1 FUNCTIONAL DOMAIN CHARACTERIZATI ON OF GW BODY MARKER PROTEIN GW182 By SONGQING LI A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2009
2 2009 Songqing Li
3 To my mother who constantly en courages me to pursue my dream; to my wife who supports me every day; especially to my high school chemistr y teacher and my Ph.D. mentor who help me to reach two of the most important milestones in my career
4 ACKNOWLEDGMENTS It is with m y greatest appr eciation and sincere thanks that I acknowledge the following individuals for their contribution to various aspe cts of my development as a scientist and as a person. Without their guidance, assistance and support, this work would not be possible. First of all, I would like to give my heartfelt thanks to the mo st important people in my life: my mother and my wife. Despite the many adversities we faced, my mother raised me to always look at the bright side of life, encouraged me to overcome th e hurdles that stood in the ways and most importantly to constantly pursue my dreams My wife was instrumental in motivating me to come to the United States to pursue my gra duate training. The utmo st support and love she provides each day keeps me going and I am lucky to have her as my life partner and to be able to call her mine. I also want to sincerely thank two of the most important mentors in my life. My high school chemistry teacher, Ms. Jieq iu Zhang, treated all her students like her children. It was she who led me to the right direction when I was aimless during my teen age years. Her spirit always assured me that I can always be better if I wa nted to. Dr. Edward Chan, my Ph.D. mentor and dear friend, has solely shaped me into the scientist I am today. He has led me to the ivory tower of science and has not only showed me the knowledge in the world of science, but also given me the tools to explore this wondrous world. His patience and guidance has taught me that scientific research is the noblest quest for truth. In addition, I am very grateful to my co mmittee: Dr. Minoru Sat oh, Dr. Rolf Renne, Dr. Brian Harfe and Dr. Hideko Kasahara for their valu able time, effort, advice and support. I would like to extend my gracious thanks to Dr. Minoru Satoh for his help in many of my experiments, data analysis and his constant encouragement al ong my way. I also woul d like to thank everyone in the Chan lab for their suppor t, assistance and friendship, thr oughout the years especially Mr.
5 John Hamel, our former lab technician, for his patience and kindness in teaching me basic laboratory techniques. I also appreciate the help of our collaborators Dr. Marvin Fr itzler and Ms. Joanna Moser for providing human sera and assistance for my e xperiments, and for reviewing my manuscripts. I want to specially thank Dr. William McArthur for generously providing me with the NIH NIDCR training grant support during my graduate career. I am al so grateful to Dr. Witold Filipowicz, Dr. Jens Lykke-Andersen, Dr. Tom Hobman, Dr. Peter Sayaski, and Dr. Gordon Chan, for generously providing reagents.
6 TABLE OF CONTENTS page ACKNOWLEDGMENTS ............................................................................................................... 4LIST OF TABLES ...........................................................................................................................9LIST OF FIGURES .......................................................................................................................10ABSTRACT ...................................................................................................................... .............12 CHAP TER 1 INTRODUCTION .................................................................................................................. 14The Discovery of GW Body and Its Marker Protein GW182 ................................................14The Functional Link between R NA Interference and GW Body ........................................... 15The Importance of GW182 in miRNA Function .................................................................... 172 IDENTIFICATION AND CHARACTERIZAT ION OF T NGW1 AS THE NOVEL ISOFORM OF GW182 AND THEIR FUNCTION IN GWB FORMATION AND TRANSLATIONAL REPRESSION ...................................................................................... 19Introduction .................................................................................................................. ...........19Materials and Methods ...........................................................................................................20Identification of TNGW1 mRNA .................................................................................... 20TNGW1 cDNA Cloning and Construction of Expression Plasmids ...............................21Generation of Antibodies Specific to the TNGW1 Isoform ............................................ 22Cell Culture and Transfection .........................................................................................22Immunoprecipitation (IP) and Wester n Blot Analysis (IP-WB) ..................................... 23Addressable Laser Bead Immunoassay ........................................................................... 24Characterization of Anti-rTNR Antibodies by Synthetic Peptide Epitope Mapping ...... 24Indirect ImmunoFluorescence Assay .............................................................................. 25Tethering Assay Using the Dual Luciferase System .......................................................26Quantification of mRNA Degradati on Using Quantitative RT-PCR ..............................26Results .....................................................................................................................................27TNGW1 Was a Novel Isof orm of Human GW182 ......................................................... 27Intracellular Localization of TNGW1 and Its Relationship with Other GWB Components ................................................................................................................. 31TNGW1 Is Not Essential for the Formation of GWB ..................................................... 32TNGW1 and GW182 Exert Strong Repre ssion Effect in Ago2 Mediated Translational Silencing ................................................................................................33Discussion .................................................................................................................... ...........35Expression of Human TNRC6 A/GW182 Gene and Its Effect on GWB Formation ....... 35Interdependence of Ago2 and TNGW1/GW182 in miRNA-Mediated Translational Repression .................................................................................................................... 36The Functional Differences of GW182 Isof orms and the Heterogeneity of GWB ......... 37
7 3 THE C-TERMINAL HALF OF AGO2 BI NDS T O MULTIPLE GW-RICH REGIONS OF GW182 AND REQUIRES GW182 TO MEDIATE SILENCING .................................. 52Introduction .................................................................................................................. ...........52Materials and Methods ...........................................................................................................53Construction of Deletion C onstructs of GW182 and Ago2 ............................................. 53Antibodies .................................................................................................................... ....55Plasmid Transfection, GST Pull-down, and Western Blot Analysis ............................... 55Indirect Immunofluorescence .......................................................................................... 56Tethering Assay Using a Dual Luciferase System ..........................................................56RNA Interference and Quantitative Real Time PCR ...................................................... 57Results .....................................................................................................................................57GW182-Ago2 Interaction Wa s Important for the Lo calization of Ago2 in Cytoplasmic Foci ......................................................................................................... 57Ago2 Bound to Multiple Non-Overla pping GW-Rich Regions of GW182 .................... 58Tryptophan Residues of GW1 1a Were Not Required for Interaction With Ago2 ........ 59The Interaction of Ago2 With GW182 Was Conserved In Other Human Ago Proteins ...................................................................................................................... ..60Tethering C-Terminal Half of Ago2 to the 3-UTR of mR NA Recapitulated Ago2Mediated Silencing Which Requi red the Presence of GW182 .................................... 60Discussion .................................................................................................................... ...........61Formation of GW182 and Ago Protein Complexes ........................................................ 61C-terminal Half of Ago2 Preserved th e Silencing Function of Ago2 Probably Because It Maintained th e Interaction With GW182 ................................................... 634 IDENTIFICATION OF TRANSLATIONAL REPRESSION DOM AINS IN GW182 ........74Introduction .................................................................................................................. ...........74Materials and Methods ...........................................................................................................75Plasmids ...................................................................................................................... .....75Antibodies .................................................................................................................... ....76Cell Culture and Plasmid Transfection ............................................................................ 76GST Pull-down and Western Blot Analysis .................................................................... 76Tethering Assay and Dual Luciferase Assay ...................................................................77Results .....................................................................................................................................78GW182 contained two putative, non-overlap ping regions harboring the repression effect in tethering assay. ..............................................................................................78Endogenous acidic ribosomal protein P0, but not P1 or P2, was specifically associated with the complexes of Ago2 and GW182 truncated constructs GW1 12 and GW1 5.................................................................................................. 79Discussion .................................................................................................................... ...........81
8 5 DISCUSSION AND CONCLUSIONS ..................................................................................89GW182 is the Repression Trigger of MiRNA-Mediated Gene Silencing .............................. 89The Redundancy of RNAi Factors and Their Po tential Link to the Different Outcomes of miRNA-Mediated Gene Regulation ............................................................................... 91Working Model of GW182 in miRNA-Mediated Gene Silencing ......................................... 93LIST OF REFERENCES ...............................................................................................................97BIOGRAPHICAL SKETCH .......................................................................................................104
9 LIST OF TABLES Table page 3-1 Primer sequences and PCR conditions fo r constructs used in the current s tudy ............... 73
10 LIST OF FIGURES Figure page 2-1 Schematics of the human TNRC6 A gene products GW 182 and TNGW1 ........................ 40 2-2 TNGW1 mRNA containing the TNR exon detected in human testis and different cell lines using RT-PCR ........................................................................................................... 41 2-3 Production and characterization of polycl onal and monoclonal antibodies specific to the TNGW1 isoform .......................................................................................................... 42 2-4 TNGW1 and GW182 were independen t products of human TNRC6A gene .................... 43 2-5 TNGW1 resided in a subset of GWB (part 1) .................................................................... 44 2-6 TNGW1 resided in a subset of GWB (part 2) .................................................................... 45 2-7 Intracellular localization of T NGW1 with other GWB components ................................. 46 2-8 Knockdown of TNGW1 has no apparent eff ect on the assembly of GWB in HeLa cells ......................................................................................................................... ...........47 2-9 Dual luciferase assay measurement a nd NHA constructs expression in 293 cells. ........... 48 2-10 Tethered TNGW1 and GW182 exerted transl ational repression to a greater extent than Ago2. ..........................................................................................................................49 2-11 Translational repression of tethered Ago2 required endogenous TNGW1 and/or GW182. ........................................................................................................................ ......50 2-12 Quantitative real time PCR showed corre sponding siRNA knock-down effect in the tethering assay. ...................................................................................................................51 3-1 Schematic of human GW182 and Ago2 dele tion constructs using in this study. ..............65 3-2 GW182 fragment GW1 (aa566-1343) recruited Ago2 to cytoplasmic foci by interacting with the C-te rminal half of Ago2. .................................................................... 66 3-3 The C-terminal half of Ago2 bound to mu ltiple non-overlapping GW-rich regions of GW182 ......................................................................................................................... ......67 3-4 Four subregions of GW1 1a are capable of binding Ago2 and the interaction between GW1 1a and PIWI was not dependent on the five tryptophan residues ............. 68 3-5 Both GW182 fragments GW1 1 and GW1 co-precipitated with other human Ago proteins .......................................................................................................................69 3-6 Gene silencing mediated by tethered C-terminal half of Ago2 required GW182 ............. 70
11 3-7 At least three non-overlapping GW-rich regions that are different from the orthologconserved GW-rich region can independently bind Ago2 ................................................. 71 3-8 Sequence of synthesized GW1 1a Mut (W>A) ................................................................ 72 4-1 Schematic of human GW182 truncated constr ucts used in this study with a summary of their interaction with Ago2 and the repression ability in the tethering assay ................ 86 4-2 Identifying two non-overlapping regions harboring repression effect in GW182 .............87 4-3 Endogenous acidic ribosomal protein P0, but not P1 or P2, was associated with GW182:Ago2 complex containing GW1 5 or GW1 12 in GST pulldown assays ..........88 5-1 A model illustrating the functional impor tance of GW182 in miRNA-mediated gene silencing ..................................................................................................................... ........95
12 Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy FUNCTIONAL DOMAINS CHARACTERIZ ATION OF GWB MARKER: GW182 By Songqing Li May 2009 Chair: Edward K. L. Chan Major: Medical Sciences -Molecular Cell Biology MicroRNAs (miRNAs) have emerged as key regulators of about 30% of all genes expressed in human. Executed by a group of ribonucleoproteins named miRNA-mediated silencing complex, most of the miRNAs induce translational repression due to its imperfect complementarity to target mRNA. Argonaute prot ein family Ago1-4 is the best characterized components in miRISC. The mechanism of miR NA-mediated translational repression remains unclear. The mRNAs targeted by miRNA were turned over in cytoplasmic foci named GW bodies. GW182, an 182kDa protein characterized by multiple glycine/tryptophan (GW) repeats, is important for GW body formation and miRNAmediated gene silencing. Here, we aim to characterize the functional domains in GW182 and understand their importance in miRNAmediated gene regulation. We first identifie d TNGW1, which contains trinucleotide repeats (TNR) in its mRNA, as a novel GW182 isoform. Knockdown of TNGW1 mRNA did not affect GW182 protein level, indicating GW182 was transcribed and translated independently from TNGW1. Using truncated constructs and glutathione-S-transferase pulldown assay, we discovered that GW182 and TNGW1 contained four non-overlapping regions which were able to interact with the C-terminal half of Ago2 and other Ago prot eins. Mutagenesis study showed that GW motif may not be im portant for some of the GW182: Ago2 interaction. Further study found that GW182 and TNGW1 acted more directly than Ago2 in translational repression. The
13 repression effect caused by tethered GW182 and TNGW1 were not dependent on Ago2 protein. Mapping of different Ago proteins revealed that their abiliti es to cause repression were associated with their interact ion with GW182. Lastly, function al studies narrowed down two non-overlapping regions in GW182 har boring inhibitory effect to luciferase reporters. These regions were also found associated with ribos omal protein RPLP0, but not RPLP1 or RPLP2, implying the presence of incomplete ribosomal stalk structures in the repression complex. Our finding proposed a model that GW182 acts as th e repression trigger in miRNA related gene regulation, where the ribosomal stalk could pot entially be the target in this event.
14 CHAPTER 1 INTRODUCTION The Discovery of GW Body and Its Marker Protein GW182 In 2002, GW182 was identified and cloned by our laboratory as a novel protein using autoimm une serum from a patient with motor a nd sensory neuropathy (Eys tathioy et al., 2002). It is a protein with a size of 182kDa and characterized by its mu ltiple glycine/tryptophan (GW) repeats along the protein sequen ce. At that time, the only known domain in GW182 based on bioinformatics is the RNA recogni tion motif (RRM) in its C-terminus. GW182 is found to be associated with a specific subset of mRNAs and consistently enriched wi thin unique cytoplasmic foci designated as GW bodies that are distinct from other known cytoplasmic organelles such as Golgi complex, endosomes, lysosomes, and peroxisomes (Eystathioy et al., 2002). Therefore, GW bodies were speculated as novel cytoplasmi c foci related with mRNA storage and/or degradation. Morphologically, GW bodies are generally small, s pherical, cytoplasmic foci of about 100-300 nm in diameter and devoi d of a lipid bilayer membrane (Yang et al. 2004). More interestingly, the number and size of GW bodies va ries in different cell types and at different stages of the cell cycle (Yang et al. 2004). In 2003, Sheth and Parker report in yeast Dcp1-positive cytoplasmic foci as processing bodies (P bodies). P bodies were found enrich of mRNA degradation in termediates in addition to the 5-3 mRNA decay factors (Sheth and Parker, 2003). Later study in mammalian cells shows RNA decay factors Dcp1 and LSm4 co-local ize with GW182 in GW bodies (Eystathioy et al., 2003c) implying GW body is the mammalian count erpart of yeast P bodies. GW bodies are also shown to contain poly (A)+ RNA a nd dynamically disappear as mRNA breakdown was abolished (Cougot et al., 2004). Therefore, GW body are considered as the mammalian
15 analogues of P bodies and as the sites for active 5-3 mRNA degradation, which are designated here provisionally as GW/P bodies (GWB). GW182 is important for the assembly of GW B. In vitro gene knockdown of GW182 using short hairpin RNA (shRNA) plasmid results in disappearance of GWB (Yang et al. 2004). Expression of a dominant negative construct of GW182 GW1 1 also disassemble GWB (Jakymiw et al., 2005). Therefor e, GW182 was proposed as the ma rker as well as the matrix protein of GWB. The Functional Link between RNA Interference and GW Body RNA interference (RNAi) is a conserved m echanism of gene regulation involved in multiple biological functions. In lower organisms, it was initially described as a genetic control mechanism implicated in virus resistance (Covey et al., 1997; Ratcliff et al., 1997), genome maintenance (Assaad et al. 1993) and developmental control (Boerjan et al. 1994). In 1998, RNAi was first characterized in C. elegans as a potent and sequence specific mechanism that silences endogenous genes by Andrew Fire and Craig Mello (Fire et al. 1998), who later received the 2006 Nobel Prize in Physiology or Medicine for thei r contribution. The classical RNAi activity is triggered by small double-stra nded RNAs including small interfering RNA (siRNA) and microRNA (miRNA). In siRNA-mediated silencing, the dsRNAs, which are formed in cells or are introduced into cells by viral infection or ar tificial expression, are processed by RNase III enzyme, Dicer, into ~20-bp double-stranded small interfering RNAs (siRNAs). The antisense strands of the siRNAs are then incorporated into siRNA-induced silencing complex (siRISC). Subsequently, siRI SC binds to and cleaves target mRNA with complete complementary sequence to the siRNA (Filipowicz et al., 2005). Argonaute 2 (Ago2) is the core component of RISC and harbors RNase H activity responsible for the cleavage of
16 target mRNA (Liu et al., 2004). Different from siRNA pathway, miRN As are endogenous ~21nt regulatory RNAs that are evolutionarily conserved in most species and are estimated to regulate ~30% of protein-encoding genes in human (Lewis et al. 2005). MiRNAs are processed from endogenous precursor molecules folded into hairpin-like structure. The maturation of miRNAs includes two steps, both catalyzed by enzymes of the RNase III family, Drosha and Dicer. Drosha, working with its co-effecter DGCR8, is responsible for the processing of primary miRNA transcripts (pri-miRNAs) to ~70-nt hair pins named precursor miRNAs (pre-miRNAs). Subsequently, Dicer, accompanied with TRBP, processes pre-miRNAs into mature miRNAs, which binds to the 3-UTR of target mRNA by the seed sequen ce and regulates gene expression by increasing instability or repressing translation of target mRNA (Filipowicz et al. 2005). The mature miRNA then incorporates into miRNA-in duced silencing complex (miRISC) and target to the mRNA in a sequence spec ific manner. If the miRNA is highly complementary to the mRNA target, it can induce mRNA cleavage as siRNA (Hornstein et al., 2005) through the slicing activity of Ago2. However, if the miRNA is only partially complementary to the target mRNA, the miRISC will induce translationa l repression. In mammal, the only well characterized components in miRISC are the fo ur Argonaute proteins, Ago1 to Ago4. However, the molecular mechanism of the miRNA-mediat ed translational repression remains unclear. GWB is closely associated with RNAi activity. Firstly, the most important RNAi factors, Ago proteins and siRNA (Jakymiw et al., 2005; Li u et al., 2005a), are f ound enriched in GWB. Secondly, the size and number of GWB are related to RNAi activity. Increase in siRNAmediated activities induced GWB formation (Lia n et al., 2007) whereas inhibition of miRNA pathway led to disassembly of GWB (Pauley et al., 2006; Eulalio et al., 2007b). Thirdly, the GWB contain enriched 5-3 mR NA degradation factors, in cluding Xrn1 (5-3exonuclease),
17 Dcp2:Dcp1 (decapping enzyme), and LSm1-7 comp lex (stimulator of mRNA decapping) (Heyer et al., 1995; Bashkirov et al., 1997; Ingelfinger et al., 2002; van Dijk et al., 2002), which are responsible for miRNA-mediated mRNA degrada tion. Therefore, GWB is proposed as the biomarker of cell RNAi ac tivity (Lian et al., 2007). The Importance of GW182 in miRNA Function Although multip le models were proposed by di fferent groups (Filipowicz et al., 2008), the mechanism of how miRNAs tri gger the translational repression remains unclear. However, because direct cleavage of the ta rget mRNA is hindered in most of the cases, more protein factors need to be recruited to assure the repression function. We are one of the first two groups to describe the importance of GW182 in RNAi f unction (Jakymiw et al., 2005; Liu et al., 2005a). The importance of GW182 is further supported by th e later studies and can be summarized as the following three aspects. First, GW182 tightly interacted with Argonaut e proteins in human (Jakymiw et al., 2005; Liu et al., 2005a), and other species incl uding Drosophila (BehmAnsmant et al., 2006), C. elegans (Ding et al., 2005) and Arabidopsis (El-Shami et al., 2007). In addition, the interaction between GW182 and Ago2 was RNA-independent (Liu et al., 2005a) and proposed to depend on evolutionally conserve d WG/GW motif (El-Shami et al., 2007; Till et al., 2007). A recent report showed the GW182:Ar gonaute interaction was essential for miRNA induced gene silencing and its subsequent mRNA decay in Drosophila (Eulalio et al., 2008b). Second, the translational repres sion effect was impaired when GW182 was knocked down (Liu et al., 2005a; Chu and Rana, 2006). Behm-Ansmant et al. also showed in Drosophila that tethering GW182 to the 3-UTR of mRNA, whic h bypassed the requirement of miRNA, led to translational repression (BehmAnsmant et al., 2006). Last GW182 was important for GWB formation as it may form an optimal microenviro nment for recruiting the RNA decay factors to
18 GWB. Tethering GW182 to mRNA induced RNA decay required 5 3 RNA decay factors in Drosophila (Behm-Ansmant et al ., 2006). In the absence of GW182, even the formation of GWB cannot be detected. Nevertheless, the inter-dependence among these proteins during translational repression remains unclear in mammalian system. With the elucidation of additional GWB components, questions are raised about how GW182, Argonaute and RNA decay factors contribute to the RNA induced gene silencing. In the current study, we hypothesize that GW182 is an important molecule in the function of miRNA-mediated translation repression. By identifying a new isoform of GW182 and characterizing its functional domai n responsible for Ago interacti on and repression, we are able gain a better insight of the mech anism of translational repression in the molecular level. More importantly, it may lead to the development of new methods to control and adjust the miRNAmediated gene regulation, which may be used to enhance RNAi-based therapy in the future.
19 CHAPTER 2 IDENTIFICATION AND CHARACTERIZATION OF T NGW1 AS THE NOVEL ISOFORM OF GW182 AND THEIR FUNCTION IN GWB FORMATION AND TRANSLATIONAL REPRESSION Introduction RNA interference (RNAi) is a poten t post-tran scriptional regulation mechanism for gene expression. It is triggered by small molecule RNAs, including small interfering RNA (siRNA) and microRNA (miRNA), and then executed by the RNA induced silencing complex (RISC) wherein targeted mRNA is degraded through the 5 3 RNA decay pathway. GW bodies (GWB), also known as mammalian processing bodies, were found closely a ssociated with RNAi and its related RNA turnover activities (Jakymiw et al., 2007; Eulalio et al., 2007a). Increase in siRNA-mediated activities induced GWB forma tion (Lian et al., 2007) whereas inhibition of miRNA pathway led to disassembly of GWB (P auley et al., 2006; Eulalio et al., 2007b). Blocking the 5 3 RNA decay before its initiation dimi nished GWB (Cougot et al., 2004) while blocking after its initiation increased the size and number of GWB (Sheth and Parker, 2003; Cougot et al., 2004; Andrei et al., 2005). A rece nt report demonstrated that the formation of GWB was the consequence of RNAi activitie s in Drosophila (Eul alio et al., 2007). GW182, one of the marker proteins of GWB, wa s first identified in 2002 as a target protein of autoantibodies from a patient with motor and sensory neuropathy (Eystathioy et al., 2002). It is an 182kDa protein characterized by multiple glycine/tryptophan (G/W) repeats. GW182 is important for GWB formation and RNA induced gene silencing function even though it had no known enzymatic activity. Our earlier studies sh owed that knock-down of GW182 significantly disassembled GWB (Yang et al., 2004; Jakymiw et al., 2005) and impaired the efficiency of siRNA functions (Jakymiw et al., 2005). Othe r investigators reported that GW182 was more
20 important in miRNA function and closely associated with transl ational repression (Liu et al., 2005a; Chu and Rana, 2006). The formation of GW B from the sub-microscopic to microscopic level may occur during the process (Franks and Lykke-Andersen, 2007). The gene name of GW182 in NCBI GenBank da tabase is trinucleot ide repeat containing 6A, or TNRC6A. Interestingly, th e GenBank database predicts an other isoform, which we have provisionally named trinucleotide GW1 (TNGW1), containing trinucleotide repeats (TNR) in its mRNA. Expansion of TNR is known to be related to a set of diseases, most notably those with neuropsychiatric features, such as Huntingtons disease (Margolis et al ., 1997). TNRC6A is one of the 20 trinucleotide repeat containing genes in the human genome but to date it has not been related to trinucleotide e xpansion diseases. In c linical studies, we showed that autoantibodies to GW182/GWB were associated with Sjgrens syndrome, mixed motor/sensory neuropathy, ataxia, and systematic lupus erythematosus (E ystathioy et al., 2003b; Bhanji et al., 2007). However, to date there are no published repor ts describing the expression of TNGW1 and therefore, we examine its expression and potential effect on tran slational repression. Materials and Methods Identification of TNGW1 mRNA To identif y the TNGW1 mRNA polymerase chain reaction (PCR) amplification was performed on cDNA from HeLa, HEp-2 and HepG2 cell lines (ATCC, Manassas, VA), and adult human normal testis (BioChain, Hayw ard, CA) using primer TNRC-1: 5ATAATGCCAAGCGAGCTACAG-3 (nt248268), and primer TNRC-2: 5AAGGGAAGTGCCATTCATACC-3 (nt1512-1492). PCR reactions used SureStart Taq DNA polymerase (Stratagene, Cedar Creek, TX ) following manufacturers protocol. The annealing temperature for PCR am plification was 54 C. The complete nucleotide sequence of the PCR products was determined in both strands using BigDye terminat or sequencing at the
21 University of Florida Interdisciplinary Cent er for Biotechnology Res earch Sequencing Core Laboratory. TNGW1 cDNA Cloning and Construction of Expression Plasmids To construct the full-length TNGW 1, PCR amplification was conducted on the human testis cDNA (BioChain) using primer TNRC-5a: 5TTTGGAAGATCTATGAGAGAATTGG AAGCTAAAGCT-3 containi ng a synthetic Bgl II site sequence (underline) immediately upstream of TNGW1 ATG transla tional start site, and primer TNRC-2, which is downstream of an internal Kpn I site (nt1252). The 1.5kb PCR product was purified and digested with Bgl II and Kpn I to generate a 1.2kb fragment that was used to replace the 5 500bp BamH I to Kpn I fr agment in the full-length GW182 cloned in the pENTR vector; the BamH I restriction site was fro m the 5 linker sequence of the pENTR vector. Both the 1.2kb Bgl II-Kpn I fragment and the BamH I and Kpn I linearized plasmid of pENTRGW182 were gel purified and then ligated at 16 C overnight to generate pENTR-TNGW1 with an 8.4kb insert. Expression vectors, enhan ced green fluorescence (EGFP) tagged and glutathione-S-transferase (GST) tagged TN GW1 were generated using pENTR-TNGW1 and respective pDEST vectors via recombination us ing LR Clonase II (Inv itrogen, Carlsbad, CA) following the manufacturers protocol. To generate a construct to express a reco mbinant polypeptide containing the TNR (rTNR, aa1-204), pENTR-TNGW1 was first digested with BamH I (nt610) and Not I (3 end linker) to release a 6.5kb fragment consist of GW182. Th e overhangs of the vector encoding the Nterminus of TNGW1 was filled-in and then liga ted at room temperature (RT) for 1 hour to generate the deletion construct pENTR-rTNR. Expression vectors for rTNR in pDEST17 (Invitrogen) and pDEST-EGFP were generated to produce 6XHis -rTNR in E. coli and EGFPrTNR for expression in ma mmalian cells respectively.
22 The tethering assay plasmids including pCln eo-NHA vector, NHA-Ago2, firefly luciferase containing 5 BoxB structures (FL-5BoxB), Renill a luciferase containing 5 BoxB structures (RL5BoxB), firefly luciferase (FL), and Renilla luciferase (RL) were gifts from Dr. Witold Filipowicz, Friedrich Miescher Institute for Bi omedical Research, Switzerland (Pillai et al., 2004). To generate NHA-GW182 and NHA-TNGW1, th e pCIneo-NHA vector was converted to gateway destination vector usi ng the Gateway Vector Conversi on System (Invitrogen). Then TNGW1 and GW182 were moved from corresp onding pENTR vectors to the pCIneo-NHA gateway vector respectively by recombination. All DNA construc ts were confirmed by direct DNA sequencing. Generation of Antibodies Sp ecific to the TNGW1 Isoform The express ion of recombinant 6XHis-rTNR protein in BL21 (DE3) E. coli and purification by nickel affinity chromatography was performed using Qi agens protocol as previously described (Eystath ioy et al., 2002). Two New Ze aland White rabbits 6225 and 6226 were used to generate polycl onal antibodies using standard pr otocol by Lampire Biological Laboratories, Pipersville, PA. Pre-immune blood samples as well as samples collected after initial and booster injections were harvested a nd analyzed for reactivity. For the production of monoclonal antibodies (mAb), hyperimmunized BALB/c mice were used to generate hybridomas carried out by the University of Fl orida Interdisciplinary Center for Biotechnology Research Hybridoma Core Laboratory. Three m ouse mAb (2E11, 5C8 and 2F11) were selected based on enzyme-linked immunosorbent assay a nd indirect immunofluores cence screening. All three mouse mAb were identified as IgG2a, antibodies. Cell Culture and Transfection HeLa and HEK 293 cells (ATCC) were cultu red in DMEM containing 10% fetal bovine serum in a 37C incubator with 5% CO2. Lipofectamine 2000 (Invitrogen) was used for
23 transient siRNA and DNA plasmid transfection fo llowing manufacturers pr otocol. Briefly, the cultured cells were grown to 40-50% confluence and transfected with 100 nM siRNA. Cells were fixed or lysed 2 days after the transfection. In 3day experiments, cells were fixed at day 1, day 2, and day 3 after transfection. The siR NA for TNGW1 (siTNR) was designed by using the online tool from the Dharmacon website and targets in the TNR re gion of TNGW1. The corresponding sequences are: sense strand 5-UCGGUAUCCUCGUGAAGU ATT-3; antisense strand 5-UACUUCACGAGGAUACCG ATT-3. The sequence of siRNA for EGFP (siGFP), Ago2 (siAgo2) and GW182 (siGW182) were reported in previous study (Lian et al., 2007). In DNA plasmid transfection experiments, cells we re maintained at 70~90% confluence for transfection and harvested 24 or 48 hours after transfection and an alyzed by immunofluorescence and/or western blot. Immunoprecipitation (IP) and We stern Blot A nalysis (IP-WB) IP-WB analysis was performed as described in detail in a previous study (Moser et al., 2007). In brief, for the IP step, human anti -GWB antibodies from the prototype serum 18033 (Mitogen Advanced Diagnostics Laboratory, Univer sity of Calgary, Calgary, AB, Canada) were chemically cross-linked to Protein A-Sepharose b eads to prevent elution in the subsequent SDSPAGE step. IP samples were resolved in a 6.5% gel SDS-PAGE with the low molecular proteins (<75kDa) run off the gel in order to achieve th e optimal separation of two isoforms of GW182 and then electrophoretically transferred to nitrocellulose memb ranes (Bio-Rad Laboratories, Hercules, CA). Primary antibodies used in the western blot step includ ed mouse antibodies antirTNR 2E11 (undiluted), 5C8 (undiluted), anti-G W182 4B6 (1:10) (Eystathioy et al., 2003a), anti-Hemaglutinin (HA) (1:1,000, Covance, Emer yville, CA), and rabbi t antibodies, anti-rTNR 6225 (1:200), 6226 (1:200), anti -GW182 5182 (1:200), 6642 (1: 200) and anti-EGFP (1:1000, Invitrogen). Secondary antibod ies included either horseradish peroxidase (HRP)-goat anti-
24 human Ig (1:20,000; Sigma, St. Louis, MO), goat anti-rabbit immunoglobu lin (1:20,000; Jackson ImmunoResearch, West Grove, PA), or goat an ti-mouse immunoglobulin (1:2,000; Santa Cruz Biotechnology, Santa Cruz, CA). Bands were de tected using the enhanced chemiluminescence kit (Amersham Biosciences, Piscataway, NJ) or Supersignal Chemiluminescent system (Pierce Chemical, Rockford, IL). When necessary, n itrocellulose membranes were stripped using stripping buffer (100 mM, 2-mercap toethanol, 2% SDS, 62.5mM Tr is, pH 6.7) for 30 minutes at 65C for further probing. Addressable Laser Bead Immunoassay A set of addressable beads bearing laser reactive dyes (Lum inex, Austin, TX) were coupled to purified rTNR polypeptide and analyzed for antibody reactivity as previously described (Eystathioy et al., 2003a). The mous e mAb and rabbit sera we re diluted in QUANTA Plex diluent (INOVA, San Diego, CA) to a final c oncentration of 1:100. Thirty microliters of QUANTA Plex diluent was added to each well followed by 10 l of the diluted sample and then incubated on an orbital shaker for 30 minutes at RT. This was followed by the addition of 40 l of phycoerythrin-conjugated species specific anti-IgG (Jackson ImmunoResearch, West Grove, PA) diluted 1:50 to each well and incubated on th e orbital shaker for an additional 30 minutes. The reactivity of the antigen-coated beads wa s determined on a Luminex 100 dual-laser flow cytometer (Luminex, Austin, TX). Each assa y included negative and positive controls and results were expressed as me dian fluorescent units (MFU). Characterization of Anti-rTNR Antibodies by Synthetic Peptide Epitope Mapping To characterize the specific reactivity of anti-rTN R antibodies generated, membranes containing in situ synthesized sequential 15mer peptides offset by five amino acids and representing the region-terminal domain of the TNGW1 protein (Table S1) were prepared (Eve Technologies, Calgary, AB, Canada) as previously de scribed (Selak et al., 2003; Eystathioy et al.,
25 2003a). The dehydrated membranes were prepared for immunoblotting by an initial 10 minutes incubation in 100% ethanol followed by rehydration in Tris-buffered saline (TBS; 10 mM Tris HCl pH 7.6, 150 mM NaCl) for 10 minutes at RT. The membranes were blocked in 2% milk/TBS overnight at 4C and incubated with various primary antibodies at the appropriate dilution for 1.5 hours at RT on a shaker. Follo wing 3 washes of 5 minutes each with 2% milk/TBS, appropriate HRP-conjugated secondary antibodies diluted in 2% milk/TBS as described above were incubated with the me mbranes for 45 minutes at RT on a shaker. Membranes were washed 3 times with TBS fo r 2 minutes each and th e bound antibodies were detected using the enhanced chemiluminescence kit (Amersham Biosciences). The same stripping method was used for the peptide membra ne as described in the western blot section. Indirect ImmunoFluorescence Assay HEp-2 slides (ImmunoConcepts, Sacramento, CA) or HeLa cells grown as a monolayer were used to perform indirect immunofluorescence assay as described (Jakymiw et al., 2005). Primary antibodies used included: mouse anti-rT NR 2F11 (undiluted culture supernatant), rabbit anti-rTNR 6226 (1:200), rabbit anti-GST (1:1000) (provided by Dr. Sayeski, University of Florida), rabbit anti-Dcp1a (1:500) (provided fr om Dr. Lykke-Andersen, University of Colorado), human anti-GWB sera 18033 (1:6000) and IC 6 (1:1000, Mitogen Advanced Diagnostics Laboratory, University of Calg ary). Secondary antibodies include Alexa Fluor 488 (1:400, Invitrogen), Alexa Fluor 568 (1:400, Invitrogen), and Cy5 (1 :100, Jackson ImmunoResearch Laboratories, West Grove, PA) conjugated goat antibodies to IgG of corresponding species (human, rabbit or mouse). Goat anti-mouse Ig G2a TRITC (1:50, Southern Biotech, Birmingham, AL) and goat anti-mouse IgG1 488 (1:400, Invitrogen) were used specifically in 2F11 and 4B6 dual staining assay. The slides were mounted by using VECTASHIELD mounting medium with 4',6-diamidino-2-phenylindole (DAPI; Vector Laboratories, Southfie ld, MI). Fluorescent images
26 were captured with a Zeiss Axiovert 200M mi croscope (Carl Zeiss, Jena, Germany and processed using Adobe Photoshop (Adobe Systems, San Jose, CA). Magenta was used as a pseudo-color in images when needed. Tethering Assay Using the Dual Luciferase System HEK 293 ce lls seeded at 75-80% confluence were transfected with 600ng DNA plasmid of NHA vector, NHA-Ago2, NHA-GW182 or NHA-TNGW1 pl us targeted luciferase (either 150ng FL-5BoxB or 10ng RL-5BoxB) and control luci ferase (50ng RL or 100ng FL) plasmid using Lipofectamine2000. Cells were harvested 48 hou rs after transfection and the FL and RL activities were measured using Dual-Luciferase Reporter Assay System (Promega, Madison, WI) following the manufacturers protocol. Rela tive luciferase activities (ratio of targeted luciferase activities over control luciferase activi ties) were first calculated as described in the laboratory of Dr. Filipowicz (Pi llai et al., 2004) and the translational repression was calculated based on a recent study (Lytle et al., 2007). Briefly, FL-5BoxB/RL activity in NHA vector transfected (control) group was regarded as 0% tr anslational repression. Th e repression levels of other experimental groups were calculated by th e percentage reduction of relative luciferase activities compared to th at in NHA control group. All data we re collected in 3 to 6 independent experiments for statistical analysis. The expr essions of all NHA constructs were monitored by western blot as shown in Fig. 2-9B. Quantification of mRNA Degradat ion Using Quantitative RT-PCR Total RNA sam ples from tethering assays were extracted from HEK 293 cell lysates by using RNeasy Mini Kit (Qiagen, Valencia, CA). RNase-Free DNase Set (Qia gen) was applied to eliminate the potential DNA contamination. Sample s were analyzed in duplicate by quantitative RT-PCR using SYBR-Green Master mix (Applied Bi osystems, Foster City, CA). The relative mRNA levels of FL-5BoxB/RL were calculated by Ct method. The melting curve in each
27 individual measurement was monitored to guard against non-specific am plification. The FL5BoxB/RL mRNA levels of NHA-Ago2, NHAGW182 and NHA-TNGW1 were compared to the mRNA level in NHA vector transfected c ontrol group, which was defined as 0% mRNA degradation, and calculated th e corresponding mRNA degradation. Sequences of primers for RL were: forward 5-TCCTACGAGCACCAAGACAAGA-3, reverse 5GATCACGTCCACGACACTCTCA-3. Sequences of primers for FL were: forward 5GCGACCAACGCCTTGATT-3, reverse 5-T CCCAGTAAGCTATGTCTCCAGAA-3. In siRNA tethering assay, the mRNA levels of Ago2 and TNGW1/GW182 were measured using TaqMan Fast Universal Master Mix (Applie d Biosystems) with the corresponding TaqMan Gene Expression Assay (Ago2, Hs00293044_m1 ; TNRC6A, Hs00379422_m1 and 18S rRNA, 4310893E, Applied Biosystems). Results TNGW1 Was a Novel Isoform of Human GW182 The TNRC6A/GW 182 gene locates on human chromosome 16p11.2. The first reported protein isoform of this gene, GW182, has distinct regions enriched in glycine (G) and tryptophan (W) repeats referred to as the GW-rich regions (Eystathioy et al., 2002). GW182 also has a glutamine/asparagine (Q/N)-rich region (Decker et al., 2007) in the middle and a classic RNA recognition motif (RRM) near the C-terminus. The predicted novel form, TNGW1 (Fig. 2-1A), is a protein of about 210 kDa c ontaining a TNR Q-repeat domain (aa93-127) in its N-terminus. The mRNA of TNGW1 contains 5 additional exons upstream of the putative AUG start codon of GW182 (Fig. 2-1B). The TNR Q-repeat domain is encoded by the 5th exon of TNGW1 mRNA and its corresponding nucleotide a nd amino acid sequence are shown in Fig. 2-1C. Interestingly, based on the genomic sequence data analysis us ing the University of California Santa Cruz Genome Browser software, the translation initiati on sites of these two isof orms are predicted to
28 be about 60kb apart (Fig. 2-1B). Sequence alignm ent analysis showed that the N-terminus of TNGW1 was conserved among human, rat, and mouse with some de gree of diversity in the Qrepeat region (Fig. 2-1D). Similar domains were not identified in Drosophila or C. elegans or the two other human homologues TNRC6B and TNRC6C. Based on the predicted sequence of TNGW1 in the NCBI GenBank database, the RT-PCR assay was designed to verify its existence at the mRNA level using primer sets flanking the unique region of TNGW1 (nt248-1492). The anti cipated 1.2kb PCR bands were amplified from cDNA samples of HeLa, HEp-2, HepG2 cells and human testis (Fig. 2-2). The 1.2kb PCR products from HeLa, HEp-2 and human testis were gel-purified, submitted for direct DNA sequencing and verified to be identical to the NCBI reference sequence (NM_014494.2). These results demonstrated that th e TNGW1 mRNA, containing an in -frame junction between the novel 5 exons and GW182, could be detected in at least 2 human cancer cell types and a normal adult tissue. After experimentally verifying that the mRNA of TNGW1 was expressed, we were interested to determine whethe r GW182 and TNGW1 proteins were both expressed. Antibodies specifically recognizing TNGW1 were developed to complement th e previously generated antiGW182 antibodies (Eystathioy et al., 2002; Eystathioy et al., 2003a). A recombinant polypeptide containing the TNR Q-repeat domai n (rTNR, aa1-204) was generated to immunize two rabbits 6225 and 6226. The polyclonal antibodies isolated from both rabbits showed strong reactivity to rTNR by addressable laser beads i mmunoassay (Fig. 2-3A) or by western blot (Fig. 2-3B) but did not cross-react with GW182 (Fig. 2-3B). Pre-immune antibodies from rabbit 6225 and 6226 as well as the two rabbit polyclona l antibodies to GW182, 5182 and 6642, did not show reactivity to rTNR (Fig. 2-3A, B). Furthermore, three mouse mAb identified as 2E11,
29 2F11, and 5C8 were generated to rTNR after init ial screening and subsequent subcloning. As demonstrated by the addressable laser bead as say, all three anti-rTNR mAb showed high MFU indicating strong reactivity to rTNR (Fig. 2-3C). In contrast, mAb to GW182 (GW182 4B6) and mAb to TNRC6B (GW2 25) had remarkably low MFU which was comparable to culture supernatant controls (Fig. 2-3C). Since there are more than 20 TNR containing genes in the human genome (Margolis et al., 1997), antibodies generated against rT NR could potentially crossreact with Q-repeat sequences of other TNR containing genes. Therefore, additional studies were performed to determine the specificity of each anti-rTNR antibody using synthetic peptide arrays spanning the first 300 amino acids of TNGW1 (Fig. 2-3D). Both the rabbit 6225 and 6226 and human anti-GWB serum 18033 recognized mu ltiple 15mer peptides including those in TNR Q-repeat domain. In contrast, all thr ee mouse anti-rTNR mAb recognized a relatively narrow set of the peptides (peptide 9-11, Table S1) resided outside of TNR Q-repeat domain. In summary, our data demonstrated that all the an ti-rTNR antibodies generated recognized specific sequence within N-terminus of TGWN1. The mo use mAb reacted highly specific to TNGW1 with a lower risk of crossreacting with other TNR-containi ng gene products. Detection of endogenous GW182 and TNGW1 pr oteins by standard western blot was challenging because the protein levels were usually very low. Hence, separating and distinguishing these high molecu lar mass proteins required careful optimization. To demonstrate the specific expression of these proteins in HeLa cells, IP-WB was carried out by using the human serum 18033, an anti-GWB serum known to contain antibodies to GW182, Ago2, and Ge-1, to enrich these protein complexes from HeLa cell lysates prior to west ern blot detection. If TNGW1 is present along with GW182 in HeLa cells as predicted from the RT-PCR data (Fig. 22), anti-GW182 antibodies will recognize two bands because TNGW1 includes the entire
30 sequence of GW182. As expected, the two rabbit pol yclonal antibodies to GW182 (5182 and 6642), and the mouse mAb to GW182 (4B6) rec ognized both forms of GW182 proteins where the faster-migrating band considered to be GW182 was the predominant isoform (Fig. 2-4A, left panel). In support of this conclusion, the an ti-rTNR mAb (2E11, Fig. 24A and 5C8, Fig. 2-4B) as well as rabbit anti-rTNR sera (6225 and 6226, Fig. 2-4B) recognized only the slowermigrating TNGW1. These data confirmed th at both TNGW1 and GW182 proteins were expressed in HeLa cells. There are three possibilities as to how TNGW1 and GW182 are expressed in cells. The first is that TNGW1 and GW182 are independently translated from two individual mRNA transcripts derived from chromosome 16 with different transcriptional start sites separated by ~60kb. The second possibility is that TNGW1 and GW182 are translated from the same mRNA with two different AUG start site s governed by respective Kozaks consensus sequences (Kozak, 1991). The third possibility is that GW182 is a post-translationally processed product of TNGW1. To address these possibilities, we designed siRNA specific for TNGW1 mRNA (siTNR) to examine the effect of suppressi ng TNGW1 mRNA on the expression of TNGW1 and GW182 protein. The knock-down effect of siTNR was initially validated by demonstrating its repression on the expression of co-transfected EGFP-TNGW1 but no effect on the expression of co-transfected EGFP-GW182 in HeLa cells (dat a not shown). Forty-eight hours after siTNR transfection, western blot analysis of cell lysates showed that only the TNGW1 band disappeared whereas the GW182 band remained the same (Fig. 2-4C, lane 1) as compared to the untreated (Fig. 2-4C, lane 4) or mock transfected cont rols (Fig. 2-4C, lane 3). These observations indicated that TNGW1 was derived from its ow n unique mRNA. However, if GW182 was posttranslationally processed from TNGW1 as c onsidered above, the GW182 band in the siTNR
31 transfected (Fig. 2-4B, lane 1) would have represented stable processed products under these experimental conditions. Contradi cting to this possibility, tran sfection of siRNA targeting the common region of TNGW1 and GW182 (siGW182) resulted in the disa ppearance of both isoforms (Fig. 2-4B, lane 2). This data s upported the conclusion that GW182 was not processed from TNGW1 and that TNGW1 and GW182 was transcribed inde pendently. In summary, our data demonstrated that TNGW1 and GW182 are dis tinct at both transcri ption and translation levels. Intracellular Localization of TNGW1 and Its Relationship w ith Other GWB Components GW182 is one of the accepted marker proteins of GWB (Eystathioy et al., 2002) and has been shown important for GWB formation (Jakym iw et al., 2005; Lian et al., 2007). Since TNGW1 shares the same amino acid sequence as GW182, except for the N-terminal domain including the TNR Q-repeat region, we were interested to examine the intracellular location of TNGW1 and its role in GW B formation. The immunofluorescence staining of mouse monoclonal anti-rTNR 2F11 on HEp2 cells showed GWB staining that was also recognized by the human anti-GWB serum 18033 (Fig. 2-5A, arrows). Notably, both mouse anti-rTNR 2F11 (Fig. 2-5A) and rabbit anti-rTNR 6226 (Fig. 2-6A) stained a subs et of about 30% GWB. In contrast, anti-GW182 mAb 4B6 stained more, although not all, GWB recognized by 18033 (Fig. 2-6B). The data implies that the amount of TNGW1 and GW182 may vary in different GWB even when they are present in GWB. We ther efore examined this hypo thesis by performing dual staining with anti-rTNR 2F11 and anti-GW182 4B6 in the same HEp-2 cells (Fig. 2-5B). Visualized by IgG subclass-specific secondary antibodies, anti-GW182 4B6 stained apparently more GWB than anti-rTNR 2F11, while all 2F11 staining colocalized with 4B6 staining. The results support the hypothesis that TNGW1 may be absent, or at leas t in very low abundance in a subset of GWB, where only GW182 is present as the predominant isoform. However, given the
32 limitation of antibodies, we could not rule out the possibility that th e immunoreactive region of these isoform were obscured by the presence of one or more additional GWB component. Nevertheless, our data indicated that TNGW1 re sided in a subset of GWB and this clearly demonstrated heterogeneity in GW B contributed by GW182 gene products. GW182 was once proposed to be a matrix protei n in GWB because it was required for the assembly of these foci (Yang et al., 2004), whic h also harbored multiple proteins including Ago2, Dcp1a, Ge-1, and RAP55 (Jakymiw et al., 2007). Therefore, we decided to determine whether the extra N-terminal domain in TNGW1 would a ffect the localization of other GWB components to GWB. Immunofluorescence assay was perfor med on HeLa cells where EGFP-Ago2 was cotransfected with either GSTGW182 or GST-TNGW1. Fig. 2-7A shows that both GST-GW182 and GST-TNGW1 are enriched in cytoplasmic foci together with EGFP-Ago2, indicating the extra N-terminal region of TNGW 1 did not interfere with the lo calization of TNGW1 or Ago2 to GWB. Furthermore, Fig. 2-7B shows that in cells transfected with EGFP-TNGW1 alone the EGFP-labeled GWB were also co-stained with anti-Dcp1a antibodies and human serum IC6 containing antibody to Ge-1 and RAP55 (Bloch et al., 2006). Notably, tr ansfected cells with either low or high expression of EGFP-TNGW1 di d not apparently affect the localization of endogenous Dcp1a and Ge-1/RAP55 to the TNGW 1-containing foci suggesting that TNGW1 could efficiently substitute for putative GW182 functions such as recruitment of Ago2 and formation of foci enriched in RNA decay factors. TNGW1 Is Not Essential for the Formation of GWB As shown in previous studies, GW 182 is e ssential for the formati on of microscopically visible GWB. However, since the existence of TNGW1 was not appreciated at that time, previous conclusions were based on the knoc k-down of both TNGW1 and GW182. Since we showed that siTNR achieved almost complete knock-down of TNGW1 without affecting the
33 level of GW182 (Fig. 2-4B), we could dete rmine whether TNGW1 knock-down affects the formation of GWB. SiTNR was transfected in to HeLa cells and the changes of GWB were monitored at day 0, 1, 2 and 3. SiGW182 was tran sfected side by side into HeLa cells as a control and the formation of GWB was monitored by co-staining with rabbit anti-Dcp1a and human anti-GWB serum 18033. Consistent with the we stern blot data (Fig. 2-4C), 2 days after transfection, the siGW182 transfection led to th e disassembly of most GWB presumably due to knock-down of both TNGW1 and GW182 (Fig. 2-8). In contrast, 2 and 3 days after siTNR transfection, microscopic GWB were still detected by both anti-Dcp1a and serum 18033. This observation demonstrated that TNGW1 was not important for GWB formation under these experimental conditions. TNGW1 and GW182 Exert Strong Repression Effect in Ago2 Mediated Translational Silencing To explore the effect of translational re pression attended by TNGW1 compared to GW182 or Ago2, we adopted the tethering assay from the wo rk of Pillai et al. (Pillai et al., 2004). An Nterminal tag N-HA (NHA) polypeptide was fused to TNGW1, GW182 and Ago2. The N tag binds the 5BoxB secondary structures harbored in the 3-UTR of the FL-5BoxB RNA resulting in a tethering effect of the ta gged protein to the 3-UTR. Th e repression effect in human 293 cells was evaluated by comparing the FL-5BoxB activities among different experimental groups relative to the untargeted RL ac tivities (Fig. 2-9A) using the me thod described in a previous study (Lytle et al., 2007). Af ter 48 hours of transfection, FL-5 BoxB activity was repressed by 46% when Ago2 was tethered to the reporter. Interestingly, teth ered TNGW1 or GW182 induced strong repression on reporter (67.6% and 65.3%, respectively) which was 46.9% or 41.3% stronger than that induced by Ago2, resp ectively (Fig. 2-10A). Comparison of the corresponding FL-5BoxB and RL mRNA levels by quantitative RT-PCR assay showed that both
34 tethered TNGW1 and GW182 was accompanied by a 23.7% and 24.5% reduction in FL-5BoxB mRNA, respectively, whereas teth ered Ago2 had an associated 50.8% reduction in FL-5BoxB mRNA (Fig. 2-10B). Therefore, the analysis of translation efficiencies of FL-5BoxB mRNA in each experimental group calculated using the formul a described in previous study (Lytle et al., 2007) showed that the tethered TNGW1 and GW182 reduced the translational efficiency of FL5BoxB mRNA (42.5% and 46.0% respectively) to a si gnificantly greater ex tent than tethered Ago2 (109.7%), which repressed the FL-5BoxB ac tivity with a lower ab undance of reporter mRNA at 48 hours (Fig. 2-10C). Previously, Pillai et al. have shown that tethering Ago2 to RL reporter with 5BoxB in 3-UTR mainly induced tr anslational repression in HeLa cells (Pillai et al., 2004). Our observations, however, indicated that tethered Ago2 repressed FL-5BoxB activity was accompanied by a reduced level of reporter mRNA in 293 cells. The discrepancy observed in RNA level may be caused by the usage of different cell lines (293 vs. HeLa), or a difference in mRNA turnover be tween FL and RL reporters. With the interesting finding that TNGW1 and GW182 exerted a stronger translational repression effect than Ago2, the next issue was to address the inte r-dependence of these proteins in translational repression. The same teth ering assay was performed in HeLa with the introduction of siRNA transfection to knock-do wn either Ago2 or GW182 prior to the cotransfection of tethering protein and reporter constructs. Surpri singly, the repression effect by NHA-Ago2 was totally abolished in cells with GW 182 knock-down (Fig 2-11A). In contrast, the repression effect of tethered GW182 or TNGW1 was not affected by knock-down of Ago2 (Fig. 2-11B). The observations that tethered Ago2 was required GW182/TNGW1 for translational suppression whereas tethered GW 182 did not require Ago2 for s uppression was reproducible when either FL-5BoxB or RL-5BoxB was used as the reporter (Fig. 2-12). Our data implied that
35 both TNGW1 and GW182 contributed further direct effect on repression than Ago2 and they were required for Ago2 mediat ed translational silencing. Discussion Expression of Human TNRC6A/GW182 Gene and Its Effect on GWB Formation In this study, we have identified TN GW1 as a novel 210kDa isoform of GW182 with both proteins highly enriched in GWB. As predicte d in the NCBI GenBank database, the amino acid sequences of TNGW1 and GW182 are identical w ith the exception that TNGW1 has an extra 253aa-polypeptide in its N-termi nus. TNGW1 was expressed in several human cancer cell lines and human testis along with the reported shor t isoform GW182. Our data demonstrated that TNGW1 and GW182 were expressed independen tly and GW182 was the predominant gene product. The expression of their mRNAs is po ssibly due to alternativ e splicing, alternative promoters and transcriptional start sites. Th e expression level of each isoform under different cellular conditions with potentially different R NAi activities and the factors determining their expression need further investigation. Both TNGW1 and GW182 were highly enriched in GWB but TNGW1 was detected only in about one third of endogenous GWB. K nock-down of TNGW1 did not noticeably disrupt GWB formation indicating that TNGW1 was not required for the formation of many GWB. Both TNGW1 and GW182 exerted similar translationa l repression effect in the tethering assay. These data implied that TNGW1 might be f unctionally redundant and GW182 plays a more important role in the formation of GWB. Howeve r, this interpretation ne eds to be solidified by determining the effect of TN GW1 on GWB formation in the absence of GW182. Since an efficient method to knock-down GW182 without affec ting TNGW1 is not currently available, it remains unclear whether TNGW1 alone can s ubstitute for all putative function for GW182. Given that TNGW1 contains the whole amino aci d sequence of GW182, it is likely that TNGW1
36 is capable of most functiona l characters of GW182. With the extended N-terminal region containing TNR Q-repeat domain, the full functi onal characteristics for TNGW1, in addition to GW182, remain to be determined. Interdependence of Ago2 and TNGW1/GW 182 in miRNA-Mediated Translational Repression In mammalian system, Ago2 is considered the most important factor in the RISC because it binds siRNA and miRNA as we ll as being the only factor ha rboring the slicing activity responsible for siRNA-induced silencing (Liu et al., 2004; Yuan et al., 2005). However, the mechanism for miRNA-mediated repression remain s unclear. In the presence of incomplete complementarity between anti-sense strand miR NA and its target mRNA, the slicing function of Ago2 is interfered, where Ago2 may need to recruit multiple factors to secure its repression effect on the targeted mRNA a nd possibly induce the subseque nt mRNA degradation. GW182 has been reported as more important in gene silencing when slicing activity is limited, for example, in miRNA-mediated silencing (Liu et al., 2005a; Chu and Rana, 2006). In the present study, we extended the understand ing of their interdependenc e by tethering TNGW1, GW182, or Ago2 to the 3-UTR of a lucife rase reporter mRNA and comparing their relative po tentials in translational repression. Our data demonstrat ed that both TNGW1 and GW182 exerted stronger translational repression than Ago2. Furthermore, the repression effect of tethered Ago2 was sensitive to the presence of TNGW1 and/or GW182, while repr ession by tethered TNGW1 or GW182 did not require Ago2. These observations suggested that either TNGW1 or GW182 has a more direct impact on translational repressi on than Ago2. Although it is possible that other Argonaute proteins could substitute for Ago2 when it was knock-down, the functional importance of TNGW1 and/or GW182 was obvious even though human cells may also have two homologues of GW182, TNRC6 B and TNRC6C. In miRNA-mediated translational silencing,
37 the miRNA loaded in Ago2 can direct GW182 to their targeting mRNA and repress its translation. Hence, when TNGW1 or GW182 was tethered to the reporter mRNA via N tag, it may bypass the requirement of miRNA-Ago2 guida nce. Consistent with this observation, Eulalio et al. have shown that the inter action between GW182 and Argonaute proteins was essential for miRNA mediated translational re pression and mRNA decay in Drosophila (Eulalio et al., 2008b). However, whether the repression effect is caused directly by TNGW1 and/or GW182 or the additional factors recruited in later stages of the process requires further investigation. In summation, our data suggest the followi ng scenario with a putative order in the sequence of events. With the guidance of miRNA, Ago2 is able to target to the 3-UTR of a specific mRNA. GW182/TNGW1 is subsequently en riched to the 3UTR due to its interaction with Ago2. Once GW182/TNGW1 is brought to th e 3-UTR, the translational suppression is triggered either directly by GW182/TNGW1 or by ot her factors further recr uited to the complex. In the tethering assay, tethered GW182/ TNGW1 did not require miRNA and Ago2 for translational suppression because their function is substituted by th e interaction of the N tag and BoxB sequence. As more studies of translatio nal repression are reported (Filipowicz et al., 2008; Eulalio et al., 2008a), the importance of GW182 in each step may be further characterized in future studies. The Functional Differences of GW182 Iso forms and the Heterogeneity of GWB It is possible that TNGW1 and GW 182 ar e redundant protein products of the human TNRC6A gene in some aspects of translational re pression process. Our data showed that they both formed cytoplasmic foci that colocalized with other RNAi related and mRNA decay factors. As demonstrated by the functional assays test ed in the current study, both isoforms induced translational repression and mRNA degradation to a similar extent. However, the hypothetical
38 redundancy between TNGW1 and GW182 may be due to our limited understanding of their functions. A similar case could also be made for the human Argonaut protein family that is comprised of at least 4 Argonaut e proteins (Ago1-4) that share over 90% sequence similarity. Except for Ago2, known as the catalytic engine of RISC (Liu et al., 2004), the biological functions and significance of Ago1, 3 and 4 are not well understood. One distinguishing feature of TNGW1 is that it is only localized to a subset of GWB in HEp-2 cells. This specific localization of T NGW1 is likely related to its unique N-terminal polypeptide domain that is not found in GW182. Th is N-terminal domain may be responsible for interacting with or recr uiting protein factors to help with tr anslational suppressi on. It is also possible that this unique N-terminal domain can affect the protein folding of TNGW1 and somehow interfere with its in teraction with Argonaute protei ns. The fact that TNGW1 functional capacity is similar to GW182 makes it important to fu rther investigate the potential differences in their functions. Our data confirmed there was heterogeneity in GWB in terms of TNGW1 and GW182 distributions, which is consistent with the observations from recen t reports (Jakymiw et al., 2007; Moser et al., 2007). The importance of this he terogeneity could be a reflection of different stages of GWB assembly. However, it could also reflect the diversity in the functional status of GWB, which are closely related to miRNA-mediat ed function. When targeted by miRNA, most mRNAs may enter the accelerated turnover proc ess whereas some may not. A recent report showed that two luciferase reporter mRNAs carrying different 3-UTR were degraded at different rates when both were translationally repressed (Eulalio et al., 2008b). Under stress conditions, the translational effi ciency of some miRNA targeted mRNAs were reported to be upregulated (Bhattacharyya et al., 2006a; Vasudeva n et al., 2007), which required these targeted
39 mRNA to remain stable during miRNA-targeting. Interestingly, the determining factors in the turnover of targeted mRNA seemed not only de pend on the miRNA targeting sequence but also on the proteins that were recr uited during the silencing process (Bhatt acharyya et al., 2006b). Since the formation of GWB was shown to be a consequence of miRNA ac tivity (Pauley et al., 2006; Eulalio et al., 2007b), it is not surprising th at there are multiple functional roles for these cytoplasmic foci. The identif ication of the TNGW1 as a novel isoform of GW182, and the heterogeneous distribution of TNGW1, GW182, and other RNAi factors in GWB (Jakymiw et al., 2007), will provide us a better understanding of the RNAi process at molecular cell biology level. *This work was published in Journal of Cell Science, 2008 Dec 15;121(Pt 24):4134-44.
40 Figure 2-1. Schematics of the human TNRC6A ge ne products GW182 and TNGW1. A) GW182 was reported as a 182 kDa protein produc t of 1709 amino acids and containing 3 GW-rich regions, a Q/N-rich region and a classical RNA recognition motif (RRM). TNGW1 is the novel 210 kDa GW182 isofor m with an extra N-terminal 253-amino acid polypeptide containing a stretch of glutamine-repeat (Q-repeat) that are translated from the CAG trinucleotide repe at (TNR). B) TNRC6A gene resides in human chromosome 16p11.2. The 5 end corresponding to the mRNA of these 2 isoforms separates ~60kb on chromosome 16. C) The TNR Q-repeat domain was encoded from the 5th exon and the co rresponding nucleotide and amino acid sequences are shown in this panel. D) Alignment of N-terminal TNGW1 sequences from mouse, rat, and human using Cl ustalW. The result showed sequence conservation within these three species with some degree of diversity (yellow highlight) in the TNR Q-repeat region.
41 Figure 2-2. TNGW1 mRNA containing the TNR exon detected in human testis and different cell lines using RT-PCR. Primer set fla nking the TNR region of TNGW1 mRNA (3621626nt) amplified 1.2 kb bands from cDNA samp les of human testis tissue, HEp-2, HeLa and HepG2 cells. The amplified 1.2 kb products from HEp-2, HeLa and human testis tissue were subsequen tly purified and their consensuses to reference sequence were confirmed by direct DNA sequencing.
42 Figure 2-3. Production and characte rization of polyclonal and monoclonal antibodies specific to the TNGW1 isoform. A) Rabbit polyclona l anti-rTNR antibodies strongly reacted with rTNR polypeptide coated in addressa ble laser bead immunoassay. Compared to 2 rabbit anti-GW182 antibodies (5182 and 6642) and the pre-immune rabbit sera, rabbit anti-rTNR sera 6225 and 6226 strongly reacted with rTNR-coated beads. B) Rabbit anti-rTNR antibodies recognized only TNGW1 but not GW182 in western blot. HeLa cells transfected with EGFP-tagged TNGW1, GW182, TNR and EGFP vector were harvested 24 hours after transf ection. These cell lysates together with recombinant 6XHis-tagged TNR were analyzed by western blot using rabbit antirTNR sera 6225 and 6226. The membranes shown in column a and c were first blotted with rabbit anti-rTNR antibodies, stri pped, and then re-blo tted with anti-EGFP antibodies to confirm the recombinant pr oteins were expressed (column b and d, respectively). C) Reactivity of mouse anti-rTNR mAb with rTNR polypeptide in addressable laser bead immunoassay. Three mouse mAb 2E11, 5C8 and 2F11 showed strong MFU to rTNR, whereas control mAb anti-GW182 4B6, anti-GW2-25 and culture medium showed little or no reactivity. Note that mAb 2E11 and 5C8 showed ~20 folds higher MFU than 2F11. D) Peptide array mapping of antibody reactivity to TNR region. Two duplicate peptide array membranes containing 59 sequential 15-amino acid peptides, each with 5-amino acid-offset, spanning the TNR region were synthesized and used to ex amine antibody specificity. Membrane #1 (left) was probed sequentially with an ti-EEA1control mAb, anti-rTNR mAb 2E11, 2F11 and rabbit anti-rTNR 6225. Membrane #2 was probed sequentially by antiGW182 mAb 4B6, anti-rTNR mAb 5C8, ra bbit anti-rTNR 6226 and human serum 18033. The black box shows that the peptide strings reside in TNR Q-repeat domain as indicated by underline in Figure 2-1C All rabbit and human sera showed reactivity to multiple peptides. The 3 m ouse anti-rTNR mAb recognized relatively narrow region outside of the TNR Q-repeat domain.
43 Figure 2-4. TNGW1 and GW182 were independen t products of human TNRC6A gene. A) Expression of both TNGW1 and GW182 were detected in HeLa cells. GWB components were enriched by IP using human anti-GWB serum 18033 and analyzed by western blot using rabbit anti-GW 182 antibodies 5182 and 6642, mouse antiGW182 mAb 4B6 and anti-rTNR mAb 2E11. All 3 anti-GW182 antibodies recognized 2 bands migrating about 210 and 180 kDa, where anti-rTNR 2E11 only recognized the 210 kDa band regarded as TNGW1, the novel isoform of GW182. B) TNGW1 was detected by multiple anti-rTNR antibodies. 18033 IP samples were also examined by rabbit polyclonal antibodies 6225, 6226, and mouse mAb 5C8 generated to the N-terminal domain rTNR. A cons entient 210 kDa band was detected by each of these antibodies regarded as T NGW1. C) TNGW1 knockdown using siRNA specific to the TNGW1 did not affect the level of GW182 in HeLa cells. HeLa cells were transfected with siRNA specific to TNGW1 (siTNR) or both forms (siGW182) using lipofectamine 2000 (LP 2000). After 48 hours of transfection, cells were lysed and analyzed by western blot using ra bbit anti-rTNR serum 6225 or anti-GW182 serum 5182. Compared to mock transf ection using lipofectamine 2000 (LP2000) and untreated control groups, siGW182 knock-down both TNGW1 and GW182 whereas siTNR only knock-down TNGW1.
44 Figure 2-5. TNGW1 resided in a subset of GW B (part1). A) Mouse monoclonal anti-rTNR antibody 2F11 stained a subset of GWB recognized by prototype serum 18033. HEp2 cells were stained with mouse mAb 2F11, which. stained a subset of GWB detected by serum 18033 (arrows) but did not recognize other GWB (arrowheads). Culture medium was used as a negative control. Bar, 10m. B) Mouse monoclonal antirTNR antibody 2F11 stained a subset of GWB recognized by monoclonal antiGW182 antibody 4B6. HEp-2 cell staining was performed similarly with anti-rTNR 2F11 (IgG2a) and anti-GW182 4B6 (IgG1) a nd different fluorochrome conjugated mouse IgG subclass specific secondary antibodies to evaluate the distribution of TNGW1 in GW182-positive GWB. Anti-rTNR 2F11 stained some of the foci detected by 4B6 (arrows) but not others (a rrowhead). The second and third row of images show controls with either 4B6 or 2F11 and stained with both secondary antibodies to demonstrate secondary anti body specificities. Th e asterisk shows nonspecific nucleolar staining detected by th e anti-IgG2a secondary antibodies. Bar 10m.
45 Figure 2-6. TNGW1 resided in a subset of GWB (p art 2). A) Rabbit polyc lonal anti-rTNR serum 6226 stained a subset of GW B in HEp-2 cells. Anti-rTNR (Post-6226) stained a subset of GWB recognized by human serum 18033 (arrows) whereas the pre-immune serum (Pre-6226) from the same rabbit did not have any GWB staining. Note some GWB (arrowheads) are not recognized by anti-rTNR. The diffuse nuclear and cytoplasmic staining for rabbit 6226 was not related to TNGW1 because similar staining was observed with the pre-immune serum. Bar, 10m. B) Anti-GW182 mAb 4B6 co-stained the majority of GWB recognized by human anti-GWB prototype serum 18033 in HEp-2 cells. Culture medium was used as a negative control. Bar, 10m.
46 Figure 2-7. Intracellular locali zation of TNGW1 with other GWB components. A) Both TNGW1 and GW182 colocalized to GWB with transfected Ago2. HeLa cells transfected with either GST tagge d GW182 or TNGW1 (red) and EGFP-Ago2 (green) were fixed and analyzed by indire ct immunofluorescence assay 24 hours after transfection. Both TNGW1 and GW182 form ed cytoplasmic foci and colocalized with Ago2. Bar, 10m. B) Transfect ed TNGW1 colocalized with RNA decay factors enriched in GWB. HeLa cells transfected with EGFPTNGW1 were fixed 24 hours after transfection and analyzed by i ndirect immunofluorescence assay using rabbit anti-Dcp1a (red) and human serum IC6 (magenta, recognizes Ge-1, RAP55 and unrelated nuclear envelop protein). The transfected EGFP-TNGW1 formed cytoplasmic foci that were co-stained by both anti-Dcp1a and IC6 antibodies. Bar, 10m.
47 Figure 2-8. Knockdown of TNGW1 has no apparent effect on the assembly of GWB in HeLa cells. HeLa cells were transfected with siTNR or siGW182 and then harvested at 0, 1, 2 or 3 days after transfection. Indir ect immunofluorescence assay was performed by using rabbit anti-Dcp1a (red) and human anti-GWB serum 18033 (green) to examine the effect of either siRNA on GWB formation. Bar, 10m.
48 Figure 2-9. Dual luciferase assay measurement a nd NHA constructs expression in 293 cells. A) Absolute reading of either FL or RL activ ities in the tethering assay performed in 293 cells (Fig. 2-10). Error bars show standa rd deviations. B) Western blotting data showed the expression of each NHA tagge d construct in th e tethering assay.
49 Figure 2-10. Tethered TNGW1 and GW182 exerted tr anslational repression to a greater extent than Ago2. A) Both tethered TNGW1 and GW182 exerted stronger repression effect than tethered Ago2. Assays for both luci ferase activities were performed 48 hours after transfection. Tethered TNGW 1 and GW182 showed 67.6% and 65.3% translation repression effect to FL-5BoxB reporter, which was significantly higher than tethered Ago2 respectively (asteris k P<0.05, t test). Measured FL-5BoxB activities were normalized to corresponding RL activities. All translation repression effects were estimated by the differences of FL-5BoxB/RL activity compared to that in NHA vector transfected group. Error bars represent standard deviations. B) Tethered TNGW1 or GW182 induced less re porter mRNA reduction than tethered Ago2. The mRNA levels from the same assay were measured by SYBR-green quantitative real time PCR. The degradatio ns of FL-5BoxB mRNA were determined by the reductions of the mR NA level between experimental groups and NHA vector transfected group. All FL-5B oxB mRNA levels were nor malized to RL mRNA to minimize the experimental errors. C) Tethered TNGW1 and GW182 strongly reduced the translation efficiency of the reporter compared to tethered Ago2. Translation efficiencies of FL-5BoxB in different groups were calculated by the ratio of relative FL-5BoxB activitie s to their mRNA levels. In 293 cells, tethered Ago2 did not significantly (ns, t test) alter the FL -5BoxB reporter translation efficiency. However, tethered TNGW1 and GW182 reduced the translation efficiency FL-5BoxB reporter by 57.5% and 54.0% respectively (a sterisk, P<0.01, t test). Error bars represent standard deviations.
50 Figure 2-11. Translational repression of tethered Ago2 required endogenous TNGW1 and/or GW182. A) The repression effect of tethered Ago2 was abolished when TNGW1 and GW182 were knock down while tethered TNGW1 or GW182 maintained repression in the absence of Ago2. HeLa cells were transfected with different siRNA 24 hours prior to the transfection of NHAtagged constructs and RL-5BoxB /FL reporters. The translati on repression effect related to tethered Ago2, TNGW1 or GW182 was determined by reduction of RL activity and the repression effect in siRNA to EGFP (siGFP) transfected group was normalized as 1. The repression effect caused by NHA-Ago2 was abolis hed when cells were treated with siGW182 (asterisk, P<0.01, t test). Knock-down of Ago2 (siAgo2) caused no significant (ns, t test) effect on tethered GW182 or TNGW1 induced tran slational repression. Error bars represent standard deviations. B) Tethered Ago2-medi ated gene silencing required TNGW1 and/or GW182 regardless either FL-5BoxB or RL-5BoxB was used as targeted reporter. Similar experiments were performed as described previously except for the use of RL-5BoxB as reporter together with FL reporter as co-trans fection quantitative control. The repression effect of tethered Ago2 was abolished consis tently when FL-5BoxB or RL-5BoxB was used as reporter while tethered GW182 maintained its repression to both reporters in the absence of Ago2.
51 Figure 2-12. Quantitative real time PCR showed corresponding siRNA knock-down effect in the tethering assay. The mRNA levels of Ago2 and TNGW1/GW182 were determined by Taqman real time RT-PCR after siRNA treatment in the tethering assay (Figure 2-11). Both Ago2 and TNGW1/GW182 we re reduced over 60% 80 hours after siRNA treatment.
52 CHAPTER 3 THE C-TERMINAL HALF OF AGO2 BINDS TO MULTIPLE GW-RICH REGIONS OF GW 182 AND REQUIRES GW182 TO MEDIATE SILENCING Introduction MicroRNA (m iRNA)-mediated gene silencing is an important post-transcriptional regulation which controls in part the half lives of mRNA targets. In this regulation, miRNA binds to the 3-UTR of mRNA leading to translational inhibition, mRNA degradation, and mRNA sequestration (Nilsen, 2007). MiRNAs are evolutionarily c onserved in most of species and are estimated to regulate ~30% of proteinencoding genes in human (Lewis et al., 2005; Filipowicz et al., 2008). There are four human Ago proteins that include Ago1 to Ago4, which are the core components of silencing effecter complexes and are known to bind single-stranded miRNA. These Ago proteins share greater than 80% iden tity and are primarily characterized by PAZ and PIWI domains. The PAZ domain contains a bi nding pocket for the 3 overhanging nucleotides of miRNA. Interestingly, despite highly c onserved sequences, only the PIWI domain of Ago2 harbors RNase H-type activity and, therefore, Ago2 also functions in siRNA-mediated slicing of mRNA targets (Liu et al., 2004; Yuan et al., 2005) Tethering Ago proteins to the 3-UTR of mRNA mimicked miRNA function and effected translational repression (Pillai et al., 2004). GW182 is important for miRNA-mediated transl ational silencing and interacts with Ago2. GW182 contains several glycine/tr yptophan-rich (GW-rich) regions a glutamine/asparagine-rich (Q/N-rich) domain, and a C-terminal RNA rec ognition motif (RRM) (Eystathioy et al., 2002; Decker et al., 2007). GW182 localized to and was essential for the fo rmation of GW bodies (GWB, also known as mammalian P bodies) (Ya ng et al., 2004; Schneider et al., 2006), cytoplasmic structures closely linked to mRNA decay (Sheth and Parker, 2003; Eystathioy et al., 2003c) and the miRNA/siRNA pathway (Jakymiw et al., 2005; Lian et al., 2006; Pauley et al.,
53 2006; Lian et al., 2007). K nockdown of GW182 greatly impair ed miRNA-mediated gene silencing and subsequent mRNA degradation (Rehwinkel et al., 2005; Liu et al., 2005a). Interestingly, Ago proteins, miRNAs, and mRNAs targeted by miRNAs all colocalized with GW182 in GWB (Jakymiw et al., 2005; Pillai et al., 2005; Sen and Blau, 2005; Liu et al., 2005b). Furthermore, GW182 interacted with A go2 and this interaction was conserved from plants to human (Ding et al., 2005; Jakymiw et al., 2005; Liu et al., 2005a; Behm-Ansmant et al., 2006; El-Shami et al., 2007; Till et al., 2007). However, the ro le of GW182 and the importance of GW182-Ago2 interaction in translational repression remain unclear. In the current study, we mapped the GW182-Ago2 interaction and investigated the possible role of this interaction in miRNA-mediated silencing in human. Materials and Methods Construction of Deletion Co nstructs of GW182 and Ago2 The details of constructing GW 1 1 (aa254-751, formerly known as GW182 1), TNR (aa1-204), Ago2 (aa1-860) and PIWI (aa478-860) were described previously (Jakymiw et al., 2005; Li et al., 2008). GW1 1a (aa254-503), GW1 1b (aa502-751), GW1 7 (aa1034-1962), GW1 12 (aa895-1211) and GW1 1a truncated constructs (W 1-2, W2-3, W-3-4, W4-5) were cloned from cDNA of full length GW182 or its deletion constructs by P CR (see Table. 3-1 for information of primers and conditions). Th e human Ago3 mutant (A go3m) in pCMV-SPORT6 vector was obtained from Invitrogen (Carlsba d, CA, Clone number: CS0DB008YP10). Ago3m sequence was cloned by PCR (see Table. 3-1 for primers and condition) to adapt to Gateway cloning system (Invitrogen). The products from the above PCR reactions were then cloned into pDONR207 (Invitrogen) using the Gateway BP recombination reac tion as per the manufacturers instructions (Invitrogen). To construct pENTR-GW1 5 (aa1670-1962), pENTR-GW182 was
54 digested with Sal I (5 end linker) and Spe I (nt5008) to release a 4.2Kb vector fragment containing the C-terminus of GW182. The overhangs of this fragment were filled in and then ligated. To construct pENTR-GW1 10 (aa566-1343), phrGFP-KIAA 1460 was digested with XhoI (5 end linker) and Sma I to release a 2.3Kb fragment, which was then subcloned into the Sal I and EcoR V sits of pENTR2B (Invitrogen). The cDNA of GW1 1a mutant (tryptophan > alanine, W>A) was directly synthesized and subcloned in pUC57 vector by GenScript Crop (Piscataway, NJ) using the sequence as shown in Fig. 3-8. The mutated gene sequence was moved to pDONR207 vector (Invitrogen) usi ng Gateway BP recombination for further applications. To construct pENTR-PAZ ( aa1-480), pENTR-Ago2 was digested with XhoI (nt1467) and XhoI (3 end linker) to generate the vector fragment containing N-terminal half of Ago2, which was purified and then ligated. To construct pENTR-Ago1, EST clone pBluescript hAgo1 was first digested with BamHI (3 end linker) and the overhang was filled in to generate a blunt 3 end. Then the digested product was cut by Kpn I (5 end linker) to generate a 4.0 kb fragment which was subcloned into the Kpn I and EcoR V sites of pENTR1A (Invitrogen). To construct pENTR-Ago4, EST clone pBlu escript hAgo4 was digested with Sma I (5 end linker) and Sca I (3 end linker) to generate a 3.5 Kb frag ment which was then subcloned into the Dra I and EcoR V sites of pENTR1A (Invitrogen). All of the variants used in current study were subcloned into Gateway compatible GST, GFP, 3xFlag or NHA (Li et al., 2008) vectors by using Gateway LR recombination reaction (Invitrogen) pIreSneo-Flag/HA Ago3 was obtained from Thomas Tuschl (Meister et al ., 2004) through Addgene. The tether ing assay plasmids including pClneo-NHA vector, NHA-Ago2, Renilla luciferase RL-5BoxB and FL were kind gifts from Dr. Witold Filipowicz, Friedrich Miescher Institute for Biomedical Research, Switzerland (Pillai et al., 2004). All DNA constructs used in this st udy were confirmed by direct DNA sequencing.
55 Antibodies Rabbit anti-Ago2 and rabbit anti-GS T were gifts from Dr. Tom Hobman (University of Alberta, Edmonton, Canada) and Dr. Peter Sayeski (University of Florida, Gainesville, USA), respectively. Mouse monoclonal anti-HA was purchased from Covance (Emeryville, CA). Mouse monoclonal anti-Flag M2 and anti-tubulin were purchased from Sigma-Aldrich (St. Louis, MO). Rabbit polyclonal anti-GFP was purchased from Invitrogen Corporation (Carlsbad, CA). Plasmid Transfection, GST Pull-do wn, and Western Blot Analysis HeLa cells were cultured in DMEM c ontaining 10% fetal bovine serum in a 37oC incubator with 5% CO2. HeLa cells were grown to 90-100% in 6-well plat e at the day of transfection. GST-tagged constr uct was singly transfected or co -transfected with other tagged constructs into HeLa cells us ing Lipofectamine 2000 (Invitrogen ) as per the manufacturers instructions for 24 h. For detection of expressi on of GST-tagged proteins in whole cell lysate, the cells were lysed in Laemmli sample buffer di rectly. For GST pull-dow n assay, the cells were lysed with NET/NP40 buffer (150mM NaCl, 5m M EDTA, 50mM Tris, pH 7.4, 0.3% NP40) with Complete Protease Cocktail Inhibitor (Roche Di agnostics) and then sonicated at 20% amplitude for 10 sec for 3 times on ice. The GST pull-dow n assay in Figure 3-4 was performed under conditions that followed a publishe d protocol (Till et al. 2007). Afterwards, the lysates were centrifuged at 13,200 rpm for 5 min. The pellets (insoluble fract ions) were lysed in Laemmli sample buffer directly. The soluble fractions were incubated with Glutathione SepharoseTM 4B (GE Healthcare) and mixed at 4oC for 2 h for GST pull-down. After the incubation, the beads were washed with NET/0.3% NP40 buffer for four times and the samples eluted in Laemmli sample buffer. The soluble fraction of cell ly sates (input), GST pull-down samples, whole cell lysates, and insoluble fractions were separated on 10% polyacr ylamide gel and transferred to
56 nitrocellulose. Western blotting was performed as described previ ously (Lian et al., 2007). The dilutions of primary antibodies were: 1:1000 for anti-GST, 1:400 for anti-Flag, 1:1000 for antiGFP, and 1:500 for anti-Ago2, 1:1000 for anti-HA. Indirect Immunofluorescence Cells were fixed and perm eabilized as describe d previously (Jakymiw et al., 2005; Lian et al., 2007). The dilution of anti-Flag antibody was 1:1000. Tethering Assay Using a Du al Luciferase Sys tem HeLa cells were grown to about 90~100% c onfluence in 24-well plate at the day of transfection. To determine the effect of tethering PIWI (aa478 860) and PAZ (aa1480), the cells were transfected with 0.1ng of constructs expressing reporter Re nilla luciferase (RL5BoxB), 100ng of control firefly luciferase (FL) plus 700ng of NHA tag, NHA-Ago2, NHAPIWI, or NHA-PAZ using Lipofectamine 2000 (Invitrogen) for 48 h as per the manufacturers instructions. Cells were harves ted 48 hour after tran sfection and the FL and RL activities were measured using Dual-Luciferase Reporter Assay System (Promega, Madison, WI) following the manufacturers protocol. Relative luciferase ac tivities (ratio of targeted luciferase activities over control luciferase ac tivities) were calculated as descri bed previously (Pillai et al., 2004). Briefly, FL/RL activity in NHA vector transf ected (control) group was regarded as 0% translational repression. The repr ession levels of other experime ntal groups were calculated by the percentage reduction of relative luciferase activities compared to that in NHA control group. The assay was performed in triplicates and repeated for 2 to 3 times. To detect the expressions of tethered NHA-tagged proteins, the above cell lysates were mi xed at 1:1 ratio with Laemmli sample buffer and western blot was performed as described in Western Blot Analysis of Materials and Methods.
57 RNA Interference and Quantitative Real Time PCR The sequence of siRNA for GW 182 or for GFP was described previously (Lian et al., 2007). To determine how GW182-knockdown affect s the Ago2or PIWI-mediated suppression, cells were grown to 30-50% confluence and we re transfected with 100 nM of siRNA for GW182, or siRNA for GFP as a negative control using Lipofectamine 2000 (Invitrogen). Thirty hours later, these cells were tr ansfected again with constructs expressing reporter RL-5BoxB, control FL and NHA tagged constructs as described above. Fortyeight hours after the second transfection, total RNA was extracted from HeLa cells using RNeasy Mini Kit (Qiagen, Valencia, CA). RNase-Free DNase Set (Qiagen) was applied to eliminate potential DNA contamination. The relative mRNA level of GW182 was measured in duplicate using Ct method (Livak and Schmittgen, 2001) and TaqMan Fast Universal Master Mix (Applied Biosystems) with the corresponding TaqM an Gene Expression Assay (TNRC6A, Hs00379422_m1, Applied Biosystems). The level of 18S rRNA was measured as internal control (18S rRNA, 4310893E, Applied Biosystems ). The melting curve in each individual measurement was monitored to guard against non-specific amplification. Results GW182-Ago2 Interaction Was Important for the Localization of Ago2 in Cytoplasmic Foci Previous studies have shown that Ago2 colo calized with GW182 in cytoplasm ic GWB and GW182 was essential for the formation of these fo ci (Yang et al., 2004; Jakymiw et al., 2005). However, the driving force for the localizat ion of Ago2 to GWB remains unknown. Based on the above data that the GW182 fragment GW1 10 formed insoluble complexes with Ago2 and PIWI, but not with PAZ, we hypothesized that the GW182-Ago2 interaction is crucial for Ago2 to localize to GWB. To examine this hypothesis, Flag-Ago2, -PIWI or -PAZ was co-expressed with GFP-GW1 10 in HeLa cells and Flag-PIWI or -PAZ were expressed alone as controls (Fig.
58 3-2). Flag-Ago2 was shown to colocalize with GFP-GW1 10 in cytoplasmic foci whereas singly expressed Flag-PIWI or -PAZ were diffusely distributed in the cytoplasm (Fig. 3-2, e, g). Interestingly, co-expression of GFP-GW1 10 with Flag-PIWI dramatically changed the distribution of Flag-PIWI, whic h was recruited to cytoplasmic foci and colocalized with GFPGW1 10 (Fig. 3-2, b, f, k). In contrast, co-expressing GFP-GW1 10 with Flag-PAZ did not recruit the diffusely distributed Flag-PAZ to cytoplasmic GFP-GW1 10-positive foci (Fig. 3-2, c, h, m). This data supported that the contention that interacti on of GW182 with the C-terminal half of Ago2 mediated the localization of Ago2 to GWB. Ago2 Bound to Multiple Non-Overlapping GW-Rich Regions of GW182 Since the G W182 fragments, GW1 1 and GW1 10, were both shown to bind Ago2 and these two fragments have overlapping 186aa, it is possible that the ove rlapping region of GW182 (aa566-751) is the primary site for the GW182-Ago2 interaction. To examine this possibility, deletion constructs GW1 1a (aa254-503) and GW1 1b (aa502-751) were generated with the latter covering the overlapping region of GW1 1 and GW1 10 (Fig. 3-1A). In addition, other deletion constructs GW1 7 (aa1034-1962) and GW1 5 (aa1670-1962) were used to investigate whether regions of GW182 other than GW1 1 and GW1 10 bound Ago2. GFP-GW1 1a, GW1 1b, -GW1 7, or -GW1 5 was co-expressed with GST-ta gged Ago2 fragment PIWI in HeLa cells and a GST pull-down as say was performed to examine the interaction. As a negative control, GFP-GW1 1 was co-expressed with GST-tagged fr agment N1, the N-terminal aa51-779 of a completely unrelated protein hZW10 (Fam ulski et al., 2008). Unexpectedly, GFP-GW1 1a, -GW1 1b, -GW1 7, and -GW1 5 were all co-precipitated with GST-PIWI (Fig. 3-3A, lanes 79, 14). Another GW182 tr uncated construct GW1 which contains the reported orthologconserved GW-rich region (aa1074-1144) (Till et al. 2007), was also able to co-precipitate Ago2
59 (Fig. 3-3B). Interestingly, GW1 1a, GW1 1b, GW1 and GW1 5 are non-overlapping fragments and thus this data s howed that at least 4 separate regions of GW182 could bind Ago2. Moreover, the deletion constr ucts that bound Ago2 all contained a GW-rich region whereas TNR, the only deletion construct that did not bi nd Ago2, lacked a GW-rich region. In summary, the GW182 deletion constructs containing GW-rich regions all bound to the C-terminal half of Ago2 indicating that multiple regions of GW182 mediated the interac tion of GW182 with Ago2 and that GW repeats might be an key element for Ago2-binding. Tryptophan Residues of GW1 1a Were Not Required for Interaction With Ago2 Previous studies from two groups have shown that short synthetic peptides containing one to two WG/GW was able to intera ct with Ago protein (El-Shami et al. 2007; Till et al. 2007). The interaction was significantly reduced when either one of the tryptophans was mutated to alanine (Till et al. 2007). We noted that GW1 1a, unlike GW1 1b and GW1 5, lacks sequence homology with the reported ort hologue-conserved sequence (see Disc ussion). It is possible that the interaction of GW1 1a with Ago2 is different from that of the other GW182 truncated constructs. To examine if any of the 5 tryptophans in GW1 1a play important roles in Ago2binding, four sequential truncated constructs were designed to span the sequence of GW1 1a and each containing two tryptophans (Fig. 3-4A). Interestingly, all these truncated constructs were co-precipitated by GST-PIWI (Fig. 3-4B). In addition, to examine whether tryptophan is essential for the GW1 1a-PIWI interaction, all of the 5 tryptophans in GW1 1a were mutated to alanine (W>A) (Fig. 3-4A). Surp risingly, the W>A mutation of GW1 1a did not abolished its ability to co-precipitate with PIWI frag ment (Fig. 3-4C). In summary, the GW1 1a fragments containing two tryptophans all bound to the C-te rminal half of Ago2. However, loss of
60 tryptophan in GW1 1a did not disrupt the GW1 1a-PIWI association, imp lying that tryptophan might not be an essential feature for me diating the interactio n of GW182 with Ago. The Interaction of Ago2 With GW182 Was Conserved In Other Human Ago Proteins There are four Ago proteins in hum an cells th at share a high degree of sequence similarity. To examine whether Ago1, Ago3, and Ago4 al so interact with GW182, GFP-Ago1, -Ago3, Ago3m, or -Ago4 was co-expressed w ith GST-tagged GW182 fragments GW1 1 or GW1 10 in HeLa cells and a GST pull-down assay was perfor med. Ago3m is a splici ng variant of Ago3 and is missing aa757-823, the C-terminal 66aa of PIWI domain. Interestingly, human Ago1, Ago3, and Ago4 bound GW1 1 and GW1 10 (Fig. 3-5A, B). Ago3m did not bind GW1 1 or GW1 10 indicating that the C-terminus of that cognate PIWI domain was required for the binding to GW182 (Fig. 3-5A, B). Notably, both GFP-Ago1 (Fig. 3-5A, lane 4) and Flag-Ago1 (Fig. 3-5B, lane 4) bound GST-GW1 1 demonstrating that different N-terminal fusion tags did not affect the binding of Ago1 with GW182. In summary, the in teraction of human GW182 with Ago2 was observed with other human Ago proteins and the C-terminal region of PIWI domain was critical for the inte raction of GW182 with Ago3. Tethering C-Terminal Half of Ago2 to the 3-UTR of mRNA Recapitulated Ago2-Mediated Silencing Which Required the P resence of GW182 It was reported that tetheri ng Ago2 to the 3-UTR of mRNA causes repression of protein synthesis (Pillai et al., 2004). Because the C-terminal half of Ago2 bound GW182 whereas the N-terminal half of Ago2 did not we examined whether the C-term inal half of Ago2 was able to mediate silencing when tethered to the 3-UTR of mRNA. The dual lu ciferase and tethering assay was used as described previously (Pillai et al., 2004). In this assay, the reporter Renilla luciferase (RL) contains five 19-nt BoxB hairpin structures in the 3-UTR of its mRNA (RL5BoxB). The N peptide, which is derived from phage and binds to BoxB structures with high
61 affinity (Legault et al., 1998), was fused to th e N-terminus of HA tagged Ago2, PIWI, or PAZ. In this way, Ago2, PIWI, or PAZ was brought directly to the 3-UTR of mRNA bypassing the requirement for miRNA. Interestingly, teth ered PIWI was attended by almost as much repression as tethered full-length Ago2 (Fig. 3-6A). In contrast, tethered PAZ was totally devoid of the repression function of Ago2 (Fig. 3-6A). This data indi cated that the functional domain mediating silencing lay within the C-terminal half of Ago2. The repression effect of other human Ago proteins was also examined. Both Ago1 and Ago4 exerted a similar repression effect on the reporter as Ago2 and PIWI (Fig. 3-6A). Interestingly, Ago3m, which lost the interaction with GW182, was not able to induce repression. Thes e data supported the conclusion that the repression effect mediat ed by the tethered construct might be associated with its interaction with GW182. To examine whether GW182 is required for Ago2or PIWI-mediated repression, siRNA was used to knockdown GW182 before Ago2, PIWI or PAZ was tethered to the reporter mRNA. Very interestingly, both Ago2and PIWI-mediated repression was significantly reduced upon GW182-knockdown (F ig. 3-6B). The GW182-knockdown was confirmed by quantitative real time PCR (Fig. 3-6C ). In summary, tethering the C-terminal half of Ago2 to the 3-UTR of mR NA recapitulated the repressi on function of Ago2 and this repression required GW182. Discussion Formation of GW182 and Ago Protein Complexes Two studies have identified th at one GW -rich region capable of binding Ago2 is conserved in the plant and yeast orthologs of GW182, and that the GW repeat within th is region is critical for GW182-Ago2 interaction (El-Shami et al. 2007; Till et al. 2007). Consistent with these studies, our data showed that multiple human GW-rich regions were able to bind Ago2 (Fig. 37). GW182 fragments GW1 10 (aa566-1343), GW1 7 (aa1034-1962), and GW1 12 (895-
62 1211) containing the ortholog-conserved GW-rich region (aa1074-1144) (Till et al. 2007) were shown to bind Ago2 (Fig. 3-7). In addition, our data showed that at least four non-overlapping regions of GW182 could independently bind Ago2 and, interestingly, three of these Ago2binding fragments are outside of the ortholog-conserved GW-rich region (Fig. 3-7). Sequence alignment analysis showed that 27aa and 23aa residues of the ortholog-conserved GW-rich region shared 40.7% and 34.8% identity with the GW1 5 and GW1 1b, respectively (Fig. 3-7). However, there was no significant sequence ident ity between the ortholog-conserved region and GW1 a. Further analysis of GW1 1a showed the 4 sub-regions, two of which did not overlap, were all capable of binding Ago2 and, surp risingly, the 5 Trp residues within GW1 1a were not apparently important for the in teraction with Ago2. Our data implied that different GW-rich regions were capable of binding Ago proteins but that the requirement for tryptophan residues varies. Future studies of th e crystal structure of the Ago2GW182 complex should enhance our understanding of how these two molecule s interact with each other. Since our data also indicated that GW182 can bind multiple Ago proteins, it is possible that the different Ago proteins incorporate into the sa me complex and contribute to the formation of functional translational silencing co mplexes. It is yet to be de termined if the function of the silencing complex depends on which Ago proteins it contains. In support of our speculation that GW182 helps to stabilize the binding of multip le Ago-miRNA complexes to the 3-UTR of target mRNA for more efficient translational re pression, it was reported th at more closely-spaced miRNA binding sites in the 3-UTR of target mRNA led to more efficient miRNA-mediated translational repression (Grimson et al. 2007). It is also possible that GW182 simultaneously binds to Ago-miRNA complexes on several different mRNAs and this GW182-Ago interaction may be the driving force for the assembly of submicroscopic and microscopic GWB. This
63 hypothesis is supported by current observations that the GW1 1 GW182 fragment or the PIWI Ago2 fragment could mediate GW182-Ago2 interaction and by our previous data that overexpression of either of thes e two constructs disassembled GW B, possibly due to disruption of GW182-Ago2 interaction by a dominant-n egative effect (Jakymiw et al. 2005). C-terminal Half of Ago2 Preserved the Silencing Function of Ago2 Probably Because It Maintained the Interaction With GW182 Our GW182-Ago2 interaction mapping showed th at the C-terminal half of Ago2 (aa478860) was sufficient for the binding with GW182 wher eas the N-terminal half (aa1-480) was not required. Interestingly, only the C-terminal half of Ago2 preserved the silencing function of Ago2 when directly brought to th e 3-UTR of target mRNA. Th e silencing function mediated by Ago2 or the C-terminal half of Ago2 was abolished upon GW182-knockdown. Our data strongly suggested that interac tion of Ago2 with GW182 is critical for the silencing process mediated by Ago2 at the 3-UTR of target mRNA. This hypothesis is also supported by two recent studies where overexpressing the Ago-bindi ng fragment of yeast or Drosophila ortholog of GW182 greatly disrupted GW182-Ago intera ction and significantly impaired miRNAmediated silencing in vitro and in vivo (Eulalio et al. 2008; Till et al. 2007). Furthermore, our recent study showed that the repression effect caused by tethering GW182 was independent of Ago2 (Li et al., 2008) a nd our data herein again suggested that Ago2 is not the final repressor because its silenc ing function relied greatl y on GW182. Interestingly, the PIWI domain of Ago2 was reported to be re sponsible for the interaction with Dicer (Tahbaz et al. 2004). It would be inte resting to determine if Dicer a nd GW182 compete for the binding to Ago2 through the PIWI domain. Based on the data from current study, we propose a model for miRNA-mediated gene silencing in which, after miRNA guides Ago-miR NA complex to the 3-UTR of target mRNA,
64 Ago protein recruits GW182 to stabilize Ago-miRNA-mRNA bindi ng and represses translation. In addition, GW182-Ago2 interaction recruits Ago2 to GWB, which accumulate many Ago2miRNA-mRNA complexes and become centers for miRNA-mediated silencing. We speculate that the ability of GW182 to potentially bi nd multiple Ago proteins may contribute to aggregation and formation of the cytoplasmic fo ci GW/P bodies, which have been implicated to be critical component s of miRNA activity. *This work was accepted by Journal RNA for publication Feb 2009 and is in press now.
65 Figure 3-1. Schematic of human GW182 and Ago2 de letion constructs using in this study. Amino acid residues of GW182 constructs ar e referenced to the TNGW1, the longer isoform of GW182 (GenBank Accession NM_0 14494.2). Q-repeat, glutamine repeat (box in white); Q/N-rich, glutamine/aspa ragine-rich region (box in magenta); RRM, RNA recognition motif (box in green); GW-rich, glycine/tryptophan-rich region (boxes in yellow); N-GW, N-terminal GW-rich region; M-GW, middle GW-rich region; C-GW, C-terminal GW-rich regi on. Human Ago2 contains two conserved domains: PAZ domain (box in blue) and PIWI domain (box in red).
66 Figure 3-2. GW182 fragment GW1 (aa566-1343) recruited Ago2 to cytoplasmic foci by interacting with the C-term inal half of Ago2.GFP-GW1 (green, a-c) was cotransfected with Flag-Ago2 (d), PIWI (aa478-860, f) or PAZ (aa1-480, h) into HeLa cells. As controls, Flag-PIWI (e) or Flag -PAZ (g) was singly transfected. The cells were stained with anti-Flag antibody (red, dh). Panels in the bottom row are the merged images (i-m). Nuclei were count erstained with DAPI (blue). Scale bar, 10 M.
67 Figure 3-3. The C-terminal half of Ago2 bound to multiple non-overlapping GW-rich regions of GW182. A) GW182 fragments co -precipitated with C terminal half of Ago2. GSTPIWI (aa478-860) was co-transfected with GFP-tagged GW1 1 (lane 1), GW1 1a (lane 2), GW1 1b (lane 3), GW1 7 (lane 4), TNR (lane 11), or GW1 5 (lane 12) into HeLa cells. Similar to positive control GW1 1 (lane 6), GW1 1a (lane 7), GW1 1b (lane 8), GW1 7 (lane 9), GFP-GW1 5 (lane 14), but no GFP-TNR (lane 13), were detected in GST-PIWI precipitates. GST-tagged N1, N-terminal fragment from an unrelated protein hZW10, was co-transfected with GFP-GW1 1 (lane 5) as a negative control and no interaction was detected (lane 10). Asterisks indicate the corresponding GFP-tagged constructs in western blot. B) GW182 fragment GW1 12 co-precipitated with Ago2. Flag-Ago2 wa s co-transfected with GST-tagged TNR (lanes 1, 3) or with GW1 12 (lanes 2, 4), which contains the conserved sequence for Ago2 interaction. Flag-Ago2 was co-p recipitated with GST-tagged GW1 12 but not TNR.
68 Figure 3-4. Four subregions of GW1 1a are capable of binding Ago2 and the interaction between GW1 1a and PIWI was not dependent on the five tryptophan residues. A) Schematics of the 4 overl apping fragments of GW1 1a each containing 2 tryptophan residues and the mutant designed with all 5 tryptophans substituted with alanine. B) Truncated constructs of GW1 1a containing any two continuous tryptophans were able to co-precipitate with PIWI. GST-PIWI was transfected with GFP tagged TNR, GW1 1a and its truncated constructs: 1a_W1-2 (aa270-346, contains the first and second tryptophans of GW1 1a), W2-3 (aa318-339, contains the second and third trytophans), W3-4 (aa 340-439, contains th e third and fourth tryptophan) and W4-5 (aa 409-495, contains the fourth and fifth tr yptophans). Compared to GFP-TNR (lane 2), all truncated c onstructs of GW1 1a (lane 3-6) co-precipi tated with GST-PIWI as full length GW1 1a did (lane 1). C) GW1 1a mutant without tryptophan still coprecipitated with PIWI. A GW1 1a mutant with all tr yptophan (W) mutated to alanine (A) (GW1 1a Mut (W>A), lane 2) was co-transfected with GST-PIWI into HeLa cell. Compared to wild type GW1 1a (lane 1), GW1 1a mutant was coprecipitated with PIWI at the comparable level. GFP-TNR (lane 3) served as a negative control in the GST pull-down assay.
69 Figure 3-5. Both GW182 fragments GW1 1 and GW1 co-precipitated with other human Ago proteins. A) Ago 1 and Ago4, but not Ago3 mutant, co-precipitated with GW182 fragments GW1 1 and GW1 GFP-tagged Ago1, Ago3m (Ago3 mutant), or Ago4 was co-transfected with GST-GW1 (lanes 1-6) or GST-GW1 1 (lanes 7-12). Ago3m is missing an exon (aa757-823), the C-terminal 66aa of the PIWI domain compared to the reference sequence ( NM_024852.2). Both Ago1 (lanes 4,10) and Ago4 (lanes 6,12) w ere pulled down by GST-GW1 or GSTGW1 1. In comparison, Ago3m was absent fr om either pull-down (lanes 5,11). B) Ago3 co-precipitated with GW182 fragment GW1 1. Flag-Ago3 (lanes 1,3) or Ago1 (lanes 2,4) was co-tra nsfected with GST-GW1 1 (lanes 1-4). GFP-Ago3m (lanes 5-6) was co-transfected with GST-GW1 1 as a negative control.
70 Figure 3-6. Gene silencing mediat ed by tethered C-terminal ha lf of Ago2 required GW182. A) Tethered PIWI (aa478-860) down-regulated pr otein synthesis to the same extends of other Ago proteins. HeLa cells were tran sfected with constructs expressing the RL5BoxB reporter, control FL reporter, and indicated NHA-tagged proteins. Bar graphs represent normalized mean values of RL/F L activities with standard errors. The RL/FL values in cells with tethered NHA-tagged Ago2, PIWI, Ago1 and Ago4 were significantly reduced compared to the value in NHA only group, which was normalized as 1. The NHA tagged PAZ, Ago3m, or HA-Ago2 did not show repression effect on the reporters. The expres sion of fusion proteins were determined by Western Blot using anti-HA mAb and are indicated below the bar graphs. The assay was repeated a minimum of 3 times. Asterisks indicate groups have significant difference with NHA only group (unpaired t test, p<0.0001). No signi ficant difference was shown between any two groups with as terisk (unpaired t test, p>0.05). B) Translational repression mediated by tether ed Ago2 or PIWI (aa478-860) was greatly impaired upon GW182-knockdown. HeLa cells were transfected with siRNA for either GW182 (siGW182) or GFP (siGFP). Th irty hours later, cells were transfected again with constructs expr essing reporter RL-5BoxB, control FL reporter, and the same NHA-tagged proteins as indicated in panel A Bar graphs represent the reduction of RL/FL in cells with tethered NHA-Ago2 or NHA-PIWI compared to those in cells with tethered NHA. The re duced values of RL/FL in cell transfected with siGFP were set as 1. Error bars indicate standard errors. The assay was performed in triplicates a nd was repeated for 2 times. *significant difference (unpaired t test, p<0.01). C) GW182 -knockdown by siRNA was confirmed by quantitative real time PCR. The bar gr aphs represent normalized mRNA level of GW182 with standard errors. The mRNA level of GW182 in cells transfected with GFP-siRNA was set as 1. The experi ment was performed in triplicates.
71 Figure 3-7. At least three non-overlapping GW-rich regions that are different from the orthologconserved GW-rich region can independen tly bind Ago2.The dot graph on top indicates the distribution of tryptophan in GW182. W, every tryptophan (magenta diamond); WG/GW, a glycine right adjacent to tryptophan (red triangle); AW/WA, an alanine right adjacent to tryptophan (blue diamond); W only, no glycine or alanine right adjacent to tryptophan (green diamond). The majority of the tryptophans are adjacent to either a glycine or alanine. The schematic of GW182 is indicated below the dot graph. GW-rich region, box in ye llow; ortholog-conserved GW-rich region (aa1074-1144, (Till et al. 2007)) box in red. Compared to GW1 which contains the ortholog-conserved region, GW1 1a, GW1 1b and GW1 5 are the three nonoverlapping regions identified in the current study that binds Ago2. The amino acid sequence alignment between GW1 5/ GW1 1b and ortholog-conserved GW-rich region was performed using ExPASy website tool. represen t sequence identity.
72 DNA sequence: (750nt) G GGG ACA AGT TTG TAC AAA AAA GCA GGC TTC atggatgctg attctgcctc cagttctgaa tcagagagaa acatcactat catggcttca gggaacacag gtggtgaaaa agatggcctt cggaatagca ctggacttgg ttcccaaaac aagtttgtag ttggtagcag cagcaataat gtgggccatg gaagtagtac tgggccagcc ggtttttccc atggagccat aataagcaca tgtcaggtct ctgtggatgc tcctgaaagc aaatctgaaa gtagcaacaa tagaatgaat gctgccggca ctgtaagttc ttcatcaaat ggagggttaa atccaagcac tttgaattca gctagcaacc atggtgccgc cccagtatta gagaacaatg gacttgccct aaaagggcct gtagggagtg gtagttctgg cattaatatt cagtgcagta ctataggcca gatgcctaac aatcagagta ttaactctaa agtgagtggt ggttctaccc atggtaccgc cggaagcctt caggaaactt gtgaatctga agtaagtggt acacagaagg tttcattcag tggtcaacct caaaatatta ccactgaaat gactggacca aataacacta ctaactttat gacctctagt ttaccaaact ccggttcagt gcagaataat gagctgccta gtagtaacac aggggccgcc cgtgtgagca caatgaatca tcctcagatg caggctccat caggtatgaa tggcacttcc TGA ACC CAG CTT TCT TGT ACA AAG TGG TCC Corresponding protein sequen ce: (aa254-503, 250aa total) MDADSASSSESERNITIMASGNTGGEKDGLRNSTGLGSQNKFVVGSSSNNVGHGSSTGPA GFSHGAIISTCQVSVDAPESKSESSNNRMNAAGTVSSSSNGGLNPSTLNSASNHGAAPVL ENNGLALKGPVGSGSSGINIQCSTIGQMPNNQSINSKVSGGSTHGTAGSLQETCESEVSG TQKVSFSGQPQNITTEMTGPNNTTNFMTSSLPNSGSVQNNELPSSNTGAARVSTMNHPQM QAPSGMNGTS Figure 3-8. Sequence of synthesized GW1 1a Mut (W>A). The underlined sequences were designed to incorporate the recombination s ites into pDONR207 vector (Invitrogen). Yellow highlights indicate se quences that were mutated from tgg to gcc in DNA sequence and tryptophan (W) to Alanine (A) in protein sequence.
73 Table 3-1. Primer sequences and PCR conditions for constructs used in the current study Constructs Primer sequence* Tm (first 2 cycles) Tm (next 30 cycles) GW a Forward GGGGACAAGTTTGTACAAAAAAGCAGGCTTCA ACG CCATGGATGCTGATTCT 55.7 C 62.5 C Reverse GGGGACCACTTTGTACAAGAAAGCTGGGT GGGAAG TGCCATTCATACCTG GW b Forward GGGGACAAGTTTGTACAAAAAAGCAGGCTTCA CTT CCCTTTCTCACCTTAGCA 55.2 C 62.4 C Reverse GGGGACCACTTTGTACAAGAAAGCTGGGT GGCCTC TGTCCCATTGTCAGT GW Forward GGGGACAAGTTTGTACAAAAAAGCAGGCTTCA AAG ACCAGCAAGCACAGGTACA 55.7 C 63.3 C Reverse GGGGACCACTTTGTACAAGAAAGCTGGGT TAGGCA ACATCAAGGCATAG GW Forward AAAGCAGGCTTC ACTTGGGGAAACAACATA 55.2 C 59.8 C Reverse AGAAAGCTGGGT TCA GGCTGGTGAGTCTCTCGAAA AAC Ago3m Forward GGGGACAAGTTTGTACAAAAAAGCAGGCTTCA TGG AAATCGGCTCCGCAGGACCC 51.0 C 57.0 C Reverse GGGGACCACTTTGTACAAGAAAGCTGGGT ATCAGA CCTTGGCCCCCAC a_W1-2 Forward AAAAAGCAGGCTCC AAAGATGGCCTTCGGAATA 53.0 C 58.9 C Reverse AGAAAGCTGGGTC TCA TGAATTCAAAGTGCTTGGA TT a_W2-3 Forward AAAAAGCAGGCTTT GGAGCCATAATAAGCACAT 51.3 C 57.6 C Reverse AGAAAGCTGGGT TCA AATACTCTGATTGTTAGGCA TC a_W3-4 Forward AAAAAGCAGGCTCT TCATCAAATGGAGGGTTA 51.9 C 57.4 C Reverse AGAAAGCTGGGT TCA GATAGTGTTATTTGGTCCAG TCA a_W4-5 Forward AAAAAGCAGGCTCT AAAGTGAGTGGTGGTTCTAC 51.7 C 57.3 C Reverse AGAAAGCTGGGT TCA AGTGCCATTCATACCTGAT The underlined sequences were designed to inco rporate the recombination sites for Gateway cloning system. Squares show the artificia l stop codons for the truncated constructs.
74 CHAPTER 4 IDENTIFICATION OF TRANSLATIONAL R EPRESSION DOMAINS IN GW182 Introduction RNA interference (RNAi) was f ound in the last d ecade as an important mechanism of gene regulation at the post-translationa l level. Given the ability to knock down essentially any gene of interest, RNAi via small interfering RNAs (siRNA) has generated a great deal of interest in both basic and applied biology. Differe nt from siRNA, which aims to regulate one gene in a time, a single microRNA (miRNA) has the pot ential to alter the expression of hundreds of proteins at the same time (Baek et al., 2008; Selbach et al ., 2008). Increasing numbers of studies have linked miRNA to many biological functions an d disease pathogenesis and this prompts the possibility of using miRNA as tools in gene therapy. Significant works were carried out by the Bartels group to characterize the sequence requirement for efficient targeti ng and regulation of miRNA (Lewis et al., 2005; Grimson et al., 2007; Friedman et al., 2009). However, the molecular basis of the miRNA-mediated translational repression is poorly understood. The Argonaute (Ago) family, including Ago1 to Ago4, was the most well characterized factor s in the miRNA-induced silencing complex (miRISC) (Peters and Meister, 2007), where they interact with miRNA and recognize the target mRNA. After that, more protein factors are re cruited to induce the s ubsequent repression. Multiple candidates are proposed to play an impor tant role in the miRNA-mediated translational repression. Among these, GW182 is thought to be a conserved factor that retains the importance in miRNA-mediated repression across different species. Knockdown of GW182 impairs miRNA function in human (Liu et al., 2005a), drosophila (Behm-Ansmant et al., 2006) and C. elegans (Ding et al., 2005). An important feature of GW1 82 in this process is its conserved ability to interact with Ago proteins (Ding et al., 2005; Jakymiw et al., 2005; Liu et al., 2005a; Behm-
75 Ansmant et al., 2006; Till et al., 2007). A study in drosophila showed that the interaction between GW182 and Ago protein is essential for miRNA-mediated translational repression (Eulalio et al., 2008b). In our previous study, we showed that GW182 is ab le to interact with Ago2 in multiple regions (Lian et al., 2009). In addition, we showed that GW182 acts downstream of Ago2 to induce the translational repression effect (Li et al., 2008; Lian et al., 2009). These bring our intere st to extend our understandi ng of how GW182 induce the repression. In the current study, we map the repression domain in GW182 and dissect the molecular mechanism of the tr anslational repression event. Materials and Methods Plasmids The cDNA of TNGW 1, GW182, TNR, GW1 1, GW1 1a, GW1 1b, GW1 10, GW1 12, GW1 7, GW1 5, Ago2 and PIWI were constructed as de scribed in previous studies (Jakymiw et al., 2005; Li et al., 2008; Li an et al., 2009). The TNGW1 N-terminal construct 1-565 is generated by restriction enzyme digestion (HpaI and SmaI) on TNGW 1 and ligation using pENTR-TNGW1. GW1 8 and GW1 11 were generated by polymerase chain reaction (PCR) using GW182 cDNA as the template. The primers for cloning GW1 8 are: forwardGGGGACAAGTTTGTACAAAAAAGCAGGCTTC aTTAGACAGAATGGCAATCC, reverseGGGGACCACTTTGTACAAGAAAGCTGGGTTA GGCAACATCAAGGCATAG. The primers for cloning GW1 11 are: forwardAAAAAGCAGGCTTCACTTGGGGAA ACAACATA, reverseAGAAAGCTGGGTTCACTCTGGTGAGTCTCTCGAAAA. A ll of the variants used in current study were subcloned into Gateway compatible GST, GFP, 3xFlag, or NHA (Li et al., 2008) vectors by using Gateway LR recombination reacti on (Invitrogen). The tethering assay plasmids
76 including pClneo-NHA vector, NHA-Ago2, Renilla lu ciferase (RL) and Fire fly luciferase (FL) with or without 5 Boxb (5Bb) structure were kind gifts from Dr. Witold Filipowicz, Friedrich Miescher Institute for Biomed ical Research, Switzerland (P illai et al., 2004). All DNA constructs used in this study were confirmed by dire ct DNA sequencing. Antibodies Rabbit anti-Ago2, rabbit anti-GST and hum an anti-P serum 6181 were gifts from Dr. Tom Hobman (University of Alberta, Edmonton, Canada), Dr. Peter Sayeski and Dr. Minoru Satoh (University of Florida, Gainesville, USA), respectively. Mouse monoclonal anti-HA was purchased from Covance (Emeryville, CA). M ouse monoclonal anti-Flag M2 was purchased from Sigma-Aldrich (St. Louis, MO). Ra bbit polyclonal anti-GFP was purchased from Invitrogen Corporation (Carlsbad, CA). Cell Culture and Plasmid Transfection HeLa and HEK293 cells were cultured in DM EM containing 10% fetal bovine serum in a 37oC incubator with 5% CO2. Cells were grown to 90-100% in 6-well or 24-well plate at the day of transfection. The plasmid transfection was performed using Lipofectamine 2000 (Invitrogen) as per the manufacturers instruction. The teth ering assays were performed in 24-well plate format, where 600ng NHA tagged construct was co-transfected with either 10ng RL-5Bb/100ng FL or 100ng FL-5Bb/10ng RL in 293 cells. Cells were harvested 48 hours after transfection for further experiments. For the GST pulldow n assay, 2ug of GST-tagged construct was cotransfected with 2ug of either GFP or Flag tagge d construct into HeLa cells. HeLa cells were harvested 24 hours after transfecti on for further experiments. GST Pull-down and Western Blot Analysis For GST pulldown assay, transfected cells were lysed with 300 ul NET/NP40 buffer (150m M NaCl, 5mM EDTA, 50mM Tris, pH 7.4, 0.3% NP40) with Complete Protease Cocktail
77 Inhibitor (Roche Diagnostics) and then sonicated at 20% amplitude for 10 sec for 3 times on ice. Afterwards, the lysates were centrifuged at 13,2 00 rpm for 5 min. A fraction of soluble cell lysate was mixed with Laemmli sample buffer as input for western blot an alysis. About 200 ul of the soluble fractions were inc ubated with Glutathione SepharoseTM 4B (GE Healthcare) and mixed at 4oC for 2 h for GST pulldown. After the incubation, the beads were washed with NET/0.3% NP40 buffer for four times and the samp les eluted in Laemmli sample buffer. The input and GST pull-down samples were separated on 10% polyacrylamide gel and transferred to nitrocellulose. Western blotting was performed as described previ ously (Lian et al., 2007). The dilutions of primary antibodies were: 1:1000 for anti-GST, 1:400 for anti-Flag, 1:1000 for antiGFP. Tethering Assay and Dual Luciferase Assay HEK293 cells were harvested 48 hou rs after transfected with te thering constructs and dual luciferase reporters. The FL a nd RL activ ities were measured using Dual-Luciferase Reporter Assay System (Promega, Madison, WI) following the manufacturers protocol. Relative luciferase activities (ratio of targeted luciferase activities over control luciferase activities) were calculated as described previously (Pillai et al ., 2004). The repression effect was estimated by the activity lost of the lucifera se containing 5Bb structure in each group compared to the one transfected with NHA vector (Li et al., 2008). The repression eff ects of each constructs were normalized to the one of TNGW1, which was standardized as 1. The assay was performed multiple times as indicated in Fig. 4-2a. For quantifying the expression levels of different NHA tagged constructs, cell lysates from representiv e luciferase assay was mixed directly with Laemmli sample buffer and separated in 4-20% HCl-Tris Ready Gel (Biorad, Hercules, CA). Samples were then transferred to membrane and performed western blot as described above. To avoid the narrow dynamic sensitivity of tradition film system, bands visualized by enhanced
78 cheminoluminescence assay were captured by Geliance 600 (PerkinElmer, Waltham MA) to obtain the most optimized image. The result s were then analyzed by GeneTools software (PerkinElmer) to quantify the amount of pr otein expressed in individual assay. Results GW182 contained two putative, non-overlapping reg ions harboring the repression effect in tethering assay. To dissect the repression function of tether ed GW182, we decided to first narrow down the region r esponsible for the repression effect in te thering assay. Multiple truncated constructs of GW182 were generated covering the full length protein (Fig. 4-1). All these constructs were adapted to the tethering assay as reported in previous studies (Pill ai et al., 2004; Li et al., 2008) and examined their repression effects accordingl y. The relative repression effects caused by different GW182 truncated constructs emerged in to three groups: 1) no repression effect, which included 1-565, TNR, GW1 1 and QN; 2) high repression eff ect comparable to full length protein, which included GW1 10, GW1 12, GW1 8, GW1 7 and GW1 5; 3) low to moderate repression effect, which included GW1 11. Interestingly, the result revealed that there was more than one region able to i nduce repression effect when it was tethered to the 3-UTR of the reporter mRNA. These regions located in the middle and the C-terminal domains of GW182. GW1 12 and GW1 5 were the smallest, non-overlapping cons tructs residing in these regions. Since GW1 12 and GW1 5 were reported to interact with Ago proteins (Chapter 3), a reasonable interpretation is that their repres sion abilities were related to Ago proteins. Contradicting to this hypothesis, GW1 1, an N-terminal truncated construct of GW182 which strongly interacted with all four human Ago proteins (Lian et al ., 2009), was not able to induced repression effect in the tethering assay. It implied that Ago proteins were not the direct effectors for repression.
79 To achieve a better characterization of the poten cies of inducing translational repression in different regions of GW182, the repression eff ects observed in the te thering assay were normalized to the expression levels of the tether ing constructs. From pr evious representative tethering assays, six cell lysates transfected with NHA tagged truncat ed constructs (TNR, GW1 1, GW1 11, GW1 12, QN and GW1 5), which covered the full length TNGW1, were selected and loaded equally as in the dual luci ferase assay and separated in SDS-PAGE. The relative expression levels of tr ansfected NHA tagged constructs were determined by western blot using the anti-HA monoclonal antib ody (Fig. 4-2B). Based on th e protein amounts of different NHA constructs, TNGW1 full length protein was much more potent, per molecule, in inducing repression when it was tethered to the mR NA of reporter (Fig. 4-2C). Although GW1 12 and GW1 5 were able to inducing repression effects in previous tethering assay, their repression potencies per molecule were only about 15% and 28%, respectively, relative to that of the full length protein of TNGW1. One potential limitation in this analysis is the low transfer efficiency of TNGW1 compared to GW1 12 and GW1 5 which are lower molecular weight truncated constructs. However, incomplete transfer of high molecular protein was not observed because there was no visible remaining dye-labeled molecular marker of 250kD a on the gel after the electrotransfer overnight. Therefore, the re pression potency of GW 182, including its isoform TNGW1, is considerably highe r than the potency of GW1 12 and GW1 5 added together. This data implied that the two regions, GW1 12 and GW1 5, may fit into better architectural positions in the full length protein to achieve an enhanced repression effect in translation. Endogenous acidic ribosomal protein P0, but no t P1 or P2, w as specifically associated with the complexes of Ago2 and GW182 truncated constructs GW1 12 and GW1 5. After identifying two non-overlapping regions in GW182 that harbored repression effect, the next question we asked is how they triggered the repression. C onserved protein domain
80 analysis based on bioinformatics tools from NC BI website indicated two well-defined domains residing in these regions. An RNA recognition motif (RRM) lo cates in the middle of GW1 5, while an Ago hook, which is a conserved peptide st retch able to bind the PIWI domain in Ago proteins (pfam10427), resides in GW1 12. In addition, a sequence string residing in GW1 12 from aa953 to aa994 showed modest similarity to 60S acidic ribosomal protein P1 (RPLP1), which is a component of ribosomal stalk and inte racts with the other two ribosomal proteins, P0 (RPLP0) and P2 (RPLP2) (Zurdo et al ., 2000; Gonzalo et al., 2001). To examine the possibility that GW182 can inte ract with ribosomal P proteins, and also distinguish it from its ability to interact with Ago proteins, we applie d the GST pulldown assay established from our previous study (Lian et al., 2009). Consistent with our previous study, GW1 1a, GW1 1b, GW1 12 and GW1 5 were associated with A go2 (Fig. 4-3A), although the interaction capabilities of GW1 1a and GW1 5 appeared to be weak er than observed when these constructs were fused to th e GFP tag (Fig. 4-3B) in our curre nt and previous study (Lian et al., 2009). Most interestingly, wh en the pulldown samples were probed with a human anti-P serum 6181, P0 was found associated with GW1 12 and GW1 5 but not with other constructs (Fig. 4-3A). In contrast to the strong inte raction of GW1 1b to Ago2 protein, GW1 1b did not pull down P0 protein. The data showed that the association of P0 to the Ago2:GW182 construct was not because of the presence of Ago2 protei n, but the specific regions of GW182 constructs. To our surprise, the rest of the ribosomal stal k components, P1 and P2, were absent in GW1 12 and GW1 5 complex (Fig. 4-3A). This implied that P0 was the only component existing in the pulldown complex, which provided strong evidence th at it was not present either as part of assembled ribosomal stalk construct, or as part of 80S ribosome. These interactions between P0 and GW1 12 or GW1 5 were also observed when we used GST-PIWI to pull down GFP tagged
81 GW182 truncated constructs (Fig 4-3B). This data again confirmed that Ago2 was not sufficient to recruit P0 to the complex. The presences of GW1 12 or GW1 5, or at least the combination of them with Ago protein, were important to interact with P0. Interestingly, GW1 12 and GW1 5 were shown as the regions harboring repr ession effects in our current study. These findings suggested that the P0 prot ein of the ribosomal stalk could be the target of translational inhibition when these GW182 truncated constructs were tethered to the 3-UTR of mRNA. Discussion Unlike that the Ago interacti ng capability identified in f our non-overlapping regions of GW 182 containing glycine and tryptophan repeats, the regions responsib le for translational repression effect are relatively concentrated in two separated regions of about 300 amino acid residing in the middle and C-terminal domains of GW182. Interestingly, there are three studies published recently about the repr ession regions of GW182 protein: drosophila GW182 (dGW182) (Eulalio et al., 2009; Chekulaev a et al., 2009) and the human TN RC6C (Zipprich et al., 2009) All of them reported that the C-terminal domai n containing RNA recognition motif to be highly important for translational repression. However, another study in drosophila found that there are three regions in dGW182 able to induce repr ession effect (Chekulaev a et al., 2009). Although this study is in agreement with the other two studies to some de gree, it actually brings out the possibility that GW182 can potentia lly interact with itself so that many regions of GW182 can trigger the repression by recruiting the endogeno us GW182 protein. Our data, which is focus on human GW182 protein, also identif y that the C-terminal domain of GW182 showing repression effect. Furthermore, another repression region is identified in the middle region of GW182. Further studies need to be carried out to char acterize the repression mechanisms of these two
82 regions. Whether their repression function de pends on the endogenous GW 182 protein will need to be clarified. In our study, we have shown that some regions in GW182 such as GW1 1 can interact with Ago proteins but unable to induce translational repression. Besides re-confirming the findings that Ago proteins alone are not sufficient to induce repression (Li et al., 2008; Lian et al., 2009), it remains unclear why GW 182 contains more regions wh ich can interact with Ago proteins than those that can induce repression. One possibility is that by having more Ago interacting region, GW182 can interact with more than one Ago protein at a time. Studies from Bartels group have showed that mRNA containing multiple miRNA target sites in its 3UTR are repressed more efficiently than the one containi ng fewer miRNA target sites (Grimson et al., 2007). More importantly, closely spaced miRNA target sites often act synergistically and result in stronger repression than those separated farther apart. It implies that a new mechanism may be trigged to enhance repression effect when miRNA sites are close to each other. With its considerable size of 182 kDa and its multiple Ago interacting sites, GW182 may play an important role in this mechanism, in which it connects miRNA:Ago2 complexes that are close to each other to form a more stable complex. It ma y also constitute a larger obstacle locating close to the 5 cap of the mRNA to hinder the transl ational initiation and ev entually induce GW body formation. Alternatively, the multiple Ago inter acting sites may aim to increase the chance, and more importantly, modulate the proper intera ction between GW182 and Ago2. More precisely, the Ago interacting abilities appears to vary in different regions of GW182. If we can imagine the wide-spread Ago interacting regions in GW182 as a funnel, where th e deepest po int is the strongest site interacting with Ago, all Ago protei ns that fall into th e funnel will eventually reach the deepest point unless it is already occupied. When the Ago protein is placed in the
83 best position along GW182, its bound mRNA will be exposed to the repression domains in GW182 and trigger the downstream event of repr ession. This hypothesis is less preferred than the first one since the enhanced effect is obser ved when miRNA target sites are placed close to each other. However, this alternative explana tion can account for having more Ago interacting sites than repression sites if GW 182 is not folded to allow interaction with multiple Ago at the same time. Our data also show reproducibly that the ri bosomal stalk protein P0, but not P1 or P2, interact specifically with GW1 12 and GW1 5, and not other GW182 deletion constructs. More interestingly, the GW182 truncat ed constructs associated with P0 are the same constructs that are able to induced repression effect in the tethering assay. It implies that the interaction with P0 is related to the repression capabil ity of GW182. The ribosomal stalk is known as a distinct lateral protuberance located in the large ribosomal subunit (60S su bunit) and is essential for the ribosome activity (Tchorzewski et al., 2003). It is composed of two heterodimers of P1 and P2 (Hagiya et al., 2005; Gr ela et al., 2008), with P1 bindi ng to P0 (Zurdo et al., 2000; Gonzalo et al., 2001), which is, in turn, attach ed to 28S RNA (Uchiumi and Kominami, 1992) and constitutes a major part of the GTPase-associ ated center in eukaryotic 60S subunit. Unlike other protein factors in the ribosome, the P1 a nd P2 in eukaryotic ribos omal stalks are highly dynamic and undergo a cyclic process of assembly and disassembly during translation (Briceno et al., 2009). A recent study has shown that alteri ng the level of P1/P2 in the ribosomal stalk can affect translational efficiency by reducing the ribosomal subunit joining capacity (MartinezAzorin et al., 2008). Since our data show that the translation inhi bitory regions of GW182 interact with P0 which is an incomplete ribosom al stalk, it is reasonable to deduce that not the whole ribosome is associated with GW182. The presence of additional protein markers from
84 either the small or large ribosomal subunit as well as rRNA will need to be examined in future studies. If only part of the ribosome is associat ed with the inhibitory regions of GW182, it will further confirm that GW182 can potentially di srupt the integrity of the 80S ribosome. It is possible that ribosomal st alk and the joining of the ribos omal subunits are the potential targets for the translational repression eff ect of GW182-containing complex. Supporting this hypothesis, a recent study has showed that eIF6, a ribosomal inhibitory protein known to prevent productive assembly of the 80S ribosome, is important for miRNA-mediated translational repression (Chendrimada et al., 2007). However, to further explore this hypothesis, more experimental evidence is needed. First, one needs to show the inte raction between GW182 and P0 in a more direct manner. Will the interact ion with P0 and the repression effect of GW182 truncated constructs depend on the presence of Ago2? If not, will it depend on other protein factors that have been reporte d important for miRNA function, such as eIF6 or RCK/p54 (Chu and Rana, 2006)? Answer to a ll these questions in further studies may provide a better understanding of this translat ional repression effect. Sec ondly, although the bioinformatics analysis predicted that GW182 shares similarity to P1 protein, we have not shown whether GW182 contributes to the repression effect by competing with P1 to interact with P0 and eventually impairs the integrity of the ribosom al stalk. However, our data has shown some evidence supporting this scenario because P1 and P2 are abse nt in the pulldown complex. Although to show the direct competition between GW182 and P1 may be difficult, it is possible to first examine which region of P0 is require d for its interaction with GW182. While the Nterminal half of P0 is known to attach to the 28s rRNA, its C-terminal half, more specifically, aa230-290, is shown important for interacting with P1/P2 (Santos and Ballesta, 1995). Therefore, it is feasible to generate truncated constructs of P0 representing the N-terminal and C-terminal
85 halves and examine their association with GW18 2. If the interaction of P0 and GW182 is through its C-terminal half and even the P1 inter acting site, it can potenti ally explain the absence of P1/P2 in the complex. If the interaction is not through the C-terminal half, it implies there is potentially other factor(s ) responsible for the absence of P1/P2. In this case, eIF6 may become a good candidate and the previous experiments exam ining the importance of eIF6 may help to interpret this issue. In summary, we have characterized the repression regions in GW182 and confirmed again that the interaction with Ago2 and the repression effect are tw o separated aspects of GW182. We have also purposed a potential mechanism that GW182 may attack the ribosomal stalk to execute translational repression effect in the miRNA-mediated repression, which opens a wide range of testable possibil ities for future studies.
86 Figure 4-1. Schematic of human GW182 truncated constructs used in this study with a summary of their interaction with A go2 and the repression ability in the tethering assay. The amino acid residues of the constructs are referenced to TNGW1, the newly reported isoform of GW182. TNR Q-repeat (aa9 3-127); Ago hook (aa1076-1144); Q/N rich (aa1264-1553) and RRM (RNA recognition motif, aa1780-1853); N-GW, M-GW and C-GW represent N-terminal, middle and C-te rminal glycine/tryptophan (GW) repeats. Asterisks show indicated the GW182 truncat ed constructs which weakly interacted with Flag-Ago2 (Fig. 4-3A) but strongly in teracted with GST-PIWI (Fig. 4-3B).
87 Figure 4-2. Identifying two non-overl apping regions harboring repr ession effect in GW182. A) Comparison of repression effects caused by GW182 full length isoforms and their truncated constructs in dual lucifera se tethering assay. Different NHA-tagged GW182 constructs were tethered to eith er FL-5Bb or RL-5Bb reporters. Their repression effects were normalized to RL or FL control, respectivel y. Then all relative repression effects were sta ndardized to the repressi on effect caused by TNGW1, which was assigned as 1. Error bars indicat e standard deviations in each group; n numbers indicate times for repeating the e xperiments; ns indicates no significance by t test. B) Quantification of protein le vels of different NHA-tagged constructs in tethering assay. 293 cell lysates from repr esentative tethering a ssays were loaded equally and separated by SDS-PAGE. Th e NHA-tagged constructs were detected by ECL western blot assay using the monoc lonal antibody to HA tag. The image was captured by Geliance 600 chemi imaging sy stem (upper inset) and analyzed by GeneTools software. The protein levels relative to the level of NHA-TNGW1 were shown in the lower inset in the figure. The protein amount of the non-specific band (NS Band) was subtracted from the measurement of GW1 1. C) Comparison of the repression potencies of different NHA-constructs based on the re pression effect per protein molecule. The repression potency of each tethering construct was calculated by its average repression eff ect (panel A) divided by its relative protein amount (panel B). The calculated potency was standardized to the potency of TNGW1, which was assigned as 1. The actual valu e of repression potency was shown on the top of its corresponding column.
88 Figure 4-3. Endogenous acidic riboso mal protein P0, but not P1 or P2, was associated with GW182:Ago2 complex containing GW1 5 or GW1 12 in GST pulldown assays. A) P0, but not P1 or P2, was co-p recipitated when GST tagged GW1 12 and GW1 5 pulled down Flag-Ago2. Different GST tagge d GW182 constructs were co-expressed with Flag-Ago2 in HeLa cells. GST pulldown assay was performed cell lysates and the GST and Flag tagged constructs were detected by corresponding antibodies in western blot. Endogenous acidic ribosomal P proteins (P0, P1 and P2) were detected by human anti-P serum 6181. Asterisks showed the corresponding GST tagged constructs in the pulldown a ssay. B) P0 was co-precipitated when GST-PIWI pulled down GFP tagged GW1 12 and GW1 5. GST-PIWI was co-expressed with different GSP tagged GW182 truncated cons tructs in HeLa cells. Similar GST pulldown assay was performed on cell lysa tes were and the GST-PIWI and GFP tagged constructs were detected by corres ponding antibodies in western blot. Human anti-P serum 6181 was used to detect ribosomal P proteins.
89 CHAPTER 5 DISCUSSION AND CONCLUSIONS GW182 is the Repression Trigger of MiRNA-Mediated Gene Silencing Although m iRNAs often silence the gene expr ession through translational repression, the detailed molecular mechanism of this process re mains unclear. In a recent review, Filipowicz and his colleagues summarized the potential mechan isms involved in the tr anslational repression (Filipowicz et al., 2008 ). One consensus among those mechanis ms is that more protein factors must be recruited to achieve silencing effect in miRNA pathway than that in siRNA pathway. Because the incomplete complementary sequence between the miRNA and the mRNA interferes with the catalytic site of the Ago2, the slicing function is hindered in miRNA-mediated gene silencing. Under this circumstance, Ago proteins function like an important guiding factor in the process by interacting with miRNA and rec ognizing its complementary mRNA. More importantly, they can recruit the downstream factor s to trigger and secure the repression effect. Multiple protein factors have been reported im portant for miRNA-mediated gene silencing, including GW182 (Liu et al., 2005a; Zhang et al., 2007; Eulalio et al., 2008b), RCK/p54 (Chu and Rana, 2006) and eIF6 (Chendrimada et al., 2007). GW182 is the only factor that has the conserved importance in miRNA func tion in mammal, Drosophila and C. elegans. It tops the list of translational inhibitory fact ors because it is the factor mo st closely associated with Ago proteins. The interaction between GW182 and Ago proteins has also been reported in different species. A conserved sequence residing in GW182, containing glycine/tryptophan motif, is reported important for the Ago:GW182 interaction (El-Shami et al., 2007; Till et al., 2007). In our study, at least 4 individual regions in GW18 2 are able to interact with Ago2 protein. Furthermore, one of the regions in N-terminal half of GW182, GW1 1a, interacts with Ago2 independent of the tryptophan in its sequence. These findings further emphasize that GW182
90 harbors a strong ability to inter act with Ago proteins. Similar interaction has not been reported in RCK or eIF6 to date. Therefore, we belie ve that GW182 and Ago protein work closely in RNAi activities. In agreement with this c oncept, the interaction between GW182 and Ago2 is reported important for miRNA-mediated transl ational repression (Eulalio et al., 2008b). The next reasonable question raised based on this finding is, whether GW182 can induce repression by itself when it is brought to the mi RNA target site in the 3-UTR of an mRNA. Using the tethering assay established by the Fili powicz laboratory (Pilla i et al., 2004), we show that GW182 can induce translatio nal repression independent on the presence of Ago proteins. In the following study, we even narrow down two regions in GW182 able to induce repression when tethered to the 3-UTR of the reporter mRNA. The repression re gions show a different pattern compared to the Ago-in teracting regions, which implies the Ago-interacting ability is distinct from the inhibitory e ffect of GW182. At the same time, the two independent inhibitory regions indentified in GW182 do not show significan t similarity in sequence. It implies that GW182 may be able to trigger the translational repression through more than one mechanism. Based on all these discussion, we propose that GW182 works downstream of the Ago proteins as the repression trigger in miRNA-mediated gene silencing. However, more studies need to be carr ied on to understand how GW182 trigger the repression effect. In our study, the two inhibitory regions of GW182 are associated with the ribosomal stalk protein P0, but not P1 or P2. It implies that the assembly of the ribosomal stalk is interfered by the presence of GW182 cons tructs. Therefore, whether GW182 silences translation by attacking the assembly of ribosomal stalk is a reasonable next question to address. More evidence needs to found to clarify whether the association between GW182 and P0 is the cause, or the result of the repression. We will be able to gain a better understanding if we can
91 further characterize how direct this interaction is. More specifically, through what regions, or domains, they interact with each other? Is the P1-like sequence in GW182 responsible for the interaction with P0? If we di srupt this interaction, will GW182 lose its repression effect? We are looking forward to examining these hypothese s and answering these questions in future studies. The relationship of GW182 and other repression factors will also need to be examined. With the knockdown-tethering assay as used in Chap ter Two and Three, it is feasible to examine whether RCK or eIF6 is required for the GW182 repression effect. Based on the results, the roles of these factors can be placed at the appr opriate stages in the cascade of translational repression. The Redundancy of RNAi Factors and Their Pote ntial Link to the Di fferent Outcomes of miRNA-Mediated Gene Regulation There are four Argonaute proteins, sharing ove r 90% identities with each other, in the mammalian system. Except for the slicing activit y of Ago2, the functional bias of each paralog remains unclear. All four Ago proteins have been reported to induce repression effect when they are tethered to luciferase repor ter as shown in our current stu dy as well as by other researchers (Pillai et al., 2004). We have al so showed that all four Ago proteins are able to interact with GW182. A recent study from the Tuschls laboratory has showed that the protein factors and the mRNA associated with different Ago proteins are highly simila r (Landthaler et al., 2008). Another study has reported that mouse embryonic st em cells deficient for Ago1-4 are completely defective in miRNA silencing function (Su et al., 2009). However, they have also showed that reintroduction of any single Ago into Ago-defici ent cells is able to rescue the endogenous miRNA silencing function. All th ese data imply that there is substantial functional redundancy within the human Ago family.
92 Similar phenomenon is observed in the GW182 family. In our current study, we have identified a novel isoform of GW182 as TNGW 1. We have showed that both TNGW1 and GW182 colocalize with other GWB component in cel ls and are able to induce repression effect in tethering assay. Studies from different groups have showed that the other two paralogs of GW182, TNRC6B and TNRC6C, are also able to i nduce translational repression in tethering assay and function in the miRNA pathway (Meister et al., 2005; Till et al ., 2007; Zipprich et al., 2009). As redundancies are observed in these importa nt RNAi factors, the miRNA-mediated activity shows a broad variety of outcome. Ev en though most mRNAs targeted by miRNAs are silenced in translation, the s ubsequent mRNA degradation can happen in very different speeds (Eulalio et al., 2008b). In addi tion, some miRNA-targeted mRNA may be re-activated into translational stage under the serum-starved stress condition (Bhattach aryya et al., 2006a; Vasudevan et al., 2007). It is suggested that certain protein factors are important for these special regulations. The Ago and GW182 familie s can probably contribute to the potential different outcomes of miRNA-mediated regulation s since they contain multiple paralogs that carry the same fundamental functi on in RNAi. The functional bias of each paralog needs to be further characterized in future study. Special attention needs to be paid to the functional importance of each paralog in different RNAi activi ties related to different cell types, different stages of cell cycle and different stress c onditions. We are the fi rst group providing evidence related to the heterogene ity of GWB in terms of its component s (Jakymiw et al., 2007). In this current study, we also showed that the novel isofor m, TNGW1, is distributed in only a subset of GWB. It implies that there ar e potential functional differences in GWBs containing different components. The cell imaging data may be more sensitive compared to biochemistry data,
93 because it describes individual variation instead of the behavior of a whole population. As novel technologies emerge rapidly nowadays, we are not surprised that distinguis hable function will be found related to each paralog in th e Ago and GW182 protein families. Working Model of GW182 in miRNA-Mediated Gene Silencing Based on our previous work and current studies, we propose a new working m odel, revised from our previous model (Jakymiw et al., 2007), that illustrates the functional importance of GW182 in miRNA-mediated gene silencing (Fig. 5-1). GW182 is an important factor in trigger the translational inhibition. Study from another group shows that GW182 is also able to induce the subsequent degradation of the targeted mR NAs (Behm-Ansmant et al ., 2006; Eulalio et al., 2008b). Therefore, GW182 is apparently im portant for both miRNA pathway and GWB formation because this cascade is disrupted if GW182 is knockdown. It is likely that more protein factors may be involved in the pathway and potentially cont ribute to the regulation of the miRNA-mediated silencing. We hypothesize that there are unknown factors, including X, Y and Z, that contribute to the miRNA-mediated gene sile ncing at different stages of the cascade. The identification of these factors may potentially fill in the blank of the poor understanding in the fine controls in miRNA-mediated gene regula tion and the reason for its potential different outcomes. Furthermore, based on the current understa nding and the findings from our study, we propose the concept of a four-stage cascade in the miRNA-mediated gene silencing pathway and correlate them to the formation of GWB. It is controversial that whet her the formation of GWB is important for miRNA function. From this model, the conclusion for this issue may depend on two major factors: the current detection limitation and which factor is selected as the detection marker of GWB. Given the he terogeneity of GWB observed (Jakymiw et al., 2007; Li et al.,
94 2008), it is even very difficult to define the inte grity of GWB. Therefore, whether the formation of GWB is absolutely important for miRNA function can still be debated. In summary, our current characterization of the functional domains in GW182 has helped to advance the understanding of miRNA-mediated gene silencing at the molecular level. We have proposed an interesting model how the tr anslational repression e ffect is triggered by GW182 and this opens a wide-range of future research to furthe r dissect the mechanism. Our findings also provide a more comprehensive m odel of miRNA-mediated regulation contributing to future translational resear ch in RNAi-based therapy.
95 Figure 5-1. A model illustrating the functional importan ce of GW182 in miRNA-mediated gene silencing. In the central dogma, genomic information in DNA was transcribed into mRNA. When the mRNA initiates translation, the 40S ribosome small subunit is loaded to the mRNA and recruits the 60S large subunit to form the 80S ribosome. The coding sequence is then translated into its functional protein. In the circumstance of miRNA-mediated gene silencing, miRNAs are transcribed from genomic DNA and transported into cytoplasm. The antisensestrand of the miRNA is incorporated into an Ago-containing complex, named miRNAinducing silencing complex (miRISC) and activat e its function. The activated miRISC can target to mRNA in sequence-specific manner. Some factors (X, such as RCK/p54) may facilitate the targeting at this stage. Note th at it is possible that more than one miRNA can target to the same mRNA in a time. Therefore, there could be multiple miRISC binding to the 3-UTR of the mRNA. The targeted mRNA is extracted from the gene expression pathway (left panel) at this stage. However, it could still return to expression under specific condition, such as stress. When GW182 is recr uited to the 3-UTR of the mRNA, it can bind to multiple Ago proteins and trigger the translational repression by interfering the joining of 80S ribosome. At this stage, the expression of mRNA is mainly silenced. There could be potentially more factors (Y, such as eIF6) i nvolved in the process and securing the repression effect at this stage. The silenced mRNA is eventually led to the degradation stage when GW182 induce the deadenylation of mRNA. Mu tilple 5-3 RNA decay factors (Pacmans) are recruited to degrade the mRNA. There may or may not be protein factors (Z) which can control the degradation speed in this step. Over all, after the miRNA is released to cytoplasm, the miRNA-mediated gene silencing can be separated into four stages, which include activation, targeting, inhibition and degradation. GW182 may play an important role in the inhibition and inducing the degradation. Along these stages, it accompanies with protein factors recruitment and aggregation, which refl ects as the formation of GWB. Knockdown of any factors before the inhibition step, including blocking the biogenesis of miRNA, can significantly impair the silencing effect as well as GWB formation. However, knockdown of the RNA decay factors may only affect the detect able GWB but not the gene silencing effect.
96 GW182 GW182 m7pppG AAAn 60S 40S Protein m7pppG AAAn Activated miRISCs m7pppG AAAn m7pppG AAAn Met m7pppG AAAn 80S miRNA Mediated Gene Silencing GW182 mRNA DNA miRNA Gene Expression Activation Targeting Degradation Inhibition Size of GW BodyDetection limitation X Z Ago RNA Decay Factors Ago Y
97 LIST OF REFERENCES Assaad, F. F., Tucker, K. L. and Signer, E. R. (1993). Epigenetic repeat-induced gene silencing (R IGS) in Arabidopsis. Plant Mol. Biol. 22, 1067-1085. Baek, D., Villen, J., Shin, C., Camargo, F. D., Gygi, S. P. and Bartel, D. P. (2008). The impact of microRNAs on protein output. Nature 455 64-71. Bashkirov, V. I., Scherthan, H., Solinger, J. A., Buerstedde, J. M. and Heyer, W. D. (1997). A mouse cytoplasmic exoribonuclease (mXRN1p) w ith preference for G4 tetraplex substrates. J. Cell Biol. 136 761-773. Behm-Ansmant, I., Rehwinkel, J., Doerks, T. Stark, A., Bork, P. and Izaurralde, E. (2006). mRNA degradation by miRNAs and GW182 re quires both CCR4:NOT deadenylase and DCP1:DCP2 decapping complexes. Genes Dev. 20, 1885-1898. Bhanji, R. A., Eystathioy, T., Chan, E. K., Bloch, D. B. and Fritzler, M. J. (2007). Clinical and serological features of patients with autoantibodies to GW/P bodies. Clin. Immunol. 125, 247-256. Bhattacharyya, S. N., Habermacher, R., Martin e, U., Closs, E. I. and Filipowicz, W. (2006a). Relief of microRNA-mediated translat ional repression in human cells subjected to stress. Cell 125, 1111-1124. Bhattacharyya, S. N., Habermacher, R., Martin e, U., Closs, E. I. and Filipowicz, W. (2006b). Stress-induced reversal of microRNA repression an d mRNA P-body localization in human cells. Cold Spring Harb. Symp. Quant. Biol. 71, 513-521. Boerjan, W., Bauw, G., Van, M. M. and Inze, D. (1994). Distinct phenotypes generated by overexpression and suppression of S-adenosyl-L-m ethionine synthetase reveal developmental patterns of gene silencing in tobacco. Plant Cell 6, 1401-1414. Briceno, V., Camargo, H., Remacha, M., Santos, C. and Ballesta, J. P. (2009). Structural and functional characterization of the amino terminal domain of the y east ribosomal stalk P1 and P2 proteins. Int. J. Biochem. Cell Biol. 41 1315-1322. Chekulaeva, M., Filipowicz, W. and Parker, R. (2009). Multiple independent domains of dGW182 function in miRNA-mediated repression in Drosophila. RNA. Chendrimada, T. P., Finn, K. J., Ji, X., Baillat, D., Gregory, R. I., Liebhaber, S. A., Pasquinelli, A. E. and Shiekhattar, R. (2007). MicroRNA silencing through RISC recruitment of eIF6. Nature 447, 823-828. Chu, C. Y. and Rana, T. M. (2006). Translation Repression in Human Cells by MicroRNAInduced Gene Silencing Requires RCK/p54. PLoS. Biol. 4, e210. Cougot, N., Babajko, S. and Seraphin, B. (2004). Cytoplasm ic foci are sites of mRNA decay in human cells. J. Cell Biol. 165, 31-40.
98 Covey, S. N., Al-Kaff, N. S., Langara, A. and Turner, D. S. (1997). Plants combat infection by gene silencing. Nature 385, 781-782. Decker, C. J., Teixeira, D. and Parker, R. (2007). Edc3p and a glutamine/asparagine-rich domain of Lsm4p function in processing body assembly in Saccharomyces cerevisiae. J. Cell Biol. 179, 437-449. Ding, L., Spencer, A., Morita, K. and Han, M. (2005). The developmental timing regulator AIN-1 interacts with miRISCs and may target the argonaute protein ALG-1 to cytoplasmic P bodies in C. elegans. Mol. Cell 19, 437-447. El-Shami, M., Pontier, D., Lahmy, S., Braun, L., Picart, C., Vega, D., Hakimi, M. A., Jacobsen, S. E., Cooke, R. and Lagrange, T. (2007). Reiterated WG/GW motifs form functionally and evolutionarily conserved ARGONAUTE-binding platforms in RNAi-related components. Genes Dev. 21, 2539-2544. Eulalio, A., Behm-Ansmant, I., Schweizer, D. and Izaurralde, E. (2007a). P-body formation is a consequence, not the cause of RNA-mediated gene silencing. Mol. Cell Biol. Eulalio, A., Huntzinger, E. and Izaurralde, E. (2008a). Getting to the root of miRNAmediated gene silencing. Cell 132, 9-14. Eulalio, A., Huntzinger, E. and Izaurralde, E. (2008b). GW182 interaction with Argonaute is essential for miRNA-mediated tran slational repression and mRNA decay. Nat. Struct. Mol. Biol. 15, 346-353. Eulalio, A., Rehwinkel, J., Stricker, M., Huntzing er, E., Yang, S. F., Doerks, T., Dorner, S., Bork, P., Boutros, M. and Izaurralde, E. (2007b). Target-specific requirements for enhancers of decapping in miRNA-mediated gene silencing. Genes Dev. 21, 2558-2570. Eulalio, A., Tritschler, F., Buttner, R., Weichenr ieder, O., Izaurralde, E. and Truffault, V. (2009). The RRM domain in GW182 proteins contributes to mi RNA-mediated gene silencing. Nucleic Acids Res. Eystathioy, T., Chan, E. K. L., Mahler, M., Luft, L. M., Fritzler, M. L. and Fritzler, M. J. (2003a). A panel of monoclonal antibodies to cytoplasmic GW bodies and the mRNA binding protein GW182. Hybrid. Hybridomics. 22, 79-86. Eystathioy, T., Chan, E. K. L., Takeuchi, K., Mahler, M., Luft, L. M., Zochodne, D. W. and Fritzler, M. J. (2003b). Clinical and serological associ ations of autoantibodies to GW bodies and a novel cytoplasmic autoantigen GW182. J. Mol. Med. 81 811-818. Eystathioy, T., Chan, E. K. L., Tenenbaum, S. A., Keene, J. D., Griffith, K. and Fritzler, M. J. (2002). A phosphorylated cytoplasmic autoantigen, GW182, associates with a unique population of human mRNAs within novel cytoplasmic speckles. Mo l. Biol. Cell 13, 1338-1351.
99 Eystathioy, T., Jakymiw, A., Chan, E. K. L., S eraphin, B., Cougot, N. and Fritzler, M. J. (2003c). The GW182 protein colocalizes with mRNA degradation associated proteins hDcp1 and hLSm4 in cytoplasmic GW bodies. RNA. 9, 1171-1173. Famulski, J. K., Vos, L., Sun, X. and Chan, G. (2008). Stable hZW10 kinetochore residency, mediated by hZwint-1 interaction, is essential for the mitotic checkpoint. J. Cell Biol. 180 507520. Filipowicz, W., Bhattacharyya, S. N. and Sonenberg, N. (2008). Mechanisms of posttranscriptional regulation by microRNAs: are the answers in sight? Nat Rev. Genet. 9, 102-114. Filipowicz, W., Jaskiewicz, L., Kolb, F. A. and Pillai, R. S. (2005). Post-transcriptional gene silencing by siRNAs and miRNAs. Curr. Opin. Struct. Biol. 15 331-341. Fire, A., Xu, S., Montgomery, M. K., Kostas S. A., Driver, S. E. and Mello, C. C. (1998). Potent and specific genetic interference by double-stranded RNA in Caenorhabditis elegans. Nature 391, 806-811. Friedman, R. C., Farh, K. K., Burge, C. B. and Bartel, D. P. (2009). Most mammalian mRNAs are conserved targets of microRNAs. Genome Res. 19, 92-105. Gonzalo, P., Lavergne, J. P. and Reboud, J. P. (2001). Pivotal role of the P1 N-terminal domain in the assembly of the mammalian ribos omal stalk and in the proteosynthetic activity. J. Biol. Chem. 276, 19762-19769. Grela, P., Sawa-Makarska, J., Gordiyenko, Y. Robinson, C. V., Grankowski, N. and Tchorzewski, M. (2008). Structural propert ies of the human acidic ribosomal P proteins forming the P1-P2 heterocomplex. J. Biochem. 143, 169-177. Grimson, A., Farh, K. K., Johnston, W. K., Garret t-Engele, P., Lim, L. P. and Bartel, D. P. (2007). MicroRNA targeting specificity in mammals: determinants beyond seed pairing. Mol. Cell 27, 91-105. Hagiya, A., Naganuma, T., Maki, Y., Ohta, J., Tohkairin, Y., Shimizu, T., Nomura, T., Hachimori, A. and Uchiumi, T. (2005). A mode of assembly of P0, P1, and P2 proteins at the GTPase-associated center in animal ribosome: in vitro analyses with P0 truncation mutants. J. Biol. Chem. 280, 39193-39199. Heyer, W. D., Johnson, A. W., Reinhart, U. and Kolodner, R. D. (1995). Regulation and intracellular localization of Saccharomyces cerevisiae strand exchange protein 1 (Sep1/Xrn1/Kem1), a multifunctional exonuclease. Mol. Cell Biol. 15, 2728-2736. Hornstein, E., Mansfield, J. H., Yekta, S ., Hu, J. K., Harfe, B. D., McManus, M. T., Baskerville, S., Bartel, D. P. and Tabin, C. J. (2005). The microRNA miR-196 acts upstream of Hoxb8 and Shh in limb development. Nature 438, 671-674.
100 Ingelfinger, D., rndt-Jovin, D. J., Luhrmann, R. and Achsel, T. (2002). The human LSm1-7 proteins colocalize with the mRNA-degrading enzymes Dcp1/2 and Xrnl in distinct cytoplasmic foci. RNA. 8 1489-1501. Jakymiw, A., Lian, S., Eystathioy, T., Li, S., Sa toh, M., Hamel, J. C., Fritzler, M. J. and Chan, E. K. L. (2005). Disruption of GW bodies impairs mammalian RNA interference. Nat. Cell Biol. 7, 1267-1274. Jakymiw, A., Pauley, K. M., Li, S., Ikeda, K., Lian, S., Eystathioy, T., Satoh, M., Fritzler, M. J. and Chan, E. K. L. (2007). The role of GW/P-bodies in RNA processing and silencing. J. Cell Sci. 120 1317-1323. Landthaler, M., Gaidatzis, D., Rothballer, A., Chen, P. Y., Soll, S. J., Dinic, L., Ojo, T., Hafner, M., Zavolan, M. and Tuschl, T. (2008). Molecular char acterization of human Argonaute-containing ribonucleoprotein co mplexes and their bound target mRNAs. RNA 14, 2580-2596. Legault, P., Li, J., Mogridge, J., Kay, L. E. and Greenblatt, J. (1998). NMR structure of the bacteriophage lambda N peptide/boxB RNA complex: recognition of a GNRA fold by an arginine-rich motif. Cell 93, 289-299. Lewis, B. P., Burge, C. B. and Bartel, D. P. (2005). Conserved seed pairing, often flanked by adenosines, indicates that thousands of human genes are microRNA targets. Cell 120, 15-20. Li, S., Lian, S. L., Moser, J. J., Fritzler, M. L., Fritzler, M. J., Satoh, M. and Chan, E. K. (2008). Identification of GW182 a nd its novel isoform TNGW1 as tr anslational repressors in Ago2-mediated silencing. J. Cell Sci. 121, 4134-4144. Lian, S., Fritzler, M. J., Katz, J., Hamazaki, T. Terada, N., Satoh, M. and Chan, E. K. L. (2007). Small interfering RNA-medi ated silencing induces target-d ependent assembly of GW/P bodies. Mol. Biol. Cell 18 3375-3387. Lian, S., Jakymiw, A., Eystathioy, T., Hamel, J. C., Fritzler, M. J. and Chan, E. K. L. (2006). GW bodies, microRNAs and the cell cycle. Cell Cycle 5, 242-245. Lian, S. L., Li, S., Abadal, G. X., Paule y, B. A., Fritzler, M. J. and Chan, E. K. (2009). The C-terminal half of human Ago2 binds to mu ltiple GW-rich regions of GW182 and requires GW182 to mediate silencing. RNA. Liu, J., Carmell, M. A., Rivas, F. V., Marsden, C. G., Thomson, J. M., Song, J. J., Hammond, S. M., Joshua-Tor, L. and Hannon, G. J. (2004). Argonaute2 is the catalytic engine of mammalian RNAi. Science 305, 1437-1441. Liu, J., Rivas, F. V., Wohlschlegel, J., Yates, J. R., Parker, R. and Hannon, G. J. (2005a). A role for the P-body component GW182 in m icroRNA function. Nat. Cell Biol. 7, 1261-1266. Liu, J., Valencia-Sanchez, M. A., Hannon, G. J. and Parker, R. (2005b). MicroRNAdependent localization of target ed mRNAs to mammalian P-bodies. Nat. Cell Biol. 7, 719-723.
101 Livak, K. J. and Schmittgen, T. D. (2001). Analysis of relative gene expression data using realtime quantitative PCR and the 2 (-Delta Delta C(T)) Method. Methods 25 402-408. Lytle, J. R., Yario, T. A. and Steitz, J. A. (2007). 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 104, 9667-9672. Martinez-Azorin, F., Remacha, M. and Ballesta, J. P. (2008). Functional characterization of ribosomal P1/P2 proteins in human cells. Biochem. J. 413, 527-534. Meister, G., Landthaler, M., Patkaniowska, A., Dorsett, Y., Teng, G. and Tuschl, T. (2004). Human Argonaute2 mediat es RNA cleavage targeted by miRNAs and siRNAs. Mol. Cell 15, 185-197. Meister, G., Landthaler, M., Peters, L., Chen, P. Y., Urlaub, H., Luhrmann, R. and Tuschl, T. (2005). Identification of novel ar gonaute-associated proteins. Curr. Biol. 15, 2149-2155. Moser, J. J., Eystathioy, T., Chan, E. K. L. and Fritzler, M. J. (2007). Markers of mRNA stabilization and degradation, and RNAi within astrocytoma GW bodies. J. Neurosci. Res. 85, 3619-3631. Nilsen, T. W. (2007). Mechanisms of microRNA-media ted gene regulation in animal cells. Trends Genet. 23, 243-249. Pauley, K. M., Eystathioy, T., Jakymiw, A., Hame l, J. C., Fritzler, M. J. and Chan, E. K. L. (2006). Formation of GW bodies is a consequence of microRNA genesis. EMBO Rep. 7, 904910. Peters, L. and Meister, G. (2007). Argonaute proteins: me diators of RNA silencing. Mol. Cell 26, 611-623. Pillai, R. S., Artus, C. G. and Filipowicz, W. (2004). Tethering of hum an Ago proteins to mRNA mimics the miRNA-mediated repression of protein synthesis. RNA. 10, 1518-1525. Pillai, R. S., Bhattacharyya, S. N., Artus, C. G., Zoller, T., Cougot, N., Basyuk, E., Bertrand, E. and Filipowicz, W. (2005). Inhibition of translational initiation by Let-7 MicroRNA in human cells. Science 309 1573-1576. Ratcliff, F., Harrison, B. D. and Baulcombe, D. C. (1997). A similarity between viral defense and gene silencing in plants. Science 276, 1558-1560. Rehw inkel, J., Behm-Ansmant, I., Gatfield, D. and Izaurralde, E. (2005). A crucial role for GW182 and the DCP1:DCP2 decapping complex in miRNA-mediated gene silencing. RNA. 11, 1640-1647. Santos, C. and Ballesta, J. P. (1995). The highly conserved prot ein P0 carboxyl end is essential for ribosome activity only in the absence of proteins P1 and P2. J. Biol. Chem. 270, 2060820614.
102 Schneider, M. D., Najand, N., Chaker, S., Pare J. M., Haskins, J., Hughes, S. C., Hobman, T. C., Locke, J. and Simmonds, A. J. (2006). Gawky is a component of cytoplasmic mRNA processing bodies required for early Drosophila development. J. Cell Biol. 174, 349-358. Selak, S., Mahler, M., Miyachi, K., Fritzler, M. L. and Fritzler, M. J. (2003). Identification of the B-cell epitopes of the early endosome antigen 1 (EEA1). Clin. Immunol. 109, 154-164. Selbach, M., Schwanhausser, B., Thierfelder, N., Fang, Z., Khanin, R. and Rajewsky, N. (2008). Widespread changes in protei n synthesis induced by microRNAs. Nature 455, 58-63. Sen, G. L. and Blau, H. M. (2005). Argonaute 2/RISC resides in sites of mammalian mRNA decay known as cytoplasmic bodies. Nat. Cell Biol. 7, 633-636. Sheth, U. and Parker, R. (2003). Decapping and decay of messenger RNA occur in cytoplasmic processing bodies. Science 300, 805-808. Su, H., Trombly, M. I., Chen, J. and Wang, X. (2009). Essential and ov erlapping functions for mammalian Argonautes in microRNA silencing. Genes Dev. 23, 304-317. Tchorzewski, M., Krokowski, D., Rzeski, W., Issinger, O. G. and Grankowski, N. (2003). The subcellular distribution of the human ribosomal "stalk" components: P1, P2 and P0 proteins. Int. J. Biochem. Cell Biol. 35, 203-211. Till, S., Lejeune, E., Thermann, R., Bortfeld, M., Hothorn, M., Enderle, D., Heinrich, C., Hentze, M. W. and Ladurner, A. G. (2007). A conserved motif in Argonaute-interacting proteins mediates functional interacti ons through the Argonaute PIWI domain. Nat Struct. Mol. Biol. 14, 897-903. Uchiumi, T. and Kominami, R. (1992). Direct evidence for interaction of the conserved GTPase domain within 28 S RNA with mammalia n ribosomal acidic phos phoproteins and L12. J. Biol. Chem. 267, 19179-19185. van Dijk, E., Cougot, N., Meyer, S., Ba bajko, S., Wahle, E. and Seraphin, B. (2002). Human Dcp2: a catalytically active mRNA decapping enzyme located in specific cytoplasmic structures. EMBO J. 21, 6915-6924. Vasudevan, S., Tong, Y. and Steitz, J. A. (2007). Switching from repression to activation: microRNAs can up-regulate translation. Science 318, 1931-1934. Yang, Z., Jakymiw, A., Wood, M. R., Eystathioy, T., Rubin, R. L., Fritzler, M. J. and Chan, E. K. L. (2004). GW182 is critical for the stability of GW bodies expressed during the cell cycle and cell proliferation. J. Cell Sci. 117 5567-5578. Yuan, Y. R., Pei, Y., Ma, J. B., Kuryavyi, V., Z hadina, M., Meister, G., Chen, H. Y., Dauter, Z., Tuschl, T. and Patel, D. J. (2005). Crystal structure of A. aeolicus argonaute, a sitespecific DNA-guided endoribonuclease, provides insights into RISC-mediated mRNA cleavage. Mol. Cell 19 405-419.
103 Zhang, L., Ding, L., Cheung, T. H., Dong, M. Q., Chen, J., Sewell, A. K., Liu, X., Yates, J. R., III and Han, M. (2007). Systematic identification of C. elegans miRISC proteins, miRNAs, and mRNA targets by their interactions with GW182 proteins AIN-1 and AIN-2. Mol. Cell 28, 598-613. Zipprich, J. T., Bhattacharyya, S., Mathys, H. and Filipowicz, W. (2009). Importance of the C-terminal domain of the human GW182 protein TNRC6C for transl ational repression. RNA. Zurdo, J., Parada, P., van den, B. A., Nusspau mer, G., Jimenez-Diaz, A., Remacha, M. and Ballesta, J. P. (2000). Assembly of Saccharomyces cerevi siae ribosomal stalk: binding of P1 proteins is required for the interaction of P2 proteins. Biochemistry 39, 8929-8934.
104 BIOGRAPHICAL SKETCH Songqing Li was born in Guangzhou, China, 1978. He studied clinical m edicine at Sun Yat-sen University, a famous medical school in China, and received Bachelor of Medicine in 2001. After he graduated, he had practiced as an OB/GYN doctor in the second affiliated hospital of the university for three years. His passion to und erstand the mechanisms behind diseases drove him to pursue a higher degree in basic science research. Songqing was admitted by the Interdisciplinary Program in Biomedical Sciences in College of Medicine at the University of Florida in 2004. At the same year, he married his love, Shang Li Lian, and established their family in America. In summ er of 2005, Songqing joined the laboratory of Dr. Edward K.L. Chan and started his scientific ad venture. With the guidance of his mentor, he devoted himself to the study of characterizing the expression and the functions of GW182, which is a protein factor that is im portant in the RNA interference pathway. He received Ph.D. degree in Medical Sciences-Molecular Cell Biology in May 2009.