK1, PLAGL2, POLD1, PPAT, PSMB5, RARA, RBM15, leukemia STAT3, TOP2A Cancer myosarcoma myosarcoma 2.01E-03 CDKN1A, FKBP1A, FOXM1, FYN, HMGA1, SMARCB1, TOP2A, TUBA1A, TUBA1C, TUBG1, UBE2C Cancer brain tumor brain tumor 2.07E-03 B4GALT5, CDCA7L, CDK6, CDKN1B, ELK1, FKBP1A, FNTB, GNA13, HIF1A, HIPK2, MARCKS, MLL3, PPARG, RB1, SEC61G, SOX11, TAOK1, TOP2A, TUBA1A, TUBA1C, TUBG1, UBE2C, VAV3 Cancer lung lung carcinoma 2.38E-03 ALDOA, CDCA5, DICER1, EIF2C1, FOXM1, FOXO3, GPI, HUWE1, carcinoma LDHA, NRAS, POLR2A, RB1, SEC61G, SMEK1, SRM, TOP2A, TUBA1A, TUBA1C, TUBG1, UBE2C, YWHAE Cancer gastrointestin gastrointestinal tumor 2.44E-03 BAX, CCDC28A, COBLL1, FKBP1A, FYN, HDAC3, MSH3, PA2G4, al tumor PRCC, PRKCQ, SEC31A, TFDP2, TMEM97, TOP2A, TUBA1A, TUBA1C, TUBG1, UBE2C Cancer lymphocytic lymphocytic 2.51E-03 BAK1, BAX, BCL2L1, CDK6, CDK7, CDKN1A, FNTB, FOXO3, FYN, leukemia leukemia HDAC3, NRAS, POLD1, PPAT, PPP3CB, PPP3R1, PSMB5, RARA, RB1, STAT3, TOP2A, TUBA1A, TUBA1C, TUBG1 Cancer stomach stomach tumor 2.66E-03 BAX, CCDC28A, COBLL1, FKBP1A, MSH3, PA2G4, PRCC, SEC31A, tumor TFDP2, TMEM97, TOP2A, TUBA1A, TUBA1C, TUBG1, UBE2C Cancer neoplasia neoplasia 2.70E-03 ABI2, ACLY, ACTR3, ADARB1, ADORA2B, AGPAT2, AHCYL1, AKAP13, ALDOA, ANXA5, ARHGDIA, ASXL1, ATF3, ATP5C1, AURKA, B4GALT5, BAG4, BAK1, BARD1, BAX, BCL2L1, BEX4, BMP2, BMPR1A, BMPR2, BRCA1, BRF2, BTRC, C13orf15, C18orf8, C6orf211, CAT, CBL, CCDC28A, CDC42, CDCA5, CDCA7L, CDK6, CDK7, CDKN1A, CDKN1B, CDKN3, CEP55, CERS2, CERS5, CETN3, CHEK2, CHMP1A, CNTNAP2, COBLL1, COPS5, CUX1, DAB2, DDHD2, DIAPH2, DICER1, DOCK11, DUSP5, DYRK2, EI24, EIF2AK1, EIF2AK2, EIF2C1, EIF4B, EIF5, ELK1, ELOVL6, ENAH, EPB41L3, EPS15, ERRFI1, ETS1, EXOSC9, FBXL2, FDFT1, FGFR1OP2, FHOD3, FKBP1A, FNTB, FOXM1, FOXN2, FOXO3, FYN, FZD3, G3BP2, GADD45A, GLUD1, GNA11, GNA13, GNAI2, GOLGA5, GOT1, GPI, GSTO1, GUCY1A3, HAND1, HDAC3, HEY1, HIF1A, HIPK2, HMGA1, HMOX1, HPS3, HS3ST3A1, HUWE1, KIF1B, KIFC1, LDHA, LMO4, LOC728622/SKP1,

192 LPP, LRP8, LRRC8D, LTN1, MAL, MAP3K1, MAPK7, MARCKS, MBNL1, MCM3, MCM5, MCM7, MELK, MGST1, MIPEP, MLF1IP, MLL3, MSH3, MSX2, MT1X, NAMPT, NBN, NCAPG (includes EG:64151), NCOA3, NCOA4, NFYC, NKX2-5, NME1, NRAS, NSD1, NUAK1, NUP214, NUP50, ODC1, PA2G4, PAK2, PDPK1, PHB (includes EG:5245), PLAGL2, POLD1, POLR2A, PPARD, PPARG, PPAT, PPP2CA, PPP3CB, PPP3R1, PPPDE1, PRCC, PRKAR1A, PRKCQ, PRPF8, PSD3, PSMB5, PTP4A1, PTPN1, RAB31, RARA, RB1, RB1CC1, RBBP8, RBM15, RBM4B, RCAN2, RND3, ROCK2, RPE, RUVBL1, SDHD, SEC22B, SEC31A, SEC61G, SEPT9, SET, SKP2, SLC35C1, SMARCB1, SMEK1, SOX11, SRD5A1, SRM, SSBP3, STAMBP, STAT3, STAT5B, STIP1, STK35, STX3, STXBP6, SUZ12, TAOK1, TFAP2A, TFDP1, TFDP2, TGIF2, TMEFF2, TMEM38B, TMEM97, TNPO3, TOM1L1, TOP2A, TPD52L1, TRIAP1, TRIM24, TRIM33, TSC1, TSC2, TUBA1A, TUBA1C, TUBG1, TXN, UBA1, UBE2C, UGCG, VAV3, WARS, WDR19, WNK1, YEATS2, YWHAE, YWHAZ, ZBTB4, ZBTB40, ZFHX3, ZMYM2, ZNF200 Cancer endocrine endocrine gland 3.26E-03 BMP2, BMPR1A, BMPR2, CDKN1A, COPS5, ETS1, FKBP1A, GNAI2, gland tumor tumor GOLGA5, HIF1A, HUWE1, MLL3, MSX2, NCOA4, NRAS, PPARD, PPARG, PRKAR1A, SDHD, TOP2A, TRIM24, TRIM33 Cancer adenocarcino adenocarcinoma 3.33E-03 AGPAT2, ALDOA, C13orf15, CDCA5, DAB2, DICER1, EIF2C1, ETS1, ma FDFT1, FKBP1A, FOXM1, FOXN2, FOXO3, FZD3, GOLGA5, GPI, HIF1A, HPS3, LDHA, LOC728622/SKP1, LRP8, NCOA4, NRAS, POLR2A, PPARG, PPP3CB, PRKAR1A, PSMB5, SMEK1, SRM, TRIM24, TRIM33, TUBA1A, TUBA1C, TUBG1, UBE2C, VAV3 Cancer urinary tract urinary tract tumor 4.56E-03 AGPAT2, AURKA, FKBP1A, HIF1A, NRAS, RB1, TOP2A, TSC1, tumor TUBA1A, TUBA1C, TUBG1, UBE2C Cancer mantle cell mantle cell 4.56E-03 CDK6, CDK7, FKBP1A, FNTB, POLD1, PSMB5, TOP2A lymphoma lymphoma Cancer cancer cancer 4.93E-03 ABI2, ACLY, ACTR3, ADARB1, ADORA2B, AGPAT2, AHCYL1, AKAP13, ALDOA, ANXA5, ARHGDIA, ASXL1, ATF3, ATP5C1, AURKA, B4GALT5, BAG4, BAK1, BARD1, BAX, BCL2L1, BEX4, BMP2, BMPR1A, BMPR2, BRCA1, BRF2, BTRC, C13orf15, C18orf8, C6orf211, CAT, CBL, CCDC28A, CDC42, CDCA5, CDCA7L, CDK6, CDK7, CDKN1A, CDKN1B, CDKN3, CEP55, CERS2, CERS5, CETN3, CHEK2, CHMP1A, CNTNAP2, COBLL1, COPS5, CUX1, DAB2, DDHD2, DICER1, DOCK11, DUSP5, DYRK2, EI24, EIF2AK1, EIF2AK2, EIF2C1, EIF4B, EIF5, ELK1, ELOVL6, ENAH, EPB41L3, EPS15, ERRFI1, ETS1, EXOSC9, FBXL2, FDFT1, FHOD3, FKBP1A, FNTB, FOXM1, FOXN2, FOXO3, FYN, FZD3, G3BP2, GADD45A, GLUD1, GNA11, GNA13, GNAI2, GOLGA5, GOT1, GPI, GSTO1, GUCY1A3, HAND1, HDAC3, HEY1, HIF1A, HIPK2, HMGA1, HMOX1, HPS3, HS3ST3A1, HUWE1, KIF1B, KIFC1, LDHA, LMO4, LOC728622/SKP1, LPP, LRP8, LRRC8D, LTN1, MAL, MAP3K1, MAPK7, MARCKS, MBNL1, MCM3, MCM5, MCM7, MELK, MGST1, MIPEP, MLF1IP, MLL3, MSH3, MSX2, MT1X, NAMPT, NBN, NCAPG (includes EG:64151), NCOA3, NCOA4, NFYC, NKX2-5, NME1, NRAS, NSD1, NUAK1, NUP214, NUP50, ODC1, PA2G4, PAK2, PDPK1, PHB (includes EG:5245), PLAGL2, POLD1, POLR2A, PPARD, PPARG, PPAT, PPP2CA, PPP3CB, PPP3R1, PPPDE1, PRCC, PRKAR1A, PRKCQ, PRPF8, PSD3, PSMB5, PTP4A1, PTPN1, RAB31, RARA, RB1, RB1CC1, RBM15, RBM4B, RCAN2, RND3, ROCK2, RPE, RUVBL1, SEC22B, SEC31A, SEC61G, SEPT9, SET, SKP2, SLC35C1, SMARCB1, SMEK1, SOX11, SRD5A1, SRM, SSBP3, STAMBP, STAT3, STAT5B, STIP1, STK35, STX3, STXBP6, SUZ12, TAOK1, TFAP2A, TFDP1, TFDP2, TGIF2, TMEFF2, TMEM38B, TMEM97, TNPO3, TOM1L1, TOP2A, TPD52L1, TRIAP1, TRIM24, TRIM33, TSC1, TUBA1A, TUBA1C, TUBG1, TXN, UBA1, UBE2C, UGCG, VAV3, WARS, WDR19, WNK1, YEATS2, YWHAE, YWHAZ, ZBTB4, ZBTB40, ZFHX3, ZMYM2, ZNF200 Cancer retinoblasto retinoblastoma 5.26E-03 CDK6, CDK7, RARA, RB1, TOP2A ma Cancer infection infection of tumor 5.36E-03 AP1B1, AP2M1, ARHGAP32, ASXL2, BRCA1, CBL, CDC42, CLDND1, cell lines CRIPAK, CRTC2, DDX50, DHX33, DNAJB1, EPS15, ERCC3, FBXW11, GABPB1, GOLPH3, HUWE1, INTS7, ISG20L2, LRRC8D, MED20, MPHOSPH6, NRBP1, NUP85 (includes EG:287830), PIP5K1A, PPP2R2A, PRKAA1, RAB5A, RAB6A, RB1CC1, RIMS4, RNF216, RNF26, SEC61G, STIP1, STX5, SYNJ1, TAOK1, TFDP2, TNPO3, TRIM5, TRMT5, UBE2E1, WASF2, WNK1, ZNRD1 Cancer infection infection of cervical 2.26E-02 AP2M1, ARHGAP32, ASXL2, BRCA1, CDC42, CLDND1, CRIPAK,

193 cancer cell lines CRTC2, DDX50, DHX33, DNAJB1, ERCC3, FBXW11, GABPB1, GOLPH3, HUWE1, INTS7, ISG20L2, LRRC8D, MED20, MPHOSPH6, NRBP1, NUP85 (includes EG:287830), PPP2R2A, PRKAA1, RAB5A, RAB6A, RB1CC1, RIMS4, RNF26, SEC61G, STIP1, STX5, TAOK1, TFDP2, TNPO3, TRMT5, UBE2E1, WNK1, ZNRD1 Cancer non-hodgkin non-hodgkin 5.63E-03 CDK6, CDK7, EIF2AK1, FKBP1A, FNTB, FYN, GADD45A, HDAC3, lymphoma lymphoma ODC1, POLD1, PPARG, PSMB5, RARA, TAOK1, TOP2A, TUBA1A, TUBA1C, TUBG1, YWHAZ Cancer disease disease of tumor cell 5.65E-03 AP1B1, AP2M1, ARHGAP32, ASXL2, BRCA1, CBL, CDC42, CDKN1A, lines CLDND1, CRIPAK, CRTC2, DDX50, DHX33, DNAJB1, EPS15, ERCC3, ETS1, FBXW11, FYN, GABPB1, GOLPH3, HMOX1, HUWE1, INTS7, ISG20L2, LRRC8D, MED20, MPHOSPH6, NME1, NRBP1, NUP85 (includes EG:287830), PDCD6IP, PIP5K1A, PPP2R2A, PRKAA1, PRKAR1A, RAB5A, RAB6A, RB1CC1, RIMS4, RNF216, RNF26, SEC61G, STIP1, STX5, SYNJ1, TAOK1, TFDP2, TNPO3, TRIM5, TRMT5, UBE2E1, WASF2, WNK1, ZNRD1 Cancer disease disease of cervical 1.93E-02 AP2M1, ARHGAP32, ASXL2, BRCA1, CDC42, CLDND1, CRIPAK, cancer cell lines CRTC2, DDX50, DHX33, DNAJB1, ERCC3, FBXW11, GABPB1, GOLPH3, HUWE1, INTS7, ISG20L2, LRRC8D, MED20, MPHOSPH6, NRBP1, NUP85 (includes EG:287830), PDCD6IP, PPP2R2A, PRKAA1, RAB5A, RAB6A, RB1CC1, RIMS4, RNF26, SEC61G, STIP1, STX5, TAOK1, TFDP2, TNPO3, TRMT5, UBE2E1, WNK1, ZNRD1 Cancer disease disease of hepatoma 4.23E-02 AP1B1, CBL, EPS15, PIP5K1A, RAB5A, SYNJ1, WASF2 cell lines Cancer bladder bladder tumor 5.83E-03 AURKA, RB1, TOP2A, TSC1, TUBA1A, TUBA1C, TUBG1, UBE2C tumor Cancer gastric gastric carcinoma 5.97E-03 BAX, CCDC28A, COBLL1, MSH3, PA2G4, PRCC, SEC31A, TFDP2, carcinoma TMEM97, TOP2A, UBE2C Cancer leukemia leukemia 6.33E-03 ADORA2B, ASXL1, BAK1, BAX, BCL2L1, CBL, CDK6, CDK7, CDKN1A, EIF2AK2, FKBP1A, FNTB, FOXO3, FYN, HDAC3, LPP, NRAS, NSD1, NUP214, PDPK1, PLAGL2, POLD1, PPAT, PPP3CB, PPP3R1, PSMB5, RARA, RB1, RBM15, STAT3, TOP2A, TUBA1A, TUBA1C, TUBG1 Cancer glioma glioma 6.53E-03 B4GALT5, CDK6, CDKN1B, ELK1, FKBP1A, GNA13, HIPK2, MARCKS, MLL3, PPARG, RB1, SEC61G, SOX11, TAOK1, TOP2A, UBE2C, VAV3 Cancer early-onset early-onset breast 6.58E-03 BARD1, BRCA1, CHEK2, EI24, RB1CC1 breast cancer cancer Cancer lung cancer lung cancer 7.00E-03 ALDOA, AURKA, BAX, BCL2L1, BTRC, CDCA5, CDK6, CDKN1A, CDKN1B, CERS2, CERS5, CHEK2, DICER1, EIF2AK1, EIF2C1, FDFT1, FKBP1A, FNTB, FOXM1, FOXO3, FYN, GPI, GUCY1A3, HDAC3, HIF1A, HIPK2, HUWE1, LDHA, NRAS, PDPK1, POLR2A, PPARG, PSMB5, RARA, RB1, SEC61G, SMEK1, SRM, STK35, TAOK1, TOP2A, TUBA1A, TUBA1C, TUBG1, UBE2C, WNK1, YWHAE Cancer benign tumor benign tumor 7.07E-03 ABI2, ADARB1, AHCYL1, ALDOA, ANXA5, CBL, CDKN1B, CEBPG, CETN3, DKC1, DYRK2, FDFT1, FKBP1A, FNTB, FYN, GPM6B, HUWE1, KIF1B, KRIT1, LTN1, MGST1, MLX, NETO2, NUP50, PPARD, PPARG, PPP3CB, PRDX6, PRPF8, RCAN2, RPE, SDHD, SEC22B, STAMBP, STX3, STXBP6, TNPO3, TOP2A, TSC2, TUBA1A, TUBA1C, TUBG1, UGCG Cancer lymphoma lymphoma 7.20E-03 CDK6, CDK7, CDKN1A, CDKN1B, EIF2AK1, FKBP1A, FNTB, FYN, GADD45A, HDAC3, ODC1, POLD1, PPARG, PSMB5, RARA, TAOK1, TOP2A, TUBA1A, TUBA1C, TUBG1, YWHAZ Cancer heart and heart and pleura 7.21E-03 FYN, PRKAR1A, RARA, TOP2A, TUBA1A, TUBA1C, TUBG1 pleura tumor tumor Cancer diffuse B- diffuse B-cell 7.42E-03 CDK6, CDK7, FKBP1A, FNTB, FYN, POLD1, PSMB5, TOP2A, YWHAZ cell lymphoma lymphoma Cancer hematologic hematologic cancer 8.87E-03 ADORA2B, ASXL1, BAK1, BAX, BCL2L1, CBL, CDK6, CDK7, cancer CDKN1A, CHEK2, EIF2AK2, FDFT1, FKBP1A, FNTB, FOXO3, FYN, HDAC3, LPP, NRAS, NSD1, NUP214, PDPK1, PLAGL2, POLD1, PPAT, PPP3CB, PPP3R1, PSMB5, RARA, RB1, RBM15, STAT3, TOP2A, TUBA1A, TUBA1C, TUBG1 Cancer malignant malignant tumor 9.22E-03 ACLY, ADORA2B, AGPAT2, ALDOA, ARHGDIA, AURKA, BAK1, tumor BAX, BMP2, BMPR1A, BMPR2, BRF2, C13orf15, C6orf211, CAT, CCDC28A, CDCA5, CDCA7L, CDK6, CDK7, CDKN1A, CDKN1B, CDKN3, CHEK2, COBLL1, COPS5, CUX1, DAB2, DDHD2, DICER1, EIF2AK1, EIF2AK2, EIF2C1, EIF4B, EPS15, ETS1, FDFT1, FKBP1A, FNTB, FOXM1, FOXN2, FOXO3, FYN, FZD3, GADD45A, GLUD1, GOLGA5, GPI, GSTO1, HAND1, HDAC3, HEY1, HIF1A, HIPK2,

194 HMGA1, HPS3, HS3ST3A1, HUWE1, KIF1B, LDHA, LOC728622/SKP1, LRP8, MBNL1, MCM5, MCM7, MIPEP, MLF1IP, MLL3, MSH3, MSX2, MT1X, NCAPG (includes EG:64151), NCOA3, NCOA4, NFYC, NKX2-5, NME1, NRAS, ODC1, PA2G4, PDPK1, POLD1, POLR2A, PPARD, PPARG, PPP3CB, PPPDE1, PRCC, PRKAR1A, PSMB5, RARA, RB1, RBM4B, RCAN2, ROCK2, SEC31A, SEC61G, SKP2, SMARCB1, SMEK1, SRD5A1, SRM, STAT3, STAT5B, STIP1, STK35, TAOK1, TFAP2A, TFDP2, TMEM97, TOP2A, TRIM24, TRIM33, TUBA1A, TUBA1C, TUBG1, TXN, UBA1, UBE2C, VAV3, WDR19, WNK1, YWHAE, YWHAZ, ZMYM2 Cancer myeloprolife myeloproliferative 9.53E-03 ASXL1, BMPR2, CBL, FDFT1, FKBP1A, FNTB, FOXO3, FYN, HDAC3, rative disorder LPP, NRAS, NSD1, NUP214, PDPK1, PLAGL2, POLD1, PPAT, PSMB5, disorder RARA, RBM15, STAT3, TOP2A Cancer hereditary hereditary gingival 9.81E-03 RB1, RBBP8, TFDP1, TFDP2 gingival fibromatosis fibromatosis Cancer angiolipoma angiolipoma 1.01E-02 FKBP1A, TSC2 Cancer epithelial epithelial tumor 1.01E-02 AGPAT2, RB1 tumor Cancer extra-adrenal extra-adrenal 1.01E-02 FKBP1A, SDHD paraganglio paraganglioma ma Cancer liver tumor liver tumor 1.10E-02 AURKA, CDKN1A, CDKN1B, CDKN3, EIF2AK2, FKBP1A, GLUD1, GSTO1, HUWE1, MLF1IP, NCAPG (includes EG:64151), PDPK1, POLD1, RARA, TOP2A, TUBA1A, TUBA1C, TUBG1, TXN, UBE2C Cancer gastric gastric cancer 1.15E-02 BAK1, BAX, BRCA1, CCDC28A, CDKN1A, COBLL1, FKBP1A, FOXM1, cancer HDAC3, MAL, MSH3, NAMPT, PA2G4, PHB (includes EG:5245), PRCC, PSMB5, SEC31A, TFDP2, TMEM97, TOP2A, TUBA1A, TUBA1C, TUBG1, UBE2C Cancer hepatocellula hepatocellular 1.21E-02 AURKA, CDKN1A, CDKN1B, CDKN3, EIF2AK2, FKBP1A, GLUD1, r carcinoma carcinoma GSTO1, HUWE1, MLF1IP, NCAPG (includes EG:64151), POLD1, RARA, TOP2A, TUBA1A, TUBA1C, TUBG1, TXN, UBE2C Cancer carcinoma carcinoma 1.46E-02 ACLY, ADORA2B, AGPAT2, ALDOA, ARHGDIA, AURKA, BAK1, BAX, BMP2, BMPR1A, BMPR2, BRF2, C13orf15, C6orf211, CAT, CCDC28A, CDCA5, CDKN1A, CDKN1B, CDKN3, COBLL1, COPS5, CUX1, DAB2, DDHD2, DICER1, EIF2AK2, EIF2C1, EIF4B, EPS15, ETS1, FDFT1, FKBP1A, FNTB, FOXM1, FOXN2, FOXO3, FYN, FZD3, GLUD1, GOLGA5, GPI, GSTO1, HAND1, HDAC3, HIF1A, HPS3, HS3ST3A1, HUWE1, LDHA, LOC728622/SKP1, LRP8, MBNL1, MCM5, MCM7, MIPEP, MLF1IP, MLL3, MSH3, MSX2, MT1X, NCAPG (includes EG:64151), NCOA4, NFYC, NKX2-5, NRAS, ODC1, PA2G4, PDPK1, POLD1, POLR2A, PPARD, PPARG, PPP3CB, PPPDE1, PRCC, PRKAR1A, PSMB5, RARA, RB1, RBM4B, RCAN2, ROCK2, SEC31A, SEC61G, SMEK1, SRD5A1, SRM, STAT5B, STIP1, TFDP2, TMEM97, TOP2A, TRIM24, TRIM33, TUBA1A, TUBA1C, TUBG1, TXN, UBA1, UBE2C, VAV3, WDR19, YWHAE, ZMYM2 Cancer bladder bladder cancer 1.65E-02 ADORA2B, AURKA, BRCA1, CDKN1A, FKBP1A, FNTB, HDAC3, cancer ODC1, PSMB5, RARA, RB1, STAT3, TOP2A, TSC1, TUBA1A, TUBA1C, TUBG1, UBE2C Cancer small-cell small-cell carcinoma 2.03E-02 CDCA5, RB1, TUBA1A, TUBA1C, TUBG1, YWHAE carcinoma Cancer thyroid gland thyroid gland tumor 2.43E-02 ETS1, GOLGA5, HUWE1, NCOA4, NRAS, PPARD, PPARG, PRKAR1A, tumor TOP2A, TRIM24, TRIM33 Cancer testicular testicular cancer 2.55E-02 FKBP1A, GNAI2, LDHA, RARA, RB1, TOP2A, TUBA1A, TUBA1C, cancer TUBG1 Cancer rhabdomyosa rhabdomyosarcoma 2.58E-02 FKBP1A, FYN, SMARCB1, TUBA1A, TUBA1C, TUBG1 rcoma Cancer large-cell large-cell lymphoma 2.68E-02 FKBP1A, FYN, POLD1, TUBA1A, TUBA1C, TUBG1, YWHAZ lymphoma Cancer thyroid thyroid cancer 2.89E-02 CDKN1B, ETS1, FKBP1A, GOLGA5, HUWE1, MCM7, NCOA4, NRAS, cancer PPARD, PPARG, PRKAR1A, STAT3, TOP2A, TRIM24, TRIM33 Cancer type M3 type M3 acute 3.03E-02 POLD1, RARA, TOP2A acute myeloid leukemia myeloid leukemia Cancer Kaposi's Kaposi's sarcoma 3.38E-02 RARA, TOP2A, TUBA1A, TUBA1C, TUBG1 sarcoma Cancer B-cell B-cell leukemia 3.38E-02 BAK1, BAX, BCL2L1, CDK6, CDK7, CDKN1A, FYN, HDAC3, NRAS, leukemia POLD1, PSMB5, RARA, STAT3, TOP2A

195 Cancer benign nerve benign nerve tumor 3.88E-02 CDKN1B, FKBP1A, FNTB, KIF1B, KRIT1, SDHD, TOP2A tumor Cancer medulloblast medulloblastoma 3.95E-02 CDCA7L, CDK6, TOP2A, TUBA1A, TUBA1C, TUBG1 oma Cancer malignant malignant cutaneous 4.33E-02 CDK6, CDK7, FKBP1A, FNTB, FYN, HDAC3, PSMB5, RARA, TUBA1A, cutaneous melanoma TUBA1C, TUBG1 melanoma Cancer chronic B- chronic B-cell 4.39E-02 BAK1, BAX, BCL2L1, CDK6, CDK7, CDKN1A, FYN, HDAC3, POLD1, cell leukemia leukemia PSMB5, RARA, STAT3, TOP2A Gastrointesti colon cancer colon cancer 2.31E-06 ACLY, AKAP13, ARHGDIA, ATF3, AURKA, BAK1, BAX, BMP2, CAT, nal Disease CDCA5, CDK7, CDKN1A, CDKN1B, CDKN3, CEP55, CHMP1A, CNTNAP2, DUSP5, EIF2AK2, EXOSC9, FKBP1A, GNA11, HAND1, HDAC3, HUWE1, KIFC1, LRP8, MCM3, MCM5, MELK, MLF1IP, MSH3, MT1X, NKX2-5, NRAS, ODC1, PDPK1, POLD1, PPARG, PPPDE1, RUVBL1, SET, SRD5A1, STIP1, STXBP6, TMEM97, TPD52L1, TRIAP1, TUBA1A, TUBA1C, TUBG1, UBA1, YWHAE Gastrointesti colon tumor colon tumor 2.70E-04 ACLY, ARHGDIA, AURKA, CAT, HAND1, HDAC3, HUWE1, LRP8, nal Disease PDPK1, POLD1, PPARG, PPPDE1, STIP1, TMEM97, TUBG1, UBA1, YWHAE Gastrointesti colon colon carcinoma 4.29E-04 ACLY, ARHGDIA, CAT, HAND1, HDAC3, HUWE1, LRP8, PDPK1, nal Disease carcinoma POLD1, PPARG, PPPDE1, STIP1, TMEM97, TUBG1, UBA1, YWHAE Gastrointesti colorectal colorectal cancer 9.26E-04 ACLY, AKAP13, ARHGDIA, ATF3, AURKA, BAK1, BAX, BEX4, BMP2, nal Disease cancer CAT, CDCA5, CDK7, CDKN1A, CDKN1B, CDKN3, CEP55, CHMP1A, CNTNAP2, COBLL1, DUSP5, EIF2AK1, EIF2AK2, ENAH, EPB41L3, ERRFI1, EXOSC9, FBXL2, FKBP1A, FNTB, GNA11, HAND1, HDAC3, HIF1A, HMOX1, HUWE1, KIFC1, LDHA, LRP8, MCM3, MCM5, MELK, MLF1IP, MSH3, MT1X, NCAPG (includes EG:64151), NFYC, NKX2-5, NRAS, NUAK1, ODC1, PDPK1, POLD1, PPARD, PPARG, PPP3CB, PPPDE1, PSMB5, RAB31, RUVBL1, SEPT9, SET, SRD5A1, SSBP3, STIP1, STK35, STXBP6, TAOK1, TFDP1, TMEFF2, TMEM97, TOP2A, TPD52L1, TRIAP1, TUBA1A, TUBA1C, TUBG1, UBA1, UBE2C, WARS, YWHAE, ZNF200 Gastrointesti digestive digestive organ 1.20E-03 ACLY, ARHGDIA, AURKA, BAX, CAT, CCDC28A, CDKN1A, CDKN1B, nal Disease organ tumor tumor CDKN3, COBLL1, COPS5, EIF2AK2, EPS15, FKBP1A, FNTB, FYN, GLUD1, GSTO1, HAND1, HDAC3, HIF1A, HUWE1, LRP8, MLF1IP, MLL3, MSH3, MSX2, MTUS1, NCAPG (includes EG:64151), PA2G4, PDPK1, POLD1, PPARG, PPP3CB, PPPDE1, PRCC, PRKCQ, PSMB5, RARA, RB1, SEC31A, STIP1, TFDP2, TMEM97, TOP2A, TUBA1A, TUBA1C, TUBG1, TXN, UBA1, UBE2C, YWHAE Gastrointesti colorectal colorectal carcinoma 1.32E-03 ACLY, ARHGDIA, CAT, COBLL1, FNTB, HAND1, HDAC3, HUWE1, nal Disease carcinoma LRP8, PDPK1, POLD1, PPARG, PPP3CB, PPPDE1, STIP1, TMEM97, TUBG1, UBA1, UBE2C, YWHAE Gastrointesti colorectal colorectal tumor 1.40E-03 ACLY, ARHGDIA, AURKA, CAT, COBLL1, FNTB, HAND1, HDAC3, nal Disease tumor HUWE1, LRP8, PDPK1, POLD1, PPARG, PPP3CB, PPPDE1, STIP1, TMEM97, TUBA1A, TUBA1C, TUBG1, UBA1, UBE2C, YWHAE Gastrointesti gastrointestin gastrointestinal tumor 2.44E-03 BAX, CCDC28A, COBLL1, FKBP1A, FYN, HDAC3, MSH3, PA2G4, nal Disease al tumor PRCC, PRKCQ, SEC31A, TFDP2, TMEM97, TOP2A, TUBA1A, TUBA1C, TUBG1, UBE2C Gastrointesti stomach stomach tumor 2.66E-03 BAX, CCDC28A, COBLL1, FKBP1A, MSH3, PA2G4, PRCC, SEC31A, nal Disease tumor TFDP2, TMEM97, TOP2A, TUBA1A, TUBA1C, TUBG1, UBE2C Gastrointesti gastric gastric carcinoma 5.97E-03 BAX, CCDC28A, COBLL1, MSH3, PA2G4, PRCC, SEC31A, TFDP2, nal Disease carcinoma TMEM97, TOP2A, UBE2C Gastrointesti liver tumor liver tumor 1.10E-02 AURKA, CDKN1A, CDKN1B, CDKN3, EIF2AK2, FKBP1A, GLUD1, nal Disease GSTO1, HUWE1, MLF1IP, NCAPG (includes EG:64151), PDPK1, POLD1, RARA, TOP2A, TUBA1A, TUBA1C, TUBG1, TXN, UBE2C Gastrointesti gastric gastric cancer 1.15E-02 BAK1, BAX, BRCA1, CCDC28A, CDKN1A, COBLL1, FKBP1A, FOXM1, nal Disease cancer HDAC3, MAL, MSH3, NAMPT, PA2G4, PHB (includes EG:5245), PRCC, PSMB5, SEC31A, TFDP2, TMEM97, TOP2A, TUBA1A, TUBA1C, TUBG1, UBE2C Gastrointesti hepatocellula hepatocellular 1.21E-02 AURKA, CDKN1A, CDKN1B, CDKN3, EIF2AK2, FKBP1A, GLUD1, nal Disease r carcinoma carcinoma GSTO1, HUWE1, MLF1IP, NCAPG (includes EG:64151), POLD1, RARA, TOP2A, TUBA1A, TUBA1C, TUBG1, TXN, UBE2C Genetic colon cancer colon cancer 2.31E-06 ACLY, AKAP13, ARHGDIA, ATF3, AURKA, BAK1, BAX, BMP2, CAT, Disorder CDCA5, CDK7, CDKN1A, CDKN1B, CDKN3, CEP55, CHMP1A, CNTNAP2, DUSP5, EIF2AK2, EXOSC9, FKBP1A, GNA11, HAND1, HDAC3, HUWE1, KIFC1, LRP8, MCM3, MCM5, MELK, MLF1IP, MSH3, MT1X, NKX2-5, NRAS, ODC1, PDPK1, POLD1, PPARG, PPPDE1, RUVBL1, SET, SRD5A1, STIP1, STXBP6, TMEM97, TPD52L1, TRIAP1, TUBA1A, TUBA1C, TUBG1, UBA1, YWHAE

196 Genetic colon tumor colon tumor 2.70E-04 ACLY, ARHGDIA, AURKA, CAT, HAND1, HDAC3, HUWE1, LRP8, Disorder PDPK1, POLD1, PPARG, PPPDE1, STIP1, TMEM97, TUBG1, UBA1, YWHAE Genetic lung lung adenocarcinoma 3.59E-04 ALDOA, CDCA5, DICER1, EIF2C1, FOXM1, FOXO3, GPI, LDHA, Disorder adenocarcino NRAS, POLR2A, SMEK1, SRM, UBE2C ma Genetic lung tumor lung tumor 3.71E-04 ALDOA, BAX, BCL2L1, CDCA5, CDKN1A, CDKN1B, CERS2, CERS5, Disorder DICER1, EIF2C1, FOXM1, FOXO3, GPI, HUWE1, LDHA, NRAS, PDPK1, POLR2A, PSMB5, RB1, SEC61G, SMEK1, SRM, TOP2A, TUBA1A, TUBA1C, TUBG1, UBE2C, YWHAE Genetic colon colon carcinoma 4.29E-04 ACLY, ARHGDIA, CAT, HAND1, HDAC3, HUWE1, LRP8, PDPK1, Disorder carcinoma POLD1, PPARG, PPPDE1, STIP1, TMEM97, TUBG1, UBA1, YWHAE Genetic tuberous tuberous sclerosis 7.02E-04 FKBP1A, HMOX1, TSC1, TSC2 Disorder sclerosis Genetic acute acute myeloid 1.86E-03 ASXL1, CBL, FKBP1A, FNTB, FYN, HDAC3, LPP, NRAS, NSD1, Disorder myeloid leukemia NUP214, PDPK1, PLAGL2, POLD1, PPAT, PSMB5, RARA, RBM15, leukemia STAT3, TOP2A Genetic lung lung carcinoma 2.38E-03 ALDOA, CDCA5, DICER1, EIF2C1, FOXM1, FOXO3, GPI, HUWE1, Disorder carcinoma LDHA, NRAS, POLR2A, RB1, SEC61G, SMEK1, SRM, TOP2A, TUBA1A, TUBA1C, TUBG1, UBE2C, YWHAE Genetic ataxia with ataxia with 3.49E-03 APTX, SETX Disorder oculomotor oculomotor apraxia apraxia Genetic Huntington's Huntington's disease 3.67E-03 AHCYL1, ALDH6A1, ATP5C1, ATP5O, ATP6V1B2, B4GALT5, BAX, Disorder disease CYFIP2 (includes EG:26999), DUSP5, FDFT1, FOXN3, GIT1, GPI, GTF3A, HMOX1, LDHA, MAP2K4, METTL9, MFN2, MLF2, MT1X, NDRG3, NME1, OSBPL8, PCDH7, PDCL, PPARG, PPARGC1B, PPRC1, PRDX6, RAB5A, RAB6A, RERE, RHOBTB3, ROCK2, SCAMP5, SCOC, SDC4, SEPHS1, SEPT6, SRD5A1, SRM, SSX2IP, SYNJ1, TBP, TRAK2, USP13, VAMP1, VAMP2, WNK1, YWHAZ, ZBTB44 Genetic early-onset early-onset breast 6.58E-03 BARD1, BRCA1, CHEK2, EI24, RB1CC1 Disorder breast cancer cancer Genetic lung cancer lung cancer 7.00E-03 ALDOA, AURKA, BAX, BCL2L1, BTRC, CDCA5, CDK6, CDKN1A, Disorder CDKN1B, CERS2, CERS5, CHEK2, DICER1, EIF2AK1, EIF2C1, FDFT1, FKBP1A, FNTB, FOXM1, FOXO3, FYN, GPI, GUCY1A3, HDAC3, HIF1A, HIPK2, HUWE1, LDHA, NRAS, PDPK1, POLR2A, PPARG, PSMB5, RARA, RB1, SEC61G, SMEK1, SRM, STK35, TAOK1, TOP2A, TUBA1A, TUBA1C, TUBG1, UBE2C, WNK1, YWHAE Genetic hereditary hereditary gingival 9.81E-03 RB1, RBBP8, TFDP1, TFDP2 Disorder gingival fibromatosis fibromatosis Genetic Charcot- Charcot-Marie-Tooth 1.01E-02 KIF1B, MFN2 Disorder Marie-Tooth disease type 2A1 disease type 2A1 Genetic type 1A type 1A maple syrup 1.93E-02 BCKDHB, DBT Disorder maple syrup urine disease urine disease Genetic lymphangiol lymphangioleiomyo 2.38E-02 FKBP1A, TSC1, TSC2 Disorder eiomyomatos matosis is Genetic type M3 type M3 acute 3.03E-02 POLD1, RARA, TOP2A Disorder acute myeloid leukemia myeloid leukemia Cellular cytokinesis cytokinesis of cell 1.05E-05 CD2AP, CEP55, GNAI2, KIFC1, MASTL, NEDD4L, NME1, SEPT6, Movement lines SEPT9, TOP2A, VAV3 Cellular cytokinesis cytokinesis of 2.90E-05 CD2AP, CEP55, GNAI2, KIFC1, MASTL, NEDD4L, SEPT6, SEPT9, Movement cervical cancer cell VAV3 lines Cellular cytokinesis cytokinesis of tumor 5.15E-05 CD2AP, CEP55, GNAI2, KIFC1, MASTL, NEDD4L, SEPT6, SEPT9, Movement cell lines TOP2A, VAV3 Cellular cytokinesis cytokinesis 5.98E-05 AURKA, CD2AP, CDKN1A, CEP55, DIAPH2, GNAI2, KIFC1, MASTL, Movement NEDD4L, NME1, SEPT6, SEPT9, SETD8, TOP2A, VAV3 Cellular invasion invasion of tumor 5.48E-03 ASAP1, ATF1, ATF3, CAT, CBL, CDC42, CDKN1B, CSK, DAB2, Movement cell lines DIAPH2, DNMT3B, ETS1, FOXM1, GPI, HIF1A, HIPK2, HMGA1, HMOX1, MAPKAPK2, MARCKS, NCOA3, NME1, NUAK1, PA2G4, PPARG, PRKAA1, RND3, SDC4, SEPT9, SKP2, SP1, STAT3, TAB1, TFAP2A, TJP1

197 Cellular invasion invasion of cell lines 6.80E-03 ASAP1, ATF1, ATF3, CAT, CBL, CDC42, CDKN1B, CSK, DAB2, Movement DIAPH2, DNMT3B, ETS1, FOXM1, GPI, HIF1A, HIPK2, HMGA1, HMOX1, MAPKAPK2, MARCKS, NCOA3, NME1, NUAK1, OTUB1, PA2G4, PPARG, PRKAA1, RND3, SDC4, SEPT9, SKP2, SP1, STAT3, TAB1, TFAP2A, TJP1 Cellular invasion invasion of cells 1.03E-02 ASAP1, ATF1, ATF3, B4GALT5, CAT, CBL, CDC42, CDKN1B, CSK, Movement DAB2, DIAPH2, DNMT3B, ETS1, FKBP1A, FOXM1, GPI, HIF1A, HIPK2, HMGA1, HMOX1, MAPKAPK2, MARCKS, NCOA3, NME1, NUAK1, OTUB1, PA2G4, PIP5K1A, PPARG, PRKAA1, RND3, SDC4, SEPT9, SKP2, SP1, STAT3, TAB1, TFAP2A, TJP1 Cellular invasion invasion of 1.19E-02 ASAP1, ATF1, ATF3, B4GALT5, CAT, CBL, CDC42, CDKN1B, CSK, Movement eukaryotic cells DAB2, DIAPH2, DNMT3B, ETS1, FKBP1A, FOXM1, GPI, HIF1A, HIPK2, HMGA1, HMOX1, MAPKAPK2, MARCKS, NCOA3, NME1, NUAK1, OTUB1, PA2G4, PPARG, PRKAA1, RND3, SDC4, SEPT9, SKP2, SP1, STAT3, TAB1, TFAP2A, TJP1 Cellular invasion invasion of colon 1.23E-02 ATF3, CSK, HIF1A, HIPK2, NME1, NUAK1, STAT3 Movement cancer cell lines Cellular invasion invasion of bladder 3.03E-02 CAT, DAB2, MAPKAPK2 Movement cancer cell lines Infectious infection infection of cell lines 1.30E-04 AKAP13, AP1B1, AP2M1, ARHGAP32, ASXL2, ATMIN, ATXN2, Disease BRCA1, C19orf50, CBL, CDC42, CLDND1, CRIPAK, CRTC2, DCP2, DDX50, DHX15, DHX33, DNAJB1, EPS15, ERCC3, EXOSC10, FBXW11, GABPB1, GATAD2A, GOLPH3, HUWE1, INTS7, ISG20L2, KARS, KAT6A, LRRC8D, MED20, MPHOSPH6, MT1X, NCKAP1, NRBP1, NUDT11, NUP214, NUP50, NUP85 (includes EG:287830), OSBPL3, PCBP2, PIP5K1A, POLR2A, POLR2C, PPP2R2A, PPP2R5E, PRKAA1, PRPF8, R3HDM1, RAB5A, RAB6A, RB1CC1, RIMS4, RNF10, RNF216, RNF26, SAP30BP, SEC61G, SF3B2, STIP1, STX5, SYNJ1, TAGLN2, TAOK1, TFDP2, TNPO3, TRIM5, TRMT5, UBE2C, UBE2E1, UBE2H, WASF2, WNK1, YBX1, ZCCHC17, ZNRD1 Infectious infection infection by virus 2.10E-04 AKAP13, AP1B1, AP2M1, ARHGAP32, ASXL2, ATMIN, ATXN2, Disease BRCA1, C19orf50, CBL, CDC42, CLDND1, CRIPAK, CRTC2, DCP2, DDX50, DHX15, DHX33, DNAJB1, EPS15, ERCC3, EXOSC10, FBXW11, FDFT1, GABPB1, GATAD2A, GOLPH3, HUWE1, INTS7, ISG20L2, KARS, KAT6A, LRRC8D, MED20, MPHOSPH6, MT1X, NCKAP1, NRBP1, NUDT11, NUP214, NUP50, NUP85 (includes EG:287830), OSBPL3, PCBP2, PIP5K1A, POLD1, POLR2A, POLR2C, PPARG, PPP2R2A, PPP2R5E, PPP3CB, PPP3R1, PRKAA1, PRPF8, R3HDM1, RAB5A, RAB6A, RARA, RB1CC1, RIMS4, RNF10, RNF216, RNF26, SAP30BP, SEC61G, SF3B2, STIP1, STX5, SYNJ1, TAGLN2, TAOK1, TFDP2, TNPO3, TOP2A, TRIM5, TRMT5, TUBA1A, TUBA1C, TUBG1, UBE2C, UBE2E1, UBE2H, UGCG, WASF2, WNK1, YBX1, ZCCHC17, ZNRD1 Infectious infection infection by 5.17E-04 AKAP13, AP2M1, ARHGAP32, ASXL2, ATMIN, ATXN2, BRCA1, Disease lentivirus C19orf50, CLDND1, CRIPAK, CRTC2, DCP2, DDX50, DHX15, DHX33, DNAJB1, ERCC3, EXOSC10, FBXW11, FDFT1, GABPB1, GATAD2A, GOLPH3, HUWE1, INTS7, ISG20L2, KARS, KAT6A, LRRC8D, MED20, MPHOSPH6, MT1X, NCKAP1, NRBP1, NUDT11, NUP214, NUP50, NUP85 (includes EG:287830), OSBPL3, POLD1, POLR2A, POLR2C, PPARG, PPP2R2A, PPP2R5E, PPP3CB, PPP3R1, PRKAA1, PRPF8, R3HDM1, RAB6A, RARA, RB1CC1, RIMS4, RNF10, RNF216, RNF26, SAP30BP, SEC61G, SF3B2, STIP1, STX5, TAGLN2, TAOK1, TFDP2, TNPO3, TOP2A, TRIM5, TRMT5, TUBA1A, TUBA1C, TUBG1, UBE2C, UBE2E1, UBE2H, UGCG, WNK1, YBX1, ZCCHC17, ZNRD1 Infectious infection infection by HIV 7.39E-04 AKAP13, AP2M1, ARHGAP32, ASXL2, ATMIN, ATXN2, BRCA1, Disease C19orf50, CLDND1, CRIPAK, CRTC2, DCP2, DDX50, DHX15, DHX33, DNAJB1, ERCC3, EXOSC10, FBXW11, FDFT1, GABPB1, GATAD2A, GOLPH3, HUWE1, INTS7, ISG20L2, KARS, KAT6A, LRRC8D, MED20, MPHOSPH6, MT1X, NCKAP1, NRBP1, NUDT11, NUP214, NUP50, NUP85 (includes EG:287830), OSBPL3, POLD1, POLR2A, POLR2C, PPARG, PPP2R2A, PPP2R5E, PPP3CB, PPP3R1, PRKAA1, PRPF8, R3HDM1, RAB6A, RARA, RB1CC1, RIMS4, RNF10, RNF216, RNF26, SAP30BP, SEC61G, SF3B2, STIP1, STX5, TAGLN2, TAOK1, TFDP2, TNPO3, TOP2A, TRMT5, TUBA1A, TUBA1C, TUBG1, UBE2C, UBE2E1, UBE2H, UGCG, WNK1, YBX1, ZCCHC17, ZNRD1 Infectious infection infection by HIV-1 7.82E-04 AKAP13, AP2M1, ARHGAP32, ASXL2, ATMIN, ATXN2, BRCA1, Disease C19orf50, CLDND1, CRIPAK, CRTC2, DCP2, DDX50, DHX15, DHX33, DNAJB1, ERCC3, EXOSC10, FBXW11, GABPB1, GATAD2A, GOLPH3, HUWE1, INTS7, ISG20L2, KARS, KAT6A, LRRC8D, MED20,

198 MPHOSPH6, MT1X, NCKAP1, NRBP1, NUDT11, NUP214, NUP50, NUP85 (includes EG:287830), OSBPL3, POLR2A, POLR2C, PPP2R2A, PPP2R5E, PRKAA1, PRPF8, R3HDM1, RAB6A, RB1CC1, RIMS4, RNF10, RNF216, RNF26, SAP30BP, SEC61G, SF3B2, STIP1, STX5, TAGLN2, TAOK1, TFDP2, TNPO3, TRMT5, UBE2C, UBE2E1, UBE2H, WNK1, YBX1, ZCCHC17, ZNRD1 Infectious infection infection of kidney 3.23E-03 AKAP13, ATMIN, ATXN2, C19orf50, DCP2, DHX15, EXOSC10, Disease cell lines GATAD2A, KARS, KAT6A, MT1X, NCKAP1, NUDT11, NUP214, NUP50, OSBPL3, PCBP2, POLR2A, POLR2C, PPP2R5E, PRPF8, R3HDM1, RNF10, RNF216, SAP30BP, SF3B2, TAGLN2, TNPO3, UBE2C, UBE2H, YBX1, ZCCHC17 Infectious infection infection of tumor 5.36E-03 AP1B1, AP2M1, ARHGAP32, ASXL2, BRCA1, CBL, CDC42, CLDND1, Disease cell lines CRIPAK, CRTC2, DDX50, DHX33, DNAJB1, EPS15, ERCC3, FBXW11, GABPB1, GOLPH3, HUWE1, INTS7, ISG20L2, LRRC8D, MED20, MPHOSPH6, NRBP1, NUP85 (includes EG:287830), PIP5K1A, PPP2R2A, PRKAA1, RAB5A, RAB6A, RB1CC1, RIMS4, RNF216, RNF26, SEC61G, STIP1, STX5, SYNJ1, TAOK1, TFDP2, TNPO3, TRIM5, TRMT5, UBE2E1, WASF2, WNK1, ZNRD1 Infectious infection infection by Dengue 1.96E-02 AP1B1, CBL, EPS15, PIP5K1A, RAB5A, SYNJ1, WASF2 Disease virus 2 Infectious infection infection of cervical 2.26E-02 AP2M1, ARHGAP32, ASXL2, BRCA1, CDC42, CLDND1, CRIPAK, Disease cancer cell lines CRTC2, DDX50, DHX33, DNAJB1, ERCC3, FBXW11, GABPB1, GOLPH3, HUWE1, INTS7, ISG20L2, LRRC8D, MED20, MPHOSPH6, NRBP1, NUP85 (includes EG:287830), PPP2R2A, PRKAA1, RAB5A, RAB6A, RB1CC1, RIMS4, RNF26, SEC61G, STIP1, STX5, TAOK1, TFDP2, TNPO3, TRMT5, UBE2E1, WNK1, ZNRD1 Infectious infectious infectious disorder of 2.30E-03 AKAP13, ATMIN, ATXN2, C19orf50, DCP2, DHX15, EXOSC10, Disease disorder embryonic cell lines GATAD2A, KARS, KAT6A, MT1X, NCKAP1, NUDT11, NUP214, NUP50, OSBPL3, PCBP2, POLR2A, POLR2C, PPP2R5E, PRPF8, R3HDM1, RNF10, RNF216, SAP30BP, SF3B2, TAGLN2, TNPO3, UBE2C, UBE2H, YBX1, ZCCHC17 Infectious infectious infectious disorder of 2.30E-03 AKAP13, ATMIN, ATXN2, C19orf50, DCP2, DHX15, EXOSC10, Disease disorder epithelial cell lines GATAD2A, KARS, KAT6A, MT1X, NCKAP1, NUDT11, NUP214, NUP50, OSBPL3, PCBP2, POLR2A, POLR2C, PPP2R5E, PRPF8, R3HDM1, RNF10, RNF216, SAP30BP, SF3B2, TAGLN2, TNPO3, UBE2C, UBE2H, YBX1, ZCCHC17 Infectious keratoconjun keratoconjunctivitis 2.87E-03 HYAL2, HYAL3, PPP3CB, PPP3R1 Disease ctivitis sicca sicca

Supplementary Table 3: Functional Enrichment Analysis of genes predicted to be miR-182 targets

Ingenuity Canonical -log(p- Ratio* Sig** Molecules Pathways value) Cell Cycle: G2/M 4.54E+00 2.29E-01 > 4SD GADD45A,YWHAE,CDK7,CDKN1A,YWHAZ,TOP2A,BTRC,BRCA1,LOC728 DNA Damage 622/SKP1,CHEK2,SKP2 (includes EG:27401) Checkpoint Regulation Role of BRCA1 in 4.07E+00 2.03E-01 > 4SD RB1,ATF1 (includes DNA Damage EG:100040260),GADD45A,CDKN1A,BARD1,RBBP8,SMARCD2,RFC5,BRCA Response 1,CHEK2,RFC3,NBN Hereditary Breast 3.97E+00 1.50E-01 > 4SD NRAS,POLR2D,TUBG1,BARD1,CDK6,SMARCD2,RFC5,NBN,POLR2G,RB1,P Cancer Signaling OLR2C,POLR2A,HDAC3,GADD45A,CDKN1A,BRCA1,CHEK2,RFC3 Cyclins and Cell Cycle 3.83E+00 1.61E-01 > 4SD TFDP1,PA2G4,PPP2CA,PPP2R2A,CDK7,CDK6,LOC728622/SKP1,SKP2 Regulation (includes EG:27401),RB1,HDAC3,CDKN1A,BTRC,PPP2R5E,CDKN1B Molecular 3.76E+00 1.03E-01 > 4SD MAP2K4,FYN,CDC42,PA2G4,BMP2,TAB2,GNA11,BMPR2,HIF1A,RB1,LAMT Mechanisms of Cancer OR3,BMPR1A,GNA13,BRCA1,HIPK2,TAB1,CHEK2,PAK2,PRKCQ,NRAS,TF DP1,CDK6,AURKA,BAX,BAK1,APH1A,NBN,GNAI2,BCL2L1,CBL,RND3,CD KN1A,FZD3,CDKN1B,ELK1,FNBP1,PRKAR1A Role of CHK Proteins 3.27E+00 2.29E-01 > 4SD RAD17 (includes EG:19356),CDKN1A,TLK2,RFC5,BRCA1,CHEK2,RFC3,NBN in Cell Cycle Checkpoint Control Protein Ubiquitination 3.12E+00 1.08E-01 > 4SD PSMB3,USP21,USP24,UBE2H,ANAPC1/LOC100286979,USO1,USP13,USP42, Pathway USP10,DNAJB1,BRCA1,NEDD4L,PSMB5,MED20,DNAJC9,UBE4B,LOC72862 2/SKP1,SKP2 (includes EG:27401),UBE2G2,DNAJC21,PSME1,CBL,CUL2,UBE2G1,BTRC,UBA1,USP3 4,UBE2E1,UBE2C

199 ATM Signaling 3.09E+00 1.85E-01 > 4SD MAP2K4,GADD45A,CDKN1A,ATF4,TP53BP1,TLK2,CREB5,BRCA1,CHEK2, NBN Aryl Hydrocarbon 3.08E+00 1.28E-01 > 4SD MGST1,TFDP1,MED1 (includes Receptor Signaling EG:19014),GSTA4,CDK6,BAX,GSTO1,NCOA3,RB1,SP1,RARA,NFIA,CDKN1 A,NFIB,CDKN1B,CHEK2,ALDH6A1,MCM7 Cell Cycle: G1/S 2.96E+00 1.69E-01 > 4SD RB1,HDAC3,TFDP1,PA2G4,CDKN1A,CDK6,BTRC,CDKN1B,LOC728622/SKP Checkpoint Regulation 1,SKP2 (includes EG:27401) Clathrin-mediated 2.63E+00 1.14E-01 > 4SD AP2M1,EPS15,RAB5A,CDC42,RAB5B,SH3GL2,AP1B1,CD2AP,CBL,ACTR3, Endocytosis Signaling WASL,PPP3CB,SYNJ1,PPP3R1,RAB5C,ARPC4,CSNK2A1,DAB2,SH3KBP1 14-3-3-mediated 2.58E+00 1.30E-01 > 2SD MAP2K4,TSC1,PRKCQ,NRAS,YWHAE,TUBG1,YWHAZ,BAX,TUBA1A,CBL, Signaling TSC2,TUBA1C,CDKN1B,PDCD6IP,ELK1 Glutamate Metabolism 2.41E+00 2.00E-01 > 3SD EARS2,GNPNAT1,GLUD1,GOT1,GCLM,GOT2,PPAT Cell Cycle Control of 2.37E+00 2.00E-01 > 4SD MCM5,MCM3,CDK7,CDK6,CHEK2,MCM7 Chromosomal Replication NRF2-mediated 2.34E+00 1.06E-01 > 4SD MAP2K4,MGST1,PRKCQ,NRAS,DNAJC9,GSTA4,MAP3K1,GSTO1,TXNRD1, Oxidative Stress MAFG,DNAJC21,HMOX1,STIP1,CAT,ATF4,MAPK7,DNAJB1,TXN (includes Response EG:116484),GCLM,FKBP5 Hypoxia Signaling in 2.34E+00 1.52E-01 > 4SD UBE2G2,UBE2H,COPS5,UBE2G1,ATF4,HIF1A,CREB5,LDHA,UBE2E1,UBE2 the Cardiovascular C System Androgen Signaling 2.24E+00 1.08E-01 > 4SD PRKCQ,POLR2D,CDK7,GNA11,TBP,POLR2G,GNAI2,POLR2A,POLR2C,ERC C3,NCOA4,GNA13,DNAJB1,PRKAR1A Glucocorticoid 2.23E+00 9.39E-02 > 3SD MAP2K4,POLR2D,NFATC3,GTF2F2,SMARCD2,POLR2C,POLR2A,PPP3CB,P Receptor Signaling PP3R1,FOXO3,PRKAA1,FKBP5,STAT5B,TAB1,NRAS,MED1 (includes EG:19014),CDK7,MAP3K1,TBP,STAT3,NCOA3,POLR2G,BCL2L1,ERCC3,CD KN1A,ELK1 PI3K Signaling in B 2.13E+00 1.15E-01 > 4SD FYN,ATF3,NRAS,ATF1 (includes EG:100040260),NFATC3,IRS4 (includes Lymphocytes EG:16370),PDPK1,CBL,PPP3CB,PPP3R1,VAV3,FOXO3,ATF4,PLEKHA1,ELK 1 Rac Signaling 2.12E+00 1.11E-01 > 4SD MAP2K4,ABI2,PAK2,NRAS,CDC42,MAP3K1,PIP5K1A,ACTR3,CYFIP2,ARPC 4,SH3RF1,ELK1,NCKAP1 PI3K/AKT Signaling 2.10E+00 1.09E-01 > 4SD TSC1,NRAS,YWHAE,PPP2CA,PPP2R2A,LIMS1,YWHAZ,PDPK1,BCL2L1,TS C2,CDKN1A,FOXO3,CDKN1B,PPP2R5E Estrogen Receptor 2.10E+00 1.12E-01 > 2SD MED23,NRAS,POLR2D,MED20,GTF2F2,MED1 (includes Signaling EG:19014),CDK7,TBP,NCOA3,POLR2G,POLR2A,HDAC3,POLR2C,ERCC3,M ED24 Polyamine Regulation 2.09E+00 2.08E-01 > 4SD PPARG,PSME1,AZIN1,PSME3,ODC1 in Colon Cancer p53 Signaling 2.07E+00 1.26E-01 > 3SD BCL2L1,RB1,PLAGL1,GADD45A,MED1 (includes EG:19014),CDKN1A,CCNK,GNL3,BAX,HIPK2,BRCA1,CHEK2 mTOR Signaling 2.07E+00 1.10E-01 > 2SD TSC1,PRKCQ,NRAS,PPP2CA,PPP2R2A,PDPK1,FKBP1A,HIF1A,HMOX1,RND 3,EIF4G2,TSC2,PRKAA1,PPP2R5E,FNBP1,EIF4B RAR Activation 2.05E+00 1.05E-01 > 4SD MAP2K4,NSD1,PRKCQ,TRIM24,RDH10,MED1 (includes EG:19014),CDK7,BMP2,CSK,MAP3K1,SMARCD2,PDPK1,RARA,ERCC3,CSN K2A1,MAPKAPK2,STAT5B,PRKAR1A ERK5 Signaling 2.03E+00 1.45E-01 > 4SD NRAS,YWHAE,FOXO3,YWHAZ,ATF4,GNA13,MAPK7,CREB5,WNK1 T Cell Receptor 1.96E+00 1.18E-01 > 1SD MAP2K4,FYN,PRKCQ,NRAS,CBL,PPP3CB,NFATC3,CSK,VAV3,PPP3R1,MA Signaling P3K1,ELK1 Nicotinate and 1.96E+00 1.25E-01 > 3SD MAP2K4,PAK2,PRKCQ,CDK7,PRKAA1,CDK6,DUSP18,NAMPT,MAPK7,EIF Nicotinamide 2AK2,IRAK1,DYRK1A Metabolism RAN Signaling 1.90E+00 2.22E-01 > 4SD KPNA6,TNPO1,KPNA1,IPO5 Mismatch Repair in 1.90E+00 2.00E-01 > 4SD MSH3 (includes EG:17686),RFC5,POLD1,RFC3 Eukaryotes Nucleotide Excision 1.81E+00 1.71E-01 > 4SD POLR2G,POLR2A,POLR2D,POLR2C,CDK7,ERCC3 Repair Pathway Regulation of IL-2 1.79E+00 1.19E-01 > 3SD MAP2K4,FYN,NRAS,PPP3CB,NFATC3,VAV3,PPP3R1,MAP3K1,TOB1,ELK1 Expression in Activated and Anergic T Lymphocytes Antiproliferative Role 1.78E+00 1.92E-01 > 4SD RB1,TOB1,CDKN1B,LOC728622/SKP1,SKP2 (includes EG:27401) of TOB in T Cell Signaling Prostate Cancer 1.75E+00 1.11E-01 > 4SD RB1,NRAS,TFDP1,PA2G4,SRD5A1,CDKN1A,ATF4,PDPK1,CDKN1B,CREB5 Signaling

200 Cardiomyocyte 1.72E+00 2.11E-01 > 4SD NKX2-5,BMPR1A,BMP2,BMPR2 Differentiation via BMP Receptors B Cell Receptor 1.68E+00 1.01E-01 > 2SD MAP2K4,ETS1,PRKCQ,NRAS,CDC42,NFATC3,CSK,MAP3K1,CREB5,BCL2L Signaling 1,PPP3CB,PPP3R1,VAV3,ATF4,ELK1 Small Cell Lung 1.63E+00 1.07E-01 > 4SD BCL2L1,PIAS3,RB1,TFDP1,PA2G4,TRAF4,CDK6,CDKN1B,SKP2 (includes Cancer Signaling EG:27401) Toll-like Receptor 1.62E+00 1.37E-01 > 4SD MAP2K4,MAP3K1,TAB2,EIF2AK2,ELK1,TAB1,IRAK1 Signaling

Supplementary Table 4: Genes involved in the ‘DNA Damage Response’, identified as significantly enriched in our miR-182 biotinylated pull-down

Gene Symbol Log2 FC adj. P-value Functional summary References ATF1 -1.27 0.0003 Belongs to the CREB family of transcription factors, which in Meyer and Habener 1993; response to growth factor/other stress signals activates expression of proto-oncogenes and cell cycle related genes ATF4 -1.09 0.0028 Belongs to the CREB/ATF family of DNA binding proteins that Liang and Hai 1997; Fung, contain leucine zipper regions. It is known to interact with several Liu et al. 2007; general transcription factors such as TFIIB and TBP and acts as a potent transcriptional activator. BARD1 -1.98 0.0002 Partners with BRCA1 to regulate transcription, cell cycle progression Scully and Livingston 2000; and maintenance of X chromosome inactivation; mutant cell lines Ganesan, Silver et al. 2002; with defective BARD1 exhibit genome instability and defective HR- Moynahan, Chiu et al. 1999; mediated DNA repair Scully, Ganesan et al. 1999; BRCA1 -1.08 0.0011 Mutations in BRCA1 predispose to breast cancer. BRCA1 deficiency Deng 2006; activates the DNA damage response, causing growth arrest and, in some cases, apoptosis. BTRC -0.64 0.0085 Is part of the E3 ubiquitin ligase complex and binding of BTRC Busino, Donzelli et al. 2003; results in ubiquitination and subsequent proteasomal degradation of its substrates. In response to DNA damage, it degrades Cdc25A leading to an intra-S-phase delay in the cell cycle. CDK6 -1.01 0.0013 Along with CDK4, is known to phosphorylate Rb, preventing Harper, Elledge et al. 1995; quiescence, senescence and cell differentiation. Upon DNA damage, Shapiro, Edwards et al. failure to inhibit these CDKs by p16 or p21 leads to accumulation of 1998; mutations and genomic instability. CDK7 -1.63 0.0002 Identified as the catalytic subunit of the CDK-activating kinase Roy, Adamczewski et al. (CAK) complex that includes cyclin H and MAT1. It is also a 1994; Kaldis 1999; component of the general transcription factor TFIIH. CDKN1A -1.373 0.001 Regulates cell cycle, apoptosis and transcription in response to DNA Dotto 2000; Coqueret 2003; Damage CDKN1B -1.48 0.0003 CDKN1B controls cell cycle progression at the G1 checkpoint by Polyak, Lee et al. 1994; inhibiting the cyclin E-CDK2 and cyclin D-CDK4 complexes. This occurs through the activation of Rb, which leads to a G1 arrest. CHEK2 -1.159 0.0009 Protein kinase activated in response to DNA damage, plays a crucial Matsuoka, Huang et al. role in cell cycle arrest. Phosphorylates and activates BRCA1, 1998; Lee, Collins et al. thereby regulating DNA Damage Response pathway 2000 CREB5 -1.25 0.0005 A member of the CREB/ATF family of DNA-binding proteins that Zu, Maekawa et al. 1993; binds to the cAMP response element as a homodimer or as a heterodimer with c-Jun, functioning as a transactivator. GADD45a -1.139 0.001 Activated upon exposure of cells to UV, ionizing radiation and Papathanasiou, Kerr et al. alkylating agents; plays a major role in maintenance of genomic 1991; Hollander and integrity and in control of centrosome duplication. Fornace 2002; HDAC3 -2.8 0.00006 Is a class I histone deacetylase which represses transcription when Bhaskara, Knutson et al. recruited to a gene promoter. It is required for efficient DNA repair 2010; and genomic stability. MAP2K4 -1.74 0.0002 Also known as MEK4, this serine/threonine kinase is a direct Lin, Minden et al. 1995; activator of MAP kinases and has been shown to activate JNK1, JNK2 and p38 in response to environmental stress or mitogenic stimuli. NBN -1.71 0.0007 It is a component of the MRN complex that governs cellular response Zhang, Zhou et al. 2006; to DNA damage. Its N-terminus binds ?H2AX, and is believed to recruit the MRN complex to the proximity of the DNA double-strand break. Also known as NBS1. NRAS -1.17 0.0007 It is a member of the Ras family of proteins that have GTPase Colicelli 2004; activity. Mutations in this gene have been associated with several cancers and Noonan syndrome.

201 PA2G4 -1.28 0.0036 It is an RNA-binding protein involved in growth regulation. It Yoo, Wang et al. 2000; Xia, interacts with the ErbB3 receptor and modulates cell signaling. It Cheng et al. 2001; also binds Rb and inhibits E2F1-mediated transcription. POLR2A -1.15 0.0042 These are four subunits of the 12-component eukaryotic DNA- Myer and Young 1998; directed RNA polymerase II that catalyzes the transcription of DNA Starita and Parvin 2003; to mRNA. BRCA1 is known to interact with RNA polymerase II and regulate gene transcription. POLR2C -1.05 0.0013 These are four subunits of the 12-component eukaryotic DNA- Myer and Young 1998; directed RNA polymerase II that catalyzes the transcription of DNA Starita and Parvin 2003; to mRNA. BRCA1 is known to interact with RNA polymerase II and regulate gene transcription. POLR2D -2.04 0.0001 These are four subunits of the 12-component eukaryotic DNA- Myer and Young 1998; directed RNA polymerase II that catalyzes the transcription of DNA Starita and Parvin 2003; to mRNA. BRCA1 is known to interact with RNA polymerase II and regulate gene transcription. POLR2G -0.83 0.0051 These are four subunits of the 12-component eukaryotic DNA- Myer and Young 1998; directed RNA polymerase II that catalyzes the transcription of DNA Starita and Parvin 2003; to mRNA. BRCA1 is known to interact with RNA polymerase II and regulate gene transcription. PPP2CA -0.88 0.0088 These are subunits of the Serine/threonine-protein phosphatase 2A Li, Cai et al. 2007; Mumby (PP2A) family. The dynamic equilibrium that is maintained between 2007; PP2A and the CDKs controls phosphorylation of the RB family of proteins. PPP2R5E is part of a sub-family that is important in the DNA damage-induced stabilization of TP53. PPP2R2A -1.08 0.0018 These are subunits of the Serine/threonine-protein phosphatase 2A Li, Cai et al. 2007; Mumby (PP2A) family. The dynamic equilibrium that is maintained between 2007; PP2A and the CDKs controls phosphorylation of the RB family of proteins. PPP2R5E is part of a sub-family that is important in the DNA damage-induced stabilization of TP53. PPP2R5E -1.03 0.009 These are subunits of the Serine/threonine-protein phosphatase 2A Li, Cai et al. 2007; Mumby (PP2A) family. The dynamic equilibrium that is maintained between 2007; PP2A and the CDKs controls phosphorylation of the RB family of proteins. PPP2R5E is part of a sub-family that is important in the DNA damage-induced stabilization of TP53. RAD17 -1.45 0.0004 A cell cycle checkpoint gene that is bound to chromatin prior to Rauen, Burtelow et al. 2000; DNA damage. Upon damage, it is phosphorylated by ATR, after Bao, Tibbetts et al. 2001; which it recruits the RAD1-RAD9a-HUS1 complex to chromatin inducing cell cycle G2 arrest RB1 -0.805 0.008 Induces cell cycle arrest at the G1/S checkpoint upon DNA Damage; Harrington, Bruce et al. important for an intra-S-phase arrest in response to DNA damage by 1998; Knudsen, Booth et al. cisplatin inducing genotoxic stress 2000; RBBP8 -0.67 0.0067 Also known as CtIP. Associates with BRCA1 and is thought to Jhanwar-Uniyal 2003; Yun regulate BRCA1-dependent transcriptional regulation, DNA repair and Hiom 2009; and cell cycle checkpoint control. Replication -1.147 0.0007 RFC is a 5-subunit containing clamp loader, which aids in PCNA Waga and Stillman 1998; Factor C assembly onto primed DNA for its replication. Hubscher, Maga et al. 2002; (RFC3, 5) Garg and Burgers 2005; SKP1 -1.92 0.0002 Is also a component of the E3 ubiquitin ligase complex and the SCF Donzelli, Squatrito et al. complex (that includes BTRC), which is responsible for degradation 2002; of Cdc25A and cell cycle delay. SKP2 -1.79 0.0001 Member of the F-box protein family that form a part of the SCF Carrano, Eytan et al. 1999; complex that is known to promote proteasomal degradation of p27kip1/CDKN1B in the S and G2 phases of the cell cycle. SMARCD2 -1.342 0.0005 Is a component of the human SWI/SNF-like chromatin-remodeling Wilson and Roberts 2011; protein complexes which possess tumour suppressive activity, with recurrent mutations of its different subunits present in various cancers TFDP1 -0.92 0.0064 Also known as DP1, this family of transcription factors Helin, Wu et al. 1993; heterodimerize with E2F proteins and enhance their DNA-binding ability, resulting in increased transcription of E2F target genes involved in cell cycle progression. TLK2 -1.39 0.0005 Member of the Tousled-like kinase family of serine/threonine kinases Sillje and Nigg 2001; that are involved in the regulation of chromatin assembly. TOP2A -0.9 0.009 A DNA topoisomerase that is involved in chromosome condensation Watt and Hickson 1994; and regulation of torsional stress during DNA transcription and replication. TP53BP1 -0.99 0.0052 It is a DNA-damage response factor involved in both homologous Wang, Matsuoka et al. 2002; recombination and nonhomologous end joining. It is also known to be an activator of p53.

202 TUBG1 -0.91 0.0022 A member of the tubulin family of proteins that are localized to the Zheng, Jung et al. 1991; centrosome where they mediate microtubule formation, nucleation and are hence required for cell cycle progression. YWHAE -2.77 0.0083 These belong to the 14-3-3 family of signal transducers that function Hermeking and Benzinger in several points of the G1S and G2M transition by modulating the 2006; function of several proteins such as Cdc25. YWHAZ -1.62 0.0014 These belong to the 14-3-3 family of signal transducers that function Hermeking and Benzinger in several points of the G1S and G2M transition by modulating the 2006; function of several proteins such as Cdc25.

Supplementary Table 5: Sample information for human breast cancer patient samples used in the miR-182 q-PCR screening analysis

Sample ID Invasive Type Size Grade Breast cancer Subtype Lymph Node Metastasis Q075 Invasive Ductal Carcinoma of No Special Type 43 3 Triple Negative N/A Q087 Invasive Ductal Carcinoma of No Special Type 28 2 HER2 Positive Positive Q132 Invasive Ductal Carcinoma of No Special Type 17 3 Luminal Positive Q151 Invasive Ductal Carcinoma of No Special Type 3 3 Triple Negative Negative Q173 Invasive Ductal Carcinoma of No Special Type 28 3 HER2 Positive Negative Q179 Invasive Ductal Carcinoma of No Special Type 20 3 HER2 Positive Positive Q184 Invasive Ductal Carcinoma of No Special Type 38 2 Triple Negative Negative Q214 Normal N/A N/A N/A N/A Q215 Normal N/A N/A N/A N/A Q218 Invasive Ductal Carcinoma of No Special Type 30 3 Triple Negative Negative Q222 Invasive Ductal Carcinoma of No Special Type 13 3 Triple Negative Positive Q227 Invasive Ductal Carcinoma of No Special Type 30 3 Triple Negative Positive Q230 Invasive Ductal Carcinoma of No Special Type 15 3 Triple Negative Negative Q232 Invasive Ductal Carcinoma of No Special Type 30 3 Luminal Positive Q234 Invasive Ductal Carcinoma of No Special Type 12 3 Triple Negative N/A Q240 Invasive Ductal Carcinoma of No Special Type 50 2 Luminal Positive Q248 Mixed ductolobular 16 2 Luminal Negative Q251 Normal N/A N/A N/A N/A Q254 Mixed ductolobular 130 2 Luminal Positive Q261 Normal N/A N/A N/A N/A Q268 Invasive Ductal Carcinoma of No Special Type 21 3 Triple Negative Negative Q279 Invasive Ductal Carcinoma of No Special Type 40 3 Triple Negative Positive Q281 Invasive Ductal Carcinoma of No Special Type 45 3 Triple Negative Negative Q293 Normal N/A N/A N/A N/A Q304 Normal N/A N/A N/A N/A Q307 Normal N/A N/A N/A N/A Q311 Invasive Ductal Carcinoma of No Special Type 45 3 HER2 Positive Positive Q356 Mixed ductolobular 60 3 Luminal Positive Q359 Invasive Ductal Carcinoma of No Special Type 45 3 Luminal Negative Q381 Invasive Lobular Carcinoma 12 1 N/A Negative Q382 Invasive Ductal Carcinoma of No Special Type 21 3 Luminal Positive Q407 Invasive Ductal Carcinoma of No Special Type 30 3 Triple Negative Positive Q437 Invasive Ductal Carcinoma of No Special Type 17 3 Triple Negative Positive Q441 Invasive Ductal Carcinoma of No Special Type 15 3 Triple Negative Negative Q453 Invasive Lobular Carcinoma 39 2 N/A Negative Q486 Invasive Ductal Carcinoma of No Special Type 45 3 Triple Negative Positive Q488 Invasive Ductal Carcinoma of No Special Type 75 3 Luminal Positive Q506 Invasive Ductal Carcinoma of No Special Type 20 3 Triple Negative Negative Q509 Invasive Ductal Carcinoma of No Special Type 25 3 Triple Negative Positive Q517 Invasive Lobular Carcinoma 19 2 N/A Positive Q543 Mixed ductolobular 40 3 Triple Negative Positive

Supplementary Table 6: Primer sequences

Oligo Name Sequence Use miR-182_pos 5’-UUUGGCAAUGGUAGAACUCACACU-Biotin (C6 leave biotinylated miRNA for pull-down blankr arm)-3’ miR-182_neg 5’-UGUGAGUUCUACCAUUGCCAUAAG-3’ passenger strand of miRNA for pull-down CHEK2_fwd AGTGGTGGGGAATAAACGCC Real time analysis of gene expression CHEK2_rev TCTGGCTTTAAGTCACGGTGTA Real time analysis of gene expression

203 TP53_fwd CCAGGGCAGCTACGGTTTC Real time analysis of gene expression TP53_rev CTCCGTCATGTGCTGTGACTG Real time analysis of gene expression BRCA1_fwd ACAAATACTCATGCCAGCTCATT Real time analysis of gene expression BRCA1_rev GGCTCCTTGCTAAGCCAGG Real time analysis of gene expression ATMIN_fwd GTGCACATGCGGCTGTCCCT Real time analysis of gene expression ATMIN_rev TGGGTCCCTGTGTTCTGCAGGT Real time analysis of gene expression FOXO1_fwd CTCGGCGGGCTGGAAGAATTCA Real time analysis of gene expression FOXO1_rev TGTTGTTGTCCATGGATGCAGCTC Real time analysis of gene expression ATF1_fwd AGATACACGGGGCAGAAAAGG Real time analysis of gene expression ATF1_rev TCCGCTGCTAGTCTGATAGATG Real time analysis of gene expression miR-182_fwd GATTCAGCTAGCGGCCGCCTGCAGGAAGGACCTTGTCGCA Forward primer to amplify miR-182 from genomic DNA miR-182_rev GGAAACAAGCTTTGTCACTTCCAGCTGCACCTGCCCTC Reverse primer to amplify miR-182 hairpin from genomic DNA MAX_fwd CTAGTGTGTGTGTGGGGGGGACTCGGCTTGTTGTTGTCGGT Top strand of luciferase construct with binding GACTTCCCCCTCCCCTTCA site MAX_rev TCGATGAAGGGGAGGGGGAAGTCACCGACAACAACAAGC Bottom strand of luciferase construct CGAGTCCCCCCCACACACACA PRKAG1_fwd CTAGACAGATTGGCACCTATGCCAATATTGCTATGGTTCGC Top strand of luciferase construct with binding ACTACCACCCCCGTCTATG site PRKAG1_rev TCGACATAGACGGGGGTGGTAGTGCGAACCATAGCAATAT Bottom strand of luciferase construct TGGCATAGGTGCCAATCTGT SMARCD3_fwd CTAGAGATTCCCCAGCGCCTCACAGCCCTGCTATTGCCCCC Top strand of luciferase construct with binding TGACCCAATTGTCATCAAC site SMARCD3_rev TCGAGTTGATGACAATTGGGTCAGGGGGCAATAGCAGGGC Bottom strand of luciferase construct TGTGAGGCGCTGGGGAATCT EIF4G2_fwd CTAGTCGTTCCTAATGAATAAAAATCAAGTGCCAAAGCTTC Top strand of luciferase construct with binding AGCCCCAGATAACTATGAT site EIF4G2_rev TCGAATCATAGTTATCTGGGGCTGAAGCTTTGGCACTTGAT Bottom strand of luciferase construct TTTTATTCATTAGGAACGA CASP7_fwd CTAGTGTTTTGGCTTTATGTGCAAAATCTGTTATAGCTTTA Top strand of luciferase construct with binding AAATATATCTGGAACTTTT site CASP7_rev TCGAAAAAGTTCCAGATATATTTTAAAGCTATAACAGATTT Bottom strand of luciferase construct TGCACATAAAGCCAAAACA PRKCD_fwd CTAGAGTGTCTGAGGTGACCGTGGGTGTGTCGGTGCTGGC Top strand of luciferase construct with binding CGAGCGCTGCAAGAAGAACA site PRKCD_rev TCGATGTTCTTCTTGCAGCGCTCGGCCAGCACCGACACACC Bottom strand of luciferase construct CACGGTCACCTCAGACACT FKBP1A_fwd CTAGAGCACCATTTATGAGTCTCAAGTTTTATTATTGCAAT Top strand of luciferase construct with binding AAAAGTGCTTTATGCCGGC site FKBP1A_rev TCGAGCCGGCATAAAGCACTTTTATTGCAATAATAAAACTT Bottom strand of luciferase construct GAGACTCATAAATGGTGCT CHEK2_fwd CTAGATATCCAGCTCCTCTACCAGCACGATGCCAAACTCCA Top strand of luciferase construct with binding GCCAGTCCTCTCACTCCAG site CHEK2_rev TCGACTGGAGTGAGAGGACTGGCTGGAGTTTGGCATCGTG Bottom strand of luciferase construct CTGGTAGAGGAGCTGGATAT PDPK1_fwd CTAGAGAAGCTGTATTTCGGCCTTAGTTATGCCAAAAATGG Top strand of luciferase construct with binding AGAACTACTTAAATATATT site PDPK1_rev TCGAAATATATTTAAGTAGTTCTCCATTTTTGGCATAACTA Bottom strand of luciferase construct AGGCCGAAATACAGCTTCT CDKN1B_fwd CTAGTTTGTAATGTGTGAAAAAGATGCCAATTATTGTTACA Top strand of luciferase construct with binding CATTAAGTAATCAATAAAG site CDKN1B_rev TCGACTTTATTGATTACTTAATGTGTAACAATAATTGGCAT Bottom strand of luciferase construct CTTTTTCACACATTACAAA RBPJ_fwd CTAGGGAGTTGAAAAATGGAAGAATTATTTGCCAAAAGAG Top strand of luciferase construct with binding GAGGACAAAAGATAATATGC site RBPJ_rev TCGAGCATATTATCTTTTGTCCTCCTCTTTTGGCAAATAATT Bottom strand of luciferase construct CTTCCATTTTTCAACTCC PRKAG2_fwd CTAGAAGCCCTGATCCTCACACCAGCAGGTGCCAAACAAA Top strand of luciferase construct with binding AGGAGACAGAAACGGAGTGA site PRKAG2_rev TCGATCACTCCGTTTCTGTCTCCTTTTGTTTGGCACCTGCTG Bottom strand of luciferase construct GTGTGAGGATCAGGGCTT NFKBIB_fwd CTAGCGACCCCCGCCCCGTGTGATTTGTTTCATTGTTAATA Top strand of luciferase construct with binding TAATTTCCAGTTTAATAAA site NFKBIB_rev TCGATTTATTAAACTGGAAATTATATTAACAATGAAACAA Bottom strand of luciferase construct ATCACACGGGGCGGGGGTCG ATF1_fwd CTAGCTAAATTTTCTAAATAACCAATAGTTGCCAATCTAAA Top strand of luciferase construct with binding TGGCAGAGAAGATGAAATT site

204 ATF1_rev TCGAAATTTCATCTTCTCTGCCATTTAGATTGGCAACTATT Bottom strand of luciferase construct GGTTATTTAGAAAATTTAG BARD1_fwd CTAGAGCAGGCTCAACAGAGAACAGCTGTTGCCAAAGCTG Top strand of luciferase construct with binding TTTGATGGATGCTACTTCTA site BARD1_rev TCGATAGAAGTAGCATCCATCAAACAGCTTTGGCAACAGC Bottom strand of luciferase construct TGTTCTCTGTTGAGCCTGCT CREB5_fwd CTAGAGCTGAAGTGGGGGGTAAGGCCAAATTGCCAACACT Top strand of luciferase construct with binding TGTTAAAAGATTAATACTCT site CREB5_rev TCGAAGAGTATTAATCTTTTAACAAGTGTTGGCAATTTGGC Bottom strand of luciferase construct CTTACCCCCCACTTCAGCT RAD17_fwd CTAGGTGTATAAAGTGTGTTTGAACATTATGCCAAATATCA Top strand of luciferase construct with binding AGATGTGAAGGACTAATTC site RAD17_rev TCGAGAATTAGTCCTTCACATCTTGATATTTGGCATAATGT Bottom strand of luciferase construct TCAAACACACTTTATACAC TP53BP1_fwd CTAGCAAGGGTTGTGTCTTCAAAAGGAAATGCCAAAAAAA Top strand of luciferase construct with binding GAATGCTCAGAAGCTATGGA site TP53BP1_rev TCGATCCATAGCTTCTGAGCATTCTTTTTTTGGCATTTCCTT Bottom strand of luciferase construct TTGAAGACACAACCCTTG

205

2. APPENDIX

Supplementary Material for Chapter 3

Supplementary Table 1: Sample information for human breast cancer patient samples used in the miR-139-5p q-PCR screening analysis

Supplementary Table 2: Significantly enriched probes, predicted to be miR-139-5p targets

probe_ID geneSymbol logFC Avg adj.P. ILMN_1663444 LIN7B -4.26 10.98 0.00 Exp Value ILMN_2157435 DYNLRB1 -3.40 13.56 0.00 ILMN_2087528 CPSF3 -5.47 12.19 0.00 ILMN_1756999 RBL2 -4.42 10.53 0.00 ILMN_1668179 HNRNPF -5.37 10.39 0.00 ILMN_1794692 DNMT3B -3.26 9.90 0.00 ILMN_2278636 CUTL1 -4.03 11.91 0.00 ILMN_1658053 DYNLRB1 -3.18 12.99 0.00 ILMN_2328972 DNMT3B -3.89 10.55 0.00 ILMN_1766762 DYNLRB1 -2.83 11.34 0.00 ILMN_1752249 FAM38A -4.10 12.58 0.00 ILMN_1659845 KIAA0355 -4.77 10.49 0.00 ILMN_1676010 SP1 -4.38 11.82 0.00 ILMN_2095840 MYST3 -3.44 11.38 0.00 ILMN_1653828 CHFR -3.83 12.07 0.00 ILMN_2307450 ZNF302 -3.63 11.88 0.00 ILMN_1705746 LSG1 -4.67 11.48 0.00 ILMN_1756402 TMEM177 -3.50 11.23 0.00 ILMN_1746408 MIDN -4.60 13.00 0.00 ILMN_1662038 LARGE -3.14 11.56 0.00

207 ILMN_1780773 LOC400027 -3.89 12.50 0.00 ILMN_1740819 STARD7 -2.71 13.67 0.00 ILMN_1754531 AP4E1 -3.68 11.72 0.00 ILMN_1781479 SUV39H1 -3.12 11.78 0.00 ILMN_2188204 ATG12 -3.11 12.90 0.00 ILMN_1764549 UBE3A -2.59 11.45 0.00 ILMN_1671933 CLCC1 -2.86 12.40 0.00 ILMN_2391141 UBE3A -2.34 11.11 0.00 ILMN_1723211 L2HGDH -2.54 10.63 0.00 ILMN_2105983 XRCC5 -3.28 10.44 0.00 ILMN_1732410 SLC16A9 -3.42 10.08 0.00 ILMN_1730888 ZNF680 -2.22 10.76 0.00 ILMN_1669696 ZNF792 -3.36 9.28 0.00 ILMN_1774604 PNKD -2.71 12.48 0.00 ILMN_1703564 DYNLRB1 -3.73 12.38 0.00 ILMN_1907095 ONECUT2 -2.12 9.61 0.00 ILMN_1684594 USP24 -3.82 10.52 0.00 ILMN_1674231 CHAF1B -3.30 10.90 0.00 ILMN_1736340 ANGEL2 -3.66 10.73 0.00 ILMN_1803745 SUOX -4.28 11.53 0.00 ILMN_2320336 CLK3 -3.77 12.27 0.00 ILMN_1670263 C1orf71 -2.76 11.22 0.00 ILMN_3307901 GAN -3.56 10.68 0.00 ILMN_1691432 PRDM4 -2.62 13.29 0.00 ILMN_1664931 GTF2E2 -3.55 13.43 0.00 ILMN_1760620 TMEM33 -2.91 9.00 0.00 ILMN_1658847 MGC61598 -3.48 12.73 0.00 ILMN_1744023 MGC18216 -2.66 11.32 0.00 ILMN_2205050 PRKX -2.66 9.64 0.00 ILMN_1797342 FNBP1 -2.71 11.36 0.00 ILMN_1800626 SESN1 -3.55 11.07 0.00 ILMN_1777881 TSPAN17 -2.28 13.10 0.00 ILMN_1719627 SLC27A3 -3.68 11.17 0.00 ILMN_2394027 CLK3 -3.52 10.38 0.00 ILMN_1788251 SNN -3.37 9.50 0.00 ILMN_2089073 ATP9A -2.38 13.81 0.00 ILMN_1697448 TXNIP -2.85 13.27 0.00 ILMN_1673898 ATG12 -3.10 9.71 0.00 ILMN_3307719 ZNF490 -3.90 9.64 0.00 ILMN_1723481 CHST3 -1.76 8.34 0.00 ILMN_1777061 ZSWIM6 -2.56 10.62 0.00 ILMN_3240321 AEN -2.68 11.37 0.00 ILMN_1786396 ZZEF1 -3.31 9.64 0.00 ILMN_1707175 NSD1 -2.12 8.85 0.00 ILMN_1665761 BCL11B -3.43 9.23 0.00 ILMN_1738821 GOLGA2 -2.15 9.28 0.00 ILMN_1672834 SSH2 -3.91 10.10 0.00 ILMN_2344216 STX2 -2.77 10.29 0.00 ILMN_1785356 DENND5A -2.31 10.78 0.00 ILMN_1709257 DSCR6 -2.16 11.17 0.00 ILMN_1662719 GPBP1L1 -2.94 10.42 0.00 ILMN_1670124 LOC150223 -1.97 8.92 0.00 ILMN_1748831 PPP1R13B -2.84 11.29 0.00 ILMN_2404049 RBM38 -1.86 12.38 0.00 ILMN_1805992 KIAA1598 -2.76 12.26 0.00 ILMN_1711005 CDC25A -2.14 11.24 0.00 ILMN_1779376 GSK3B -3.06 13.53 0.00 ILMN_1775761 TSR1 -2.35 9.83 0.00 ILMN_1764177 JARID2 -2.69 10.95 0.00 ILMN_2206812 CCNJ -1.98 8.69 0.00 ILMN_2165867 DHCR7 -2.92 12.65 0.00 ILMN_1741334 IKZF4 -2.11 8.67 0.00 ILMN_1693421 RPN2 -2.95 10.20 0.00 ILMN_1810712 ARHGEF12 -2.57 8.84 0.00 ILMN_1805271 ZNF721 -3.21 12.06 0.00 ILMN_3238048 LOC730324 -2.72 11.44 0.00 ILMN_3245413 DENND5A -2.23 10.09 0.00 ILMN_1681304 PAN3 -2.73 11.27 0.00 ILMN_1815626 DHCR7 -2.71 12.61 0.00 ILMN_1663532 RIC8B -1.89 10.25 0.00 ILMN_1735499 DCBLD2 -3.27 11.19 0.00 ILMN_1742705 SLC39A11 -2.48 9.65 0.00 ILMN_1782829 GLTSCR1 -3.83 9.90 0.00 ILMN_1710078 TMEM181 -3.49 11.70 0.00 ILMN_1814282 ISG20L1 -2.37 12.34 0.00 ILMN_1676629 INSIG2 -2.63 10.01 0.00 ILMN_1718297 EML4 -2.79 11.93 0.00 ILMN_1682864 SPSB3 -1.87 10.78 0.00 ILMN_2339655 MAPK7 -2.00 8.70 0.00 ILMN_1664034 ZNF485 -3.34 10.59 0.00 ILMN_1687567 CUX1 -2.79 8.95 0.00 ILMN_1711543 C14orf169 -3.54 11.99 0.00 ILMN_1733690 AKAP7 -2.40 8.88 0.00 ILMN_1775182 GSR -2.47 10.73 0.00 ILMN_1696675 CES2 -3.14 10.02 0.00 ILMN_1760303 PIK3R1 -2.68 11.49 0.00 ILMN_1669905 DCP2 -2.81 12.38 0.00 ILMN_1660079 RNF44 -2.68 10.60 0.00 ILMN_2186061 PFKFB3 -3.43 11.11 0.00 ILMN_1803652 C9orf91 -2.20 11.22 0.00 ILMN_1735360 SDAD1 -3.03 13.11 0.00 ILMN_2164081 KLHL12 -1.79 11.21 0.00 ILMN_1705310 VEZF1 -2.64 13.00 0.00 ILMN_1814737 LNPEP -2.71 9.58 0.00 ILMN_1719064 KCTD10 -2.91 10.10 0.00 ILMN_2126423 ZNF480 -2.24 11.45 0.00 ILMN_2374244 DYRK2 -2.54 13.03 0.00 ILMN_1708081 LCLAT1 -3.70 10.25 0.00 ILMN_1739222 ETV5 -4.30 11.25 0.00 ILMN_2290118 MEGF9 -2.88 9.73 0.00 ILMN_1696311 IMPAD1 -3.52 11.07 0.00 ILMN_1807873 SNX6 -2.35 10.97 0.00 ILMN_1705985 PIGA -3.52 12.71 0.00 ILMN_1801605 BIRC6 -2.44 9.61 0.00 ILMN_1682699 PBX2 -2.71 10.09 0.00 ILMN_2362681 CES2 -2.74 8.90 0.00 ILMN_1747775 STX2 -3.14 10.86 0.00 ILMN_1673950 STBD1 -2.01 8.60 0.00 ILMN_2405602 OSBPL1A -2.88 11.45 0.00 ILMN_1730539 NPHP3 -2.34 8.61 0.00 ILMN_2055760 KIAA1715 -3.07 10.77 0.00 ILMN_1711894 MYB -2.81 12.69 0.00 ILMN_1766094 MOSPD2 -3.17 11.31 0.00 ILMN_1809484 TMOD3 -1.71 9.40 0.00 ILMN_1773865 HSPA5 -3.04 10.05 0.00 ILMN_2256359 HSZFP36 -3.13 12.05 0.00 ILMN_1786125 CCNA2 -3.07 12.72 0.00 ILMN_1705151 SF3A3 -2.20 11.43 0.00 ILMN_1658425 DAG1 -2.87 10.91 0.00 ILMN_1673682 GATAD2A -1.87 13.86 0.00 ILMN_2053567 FASTKD2 -2.71 9.99 0.00 ILMN_1763228 MEF2D -1.82 8.98 0.00 ILMN_1709623 MAPK7 -2.28 10.07 0.00 ILMN_1712298 ANKRD46 -2.49 11.30 0.00 ILMN_1798354 PAPOLA -2.71 12.53 0.00 ILMN_1755954 CPEB3 -2.48 9.69 0.00 ILMN_1783350 PCNXL3 -2.97 11.47 0.00 ILMN_1672660 MBP -3.21 10.26 0.00 ILMN_1804562 SLC31A1 -4.04 10.79 0.00 ILMN_2174127 DCBLD2 -3.11 11.68 0.00

208 ILMN_1654385 ASB13 -2.50 10.08 0.00 ILMN_1786834 PRKX -2.11 8.96 0.00 ILMN_1717094 ZNF618 -1.53 9.12 0.00 ILMN_1751615 COQ10B -1.64 12.82 0.00 ILMN_1794588 DYRK2 -2.49 10.49 0.00 ILMN_1811579 HOMER3 -1.56 10.39 0.00 ILMN_1738383 EEF2 -2.11 14.35 0.00 ILMN_3186473 LCLAT1 -3.30 9.26 0.00 ILMN_1735151 EIF5A2 -1.68 9.29 0.00 ILMN_1779448 EFHD1 -2.22 11.89 0.00 ILMN_1700231 IP6K1 -1.75 10.16 0.00 ILMN_1773427 KANK1 -1.97 9.26 0.00 ILMN_1733155 GIT1 -2.79 11.24 0.00 ILMN_1669064 ZNF823 -3.11 10.85 0.00 ILMN_2196569 NUP93 -1.87 10.37 0.00 ILMN_1820244 -3.37 10.15 0.00 ILMN_1709479 YAP1 -2.48 9.47 0.00 ILMN_2173294 FLJ10916 -2.09 10.13 0.00 ILMN_2196337 C12orf11 -1.87 10.29 0.00 ILMN_1754130 TRIM52 -2.88 9.14 0.00 ILMN_1746561 BCL2L2 -2.44 11.88 0.00 ILMN_2337263 PKIB -2.30 14.25 0.00 ILMN_1776582 PDK3 -2.54 10.79 0.00 ILMN_2358784 ASB3 -2.74 10.46 0.00 ILMN_2076940 C1orf149 -1.85 9.38 0.00 ILMN_2358783 ASB3 -1.99 9.38 0.00 ILMN_1743747 RUSC1 -2.07 14.07 0.00 ILMN_1804820 ZNF431 -1.72 8.75 0.00 ILMN_1851492 -2.25 12.06 0.00 ILMN_2311518 TROVE2 -1.84 12.58 0.00 ILMN_1772588 C6orf97 -2.60 9.60 0.00 ILMN_2272074 TROVE2 -1.81 9.24 0.00 ILMN_1726306 HMBS -2.54 12.19 0.00 ILMN_1737084 TXLNA -1.52 10.45 0.00 ILMN_1848552 -2.43 11.53 0.00 ILMN_1768004 PDCD4 -1.65 8.64 0.00 ILMN_1743582 NUDT22 -2.63 11.11 0.00 ILMN_1771957 MAN1B1 -1.66 10.54 0.00 ILMN_2214197 TP53INP1 -2.73 9.34 0.00 ILMN_1795285 PHF15 -1.44 10.47 0.00 ILMN_1760441 MRPS5 -2.53 10.81 0.00 ILMN_1778173 AK3 -2.23 12.52 0.00 ILMN_1715392 PRPF3 -2.37 12.30 0.00 ILMN_1732089 MRI1 -2.90 11.18 0.00 ILMN_1695961 CLK3 -2.25 8.74 0.00 ILMN_2374383 TSPAN17 -2.40 10.95 0.00 ILMN_1772876 ZNF395 -1.76 11.00 0.00 ILMN_1782788 CSDA -1.54 14.71 0.00 ILMN_1781791 PRRG1 -1.96 9.10 0.00 ILMN_1687884 ZNF2 -1.76 8.99 0.00 ILMN_1761531 SGPL1 -2.74 9.99 0.00 ILMN_1682775 EDN1 -3.11 11.53 0.00 ILMN_1783023 C5orf51 -1.79 11.93 0.00 ILMN_2408430 LARGE -2.51 9.38 0.00 ILMN_1658834 ZC3H18 -2.54 9.49 0.00 ILMN_1744713 PARK7 -2.80 13.91 0.00 ILMN_1727970 ONECUT1 -1.99 8.80 0.00 ILMN_1768743 FIP1L1 -1.69 9.99 0.00 ILMN_1790973 CDS2 -3.62 9.89 0.00 ILMN_1745112 FAM102A -1.82 12.55 0.00 ILMN_1658351 FIS1 -2.47 12.75 0.00 ILMN_2401779 FAM102A -1.63 12.61 0.00 ILMN_2194627 GMCL1 -2.13 10.04 0.00 ILMN_1769637 RNMT -2.40 10.17 0.00 ILMN_1847308 -2.89 11.18 0.00 ILMN_1751234 C1GALT1C -1.95 9.74 0.00 ILMN_2390162 PHF11 -2.41 11.22 0.00 1 ILMN_1789599 NBL1 -2.12 12.48 0.00 ILMN_1804327 NAP1L4 -1.87 12.22 0.00 ILMN_2153495 WNT7B -2.41 10.17 0.00 ILMN_1688629 ZNF274 -2.04 10.54 0.00 ILMN_1813635 KIAA1429 -2.71 9.95 0.00 ILMN_1726842 TYW3 -2.05 9.48 0.00 ILMN_2158242 SHOC2 -2.27 10.88 0.00 ILMN_1720373 SLC7A5 -1.78 14.58 0.00 ILMN_1782412 IRX2 -1.71 12.18 0.00 ILMN_1773751 HRAS -1.65 11.87 0.00 ILMN_1706376 OSBP -2.48 11.82 0.00 ILMN_2367782 STARD7 -3.09 10.97 0.00 ILMN_1806828 MRI1 -2.80 9.44 0.00 ILMN_1815519 EPN2 -1.46 9.45 0.00 ILMN_1726064 PAK1IP1 -2.42 11.87 0.00 ILMN_1712517 ZNF696 -1.97 9.96 0.00 ILMN_1782069 TRAK1 -2.70 10.60 0.00 ILMN_1744665 EP300 -1.93 8.79 0.00 ILMN_1701131 C2orf49 -3.60 11.71 0.00 ILMN_1678494 ZNF438 -1.55 8.66 0.00 ILMN_2383455 SUOX -3.87 9.87 0.00 ILMN_1737715 OSR2 -1.76 11.09 0.00 ILMN_2410965 MRI1 -2.67 10.54 0.00 ILMN_3251207 TMEM19 -1.93 9.04 0.00 ILMN_1715069 TANK -2.86 10.94 0.00 ILMN_1719696 PLD1 -1.36 8.14 0.00 ILMN_2396272 PDCD4 -2.42 12.92 0.00 ILMN_1690320 TTC23 -1.85 9.31 0.00 ILMN_1723632 PIGC -2.26 12.36 0.00 ILMN_1693014 CEBPB -1.54 14.63 0.00 ILMN_1812580 YDJC -3.38 11.62 0.00 ILMN_1792076 TRERF1 -1.90 9.07 0.00 ILMN_1727996 BAG4 -1.68 9.47 0.00 ILMN_3235593 ZNF841 -1.91 9.54 0.00 ILMN_3199647 LOC645251 -1.92 8.98 0.00 ILMN_1794063 ANKRD27 -2.13 8.85 0.00 ILMN_1721170 GREB1 -1.85 11.40 0.00 ILMN_1784860 RFC3 -1.81 10.85 0.00 ILMN_1657550 MVD -3.26 10.25 0.00 ILMN_1706706 WDR68 -2.49 11.12 0.00 ILMN_2128770 CDR2L -1.93 12.12 0.00 ILMN_2122953 CISD1 -3.17 12.97 0.00 ILMN_1703852 EFNB2 -1.72 10.71 0.00 ILMN_3246910 LOC100190 -1.37 8.69 0.00 ILMN_1862521 -1.70 8.69 0.00 986 ILMN_1699989 BNIPL -1.62 8.97 0.00 ILMN_1670532 GMCL1 -1.94 10.05 0.00 ILMN_2392189 CTDSPL -1.63 9.20 0.00 ILMN_2284591 OPA3 -1.90 10.08 0.00 ILMN_3245066 DENND4B -3.97 12.00 0.00 ILMN_1662316 VPS33A -2.16 9.37 0.00 ILMN_1743204 DUSP8 -2.87 10.15 0.00 ILMN_2342240 MGAT2 -3.00 11.75 0.00 ILMN_1688246 LOC642852 -2.20 9.33 0.00 ILMN_2331636 ACACA -1.85 10.14 0.00 ILMN_3187328 IP6K1 -1.83 10.15 0.00 ILMN_1776723 PHF11 -2.48 11.45 0.00 ILMN_1714622 TNRC6A -2.09 10.30 0.00 ILMN_1770244 CBX1 -2.21 8.76 0.00 ILMN_1708619 SEH1L -1.67 10.44 0.00 ILMN_3262345 HEATR7A -1.47 8.96 0.00

209 ILMN_1765829 NUFIP2 -1.85 10.17 0.00 ILMN_1673966 POLR3F -2.25 10.66 0.00 ILMN_1827736 -2.00 11.28 0.00 ILMN_1806804 USP14 -1.30 8.99 0.00 ILMN_2293758 AKAP11 -1.39 10.06 0.00 ILMN_1761463 EFHD2 -3.13 12.80 0.00 ILMN_1677292 C5orf30 -2.04 10.94 0.00 ILMN_1773063 OSBPL1A -1.52 9.10 0.00 ILMN_1691097 HSP90AA1 -2.98 13.28 0.00 ILMN_1809894 TMEM117 -1.51 8.39 0.00 ILMN_1714854 MEAF6 -1.33 8.39 0.00 ILMN_3247821 TMEM206 -2.18 9.11 0.00 ILMN_1680782 PATL1 -1.87 11.24 0.00 ILMN_1709132 ELP2 -1.76 10.53 0.00 ILMN_2352293 PRDM10 -1.73 8.95 0.00 ILMN_1714384 PCCA -1.83 8.67 0.00 ILMN_1803045 TUBGCP5 -1.77 8.31 0.00 ILMN_2217935 RFC1 -1.70 10.04 0.00 ILMN_1741780 DUSP28 -2.24 10.07 0.00 ILMN_2230683 CDCA7L -1.93 9.03 0.00 ILMN_1718718 MKKS -2.08 12.04 0.00 ILMN_1689868 TMEM80 -2.06 8.79 0.00 ILMN_1722781 EGR3 -1.27 8.12 0.00 ILMN_1672504 PDXK -2.41 12.71 0.00 ILMN_1831566 -1.55 8.55 0.00 ILMN_2075440 PDIA3P -1.27 8.84 0.00 ILMN_1687107 RFWD3 -1.56 12.15 0.00 ILMN_1724410 USP46 -2.56 9.34 0.00 ILMN_1789233 VPS37C -1.41 13.47 0.00 ILMN_3238751 PMS2L4 -1.13 11.77 0.00 ILMN_1832208 -2.93 11.71 0.00 ILMN_2102693 NUFIP2 -1.64 10.00 0.00 ILMN_1837935 TNPO1 -2.39 9.05 0.00 ILMN_1743499 POLDIP2 -2.43 10.18 0.00 ILMN_2405078 OSBPL8 -2.78 10.24 0.00 ILMN_1706434 LOC440359 -1.86 13.01 0.00 ILMN_1777096 TDG -1.67 13.39 0.00 ILMN_1889752 -1.98 9.75 0.00 ILMN_1793040 ADAMTSL5 -1.41 8.33 0.00 ILMN_1787256 HCN3 -1.18 8.08 0.00 ILMN_2132809 ARHGEF10 -1.85 9.88 0.00 ILMN_1663664 MRPS10 -1.30 12.05 0.00 ILMN_2405009 NBL1 -2.23 11.75 0.00 ILMN_1718961 BNIP3L -1.58 9.37 0.00 ILMN_1676062 DIP2C -2.06 9.52 0.00 ILMN_1797341 ARID1A -2.08 9.83 0.00 ILMN_3238058 LOC151162 -1.48 12.29 0.00 ILMN_1856861 -1.32 9.42 0.00 ILMN_1735058 CHST12 -1.32 8.70 0.00 ILMN_1815361 XIAP -1.32 8.13 0.00 ILMN_1678535 ESR1 -2.01 12.60 0.00 ILMN_1733042 BCAS1 -1.68 9.83 0.00 ILMN_2386008 MPZL1 -2.59 10.33 0.00 ILMN_1756049 NT5DC3 -1.30 9.92 0.00 ILMN_2186626 ZNF485 -2.22 8.86 0.00 ILMN_2397571 PIGC -2.07 9.84 0.00 ILMN_1682264 DCAF7 -2.08 12.52 0.00 ILMN_1710738 RC3H2 -2.04 11.46 0.00 ILMN_1714170 SPSB1 -1.93 10.20 0.00 ILMN_1689817 LCOR -1.25 10.00 0.00 ILMN_1758105 ZNF791 -1.71 10.36 0.00 ILMN_3238006 LOC730323 -1.09 10.78 0.00 ILMN_2390416 BAT3 -1.59 10.23 0.00 ILMN_1708482 TMEM80 -1.82 8.96 0.00 ILMN_2073543 C15orf63 -1.44 12.26 0.00 ILMN_1685602 TMEM41A -1.51 9.68 0.00 ILMN_1676946 AP3M2 -1.53 9.95 0.00 ILMN_1784540 KBTBD2 -1.39 11.93 0.00 ILMN_1723971 SLC29A1 -2.54 9.76 0.00 ILMN_1783771 UBE2Z -2.07 10.54 0.00 ILMN_1655068 TOM1L2 -1.72 8.52 0.00 ILMN_1710427 VPS37B -1.63 10.47 0.00 ILMN_1837428 -1.29 11.69 0.00 ILMN_1807031 C14orf28 -1.74 9.38 0.00 ILMN_1674380 TRPC1 -2.02 8.45 0.00 ILMN_2060145 GRHL2 -2.19 10.87 0.00 ILMN_1808356 FAM3A -1.56 13.13 0.00 ILMN_1690442 C18orf45 -1.68 11.51 0.00 ILMN_1655206 ZBTB34 -1.91 9.80 0.00 ILMN_1768662 UCK2 -2.55 11.53 0.00 ILMN_1668535 JOSD1 -1.34 10.61 0.00 ILMN_1694147 PUS3 -2.33 10.25 0.00 ILMN_1726388 ACBD7 -1.44 9.10 0.00 ILMN_2252136 YWHAE -2.56 11.68 0.00 ILMN_1654392 KHNYN -2.84 9.93 0.00 ILMN_2046470 DAAM1 -1.36 9.59 0.00 ILMN_1669201 ABCF2 -1.75 9.19 0.00 ILMN_2096191 AASDHPPT -1.14 9.09 0.00 ILMN_1670439 FYTTD1 -1.35 10.25 0.00 ILMN_1771824 DNAJC30 -1.04 8.46 0.00 ILMN_1673024 RBM15B -1.21 8.89 0.00 ILMN_2126706 LMNB1 -1.44 9.47 0.00 ILMN_1746359 RERG -1.51 11.86 0.00 ILMN_1720998 CA12 -1.55 11.38 0.00 ILMN_1666453 STK3 -1.95 11.51 0.00 ILMN_2139816 GPSM2 -1.16 11.44 0.00 ILMN_1660111 UCHL3 -3.72 11.92 0.00 ILMN_3224204 PSMG4 -1.35 11.50 0.00 ILMN_3243961 ZNF252 -2.00 12.07 0.00 ILMN_1743316 FAM109A -1.17 8.70 0.00 ILMN_1786852 ZCCHC3 -1.46 9.26 0.00 ILMN_1810436 DNAJC27 -1.94 10.07 0.00 ILMN_2122952 CISD1 -3.10 12.73 0.00 ILMN_1801443 TSKU -2.24 12.57 0.00 ILMN_1709549 PLEKHM1 -2.19 8.67 0.00 ILMN_1706118 HN1L -2.24 9.92 0.00 ILMN_1662161 TBC1D13 -1.26 9.74 0.00 ILMN_1794677 TMC6 -1.44 9.97 0.00 ILMN_1742578 MKLN1 -1.44 10.89 0.00 ILMN_1699703 ARCN1 -1.49 12.69 0.00 ILMN_1659937 ZBTB24 -1.57 10.61 0.00 ILMN_1704079 RBM38 -1.64 10.53 0.00 ILMN_1809400 FAM49B -2.01 8.53 0.00 ILMN_1762678 NMT1 -1.15 9.19 0.00 ILMN_1789839 GTF3C1 -1.52 9.16 0.00 ILMN_2106227 KIAA2026 -1.33 8.62 0.00 ILMN_1666761 PPP2R5E -1.72 13.75 0.00 ILMN_1709630 CCDC107 -1.06 8.15 0.00 ILMN_2097185 PUS3 -2.20 9.36 0.00 ILMN_1784630 KBTBD11 -1.47 8.20 0.00 ILMN_1787808 CEP63 -1.84 10.61 0.00 ILMN_1748432 ZNF525 -2.59 10.38 0.00 ILMN_3237850 C5orf51 -1.73 11.57 0.00 ILMN_1777318 C9orf64 -1.89 10.21 0.00 ILMN_1763666 ALDH3B2 -1.71 11.62 0.00 ILMN_3298694 TYW1B -1.34 10.00 0.00 ILMN_1760280 NXT1 -2.10 12.93 0.00 ILMN_1727809 STK35 -1.50 9.61 0.00 ILMN_3244019 LOC647886 -1.62 8.49 0.00 ILMN_1702501 RPS6KA2 -1.15 8.31 0.00

210 ILMN_1703946 ADORA2B -2.20 9.01 0.00 ILMN_1673892 GK5 -1.09 9.14 0.00 ILMN_1684591 ZNF434 -1.79 10.20 0.00 ILMN_3226769 LOC730074 -2.30 10.30 0.00 ILMN_1803824 ZDHHC9 -1.15 12.43 0.00 ILMN_1677534 SCAP -1.16 10.41 0.00 ILMN_1746314 EVI5 -1.50 9.52 0.00 ILMN_2338963 SLC29A1 -2.25 8.87 0.00 ILMN_1714108 TP53INP1 -3.14 10.77 0.00 ILMN_2316918 PANK1 -1.90 10.44 0.00 ILMN_2224143 MCM3 -1.73 12.60 0.00 ILMN_1786606 LOC729764 -1.53 8.25 0.00 ILMN_2325112 C22orf40 -1.43 8.08 0.00 ILMN_1660847 PFKFB3 -1.45 8.56 0.00 ILMN_1743456 ZCCHC14 -1.15 10.05 0.00 ILMN_1656165 USP9X -1.70 10.50 0.00 ILMN_1734231 DDOST -1.81 12.13 0.00 ILMN_1667476 LTBR -1.30 12.90 0.00 ILMN_1713846 PPM1H -1.03 10.94 0.00 ILMN_2083334 PMS2L5 -1.47 9.23 0.00 ILMN_1841334 -1.49 10.79 0.00 ILMN_1715969 SLC25A37 -1.01 9.66 0.00 ILMN_1726520 TDP1 -1.47 8.87 0.00 ILMN_2159152 TP53TG3 -1.11 7.99 0.00 ILMN_2338997 PTP4A2 -1.15 14.04 0.00 ILMN_1664646 NSUN6 -1.05 8.27 0.00 ILMN_2361570 SNX14 -1.40 9.37 0.00 ILMN_1669599 DENND4C -1.43 10.55 0.00 ILMN_1788738 ZNRF3 -1.25 8.10 0.00 ILMN_1697153 ZDHHC17 -1.13 8.96 0.00 ILMN_1724658 BNIP3 -1.34 13.43 0.00 ILMN_1797594 NFAT5 -1.48 8.78 0.00 ILMN_1792910 MNT -1.21 8.90 0.00 ILMN_2115669 SEMA4C -1.04 8.90 0.00 ILMN_2144573 CTBS -1.65 9.09 0.00 ILMN_1765044 CUTC -2.79 11.02 0.00 ILMN_1664921 PPP6C -2.17 14.02 0.00 ILMN_2118472 C10orf58 -1.17 14.59 0.00 ILMN_2112493 DAP -1.73 12.19 0.00 ILMN_2043918 DLEU1 -2.50 10.33 0.00 ILMN_1808789 MYO5C -1.25 12.24 0.00 ILMN_2077886 C1orf109 -2.26 9.47 0.00 ILMN_1704383 TRIM37 -2.91 12.53 0.00 ILMN_1691117 DNTTIP1 -2.78 11.20 0.00 ILMN_3240685 INO80D -1.47 10.15 0.00 ILMN_1735275 WDSUB1 -1.77 9.91 0.00 ILMN_1794333 POU2F1 -1.24 9.05 0.00 ILMN_3230435 LOC729086 -1.69 9.11 0.00 ILMN_1813489 RAF1 -1.39 12.10 0.00 ILMN_2096405 WDR37 -1.46 9.70 0.00 ILMN_2221673 ASNSD1 -2.08 12.79 0.00 ILMN_1796130 LOC221710 -1.66 10.44 0.00 ILMN_1772522 ZFP161 -1.69 10.13 0.00 ILMN_2329914 SPRY1 -1.99 8.68 0.00 ILMN_1791067 TESK1 -1.47 11.26 0.00 ILMN_1765636 FLJ22184 -1.32 8.79 0.00 ILMN_1771333 CD47 -2.02 9.96 0.00 ILMN_1740523 KTN1 -2.25 8.85 0.00 ILMN_1798790 IL17RC -1.55 8.54 0.00 ILMN_2399174 TRAK1 -1.54 8.89 0.00 ILMN_2181363 UBE3C -1.18 13.20 0.00 ILMN_1786470 C1orf74 -1.45 9.88 0.00 ILMN_1702279 KIF3B -1.99 9.89 0.00 ILMN_1802819 DEPDC1 -1.32 9.63 0.00 ILMN_1702384 ZNF706 -1.06 9.78 0.00 ILMN_1741942 STX16 -1.32 12.50 0.00 ILMN_1772123 ACACA -1.57 9.76 0.00 ILMN_1725969 RPP14 -1.04 8.10 0.00 ILMN_1794967 EIF4ENIF1 -1.16 9.06 0.00 ILMN_1838320 ONECUT2 -1.36 9.57 0.00 ILMN_1789436 C1orf218 -1.32 11.08 0.00 ILMN_1726786 TNRC6B -1.04 8.92 0.00 ILMN_1794294 DSTN -1.45 9.94 0.00 ILMN_1741300 ZNF407 -1.15 8.52 0.00 ILMN_1807515 CSTF2T -1.00 7.91 0.00 ILMN_2063586 CLIC4 -1.13 10.37 0.00 ILMN_2339202 KTN1 -2.43 11.05 0.00 ILMN_1774083 TRIAP1 -1.54 12.15 0.00 ILMN_2392286 IP6K1 -1.64 9.74 0.00 ILMN_1734476 KIF2A -2.39 10.20 0.00 ILMN_1763208 ZNF10 -1.08 7.91 0.00 ILMN_1750967 ZKSCAN3 -1.09 8.28 0.00 ILMN_1764794 PSMB2 -3.34 13.03 0.00 ILMN_2054510 BCAS3 -1.31 11.62 0.00 ILMN_2094776 CCNL1 -2.17 9.97 0.00 ILMN_2210482 MRPS34 -2.43 10.39 0.00 ILMN_1733519 HMGB3 -1.32 10.15 0.00 ILMN_1804929 OXTR -1.54 9.13 0.00 ILMN_2359789 RAC1 -1.38 14.30 0.00 ILMN_1654287 ADCY9 -1.02 9.22 0.00 ILMN_1880983 -2.17 11.05 0.00 ILMN_1674506 MED23 -1.04 8.18 0.00 ILMN_1683494 TMEM154 -1.53 9.35 0.00 ILMN_3236428 TMEM170B -1.18 9.00 0.00 ILMN_1751258 NDUFA4 -2.37 13.22 0.00 ILMN_1810085 ABCF3 -1.73 8.77 0.00 ILMN_1746012 MBD6 -1.25 11.49 0.00 ILMN_1670901 COX10 -1.24 11.19 0.00 ILMN_1694686 KIAA0194 -2.53 10.36 0.00 ILMN_2325347 B3GALNT1 -1.60 12.72 0.00 ILMN_1714159 LUZP1 -1.33 9.40 0.00 ILMN_1711109 TOM1L2 -1.11 8.12 0.00 ILMN_1802780 M160 -1.28 9.53 0.00 ILMN_2088124 TMEM154 -1.45 9.56 0.00 ILMN_1704342 UBE3C -1.27 11.02 0.00 ILMN_2246956 BCL2 -1.14 8.36 0.00 ILMN_2342437 KLHL5 -2.71 9.94 0.00 ILMN_1797903 ZNF544 -1.01 12.10 0.00 ILMN_1707326 TASP1 -1.90 8.57 0.00 ILMN_1808202 C19orf22 -1.34 12.06 0.00 ILMN_1782057 ATP8B2 -1.04 9.11 0.00 ILMN_1673795 HSD17B4 -2.65 11.00 0.00 ILMN_2346831 MGAT2 -1.80 8.70 0.00 ILMN_1758719 NEDD9 -1.19 8.10 0.00 ILMN_2081398 KIF3B -2.21 9.82 0.00 ILMN_1798210 E2F7 -1.29 9.65 0.00 ILMN_2212763 ICAM3 -1.37 11.64 0.00 ILMN_1696827 PARS2 -1.12 8.25 0.00 ILMN_1738579 VHL -1.39 12.83 0.00 ILMN_2414014 RBM10 -1.79 12.42 0.00 ILMN_1661599 DDIT4 -2.97 12.66 0.00 ILMN_1811373 FAM20B -1.33 12.71 0.00 ILMN_1663042 SDC4 -1.39 11.31 0.00 ILMN_2173004 RAB8B -2.14 9.30 0.00 ILMN_1701512 KIAA0391 -1.49 11.73 0.00 ILMN_1813625 TRIM25 -1.72 9.73 0.00 ILMN_3241510 FBXO45 -1.29 9.36 0.00 ILMN_2387471 FLJ22184 -1.22 10.60 0.00 ILMN_1694759 C19orf42 -1.41 11.29 0.00 ILMN_3237156 LOC100009 -1.11 8.15 0.00

211 676 ILMN_2064606 TBC1D2B -1.23 9.05 0.00 ILMN_1760320 GNB1 -1.06 13.41 0.00 ILMN_3228822 TMEM194A -1.19 11.21 0.00 ILMN_2255579 RAB37 -1.38 8.63 0.00 ILMN_1816244 -1.02 8.58 0.00 ILMN_1686626 BAT1 -1.93 12.74 0.00 ILMN_1781745 C9orf152 -1.79 11.00 0.00 ILMN_2122374 FAM49B -1.48 8.26 0.00 ILMN_1681641 DLEU1 -2.08 8.61 0.00 ILMN_1803939 YIPF6 -1.43 11.42 0.00 ILMN_1772658 KIAA0947 -1.45 8.27 0.00 ILMN_1765520 MTIF2 -1.82 11.19 0.00 ILMN_2243687 LAMP2 -1.24 10.18 0.00 ILMN_1768470 EIF4G1 -1.60 11.92 0.00 ILMN_2387919 PRKAG2 -1.22 9.43 0.00 ILMN_1699112 COPB1 -1.73 11.42 0.00 ILMN_1692219 RAB11FIP1 -1.12 10.58 0.00 ILMN_2373515 HSP90AA1 -1.01 14.81 0.00 ILMN_2136455 C3orf64 -1.20 8.01 0.00 ILMN_1688755 AAK1 -1.88 9.26 0.00 ILMN_1767475 CERK -1.63 9.13 0.00 ILMN_2091792 ENTPD6 -1.51 10.71 0.00 ILMN_2381138 SEH1L -1.80 10.86 0.00 ILMN_1747183 GXYLT1 -1.83 10.13 0.00 ILMN_1666545 GCNT1 -1.61 9.96 0.00 ILMN_1695719 EIF2C2 -1.09 13.39 0.00 ILMN_2104830 ACP2 -1.10 9.19 0.00 ILMN_2140974 TPM4 -1.72 10.58 0.00 ILMN_1678504 RHOT1 -1.06 11.46 0.00 ILMN_1706687 KLHL5 -2.90 11.92 0.00 ILMN_1749809 ZNF813 -1.02 8.42 0.00 ILMN_2053415 LDLR -1.18 12.04 0.00 ILMN_1710092 ZBTB46 -1.86 8.89 0.00 ILMN_1687315 RXRA -1.01 12.47 0.00 ILMN_1741005 RG9MTD2 -1.20 8.23 0.00 ILMN_1721349 MAGT1 -1.21 11.95 0.00 ILMN_1811489 OXSR1 -1.52 11.76 0.00 ILMN_1813191 LOC653080 -1.09 8.93 0.00 ILMN_2050790 C11orf52 -1.04 8.49 0.00 ILMN_1746368 SELT -1.15 9.90 0.00 ILMN_1778991 NFIB -1.08 12.05 0.00 ILMN_1754553 MED19 -1.53 10.16 0.00 ILMN_1781151 ARMC8 -1.77 9.36 0.00 ILMN_2105966 SLC35A4 -2.20 11.42 0.00 ILMN_2161832 VPS37A -1.12 10.39 0.00 ILMN_1692168 UBE2Z -1.66 11.30 0.00 ILMN_1693220 AKAP11 -1.12 9.44 0.00 ILMN_1714965 NFKB1 -1.31 11.79 0.00 ILMN_1771903 NUP37 -1.77 11.35 0.00 ILMN_1668822 BATF -1.72 11.11 0.00 ILMN_1657631 STAP2 -2.38 9.18 0.00 ILMN_1693410 BRI3BP -1.26 12.95 0.00 ILMN_1790136 C20orf20 -1.25 13.36 0.00 ILMN_1789492 ZDHHC8 -1.47 10.14 0.00 ILMN_1705468 PIK3CA -1.12 8.07 0.00 ILMN_2388975 CERK -1.25 8.46 0.00 ILMN_2408946 PDE8A -1.26 8.50 0.00 ILMN_2067709 TFB2M -1.96 9.42 0.00 ILMN_1729650 PEX7 -1.04 9.45 0.00 ILMN_1787567 TSC22D1 -2.52 11.87 0.00 ILMN_1724309 FAM35A -1.01 7.93 0.00 ILMN_1791106 HEATR6 -1.83 11.70 0.00 ILMN_1665455 DCUN1D3 -1.31 10.56 0.00 ILMN_2194649 TADA1L -1.01 9.40 0.00 ILMN_2404135 RIOK3 -1.14 11.21 0.00 ILMN_1683305 COMMD2 -2.15 11.22 0.00 ILMN_1797693 BRI3BP -1.10 11.04 0.00 ILMN_1712320 DDX50 -1.03 10.13 0.00 ILMN_1741613 SERINC1 -1.29 9.63 0.00 ILMN_1746465 FJX1 -1.59 9.40 0.00 ILMN_1705814 KRT80 -1.28 14.64 0.00 ILMN_1748770 CKAP5 -1.29 11.73 0.00 ILMN_1719392 FH -2.20 10.35 0.00 ILMN_1774074 RXRB -1.48 9.27 0.00 ILMN_1773228 DLST -1.41 10.68 0.00 ILMN_3250585 KIAA0194 -2.67 10.43 0.00 ILMN_1711327 TRIM37 -2.37 10.64 0.00 ILMN_1802973 ANAPC4 -1.85 10.41 0.00 ILMN_1689908 ANKRD13A -1.47 9.79 0.00 ILMN_1722239 TIMM8A -1.07 9.73 0.00 ILMN_3228595 LOC729768 -1.05 12.88 0.00 ILMN_1802799 AKIRIN1 -1.13 11.34 0.00 ILMN_2044027 C20orf191 -1.48 9.94 0.00 ILMN_2376625 VHL -1.39 9.37 0.00 ILMN_1717799 PRKCE -1.07 8.04 0.00 ILMN_2347349 CCNB1IP1 -1.68 11.61 0.00 ILMN_1797933 MRPL17 -1.95 12.56 0.00 ILMN_3240541 EML2 -1.85 8.57 0.00 ILMN_1765082 RBM10 -1.86 9.55 0.00 ILMN_3305055 TP63 -1.96 9.63 0.00 ILMN_3248026 RBM27 -1.30 9.50 0.00 ILMN_1797534 RIOK1 -1.75 9.84 0.00 ILMN_1739601 MSI1 -1.35 8.71 0.00 ILMN_1708611 RDX -2.10 12.02 0.00 ILMN_2123871 TMEM18 -1.27 8.88 0.00 ILMN_2136147 BCAS1 -1.42 9.53 0.00 ILMN_1779517 RASEF -1.06 10.01 0.00 ILMN_1752394 CCNB1IP1 -1.59 10.64 0.00 ILMN_2352574 ZNF274 -1.44 8.99 0.00 ILMN_1806607 SFN -1.49 11.19 0.00 ILMN_2155322 ZNF652 -1.05 12.52 0.00 ILMN_1759495 XPO5 -1.44 10.29 0.00 ILMN_1797822 SEL1L3 -1.42 10.05 0.00 ILMN_2101832 LAPTM4B -1.77 13.29 0.00 ILMN_1748291 C1orf55 -1.04 9.31 0.00 ILMN_1752226 P2RY11 -1.17 8.56 0.00 ILMN_1721713 EXOSC9 -1.08 10.05 0.00 ILMN_1699254 PLEKHH1 -1.55 8.99 0.00 ILMN_1676322 C1orf172 -1.31 9.55 0.00 ILMN_1711878 ENOPH1 -1.58 12.65 0.00 ILMN_2116127 NPEPPS -1.82 9.75 0.00 ILMN_2366703 SGK3 -2.11 10.85 0.00 ILMN_2115696 USP42 -1.68 10.16 0.00 ILMN_1750008 SUPV3L1 -2.18 10.11 0.00 ILMN_1798360 CXCR7 -1.28 12.03 0.00 ILMN_1801101 ZBTB48 -1.19 8.36 0.00 ILMN_2412807 DCTN1 -1.80 10.44 0.00 ILMN_1718988 DAZAP2 -1.35 13.39 0.00 ILMN_1664012 CANT1 -1.30 10.27 0.00 ILMN_1664922 FLNB -1.73 10.27 0.00 ILMN_1706413 C1orf66 -1.10 8.86 0.00 ILMN_1770719 KIAA0664 -1.96 9.75 0.00 ILMN_2067708 TFB2M -2.15 10.12 0.00 ILMN_2061950 RABGAP1 -1.38 12.02 0.00 ILMN_1728626 WDR5 -1.04 10.22 0.00 ILMN_1810832 ZNF343 -1.44 8.64 0.00 ILMN_1764813 B3GALTL -1.22 8.51 0.00 ILMN_1653896 ATRIP -1.11 8.77 0.00 ILMN_1655482 TRIM27 -1.46 9.17 0.00

212 ILMN_1694111 PNKP -1.83 10.41 0.00 ILMN_1719883 CYP4F11 -1.02 8.18 0.01 ILMN_3245057 ASAP1 -1.68 12.09 0.00 ILMN_1667711 HRASLS3 -1.20 10.61 0.01 ILMN_2128668 FOXJ3 -1.04 10.80 0.00 ILMN_1724207 IVD -1.05 8.33 0.01 ILMN_2038775 TUBB2A -1.95 9.58 0.00 ILMN_2172174 NP -1.73 10.95 0.01 ILMN_1681829 ZNF606 -1.22 8.44 0.00 ILMN_2377980 PPP1CA -1.67 10.59 0.01 ILMN_1652082 ELF4 -1.16 11.63 0.00 ILMN_3244286 TANC2 -1.07 10.19 0.01 ILMN_1676600 SEC24C -1.10 11.11 0.00 ILMN_1772700 TMEM18 -1.03 8.48 0.01 ILMN_2096012 UHMK1 -1.15 10.32 0.00 ILMN_1677432 SRGAP1 -1.15 10.86 0.01 ILMN_1676075 KLHL12 -1.23 9.30 0.00 ILMN_1742379 IFT122 -1.15 9.15 0.01 ILMN_1846771 -1.00 7.98 0.00 ILMN_3245625 RFX7 -1.07 10.79 0.01 ILMN_1681118 CAPRIN2 -1.06 8.57 0.00 ILMN_1793017 DGKQ -1.17 9.18 0.01 ILMN_2384237 STAP2 -2.23 8.84 0.00 ILMN_1675669 IBTK -1.11 10.30 0.01 ILMN_2366445 KRT80 -1.34 11.66 0.00 ILMN_1728224 OGFR -2.59 10.34 0.01 ILMN_3240721 LOC645233 -1.13 9.51 0.00 ILMN_1746720 TTC39C -1.05 8.80 0.01 ILMN_1794470 ANKFY1 -1.18 9.61 0.00 ILMN_1774086 CBX3 -1.16 8.41 0.01 ILMN_1787251 DAAM1 -1.58 10.15 0.00 ILMN_1677843 RAB24 -1.22 11.42 0.01 ILMN_1797531 PRKAG2 -1.53 10.57 0.00 ILMN_1672042 DOLPP1 -1.19 9.97 0.01 ILMN_1849494 -1.07 10.09 0.00 ILMN_1785646 PMP22 -1.63 10.85 0.01 ILMN_1706246 CCT5 -1.16 9.73 0.00 ILMN_2110281 UFC1 -1.15 11.81 0.01 ILMN_1653180 TPM4 -1.79 10.68 0.00 ILMN_1772492 MCART1 -1.17 12.83 0.01 ILMN_1674024 IKZF5 -1.10 8.84 0.00 ILMN_2193706 HRK -1.53 8.88 0.01 ILMN_1784847 CREBZF -1.10 9.68 0.00 ILMN_1766408 CBFB -1.13 10.83 0.01 ILMN_1680673 NT5DC1 -1.06 9.31 0.00 ILMN_1713706 ZNF786 -1.01 9.54 0.01 ILMN_1799642 TRIM24 -1.60 9.49 0.00 ILMN_2246510 TSC1 -1.08 9.16 0.01 ILMN_3242540 CD163L1 -1.04 8.72 0.00 ILMN_1792072 FUT4 -1.32 8.66 0.01 ILMN_1753008 REXO1 -1.44 8.77 0.00 ILMN_1686367 HSPA8 -1.62 13.54 0.01 ILMN_2361695 BAG5 -1.29 12.57 0.00 ILMN_1774161 ARL15 -1.12 8.06 0.01 ILMN_2382942 CA12 -1.04 9.24 0.00 ILMN_1671885 MLF2 -1.45 11.94 0.01 ILMN_1771411 ALG6 -1.08 9.11 0.00 ILMN_1755235 XPO6 -1.20 11.51 0.01 ILMN_1729051 MSH6 -1.03 11.01 0.00 ILMN_1690122 CRKL -1.40 12.10 0.01 ILMN_1660533 RPN1 -1.05 13.21 0.00 ILMN_2370772 EIF4G1 -1.20 11.68 0.01 ILMN_2311826 USP6NL -1.11 9.26 0.00 ILMN_1758806 C21orf2 -1.12 8.81 0.01 ILMN_1744395 LOC284371 -1.01 8.54 0.00 ILMN_1732296 ID3 -1.99 11.15 0.01 ILMN_1658494 C13orf15 -1.46 8.91 0.00 ILMN_2384241 TGFBR2 -1.78 9.07 0.01 ILMN_1793563 DCTN1 -1.64 8.66 0.00 ILMN_1724040 ANKRD57 -1.10 11.10 0.01 ILMN_1706301 RET -1.08 9.98 0.00 ILMN_3200330 LOC399988 -1.59 12.47 0.01 ILMN_1690963 ASAP1 -1.71 11.39 0.00 ILMN_1707475 UBE2E2 -1.19 11.98 0.01 ILMN_2363880 ALDH3B2 -1.02 9.87 0.00 ILMN_1735553 MAP3K9 -1.11 8.30 0.01 ILMN_1790650 C16orf63 -1.05 12.48 0.00 ILMN_1687375 ATP2A2 -1.13 9.21 0.01 ILMN_1798804 SRPK1 -1.08 10.84 0.00 ILMN_1658950 LOC400890 -1.16 9.89 0.01 ILMN_1727574 ZNF827 -1.14 8.67 0.00 ILMN_1793290 WDR60 -1.06 8.25 0.01 ILMN_2383419 GMEB1 -1.01 8.71 0.00 ILMN_1739914 ZNF618 -1.01 8.10 0.01 ILMN_1806818 MCM3 -1.42 9.42 0.00 ILMN_1766010 YARS -1.93 12.15 0.01 ILMN_1777526 MED20 -1.14 11.15 0.00 ILMN_2415170 VPS8 -1.50 8.78 0.01 ILMN_1738523 MYD88 -2.07 9.86 0.00 ILMN_3248975 PPP4C -1.32 13.06 0.01 ILMN_2397880 CSTF3 -2.34 10.22 0.00 ILMN_1775579 ACAD9 -1.01 9.70 0.01 ILMN_2123567 SENP2 -1.05 11.44 0.00 ILMN_2062381 LCOR -1.08 9.29 0.01 ILMN_1726755 COPS4 -1.27 9.76 0.00 ILMN_1769787 SELO -1.53 8.51 0.01 ILMN_1704253 C6orf106 -1.07 10.27 0.00 ILMN_1654653 KLC1 -1.24 11.29 0.01 ILMN_1674366 LHX4 -1.08 8.79 0.00 ILMN_1745784 ZNF324 -1.30 8.97 0.01 ILMN_1696294 SNX6 -1.20 8.06 0.00 ILMN_1680196 LAPTM4B -1.64 13.11 0.01 ILMN_1705364 BAT3 -1.14 9.23 0.00 ILMN_1749521 SLC35E3 -1.11 10.44 0.01 ILMN_1741392 SLC25A20 -1.19 8.52 0.00 ILMN_1790909 NFE2L2 -1.35 10.46 0.01 ILMN_2119774 CYP2R1 -1.24 8.91 0.01 ILMN_3233135 FAM178A -1.08 9.67 0.01 ILMN_1694174 TRIM68 -1.32 8.60 0.01 ILMN_1809590 GINS2 -1.30 13.49 0.01 ILMN_1733931 PDCD6 -1.46 12.79 0.01 ILMN_1795639 MGMT -1.95 11.17 0.01 ILMN_1797074 EMG1 -1.20 11.15 0.01 ILMN_1728514 BAG5 -1.07 10.05 0.01 ILMN_1700109 PTOV1 -2.54 12.23 0.01 ILMN_1658835 CAV2 -1.86 8.94 0.01 ILMN_1711270 SFRS14 -1.40 11.09 0.01 ILMN_1678268 VPS8 -1.15 9.64 0.01 ILMN_1770084 TACC1 -1.43 9.86 0.01 ILMN_1772821 KIAA1671 -1.17 9.83 0.01 ILMN_2371458 CXCR7 -1.45 10.62 0.01 ILMN_1762835 HELZ -1.00 10.63 0.01 ILMN_2278433 LOC285074 -1.23 9.51 0.01 ILMN_1658071 ATP1B1 -1.27 11.79 0.01 ILMN_2175112 KCNS3 -1.69 9.83 0.01 ILMN_1709377 ZNF28 -1.01 8.50 0.01 ILMN_1783497 PANK1 -1.25 8.53 0.01 ILMN_3294365 LOC646993 -1.11 10.22 0.01 ILMN_1739450 NFE2L1 -1.46 10.51 0.01 ILMN_1661432 NUP43 -1.07 10.57 0.01

213 ILMN_1784602 CDKN1A -1.77 13.92 0.01 ILMN_2368068 TCF20 -1.12 9.54 0.02 ILMN_1761594 FAM35A -1.18 8.82 0.01 ILMN_1798886 NUDT21 -1.13 9.00 0.02 ILMN_1803279 TMED5 -1.80 11.21 0.01 ILMN_2322996 EYA2 -1.05 8.83 0.02 ILMN_1909886 -1.31 8.19 0.01 ILMN_1695827 PPP1CA -1.14 12.03 0.02 ILMN_1672884 MTFMT -1.30 9.45 0.01 ILMN_1718537 HPS6 -1.60 10.80 0.02 ILMN_1658160 FAM156A -1.12 8.58 0.01 ILMN_1655921 GTF2E1 -1.26 9.92 0.02 ILMN_1815610 SYT12 -1.19 8.81 0.01 ILMN_1798172 IPO4 -1.17 9.32 0.02 ILMN_1697864 CXorf38 -1.10 9.72 0.01 ILMN_1726901 KLC1 -1.06 10.13 0.02 ILMN_1762002 CSTF3 -2.19 8.99 0.01 ILMN_2366246 SEC23B -1.34 10.26 0.02 ILMN_1691436 BLVRA -1.96 10.45 0.01 ILMN_1781374 TUFT1 -1.33 11.32 0.02 ILMN_1796464 WDR37 -1.08 8.61 0.01 ILMN_1803464 PHTF1 -1.81 10.60 0.02 ILMN_1733615 MTF2 -1.06 8.82 0.01 ILMN_1743538 MLLT10 -1.07 8.89 0.02 ILMN_3251506 ZNF69 -1.15 8.55 0.01 ILMN_1801121 SENP2 -1.26 9.75 0.02 ILMN_1759436 NOSIP -1.19 11.00 0.01 ILMN_3187852 KIAA1310 -1.28 11.93 0.02 ILMN_1672940 ZNF562 -1.17 11.20 0.01 ILMN_3244110 FAM156B -1.01 9.01 0.02 ILMN_1760954 DENR -1.15 12.93 0.01 ILMN_1685415 HBP1 -1.01 9.43 0.02 ILMN_1678619 DAK -1.07 8.37 0.01 ILMN_3290261 LOC644877 -1.21 9.42 0.02 ILMN_3229033 LOC732360 -1.03 10.89 0.01 ILMN_1693136 VTI1B -1.09 11.51 0.02 ILMN_1746031 RIMS4 -1.36 8.71 0.01 ILMN_2131336 TMEM194 -1.07 9.57 0.02 ILMN_2407824 ATP1B1 -1.13 12.54 0.01 ILMN_1698231 RRM2B -1.12 8.28 0.02 ILMN_2130635 FOXRED2 -1.02 10.54 0.01 ILMN_1726245 TGFBR2 -1.50 9.14 0.02 ILMN_2413084 HSPA8 -1.43 13.28 0.01 ILMN_1657483 SEC23B -1.17 10.83 0.02 ILMN_1720114 GMNN -1.66 9.19 0.01 ILMN_1674859 LOC651576 -1.00 8.95 0.03 ILMN_1747020 SGK3 -1.04 9.40 0.01 ILMN_1744442 TTPAL -1.00 9.94 0.03 ILMN_1687495 SLC37A1 -1.37 10.71 0.02 ILMN_1745217 FLJ10081 -1.06 9.93 0.03 ILMN_1787345 FKBP11 -1.76 10.53 0.02 ILMN_1859946 -1.05 12.65 0.03 ILMN_1718023 APEH -1.78 10.98 0.02 ILMN_1720829 ZFP36 -1.11 11.93 0.03 ILMN_1742432 COBRA1 -1.21 11.22 0.02 ILMN_3281195 LOC440459 -1.06 8.21 0.03 ILMN_2379718 RAB24 -1.08 10.14 0.02 ILMN_1667893 TNS3 -1.40 8.48 0.03 ILMN_1808587 ZFHX3 -1.34 10.93 0.02 ILMN_1815874 NANS -1.57 10.70 0.03 ILMN_1791873 COG1 -1.43 8.61 0.02 ILMN_2321634 RAD17 -1.01 8.72 0.03 ILMN_1744963 ERO1L -1.25 10.49 0.02 ILMN_1763523 HARS -1.01 11.63 0.03 ILMN_1681591 PTPN1 -1.04 12.70 0.02 ILMN_1723480 BST2 -1.36 8.31 0.03 ILMN_2224031 CETN3 -1.01 10.48 0.02 ILMN_1750641 SRCAP -1.22 8.38 0.03 ILMN_1662438 SOD1 -1.14 14.12 0.02 ILMN_2166865 ENY2 -1.26 13.24 0.03 ILMN_1670130 ARID3A -1.20 10.38 0.02 ILMN_3306440 TMEM194A -1.09 10.05 0.03 ILMN_1798308 AHSA2 -1.01 8.45 0.02 ILMN_1697420 TINF2 -1.17 9.58 0.04 ILMN_2364131 TTPAL -1.27 10.53 0.02 ILMN_2138801 TP73L -1.15 8.41 0.04 ILMN_1690621 GPR98 -1.01 7.91 0.02 ILMN_1772487 SFRS14 -1.32 9.34 0.04 ILMN_1747281 EVI5L -1.47 9.61 0.02 ILMN_1713266 FAM46C -1.05 10.57 0.04 ILMN_1700915 BMI1 -1.07 13.01 0.02 ILMN_1689007 SFRS14 -1.17 8.67 0.05 ILMN_1800276 RCN1 -1.10 9.58 0.02 ILMN_1657395 HMGCR -1.07 11.25 0.05 ILMN_1721626 ARID5B -1.12 10.11 0.02 ILMN_1724753 NIN -1.01 8.98 0.02 ILMN_1726574 CACYBP -1.43 10.79 0.02

214 Supplementary Table 3: Functional Enrichment Analysis of genes predicted to be miR- 139-5p targets

Ingenuity Canonical -log10(p- Ratio Molecules Pathways value) Glucocorticoid Receptor 3.39E+00 7.94E-02 RAF1,PIK3CA,PIK3R1,RAC1,HRAS,GTF2E2,CEBPB (includes Signaling EG:1051),NFKB1,HSPA5,EP300,BCL2,TGFBR2,HSPA8,POU2F1,NFAT5,NCOR1P1,G TF2E1,CDKN1A,PRKAG2,HSP90AA1,ESR1,CREBZF PI3K/AKT Signaling 3.35E+00 1.01E-01 RAF1,TSC1,PIK3CA,YWHAE,PIK3R1,HRAS,NFKB1,BCL2,CDKN1A,HSP90AA1,GS K3B,PPP2R5E,SFN Cleavage and 3.09E+00 3.33E-01 PAPOLA,NUDT21,CPSF3,CSTF3 Polyadenylation of Pre- mRNA p53 Signaling 3.08E+00 1.16E-01 PIK3CA,TP53INP1,TP63,PPP1R13B,PIK3R1,CDKN1A,RRM2B,GSK3B,SFN,BCL2,EP 300 Prostate Cancer Signaling 3.08E+00 1.11E-01 RAF1,PIK3CA,PIK3R1,SUV39H1,CDKN1A,HSP90AA1,HRAS,GSK3B,NFKB1,BCL2 mTOR Signaling 3.02E+00 9.59E-02 TSC1,PIK3CA,DDIT4,PIK3R1,RAC1,HRAS,EIF4G1,PLD1,RHOT1,PRKAG2,PRKCE, PPP2R5E,RPS6KA2,FNBP1 N-Glycan Biosynthesis 3.01E+00 1.67E-01 DOLPP1,RPN2,DDOST,MAN1B1,RPN1,MGAT2,FUT4 Breast Cancer Regulation 2.97E+00 8.63E-02 ADCY9,RAF1,PIK3CA,ARHGEF12,PIK3R1,TUBB2A,RAC1,HRAS,GNB1,ARHGEF1 by Stathmin1 0,CDKN1A,PRKAG2,PRKCE,UHMK1,PPP2R5E,PPP1CA,PRKAR1A Melanocyte Development 2.87E+00 1.15E-01 ADCY9,RAF1,PIK3CA,PIK3R1,PRKAG2,HRAS,RPS6KA2,BCL2,PRKAR1A,EP300 and Pigmentation Signaling Chronic Myeloid 2.57E+00 9.80E-02 TGFBR2,RAF1,PIK3CA,RBL2 (includes Leukemia Signaling EG:100331892),CRKL,PIK3R1,SUV39H1,CDKN1A,HRAS,NFKB1 14-3-3-mediated 2.40E+00 9.57E-02 RAF1,TSC1,PIK3CA,YWHAE,PIK3R1,TUBB2A,YAP1 (includes Signaling EG:10413),PRKCE,HRAS,GSK3B,SFN Ovarian Cancer Signaling 2.35E+00 8.70E-02 RAF1,PIK3CA,EDN1,PIK3R1,WNT7B,SUV39H1,MSH6,PRKAG2,HRAS,GSK3B,BCL 2,PRKAR1A HGF Signaling 2.34E+00 9.80E-02 RAF1,MAP3K9,ELF4,PIK3CA,CRKL,PIK3R1,CDKN1A,RAC1,PRKCE,HRAS P2Y Purigenic Receptor 2.34E+00 9.02E-02 ADCY9,GNB1,RAF1,PPAN,PIK3CA,PIK3R1,PRKAG2,PRKCE,HRAS,NFKB1,PRKA Signaling Pathway R1A Renal Cell Carcinoma 2.33E+00 1.13E-01 RAF1,PIK3CA,PIK3R1,RAC1,HRAS,FH,VHL,EP300 Signaling Aryl Hydrocarbon 2.30E+00 8.51E-02 CCNA2,ALDH3B2,SP1,CDKN1A,HSP90AA1,NFIB,RXRB,RXRA,NFKB1,NFE2L2,ES Receptor Signaling R1,EP300 Protein Ubiquitination 2.27E+00 7.06E-02 USP24,UCHL3,USP14,MED20,DNAJC27,USP9X,BIRC6,HSPA5,UBE3A,XIAP,HSPA Pathway 8,ANAPC4,PSMB2,USP42,HSP90AA1,UBE2E2,USP46,DNAJC30,VHL Prolactin Signaling 2.18E+00 1.04E-01 RAF1,PIK3CA,SP1,PIK3R1,PRKCE,HRAS,CEBPB (includes EG:1051),EP300 Pancreatic 2.16E+00 8.77E-02 TGFBR2,RAF1,PIK3CA,PIK3R1,SUV39H1,CDKN1A,RAC1,NFKB1,PLD1,BCL2 Adenocarcinoma Signaling Insulin Receptor 2.12E+00 8.73E-02 RAF1,TSC1,PIK3CA,CRKL,PIK3R1,PTPN1,PRKAG2,HRAS,GSK3B,PPP1CA,PRKAR Signaling 1A Biosynthesis of Steroids 2.12E+00 1.90E-01 MVD,DHCR7,COX10 (includes EG:1352),HMGCR Estrogen-Dependent 2.11E+00 1.08E-01 PIK3CA,SP1,PIK3R1,HRAS,NFKB1,HSD17B4,ESR1 Breast Cancer Signaling Renin-Angiotensin 2.11E+00 8.93E-02 ADCY9,RAF1,PIK3CA,PIK3R1,RAC1,PRKAG2,PRKCE,HRAS,NFKB1,PRKAR1A Signaling PTEN Signaling 2.11E+00 8.33E-02 TGFBR2,RAF1,PIK3CA,PIK3R1,CDKN1A,RAC1,HRAS,GSK3B,NFKB1,BCL2 Role of CHK Proteins in 2.10E+00 1.43E-01 RAD17 (includes EG:19356),CDKN1A,RFC1,CDC25A,RFC3 Cell Cycle Checkpoint Control Xenobiotic Metabolism 2.08E+00 6.95E-02 RAF1,MAP3K9,PIK3CA,MGMT,PIK3R1,HRAS,CHST12,CES2,NFKB1,EP300,ALDH Signaling 3B2,CHST3,HSP90AA1,PRKCE,MAPK7,PPP2R5E,RXRA,NFE2L2 HMGB1 Signaling 2.04E+00 9.28E-02 PIK3CA,SP1,RHOT1,KAT6A,PIK3R1,RAC1,HRAS,NFKB1,FNBP1 Molecular Mechanisms of 2.02E+00 6.13E-02 ADCY9,RAF1,PIK3CA,ARHGEF12,SUV39H1,PIK3R1,RAC1,HRAS,NFKB1,XIAP,EP Cancer 300,BCL2,TGFBR2,ARHGEF10,RHOT1,CDKN1A,PRKAG2,PRKCE,GSK3B,FNBP1,P RKAR1A,CDC25A Ceramide Signaling 2.01E+00 9.76E-02 RAF1,PIK3CA,PIK3R1,HRAS,CERK,PPP2R5E,NFKB1,BCL2 RAR Activation 1.96E+00 7.60E-02 ADCY9,NSD1,PIK3CA,TRIM24,PIK3R1,RAC1,NFKB1,EP300,PRKAG2,PRKCE,RXR B,RXRA,PRKAR1A

215 Cardiac Î²-adrenergic 1.93E+00 8.03E-02 PDE8A,ADCY9,GNB1,PKIB,PRKAG2,PPP2R5E,PPP1CA,ATP2A2,AKAP7,AKAP11,P Signaling RKAR1A Non-Small Cell Lung 1.93E+00 9.59E-02 RAF1,PIK3CA,PIK3R1,SUV39H1,HRAS,RXRB,RXRA Cancer Signaling EIF2 Signaling 1.89E+00 8.79E-02 RAF1,PIK3CA,PIK3R1,HRAS,EIF2C2,GSK3B,EIF4G1,PPP1CA Phospholipase C 1.87E+00 6.58E-02 ADCY9,RAF1,ARHGEF12,RAC1,HRAS,NFKB1,PLD1,EP300,GNB1,ARHGEF10,NFA Signaling T5,RHOT1,MEF2D,PRKCE,PPP1CA,FNBP1 Macropinocytosis 1.83E+00 9.21E-02 ANKFY1,PIK3CA,PIK3R1,USP6NL,RAC1,PRKCE,HRAS Signaling p70S6K Signaling 1.82E+00 8.20E-02 RAF1,PIK3CA,YWHAE,EEF2,PIK3R1,PRKCE,HRAS,PPP2R5E,SFN,PLD1 TR/RXR Activation 1.80E+00 8.99E-02 PIK3CA,LDLR,PIK3R1,ACACA,RXRB,RXRA,SYT12,EP300 PPARÎ±/RXRÎ± 1.76E+00 7.27E-02 ADCY9,TGFBR2,RAF1,MED23,CKAP5,PRKAG2,HSP90AA1,HRAS,RXRA,NFKB1,P Activation RKAR1A,EP300 Role of NFAT in Cardiac 1.75E+00 6.95E-02 ADCY9,RAF1,PIK3CA,PIK3R1,HRAS,EP300,GNB1,TGFBR2,MEF2D,PRKAG2,PRK Hypertrophy CE,GSK3B,PRKAR1A IL-3 Signaling 1.73E+00 9.59E-02 RAF1,PIK3CA,CRKL,PIK3R1,RAC1,PRKCE,HRAS Small Cell Lung Cancer 1.73E+00 8.33E-02 PIK3CA,PIK3R1,SUV39H1,RXRB,RXRA,NFKB1,BCL2 Signaling GNRH Signaling 1.71E+00 7.63E-02 ADCY9,RAF1,MAP3K9,RAC1,PRKAG2,PRKCE,HRAS,MAPK7,NFKB1,PRKAR1A Role of BRCA1 in DNA 1.71E+00 1.02E-01 RBL2 (includes EG:100331892),POU2F1,CDKN1A,MSH6,RFC1,RFC3 Damage Response IL-17 Signaling 1.70E+00 9.46E-02 PIK3CA,PIK3R1,IL17RC,HRAS,CEBPB (includes EG:1051),GSK3B,NFKB1 Melanoma Signaling 1.67E+00 1.14E-01 RAF1,PIK3CA,PIK3R1,CDKN1A,HRAS LPS-stimulated MAPK 1.67E+00 8.86E-02 RAF1,PIK3CA,PIK3R1,RAC1,PRKCE,HRAS,NFKB1 Signaling fMLP Signaling in 1.67E+00 7.69E-02 GNB1,RAF1,PIK3CA,NFAT5,PIK3R1,RAC1,PRKCE,HRAS,NFKB1 Neutrophils Mismatch Repair in 1.67E+00 1.50E-01 MSH6,RFC1,RFC3 Eukaryotes Cardiac Hypertrophy 1.65E+00 6.58E-02 ADCY9,RAF1,MAP3K9,PIK3CA,PIK3R1,HRAS,EP300,TGFBR2,GNB1,RHOT1,MEF Signaling 2D,PRKAG2,GSK3B,FNBP1,PRKAR1A Myc Mediated Apoptosis 1.64E+00 1.00E-01 PIK3CA,YWHAE,PIK3R1,HRAS,SFN,BCL2 Signaling Glioblastoma Multiforme 1.63E+00 7.01E-02 RAF1,TSC1,PIK3CA,RHOT1,PIK3R1,WNT7B,CDKN1A,RAC1,HRAS,GSK3B,FNBP1 Signaling *Ratio of miR-139 targets in the pathway to the total number of genes in the pathway

Supplementary Table 4: Transcripts per million for miRNAs assessed by small RNA sequencing miRNA (tpm) MDA-MB- MCF7 hsa-let-7f-1-5p 1003.356053 1767.410792 231 hsa-let-7f-2-3p 69.19696917 112.8134548 hsa-let-7a-1-3p 345.9848459 394.8470919 hsa-let-7f-2-5p 1003.356053 1673.39958 hsa-let-7a-1-5p 6435.318133 5227.023407 hsa-let-7g-3p 622.7727226 376.0448494 hsa-let-7a-1-amb 34.59848459 112.8134548 hsa-let-7g-5p 2248.901498 1222.145761 hsa-let-7a-2-3p 518.9772688 0 hsa-let-7g-amb 34.59848459 0 hsa-let-7a-2-5p 6539.113587 5678.277226 hsa-let-7i-3p 380.5833305 319.638122 hsa-let-7a-2-amb 34.59848459 0 hsa-let-7i-5p 1902.916652 1428.970428 hsa-let-7a-3-3p 276.7878767 282.0336371 hsa-miR-1-1-3p 34.59848459 0 hsa-let-7a-3-5p 5985.537833 5358.639104 hsa-miR-1-2-3p 34.59848459 0 hsa-let-7b-3p 449.7802996 451.2538193 hsa-miR-100-3p 138.3939383 0 hsa-let-7b-5p 1107.151507 1522.98164 hsa-miR-100-5p 6400.719648 37.60448494 hsa-let-7b-amb 0 18.80224247 hsa-miR-101-1-3p 726.5681763 2275.071339 hsa-let-7c-3p 69.19696917 56.40672742 hsa-miR-101-1-5p 103.7954538 18.80224247 hsa-let-7c-5p 899.5605992 1052.925578 hsa-miR-101-2-3p 899.5605992 2444.291521 hsa-let-7d-3p 795.7651455 1203.343518 hsa-miR-101-2-5p 0 75.20896989 hsa-let-7d-5p 1107.151507 1767.410792 hsa-miR-103a-1-3p 14981.14383 9156.692084 hsa-let-7e-3p 311.3863613 244.4291521 hsa-miR-103a-1-5p 138.3939383 131.6156973 hsa-let-7e-5p 899.5605992 1015.321093 hsa-miR-103a-2-3p 15742.31049 9607.945903 hsa-let-7e-amb 103.7954538 18.80224247 hsa-miR-103a-2-5p 207.5909075 300.8358795 hsa-let-7f-1-3p 172.9924229 733.2874564 hsa-miR-103a-2-amb 0 18.80224247

216 hsa-miR-105-1-3p 34.59848459 0 hsa-miR-1266-5p 103.7954538 206.8246672 hsa-miR-105-1-5p 34.59848459 37.60448494 hsa-miR-1267-5p 34.59848459 0 hsa-miR-105-2-3p 34.59848459 0 hsa-miR-1268a-3p 345.9848459 131.6156973 hsa-miR-105-2-5p 34.59848459 37.60448494 hsa-miR-1268a-5p 1937.515137 169.2201822 hsa-miR-106a-5p 345.9848459 658.0784865 hsa-miR-1268a-amb 0 18.80224247 hsa-miR-106b-3p 5016.780265 3835.657464 hsa-miR-1268b-3p 0 56.40672742 hsa-miR-106b-5p 1556.931806 3760.448494 hsa-miR-1268b-5p 2283.499983 357.242607 hsa-miR-106b-amb 207.5909075 94.01121236 hsa-miR-1269a-3p 34.59848459 75.20896989 hsa-miR-107-3p 13493.40899 8272.986688 hsa-miR-1269b-5p 172.9924229 0 hsa-miR-107-5p 138.3939383 112.8134548 hsa-miR-127-3p 34.59848459 0 hsa-miR-10a-3p 34.59848459 0 hsa-miR-1270-1-5p 276.7878767 0 hsa-miR-10a-5p 622.7727226 169.2201822 hsa-miR-1270-2-5p 276.7878767 0 hsa-miR-10a-amb 69.19696917 0 hsa-miR-1271-3p 103.7954538 0 hsa-miR-10b-5p 0 37.60448494 hsa-miR-1271-5p 242.1893921 0 hsa-miR-1178-5p 0 18.80224247 hsa-miR-1273a-3p 3044.666644 2218.664612 hsa-miR-1179-5p 0 18.80224247 hsa-miR-1273a-5p 380.5833305 413.6493344 hsa-miR-1180-3p 242.1893921 300.8358795 hsa-miR-1273a-amb 172.9924229 150.4179398 hsa-miR-1180-5p 0 37.60448494 hsa-miR-1273c-3p 0 37.60448494 hsa-miR-1181-5p 69.19696917 18.80224247 hsa-miR-1273c-5p 172.9924229 75.20896989 hsa-miR-122-5p 34.59848459 0 hsa-miR-1273d-3p 484.3787842 357.242607 hsa-miR-1224-3p 0 56.40672742 hsa-miR-1273d-5p 103.7954538 112.8134548 hsa-miR-1224-5p 0 150.4179398 hsa-miR-1273d-amb 207.5909075 206.8246672 hsa-miR-1226-5p 34.59848459 56.40672742 hsa-miR-1273e-3p 1037.954538 733.2874564 hsa-miR-1226-amb 0 37.60448494 hsa-miR-1273e-5p 345.9848459 188.0224247 hsa-miR-1227-3p 138.3939383 225.6269097 hsa-miR-1273e-amb 345.9848459 131.6156973 hsa-miR-1228-3p 242.1893921 18.80224247 hsa-miR-1273f-3p 69.19696917 37.60448494 hsa-miR-1228-5p 34.59848459 0 hsa-miR-1273f-5p 345.9848459 94.01121236 hsa-miR-1229-3p 553.5757534 770.8919413 hsa-miR-1273f-amb 276.7878767 131.6156973 hsa-miR-1233-1-3p 34.59848459 18.80224247 hsa-miR-1273g-3p 1383.939383 940.1121236 hsa-miR-1233-2-3p 34.59848459 18.80224247 hsa-miR-1273g-5p 103.7954538 94.01121236 hsa-miR-1234-5p 172.9924229 1034.123336 hsa-miR-1275-5p 276.7878767 37.60448494 hsa-miR-1236-3p 103.7954538 37.60448494 hsa-miR-1276-5p 0 18.80224247 hsa-miR-1237-3p 276.7878767 131.6156973 hsa-miR-1277-5p 0 18.80224247 hsa-miR-1244-1-5p 0 18.80224247 hsa-miR-1277-amb 34.59848459 0 hsa-miR-1244-2-5p 0 18.80224247 hsa-miR-1278-3p 0 37.60448494 hsa-miR-1244-3-5p 0 18.80224247 hsa-miR-128-1-3p 761.1666609 1466.574913 hsa-miR-1246-5p 69.19696917 56.40672742 hsa-miR-128-1-5p 34.59848459 75.20896989 hsa-miR-1247-3p 0 37.60448494 hsa-miR-128-2-3p 657.3712071 1334.959215 hsa-miR-1247-5p 0 620.4740016 hsa-miR-1280-5p 34.59848459 0 hsa-miR-1248-5p 103.7954538 150.4179398 hsa-miR-1283-1-5p 0 56.40672742 hsa-miR-1249-3p 0 94.01121236 hsa-miR-1283-2-3p 69.19696917 18.80224247 hsa-miR-1250-5p 0 37.60448494 hsa-miR-1283-2-5p 0 56.40672742 hsa-miR-1251-3p 0 18.80224247 hsa-miR-1284-3p 0 94.01121236 hsa-miR-1254-1-3p 311.3863613 300.8358795 hsa-miR-1284-5p 0 56.40672742 hsa-miR-1254-1-5p 242.1893921 695.6829715 hsa-miR-1285-1-3p 138.3939383 1165.739033 hsa-miR-1254-2-5p 207.5909075 526.4627892 hsa-miR-1285-1-5p 34.59848459 376.0448494 hsa-miR-1254-2-amb 34.59848459 37.60448494 hsa-miR-1285-1-amb 276.7878767 131.6156973 hsa-miR-1255a-3p 34.59848459 94.01121236 hsa-miR-1285-2-3p 138.3939383 1090.530063 hsa-miR-1255a-5p 0 18.80224247 hsa-miR-1285-2-5p 138.3939383 94.01121236 hsa-miR-1256-5p 0 18.80224247 hsa-miR-1285-2-amb 0 18.80224247 hsa-miR-1257-3p 0 18.80224247 hsa-miR-1286-3p 0 37.60448494 hsa-miR-1257-5p 242.1893921 150.4179398 hsa-miR-1286-5p 0 18.80224247 hsa-miR-125a-3p 484.3787842 488.8583043 hsa-miR-1287-3p 34.59848459 37.60448494 hsa-miR-125a-5p 2456.492406 5640.672742 hsa-miR-1287-5p 138.3939383 112.8134548 hsa-miR-125b-1-3p 1591.530291 0 hsa-miR-1289-1-5p 34.59848459 0 hsa-miR-125b-1-5p 17368.43926 2669.918431 hsa-miR-129-1-3p 0 18.80224247 hsa-miR-125b-1-amb 69.19696917 0 hsa-miR-129-1-5p 0 18.80224247 hsa-miR-125b-2-3p 103.7954538 1109.332306 hsa-miR-129-2-3p 69.19696917 75.20896989 hsa-miR-125b-2-5p 17472.23472 2632.313946 hsa-miR-129-2-5p 0 37.60448494 hsa-miR-126-3p 345.9848459 977.7166085 hsa-miR-1290-3p 69.19696917 0 hsa-miR-126-5p 138.3939383 188.0224247 hsa-miR-1291-5p 0 37.60448494 hsa-miR-1260a-5p 484.3787842 582.8695166 hsa-miR-1292-3p 34.59848459 0 hsa-miR-1260b-5p 34.59848459 188.0224247 hsa-miR-1292-5p 242.1893921 131.6156973 hsa-miR-1261-5p 0 18.80224247 hsa-miR-1293-3p 34.59848459 112.8134548 hsa-miR-1266-3p 34.59848459 18.80224247 hsa-miR-1293-5p 34.59848459 75.20896989

217 hsa-miR-1293-amb 0 18.80224247 hsa-miR-150-5p 0 18.80224247 hsa-miR-1294-3p 69.19696917 131.6156973 hsa-miR-151a-3p 74248.34792 36382.33918 hsa-miR-1294-5p 34.59848459 0 hsa-miR-151a-5p 4601.59845 3572.42607 hsa-miR-1296-3p 69.19696917 0 hsa-miR-151b-3p 3978.825727 3177.578978 hsa-miR-1296-5p 2075.909075 338.4403645 hsa-miR-152-3p 276.7878767 545.2650317 hsa-miR-1297-3p 34.59848459 394.8470919 hsa-miR-152-5p 0 18.80224247 hsa-miR-1301-3p 311.3863613 1993.037702 hsa-miR-153-1-3p 34.59848459 0 hsa-miR-1301-5p 69.19696917 338.4403645 hsa-miR-153-1-5p 34.59848459 0 hsa-miR-1302-8-amb 0 37.60448494 hsa-miR-153-2-3p 34.59848459 0 hsa-miR-1303-3p 34.59848459 75.20896989 hsa-miR-1537-3p 34.59848459 0 hsa-miR-1303-5p 657.3712071 1372.5637 hsa-miR-1538-3p 34.59848459 0 hsa-miR-1304-3p 207.5909075 131.6156973 hsa-miR-1538-5p 34.59848459 18.80224247 hsa-miR-1304-5p 103.7954538 75.20896989 hsa-miR-15a-3p 380.5833305 263.2313946 hsa-miR-1306-3p 242.1893921 37.60448494 hsa-miR-15a-5p 1280.14393 1635.795095 hsa-miR-1306-5p 345.9848459 188.0224247 hsa-miR-15b-3p 415.181815 883.7053962 hsa-miR-1307-3p 1799.121198 1579.388368 hsa-miR-15b-5p 4221.01512 5509.057044 hsa-miR-1307-5p 968.7575684 714.4852139 hsa-miR-16-1-3p 103.7954538 94.01121236 hsa-miR-130a-3p 2767.878767 75.20896989 hsa-miR-16-1-5p 2525.689375 6919.22523 hsa-miR-130a-5p 2387.295436 0 hsa-miR-16-2-3p 345.9848459 394.8470919 hsa-miR-130a-amb 34.59848459 0 hsa-miR-16-2-5p 2387.295436 6731.202805 hsa-miR-130b-3p 1418.537868 1579.388368 hsa-miR-17-3p 761.1666609 244.4291521 hsa-miR-130b-5p 242.1893921 94.01121236 hsa-miR-17-5p 4463.204512 2745.127401 hsa-miR-132-3p 518.9772688 75.20896989 hsa-miR-181a-1-3p 484.3787842 282.0336371 hsa-miR-132-5p 207.5909075 56.40672742 hsa-miR-181a-1-5p 2318.098467 3440.810372 hsa-miR-132-amb 34.59848459 0 hsa-miR-181a-2-3p 518.9772688 0 hsa-miR-1322-3p 34.59848459 18.80224247 hsa-miR-181a-2-5p 588.174238 1203.343518 hsa-miR-133a-1-3p 0 37.60448494 hsa-miR-181a-2-amb 1522.333322 2275.071339 hsa-miR-133a-2-3p 0 37.60448494 hsa-miR-181b-1-3p 415.181815 244.4291521 hsa-miR-1343-3p 138.3939383 112.8134548 hsa-miR-181b-1-5p 2145.106044 2519.500491 hsa-miR-1343-5p 34.59848459 18.80224247 hsa-miR-181b-1-amb 1453.136353 1128.134548 hsa-miR-135a-1-5p 0 112.8134548 hsa-miR-181b-2-3p 69.19696917 18.80224247 hsa-miR-135a-2-5p 0 112.8134548 hsa-miR-181b-2-5p 3010.068159 3121.17225 hsa-miR-135b-3p 69.19696917 0 hsa-miR-181c-3p 207.5909075 112.8134548 hsa-miR-135b-5p 69.19696917 94.01121236 hsa-miR-181c-5p 207.5909075 94.01121236 hsa-miR-137-3p 34.59848459 0 hsa-miR-181d-3p 138.3939383 18.80224247 hsa-miR-137-5p 34.59848459 0 hsa-miR-181d-5p 1072.553022 244.4291521 hsa-miR-138-1-3p 1868.318168 0 hsa-miR-182-3p 34.59848459 188.0224247 hsa-miR-138-1-5p 57295.09048 0 hsa-miR-182-5p 1280.14393 6693.59832 hsa-miR-138-2-3p 0 18.80224247 hsa-miR-182-amb 0 18.80224247 hsa-miR-138-2-5p 56257.13594 18.80224247 hsa-miR-183-3p 103.7954538 639.276244 hsa-miR-139-3p 207.5909075 0 hsa-miR-183-5p 553.5757534 2688.720673 hsa-miR-139-5p 311.3863613 131.6156973 hsa-miR-184-3p 0 37.60448494 hsa-miR-140-3p 2352.696952 1635.795095 hsa-miR-184-5p 0 18.80224247 hsa-miR-140-5p 103.7954538 94.01121236 hsa-miR-185-3p 0 56.40672742 hsa-miR-140-amb 34.59848459 0 hsa-miR-185-5p 380.5833305 1504.179398 hsa-miR-141-3p 103.7954538 2068.246672 hsa-miR-186-3p 34.59848459 56.40672742 hsa-miR-141-5p 0 282.0336371 hsa-miR-186-5p 311.3863613 526.4627892 hsa-miR-141-amb 0 244.4291521 hsa-miR-187-3p 69.19696917 18.80224247 hsa-miR-142-3p 103.7954538 0 hsa-miR-188-3p 103.7954538 75.20896989 hsa-miR-142-5p 103.7954538 0 hsa-miR-188-5p 5431.96208 5885.101894 hsa-miR-143-3p 34.59848459 56.40672742 hsa-miR-18a-3p 311.3863613 56.40672742 hsa-miR-144-5p 0 37.60448494 hsa-miR-18a-5p 1383.939383 413.6493344 hsa-miR-145-5p 34.59848459 75.20896989 hsa-miR-18b-5p 276.7878767 37.60448494 hsa-miR-146a-5p 9030.204477 0 hsa-miR-1908-3p 69.19696917 150.4179398 hsa-miR-146b-5p 69.19696917 169.2201822 hsa-miR-1908-5p 172.9924229 56.40672742 hsa-miR-147b-3p 276.7878767 18.80224247 hsa-miR-1909-3p 69.19696917 18.80224247 hsa-miR-147b-5p 0 37.60448494 hsa-miR-1909-5p 138.3939383 18.80224247 hsa-miR-148a-3p 345.9848459 545.2650317 hsa-miR-190a-3p 0 18.80224247 hsa-miR-148a-5p 34.59848459 37.60448494 hsa-miR-190a-5p 0 37.60448494 hsa-miR-148b-3p 172.9924229 620.4740016 hsa-miR-191-3p 276.7878767 244.4291521 hsa-miR-148b-5p 207.5909075 958.9143661 hsa-miR-191-5p 3217.659067 4963.792013 hsa-miR-149-3p 0 75.20896989 hsa-miR-1910-3p 138.3939383 507.6605467 hsa-miR-149-5p 449.7802996 2857.940856 hsa-miR-1910-5p 172.9924229 714.4852139 hsa-miR-149-amb 0 18.80224247 hsa-miR-1911-3p 0 18.80224247 hsa-miR-150-3p 0 18.80224247 hsa-miR-1911-5p 0 56.40672742

218 hsa-miR-1913-3p 0 56.40672742 hsa-miR-2116-3p 0 56.40672742 hsa-miR-1913-5p 34.59848459 56.40672742 hsa-miR-212-3p 380.5833305 112.8134548 hsa-miR-1914-3p 138.3939383 112.8134548 hsa-miR-212-5p 69.19696917 37.60448494 hsa-miR-1914-5p 172.9924229 94.01121236 hsa-miR-214-3p 34.59848459 0 hsa-miR-1915-3p 34.59848459 601.6717591 hsa-miR-215-5p 172.9924229 432.4515769 hsa-miR-1915-5p 242.1893921 244.4291521 hsa-miR-216a-3p 34.59848459 0 hsa-miR-192-3p 172.9924229 131.6156973 hsa-miR-216a-5p 34.59848459 0 hsa-miR-192-5p 518.9772688 789.6941838 hsa-miR-216b-5p 69.19696917 0 hsa-miR-193a-3p 795.7651455 4174.097829 hsa-miR-217-5p 34.59848459 0 hsa-miR-193a-5p 207.5909075 169.2201822 hsa-miR-218-1-3p 138.3939383 0 hsa-miR-193b-3p 968.7575684 6975.631957 hsa-miR-218-1-5p 345.9848459 75.20896989 hsa-miR-193b-5p 0 112.8134548 hsa-miR-218-2-5p 311.3863613 75.20896989 hsa-miR-194-1-5p 69.19696917 150.4179398 hsa-miR-219-1-3p 0 112.8134548 hsa-miR-194-2-3p 69.19696917 37.60448494 hsa-miR-219-1-5p 0 37.60448494 hsa-miR-194-2-5p 103.7954538 225.6269097 hsa-miR-22-3p 657.3712071 1184.541276 hsa-miR-195-3p 69.19696917 75.20896989 hsa-miR-22-5p 138.3939383 56.40672742 hsa-miR-195-5p 207.5909075 470.0560618 hsa-miR-221-3p 12801.4393 432.4515769 hsa-miR-196a-1-5p 34.59848459 56.40672742 hsa-miR-221-5p 726.5681763 37.60448494 hsa-miR-196a-2-3p 0 56.40672742 hsa-miR-222-3p 19513.54531 451.2538193 hsa-miR-196a-2-5p 34.59848459 75.20896989 hsa-miR-222-5p 2387.295436 56.40672742 hsa-miR-197-3p 2594.886344 3252.787948 hsa-miR-224-3p 311.3863613 0 hsa-miR-197-5p 69.19696917 75.20896989 hsa-miR-224-5p 553.5757534 0 hsa-miR-1972-1-3p 138.3939383 188.0224247 hsa-miR-2276-3p 0 282.0336371 hsa-miR-1972-1-5p 0 75.20896989 hsa-miR-2276-5p 0 225.6269097 hsa-miR-1972-2-3p 138.3939383 188.0224247 hsa-miR-2277-3p 207.5909075 94.01121236 hsa-miR-1972-2-5p 0 75.20896989 hsa-miR-2277-5p 34.59848459 56.40672742 hsa-miR-1973-3p 0 37.60448494 hsa-miR-2278-3p 0 18.80224247 hsa-miR-1973-amb 69.19696917 18.80224247 hsa-miR-2278-5p 172.9924229 150.4179398 hsa-miR-1976-3p 103.7954538 18.80224247 hsa-miR-2278-amb 0 56.40672742 hsa-miR-1976-5p 34.59848459 0 hsa-miR-2355-3p 0 56.40672742 hsa-miR-199a-1-3p 34.59848459 56.40672742 hsa-miR-2355-5p 34.59848459 75.20896989 hsa-miR-199a-1-5p 34.59848459 18.80224247 hsa-miR-23a-3p 10725.53022 13594.02131 hsa-miR-199a-2-3p 34.59848459 56.40672742 hsa-miR-23a-5p 242.1893921 56.40672742 hsa-miR-199a-2-5p 34.59848459 18.80224247 hsa-miR-23b-3p 8580.424177 11224.93876 hsa-miR-199b-3p 34.59848459 56.40672742 hsa-miR-23b-5p 380.5833305 169.2201822 hsa-miR-199b-5p 34.59848459 112.8134548 hsa-miR-23b-amb 0 18.80224247 hsa-miR-19a-3p 380.5833305 451.2538193 hsa-miR-23c-3p 484.3787842 752.0896989 hsa-miR-19a-5p 69.19696917 0 hsa-miR-24-1-3p 11901.8787 16658.78683 hsa-miR-19b-1-3p 761.1666609 1184.541276 hsa-miR-24-1-5p 380.5833305 526.4627892 hsa-miR-19b-1-5p 69.19696917 56.40672742 hsa-miR-24-2-3p 10275.74992 15493.0478 hsa-miR-19b-2-3p 830.3636301 1203.343518 hsa-miR-24-2-5p 657.3712071 1015.321093 hsa-miR-200a-3p 138.3939383 958.9143661 hsa-miR-25-3p 1349.340899 2613.511704 hsa-miR-200a-5p 69.19696917 225.6269097 hsa-miR-25-5p 172.9924229 300.8358795 hsa-miR-200b-3p 172.9924229 996.518851 hsa-miR-2682-3p 103.7954538 0 hsa-miR-200b-5p 34.59848459 244.4291521 hsa-miR-2682-5p 69.19696917 0 hsa-miR-200c-3p 69.19696917 2763.929643 hsa-miR-26a-1-3p 172.9924229 37.60448494 hsa-miR-200c-5p 0 131.6156973 hsa-miR-26a-1-5p 1799.121198 6693.59832 hsa-miR-200c-amb 0 18.80224247 hsa-miR-26a-2-3p 34.59848459 112.8134548 hsa-miR-203-3p 0 2143.455642 hsa-miR-26a-2-5p 1868.318168 6204.740016 hsa-miR-203-5p 0 150.4179398 hsa-miR-26b-3p 172.9924229 338.4403645 hsa-miR-204-3p 34.59848459 0 hsa-miR-26b-5p 2837.075736 6054.322076 hsa-miR-204-5p 207.5909075 0 hsa-miR-27a-3p 9030.204477 10510.45354 hsa-miR-205-5p 172.9924229 206.8246672 hsa-miR-27a-5p 276.7878767 131.6156973 hsa-miR-20a-3p 415.181815 488.8583043 hsa-miR-27a-amb 2179.704529 94.01121236 hsa-miR-20a-5p 553.5757534 300.8358795 hsa-miR-27b-3p 7058.090856 8536.218082 hsa-miR-20b-5p 69.19696917 37.60448494 hsa-miR-27b-5p 242.1893921 300.8358795 hsa-miR-21-3p 1799.121198 3008.358795 hsa-miR-27b-amb 34.59848459 18.80224247 hsa-miR-21-5p 6435.318133 34784.14857 hsa-miR-28-3p 864.9621147 413.6493344 hsa-miR-21-amb 0 18.80224247 hsa-miR-28-5p 276.7878767 376.0448494 hsa-miR-210-3p 2975.469674 958.9143661 hsa-miR-296-3p 0 864.9031537 hsa-miR-210-5p 311.3863613 56.40672742 hsa-miR-296-5p 0 244.4291521 hsa-miR-211-5p 34.59848459 0 hsa-miR-2964a-5p 69.19696917 37.60448494 hsa-miR-2110-3p 69.19696917 357.242607 hsa-miR-29a-3p 19271.35591 902.5076386 hsa-miR-2110-5p 207.5909075 263.2313946 hsa-miR-29a-5p 172.9924229 18.80224247 hsa-miR-2114-5p 0 37.60448494 hsa-miR-29b-1-3p 1660.72726 601.6717591

219 hsa-miR-29b-1-5p 345.9848459 75.20896989 hsa-miR-3158-2-5p 0 112.8134548 hsa-miR-29b-2-3p 1764.522714 601.6717591 hsa-miR-3159-3p 553.5757534 1711.004065 hsa-miR-29b-2-5p 311.3863613 169.2201822 hsa-miR-3159-5p 0 18.80224247 hsa-miR-29c-3p 553.5757534 300.8358795 hsa-miR-3159-amb 0 37.60448494 hsa-miR-29c-5p 726.5681763 432.4515769 hsa-miR-3160-1-3p 0 18.80224247 hsa-miR-301a-3p 15915.30291 64811.3298 hsa-miR-3160-2-3p 0 18.80224247 hsa-miR-301a-5p 242.1893921 188.0224247 hsa-miR-3161-5p 0 56.40672742 hsa-miR-301b-3p 1764.522714 1560.586125 hsa-miR-3168-3p 34.59848459 0 hsa-miR-301b-5p 103.7954538 37.60448494 hsa-miR-3171-5p 0 18.80224247 hsa-miR-3064-3p 0 75.20896989 hsa-miR-3173-3p 0 37.60448494 hsa-miR-3064-5p 34.59848459 18.80224247 hsa-miR-3173-5p 69.19696917 112.8134548 hsa-miR-3065-3p 103.7954538 75.20896989 hsa-miR-3174-3p 0 37.60448494 hsa-miR-3065-5p 138.3939383 169.2201822 hsa-miR-3174-5p 34.59848459 56.40672742 hsa-miR-3074-3p 172.9924229 150.4179398 hsa-miR-3175-3p 34.59848459 56.40672742 hsa-miR-3074-5p 34.59848459 56.40672742 hsa-miR-3175-5p 34.59848459 37.60448494 hsa-miR-30a-3p 2629.484829 112.8134548 hsa-miR-3176-3p 69.19696917 507.6605467 hsa-miR-30a-5p 46327.37086 3986.075404 hsa-miR-3176-5p 0 18.80224247 hsa-miR-30b-3p 138.3939383 37.60448494 hsa-miR-3177-3p 207.5909075 206.8246672 hsa-miR-30b-5p 2075.909075 2481.896006 hsa-miR-3177-5p 0 18.80224247 hsa-miR-30c-1-3p 69.19696917 18.80224247 hsa-miR-3180-1-3p 34.59848459 18.80224247 hsa-miR-30c-1-5p 2075.909075 1109.332306 hsa-miR-3180-1-5p 0 56.40672742 hsa-miR-30c-2-3p 207.5909075 0 hsa-miR-3180-2-3p 34.59848459 18.80224247 hsa-miR-30c-2-5p 1868.318168 902.5076386 hsa-miR-3180-2-5p 0 56.40672742 hsa-miR-30c-2-amb 34.59848459 0 hsa-miR-3180-3-3p 34.59848459 18.80224247 hsa-miR-30d-3p 103.7954538 56.40672742 hsa-miR-3180-3-5p 0 56.40672742 hsa-miR-30d-5p 12836.03778 21096.11605 hsa-miR-3180-4-5p 34.59848459 18.80224247 hsa-miR-30e-3p 207.5909075 169.2201822 hsa-miR-3180-5-5p 34.59848459 18.80224247 hsa-miR-30e-5p 2664.083313 3534.821585 hsa-miR-3182-5p 34.59848459 225.6269097 hsa-miR-30e-amb 69.19696917 0 hsa-miR-3183-5p 103.7954538 56.40672742 hsa-miR-31-3p 0 18.80224247 hsa-miR-3185-amb 0 18.80224247 hsa-miR-31-5p 0 150.4179398 hsa-miR-3187-3p 691.9696917 1316.156973 hsa-miR-3116-1-5p 0 75.20896989 hsa-miR-3187-5p 69.19696917 18.80224247 hsa-miR-3116-2-5p 0 75.20896989 hsa-miR-3188-3p 0 37.60448494 hsa-miR-3127-3p 34.59848459 225.6269097 hsa-miR-3189-5p 0 18.80224247 hsa-miR-3127-5p 0 56.40672742 hsa-miR-3190-3p 34.59848459 37.60448494 hsa-miR-3128-5p 0 37.60448494 hsa-miR-3190-5p 0 18.80224247 hsa-miR-3130-1-3p 0 37.60448494 hsa-miR-3191-3p 34.59848459 0 hsa-miR-3130-1-5p 34.59848459 169.2201822 hsa-miR-3193-5p 0 18.80224247 hsa-miR-3130-2-3p 0 37.60448494 hsa-miR-3194-3p 0 18.80224247 hsa-miR-3130-2-5p 69.19696917 169.2201822 hsa-miR-3194-5p 622.7727226 658.0784865 hsa-miR-3131-5p 69.19696917 0 hsa-miR-3195-3p 34.59848459 263.2313946 hsa-miR-3135a-3p 138.3939383 56.40672742 hsa-miR-3195-5p 69.19696917 18.80224247 hsa-miR-3135a-5p 0 37.60448494 hsa-miR-3196-5p 415.181815 18.80224247 hsa-miR-3135a-amb 0 18.80224247 hsa-miR-3197-3p 0 37.60448494 hsa-miR-3135b-3p 34.59848459 150.4179398 hsa-miR-32-3p 103.7954538 188.0224247 hsa-miR-3136-5p 34.59848459 37.60448494 hsa-miR-32-5p 172.9924229 206.8246672 hsa-miR-3138-3p 34.59848459 94.01121236 hsa-miR-32-amb 172.9924229 75.20896989 hsa-miR-3138-5p 0 37.60448494 hsa-miR-3200-3p 242.1893921 206.8246672 hsa-miR-3139-5p 0 37.60448494 hsa-miR-3200-5p 0 131.6156973 hsa-miR-3143-3p 69.19696917 18.80224247 hsa-miR-320a-3p 7715.462063 6167.135531 hsa-miR-3143-5p 69.19696917 0 hsa-miR-320a-5p 103.7954538 0 hsa-miR-3143-amb 0 18.80224247 hsa-miR-320a-amb 69.19696917 18.80224247 hsa-miR-3144-3p 0 37.60448494 hsa-miR-320b-1-3p 2733.280282 2444.291521 hsa-miR-3145-5p 0 18.80224247 hsa-miR-320b-2-3p 2525.689375 2557.104976 hsa-miR-3146-3p 0 18.80224247 hsa-miR-320c-1-3p 1660.72726 2049.444429 hsa-miR-3149-3p 172.9924229 37.60448494 hsa-miR-320c-2-3p 1383.939383 1428.970428 hsa-miR-3149-amb 0 37.60448494 hsa-miR-320d-1-3p 0 37.60448494 hsa-miR-3150a-3p 0 18.80224247 hsa-miR-320d-2-3p 0 56.40672742 hsa-miR-3150a-5p 34.59848459 56.40672742 hsa-miR-320e-3p 138.3939383 282.0336371 hsa-miR-3155b-5p 0 18.80224247 hsa-miR-323a-3p 0 18.80224247 hsa-miR-3157-3p 0 37.60448494 hsa-miR-324-3p 4670.795419 1146.936791 hsa-miR-3157-5p 69.19696917 56.40672742 hsa-miR-324-5p 6020.136318 2688.720673 hsa-miR-3158-1-3p 34.59848459 18.80224247 hsa-miR-324-amb 34.59848459 0 hsa-miR-3158-1-5p 0 112.8134548 hsa-miR-326-3p 864.9621147 1015.321093 hsa-miR-3158-2-3p 34.59848459 18.80224247 hsa-miR-326-5p 34.59848459 18.80224247

220 hsa-miR-328-3p 1522.333322 545.2650317 hsa-miR-3648-5p 657.3712071 263.2313946 hsa-miR-328-amb 0 37.60448494 hsa-miR-3648-amb 34.59848459 37.60448494 hsa-miR-329-1-3p 0 37.60448494 hsa-miR-3651-3p 69.19696917 0 hsa-miR-329-2-3p 0 37.60448494 hsa-miR-3652-5p 0 18.80224247 hsa-miR-330-3p 1383.939383 883.7053962 hsa-miR-3652-amb 34.59848459 0 hsa-miR-330-5p 207.5909075 56.40672742 hsa-miR-3653-5p 103.7954538 131.6156973 hsa-miR-331-3p 1280.14393 3647.63504 hsa-miR-3656-3p 34.59848459 0 hsa-miR-331-5p 1210.946961 582.8695166 hsa-miR-3657-5p 34.59848459 18.80224247 hsa-miR-331-amb 0 37.60448494 hsa-miR-365a-3p 103.7954538 695.6829715 hsa-miR-335-3p 207.5909075 244.4291521 hsa-miR-365a-5p 69.19696917 112.8134548 hsa-miR-335-5p 103.7954538 376.0448494 hsa-miR-365b-3p 103.7954538 695.6829715 hsa-miR-337-5p 0 18.80224247 hsa-miR-365b-5p 69.19696917 169.2201822 hsa-miR-338-3p 0 94.01121236 hsa-miR-3661-5p 69.19696917 18.80224247 hsa-miR-338-5p 0 18.80224247 hsa-miR-3663-3p 0 37.60448494 hsa-miR-339-3p 1037.954538 4568.944921 hsa-miR-3664-3p 34.59848459 56.40672742 hsa-miR-339-5p 7646.265094 15173.40967 hsa-miR-3664-5p 0 112.8134548 hsa-miR-339-amb 0 18.80224247 hsa-miR-3665-5p 795.7651455 300.8358795 hsa-miR-33a-3p 242.1893921 169.2201822 hsa-miR-3676-3p 1037.954538 789.6941838 hsa-miR-33a-5p 69.19696917 94.01121236 hsa-miR-3676-5p 0 188.0224247 hsa-miR-33a-amb 553.5757534 37.60448494 hsa-miR-3676-amb 276.7878767 658.0784865 hsa-miR-33b-3p 3494.446943 2500.698249 hsa-miR-3677-3p 69.19696917 75.20896989 hsa-miR-33b-5p 415.181815 188.0224247 hsa-miR-3677-5p 207.5909075 131.6156973 hsa-miR-340-3p 138.3939383 206.8246672 hsa-miR-3679-3p 0 37.60448494 hsa-miR-340-5p 34.59848459 94.01121236 hsa-miR-3679-5p 103.7954538 206.8246672 hsa-miR-342-3p 553.5757534 33035.54002 hsa-miR-3680-1-3p 0 18.80224247 hsa-miR-342-5p 69.19696917 225.6269097 hsa-miR-3680-1-5p 0 37.60448494 hsa-miR-342-amb 0 75.20896989 hsa-miR-3680-2-3p 0 18.80224247 hsa-miR-345-3p 69.19696917 75.20896989 hsa-miR-3680-2-5p 0 37.60448494 hsa-miR-345-5p 2525.689375 5057.803225 hsa-miR-3681-3p 34.59848459 0 hsa-miR-346-3p 34.59848459 0 hsa-miR-3681-5p 34.59848459 0 hsa-miR-34a-3p 311.3863613 394.8470919 hsa-miR-3682-5p 34.59848459 0 hsa-miR-34a-5p 1314.742414 1165.739033 hsa-miR-3683-5p 0 18.80224247 hsa-miR-34a-amb 34.59848459 131.6156973 hsa-miR-3687-3p 518.9772688 94.01121236 hsa-miR-34b-3p 207.5909075 0 hsa-miR-3687-5p 138.3939383 18.80224247 hsa-miR-34b-5p 207.5909075 18.80224247 hsa-miR-3687-amb 103.7954538 18.80224247 hsa-miR-34c-3p 242.1893921 0 hsa-miR-3688-1-5p 0 18.80224247 hsa-miR-34c-5p 172.9924229 0 hsa-miR-3688-2-5p 0 18.80224247 hsa-miR-3545-3p 0 75.20896989 hsa-miR-3688-2-amb 0 18.80224247 hsa-miR-3605-3p 311.3863613 56.40672742 hsa-miR-3691-3p 34.59848459 0 hsa-miR-3605-5p 103.7954538 0 hsa-miR-3691-5p 0 18.80224247 hsa-miR-3605-amb 34.59848459 0 hsa-miR-3692-5p 34.59848459 18.80224247 hsa-miR-3607-3p 34.59848459 37.60448494 hsa-miR-373-3p 34.59848459 0 hsa-miR-3607-5p 0 56.40672742 hsa-miR-374a-3p 69.19696917 37.60448494 hsa-miR-3607-amb 0 18.80224247 hsa-miR-374a-5p 69.19696917 206.8246672 hsa-miR-3609-3p 34.59848459 112.8134548 hsa-miR-374b-3p 69.19696917 131.6156973 hsa-miR-361-3p 380.5833305 676.880729 hsa-miR-374b-5p 242.1893921 1748.60855 hsa-miR-361-5p 449.7802996 789.6941838 hsa-miR-374c-5p 0 131.6156973 hsa-miR-3610-3p 34.59848459 18.80224247 hsa-miR-375-3p 0 507.6605467 hsa-miR-3613-3p 34.59848459 1090.530063 hsa-miR-375-5p 0 112.8134548 hsa-miR-3613-5p 0 188.0224247 hsa-miR-376b-3p 0 18.80224247 hsa-miR-3614-3p 69.19696917 18.80224247 hsa-miR-376c-3p 0 18.80224247 hsa-miR-3614-5p 69.19696917 94.01121236 hsa-miR-378a-3p 22834.99983 42756.29938 hsa-miR-3615-3p 415.181815 263.2313946 hsa-miR-378a-5p 242.1893921 206.8246672 hsa-miR-3616-3p 0 37.60448494 hsa-miR-378b-3p 0 169.2201822 hsa-miR-3617-3p 69.19696917 0 hsa-miR-378c-5p 22523.61347 42173.42986 hsa-miR-3617-5p 69.19696917 0 hsa-miR-378d-1-3p 22454.4165 41778.58277 hsa-miR-3619-3p 103.7954538 75.20896989 hsa-miR-378d-2-5p 22523.61347 42229.83659 hsa-miR-3619-5p 311.3863613 507.6605467 hsa-miR-378e-3p 172.9924229 488.8583043 hsa-miR-362-3p 103.7954538 263.2313946 hsa-miR-378f-3p 69.19696917 75.20896989 hsa-miR-362-5p 172.9924229 300.8358795 hsa-miR-378g-5p 34.59848459 112.8134548 hsa-miR-3620-3p 380.5833305 357.242607 hsa-miR-378i-5p 103.7954538 56.40672742 hsa-miR-3620-5p 34.59848459 0 hsa-miR-383-3p 0 18.80224247 hsa-miR-3622a-3p 34.59848459 18.80224247 hsa-miR-3908-3p 0 18.80224247 hsa-miR-363-3p 0 18.80224247 hsa-miR-3909-3p 34.59848459 169.2201822 hsa-miR-3648-3p 588.174238 300.8358795 hsa-miR-3909-5p 103.7954538 37.60448494

221 hsa-miR-3909-amb 0 18.80224247 hsa-miR-4429-5p 622.7727226 827.2986688 hsa-miR-3911-5p 34.59848459 0 hsa-miR-4430-3p 172.9924229 56.40672742 hsa-miR-3912-3p 34.59848459 37.60448494 hsa-miR-4430-amb 0 18.80224247 hsa-miR-3913-1-3p 0 56.40672742 hsa-miR-4435-1-3p 34.59848459 0 hsa-miR-3913-1-5p 0 75.20896989 hsa-miR-4435-1-5p 34.59848459 18.80224247 hsa-miR-3913-2-3p 0 56.40672742 hsa-miR-4435-2-3p 34.59848459 0 hsa-miR-3913-2-5p 0 75.20896989 hsa-miR-4435-2-5p 34.59848459 18.80224247 hsa-miR-3916-5p 34.59848459 37.60448494 hsa-miR-4440-3p 34.59848459 0 hsa-miR-3917-3p 0 18.80224247 hsa-miR-4440-5p 0 18.80224247 hsa-miR-3917-5p 34.59848459 0 hsa-miR-4443-5p 0 18.80224247 hsa-miR-3917-amb 0 18.80224247 hsa-miR-4444-1-3p 0 18.80224247 hsa-miR-3922-5p 172.9924229 0 hsa-miR-4444-2-3p 0 18.80224247 hsa-miR-3926-1-5p 0 18.80224247 hsa-miR-4446-3p 0 319.638122 hsa-miR-3927-5p 0 18.80224247 hsa-miR-4446-5p 0 94.01121236 hsa-miR-3928-3p 103.7954538 75.20896989 hsa-miR-4448-3p 34.59848459 37.60448494 hsa-miR-3928-5p 0 18.80224247 hsa-miR-4449-3p 0 37.60448494 hsa-miR-3929-3p 103.7954538 112.8134548 hsa-miR-4449-5p 276.7878767 206.8246672 hsa-miR-3929-5p 103.7954538 282.0336371 hsa-miR-4451-3p 0 18.80224247 hsa-miR-3929-amb 553.5757534 507.6605467 hsa-miR-4452-5p 0 18.80224247 hsa-miR-3934-5p 0 18.80224247 hsa-miR-4454-5p 207.5909075 112.8134548 hsa-miR-3935-3p 0 18.80224247 hsa-miR-4459-3p 34.59848459 18.80224247 hsa-miR-3939-3p 34.59848459 18.80224247 hsa-miR-4459-5p 69.19696917 94.01121236 hsa-miR-3940-3p 276.7878767 75.20896989 hsa-miR-4459-amb 103.7954538 94.01121236 hsa-miR-3940-5p 172.9924229 18.80224247 hsa-miR-4461-3p 276.7878767 545.2650317 hsa-miR-3943-5p 34.59848459 0 hsa-miR-4461-5p 69.19696917 0 hsa-miR-3944-3p 69.19696917 0 hsa-miR-4463-3p 0 18.80224247 hsa-miR-3944-5p 69.19696917 56.40672742 hsa-miR-4466-5p 345.9848459 131.6156973 hsa-miR-3960-3p 5950.939349 2575.907219 hsa-miR-4467-5p 0 18.80224247 hsa-miR-3960-5p 69.19696917 0 hsa-miR-4469-3p 34.59848459 0 hsa-miR-409-3p 0 18.80224247 hsa-miR-4470-3p 0 18.80224247 hsa-miR-421-3p 795.7651455 3403.205887 hsa-miR-4472-2-3p 34.59848459 37.60448494 hsa-miR-422a-5p 34.59848459 75.20896989 hsa-miR-4472-2-amb 0 18.80224247 hsa-miR-423-3p 3909.628758 2632.313946 hsa-miR-4473-3p 0 37.60448494 hsa-miR-423-5p 380.5833305 357.242607 hsa-miR-4474-5p 0 18.80224247 hsa-miR-424-3p 1280.14393 432.4515769 hsa-miR-4479-3p 103.7954538 56.40672742 hsa-miR-424-5p 1487.734837 2049.444429 hsa-miR-448-3p 0 18.80224247 hsa-miR-425-3p 1245.545445 977.7166085 hsa-miR-448-5p 0 18.80224247 hsa-miR-425-5p 2041.310591 7972.150808 hsa-miR-4484-3p 138.3939383 526.4627892 hsa-miR-4254-3p 0 18.80224247 hsa-miR-4485-3p 1902.916652 1616.992853 hsa-miR-4279-3p 276.7878767 225.6269097 hsa-miR-4485-5p 726.5681763 470.0560618 hsa-miR-4284-5p 1072.553022 2594.709461 hsa-miR-4488-5p 172.9924229 0 hsa-miR-4286-5p 1453.136353 282.0336371 hsa-miR-4492-3p 0 37.60448494 hsa-miR-429-3p 103.7954538 658.0784865 hsa-miR-4492-5p 34.59848459 37.60448494 hsa-miR-429-5p 0 18.80224247 hsa-miR-4498-5p 0 18.80224247 hsa-miR-4296-5p 34.59848459 0 hsa-miR-449a-5p 207.5909075 18.80224247 hsa-miR-4300-3p 0 18.80224247 hsa-miR-449b-3p 34.59848459 0 hsa-miR-4300-amb 0 94.01121236 hsa-miR-449b-5p 103.7954538 18.80224247 hsa-miR-4301-5p 207.5909075 300.8358795 hsa-miR-449c-5p 138.3939383 18.80224247 hsa-miR-4313-3p 0 18.80224247 hsa-miR-4500-3p 0 18.80224247 hsa-miR-4324-3p 138.3939383 94.01121236 hsa-miR-4502-3p 34.59848459 0 hsa-miR-4325-3p 0 56.40672742 hsa-miR-4504-3p 0 18.80224247 hsa-miR-4326-5p 34.59848459 0 hsa-miR-4504-5p 34.59848459 0 hsa-miR-4326-amb 138.3939383 0 hsa-miR-4508-5p 0 18.80224247 hsa-miR-4419a-3p 0 18.80224247 hsa-miR-450a-1-3p 34.59848459 0 hsa-miR-4419a-5p 242.1893921 112.8134548 hsa-miR-450a-1-5p 103.7954538 413.6493344 hsa-miR-4419a-amb 380.5833305 206.8246672 hsa-miR-450a-2-5p 103.7954538 413.6493344 hsa-miR-4419b-3p 103.7954538 37.60448494 hsa-miR-450b-5p 34.59848459 56.40672742 hsa-miR-4419b-amb 138.3939383 37.60448494 hsa-miR-4512-3p 0 56.40672742 hsa-miR-4420-3p 103.7954538 0 hsa-miR-4512-5p 484.3787842 770.8919413 hsa-miR-4421-3p 69.19696917 0 hsa-miR-4516-5p 138.3939383 37.60448494 hsa-miR-4421-amb 34.59848459 0 hsa-miR-4516-amb 0 56.40672742 hsa-miR-4422-5p 0 18.80224247 hsa-miR-4517-5p 0 18.80224247 hsa-miR-4423-5p 34.59848459 0 hsa-miR-4519-5p 34.59848459 0 hsa-miR-4424-3p 0 18.80224247 hsa-miR-451a-5p 0 37.60448494 hsa-miR-4428-3p 0 18.80224247 hsa-miR-451a-amb 0 37.60448494

222 hsa-miR-452-3p 138.3939383 0 hsa-miR-4710-5p 69.19696917 0 hsa-miR-452-5p 207.5909075 0 hsa-miR-4713-5p 0 282.0336371 hsa-miR-4521-3p 0 18.80224247 hsa-miR-4714-3p 0 18.80224247 hsa-miR-4521-5p 380.5833305 451.2538193 hsa-miR-4714-5p 0 18.80224247 hsa-miR-4523-5p 34.59848459 94.01121236 hsa-miR-4715-amb 0 18.80224247 hsa-miR-4525-3p 0 300.8358795 hsa-miR-4717-3p 0 18.80224247 hsa-miR-4525-5p 0 18.80224247 hsa-miR-4717-5p 34.59848459 37.60448494 hsa-miR-4526-3p 0 56.40672742 hsa-miR-4720-5p 0 18.80224247 hsa-miR-4532-5p 622.7727226 37.60448494 hsa-miR-4721-5p 34.59848459 0 hsa-miR-4536-1-3p 0 37.60448494 hsa-miR-4722-3p 34.59848459 18.80224247 hsa-miR-454-3p 588.174238 3534.821585 hsa-miR-4723-3p 34.59848459 56.40672742 hsa-miR-454-5p 138.3939383 188.0224247 hsa-miR-4723-5p 0 37.60448494 hsa-miR-455-3p 21347.26499 3722.844009 hsa-miR-4724-5p 0 18.80224247 hsa-miR-455-5p 380.5833305 94.01121236 hsa-miR-4725-3p 0 75.20896989 hsa-miR-4632-3p 34.59848459 0 hsa-miR-4725-5p 69.19696917 206.8246672 hsa-miR-4633-5p 0 18.80224247 hsa-miR-4726-3p 0 37.60448494 hsa-miR-4635-3p 0 18.80224247 hsa-miR-4726-5p 69.19696917 37.60448494 hsa-miR-4638-3p 138.3939383 94.01121236 hsa-miR-4728-3p 138.3939383 300.8358795 hsa-miR-4638-5p 0 18.80224247 hsa-miR-4730-5p 34.59848459 56.40672742 hsa-miR-4640-3p 34.59848459 37.60448494 hsa-miR-4731-3p 138.3939383 37.60448494 hsa-miR-4640-5p 0 75.20896989 hsa-miR-4731-5p 34.59848459 0 hsa-miR-4644-5p 0 18.80224247 hsa-miR-4732-5p 0 18.80224247 hsa-miR-4645-3p 0 18.80224247 hsa-miR-4733-5p 0 37.60448494 hsa-miR-4646-3p 34.59848459 56.40672742 hsa-miR-4737-amb 0 37.60448494 hsa-miR-4647-5p 69.19696917 0 hsa-miR-4738-3p 0 18.80224247 hsa-miR-4647-amb 0 18.80224247 hsa-miR-4741-3p 103.7954538 0 hsa-miR-4648-5p 0 37.60448494 hsa-miR-4741-5p 69.19696917 37.60448494 hsa-miR-4652-3p 0 56.40672742 hsa-miR-4742-3p 34.59848459 18.80224247 hsa-miR-4653-3p 0 18.80224247 hsa-miR-4742-5p 34.59848459 18.80224247 hsa-miR-4655-3p 0 37.60448494 hsa-miR-4743-5p 69.19696917 0 hsa-miR-4655-5p 0 37.60448494 hsa-miR-4745-3p 0 94.01121236 hsa-miR-4659a-3p 0 18.80224247 hsa-miR-4745-5p 69.19696917 282.0336371 hsa-miR-4659a-5p 0 37.60448494 hsa-miR-4746-3p 0 37.60448494 hsa-miR-466-3p 0 37.60448494 hsa-miR-4746-5p 138.3939383 75.20896989 hsa-miR-466-5p 0 18.80224247 hsa-miR-4747-3p 0 37.60448494 hsa-miR-4660-5p 0 18.80224247 hsa-miR-4748-5p 34.59848459 0 hsa-miR-4661-5p 34.59848459 18.80224247 hsa-miR-4749-3p 138.3939383 37.60448494 hsa-miR-4662a-5p 34.59848459 0 hsa-miR-4749-5p 34.59848459 0 hsa-miR-4664-3p 34.59848459 18.80224247 hsa-miR-4750-5p 69.19696917 37.60448494 hsa-miR-4664-5p 0 56.40672742 hsa-miR-4754-3p 0 37.60448494 hsa-miR-4665-3p 34.59848459 18.80224247 hsa-miR-4756-3p 0 37.60448494 hsa-miR-4667-3p 69.19696917 94.01121236 hsa-miR-4756-5p 0 37.60448494 hsa-miR-4667-5p 34.59848459 18.80224247 hsa-miR-4757-3p 34.59848459 0 hsa-miR-4669-3p 0 18.80224247 hsa-miR-4757-5p 34.59848459 0 hsa-miR-4673-5p 0 18.80224247 hsa-miR-4758-3p 242.1893921 507.6605467 hsa-miR-4674-5p 0 18.80224247 hsa-miR-4758-5p 34.59848459 94.01121236 hsa-miR-4676-3p 34.59848459 0 hsa-miR-4758-amb 0 18.80224247 hsa-miR-4676-5p 34.59848459 18.80224247 hsa-miR-4760-3p 0 37.60448494 hsa-miR-4677-3p 0 18.80224247 hsa-miR-4762-5p 0 18.80224247 hsa-miR-4683-3p 34.59848459 56.40672742 hsa-miR-4763-3p 0 75.20896989 hsa-miR-4685-3p 34.59848459 0 hsa-miR-4763-5p 34.59848459 94.01121236 hsa-miR-4687-3p 69.19696917 18.80224247 hsa-miR-4765-5p 0 18.80224247 hsa-miR-4689-5p 34.59848459 37.60448494 hsa-miR-4767-5p 103.7954538 37.60448494 hsa-miR-4690-3p 172.9924229 131.6156973 hsa-miR-4769-3p 34.59848459 0 hsa-miR-4691-3p 0 18.80224247 hsa-miR-4783-3p 0 56.40672742 hsa-miR-4695-3p 0 37.60448494 hsa-miR-4786-3p 34.59848459 0 hsa-miR-4698-5p 0 18.80224247 hsa-miR-4787-3p 138.3939383 37.60448494 hsa-miR-4700-5p 34.59848459 0 hsa-miR-4787-5p 69.19696917 18.80224247 hsa-miR-4701-5p 0 37.60448494 hsa-miR-4789-3p 34.59848459 37.60448494 hsa-miR-4706-5p 34.59848459 18.80224247 hsa-miR-4791-5p 0 37.60448494 hsa-miR-4707-3p 103.7954538 131.6156973 hsa-miR-4795-5p 0 18.80224247 hsa-miR-4707-5p 207.5909075 75.20896989 hsa-miR-4797-5p 0 18.80224247 hsa-miR-4707-amb 34.59848459 0 hsa-miR-4798-5p 0 18.80224247 hsa-miR-4708-3p 0 18.80224247 hsa-miR-4799-5p 34.59848459 37.60448494 hsa-miR-4708-5p 0 18.80224247 hsa-miR-4800-3p 0 18.80224247

223 hsa-miR-4802-3p 0 18.80224247 hsa-miR-5189-3p 0 18.80224247 hsa-miR-4802-5p 0 18.80224247 hsa-miR-518a-1-3p 0 18.80224247 hsa-miR-483-3p 0 56.40672742 hsa-miR-518a-1-5p 0 18.80224247 hsa-miR-483-5p 0 75.20896989 hsa-miR-518a-2-3p 0 18.80224247 hsa-miR-484-5p 1660.72726 3440.810372 hsa-miR-518a-2-5p 0 18.80224247 hsa-miR-486-3p 138.3939383 18.80224247 hsa-miR-518b-3p 0 18.80224247 hsa-miR-486-5p 1210.946961 206.8246672 hsa-miR-518c-3p 0 18.80224247 hsa-miR-487b-3p 0 18.80224247 hsa-miR-518e-3p 0 18.80224247 hsa-miR-489-3p 3667.439366 60543.22076 hsa-miR-5196-3p 69.19696917 18.80224247 hsa-miR-489-5p 69.19696917 18.80224247 hsa-miR-519a-1-3p 34.59848459 18.80224247 hsa-miR-491-3p 69.19696917 18.80224247 hsa-miR-519a-1-5p 34.59848459 18.80224247 hsa-miR-491-5p 103.7954538 94.01121236 hsa-miR-519a-2-3p 34.59848459 18.80224247 hsa-miR-491-amb 34.59848459 0 hsa-miR-519a-2-5p 0 18.80224247 hsa-miR-492-3p 0 56.40672742 hsa-miR-519b-3p 0 18.80224247 hsa-miR-492-5p 0 56.40672742 hsa-miR-519c-3p 0 18.80224247 hsa-miR-494-3p 0 18.80224247 hsa-miR-519d-3p 0 18.80224247 hsa-miR-495-3p 0 18.80224247 hsa-miR-520b-5p 0 18.80224247 hsa-miR-497-3p 0 18.80224247 hsa-miR-520d-3p 0 18.80224247 hsa-miR-497-5p 207.5909075 300.8358795 hsa-miR-520d-5p 0 18.80224247 hsa-miR-5000-5p 0 18.80224247 hsa-miR-520f-3p 0 18.80224247 hsa-miR-5001-3p 34.59848459 37.60448494 hsa-miR-520g-3p 34.59848459 37.60448494 hsa-miR-5001-5p 0 56.40672742 hsa-miR-520g-5p 0 18.80224247 hsa-miR-5008-3p 103.7954538 150.4179398 hsa-miR-520h-3p 0 18.80224247 hsa-miR-5008-5p 69.19696917 75.20896989 hsa-miR-520h-5p 0 18.80224247 hsa-miR-500a-3p 138.3939383 338.4403645 hsa-miR-521-1-3p 34.59848459 18.80224247 hsa-miR-500a-5p 207.5909075 357.242607 hsa-miR-521-2-3p 34.59848459 18.80224247 hsa-miR-500b-3p 34.59848459 37.60448494 hsa-miR-522-3p 0 18.80224247 hsa-miR-500b-5p 207.5909075 357.242607 hsa-miR-525-3p 0 18.80224247 hsa-miR-501-3p 172.9924229 244.4291521 hsa-miR-526b-3p 0 18.80224247 hsa-miR-501-5p 345.9848459 451.2538193 hsa-miR-526b-5p 0 37.60448494 hsa-miR-5010-3p 0 18.80224247 hsa-miR-527-5p 0 18.80224247 hsa-miR-5010-5p 34.59848459 0 hsa-miR-532-3p 588.174238 1240.948003 hsa-miR-502-3p 138.3939383 376.0448494 hsa-miR-532-5p 1037.954538 2105.851157 hsa-miR-502-5p 138.3939383 282.0336371 hsa-miR-542-3p 103.7954538 658.0784865 hsa-miR-503-3p 553.5757534 37.60448494 hsa-miR-542-5p 138.3939383 75.20896989 hsa-miR-503-5p 6504.515102 1203.343518 hsa-miR-544b-3p 0 37.60448494 hsa-miR-504-3p 0 18.80224247 hsa-miR-544b-5p 0 18.80224247 hsa-miR-504-5p 0 75.20896989 hsa-miR-545-3p 138.3939383 112.8134548 hsa-miR-505-3p 1037.954538 1165.739033 hsa-miR-545-5p 69.19696917 112.8134548 hsa-miR-505-5p 172.9924229 357.242607 hsa-miR-548a-1-3p 34.59848459 18.80224247 hsa-miR-5087-3p 0 18.80224247 hsa-miR-548a-2-3p 34.59848459 0 hsa-miR-5090-3p 69.19696917 56.40672742 hsa-miR-548a-2-5p 0 18.80224247 hsa-miR-5090-5p 103.7954538 131.6156973 hsa-miR-548a-3-3p 69.19696917 0 hsa-miR-5094-5p 34.59848459 56.40672742 hsa-miR-548aa-1-3p 311.3863613 564.0672742 hsa-miR-5095-3p 69.19696917 545.2650317 hsa-miR-548aa-2-3p 311.3863613 582.8695166 hsa-miR-5095-5p 207.5909075 394.8470919 hsa-miR-548ab-5p 0 37.60448494 hsa-miR-5096-3p 518.9772688 733.2874564 hsa-miR-548ac-5p 0 18.80224247 hsa-miR-5096-5p 172.9924229 94.01121236 hsa-miR-548ad-5p 0 75.20896989 hsa-miR-5096-amb 276.7878767 300.8358795 hsa-miR-548ae-1-3p 0 18.80224247 hsa-miR-5100-3p 0 75.20896989 hsa-miR-548ae-2-3p 0 18.80224247 hsa-miR-512-1-3p 0 18.80224247 hsa-miR-548ag-2-3p 0 18.80224247 hsa-miR-512-1-5p 0 56.40672742 hsa-miR-548ai-5p 0 56.40672742 hsa-miR-512-2-3p 0 18.80224247 hsa-miR-548aj-1-3p 0 18.80224247 hsa-miR-512-2-5p 0 56.40672742 hsa-miR-548aj-2-3p 0 18.80224247 hsa-miR-513b-5p 0 18.80224247 hsa-miR-548aj-2-5p 0 37.60448494 hsa-miR-515-1-3p 0 37.60448494 hsa-miR-548ak-3p 0 37.60448494 hsa-miR-515-2-3p 0 37.60448494 hsa-miR-548ak-5p 0 37.60448494 hsa-miR-516a-1-3p 0 18.80224247 hsa-miR-548al-3p 34.59848459 37.60448494 hsa-miR-516a-2-3p 0 18.80224247 hsa-miR-548am-3p 0 18.80224247 hsa-miR-516b-1-3p 0 18.80224247 hsa-miR-548am-5p 0 18.80224247 hsa-miR-516b-2-3p 0 18.80224247 hsa-miR-548an-3p 0 37.60448494 hsa-miR-517a-3p 34.59848459 18.80224247 hsa-miR-548ao-5p 34.59848459 0 hsa-miR-517b-3p 34.59848459 18.80224247 hsa-miR-548ap-3p 34.59848459 0 hsa-miR-517c-3p 0 18.80224247 hsa-miR-548aq-5p 34.59848459 0 hsa-miR-5187-5p 0 18.80224247 hsa-miR-548au-5p 0 37.60448494

224 hsa-miR-548aw-3p 0 18.80224247 hsa-miR-566-amb 103.7954538 94.01121236 hsa-miR-548aw-5p 69.19696917 0 hsa-miR-5684-3p 207.5909075 94.01121236 hsa-miR-548b-3p 34.59848459 18.80224247 hsa-miR-5684-5p 242.1893921 413.6493344 hsa-miR-548b-5p 34.59848459 0 hsa-miR-5689-3p 0 56.40672742 hsa-miR-548c-3p 34.59848459 18.80224247 hsa-miR-5689-5p 34.59848459 18.80224247 hsa-miR-548c-5p 34.59848459 37.60448494 hsa-miR-5690-5p 0 37.60448494 hsa-miR-548d-1-3p 0 37.60448494 hsa-miR-5691-5p 0 18.80224247 hsa-miR-548d-2-3p 0 37.60448494 hsa-miR-5692a-1-3p 0 37.60448494 hsa-miR-548d-2-5p 0 18.80224247 hsa-miR-5692a-2-3p 0 18.80224247 hsa-miR-548e-3p 0 18.80224247 hsa-miR-5695-3p 34.59848459 75.20896989 hsa-miR-548f-1-5p 0 18.80224247 hsa-miR-5696-3p 0 18.80224247 hsa-miR-548f-2-3p 0 18.80224247 hsa-miR-5699-3p 34.59848459 56.40672742 hsa-miR-548f-3-3p 0 18.80224247 hsa-miR-5699-5p 103.7954538 357.242607 hsa-miR-548g-5p 34.59848459 0 hsa-miR-570-3p 0 37.60448494 hsa-miR-548h-2-3p 103.7954538 394.8470919 hsa-miR-570-5p 0 56.40672742 hsa-miR-548h-3-3p 34.59848459 56.40672742 hsa-miR-5708-3p 172.9924229 188.0224247 hsa-miR-548h-4-3p 69.19696917 56.40672742 hsa-miR-572-3p 0 37.60448494 hsa-miR-548i-1-3p 0 18.80224247 hsa-miR-572-5p 0 18.80224247 hsa-miR-548i-1-5p 34.59848459 37.60448494 hsa-miR-573-5p 34.59848459 37.60448494 hsa-miR-548i-2-3p 0 18.80224247 hsa-miR-574-3p 7369.477217 3365.601402 hsa-miR-548i-2-5p 0 18.80224247 hsa-miR-574-5p 2145.106044 902.5076386 hsa-miR-548i-3-3p 0 18.80224247 hsa-miR-576-3p 0 18.80224247 hsa-miR-548i-3-5p 0 18.80224247 hsa-miR-576-5p 0 75.20896989 hsa-miR-548i-4-5p 34.59848459 56.40672742 hsa-miR-577-5p 34.59848459 18.80224247 hsa-miR-548j-5p 34.59848459 0 hsa-miR-579-3p 0 37.60448494 hsa-miR-548n-3p 69.19696917 0 hsa-miR-579-5p 69.19696917 56.40672742 hsa-miR-548o-2-3p 34.59848459 18.80224247 hsa-miR-582-3p 34.59848459 0 hsa-miR-548o-2-5p 34.59848459 37.60448494 hsa-miR-582-5p 103.7954538 37.60448494 hsa-miR-548o-3p 34.59848459 18.80224247 hsa-miR-584-3p 311.3863613 0 hsa-miR-548p-3p 0 56.40672742 hsa-miR-584-5p 622.7727226 0 hsa-miR-548q-5p 34.59848459 150.4179398 hsa-miR-584-amb 34.59848459 0 hsa-miR-548t-3p 311.3863613 545.2650317 hsa-miR-588-5p 69.19696917 18.80224247 hsa-miR-548t-5p 0 18.80224247 hsa-miR-589-3p 172.9924229 451.2538193 hsa-miR-548v-3p 0 37.60448494 hsa-miR-589-5p 103.7954538 131.6156973 hsa-miR-548v-5p 0 56.40672742 hsa-miR-590-3p 34.59848459 56.40672742 hsa-miR-548w-3p 34.59848459 0 hsa-miR-590-5p 138.3939383 545.2650317 hsa-miR-548w-5p 0 18.80224247 hsa-miR-590-amb 69.19696917 0 hsa-miR-548z-3p 69.19696917 94.01121236 hsa-miR-597-3p 0 37.60448494 hsa-miR-549-3p 34.59848459 0 hsa-miR-597-5p 34.59848459 18.80224247 hsa-miR-549-5p 34.59848459 0 hsa-miR-598-3p 345.9848459 244.4291521 hsa-miR-550a-1-3p 207.5909075 150.4179398 hsa-miR-598-5p 103.7954538 37.60448494 hsa-miR-550a-1-5p 588.174238 1316.156973 hsa-miR-600-3p 34.59848459 0 hsa-miR-550a-2-3p 207.5909075 150.4179398 hsa-miR-602-5p 103.7954538 169.2201822 hsa-miR-550a-2-5p 588.174238 1316.156973 hsa-miR-603-3p 34.59848459 0 hsa-miR-550a-3-3p 207.5909075 150.4179398 hsa-miR-603-5p 0 18.80224247 hsa-miR-550a-3-5p 553.5757534 1109.332306 hsa-miR-608-5p 0 18.80224247 hsa-miR-550b-1-5p 69.19696917 0 hsa-miR-609-5p 0 18.80224247 hsa-miR-550b-2-5p 69.19696917 0 hsa-miR-610-3p 34.59848459 0 hsa-miR-551a-3p 69.19696917 263.2313946 hsa-miR-610-5p 0 18.80224247 hsa-miR-551b-3p 830.3636301 75.20896989 hsa-miR-611-5p 0 18.80224247 hsa-miR-551b-5p 69.19696917 0 hsa-miR-611-amb 34.59848459 0 hsa-miR-556-3p 0 18.80224247 hsa-miR-615-3p 276.7878767 733.2874564 hsa-miR-556-5p 0 18.80224247 hsa-miR-615-5p 34.59848459 0 hsa-miR-558-5p 69.19696917 75.20896989 hsa-miR-616-5p 34.59848459 56.40672742 hsa-miR-5580-3p 0 18.80224247 hsa-miR-618-5p 0 56.40672742 hsa-miR-5581-3p 0 56.40672742 hsa-miR-619-5p 1003.356053 1184.541276 hsa-miR-5582-5p 34.59848459 18.80224247 hsa-miR-622-3p 0 18.80224247 hsa-miR-5585-3p 103.7954538 94.01121236 hsa-miR-622-5p 0 56.40672742 hsa-miR-5587-3p 0 112.8134548 hsa-miR-624-3p 34.59848459 56.40672742 hsa-miR-5587-5p 34.59848459 56.40672742 hsa-miR-624-5p 138.3939383 169.2201822 hsa-miR-559-3p 0 18.80224247 hsa-miR-625-3p 2006.712106 2519.500491 hsa-miR-559-5p 0 18.80224247 hsa-miR-625-5p 34.59848459 131.6156973 hsa-miR-564-3p 0 18.80224247 hsa-miR-627-3p 0 75.20896989 hsa-miR-566-3p 415.181815 695.6829715 hsa-miR-627-5p 34.59848459 75.20896989 hsa-miR-566-5p 138.3939383 94.01121236 hsa-miR-628-3p 34.59848459 75.20896989

225 hsa-miR-628-5p 69.19696917 94.01121236 hsa-miR-874-3p 622.7727226 37.60448494 hsa-miR-629-3p 207.5909075 244.4291521 hsa-miR-874-5p 207.5909075 0 hsa-miR-629-5p 1280.14393 1955.433217 hsa-miR-877-3p 795.7651455 676.880729 hsa-miR-631-5p 0 37.60448494 hsa-miR-877-5p 3252.257551 3666.437282 hsa-miR-632-3p 0 18.80224247 hsa-miR-885-5p 34.59848459 18.80224247 hsa-miR-636-3p 138.3939383 244.4291521 hsa-miR-887-3p 553.5757534 94.01121236 hsa-miR-636-5p 0 18.80224247 hsa-miR-9-1-3p 0 112.8134548 hsa-miR-639-3p 0 18.80224247 hsa-miR-9-1-5p 34.59848459 56.40672742 hsa-miR-639-5p 0 18.80224247 hsa-miR-9-2-3p 0 112.8134548 hsa-miR-641-5p 103.7954538 112.8134548 hsa-miR-9-2-5p 34.59848459 56.40672742 hsa-miR-642a-3p 69.19696917 18.80224247 hsa-miR-9-3-3p 0 112.8134548 hsa-miR-642a-5p 34.59848459 0 hsa-miR-9-3-5p 34.59848459 56.40672742 hsa-miR-642b-3p 34.59848459 18.80224247 hsa-miR-92a-1-3p 3736.636335 2406.687036 hsa-miR-643-3p 0 56.40672742 hsa-miR-92a-1-5p 103.7954538 94.01121236 hsa-miR-643-5p 0 18.80224247 hsa-miR-92a-2-3p 3252.257551 2162.257884 hsa-miR-644b-3p 138.3939383 94.01121236 hsa-miR-92b-3p 2664.083313 1993.037702 hsa-miR-651-5p 0 56.40672742 hsa-miR-92b-5p 345.9848459 188.0224247 hsa-miR-652-3p 1107.151507 3233.985705 hsa-miR-93-3p 553.5757534 1128.134548 hsa-miR-652-5p 761.1666609 2030.642187 hsa-miR-93-5p 2525.689375 6373.960198 hsa-miR-653-3p 0 225.6269097 hsa-miR-933-3p 69.19696917 18.80224247 hsa-miR-653-5p 34.59848459 206.8246672 hsa-miR-934-3p 0 18.80224247 hsa-miR-657-3p 0 18.80224247 hsa-miR-935-3p 138.3939383 112.8134548 hsa-miR-659-3p 34.59848459 0 hsa-miR-937-3p 172.9924229 376.0448494 hsa-miR-659-5p 34.59848459 18.80224247 hsa-miR-937-5p 69.19696917 37.60448494 hsa-miR-660-3p 276.7878767 94.01121236 hsa-miR-938-5p 0 37.60448494 hsa-miR-660-5p 172.9924229 394.8470919 hsa-miR-939-3p 69.19696917 37.60448494 hsa-miR-662-3p 34.59848459 0 hsa-miR-939-5p 69.19696917 75.20896989 hsa-miR-663a-3p 657.3712071 37.60448494 hsa-miR-940-3p 4497.802996 7539.699231 hsa-miR-663a-5p 276.7878767 0 hsa-miR-940-5p 0 94.01121236 hsa-miR-663a-amb 138.3939383 18.80224247 hsa-miR-941-1-3p 1037.954538 1447.77267 hsa-miR-663b-3p 103.7954538 37.60448494 hsa-miR-941-1-5p 657.3712071 1015.321093 hsa-miR-663b-5p 34.59848459 0 hsa-miR-941-2-3p 1037.954538 1447.77267 hsa-miR-664-3p 138.3939383 94.01121236 hsa-miR-941-2-5p 1037.954538 1447.77267 hsa-miR-664-5p 34.59848459 37.60448494 hsa-miR-941-3-3p 1037.954538 1447.77267 hsa-miR-671-3p 588.174238 470.0560618 hsa-miR-941-3-5p 1037.954538 1447.77267 hsa-miR-671-5p 1245.545445 1955.433217 hsa-miR-941-4-3p 1037.954538 1447.77267 hsa-miR-675-3p 0 244.4291521 hsa-miR-941-4-5p 1037.954538 1447.77267 hsa-miR-675-5p 0 206.8246672 hsa-miR-942-3p 69.19696917 75.20896989 hsa-miR-676-3p 34.59848459 37.60448494 hsa-miR-942-5p 34.59848459 75.20896989 hsa-miR-676-5p 0 18.80224247 hsa-miR-942-amb 0 18.80224247 hsa-miR-7-1-3p 484.3787842 376.0448494 hsa-miR-943-3p 69.19696917 18.80224247 hsa-miR-7-1-5p 553.5757534 1974.23546 hsa-miR-944-3p 0 18.80224247 hsa-miR-7-2-3p 0 18.80224247 hsa-miR-944-5p 0 18.80224247 hsa-miR-7-2-5p 449.7802996 1466.574913 hsa-miR-95-3p 34.59848459 188.0224247 hsa-miR-7-3-5p 415.181815 1692.201822 hsa-miR-96-3p 69.19696917 169.2201822 hsa-miR-708-5p 34.59848459 150.4179398 hsa-miR-96-5p 103.7954538 1541.783883 hsa-miR-720-5p 4636.196935 1015.321093 hsa-miR-98-3p 138.3939383 244.4291521 hsa-miR-744-3p 1314.742414 1334.959215 hsa-miR-98-5p 934.1590838 996.518851 hsa-miR-744-5p 449.7802996 996.518851 hsa-miR-99a-3p 172.9924229 225.6269097 hsa-miR-760-3p 553.5757534 507.6605467 hsa-miR-99a-5p 276.7878767 1203.343518 hsa-miR-760-5p 138.3939383 206.8246672 hsa-miR-99b-3p 1349.340899 1109.332306 hsa-miR-762-3p 34.59848459 37.60448494 hsa-miR-99b-5p 1972.113621 3027.161038 hsa-miR-766-3p 5155.174203 94.01121236 hsa-miR-766-5p 69.19696917 0 hsa-miR-769-3p 1418.537868 225.6269097 hsa-miR-769-5p 657.3712071 507.6605467

Supplementary Table 5: Primer Sequences

Oligo Name Sequence Use mir139_Fwd_gDNA ATAGAAGGATCCACTGGCTTGGGAAGGGCGAGAGG Forward primer to amplify miR-139- 5p from genomic DNA mir139_Rev_gDNA TTAAAGAAGCTTCCTGCCAGAGACCTTTCCTCCTCCCT Reverse primer to amplify miR-139- 5p from genomic DNA BMI1_fwd CTA GTT GTT ATT ACG CTG TTT TGT GAA CCT GTA GAA Top strand of luciferase construct

226 AAC AAG TGC TTT TTA TCT TGA AAT with binding site BMI1__rev AGC TAT TTC AAG ATA AAA AGC ACT TGT TTT CTA CAG Bottom strand of luciferase construct GTT CAC AAA ACA GCG TAA TAA CAA HRAS_fwd CTA GGC CCC GGC TGC ATG AGC TGC AAG TGT GTG CTC Top strand of luciferase construct TCC TGA CGC AGG TGA GGG GGA CTC with binding site HRAS_rev AGC TGA GTC CCC CTC ACC TGC GTC AGG AGA GCA Bottom strand of luciferase construct CAC ACT TGC AGC TCA TGC AGC CGG GGC NFKB1_fwd CTA GGG AAA ACA CTG TGA GGA TGG GAT CTG CAC Top strand of luciferase construct TGT AAC TGC TGG ACC CAA GGA CAT GGT with binding site NFKB1_rev AGC TAC CAT GTC CTT GGG TCC AGC AGT TAC AGT GCA Bottom strand of luciferase construct GAT CCC ATC CTC ACA GTG TTT TCC NFKB1_fwd CTA GGC CAT GAT GAG CAA TAG CCT GCC ATG TTT GCT Top strand of luciferase construct GCT GCT GGT GGC CGC TGG GGC TGA with binding site NFKB1_rev AGC TTC AGC CCC AGC GGC CAC CAG CAG CAG CAA Bottom strand of luciferase construct ACA TGG CAG GCT ATT GCT CAT CAT GGC PIK3CA_fwd CTA GAA GTA ATA AAA ATA ATT TTA AAC ACA CTG Top strand of luciferase construct TAG TAA GAA ATG ACT GTT GGA AAA TTA with binding site PIK3CA_rev AGC TTA ATT TTC CAA CAG TCA TTT CTT ACT ACA GTG Bottom strand of luciferase construct TGT TTA AAA TTA TTT TTA TTA CTT PRKCE_fwd CTA GTA TTG AAG GGA TGT GAC ATT ACC TCC TGT AGA Top strand of luciferase construct TAT GCT AAC AGT GTT ATT CTT TCA with binding site PRKCE_rev AGC TTG AAA GAA TAA CAC TGT TAG CAT ATC TAC Bottom strand of luciferase construct AGG AGG TAA TGT CAC ATC CCT TCA ATA RAF_fwd CTA GCC CCA TGC TCA AGG CCC AGC CTT CTG TAG ATG Top strand of luciferase construct CGC AAG TGG ATG TTG ATG GTA GTA with binding site RAF_rev AGC TTA CTA CCA TCA ACA TCC ACT TGC GCA TCT ACA Bottom strand of luciferase construct GAA GGC TGG GCC TTG AGC ATG GGG RAF_fwd CTA GAC TTG TGG CTA CAA ATT TCA TGA GCA CTG TAG Top strand of luciferase construct CAC CAA AGT ACC TAC TAT GTG TGT with binding site RAF_rev AGC TAC ACA CAT AGT AGG TAC TTT GGT GCT ACA GTG Bottom strand of luciferase construct CTC ATG AAA TTT GTA GCC ACA AGT RHOT1_fwd CTA GTC AAG TAT TTA AAT AAA TAA CTG CTG TGT ACT Top strand of luciferase construct GTG ATC TTG AGT TCT TTT GTC ATC with binding site RHOT1_fwd AGC TGA TGA CAA AAG AAC TCA AGA TCA CAG TAC Bottom strand of luciferase construct ACA GCA GTT ATT TAT TTA AAT ACT TGA

227

3. APPENDIX

Supplementary Material for Chapter 4

Appendix 3.1: Summary of miRNAs that play a significant role in regulation of the cell cycle

miRNA Role in cell cycle Target Reference miR-15, 16 Regulates G1-S transition Cyclins D1, D3, E1, CDK6, CDC27, [183, 316] E2F3, WEE1 Let-7b Cell cycle progression Cyclins D1, D2, D3, A, CDK4, [454] CDK6, CDC25A, CDC34, Myc miR-17 family Regulates G1-S transition E2F1, CDKN1A, Cyclin D1, [258, 298] MYCN, p21CIP1, Rb miR19a Regulates G1-S transition Cyclin D1 [455] miR-21 Regulates G1-S transition CDC25A [456] miR-24 Regulates cell cycle progression p16INK4a, E2F2 [365, 457] miR-25 G1-S transition P57Kip2 [458] miR-26a Regulates cell cycle progression Cyclins D2, E2 [459] miR-31 Regulates cell cycle progression P16INK4a, p19INK4d [460] miR-34a Regulates G1-S transition Cyclins D1, E2, CDK4, CDK6, [188, 461] E2F1, E2F3, Myc miR-100 Regulates mitosis PLK1 [462] miR-124a Regulates G1-S transition CDK6 [463] miR-125b Regulates G1-S checkpoint CDK6, CDC25A, E2F3, Cyclin A, [464, 465] miR-128a Regulates G2-M transition WEE1 [466] miR-129 Regulates G1-S transition p21Cip1, CDK6 [467, 468] miR-137 Regulates G1-S transition CDK6 [463] miR-145 Enables p53 regulation of Myc Myc [469] miR-155 Regulates G2-M transition WEE1 [466] miR-181 family Regulates G1-S transition p27Kip1 [470] miR-210 Cell cycle regulation in ovarian cancer E2F3 [471] miR-221 Enables G1-S transition p27Kip1, p57Kip2 [164] miR-331 Blocks G1-S transition E2F1 [472] miR-322/424 Promotes cell cycle quiescence CDC25A [473] miR-449a/b Blocks G1-S transition CDC25A, CDK6 [474] miR-503 Promotes cell cycle quiescence CDC25A [473] miR-504 Blocks p53 mediates cell cycle arrest p53 [475] miR-516a Enables G2-M transition WEE1 [466] miR-1285 Enables cell cycle progression p53 [462] miR-1296 Regulates S phase MCM2 [476] miR-365b-3p Cell cycle progression p21, p27, cdc2, Cyclin D1, PAX6 [477] miR-142-3p Blocks G1-S progression CD133, ABCG2, Lgr5, CyclinD1 [478] miR-223 Induces G1-S block Ect2 [479] miR-200b Blocks cell cycle progression GATA4, CyclinD1 [479] miR-497-195 cluster Induce G1 arrest Cyclin E1, D3, CDC25A, CDK4 [480] miR-32 Promotes cell cycle progression PTEN [481]

229 Appendix 3.2: Depth of sequencing for the samples studied here. Numbers represent the number of tags aligned to miRBase derived pre-miRNA hairpins

Cell line Synchronized Phase Effective Library Sizes MCF7 G1 733020.4316 MCF7 S 787469.3512 MCF7 G2M 817938.8817 HeLa G1 445658.0818 HeLa S 309720.9925 HeLa G2M 873672.2233

Appendix 3.3: List of hairpin miRNAs and their corresponding potential cross-mapping miRNAs miRNA Cross mapping miRNAs* 5p,miR-548au-3p,miR-548d-5p,miR-548o- mir-520f miR-520f,miR-519c-5p,miR-522-5p,miR-519a- 5p,miR-548am-5p 5p,miR-526b-3p,miR-520b,miR-520c-3p,miR- mir-548ae- miR-548aj-5p,miR-548aq-3p,miR-548ay-5p,miR- 523-5p,miR-519b-5p,miR-518e-5p 2 548x-5p,miR-548x-3p,miR-548d-5p,miR-548az- mir-1269a miR-1269a,miR-1269b 5p,miR-548ae,miR-548n,miR-548g-5p, mir-23c miR-23c,miR-23b-3p mir-548aj-2 miR-548aj-5p,miR-548ay-5p,miR-548x-5p,miR- mir-301b miR-301b,miR-301a-3p 548x-3p,miR-548d-5p,miR-548o-5p,miR-548g- mir-320a miR-320a,miR-320c,miR-320b 5p,miR-548au-5p,miR-548aq-3p,miR-548c- 5p,miR-548ar-5p,miR-548aj-3p,miR-548am-5p mir-3689c miR-3689c,miR-3689b-3p,miR-3689a-3p,miR- mir-548aq miR-548t-3p,miR-548ae,miR-548h-3p,miR- 3689b-5p,miR-3689a-5p,miR-3689e 548aq-3p,miR-548am-3p,miR-548aa,miR-548ah- mir-378g miR-378g,miR-378c 3p,miR-548z,miR-548aj-3p,miR-548az-3p,miR- mir-4477a miR-4477a,miR-4477b 548u,miR-548aq-5p, mir-4477b miR-4477b,miR-4477a mir-548ay miR-548aj-5p,miR-548ay-5p,miR-548x-5p,miR- mir-449a miR-449a,miR-449b-5p 548d-5p,miR-548o-5p,miR-548ay-3p,miR-548g- mir-4662b miR-4662b,miR-4662a-3p,miR-4662a-5p 5p,miR-548au-5p,miR-548c-5p,miR-548ar- mir-484 miR-484,miR-4270 5p,miR-548ab,miR-548b-5p,miR-548az-5p,miR- mir-548ac miR-548ac,miR-548ay-5p,miR-548d-5p,miR- 548n,miR-548am-5p 548az-5p mir-548d-1 miR-548aj-5p,miR-548ay-5p,miR-548x-5p,miR- mir-548ak miR-548ak,miR-548w,miR-548h-3p,miR-548z 548d-5p,miR-548o-5p,miR-548g-5p,miR-548au- mir-548j miR-548j,miR-548ap-5p,miR-548q 5p,miR-548d-3p,miR-548c-5p,miR-548ar-5p,miR- mir-548l miR-548l,miR-548p,miR-548am-3p,miR-548ah- 548ab,miR-548b-5p,miR-548az-5p,miR- 3p 548n,miR-548am-5p mir-548v miR-548v,miR-548ax, mir-548d-2 miR-548aj-5p,miR-548ay-5p,miR-548x-5p,miR- mir-5681b miR-5681b,miR-5681a, 548d-5p,miR-548o-5p,miR-548g-5p,miR-548au- mir-5692b miR-5692b,miR-5692c, 5p,miR-548d-3p,miR-548c-5p,miR-548ar-5p,miR- mir-598 miR-598,miR-1322, 548ab,miR-548b-5p,miR-548az-5p,miR- mir-548h-3 miR-548t-3p,miR-548ae,miR-548h-5p,miR- 548n,miR-548am-5p 548p,miR-548h-3p,miR-548aq-3p,miR-548am- mir-548g miR-548aj-5p,miR-548ay-5p,miR-548x-5p,miR- 3p,miR-548aa,miR-548ah-3p,miR-548z,miR 548d-5p,miR-548g-3p,miR-548o-5p,miR-548g- 548aj-3p,miR-548az-3p,miR-548u 5p,miR-548au-5p,miR-548c-5p,miR-548ar- mir-548am miR-548aj-5p,miR-548i,miR-548av-3p,miR- 5p,miR-548am-5p 548ay-5p,miR-548x-5p,miR-548d-5p,miR-548o- mir-548x miR-548aj-5p,miR-548ay-5p,miR-548x-5p,miR- 5p,miR-548g-5p,miR-548au-5p,miR-548w,miR- 548x-3p,miR-548d-5p,miR-548ae,miR-548o- 548p,miR-548c-5p,miR-548am-3p,miR-548aq- 5p,miR-548g-5p,miR-548au-5p,miR-548aq- 3p,miR-548ar-5p,miR-548o-3p,miR-548as- 3p,miR-548c-5p,miR-548ar-5p,miR-548aj- 5p,miR-548ah-3p, 3p,miR-548am-5p, mir-548c miR-548aj-5p,miR-548i,miR-548ay-5p,miR-548x- mir-548al miR-548au-5p,miR-548aj-5p,miR-548c-5p,miR- 5p,miR-548d-5p,miR-548o-5p,miR-548c-3p,miR- 548ar-5p,miR-548y,miR-548x-5p,miR-548al,miR- 548g-5p,miR-548au-5p,miR-548w,miR-548c- 548o-5p,miR-548aq-5p,miR-548am-5p,miR-548g- 5p,miR-548ar-5p,miR-548as-5p,miR-548am-5p, 5p mir-548o-2 miR-548aj-5p,miR-548i,miR-548av-3p,miR- mir-520c miR-520f,miR-520e,miR-520c-5p,miR-519c- 548ay-5p,miR-548x-5p,miR-548d-5p,miR-548o- 5p,miR-522-5p,miR-526a,miR-519a-5p,miR- 5p,miR-548g-5p,miR-548au-5p,miR-548w,miR- 518d-5p,miR-518f-5p,miR-526b-3p,miR- 548am-3p,miR-548c-5p,miR-548ar-5p,miR-548o- 520b,miR-526b-5p,miR-520c-3p,miR-523- 3p,miR-548as-5p,miR-548ah-3p,miR-548am-5p 5p,miR-519b-5p,miR-518e-5p, mir-548au miR-548au-5p,miR-548w,miR-548i,miR-548c- mir-526a-1 miR-520c-5p,miR-519c-5p,miR-522-5p,miR- 5p,miR-548ar-5p,miR-548as-5p,miR-548ay- 526a,miR-519a-5p,miR-526b-5p,miR-518d-

230 5p,miR-518f-3p,miR-518f-5p,miR-523-5p,miR- mir-10a miR-10b-5p,miR-10a-5p,miR-10a-3p 519b-5p,miR-526b-3p,miR-518e-5p mir-10b miR-10b-5p,miR-10b-3p,miR-10a-5p mir-520b miR-520e,miR-519c-5p,miR-522-5p,miR-519a- mir-1178 miR-1178-3p 5p,miR-520c-3p,miR-523-5p,miR-519b-5p,miR- mir-1184-1 miR-1184 526b-3p,miR-518e-5p,miR-520b, mir-1184-2 miR-1184 mir-518e miR-520c-5p,miR-519c-5p,miR-522-5p,miR- mir-1184-3 miR-1184 526a,miR-519a-5p,miR-518d-5p,miR-518f- mir-1185-1 miR-1185-5p,miR-1185-1-3p,miR-1185-2-3p 5p,miR-518e-3p,miR-523-5p,miR-519b-5p,miR- mir-1185-2 miR-1185-5p,miR-1185-1-3p,miR-1185-2-3p 518e-5p mir-1207 miR-1207-5p mir-519c miR-520c-5p,miR-519c-5p,miR-522-5p,miR- mir-122 miR-122-5p,miR-122-3p, 526a,miR-519a-5p,miR-519c-3p,miR-518d- mir-1224 miR-1224-3p,miR-1224-5p, 5p,miR-518f-5p,miR-523-5p,miR-519b-5p,miR- 518e-5p mir-1225 miR-1225-5p,miR-1225-3p, mir-522 miR-522-3p,miR-520c-5p,miR-519c-5p,miR-522- mir-1226 miR-1226-5p,miR-1226-3p, 5p,miR-526a,miR-519a-5p,miR-518d-5p,miR- mir-1228 miR-1228-5p,miR-1228-3p, 518f-5p,miR-523-5p,miR-519b-5p,miR-518e-5p mir-1229 miR-1229-3p,miR-1229-5p, mir-523 miR-520c-5p,miR-519c-5p,miR-522-5p,miR- mir-1233-1 miR-1233-1-5p 526a,miR-519a-5p,miR-518d-5p,miR-518f- mir-1233-2 miR-1233-1-5p 5p,miR-523-3p,miR-523-5p,miR-519b-5p,miR- mir-1234 miR-1234-3p,miR-1234-5p, 518e-5p mir-1236 miR-1236-5p,miR-1236-3p, mir-526a-2 miR-520c-5p,miR-519c-5p,miR-522-5p,miR- mir-1237 miR-1237-5p,miR-1237-3p, 526a,miR-519a-5p,miR-518d-5p,miR-518f- mir-1238 miR-1238-3p,miR-1238-5p, 5p,miR-526b-5p,miR-523-5p,miR-519b-5p,miR- mir-124-1 miR-124-5p,miR-124-3p, 518e-5p mir-124-2 miR-124-5p,miR-124-3p, mir-519a-1 miR-520c-5p,miR-519c-5p,miR-522-5p,miR- mir-124-3 miR-124-5p,miR-124-3p, 526a,miR-519a-5p,miR-519e-3p,miR-519a- mir-1244-1 miR-1244 3p,miR-518d-5p,miR-518f-5p,miR-519b-3p,miR- mir-1244-2 miR-1244 523-5p,miR-519b-5p,miR-518e-5p mir-1244-3 miR-1244 mir-519b miR-520c-5p,miR-519c-5p,miR-522-5p,miR- mir-1245b miR-1245b-5p,miR-1245b-3p, 526a,miR-519a-5p,miR-519a-3p,miR-518d- mir-1247 miR-1247-5p,miR-1247-3p, 5p,miR-518f-5p,miR-526b-3p,miR-519b-3p,miR- 523-5p,miR-519b-5p,miR-518e-5p mir-1254-1 miR-1254 mir-518f miR-518c-3p,miR-520c-5p,miR-519c-5p,miR- mir-1254-2 miR-1254 522-5p,miR-526a,miR-519a-5p,miR-526b- mir-1255b- miR-1255b-5p 5p,miR-518d-5p,miR-518f-3p,miR-518f-5p,miR- 1 523-5p,miR-519b-5p,miR-518e-5p mir-1255b- miR-1255b-2-3p,miR-1255b-5p, mir-518d miR-520c-5p,miR-519c-5p,miR-522-5p,miR- 2 526a,miR-519a-5p,miR-526b-5p,miR-518d- mir-125a miR-125a-5p,miR-125a-3p, 5p,miR-518f-5p,miR-523-5p,miR-518d-3p,miR- mir-125b-1 miR-125b-1-3p,miR-125b-5p, 519b-5p,miR-518e-5p, mir-125b-2 miR-125b-2-3p,miR-125b-5p, let-7a-1 let-7a-5p,let-7f-2-3p,let-7c,let-7f-5p,let-7a-3p,let- mir-126 miR-126-5p,miR-126-3p, 7e-5p, mir-1269b miR-1269a,miR-1269b, let-7a-2 let-7a-5p,let-7c,let-7a-2-3p,let-7f-5p,let-7e-5p mir-127 miR-127-3p,miR-127-5p, let-7a-3 let-7a-5p,let-7f-2-3p,let-7c,let-7f-5p,let-7a-3p,let- mir-1270-1 miR-1270 7e-5p, mir-1270-2 miR-1270 let-7b let-7b-3p,let-7c,let-7b-5p mir-1271 miR-1271-5p,miR-1271-3p, let-7c let-7a-5p,let-7c,let-7b-5p mir-1273g miR-1273g-3p,miR-1273g-5p, let-7d let-7d-3p,let-7d-5p, mir-1277 miR-1277-3p,miR-1277-5p, let-7e let-7a-5p,let-7e-3p,let-7e-5p mir-128-1 miR-128 let-7f-1 let-7a-5p,let-7f-1-3p,let-7f-5p mir-128-2 miR-128 let-7f-2 let-7a-5p,let-7f-2-3p,let-7f-5p,let-7a-3p, mir-1283-1 miR-1283 let-7g let-7g-3p,let-7g-5p, mir-1283-2 miR-1283 let-7i let-7i-3p,let-7i-5p, mir-1285-1 miR-1285-3p,miR-1285-5p, mir-1-1 miR-1 mir-1285-2 miR-1285-3p mir-1-2 miR-1 mir-1289-1 miR-1289 mir-100 miR-100-3p,miR-99a-5p,miR-100-5p mir-1289-2 miR-1289 mir-101-1 miR-101-3p,miR-101-5p, mir-129-1 miR-129-2-3p,miR-129-1-3p,miR-129-5p mir-101-2 miR-101-3p mir-129-2 miR-129-2-3p,miR-129-1-3p,miR-129-5p mir-103a-1 miR-103a-2-5p,miR-103a-3p,miR-107 mir-1292 miR-1292-3p,miR-1292-5p, mir-103a-2 miR-103a-3p,miR-103a-2-5p,miR-107 mir-1295b miR-1295b-5p,miR-1295b-3p, mir-103b-1 miR-103b mir-1302-1 miR-1302 mir-103b-2 miR-103b mir-1302- miR-1302 mir-105-1 miR-105-3p,miR-105-5p, 10 mir-105-2 miR-105-3p,miR-105-5p, mir-1302- miR-1302 mir-106a miR-106a-5p,miR-106a-3p,miR-17-5p 11 mir-106b miR-106b-3p,miR-106b-5p, mir-1302-2 miR-1302 mir-107 miR-103a-3p,miR-103a-2-5p,miR-107 mir-1302-3 miR-1302 mir-1302-4 miR-1302

231 mir-1302-5 miR-1302 mir-193a miR-193a-3p,miR-193a-5p, mir-1302-6 miR-1302 mir-193b miR-193b-5p,miR-193b-3p, mir-1302-7 miR-1302 mir-194-1 miR-194-5p mir-1302-8 miR-1302 mir-194-2 miR-194-3p,miR-194-5p, mir-1302-9 miR-1302 mir-195 miR-195-3p,miR-195-5p, mir-1304 miR-1304-5p,miR-1304-3p, mir-196a-1 miR-196b-5p,miR-196a-5p, mir-1306 miR-1306-5p mir-196a-2 miR-196b-5p,miR-196a-3p,miR-196a-5p mir-1307 miR-1307-5p,miR-1307-3p, mir-196b miR-196b-5p,miR-196b-3p,miR-196a-5p mir-130a miR-130a-5p,miR-130a-3p, mir-197 miR-197-3p,miR-197-5p, mir-130b miR-130b-5p,miR-130b-3p, mir-1972-1 miR-1972 mir-132 miR-132-5p,miR-132-3p, mir-1972-2 miR-1972 mir-133a-1 miR-133a,miR-133b, mir-199a-1 miR-199b-3p,miR-199a-3p,miR-199a-5p mir-133a-2 miR-133a,miR-133b, mir-199a-2 miR-199b-3p,miR-199a-3p,miR-199a-5p mir-133b miR-133a,miR-133b, mir-199b miR-199b-3p,miR-199a-3p,miR-199b-5p mir-135a-1 miR-135a-5p,miR-135b-5p,miR-135a-3p mir-19a miR-19b-3p,miR-19a-3p,miR-19a-5p mir-135a-2 miR-135a-5p,miR-135b-5p, mir-19b-1 miR-19b-3p,miR-19b-2-5p,miR-19a-3p,miR-19b- mir-135b miR-135b-3p,miR-135a-5p,miR-135b-5p 1-5p, mir-136 miR-136-3p,miR-136-5p, mir-19b-2 miR-19b-3p,miR-19b-2-5p,miR-19a-3p,miR-19b- mir-138-1 miR-138-5p,miR-138-1-3p, 1-5p, mir-138-2 miR-138-5p,miR-138-2-3p, mir-200a miR-200a-3p,miR-141-3p,miR-200a-5p mir-139 miR-139-5p,miR-139-3p, mir-200b miR-200b-5p,miR-200c-3p,miR-200b-3p mir-140 miR-140-3p,miR-140-5p, mir-200c miR-200c-5p,miR-200b-3p,miR-200c-3p mir-141 miR-200a-3p,miR-141-3p,miR-141-5p mir-202 miR-202-3p,miR-202-5p, mir-142 miR-142-3p,miR-142-5p, mir-203a miR-203b-3p,miR-203a, mir-143 miR-143-5p,miR-143-3p, mir-203b miR-203b-3p,miR-203b-5p, mir-144 miR-144-5p,miR-144-3p, mir-204 miR-204-3p,miR-204-5p, mir-145 miR-145-5p,miR-145-3p, mir-205 miR-205-5p,miR-205-3p, mir-146a miR-146a-3p,miR-146a-5p, mir-20a miR-20a-3p,miR-20a-5p, mir-146b miR-146b-3p,miR-146b-5p, mir-20b miR-20b-5p,miR-20b-3p, mir-147a miR-147b,miR-147a, mir-21 miR-21-5p,miR-21-3p, mir-148a miR-148a-3p,miR-148b-3p,miR-148a-5p mir-211 miR-211-5p,miR-211-3p, mir-148b miR-148a-3p,miR-148b-5p,miR-148b-3p mir-2114 miR-2114-5p,miR-2114-3p, mir-149 miR-149-5p,miR-149-3p, mir-2115 miR-2115-5p,miR-2115-3p, mir-150 miR-150-5p,miR-150-3p, mir-2116 miR-2116-3p,miR-2116-5p, mir-151a miR-151a-5p,miR-151a-3p, mir-212 miR-212-3p,miR-212-5p, mir-151b miR-151a-5p mir-214 miR-214-3p,miR-214-5p, mir-153-1 miR-153 mir-216a miR-216a-5p,miR-216a-3p, mir-153-2 miR-153 mir-218-1 miR-218-5p,miR-218-1-3p, mir-154 miR-154-5p,miR-154-3p, mir-218-2 miR-218-5p,miR-218-2-3p, mir-155 miR-155-5p,miR-155-3p, mir-219-1 miR-219-1-3p,miR-219-5p, mir-15a miR-15a-3p,miR-15a-5p, mir-219-2 miR-219-5p,miR-219-2-3p, mir-15b miR-15b-3p,miR-15b-5p, mir-22 miR-22-3p,miR-22-5p, mir-16-1 miR-16-5p,miR-16-1-3p, mir-221 miR-221-3p,miR-221-5p, mir-16-2 miR-16-2-3p,miR-16-5p, mir-222 miR-222-5p,miR-222-3p, mir-17 miR-17-3p,miR-106a-5p,miR-17-5p mir-223 miR-223-5p,miR-223-3p, mir-181a-1 miR-181a-5p,miR-181a-3p, mir-224 miR-224-3p,miR-224-5p, mir-181a-2 miR-181a-5p,miR-181a-2-3p, mir-2277 miR-2277-5p,miR-2277-3p, mir-181b-1 miR-181b-5p,miR-181b-3p,miR-181d mir-2355 miR-2355-5p,miR-2355-3p, mir-181b-2 miR-181b-5p,miR-181d,miR-181b-3p mir-23a miR-23a-5p,miR-23a-3p,miR-23b-3p mir-181c miR-181c-3p,miR-181c-5p, mir-23b miR-23c,miR-23b-5p,miR-23a-3p,miR-23b-3p, mir-181d miR-181b-5p,miR-181d, mir-24-1 miR-24-1-5p,miR-24-3p, mir-182 miR-182-5p,miR-182-3p, mir-24-2 miR-24-3p,miR-24-2-5p, mir-183 miR-183-3p,miR-183-5p, mir-2467 miR-2467-3p,miR-2467-5p, mir-185 miR-185-3p,miR-185-5p, mir-25 miR-25-3p,miR-25-5p, mir-186 miR-186-5p,miR-186-3p, mir-2681 miR-2681-3p,miR-2681-5p, mir-187 miR-187-3p,miR-187-5p, mir-2682 miR-2682-5p,miR-2682-3p, mir-188 miR-188-5p,miR-188-3p, mir-26a-1 miR-26a-1-3p,miR-26a-5p, mir-18a miR-18b-5p,miR-18a-5p,miR-18a-3p mir-26a-2 miR-26a-2-3p,miR-26a-5p, mir-18b miR-18b-3p,miR-18b-5p,miR-18a-5p mir-26b miR-26b-3p,miR-26b-5p, mir-1909 miR-1909-5p,miR-1909-3p, mir-27a miR-27a-5p,miR-27b-3p,miR-27a-3p mir-191 miR-191-3p,miR-191-5p, mir-27b miR-27b-3p,miR-27b-5p,miR-27a-3p mir-1911 miR-1911-5p,miR-1911-3p, mir-28 miR-28-5p,miR-28-3p, mir-1914 miR-1914-5p,miR-1914-3p, mir-296 miR-296-5p,miR-296-3p, mir-1915 miR-1915-5p,miR-1915-3p, mir-2964a miR-2964a-3p,miR-2964a-5p, mir-192 miR-192-5p,miR-192-3p, mir-299 miR-299-5p,miR-299-3p,

232 mir-29a miR-29a-3p,miR-29a-5p,miR-29c-3p mir-3180-1 miR-3180,miR-3180-3p,miR-3180-5p mir-29b-1 miR-29b-1-5p,miR-29b-3p, mir-3180-2 miR-3180,miR-3180-3p,miR-3180-5p mir-29b-2 miR-29b-3p,miR-29b-2-5p, mir-3180-3 miR-3180,miR-3180-3p,miR-3180-5p mir-29c miR-29a-3p,miR-29c-5p,miR-29c-3p mir-3180-4 miR-3180,miR-3180-3p, mir-301a miR-301a-5p,miR-301b,miR-301a-3p mir-3180-5 miR-3180,miR-3180-3p, mir-302a miR-302b-3p,miR-302c-3p,miR-302a-5p,miR- mir-3184 miR-3184-3p,miR-3184-5p, 302a-3p, mir-3186 miR-3186-3p,miR-3186-5p, mir-302b miR-302c-3p,miR-302b-3p,miR-302d-3p,miR- mir-3187 miR-3187-3p,miR-3187-5p, 302b-5p,miR-302a-3p mir-3189 miR-3189-5p,miR-3189-3p, mir-302c miR-302c-5p,miR-302b-3p,miR-302c-3p,miR- mir-3190 miR-3190-5p,miR-3190-3p, 302d-3p,miR-302a-3p mir-3191 miR-3191-5p,miR-3191-3p, mir-302d miR-302b-3p,miR-302c-3p,miR-302d-3p,miR- mir-3194 miR-3194-5p,miR-3194-3p, 302d-5p, mir-3198-1 miR-3198 mir-302e miR-302c-3p mir-3198-2 miR-3198 mir-3064 miR-3064-3p,miR-3064-5p, mir-3199-1 miR-3199 mir-3065 miR-3065-3p,miR-3065-5p, mir-3199-2 miR-3199 mir-3074 miR-3074-3p,miR-3074-5p, mir-32 miR-32-3p,miR-32-5p, mir-30a miR-30e-5p,miR-30d-5p,miR-30e-3p,miR-30a- mir-3200 miR-3200-5p,miR-3200-3p, 3p,miR-30a-5p mir-3202-1 miR-3202 mir-30b miR-30b-5p,miR-30c-5p,miR-30b-3p mir-3202-2 miR-3202 mir-30c-1 miR-30b-5p,miR-30c-5p,miR-30c-1-3p mir-320b-1 miR-320a,miR-320c,miR-320b mir-30c-2 miR-30b-5p,miR-30c-2-3p,miR-30c-5p mir-320b-2 miR-320a,miR-320c,miR-320b mir-30d miR-30e-5p,miR-30d-5p,miR-30d-3p,miR-30a-5p, mir-320c-1 miR-320a,miR-320c,miR-320b mir-30e miR-30e-5p,miR-30d-5p,miR-30a-5p,miR-30e- mir-323a miR-323a-5p,miR-323a-3p, 3p,miR-30a-3p mir-323b miR-323b-3p,miR-323b-5p, mir-31 miR-31-3p,miR-31-5p, mir-324 miR-324-3p,miR-324-5p, mir-3116-1 miR-3116 mir-329-1 miR-329 mir-3116-2 miR-3116 mir-329-2 miR-329 mir-3117 miR-3117-5p,miR-3117-3p, mir-330 miR-330-5p,miR-330-3p, mir-3118-1 miR-3118 mir-331 miR-331-3p,miR-331-5p, mir-3118-2 miR-3118 mir-335 miR-335-5p,miR-335-3p, mir-3118-3 miR-3118 mir-337 miR-337-5p,miR-337-3p, mir-3118-4 miR-3118 mir-338 miR-338-5p,miR-338-3p, mir-3118-5 miR-3118 mir-339 miR-339-5p,miR-339-3p, mir-3118-6 miR-3118 mir-33a miR-33a-3p,miR-33a-5p, mir-3119-1 miR-3119 mir-33b miR-33b-3p,miR-33b-5p, mir-3119-2 miR-3119 mir-340 miR-340-5p,miR-340-3p, mir-3120 miR-3120-5p,miR-3120-3p, mir-342 miR-342-3p,miR-342-5p, mir-3121 miR-3121-5p,miR-3121-3p, mir-345 miR-345-5p,miR-345-3p, mir-3124 miR-3124-5p,miR-3124-3p, mir-34a miR-34a-3p,miR-34a-5p, mir-3126 miR-3126-5p,miR-3126-3p, mir-34b miR-34b-3p,miR-34b-5p, mir-3127 miR-3127-3p,miR-3127-5p, mir-34c miR-34c-3p,miR-34c-5p, mir-3129 miR-3129-5p,miR-3129-3p, mir-3529 miR-3529-5p,miR-3529-3p, mir-3130-1 miR-3130-5p,miR-3130-3p, mir-3591 miR-3591-3p,miR-3591-5p, mir-3130-2 miR-3130-5p,miR-3130-3p, mir-3605 miR-3605-3p,miR-3605-5p, mir-3136 miR-3136-3p,miR-3136-5p, mir-3606 miR-3606-3p,miR-3606-5p, mir-3140 miR-3140-5p,miR-3140-3p, mir-3607 miR-3607-3p,miR-3607-5p, mir-3144 miR-3144-5p,miR-3144-3p, mir-361 miR-361-5p,miR-361-3p, mir-3145 miR-3145-5p,miR-3145-3p, mir-3613 miR-3613-3p,miR-3613-5p, mir-3150a miR-3150a-3p,miR-3150a-5p, mir-3614 miR-3614-3p,miR-3614-5p, mir-3150b miR-3150b-5p,miR-3150b-3p, mir-3616 miR-3616-5p,miR-3616-3p, mir-3152 miR-3152-3p,miR-3152-5p, mir-3617 miR-3617-5p,miR-3617-3p, mir-3156-1 miR-3156-3p,miR-3156-5p, mir-3619 miR-3619-5p,miR-3619-3p, mir-3156-2 miR-3156-3p,miR-3156-5p, mir-362 miR-362-3p,miR-362-5p, mir-3156-3 miR-3156-3p,miR-3156-5p, mir-3620 miR-3620-5p,miR-3620-3p, mir-3157 miR-3157-5p,miR-3157-3p, mir-3622a miR-3622a-3p,miR-3622a-5p, mir-3158-1 miR-3158-5p,miR-3158-3p, mir-3622b miR-3622b-3p,miR-3622b-5p, mir-3158-2 miR-3158-5p,miR-3158-3p, mir-363 miR-363-3p,miR-363-5p, mir-3160-1 miR-3160-5p,miR-3160-3p, mir-365a miR-365b-3p,miR-365a-3p,miR-365a-5p mir-3160-2 miR-3160-5p,miR-3160-3p, mir-365b miR-365b-3p,miR-365a-3p,miR-365b-5p mir-3162 miR-3162-5p,miR-3162-3p, mir-3663 miR-3663-3p,miR-3663-5p, mir-3173 miR-3173-3p,miR-3173-5p, mir-3664 miR-3664-3p,miR-3664-5p, mir-3177 miR-3177-5p,miR-3177-3p, mir-3667 miR-3667-5p,miR-3667-3p, mir-3179-1 miR-3179 mir-367 miR-367-3p,miR-367-5p, mir-3179-2 miR-3179 mir-3670-1 miR-3670 mir-3179-3 miR-3179

233 mir-3670-2 miR-3670 mir-409 miR-409-5p,miR-409-3p, mir-3675 miR-3675-5p,miR-3675-3p, mir-411 miR-411-5p,miR-379-3p,miR-411-3p mir-3676 miR-5100,miR-3676-3p, mir-423 miR-423-3p,miR-423-5p, mir-3677 miR-3677-5p,miR-3677-3p, mir-424 miR-424-3p,miR-424-5p, mir-3678 miR-3678-3p,miR-3678-5p, mir-425 miR-425-3p,miR-425-5p, mir-3679 miR-3679-3p,miR-3679-5p, mir-4270 miR-484,miR-4270, mir-3680-1 miR-3680-5p,miR-3680-3p, mir-4290 miR-483-5p,miR-4290, mir-3680-2 miR-3680-3p,miR-3680-5p, mir-431 miR-431-3p,miR-431-5p, mir-3681 miR-3681-5p,miR-3681-3p, mir-432 miR-432-5p,miR-432-3p, mir-3682 miR-3682-5p,miR-3682-3p, mir-4323 miR-766-3p mir-3688-1 miR-3688-3p,miR-3688-5p, mir-4423 miR-4423-3p,miR-4423-5p, mir-3688-2 miR-3688-3p,miR-3688-5p, mir-4433 miR-4433-5p,miR-4433-3p, mir-3689a miR-3689c,miR-3689b-5p,miR-3689a-5p,miR- mir-4435-1 miR-4435 3689b-3p,miR-3689e,miR-3689f, mir-4435-2 miR-4435 mir-3689b miR-3689c,miR-3689b-5p,miR-3689a-5p,miR- mir-4436b- miR-4436b-5p,miR-4436b-3p, 3689b-3p,miR-3689e,miR-3689a-3p, 1 mir-3689d- miR-3689d mir-4436b- miR-4436b-5p,miR-4436b-3p, 1 2 mir-3689d- miR-3689d mir-4445 miR-4445-3p,miR-4445-5p, 2 mir-4446 miR-4446-5p,miR-4446-3p, mir-3689e miR-3689a-3p,miR-3689b-5p,miR-3689a-5p,miR- mir-4474 miR-4474-5p,miR-4474-3p, 3689e, mir-4482 miR-4482-5p,miR-4482-3p, mir-3689f miR-3689d,miR-3689f, mir-449b miR-449a,miR-449b-3p,miR-449b-5p mir-369 miR-369-5p,miR-369-3p, mir-449c miR-449c-3p,miR-449c-5p, mir-3690-1 miR-3690 mir-4509-1 miR-4509 mir-3690-2 miR-3690 mir-4509-2 miR-4509 mir-3691 miR-3691-3p,miR-3691-5p, mir-4509-3 miR-4509 mir-3692 miR-3692-5p,miR-3692-3p, mir-450a-1 miR-450a-5p mir-371a miR-371a-3p,miR-371a-5p, mir-450a-2 miR-450a-5p,miR-450a-3p, mir-371b miR-371b-3p,miR-371b-5p, mir-450b miR-450b-3p,miR-450b-5p, mir-373 miR-373-5p,miR-373-3p, mir-452 miR-452-3p,miR-452-5p, mir-374a miR-374a-3p,miR-374a-5p, mir-4520a miR-4520a-3p,miR-4520a-5p,miR-4520b-5p,miR- mir-374b miR-374c-3p,miR-374c-5p,miR-374b-3p,miR- 4520b-3p, 374b-5p, mir-4520b miR-4520a-3p,miR-4520a-5p,miR-4520b-5p,miR- mir-374c miR-374c-5p,miR-374b-5p,miR-374c-3p 4520b-3p, mir-376a-1 miR-376a-5p,miR-376b-3p,miR-376a-3p mir-4524a miR-4524a-5p,miR-4524a-3p, mir-376a-2 miR-376a-2-5p,miR-376b-3p,miR-376a-3p mir-4524b miR-4524b-3p,miR-4524b-5p, mir-376b miR-376c-5p,miR-376b-5p,miR-376b-3p,miR- mir-4529 miR-4529-3p,miR-4529-5p, 376a-3p, mir-4536-1 miR-4536-3p,miR-4536-5p, mir-376c miR-376c-5p,miR-376b-5p,miR-376c-3p mir-4536-2 miR-4536-5p,miR-4536-3p, mir-377 miR-377-5p,miR-377-3p, mir-454 miR-454-3p,miR-454-5p, mir-378a miR-378a-5p,miR-378a-3p,miR-378d,miR- mir-455 miR-455-5p,miR-455-3p, 378i,miR-378c mir-4632 miR-4632-3p,miR-4632-5p, mir-378c miR-378a-3p,miR-378d,miR-378c mir-4633 miR-4633-5p,miR-4633-3p, mir-378d-1 miR-378a-3p,miR-378d,miR-378c mir-4638 miR-4638-3p,miR-4638-5p, mir-378d-2 miR-378a-3p,miR-378d,miR-378c mir-4639 miR-4639-3p,miR-4639-5p, mir-378i miR-378a-3p,miR-378i, mir-4640 miR-4640-3p,miR-4640-5p, mir-379 miR-411-3p,miR-379-5p,miR-379-3p mir-4645 miR-4645-3p,miR-4645-5p, mir-380 miR-380-5p,miR-380-3p, mir-4646 miR-4646-3p,miR-4646-5p, mir-381 miR-381-3p,miR-381-5p, mir-4649 miR-4649-5p,miR-4649-3p, mir-382 miR-382-5p,miR-382-3p, mir-4650-1 miR-4650-5p,miR-4650-3p, mir-3910-1 miR-3910 mir-4650-2 miR-4650-5p,miR-4650-3p, mir-3910-2 miR-3910 mir-4652 miR-4652-3p,miR-4652-5p, mir-3913-1 miR-3913-5p,miR-3913-3p, mir-4653 miR-4653-5p,miR-4653-3p, mir-3913-2 miR-3913-5p,miR-3913-3p, mir-4655 miR-4655-3p,miR-4655-5p, mir-3914-1 miR-3914 mir-4659a miR-4659a-3p,miR-4659b-3p,miR-4659a-5p mir-3914-2 miR-3914 mir-4659b miR-4659a-3p,miR-4659b-3p,miR-4659a-5p,miR- mir-3922 miR-3922-3p,miR-3922-5p, 4659b-5p, mir-3925 miR-3925-3p,miR-3925-5p, mir-4661 miR-4661-3p,miR-4661-5p, mir-3926-1 miR-3926 mir-4662a miR-4662b,miR-4662a-3p,miR-4662a-5p mir-3926-2 miR-3926 mir-4664 miR-4664-5p,miR-4664-3p, mir-3927 miR-3927-5p,miR-3927-3p, mir-4665 miR-4665-3p,miR-4665-5p, mir-3934 miR-3934-3p,miR-3934-5p, mir-4666a miR-4666a-5p,miR-4666a-3p, mir-3940 miR-3940-5p,miR-3940-3p, mir-4667 miR-4667-3p,miR-4667-5p, mir-3942 miR-3942-5p,miR-3942-3p, mir-4668 miR-4668-3p,miR-4668-5p, mir-3944 miR-3944-5p,miR-3944-3p, mir-4670 miR-4670-5p,miR-4670-3p,

234 mir-4671 miR-4671-5p,miR-4671-3p, mir-4771-2 miR-4771 mir-4676 miR-4676-5p,miR-4676-3p, mir-4772 miR-4772-5p,miR-4772-3p, mir-4677 miR-4677-5p,miR-4677-3p, mir-4773-1 miR-4773 mir-4679-1 miR-4679 mir-4773-2 miR-4773 mir-4679-2 miR-4679 mir-4774 miR-4774-5p,miR-4774-3p, mir-4680 miR-4680-5p,miR-4680-3p, mir-4776-1 miR-4776-5p,miR-4776-3p, mir-4684 miR-4684-5p,miR-4684-3p, mir-4776-2 miR-4776-5p,miR-4776-3p, mir-4685 miR-4685-5p,miR-4685-3p, mir-4777 miR-4777-5p,miR-4777-3p, mir-4687 miR-4687-3p,miR-4687-5p, mir-4778 miR-4778-3p,miR-4778-5p, mir-4690 miR-4690-3p,miR-4690-5p, mir-4781 miR-4781-3p,miR-4781-5p, mir-4691 miR-4691-5p,miR-4691-3p, mir-4782 miR-4782-5p,miR-4782-3p, mir-4693 miR-4693-5p,miR-4693-3p, mir-4783 miR-4783-3p,miR-4783-5p, mir-4694 miR-4694-3p,miR-4694-5p, mir-4786 miR-4786-5p,miR-4786-3p, mir-4695 miR-4695-3p,miR-4695-5p, mir-4787 miR-4787-5p,miR-4787-3p, mir-4697 miR-4697-5p,miR-4697-3p, mir-4789 miR-4789-5p,miR-4789-3p, mir-4699 miR-4699-5p,miR-4699-3p, mir-4790 miR-4790-3p,miR-4790-5p, mir-4700 miR-4700-3p,miR-4700-5p, mir-4793 miR-4793-3p,miR-4793-5p, mir-4701 miR-4701-3p,miR-4701-5p, mir-4795 miR-4795-5p,miR-4795-3p, mir-4703 miR-4703-3p,miR-4703-5p, mir-4796 miR-4796-5p,miR-4796-3p, mir-4704 miR-4704-3p,miR-4704-5p, mir-4797 miR-4797-5p,miR-4797-3p, mir-4707 miR-4707-3p,miR-4707-5p, mir-4798 miR-4798-3p,miR-4798-5p, mir-4708 miR-4708-3p,miR-4708-5p, mir-4799 miR-4799-5p,miR-4799-3p, mir-4709 miR-4709-5p,miR-4709-3p, mir-4800 miR-4800-3p,miR-4800-5p, mir-4711 miR-4711-5p mir-4802 miR-4802-5p,miR-4802-3p, mir-4712 miR-4712-5p,miR-4712-3p, mir-4804 miR-4804-5p,miR-4804-3p, mir-4713 miR-4713-3p,miR-4713-5p, mir-483 miR-483-3p,miR-483-5p,miR-4290 mir-4714 miR-4714-3p,miR-4714-5p, mir-485 miR-485-5p,miR-485-3p, mir-4715 miR-4715-3p,miR-4715-5p, mir-486 miR-486-3p,miR-486-5p, mir-4716 miR-4716-5p,miR-4716-3p, mir-488 miR-488-5p,miR-488-3p, mir-4717 miR-4717-3p,miR-4717-5p, mir-490 miR-490-3p,miR-490-5p, mir-4720 miR-4720-3p,miR-4720-5p, mir-491 miR-491-3p,miR-491-5p, mir-4722 miR-4722-3p,miR-4722-5p, mir-493 miR-493-5p,miR-493-3p, mir-4723 miR-4723-3p,miR-4723-5p, mir-495 miR-495-3p,miR-495-5p, mir-4724 miR-4724-3p,miR-4724-5p, mir-497 miR-497-3p,miR-497-5p, mir-4725 miR-4725-5p,miR-4725-3p, mir-4999 miR-4999-5p,miR-4999-3p, mir-4726 miR-4726-3p,miR-4726-5p, mir-499a miR-499a-3p,miR-499a-5p, mir-4727 miR-4727-5p,miR-4727-3p, mir-499b miR-499b-3p,miR-499b-5p, mir-4728 miR-4728-5p,miR-4728-3p, mir-5000 miR-5000-5p,miR-5000-3p, mir-4731 miR-4731-3p,miR-4731-5p, mir-5001 miR-5001-5p,miR-5001-3p, mir-4732 miR-4732-3p,miR-4732-5p, mir-5002 miR-5002-3p,miR-5002-5p, mir-4733 miR-4733-5p,miR-4733-3p, mir-5003 miR-5003-3p,miR-5003-5p, mir-4735 miR-4735-5p,miR-4735-3p, mir-5004 miR-5004-3p,miR-5004-5p, mir-4738 miR-4738-5p,miR-4738-3p, mir-5006 miR-5006-3p,miR-5006-5p, mir-4740 miR-4740-5p,miR-4740-3p, mir-5007 miR-5007-3p,miR-5007-5p, mir-4742 miR-4742-5p,miR-4742-3p, mir-5008 miR-5008-3p,miR-5008-5p, mir-4743 miR-4743-3p,miR-4743-5p, mir-5009 miR-5009-3p,miR-5009-5p, mir-4745 miR-4745-5p,miR-4745-3p, mir-500a miR-501-3p,miR-502-3p,miR-500a-5p,miR-500a- mir-4746 miR-4746-3p,miR-4746-5p, 3p, mir-4747 miR-4747-5p,miR-4747-3p, mir-500b miR-500a-5p mir-4749 miR-4749-5p mir-501 miR-501-3p,miR-502-3p,miR-501-5p,miR-500a- mir-4750 miR-4750-5p,miR-4750-3p, 3p, mir-4753 miR-4753-5p,miR-4753-3p, mir-5010 miR-5010-5p,miR-5010-3p, mir-4755 miR-4755-3p,miR-4755-5p, mir-5011 miR-5011-5p,miR-5011-3p, mir-4756 miR-4756-3p,miR-4756-5p, mir-502 miR-502-5p,miR-501-3p,miR-502-3p,miR-500a- mir-4757 miR-4757-3p,miR-4757-5p, 3p, mir-4758 miR-4758-3p,miR-4758-5p, mir-503 miR-503-5p,miR-503-3p, mir-4760 miR-4760-5p,miR-4760-3p, mir-505 miR-505-3p,miR-505-5p, mir-4761 miR-4761-5p,miR-4761-3p, mir-506 miR-506-5p,miR-506-3p, mir-4762 miR-4762-5p,miR-4762-3p, mir-508 miR-508-3p,miR-508-5p, mir-4763 miR-4763-3p,miR-4763-5p, mir-5089 miR-5089-3p,miR-5089-5p, mir-4764 miR-4764-3p,miR-4764-5p, mir-509-1 miR-509-5p,miR-509-3p, mir-4766 miR-4766-5p,miR-4766-3p, mir-509-2 miR-509-5p,miR-509-3p, mir-4768 miR-4768-5p,miR-4768-3p, mir-509-3 miR-509-3-5p,miR-509-3p, mir-4769 miR-4769-3p,miR-4769-5p, mir-511-1 miR-511 mir-4771-1 miR-4771 mir-511-2 miR-511

235 mir-512-1 miR-512-3p,miR-512-5p, mir-548aa- miR-548t-3p,miR-548ap-3p,miR-548h-3p,miR- mir-512-2 miR-512-3p,miR-512-5p, 1 548aq-3p,miR-548aa,miR-548z, mir-513a-1 miR-513a-3p,miR-513b,miR-513c-3p mir-548aa- miR-548t-3p,miR-548ap-3p,miR-548h-3p,miR- mir-513a-2 miR-513a-3p,miR-513b,miR-513c-3p 2 548aq-3p,miR-548aa,miR-548z, mir-513c miR-513c-5p,miR-513a-3p,miR-513c-3p mir-548ab miR-548i,miR-548ab,miR-548ay-5p,miR-548d- mir-514a-1 miR-514a-5p,miR-514a-3p, 5p,miR-548n mir-514a-2 miR-514a-5p,miR-514a-3p, mir-548ad miR-548aj-5p,miR-548ay-5p,miR-548x-5p,miR- mir-514a-3 miR-514a-5p,miR-514a-3p, 548d-5p,miR-548g-5p,miR-548az-5p,miR- mir-514b miR-514b-3p,miR-514b-5p, 548ad,miR-548n, mir-515-1 miR-515-5p,miR-515-3p,miR-519e-5p mir-548ae- miR-548x-3p,miR-548ae,miR-548aq-3p 1 mir-515-2 miR-515-5p,miR-515-3p,miR-519e-5p mir-548ag- miR-548ag mir-516a-1 miR-520f,miR-516a-5p,miR-520c-3p,miR-526b- 1 3p,miR-520b mir-548ag- miR-548ag mir-516a-2 miR-520f,miR-516a-5p,miR-526b-3p,miR- 2 520b,miR-520c-3p mir-548ah miR-548av-3p,miR-548ah-5p,miR-548p,miR- mir-516b-1 miR-520f,miR-526b-3p,miR-520b,miR-516b- 548aq-3p,miR-548am-3p,miR-548o-3p,miR- 5p,miR-520c-3p 548ah-3p,miR-548au-3p, mir-516b-2 miR-520f,miR-526b-3p,miR-520b,miR-516b- mir-548ai miR-548aj-5p,miR-548ai,miR-548t-3p,miR- 5p,miR-520c-3p 548aa,miR-570-5p,miR-548x-5p,miR-548g-5p mir-517a miR-517-5p,miR-517b-3p,miR-517c-3p,miR- mir-548aj-1 miR-548x-3p,miR-548aq-3p,miR-548aj-3p 517a-3p, mir-548an miR-548h-3p,miR-548an,miR-548z,miR-548ay- mir-517b miR-517b-3p,miR-517c-3p,miR-517-5p,miR- 5p,miR-548d-5p 517a-3p, mir-548ao miR-548ao-3p,miR-548ao-5p, mir-517c miR-517-5p,miR-517b-3p,miR-517c-3p,miR- 517a-3p, mir-548ap miR-548j,miR-548ap-5p,miR-548t-3p,miR-548ap- 3p,miR-548aa mir-5187 miR-5187-3p,miR-5187-5p, mir-548ar miR-548ay-5p,miR-548d-5p,miR-548ar-3p,miR- mir-518a-1 miR-518a-3p,miR-527,miR-518a-5p 548o-5p,miR-548au-5p,miR-548c-5p,miR-548ar- mir-518a-2 miR-518a-3p,miR-527,miR-518a-5p 5p,miR-548am-5p, mir-518b miR-519c-5p,miR-522-5p,miR-519a-5p,miR- mir-548as miR-548au-5p,miR-548i,miR-548c-5p,miR-548as- 520a-5p,miR-518b,miR-523-5p,miR-519b- 5p,miR-548as-3p,miR-548o-5p,miR-548am-5p 5p,miR-518e-5p, mir-548at miR-548at-3p,miR-548at-5p, mir-518c miR-520c-5p,miR-526a,miR-518d-5p,miR-518f- mir-548av miR-548am-3p,miR-548o-3p,miR-548ah-3p 5p,miR-518c-3p,miR-526b-5p,miR-518c-5p,miR- mir-548ax miR-548g-3p,miR-548h-3p,miR-548ax,miR-548z, 518f-3p, mir-5195 miR-5195-5p,miR-5195-3p, mir-548az miR-548ay-5p,miR-548d-5p,miR-548t-5p,miR- 548aq-3p,miR-548az-3p,miR-548az-5p, mir-5196 miR-5196-3p,miR-5196-5p, mir-548b miR-548ay-5p,miR-548d-5p,miR-548b-3p,miR- mir-5197 miR-5197-3p,miR-5197-5p, 548b-5p, mir-519a-2 miR-519c-5p,miR-522-5p,miR-519a-5p,miR- mir-548f-1 miR-548aj-5p,miR-548x-5p,miR-548f,miR-548g- 519e-3p,miR-519a-3p,miR-519b-3p,miR-523- 3p,miR-548g-5p,miR-548aq-3p, 5p,miR-519b-5p,miR-518e-5p mir-548f-2 miR-548t-3p,miR-548f,miR-548g-3p,miR-548aq- mir-519d miR-519e-3p,miR-520g,miR-519d,miR-520h, 3p,miR-548aa mir-519e miR-519e-3p,miR-515-5p,miR-519a-3p,miR- mir-548f-3 miR-548t-3p,miR-548f,miR-548g-3p,miR-548aq- 519d,miR-519e-5p 3p,miR-548aa mir-520a miR-520a-5p,miR-520a-3p, mir-548f-4 miR-548aq-3p,miR-548ay-5p,miR-548f,miR- mir-520d miR-520d-5p,miR-520d-3p,miR-524-5p 548d-5p,miR-548g-3p mir-520e miR-520f,miR-520e,miR-526b-3p,miR-520b,miR- mir-548f-5 miR-548f,miR-548g-3p, 520c-3p mir-548h-1 miR-548h-5p mir-520g miR-519d,miR-520g,miR-520h mir-548h-2 miR-548h-5p,miR-548t-3p,miR-548aa mir-520h miR-520g,miR-519d,miR-520h mir-548h-4 miR-548h-5p,miR-548h-3p,miR-548aq-3p,miR- mir-521-1 miR-521 548t-3p,miR-548aa,miR-548z, mir-521-2 miR-521 mir-548h-5 miR-548h-5p mir-524 miR-524-5p,miR-524-3p, mir-548i-1 miR-548i,miR-548k,miR-548o-5p,miR-548au- mir-525 miR-525-3p,miR-525-5p, 5p,miR-548c-5p,miR-548ab,miR-548as-5p,miR- mir-526b miR-520f,miR-520c-5p,miR-526a,miR-518d- 548n,miR-548am-5p 5p,miR-518f-5p,miR-526b-3p,miR-520b,miR- mir-548i-2 miR-548i,miR-548k,miR-548o-5p,miR-548au- 526b-5p,miR-520c-3p 5p,miR-548c-5p,miR-548ab,miR-548as-5p,miR- mir-527 miR-520a-3p,miR-527,miR-518a-5p 548n,miR-548am-5p mir-532 miR-532-5p,miR-532-3p, mir-548i-3 miR-548au-5p,miR-548i,miR-548k,miR-548c- mir-539 miR-539-3p,miR-539-5p, 5p,miR-548ab,miR-548as-5p,miR-548o-5p,miR- mir-541 miR-541-5p,miR-541-3p, 548n,miR-548am-5p mir-542 miR-542-5p,miR-542-3p, mir-548i-4 miR-548i,miR-548k,miR-548o-5p,miR-548au- mir-545 miR-545-3p,miR-545-5p, 5p,miR-548c-5p,miR-548ab,miR-548as-5p,miR- mir-548a-1 miR-548aj-5p,miR-548a-3p,miR-548ay-5p,miR- 548n,miR-548am-5p 548x-5p,miR-548d-5p,miR-548g-5p,miR-548ar- mir-548k miR-548h-3p,miR-548i,miR-548k,miR-548z, 5p mir-548n miR-548h-3p,miR-548ab,miR-548z,miR-548n, mir-548a-2 miR-548a-3p mir-548o miR-548av-3p,miR-548am-3p,miR-548o-3p,miR- mir-548a-3 miR-548a-3p,miR-548a-5p, 548ah-3p,

236 mir-548p miR-548av-3p,miR-548p,miR-548aq-3p,miR- mir-6089-2 miR-6089 548am-3p,miR-548o-3p,miR-548ah-3p, mir-615 miR-615-5p,miR-615-3p, mir-548q miR-548o-5p,miR-548au-5p,miR-548c-5p,miR- mir-616 miR-616-5p,miR-616-3p, 548q,miR-548am-5p mir-624 miR-624-5p,miR-624-3p, mir-548s miR-548aj-5p,miR-548s,miR-548x-5p,miR-548ar- mir-625 miR-625-5p,miR-625-3p, 3p,miR-548g-5p,miR-548ar-5p,miR-548o-3p mir-628 miR-628-3p,miR-628-5p, mir-548t miR-548ap-3p,miR-548h-3p,miR-548aq-3p,miR- mir-629 miR-629-3p,miR-629-5p, 548t-3p,miR-548aa,miR-548z,miR-548az-5p,miR- mir-642a miR-642a-3p,miR-642b-5p,miR-642a-5p,miR- 548t-5p, 642b-3p, mir-548u miR-548aq-3p,miR-548u, mir-642b miR-642a-3p,miR-642b-5p,miR-642a-5p,miR- mir-548w miR-548au-5p,miR-548w,miR-548t-3p,miR-548c- 642b-3p, 5p,miR-548aa,miR-548ak,miR-548o-5p,miR- mir-6499 miR-6499-5p,miR-6499-3p, 548am-5p, mir-6500 miR-6500-3p,miR-6500-5p, mir-548x-2 miR-548ab,miR-548ay-5p,miR-548aj-3p,miR- mir-6501 miR-6501-3p,miR-6501-5p, 548x-3p,miR-548d-5p,miR-548ae,miR-548n mir-6502 miR-6502-5p,miR-6502-3p, mir-548y miR-548h-3p,miR-548t-3p,miR-548y,miR- mir-6503 miR-6503-5p,miR-6503-3p, 548aa,miR-548z mir-6504 miR-6504-3p,miR-6504-5p, mir-548z miR-548t-3p,miR-548h-3p,miR-548aq-3p,miR- 548aa,miR-548z mir-6505 miR-6505-3p,miR-6505-5p, mir-550a-1 miR-550a-5p,miR-550a-3p,miR-550a-3-5p mir-6506 miR-6506-5p,miR-6506-3p, mir-550a-2 miR-550a-3p,miR-550a-5p,miR-550a-3-5p mir-6507 miR-6507-3p,miR-6507-5p, mir-550a-3 miR-550a-3p,miR-550a-5p,miR-550a-3-5p mir-6508 miR-6508-3p,miR-6508-5p, mir-550b-1 miR-550b-3p,miR-550b-2-5p, mir-6509 miR-6509-5p,miR-6509-3p, mir-550b-2 miR-550b-3p,miR-550b-2-5p, mir-6510 miR-6510-3p,miR-6510-5p, mir-551b miR-551b-3p,miR-551b-5p, mir-6511a- miR-6511a-3p,miR-6511b-3p,miR-6511a-5p,miR- 1 6511b-5p, mir-556 miR-556-3p,miR-556-5p, mir-6511a- miR-6511a-3p,miR-6511b-3p,miR-6511a-5p,miR- mir-5571 miR-5571-5p,miR-5571-3p, 2 6511b-5p, mir-5579 miR-5579-3p,miR-5579-5p, mir-6511a- miR-6511a-3p,miR-6511b-3p,miR-6511a-5p,miR- mir-5580 miR-5580-3p,miR-5580-5p, 3 6511b-5p, mir-5581 miR-5581-5p,miR-5581-3p, mir-6511a- miR-6511a-3p,miR-6511b-3p,miR-6511a-5p,miR- mir-5582 miR-5582-5p,miR-5582-3p, 4 6511b-5p, mir-5583-1 miR-5583-5p,miR-5583-3p, mir-6511b- miR-6511a-3p,miR-6511b-3p,miR-6511a-5p,miR- mir-5583-2 miR-5583-5p,miR-5583-3p, 1 6511b-5p, mir-5584 miR-5584-5p,miR-5584-3p, mir-6512 miR-6512-3p,miR-6512-5p, mir-5585 miR-5585-5p,miR-5585-3p, mir-6513 miR-6513-3p,miR-6513-5p, mir-5586 miR-5586-5p,miR-5586-3p, mir-6514 miR-6514-3p,miR-6514-5p, mir-5587 miR-5587-3p mir-6515 miR-6515-5p mir-5588 miR-1234-5p,miR-5588-5p, mir-652 miR-652-3p,miR-652-5p, mir-5589 miR-5589-3p,miR-5589-5p, mir-654 miR-654-5p,miR-654-3p, mir-5590 miR-5590-5p,miR-5590-3p, mir-659 miR-659-5p,miR-659-3p, mir-5591 miR-5591-3p,miR-5591-5p, mir-660 miR-660-3p,miR-660-5p, mir-561 miR-561-5p,miR-561-3p, mir-664a miR-664a-3p,miR-664a-5p, mir-5681a miR-5681b,miR-5681a, mir-664b miR-664b-5p,miR-664b-3p, mir-5692a- miR-5692a mir-671 miR-671-3p,miR-671-5p, 1 mir-6715a miR-6715a-3p mir-5692a- miR-5692a mir-6715b miR-6715b-3p,miR-6715b-5p, 2 mir-6716 miR-6716-3p,miR-6716-5p, mir-5692c- miR-5692c,miR-5692b, mir-6717 miR-6717-5p 1 mir-6718 miR-6718-5p mir-5692c- miR-5692b,miR-5692c, mir-6719 miR-6719-3p 2 mir-6720 miR-6720-3p mir-570 miR-548aj-5p,miR-548ai,miR-570-5p,miR-548x- 5p,miR-570-3p,miR-548g-5p, mir-6721 miR-6721-5p mir-5701-1 miR-5701 mir-6722 miR-6722-5p,miR-6722-3p, mir-5701-2 miR-5701 mir-6723 miR-6723-5p mir-5708 miR-1972,miR-5708, mir-6724 miR-6724-5p mir-574 miR-574-3p,miR-574-5p, mir-675 miR-675-5p,miR-675-3p, mir-576 miR-576-3p,miR-576-5p, mir-676 miR-676-3p,miR-676-5p, mir-582 miR-582-3p,miR-582-5p, mir-7-1 miR-7-1-3p,miR-7-5p, mir-584 miR-584-3p,miR-584-5p, mir-7-2 miR-7-5p,miR-7-2-3p, mir-589 miR-589-3p,miR-589-5p, mir-7-3 miR-7-5p mir-590 miR-590-5p,miR-590-3p, mir-708 miR-708-3p,miR-708-5p, mir-593 miR-593-3p,miR-593-5p, mir-744 miR-744-3p,miR-744-5p, mir-6077-1 miR-6077 mir-758 miR-758-3p,miR-758-5p, mir-6077-2 miR-6077 mir-766 miR-766-5p,miR-766-3p, mir-6089-1 miR-6089 mir-767 miR-767-5p,miR-767-3p, mir-769 miR-769-5p,miR-769-3p,

237 mir-770 miR-770-5p mir-92a-2 miR-92a-3p,miR-92a-2-5p, mir-873 miR-873-3p,miR-873-5p, mir-92b miR-92b-3p,miR-92b-5p, mir-875 miR-875-3p,miR-875-5p, mir-93 miR-93-5p,miR-93-3p, mir-876 miR-876-5p,miR-876-3p, mir-937 miR-937-5p,miR-937-3p, mir-877 miR-877-3p,miR-877-5p, mir-939 miR-939-3p,miR-939-5p, mir-885 miR-885-5p,miR-885-3p, mir-941-1 miR-941 mir-888 miR-888-5p,miR-888-3p, mir-941-2 miR-941 mir-892a miR-892c-5p,miR-892a, mir-941-3 miR-941 mir-892c miR-892c-3p,miR-892c-5p, mir-941-4 miR-941 mir-9-1 miR-9-3p,miR-9-5p, mir-96 miR-96-3p,miR-96-5p mir-9-2 miR-9-3p,miR-9-5p, mir-98 miR-98-5p,miR-98-3p mir-9-3 miR-9-3p,miR-9-5p, mir-99a miR-99a-3p,miR-99a-5p,miR-100-5p mir-92a-1 miR-92a-1-5p,miR-92a-3p, mir-99b miR-99b-5p,miR-99b-3p

Appendix 3.4: Expression of miRNAs that are dynamically expressed in HeLa and MCF7

MCF7 HeLa miRNAs G1G0 phase S phase G2M phase S phase G2M phase G1G0 phase hsa-mir-760-3p 12.06232169 28.79623605 35.733373 12.76459039 7.19708252 27.45402352 hsa-mir-6127-3p 3.454460851 25.28317998 19.35772214 12.76459039 6.063014107 19.19107957 hsa-mir-579-3p 7.365254973 12.47976575 19.35772214 2.408168808 13.14185089 19.19107957 hsa-mir-5095-3p 5.555220935 12.47976575 12.37378789 479.3479974 174.9908388 161.7271747 hsa-mir-503-5p 144.5352125 346.4061392 232.2780811 27.75186474 61.85991073 21.86980984 hsa-mir-491-5p 9.808235237 19.25054078 0 16.49929082 6.901533728 3.973776348 hsa-mir-4731-3p 26.56664015 12.47976575 15.74945912 16.49929082 1.775285613 6.588626413 hsa-mir-451a-5p 464.3984812 1160.405289 776.9126178 435.389122 8463.374391 539.6355981 hsa-mir-4485 12.06232169 16.96932302 28.68844341 2959.069365 722.6413432 500.9612173 hsa-mir-4472-2-5p 17.02352497 6.901533728 17.57739354 5.925868684 5.414978892 10.87801184 hsa-mir-4461-3p 17.02352497 28.79623605 54.64718167 38.9366264 18.20469715 16.96932302 hsa-mir-4319-5p 15.65182036 13.85899936 7.235712428 32.28203593 17.24268554 14.99714146 hsa-mir-4275-5p 7.365254973 16.96932302 22.90389448 5.925868684 12.21531703 35.84572274 hsa-mir-4263-3p 331.1870844 86.11761011 166.3979328 12.76459039 3.520423995 25.28317998 hsa-mir-4254-3p 5.555220935 15.13217853 10.73942422 16.49929082 4.896212907 1.775285613 hsa-mir-4251-3p 1.40533462 10.6139014 7.235712428 0 4.051036165 10.87801184 hsa-mir-421-3p 1902.913564 792.5027497 1160.405289 1204.46625 504.4351209 1084.376865 hsa-mir-3975-3p 7.365254973 55.12021658 133.6559941 3364.683276 1587.056643 29.27167894 hsa-mir-3940-3p 1.40533462 10.6139014 19.35772214 182.5367191 90.56341406 222.9296522 hsa-mir-3908-5p 672.1391295 742.0782567 1397.768541 32.28203593 9.502153338 14.99714146 hsa-mir-3619-5p 21.63629357 12.47976575 38.46508822 56.73645198 10.87801184 19.19107957 hsa-mir-3609-3p 21.63629357 12.47976575 26.91177318 683.6070318 144.5352125 149.3497026 hsa-mir-3607-5p 454.9515476 454.9515476 2238.256233 3443.931007 1033.856472 776.9126178 hsa-mir-338-3p 34.95775308 22.82050905 15.74945912 78.49640148 208.1918988 162.9516609 hsa-mir-331-5p 75.83474817 40.45220884 85.35884686 27.75186474 16.15418281 6.588626413 hsa-mir-3200-3p 17.02352497 20.6677871 9.234176338 12.76459039 5.414978892 8.946392223 hsa-mir-3196-5p 121.8498074 36.26972128 216.8646298 2.408168808 5.414978892 45.64292002 hsa-mir-3195-5p 184.6345363 42.85439231 101.1881681 9.436190194 3.520423995 12.76459039 hsa-mir-3195-3p 306.4945739 156.4369552 358.2693807 0 3.520423995 21.86980984 hsa-mir-3159-3p 7.365254973 21.86980984 26.91177318 890.5533074 316.8402022 423.525475 hsa-mir-3149-3p 13.94242081 5.19250601 17.57739354 63.99197567 24.91730951 29.27167894 hsa-mir-222-5p 3.454460851 10.6139014 5.925868684 27.75186474 24.91730951 61.6249153 hsa-mir-211-3p 3.454460851 8.691102594 15.74945912 20.02993942 8.374487244 3.973776348 hsa-mir-210-5p 13.94242081 5.19250601 9.234176338 5.925868684 7.19708252 16.96932302 hsa-mir-20a-3p 12.06232169 27.36461335 24.77847397 12.76459039 75.83474817 82.95894747 hsa-mir-192-3p 3.454460851 13.85899936 13.94242081 2.408168808 4.051036165 16.96932302 hsa-mir-190a-3p 1.40533462 19.25054078 17.57739354 5.925868684 9.502153338 16.96932302 hsa-mir-185-5p 1778.668825 2387.155083 629.5079415 49.26349562 78.27681935 109.3698654 hsa-mir-15b-3p 88.85183792 427.0150708 341.4195047 85.73822848 326.0837354 365.8906818 hsa-mir-15a-3p 52.80033202 173.8690146 195.7820947 23.83758649 57.93701336 48.89083798 hsa-mir-1468-3p 12.06232169 5.19250601 7.235712428 121.8498074 22.97406628 35.84572274 hsa-mir-143-3p 3.454460851 10.6139014 9.234176338 60.68458879 1058.414274 45.64292002 hsa-mir-1304-3p 7.365254973 12.47976575 2.855975236 75.4163265 33.5467733 68.70001877

238 hsa-mir-1302-5-3p 23.99401305 10.6139014 13.94242081 12.76459039 5.414978892 10.87801184 hsa-mir-1291-5p 12.06232169 10.6139014 5.925868684 0 3.520423995 12.76459039 hsa-mir-1256-5p 13.94242081 23.99401305 7.235712428 2.408168808 7.19708252 31.17931184 hsa-mir-1248-5p 21.63629357 49.67558037 49.0771668 186.3711373 8.85043643 14.99714146 hsa-mir-1247-5p 45.04465007 86.11761011 96.68933084 116.394709 29.07284912 21.86980984 hsa-mir-1246-5p 13.94242081 19.25054078 32.42511018 226.398542 15.05491872 35.84572274 hsa-mir-1234-5p 6576.023125 2699.734622 5160.096806 1186.640347 497.4873137 1033.856472 hsa-mir-1229-3p 21.63629357 6.901533728 22.90389448 16.49929082 5.414978892 3.973776348 hsa-mir-1180-3p 5.555220935 22.82050905 22.90389448 153.3726355 51.4622954 57.82377913 hsa-mir-106b-3p 273.2439111 140.4574015 301.6864157 589.3979019 275.6774351 381.1021305 hsa-let-7d-5p 1445.121007 2323.808068 575.3725301 85.73822848 154.3847464 191.9831302 hsa-let-7d-3p 205.7761542 216.8646298 82.52143755 70.86429081 12.21531703 48.89083798

Appendix 3.5: (A) miRNAs previously shown to be cell cycle regulators with cyclic expression across cell cycle phases

HeLa S Phase G2M Phase G1G0 Phase hsa-mir-145-5p 1482.25 9497.91 1902.91 hsa-mir-15a-5p 101.98 229.10 328.63 hsa-mir-195-5p 80.23 197.63 339.01 hsa-mir-331-3p 5597.73 2184.50 2534.57 hsa-mir-424-5p 116.39 672.13 572.46 MCF7 S Phase G2M Phase G1G0 Phase hsa-mir-200b-5p 65.44 127.81 141.45 hsa-mir-503-5p 346.40 232.27 144.53

(B) miRNAs previously shown to be cell cycle regulators but not cyclic in this study

miRNAs MCF7_S MCF7_G2M MCF7_G1G0 HeLa_S HeLa_G2M HeLa_G1G0 hsa-mir-17-5p 2604.81 2959.06 3047.68 16229.24 16229.24 17370.86 hsa-mir-31-5p 68.28 80.89 103.20 53836.60 67962.10 67962.10 hsa-mir-210-5p 13.94 5.19 9.23 5.925 7.19 16.96

miRNA Start End MCF7 MCF7 MCF7 HeLa HeLa HeLa Position Position S G2M G1G0 S G2M G1G0 Sum of reads mapping to mature miRNA - 14 37 247.69 331.74 314.87 2941. 2697.3 3197.9 hsa-mir-17-5p 10 0 3 Sum of reads to mapping to isomirs 11-20 32-51 2212.4 3053.52 2860.74 6088. 10187. 9481.7 3 80 75 0 Sum of isomirs shorter than mature 11-15 32-36 1812.2 2341.60 2158.98 2659. 4869.7 4473.3 miRNA (in the 3' end) 1 18 90 2 0 Sum of isomirs longer than mature 11-20 38-51 302.26 514.57 491.84 2482. 3780.4 3629.4 miRNA (in the 3' end) 00 2 0

(D) miRNAs identified to be cyclic in the Zhou et. al (2009)

miRNAs HeLa_S HeLa_G2M HeLa_G1G0 Potential Dynamic across to Cross- cross hybridize? hybridizing miRNA? hsa-mir-19a-5p 0 5.414978892 6.588626413 yes yes hsa-mir-34b-5p 0 1.775285613 3.973776348 yes yes hsa-mir-34c-3p 0 0 0 yes yes hsa-mir-933-5p 2.408168808 4.051036165 6.588626413 hsa-mir-329-3p 0 0 0 yes yes hsa-mir-519d-5p 0 0 0 yes yes hsa-mir-590-3p 2.408168808 29.07284912 61.6249153 hsa-mir-618-5p 0 0 0 hsa-mir-126-5p 32.28203593 563.2081695 123.698712

239 hsa-mir-147b-5p 0 1.775285613 1.775285613 yes yes hsa-mir-224-5p 2815.209335 2869.838662 2815.209335 hsa-mir-299-5p 0 57.15177807 10.87801184 hsa-mir-524-5p 0 0 0 yes yes hsa-mir-553-5p 0 0 0 hsa-mir-582-5p 536.5001919 1832.812821 2184.505937 hsa-mir-653-5p 0 0 0 yes yes has-mir-924-5p 0 0 0 hsa-mir-34a-5p 1046.135373 1445.121007 1778.668825 hsa-mir-30b-5p 4660.366383 4979.755206 3443.931007 yes yes hsa-mir-22-5p 273.2439111 200.6373135 219.6043546 hsa-mir-510-5p 0 0 0 yes no has-let-7a 711.0180338 1102.727249 913.7149604 yes yes hsa-mir-21-5p 23844.11968 92303.10664 92303.10664 hsa-mir-221-3p 968.038206 698.8089319 609.5658477 hsa-mir-222-3p 1397.768541 905.7753959 1058.414274

Appendix 3.6: List of miRNA probes and their corresponding potential cross-hybridizing miRNAs

miRNAs Cross-hybridizing miRNAs* 5p,miR-524-5p,miR-519b-5p,miR-518e-5p, miR-453 miR-323a-5p,miR-323b-5p, miR-513 miR-513a-5p,miR-513c-5p,miR-513b, miR-376b miR-376b-3p,miR-376a-3p,miR-376c-3p, miR-500 miR-501-3p,miR-502-3p,miR-500a-3p, miR-146a miR-146a-5p,miR-146b-5p, miR-518e miR-520f,miR-519e-3p,miR-523-3p,miR-521,miR- miR-582 miR-582-5p, 525-3p,miR-526b-3p,miR-518e-3p,miR-518c- miR-422b miR-378j,miR-378a-3p,miR-378d,miR-6128,miR- 3p,miR-518b,miR-3158-5p,miR-518f-3p,miR- 378i,miR-378e,miR-378b,miR-422a,miR- 518d-3p,miR-524-3p, 378g,miR-378f,miR-378h,miR-378c, miR-515-5p miR-520d-5p,miR-520c-5p,miR-526a,miR-520a- miR- miR-519d,miR-520f,miR-520e,miR-519c-3p,miR- 5p,miR-515-5p,miR-518d-5p,miR-518f-5p,miR- 520f_miR- 519b-3p,miR-519e-3p,miR-519a-3p,miR-520c- 1283,miR-519e-5p,miR-525-5p,miR-527,miR- 520c 3p,miR-516a-3p,miR-520g,miR-516b-3p,miR- 518a-5p,miR-524-5p, 520h,miR-526b-3p,miR-520b, miR-301 miR-301b,miR-301a-3p, miR-376a* miR-376c-5p,miR-376a-5p,miR-376a-2-5p, miR-144 miR-144-3p, miR-510 miR-514a-5p,miR-510, miR-125b miR-4324,miR-125b-5p, miR-200b miR-200a-3p,miR-141-3p,miR-200b-3p,miR-200c- miR-338 miR-338-3p, 3p, miR-187 miR-187-3p, miR-30e-3p miR-30d-3p,miR-30e-3p,miR-30a-3p, miR-384 miR-384, miR-628 miR-628-3p, miR-574 miR-574-3p, miR-520b miR-520f,miR-520e,miR-519e-3p,miR-520a- miR-653 miR-24-2-5p,miR-4684-5p,miR-653, 3p,miR-519a-3p,miR-520g,miR-520d-3p,miR-523- miR-302a miR-520e,miR-302b-3p,miR-302c-3p,miR-302d- 3p,miR-525-3p,miR-526b-3p,miR-518e-3p,miR- 3p,miR-302a-3p, 520b,miR-518c-3p,miR-519d,miR-522-3p,miR- miR-491 miR-491-5p, 519c-3p,miR-518b,miR-519b-3p,miR-517c- miR-483 miR-483-3p, 3p,miR-520c-3p,miR-518f-3p,miR-518d-3p,miR- miR-625 miR-625-5p, 516a-3p,miR-516b-3p,miR-520h,miR-524-3p, miR-138 miR-138-5p, miR-147 miR-147b,miR-147a, miR-126 miR-126-3p, miR-32 miR-32-5p, miR-429 miR-429,miR-141-3p, miR-642 miR-642b-5p,miR-642a-5p, let-7d let-7a-5p,miR-98-5p,let-7c,let-7b-5p,let-7f-5p,let- miR-519c miR-520f,miR-520e,miR-519e-3p,miR-519a- 7e-5p,let-7d-5p,let-7g-5p, 3p,miR-520g,miR-520d-3p,miR-526b-3p,miR- miR-503 miR-503-5p, 520b,miR-519d,miR-518c-3p,miR-519c-3p,miR- miR-29a miR-29b-3p,miR-29a-3p,miR-29c-3p, 519b-3p,miR-517c-3p,miR-520c-3p,miR-518f- miR-488 miR-488-5p, 3p,miR-520h, miR-766 miR-766-3p,miR-4323, miR-16 miR-16-5p,miR-195-5p, miR-192 miR-192-5p,miR-215, miR-18b miR-18b-5p,miR-18a-5p,miR-20a-5p, miR-570 miR-548av-3p,miR-548t-3p,miR-548ac,miR-548g- miR-17-5p miR-20b-5p,miR-106b-5p,miR-20a-5p,miR-106a- 3p,miR-570-3p,miR-548h-3p,miR-548aq-3p,miR- 5p,miR-17-5p, 548aa,miR-548z,miR-548aj-3p,miR-548az-3p,miR- miR-548d miR-548at-3p,miR-548av-3p,miR-548t-3p,miR- 548ad, 548c-3p,miR-548ap-3p,miR-548h-3p,miR-548d- miR-105 miR-105-5p, 3p,miR-548aq-3p,miR-548am-3p,miR-548b- 3p,miR-548aa,miR-548o-3p,miR-548ah-3p,miR- let-7g let-7a-5p,miR-98-5p,let-7c,let-7i-5p,let-7b-5p,let- 548z, 7f-5p,let-7e-5p,let-7d-5p,let-7g-5p, miR-525 miR-520d-5p,miR-520c-5p,miR-519c-5p,miR-522- miR-9 miR-9-5p, 5p,miR-526a,miR-519a-5p,miR-520a-5p,miR- miR-517* miR-520c-5p,miR-519c-5p,miR-522-5p,miR- 518d-5p,miR-518f-5p,miR-526b-5p,miR-518c- 526a,miR-519a-5p,miR-520a-5p,miR-518d- 5p,miR-523-5p,miR-525-5p,miR-527,miR-518a- 5p,miR-517-5p,miR-518f-5p,miR-526b-5p,miR-

240 518c-5p,miR-523-5p,miR-525-5p,miR-519b- miR-518d miR-520e,miR-520a-3p,miR-520g,miR-520d- 5p,miR-518e-5p, 3p,miR-523-3p,miR-521,miR-525-3p,miR-526b- miR-133a miR-133a,miR-133b, 3p,miR-518e-3p,miR-520b,miR-519d,miR-518c- miR-155 miR-155-5p, 3p,miR-518a-3p,miR-518b,miR-520c-3p,miR-518f- miR-520g miR-520f,miR-520e,miR-519e-3p,miR-520a- 3p,miR-518d-3p,miR-520h,miR-524-3p, 3p,miR-519a-3p,miR-520g,miR-520d-3p,miR-523- miR-139 miR-139-5p, 3p,miR-521,miR-525-3p,miR-526b-3p,miR-518e- miR-200a miR-200a-3p,miR-141-3p,miR-200c-3p,miR-200b- 3p,miR-520b,miR-519d,miR-518c-3p,miR-522- 3p, 3p,miR-518b,miR-519c-3p,miR-517b-3p,miR- miR-30d miR-30e-5p,miR-30d-5p,miR-30a-5p, 519b-3p,miR-517c-3p,miR-520c-3p,miR-518f- miR-629 miR-629-3p, 3p,miR-518d-3p,miR-520h,miR-524-3p,miR-517a- miR-93 miR-93-5p, 3p, miR-193b miR-193b-3p, miR-34b miR-34c-5p,miR-34b-5p, miR-371 miR-371a-3p, miR-616 miR-616-5p, miR-27a miR-27b-3p,miR-27a-3p, miR-520d* miR-520d-5p,miR-520c-5p,miR-519c-5p,miR-522- miR-518c* miR-520c-5p,miR-519c-5p,miR-522-5p,miR- 5p,miR-526a,miR-519a-5p,miR-520a-5p,miR-515- 526a,miR-519a-5p,miR-520a-5p,miR-518d- 5p,miR-518d-5p,miR-518f-5p,miR-1283,miR-519e- 5p,miR-518f-5p,miR-526b-5p,miR-518c-5p,miR- 5p,miR-523-5p,miR-525-5p,miR-527,miR-518a- 523-5p,miR-525-5p,miR-519b-5p,miR-518e-5p, 5p,miR-524-5p,miR-519b-5p,miR-518e-5p, miR-590 miR-590-5p, miR-222 miR-222-3p, miR-523 miR-519c-5p,miR-522-5p,miR-519a-5p,miR-523- miR-181a miR-181b-5p,miR-181a-5p, 3p,miR-521,miR-525-3p,miR-526b-3p,miR-518e- miR-432 miR-432-5p, 3p,miR-520b,miR-518c-3p,miR-518b,miR-520c- miR-92 miR-25-3p,miR-92b-3p,miR-92a-3p, 3p,miR-523-5p,miR-518f-3p,miR-518d-3p,miR- miR-593 miR-593-5p, 519b-5p,miR-524-3p,miR-518e-5p, miR-518f* miR-520c-5p,miR-519c-5p,miR-522-5p,miR- let-7i let-7a-5p,miR-98-5p,let-7c,let-7i-5p,let-7b-5p,let- 526a,miR-519a-5p,miR-520a-5p,miR-518d- 7f-5p,let-7g-5p, 5p,miR-517-5p,miR-518f-5p,miR-526b-5p,miR- miR-335 miR-335-5p, 518c-5p,miR-523-5p,miR-525-5p,miR-524-5p,miR- miR-422a miR-378g,miR-378f,miR-378h,miR-378a-3p,miR- 519b-5p,miR-518e-5p, 378d,miR-378i,miR-422a,miR-378c, miR-151 miR-151a-3p, miR-152 miR-148a-3p,miR-152,miR-148b-3p, miR-208 miR-208b,miR-208a, miR-23b miR-23c,miR-23a-3p,miR-23b-3p, miR-302c* miR-302c-5p, miR-363* miR-363-5p, miR-526c miR-520c-5p,miR-519c-5p,miR-522-5p,miR- miR-188 miR-188-5p, 526a,miR-519a-5p,miR-518d-5p,miR-518f-5p,miR- miR-101 miR-101-3p, 526b-5p,miR-518c-5p,miR-523-5p,miR-519b- miR-452 miR-452-5p, 5p,miR-518e-5p, miR-339 miR-339-5p,miR-1233-3p, miR-487a miR-487a,miR-487b, miR-539 miR-2355-3p,miR-539-5p, miR-221 miR-221-3p, miR-28 miR-151a-5p,miR-28-5p, miR-204 miR-211-5p,miR-204-5p, miR-342 miR-342-3p, miR-195 miR-16-5p,miR-195-5p, miR-296 miR-296-5p, miR-487b miR-487a,miR-487b, miR-33 miR-33a-5p,miR-33b-5p, miR-96 miR-96-5p, miR-33b miR-33a-5p,miR-33b-5p, miR-519d miR-520f,miR-520e,miR-519e-3p,miR-520a- miR-519b miR-520f,miR-520e,miR-519e-3p,miR-519a- 3p,miR-519a-3p,miR-520g,miR-521,miR-525- 3p,miR-526b-3p,miR-520b,miR-519d,miR-519c- 3p,miR-526b-3p,miR-518e-3p,miR-520b,miR- 3p,miR-519b-3p,miR-517c-3p,miR-520c-3p, 519d,miR-518c-3p,miR-522-3p,miR-518b,miR- miR-31 miR-31-5p, 519c-3p,miR-517b-3p,miR-519b-3p,miR-517c- miR-124a miR-124-3p, 3p,miR-520c-3p,miR-518f-3p,miR-518d-3p,miR- miR-526b miR-520c-5p,miR-519c-5p,miR-522-5p,miR- 520h,miR-517a-3p, 526a,miR-519a-5p,miR-520a-5p,miR-518d- miR-545 miR-421,miR-545-3p, 5p,miR-518f-5p,miR-526b-5p,miR-518c-5p,miR- miR-181d miR-181b-5p,miR-181d, 523-5p,miR-525-5p,miR-519b-5p,miR-518e-5p, miR-135b miR-135a-5p,miR-135b-5p, miR-99a miR-99b-5p,miR-99a-5p,miR-100-5p, miR-214 miR-214-3p, miR-133b miR-133a,miR-133b, miR-514 miR-514b-3p,miR-514a-3p, let-7b let-7a-5p,miR-98-5p,let-7c,let-7i-5p,let-7b-5p,let- miR-377 miR-377-3p, 7f-5p,let-7e-5p,miR-4510,let-7d-5p,let-7g-5p, miR-27b miR-27b-3p,miR-27a-3p, miR-544 miR-544a, miR-548c miR-548av-3p,miR-548t-3p,miR-548a-3p,miR- miR-361 miR-361-5p, 548ae,miR-548c-3p,miR-548ap-3p,miR-548h- miR-520a miR-520f,miR-520e,miR-519e-3p,miR-520a- 3p,miR-548d-3p,miR-548aq-3p,miR-548am- 3p,miR-519a-3p,miR-520g,miR-520d-3p,miR- 3p,miR-548b-3p,miR-548aa,miR-548ah-3p,miR- 526b-3p,miR-518e-3p,miR-520b,miR-519d,miR- 548z,miR-548aj-3p,miR-548az-3p,miR-548u, 522-3p,miR-520c-3p,miR-518d-3p,miR-520h,miR- miR-549 miR-549a, 524-3p, miR-34c miR-34c-5p,miR-34b-5p, miR-143 miR-143-3p, miR-520e miR-520f,miR-520e,miR-519e-3p,miR-520a- miR-199b miR-199b-5p,miR-199a-5p, 3p,miR-519a-3p,miR-520g,miR-520d-3p,miR-525- miR-330 miR-330-3p, 3p,miR-526b-3p,miR-520b,miR-519d,miR-519c- miR-136 miR-136-5p, 3p,miR-519b-3p,miR-373-3p,miR-520c-3p,miR- miR-551a miR-551a,miR-551b-3p, 518f-3p,miR-518d-3p,miR-520h,miR-524-3p,

241 miR-615 miR-615-3p, 603,miR-548az-3p,miR-548u, miR-320 miR-320a,miR-320c,miR-4429,miR-320e,miR- miR-527 miR-520d-5p,miR-520c-5p,miR-519c-5p,miR-522- 320d,miR-320b, 5p,miR-526a,miR-519a-5p,miR-520a-5p,miR- miR-449 miR-449a,miR-449c-5p,miR-449b-5p, 518d-5p,miR-518f-5p,miR-1283,miR-523-5p,miR- miR-223 miR-223-3p, 525-5p,miR-527,miR-518a-5p,miR-524-5p,miR- miR-141 miR-200a-3p,miR-429,miR-141-3p,miR-200c- 519b-5p,miR-518e-5p, 3p,miR-200b-3p, miR-185 miR-4306,miR-185-5p, miR-508 miR-508-3p, miR-423 miR-423-3p, let-7a let-7a-5p,miR-98-5p,let-7c,miR-6129,let-7i-5p,let- miR-20b miR-93-5p,miR-20b-5p,miR-106b-5p,miR-20a- 7b-5p,miR-6134,let-7f-5p,let-7e-5p,miR-4510,let- 5p,miR-106a-5p,miR-17-5p, 7d-5p,let-7g-5p, miR-631 miR-631,miR-3913-3p, miR-576 miR-576-5p, miR-331 miR-331-3p, miR-378 miR-378a-5p, miR-106a miR-20b-5p,miR-106b-5p,miR-20a-5p,miR-106a- miR-518c miR-520f,miR-519e-3p,miR-520a-3p,miR-519a- 5p,miR-17-5p, 3p,miR-520g,miR-520d-3p,miR-523-3p,miR- miR-519a miR-520f,miR-519e-3p,miR-519a-3p,miR-526b- 521,miR-525-3p,miR-526b-3p,miR-518e-3p,miR- 3p,miR-520b,miR-519c-3p,miR-517b-3p,miR- 520b,miR-519d,miR-518c-3p,miR-522-3p,miR- 519b-3p,miR-517c-3p,miR-520c-3p,miR-517a-3p, 518b,miR-519c-3p,miR-520c-3p,miR-518f-3p,miR- miR-199a* miR-199b-3p,miR-199a-3p, 518d-3p,miR-520h,miR-524-3p, miR-18a* miR-18a-3p,miR-18b-3p, miR-526b* miR-520f,miR-520e,miR-519e-3p,miR-520a- miR-149 miR-149-5p, 3p,miR-519a-3p,miR-520g,miR-520d-3p,miR-525- miR-516-5p miR-516a-5p,miR-516b-5p, 3p,miR-526b-3p,miR-518e-3p,miR-520b,miR- miR-92b miR-25-3p,miR-92b-3p,miR-92a-3p, 519d,miR-518c-3p,miR-522-3p,miR-518b,miR- miR-130b miR-130a-3p,miR-130b-3p, 519c-3p,miR-519b-3p,miR-517c-3p,miR-520c- miR-486 miR-486-5p, 3p,miR-518f-3p,miR-518d-3p,miR-520h, miR-542-5p miR-542-5p, miR-190 miR-190a, miR-532 miR-532-5p, miR-98 let-7a-5p,miR-98-5p,let-7c,let-7i-5p,let-7b-5p,let- miR-663 miR-663a, 7f-5p,let-7e-5p,miR-4510,let-7d-5p,let-7g-5p, miR-99b miR-99b-5p,miR-99a-5p,miR-100-5p, miR-203 miR-203b-3p,miR-203a, miR-519e miR-520f,miR-520e,miR-519e-3p,miR-520a- miR-302a* miR-302a-5p, 3p,miR-519a-3p,miR-520g,miR-526b-3p,miR-515- miR-561 miR-561-3p, 3p,miR-518e-3p,miR-520b,miR-519d,miR-522- miR-191 miR-191-5p, 3p,miR-519c-3p,miR-517b-3p,miR-519b-3p,miR- miR-381 miR-300,miR-381-3p, 517c-3p,miR-520c-3p,miR-520h,miR-517a-3p, miR-127 miR-127-3p, miR-550 miR-550b-3p,miR-550a-3p, miR-519e* miR-520d-5p,miR-520c-5p,miR-526a,miR-520a- miR-520a* miR-520d-5p,miR-520c-5p,miR-519c-5p,miR-522- 5p,miR-515-5p,miR-518d-5p,miR-518f-5p,miR- 5p,miR-526a,miR-519a-5p,miR-520a-5p,miR- 519e-5p,miR-525-5p,miR-527,miR-518a-5p,miR- 518d-5p,miR-518f-5p,miR-526b-5p,miR-518c- 524-5p, 5p,miR-523-5p,miR-525-5p,miR-527,miR-518a- miR-211 miR-211-5p,miR-204-5p, 5p,miR-524-5p,miR-519b-5p,miR-518e-5p, miR-181a* miR-181a-3p,miR-181c-3p,miR-181a-2-3p, let-7e let-7a-5p,miR-98-5p,let-7c,let-7b-5p,let-7f-5p,miR- miR-19b miR-19b-3p,miR-19a-3p, 6133,let-7e-5p,miR-4510,let-7d-5p,let-7g-5p, miR-432* miR-432-3p, miR-499 miR-499b-5p,miR-499a-5p, miR-380-3p miR-379-3p,miR-380-3p, miR-30a-5p miR-30e-5p,miR-30d-5p,miR-30a-5p, miR-194 miR-194-5p, miR-520h miR-520f,miR-520e,miR-519e-3p,miR-520a- miR-502 miR-502-5p,miR-500b,miR-500a-5p, 3p,miR-519a-3p,miR-520g,miR-520d-3p,miR-523- miR-150 miR-150-5p, 3p,miR-521,miR-525-3p,miR-526b-3p,miR-518e- miR-30c miR-30b-5p,miR-30c-5p, 3p,miR-520b,miR-519d,miR-518c-3p,miR-522- miR-181c miR-181c-5p, 3p,miR-518b,miR-519c-3p,miR-517b-3p,miR- miR-526a miR-520d-5p,miR-520c-5p,miR-519c-5p,miR-522- 519b-3p,miR-517c-3p,miR-520c-3p,miR-518f- 5p,miR-526a,miR-519a-5p,miR-520a-5p,miR- 3p,miR-518d-3p,miR-520h,miR-524-3p,miR-517a- 518d-5p,miR-517-5p,miR-518f-5p,miR-526b- 3p, 5p,miR-518c-5p,miR-523-5p,miR-525-5p,miR- miR-758 miR-758-3p,miR-379-3p, 527,miR-518a-5p,miR-524-5p,miR-519b-5p,miR- miR-216 miR-216a-5p,miR-216b, 518e-5p, miR-548a miR-548av-3p,miR-548t-3p,miR-548a-3p,miR- miR-451 miR-451a, 548ac,miR-548ae,miR-548c-3p,miR-548h-3p,miR- miR-556 miR-556-5p, 548aq-3p,miR-548am-3p,miR-548aa,miR-548ah- miR-374 miR-374c-5p,miR-374a-5p,miR-374b-5p, 3p,miR-548z,miR-548au-3p,miR-548aj-3p,miR- miR-18a miR-18b-5p,miR-18a-5p,miR-20a-5p, 548az-3p,miR-548u, miR-205 miR-205-5p, miR-522 miR-520f,miR-519e-3p,miR-522-3p, miR-224 miR-224-5p, miR-30e-5p miR-30e-5p,miR-30d-5p,miR-30a-5p, miR-769-5p miR-769-5p, miR-10b miR-10b-5p,miR-10a-5p,miR-100-5p, miR-516-3p miR-520f,miR-516a-3p,miR-516b-3p, let-7c let-7a-5p,miR-98-5p,let-7c,let-7i-5p,let-7b-5p,let- 7f-5p,miR-6130,let-7e-5p,miR-4510,let-7d-5p,let- miR-659 miR-659-3p, 7g-5p, miR-603 miR-548av-3p,miR-548t-3p,miR-548f,miR-548x- miR-431 miR-431-3p,miR-431-5p, 3p,miR-548g-3p,miR-548ae,miR-548h-3p,miR- 548aq-3p,miR-548am-3p,miR-548aa,miR-548o- miR-197 miR-197-3p, 3p,miR-548ah-3p,miR-548z,miR-548aj-3p,miR- miR-520d miR-520f,miR-520d-3p,

242 miR-103 miR-103a-3p,miR-4289,miR-107, miR-15a miR-497-5p,miR-15a-5p,miR-15b-5p, miR-202* miR-202-5p, miR-652 miR-652-3p, miR-24 miR-24-3p, miR-323 miR-323b-3p,miR-323a-3p, miR-518f miR-520f,miR-520e,miR-520g,miR-520d-3p,miR- miR-129 miR-129-5p, 523-3p,miR-521,miR-525-3p,miR-526b-3p,miR- miR-193a miR-193a-3p,miR-193b-3p, 518e-3p,miR-520b,miR-519d,miR-518c-3p,miR- miR-125a miR-125a-5p, 518b,miR-519c-3p,miR-519b-3p,miR-520c- miR-128a miR-128, 3p,miR-518f-3p,miR-518d-3p,miR-520h,miR-524- miR-20a miR-93-5p,miR-20b-5p,miR-106b-5p,miR-20a- 3p, 5p,miR-106a-5p,miR-17-5p, miR-100 miR-99b-5p,miR-10a-5p,miR-99a-5p,miR-100-5p, miR-324-3p miR-18a-3p,miR-324-3p, miR-660 miR-660-5p, miR-183 miR-182-5p,miR-183-5p, miR-148a miR-148a-3p,miR-152,miR-148b-3p, miR-1 miR-206,miR-1, miR-379 miR-379-5p, miR-490 miR-490-3p, miR-367 miR-367-3p, miR-376a miR-376b-3p,miR-376a-3p,miR-376c-3p, miR-369-5p miR-369-5p, miR-455 miR-455-5p, miR-7 miR-7-5p, miR-373 miR-520e,miR-373-3p,miR-302d-3p,miR-302a-3p, miR-382 miR-382-5p,miR-323b-5p, miR-200c miR-200a-3p,miR-141-3p,miR-200c-3p,miR-200b- miR-182 miR-182-5p,miR-183-5p, 3p, miR-30b miR-30b-5p,miR-30c-5p, miR-368 miR-376b-3p,miR-376a-3p,miR-376c-3p, miR-219 miR-219-5p, miR-196b miR-196b-5p,miR-196a-5p, miR-345 miR-345-5p, miR-126* miR-126-5p, miR-25 miR-25-3p,miR-92b-3p,miR-92a-3p, miR-363 miR-363-3p, miR-424 miR-424-5p, miR-337 miR-337-3p, miR-186 miR-186-5p, miR-130a miR-130a-3p,miR-130b-3p, miR-506 miR-506-3p, miR-145 miR-145-5p, miR-23a miR-23c,miR-23a-3p,miR-23b-3p, miR-524* miR-520d-5p,miR-520c-5p,miR-519c-5p,miR-522- miR-302b miR-302b-3p,miR-302c-3p,miR-302d-3p,miR- 5p,miR-526a,miR-519a-5p,miR-520a-5p,miR-515- 302a-3p, 5p,miR-518d-5p,miR-518f-5p,miR-1283,miR-519e- miR-181b miR-181b-5p,miR-181a-5p,miR-181d, 5p,miR-526b-5p,miR-523-5p,miR-525-5p,miR- miR-365 miR-365b-3p,miR-365a-3p, 527,miR-518a-5p,miR-524-5p,miR-519b-5p,miR- miR-449b miR-449a,miR-449c-5p,miR-449b-5p, 518e-5p, miR-10a miR-10b-5p,miR-10a-5p,miR-99a-5p,miR-100-5p, miR-148b miR-148a-3p,miR-152,miR-148b-3p, miR-21 miR-21-5p, miR-767-5p miR-767-5p,miR-767-3p, miR-199a miR-199b-5p,miR-199a-5p, miR-340 miR-340-3p, miR-140 miR-140-5p, miR-373* miR-373-5p, miR-302c miR-302b-3p,miR-302c-3p,miR-302d-3p,miR- miR-515-3p miR-519e-3p,miR-515-3p, 302a-3p, miR-26b miR-26b-5p,miR-26a-5p, miR-655 miR-329,miR-655,miR-369-3p, miR-106b miR-20b-5p,miR-106b-5p,miR-20a-5p,miR-106a- miR-654 miR-654-5p, 5p,miR-17-5p, miR-485-5p miR-485-5p,miR-377-5p, miR-518b miR-520f,miR-520e,miR-520g,miR-523-3p,miR- miR-202 miR-202-3p, 521,miR-525-3p,miR-526b-3p,miR-518e-3p,miR- miR-548b miR-548t-3p,miR-548c-3p,miR-548h-3p,miR-548d- 520b,miR-519d,miR-518c-3p,miR-518b,miR-519b- 3p,miR-548am-3p,miR-548b-3p,miR-548aa,miR- 3p,miR-520c-3p,miR-518f-3p,miR-518d-3p,miR- 548ah-3p,miR-548z, 520h,miR-524-3p, miR-501 miR-501-5p,miR-500a-5p, miR-497 miR-497-5p,miR-15a-5p, miR-606 miR-606, let-7f let-7a-5p,miR-98-5p,let-7c,let-7i-5p,miR-1827,let- miR-146b miR-146a-5p,miR-146b-5p, 7b-5p,let-7f-5p,miR-6130,let-7e-5p,miR-4510,let- miR-521 miR-523-3p,miR-521,miR-525-3p,miR-518e- 7d-5p,let-7g-5p, 3p,miR-518c-3p,miR-518b,miR-517b-3p,miR-518f- miR-369-3p miR-154-3p,miR-655,miR-369-3p, 3p,miR-518d-3p,miR-524-3p,miR-517a-3p, miR-524 miR-520a-3p,miR-523-3p,miR-521,miR-525- miR-212 miR-212-3p,miR-132-3p, 3p,miR-518e-3p,miR-518c-3p,miR-518a-3p,miR- miR-30a-3p miR-30d-3p,miR-30e-3p,miR-30a-3p, 518b,miR-518f-3p,miR-518d-3p,miR-524-3p, miR-206 miR-206,miR-1, miR-509 miR-514a-3p,miR-509-3p, miR-329 miR-329,miR-655, miR- miR-519e-3p,miR-519a-3p,miR-517b-3p,miR- 517a_miR- 517c-3p,miR-517a-3p, miR-189 miR-24-2-5p,miR-24-1-5p, 517b miR-196a miR-196b-5p,miR-196a-5p, miR-302d miR-302e,miR-302b-3p,miR-302c-3p,miR- miR-525* miR-520g,miR-523-3p,miR-521,miR-525-3p,miR- 302f,miR-302d-3p,miR-302a-3p, 526b-3p,miR-518e-3p,miR-520b,miR-519d,miR- miR-517c miR-520f,miR-519e-3p,miR-519a-3p,miR-526b- 518c-3p,miR-518b,miR-520c-3p,miR-518f-3p,miR- 3p,miR-520b,miR-519c-3p,miR-517b-3p,miR- 518d-3p,miR-520h,miR-524-3p, 519b-3p,miR-517c-3p,miR-520c-3p,miR-517a-3p, miR-107 miR-103a-3p,miR-4289,miR-107, miR-624 miR-624-5p, miR-19a miR-19b-3p,miR-19a-3p, miR-218 miR-218-5p, miR-128b miR-128, miR-302b* miR-302d-5p,miR-302b-5p, miR-411 miR-411-5p, miR-551b miR-551a,miR-551b-3p, miR-518a miR-520d-3p,miR-518a-3p,miR-518d-3p,miR-524- miR-326 miR-1296, 3p, miR-362 miR-362-5p,

243 miR-15b miR-15a-5p,miR-15b-5p, miR-589 miR-589-3p, miR-505 miR-214-5p,miR-505-3p, miR-215 miR-192-5p,miR-215, miR-154 miR-154-5p,miR-323b-5p, miR-154* miR-154-3p,miR-655,miR-369-3p, miR-662 miR-662,miR-1285-5p, miR-512-5p miR-4287,miR-512-5p, miR-26a miR-26b-5p,miR-26a-5p, miR-34a miR-34a-5p, miR-122a miR-3591-5p,miR-122-5p, miR-671 miR-671-5p, miR-495 miR-495-3p, miR-584 miR-584-5p, miR-200a* miR-200a-5p, miR-135a miR-135a-5p,miR-135b-5p, miR-29c miR-29b-3p,miR-29a-3p,miR-29c-3p, miR-29b miR-29b-3p,miR-29a-3p,miR-29c-3p, miR-132 miR-132-3p,

* miRNAs whose probes do not cross hybridize to other miRNAs have been removed from this list

Appendix 3.7: Oscillating miRNAs in HeLa and MCF7 with upstream transcription factor binding sites for known cell cycle regulators

(A) Phasic miRNAs

miRNA Phasic Phasic Transcription Cell Coordinates of TF Distance Score H3K27ac H3K4me3 HeLa MCF7 Factor Line binding in kb* hsa-let-7d-5p Y Y None hsa-mir-106b-3p Y Y HA-E2F1 HeLa chr7:99698107- 6.5 1000 Y Y 99698808 MCF7 chr7:99698837- 7.2 1000 Y Y 99699404 CEBPB HeLa chr7:99699195- 7.5 98 Y Y 99699466 c-Myc MCF7 chr7:99698072- 6.5 1000 Y Y 99698341 HeLa chr7:99698200- 7 721 Y Y 99698533 ELK4 HeLa chr7:99698207- 6.6 711 Y Y 99698590 E2F4 HeLa chr7:99698322- 6.7 399 Y Y 99698661 E2F6 HeLa chr7:99698363- 6.7 508 Y Y 99698692 E2F1 HeLa hr7:99698358- 6.7 666 Y Y 99698633 ELF1 MCF7 chr7:99697879- 6.8 517 Y Y 99698803 Ini1 HeLa chr7:99698933- 7.3 806 Y Y 99699252 USF2 HeLa chr7:99691270- 0.1 534 Y Y 99691849 chr7:99698962- 7.3 1000 Y Y 99699149 EGR1 MCF7 chr7:99697864- 6.2 133 Y Y 99698720 GATA3 MCF7 chr7:99697898- 6.2 38 Y Y 99698749 FOXM1 MCF7 chr7:99697898- 6.2 65 Y Y 99698749 JUND MCF7 chr7:99697893- 6.2 111 Y Y 99699827 MAX MCF7 chr7:99697805- 6.3 958 Y Y 99699801 SRF MCF7 chr7:99698005- 6.5 79 Y Y 99699829 Mxi1 HeLa chr7:99698205- 6.6 469 Y Y 99698524 HA-E2F1 MCF7 chr8:145634440- 9 314 Y Y 145634623

244 hsa-mir-1234-5p Y Y HeLa chr8:145634420- 9 101 Y Y 145635112 c-Myc MCF7 chr8:145634613- 9.2 773 Y Y 145634882 HeLa chr8:145634462- 9 110 Y Y 145635414 NRF1 HeLa chr8:145634642- 9.2 605 Y Y 145634891 EGR1 MCF7 chr8:145634093- 8.6 125 Y Y 145635025 FOXM1 MCF7 chr8:145634462- 9 39 Y Y 145635455 JUND MCF7 chr8:145634185- 8.8 57 Y Y 145635565 MAX MCF7 chr8:145633899- 8.5 357 Y Y 145635587 HeLa chr8:145634523- 9.1 1000 Y Y 145635011 SRF MCF7 chr8:145634405- 9 36 Y Y 145635556 ELK1 HeLa chr8:145634577- 9.1 1000 Y Y 145634982 INI1 HeLa chr8:145634416- 9 242 Y Y 145635322 GATA3 MCF7 chr8:145634302- 8.9 45 Y Y 145635063 E2F1 HeLa chr3:160118681- 3.5 1000 Y Y 160118956 hsa-mir-15b-3p Y Y HA-E2F1 HeLa chr3:160118655- 3.5 581 Y Y 160118999 MCF7 chr3:160118655- 3.5 563 Y Y 160118999 CEBPB HeLa chr3:160118234- 4.3 462 Y Y 160118489 USF2 HeLa chr3:160119147- 3.3 675 Y Y 160119350 ELK1 HeLa chr3:160118704- 3.7 483 Y Y 160118858 E2F4 HeLa chr3:160116843- 5.1 652 Y Y 160118337 STAT1 HeLa chr3:160116725- 5.2 356 Y Y 160117084 GATA3 MCF7 chr3:160117179- 5.1 528 Y Y 160117490 MAX MCF7 chr3:160117156- 5.3 38 Y Y 160118397 NRF1 HeLa chr3:160117098- 5.1 170 Y Y 160117519 INI1 HeLa chr3:160116804- 4.3 191 Y Y 160119025 FOXM1 MCF7 chr3:160116778- 4.3 37 Y Y 160117727 Mxi1 HeLa chr3:160116855- 4.2 413 Y Y 160117104 SRF MCF7 chr3:160116769- 4.3 31 Y Y 160117767 ZNF143 HeLa chr3:160118184- 4.2 1000 Y Y 160118383 HA-E2F1 MCF7 chr22:20014263- 6.4 461 Y N 20014377 hsa-mir-185-5p Y Y STAT1 HeLa chr5:85913437- 2.8 176 Y Y 85913814 hsa-mir-3607-5p Y Y ELK4 HeLa chr5:85913453- 2.8 414 Y Y 85913816 HA-E2F1 MCF7 chr5:85913599- 2.7 165 Y Y 85914048 SRF MCF7 chr5:85913370- 2.8 52 Y Y 85914409

245 ELK1 HeLa chr5:85913259- 2.9 400 Y Y 85914252 MAZ HeLa chr5:85913513- 2.7 518 Y Y 85914365 USF2 HeLa chr5:85913526- 2.7 592 Y Y 85914008 ZNF143 HeLa chr5:85913426- 2.8 1000 Y Y 85914130 GATA3 MCF7 chr5:85913055- 3.2 498 Y Y 85914565 c-Myc MCF7 chr5:85913702- 2.5 1000 Y Y 85914273 NR2F2 MCF7 chr5:85913539- 2.8 89 Y Y 85914657 ELF1 MCF7 chr5:85913352- 2.8 691 Y Y 85914465 ELK1 HeLa chr18:33161918- 9.7 392 Y Y 33162077 MAX HeLa chr18:33161918- 9.7 501 Y Y 33162336 hsa-mir-3975-3p Y Y c-Myc MCF7 chr18:33161972- 9.7 706 Y Y 33162276 CEBPB MCF7 chrX:73439724- 1.6 371 Y N 73440008 c-Myc MCF7 chr17:27188732- 0.4 572 Y N 27188736 hsa-mir-421-3p Y Y CEBPB MCF7 chrX:73439724 1.6 371 Y N -73440008 hsa-mir-451a-5p Y Y c-Myc MCF7 chr17:2718873 0.4 572 Y N 2-27188736

(B) Non-Phasic miRNAs

Non-phasic Phasic Phasic Transcription Cell Coordinates of Distance Score H3K27ac H3K4me3 miRNAs HeLa MCF7 Factor Line TF binding in kb hsa-mir-378a-3p N N c-Myc MCF7 chr5:149109767- 2.6 378 Y Y 149109976 HeLa chr5:149109681- 2.7 103 Y Y 149111585 HA-E2F1 MCF7 chr5:149111386- 1 389 Y Y 149112048 HeLa chr5:149110106- 2.2 326 Y Y 149110555 EGR1 MCF7 chr5:149109402- 2.9 38 Y Y 149109963 MAX MCF7 chr5:149109384- 2.9 955 Y Y 149111983 AP-2g HeLa chr5:149109618- 2.7 103 Y Y 149110786 ELK1 HeLa chr5:149110817- 1.5 399 Y Y 149110865 INI1 HeLa chr5:149109484- 2.8 298 Y Y 149111954 MAZ HeLa chr5:149109596- 2.7 1000 Y Y 149111360 USF2 HeLa chr5:149111201- 1.2 410 Y Y 149111334 ZNF143 HeLa chr5:149110545- 1.8 390 Y Y 149110697 hsa-mir-34a-5p N N None hsa-mir-186-5p N N None hsa-mir-484-5p N N PRDM1 HeLa chr16:15736345- 0.8 116 Y Y 15736588 STAT1 HeLa chr16:15736387- 0.85 102 Y Y 15736746 SMC3 HeLa chr16:15736485-0.95 370 Y Y

246 15736762 MAX MCF7 chr16:15736341- 0.8 88 Y Y 15737714 HeLa chr16:15736276- 0.75 1000 Y Y 15737753 ELK1 HeLa chr16:15736380- 0.85 570 Y Y 15737532 MAZ HeLa chr16:15736308- 0.75 1000 Y Y 15737712 p300 HeLa chr16:15735947- 1.1 384 Y Y 15736548 USF2 HeLa chr16:15736281- 0.75 395 Y Y 15736573 ZNF143 HeLa chr16:15736459- 0.95 1000 Y Y 15737681 c-Myc MCF7 chr16:15736729- 0.7 757 Y Y 15737158 hsa-mir-505-3p N N HA-E2F1 HeLa chrX:139014150- 7.8 252 Y Y 139014433 MAX HeLa chrX:139013617- 7.2 1000 Y Y 139016005 MCF7 chrX:139015632- 9.2 33 Y Y 139016018 ELK1 HeLa chrX:139014987- 8.6 478 Y Y 139015808 INI1 HeLa chrX:139013902- 7.6 160 Y Y 139016316 MAZ HeLa chrX:139014122- 7.8 1000 Y Y 139016025 ZNF143 HeLa chrX:139014055- 7.7 435 Y Y 139014468 c-Myc MCF7 chrX:139013493- 7.1 583 Y Y 139013573 HeLa chrX:139015332- 7.15 812 Y Y 139015677 hsa-mir-222-3p N N SMC3 HeLa chrX:45608826- 4.4 659 Y N 45609065 hsa-mir-18b-5p N N c-Myc MCF7 chrX:133306417- 2.4 597 N Y 133306486 hsa-mir-100-5p N N None hsa-mir-30d-5p N N STAT1 HeLa chr8:135822525- 5.9 116 Y Y 135823208

* Distance from miRNAs in Kilobases

Appendix 3.8: List of molecular and cellular functions associated with TargetScan predicted targets of miR-3607-5p

Category p-value Molecules Tissue 5.1E-07- ABCD1,GLDN,VTI1A,FGF2,PIK3R1,BMP2,SSBP2,DICER1,MYH11,ATP2A2,MMP24,COL5A1,HMGB1,FOXO Morphology 1.3E-02 3,MARCKS,SIGLEC1,TCF7L1,GDF11,APC,GNAI2,CCM2,GNAI3,GINS1,FOXO1,NDEL1,BMF,RASSF2,SNX1, ARHGAP35,CAPRIN1,KDR,NDST1 Cellular 3E-05- EPS8,FGF2,PIK3R1,TXLNA,BMP2,SSBP2,DICER1,MYH11,HSPA5,ATP2A2,TRPC3,SEMA3D,HMGB1,IKZF2, Development 1.15E-02 FOXO3,ULK2,MAP3K2,SIGLEC1,CNOT6L,TCF7L1,TUSC3,PDE4D,NFE2L1,GDF11,APC,CCM2,GNAI2,EMP1 ,COL1A1,FABP2,NDEL1,FOXO1,RASSF2,MBNL1,CAPRIN1,NUP98,KDR,NDST1,FZD7 Cellular 3E-05- FGF2,BMP2,PIK3R1,TXLNA,MYH11,DICER1,HSPA5,TRPC3,ATP2A2,IKZF2,HMGB1,FOXO3,MAP3K2,SIGL Growth and 1.15E-02 EC1,SERTAD2,CNOT6L,TUSC3,PDE4D,GDF11,APC,GNAI2,EMP1,COL1A1,FOXO1,RSU1,CAPRIN1,SETBP1 Proliferation ,KDR,NUP98 DNA 3.45E-05- ABCD1,GNAI2,GNAI3,HMGB1,RAB2A,HSPA5,PDE4D Replication, 1.29E-02 Recombinati on, and Repair Cellular 3.55E-05- SLAIN2,EPS8,RAB2A,FGF2,TRIM9,BMP2,PIK3R1,MARK1,DICER1,MYH11,HSPA5,ATP2A2,COL5A1,HMG

247 Function and 1.23E-02 B1,SEMA3D,FOXO3,MARCKS,NAPB,ULK2,SIGLEC1,SERTAD2,APC,GDF11,GNAI2,COL1A1,GNAI3,FOXO Maintenance 1,NDEL1,BMF,SNX1,ARHGAP35,RSU1,KDR,FZD7 Embryonic 3.99E-05- EPS8,FGF2,PIK3R1,BMP2,DICER1,MYH11,HSPA5,ATP2A2,COL5A1,SEMA3D,IKZF2,HMGB1,FOXO3,MAR Development 1.23E-02 CKS,ULK2,TCF7L1,GDF11,ARL3,APC,EMP1,CCM2,GNAI3,COL1A1,FABP2,TENC1,GINS1,FOXO1,NDEL1, RASSF2,SNX1,ARHGAP35,MBNL1,KDR,NDST1,FZD7 Cellular 4.19E-05- FOXO1,FGF2,BMP2,FOXO3,DICER1,HSPA5,ATP2A2 Compromise 6.5E-03 Tissue 8.01E-05- EPS8,FGF2,PIK3R1,BMP2,MYH11,DICER1,HSPA5,SDK1,ATP2A2,COL5A1,HMGB1,SEMA3D,FOXO3,MARC Development 1.2E-02 KS,ULK2,TCF7L1,ARL3,GDF11,APC,GNAI2,CCM2,EMP1,GNAI3,COL1A1,FABP2,TENC1,GINS1,FOXO1,ND EL1,RASSF2,ARHGAP35,MBNL1,KDR,NDST1 Cell Cycle 8.47E-05- EPS8,FGF2,BMP2,PIK3R1,SSBP2,ZNF655,DICER1,HMGB1,FOXO3,MARCKS,TFDP2,PFDN4,RAB11FIP3,MA 1.17E-02 PK6,APC,ARL3,GDF11,GNAI2,EMP1,GNAI3,COL1A1,FOXO1,RASSF2,RSU1,KDR Cancer 9.3E-05- EPS8,VTI1A,TRIM9,PIK3R1,SLC9A9,MYH15,MARK1,SSBP2,TMEM169,DICER1,HSPA5,ATP2A2,TOMM70A 1.23E-02 ,MMP24,HMGB1,IKZF2,ETF1,TFDP2,ANKFN1,NOVA1,FAM26E,ZC3H6,DNALI1,TUSC3,WWC3,APC,GDF11 ,GINS1,SNX1,ARHGAP35,RSU1,CAPRIN1,AK2,NUP98,SETBP1,DPP10,IQSEC1,RREB1,FGF2,ARL4C,BMP2, TXLNA,DGKB,ZNF655,MYH11,SDK1,KCNT2,FOXO3,DIP2B,ABHD2,CD109,MARCKS,BTBD3,WDR1,KPNA 3,SIGLEC1,MCMBP,SERTAD2,MAPK6,SLCO2A1,TCF7L1,PDE4D,INO80D,NFE2L1,GNAI2,CCM2,EMP1,CO L1A1,TENC1,FOXO1,BMF,FAM13A,MBNL1,KDR,A1CF,FZD7 Cellular 1.1E-04- EPS8,FGF2,PIK3R1,BMP2,MARK1,DICER1,MYH11,HSPA5,HMGB1,FOXO3,ABHD2,MARCKS,RAB11FIP3, Movement 1.17E-02 MAP3K2,PDE4D,TUSC3,ARL3,APC,GNAI2,GNAI3,COL1A1,TENC1,FOXO1,NDEL1,ARHGAP35,KDR,NDST 1 Cell 1.12E-04- SLAIN2,EPS8,FGF2,PIK3R1,BMP2,DICER1,HSPA5,TRPC3,COL5A1,HMGB1,SEMA3D,FOXO3,ULK2,GDF11, Morphology 1.23E-02 APC,COL1A1,GNAI3,PHLPP2,FABP2,NDEL1,FOXO1,ARHGAP35,RSU1,KDR,NDST1 Hematologic 1.7E-04- SIGLEC1,FGF2,PIK3R1,BMP2,SSBP2,DICER1,MYH11,PDE4D,ATP2A2,APC,NFE2L1,GNAI2,COL1A1,GNAI3 al System 1.17E-02 ,HMGB1,FOXO1,IKZF2,RASSF2,BMF,FOXO3,KDR,NUP98,MAP3K2,FZD7 Development and Function Hematopoies 1.7E-04- FGF2,BMP2,PIK3R1,SSBP2,DICER1,MYH11,NFE2L1,GNAI2,FOXO1,BMF,RASSF2,FOXO3,NUP98,KDR is 8.49E-03 Tumor 2.49E-04- FGF2,TXLNA,BMP2,DICER1,PDE4D,HSPA5,TUSC3,ATP2A2,APC,COL1A1,FOXO1,HMGB1,FOXO3,KDR Morphology 6.5E-03

248