Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Characterization of alternative reading frame selection by a viral internal ribosome entry site Ren, Qian 2014

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2014_september_ren_qian.pdf [ 6.8MB ]
Metadata
JSON: 24-1.0167579.json
JSON-LD: 24-1.0167579-ld.json
RDF/XML (Pretty): 24-1.0167579-rdf.xml
RDF/JSON: 24-1.0167579-rdf.json
Turtle: 24-1.0167579-turtle.txt
N-Triples: 24-1.0167579-rdf-ntriples.txt
Original Record: 24-1.0167579-source.json
Full Text
24-1.0167579-fulltext.txt
Citation
24-1.0167579.ris

Full Text

CHARACTERIZATION OF ALTERNATIVE READING FRAME SELECTION BY A VIRAL INTERNAL RIBOSOME ENTRY SITE    by Qian Ren  M.Sc., Chinese Academy of Sciences, 2009  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Biochemistry and Molecular Biology)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)   July 2014  © Qian Ren, 2014 ii  Abstract  Dicistroviruses possess a positive-sense, monopartite single-strand RNA genome that encodes two open reading frames containing the nonstructural and structural polyproteins (ORF1 and ORF2) separated by the intergenic region (IGR) internal ribosome entry site (IRES). Translation of each ORF is directed by distinct IRESs, a 5’ untranslated region (UTR) and an IGR IRES. Previous bioinformatic studies have shown that a subset of dicistroviruses contain an overlapping gene in the +1 translational reading frame within the structural polyprotein gene near the IGR IRES region. We hypothesize that the IGR IRES directs translation of two overlapping ORFs, a novel +1 frame ORFx and the 0 frame ORF which encodes the viral structural polyprotein. In this thesis, using Israeli acute paralysis virus (IAPV) as a model, the existence and start site of ORFx were identified using mutagenesis and Mass Spectrometry analyses. In addition, the structural elements within the IAPV IGR IRES that determine alternative reading frame translation initiation were explored. Lastly, the localization of overexpressed tagged-ORFx in Drosophila S2 cells was examined to gain insights of its function. Summarizing, we have discovered a novel mechanism that increases the coding capacity of a virus through an IGR IRES. These studies of IAPV IGR-IRES will further our understanding of IRESs mediated translation initiation and reading frame decoding.  iii  Preface   A version of chapter 2 has been published. Ren, Q. *, Wang, Q.S. *, Firth, A.E., Chan, M.M.Y., Gouw, J.W., Guarna, M.M., Foster, L.J., Atkins, J.F., and Jan, E. (2012). Alternative reading frame selection mediated by a tRNA-like domain of an internal ribosome entry site. Proceedings of the National Academy of Sciences of the United States of America 109, E630-E639 (*shared first authorship). I conducted the experiments and data analyses on identification of the existence and start site of ORFx, and characterization of the U-G mediated +1 frame translation mechanism in in vitro translation extracts. Wang, Q.S. constructed the pIAPV-WT bicistronic reporter and some of the mutants and performed the preliminary activity tests in RRL. The mass spectrometry analysis to identify the start site of ORFx and the multiple reaction monitoring analysis was performed in collaboration with Chan, M.M.Y., Gouw, J.W., Guarna, M.M., Foster, L.J (Figure 2.6 and 2.8). I contributed to the samples preparation for mass spectrometry analysis. A version of chapter 3 is in press. Ren, Q., Au, H.H.T., Wang Q.S., Lee, S., and Jan, E. (2014) Structural determinants of an internal ribosome entry site that direct translational reading frame selection. Nucleic Acids Res. In press. Preliminary experiments of Figure 3.12 were performed by Hilda Au (Shown in Appendix Figure F). Some of the mutagenesis assays were completed by Seonghoon Lee (Listed in Appendix G). Results are shown with permission.  Copyrighted materials presented herein have been reproduced or adapted with permission, as indicated. iv  Table of Contents Abstract ................................................................................................................................. ii Preface .................................................................................................................................. iii Table of Contents ................................................................................................................. iv List of Tables ........................................................................................................................ ix List of Figures ....................................................................................................................... x List of Symbols and Abbreviations .................................................................................. xiii Acknowledgements ............................................................................................................ xvi Dedication .......................................................................................................................... xvii Chapter 1: Introduction ....................................................................................................... 1 1.1 The canonical translation ........................................................................................ 1 1.1.1 Eukaryotic cap-dependent translation initiation ................................................. 1 1.1.2 Eukaryotic translation elongation ....................................................................... 4 1.2 Non-canonical translation ....................................................................................... 7 1.2.1 Cap-independent translation initiation by an internal ribosome entry site ......... 7 1.2.2 Recoding mechanisms ...................................................................................... 13 1.2.2.1 Programmed eukaryotic -1 ribosomal frameshifting ................................ 13 1.2.2.2 Programmed +1 translational frameshifting ............................................. 16 1.2.2.2.1 The Ty1 and Ty3 retrotransposons in yeast S. cerevisiae ................... 16 1.2.2.2.2 Antizyme and +1 frameshift suppressor tRNA ................................... 19 1.3 The Dicistroviridae viral family ........................................................................... 20 1.3.1 Genomic organization....................................................................................... 23 v  1.3.2 Modulation of host translation machinery ........................................................ 26 1.4 The Dicistroviridae intergenic region internal ribosome entry site ..................... 28 1.4.1 Structure of the IGR IRES ................................................................................ 28 1.4.2 Function of the IGR IRES ................................................................................ 37 1.5 Thesis investigation .............................................................................................. 39 Chapter 2: Identification of the +1 Frame ORFx ............................................................ 41 2.1 Introduction .......................................................................................................... 41 2.2 Material and methods ........................................................................................... 43 2.2.1 DNA constructs and reagents ........................................................................... 43 2.2.2 In vitro translation assay ................................................................................... 43 2.2.3 Western Blot analysis ....................................................................................... 44 2.2.4 Northern Blot analysis ...................................................................................... 44 2.2.5 Mass spectrometry analysis .............................................................................. 45 2.2.6 Virus infection in honey bee ............................................................................. 45 2.2.7 RT-PCR detection assay ................................................................................... 46 2.3 Results .................................................................................................................. 46 2.3.1 ORFx is expressed from a Full-Length IAPV cDNA in in-vitro insect Sf21 extracts .......................................................................................................................... 46 2.3.2 Mutagenesis studies reveal the start site of ORFx in a Full-Length IAPV cDNA clone .................................................................................................................. 51 2.3.3 U6562-G6618 base pairing directs +1 frame ORFx translation ....................... 55 2.3.4 Mass spectrometry analysis confirms the start site of ORFx expressed in Sf21 extracts in-vitro ............................................................................................................. 60 vi  2.3.5 Detection of ORFx in virally infected honey bees by mass spectrometry ....... 62 2.4 Discussion ............................................................................................................. 65 Chapter 3: The mechanism of ORFx translation frame selection ................................. 68 3.1 Introduction .......................................................................................................... 68 3.2 Materials and methods .......................................................................................... 69 3.2.1 DNA constructs and reagents ........................................................................... 69 3.2.2 In vitro transcription and translation assay ....................................................... 69 3.2.3 5’ End labeling of RNA and primers ................................................................ 70 3.2.4 SHAPE analysis of IRES and IRES–ribosome complexes .............................. 71 3.2.5 40S and 60S ribosomal subunit purification .................................................... 72 3.2.6 Toeprinting analysis of ribosomal complexes using purified components ...... 73 3.3 Results .................................................................................................................. 74 3.3.1 U6562-G6618 base pairing is not strictly necessary for +1 frame translation in specific mutant IRES contexts ...................................................................................... 74 3.3.2 Structure probing analysis of WT and mutants IAPV IGR IRESs in solution . 80 3.3.3 Mutations within PKI codon-anticodon base pairing region affect 0 and +1 frame translation ........................................................................................................... 85 3.3.4 Single mutant within IGR IRES that determines 0 or +1 reading frame translation ..................................................................................................................... 88 3.3.5 Effects of SLIII on 0 and +1 frame translation ................................................. 95 3.3.6 Nucleotide identity within the VLR affects IRES translation .......................... 97 3.3.7 Structure probing analysis of mutant IRESs that directs 0 or +1 reading frame translation ..................................................................................................................... 97 vii  3.3.8 Structural changes in the PKI domain of mutant IRESs upon ribosome binding 101 3.3.9 Positioning of the 80S ribosomes on the mutant IRESs ................................. 108 3.4 Discussion ........................................................................................................... 111 Chapter 4: The localization of tagged-ORFx in Drosophila S2 cells ............................ 115 4.1 Introduction ........................................................................................................ 115 4.2 Materials and methods ........................................................................................ 117 4.2.1 DNA constructs and reagents ......................................................................... 117 4.2.2 Cell culture and transfection ........................................................................... 117 4.2.3 Cell viability assay ......................................................................................... 117 4.2.4 Western Blot analysis ..................................................................................... 118 4.2.5 Immunofluorenscence and subcellular localization analysis.......................... 118 4.2.6 Subcellular fractionation................................................................................. 119 4.2.7 CrPV propogation, infection and viral titer .................................................... 120 4.3 Results ................................................................................................................ 121 4.3.1 Overexpression of tagged-ORFx in Drosophila S2 cells has little effect on cell viability ....................................................................................................................... 121 4.3.2 Subcellular localization of ORFx in Drosophila S2 cells .............................. 123 4.3.3 Localization of ORFx in CrPV infected S2 cells ........................................... 130 4.3.4 ORFx expression has little effect on CrPV propagation in S2 cells ............... 135 4.4 Discussion ........................................................................................................... 136 Chapter 5: Summary and future directions ................................................................... 141 References.......................................................................................................................... 148 viii  Appendices ........................................................................................................................ 166 Appendix A Methods ..................................................................................................... 166 A.1 DNA constructs, cell culture and transfection ................................................ 166 A.2 Translation assay ............................................................................................ 166 A.3 Western Blot analysis ..................................................................................... 166 A.4 Mass spectrometry analysis ............................................................................ 167 A.5 MRM assay ..................................................................................................... 168 Appendix B Sequences of plasmid and primers ............................................................. 169 B.1 Sequence inserted in the IGR region of pIAPV-WT ...................................... 169 B.2 List of primers used for site-directed mutagenesis ......................................... 170 Appendix C SHAPE analysis of wild-type and mutant IAPV IGR IRESs .................... 174 Appendix D Prediction of the structure and function of ORFx ...................................... 176 Appendix E Overexpression of ORFx has little effect on cotransfected reporter translation or host translation ......................................................................................... 178 Appendix F Positioning of the 80S ribosomes on the mutant IRESs ............................. 181 Appendix G Translational activities of mutant IRESs ................................................... 182  ix  List of Tables Table 1.1  Dicistroviruses and their natural hosts ................................................................ 22 Table 3.1 IRES mutations that confer 0 or +1 frame translation. ......................................... 94  x  List of Figures Figure 1.1 Model of cap-dependent translation initiation in eukaryotes. ............................... 3 Figure 1.2 Model of eukaryotic translation elongation cycle. ................................................ 6 Figure 1.3 A scheme of four viral IRESs groups. ................................................................ 12 Figure 1.4 The simultaneous slippage models of programmed -1 ribosomal frameshifting. .............................................................................................................................................. 15 Figure 1.5 Models for programmed +1 frameshifting in yeast Ty1 elements. ..................... 17 Figure 1.6 Models for programmed +1 frameshifting in yeast Ty3 elements. ..................... 19 Figure 1.7 Genome organization of dicistroviruses. ............................................................ 23 Figure 1.8 The bioinformatic evidence for an overlapping ORFx in dicistroviruses. .......... 26 Figure 1.9 The secondary structure of the IGR-IRES of 14 dicistroviruses......................... 31 Figure 1.10 Representative type I and II IGR IRES secondary structures. .......................... 34 Figure 1.11 The crystal structure of the type I IGR IRES at 3.1 Å resolution. .................... 35 Figure 1.12 Mimicry of a tRNA by PKI of the IGR IRES and a comparison with an authentic tRNA. .................................................................................................................... 36 Figure 1.13 Cryo-EM maps of the vacant 40S subunit and the 40S-CrPV IRES complex. 38 Figure 2.1 Identification of the initiation codon of ORFx translation in IAPV genomic cDNA. ................................................................................................................................... 50 Figure 2.2 Compensatory mutations restore +1 frame ORFx translation. ........................... 54 Figure 2.3 U6562∕G6618 base pairing directs +1 frame ORFx translation in IAPV genomic cDNA. .................................................................................................................... 56 Figure 2.4 Diagram of a bicistronic reporter construct, pIAPV-WT. .................................. 57 xi  Figure 2.5 U6562∕G6618 base pairing directs +1 frame ORFx translation in bicistronic reporter.................................................................................................................................. 60 Figure 2.6 Identification of the start site of ORFx by mass spectrometry............................ 61 Figure 2.7 Detection of KBV, ABPV, and IAPV from virus-infected honey bee pupae by RT-PCR using virus-specific primers................................................................................... 63 Figure 2.8 Detection of ORFx in virus-infected honey bee pupae. ...................................... 64 Figure 3.1 Mutagenesis of the U6562/G6618 base pair adjacent to PKI. ............................ 77 Figure 3.2 Translation of C6620G mutant IRES. ................................................................. 79 Figure 3.3 SHAPE analysis of the PKI domain of the WT and mutant IAPV IGR IRES. .. 83 Figure 3.4 Translational activities of mutant IGR IRESs within PKI codon-anticodon base pairing region ........................................................................................................................ 87 Figure 3.5 Specific nucleotides direct IAPV IGR IRES-mediated 0 or +1 frame translation. .............................................................................................................................................. 90 Figure 3.6 Translational activity of the Type II IGR IRESs. ............................................... 92 Figure 3.7 Effects of the SLIII and VLR on 0 and +1 frame translation ............................. 96 Figure 3.8 SHAPE analysis of more mutant IAPV IGR IRESs. .......................................... 99 Figure 3.9 SHAPE analysis of WT IAPV IGR IRES in the unbound and 80S-bound forms. ............................................................................................................................................ 102 Figure 3.10 SHAPE analysis of mutant IAPV IGR IRESs bound to 80S ribosomes......... 107 Figure 3.11 Toeprinting analysis of 80S ribosomes on mutant IAPV IGR IRESs. ........... 110 Figure 4.1 ORFx alignment in honey bee viruses. ............................................................. 116 Figure 4.2 The effect of ORFx on cell viability by propidium iodide................................ 122 Figure 4.3 Subcellular localization of tagged-ORFx in Drosophila S2 cells. .................... 124 xii  Figure 4.4 Localization of tagged-ORFx in Drosophila S2 cells with subcellular markers. ............................................................................................................................................ 127 Figure 4.5 Subcellular fractionation of S2 cells expressing tagged-ORFx. ....................... 128 Figure 4.6. Localization of mCherry-NLS in GFP-tagged-ORFx expressing S2 cells. ..... 129 Figure 4.7 Localization of GFP tagged-ORFx and nuclear lamin Dm0 during CrPV infection in Drosophila S2 cells. ........................................................................................ 132 Figure 4.8 Localization of GFP tagged-ORFx and ER marker KDEL during CrPV infection in Drosophila S2 cells. ....................................................................................................... 134 Figure 4.9 Viral titer of CrPV infected tagged-ORFx transfected S2 cells. ....................... 135  xiii  List of Symbols and Abbreviations A site: acceptor site Å: Angstrom 10-10 meters aa-tRNA: aminoacyl-tRNA ABPV: acute bee paralysis virus  ATP: adenosine triphosphate BiP: human immune-globulin heavy chain binding protein CCD: colony collapse disorder CMV: cytomegalovirus CrPV: cricket paralysis virus cryo-EM: cryo-electron microscopy CSFV: classical swine fever virus DCV: Drosophila C virus DNA: deoxyribonucleic acid DTT: di-thiothreitol, a reducing agent E site: exit site EDTA: ethylene-di-amine-tetra-acetic acid eIFs: eukaryotic initiation factors EMCV: encephalomyocarditis virus FGF-2: fibroblast growth factor 2 FLuc: Firefly luciferase FMDV: foot and mouth disease virus GTP: guanosine triphosphate HCV: hepatitis C virus HSV-1: herpes simplex virus type-1 IAPV: Israeli acute paralysis virus IC: initiation complex  IGR: intergenic region INM: inner nuclear membrane xiv  IRES: internal ribosome entry site ITAFs: IRES trans-activating factors KBV: Kashmir bee virus  kDa: kilodalton = 1000 gram/mole LC-MS: liquid chromatography mass spectrometry Met-tRNAi: initiator methionyl-tRNA MRM: multiple reaction monitoring mRNAs: messenger RNAs mRNP: messenger ribonucleoprotein MS: mass spectrometry NLS: nuclear localization signals NMIA: N-methylisotoic anhydride NMR: nuclear magnetic resonance spectroscopy NPCs: nuclear pore complexes ORF: open reading frame P site: peptidyl-tRNA site PABP: poly (A)-binding protein PAGE: poly-acrylamide gel electrophoresis PIC: pre-initiation complex PK: pseudoknot PSIV: Plautia stali intestine virus PTV-1: porcine teschovirus 1 PV: Poliovirus RdRP: RNA-dependent RNA polymerase RhPV: Rhopalosiphum padi virus  RLuc: Renilla luciferase RNA: ribonucleic acid RRL: rabbit reticulocyte lysate RT-PCR: Reverse transcription polymerase chain reaction SDS: sodium dodecyl sulfate xv  SG: stress granules SHAPE: selective 2'-hydroxyl acylation analyzed by primer extension analysis SINV-1: Solenopsis invicta virus  siRNA: small interfering RNA SL: stem loop TC: ternary complex TMEV: Theiler’s Murine encephalomyelitis virus TSV: Taura syndrome virus UTR: untranslated region VEGF: vascular endothelial growth factor VLR: variable loop region  Δ: delta               xvi  Acknowledgements I would like to thank my supervisor, Dr. Eric Jan, for his guidance to study in his lab. I feel extremely honored to have him as my mentor that he provided research opportunities and learning experiences in my projects. I am so grateful that he is always there to give me the constructive criticism or praise. Thank you for all your insightful suggestions. My supervisory committee members, Prof. George A. Mackie, Prof. Leonard J. Foster, Prof. François Jean and Prof. David Theilmann, have been a phenomenal source of inspiration and wisdom. Thank you all for the insightful discussion during my Ph.D. training study. Particularly, I would like to extend my deepest gratitude to Dr. George A. Mackie and Dr. Leonard J. Foster for all their precious advice and encouragement when I need one. I would also like to express my sincere thanks to Dr. François Jean and Dr. David Theilmann for sharing their expertise during my studies. I would like to thank all the past and present members of the Jan lab, including: Dr. Julianne Garrey, Dr. Qing Wang, Dr. Christopher Jang, Dr. Jennifer M. Bonderoff, Yun-Young Lee, Hilda Au, Anthony Khong, Seonghoon Lee, Julienne Jagdeo, Craig Kerr, Lisa-Marie Rauschendorfer and Min Ju Kim, for all the technical assistance, scientific discussion and emotional support.   I would also like to acknowledge the University of British Columbia and China Scholarship Council for financial supports. Lastly, I offer my greatest gratitude to my dearest family and friends for their love. I could not have finished my graduate work without your support. xvii  Dedication          This work is dedicated to my parents, Deyin Ren and Zonglan Xu.       This work is also dedicated to my dearest friend, Huan Bao.        谨以此文献给我亲爱的父母和家人。on 1  Chapter 1: Introduction 1.1 The canonical translation  1.1.1 Eukaryotic cap-dependent translation initiation Eukaryotic translation can be roughly divided into the following stages: initiation, elongation and termination/recycling. Although all stages can be the target of regulation, translational control is most commonly exerted during the initiation stage. In eukaryotes, translation initiation of the majority of messenger RNAs (mRNAs) occurs by a 5’ cap-dependent scanning mechanism. It is a highly regulated process which requires the participation of at least 12 eukaryotic initiation factors (eIFs) (Hinnebusch and Lorsch, 2012).  Figure 1.1 summarizes the basic outline of cap-dependent initiation (Aitken and Lorsch, 2012; Hinnebusch, 2011; Lorsch and Dever, 2010; Parsyan et al., 2011). The first step in initiation is the formation of the ternary complex (TC), initiator methionyl-tRNA (Met-tRNAi) -eIF2-GTP. With the help of eIF1, eIF1A, eIF3 and eIF5 binding to the ribosomal 40S subunit and inducing an ‘open’ conformation, the TC is recruited to assemble a 43S pre-initiation complex (PIC). Next, the 43S PIC is recruited to the mRNA near the 5’-7-methylguanosine cap by poly (A)-binding protein (PABP), eIF3, eIF4B and the eIF4F complex, which includes the cap binding protein eIF4E, an RNA helicase, eIF4A, and eIF4G. eIF4G is a scaffold protein which binds to eIF4E, eIF4A, PABP and eIF3. Both yeast and human eIF4G bind RNA. Their recruitment enables eIF4A, in conjunction with eIF4B, to prepare a single-stranded attachment site for the incoming PIC by unwinding of the cap-proximal region structure by the helicase activity of eIF4A. 2  eIF4G, via its interactions with eIF 4E and PABP, can bind to the cap and the poly (A) tail of mRNA to form a stable circular messenger ribonucleoprotein (mRNP), also termed as the “closed-loop” structure. The interaction between eIF4G and eIF3 brings the activated mRNP onto the 43S PIC.  Once loaded onto the transcript, the 43S PIC scans the 5’ untranslated region (UTR) for an AUG start codon. Base pairing forms between the anticodon of Met-tRNAi and the AUG codon in the peptidyl-tRNA (P) site. Once the 43S PIC recognizes the start codon, eIF1 is ejected. This triggers hydrolysis of eIF2-bound GTP to its GDP-bound form via gated phosphate Pi release and the action of the GTPase-activating protein factor eIF5 (Cheung et al., 2007; Maag et al., 2006; Maag et al., 2005; Unbehaun et al., 2004). eIF1 also contributes to the fidelity of initiation codon recognition. These events are associated with the transition of the PIC to a ‘closed’ conformation which stabilizes its interaction with Met-tRNAi and mRNA. Once the PIC adopts a closed conformation, eIF2-GDP and eIF5 dissociate. eIF5B, a ribosome-dependent GTPase, together with eIF1A mediates joining of the 60S subunit to form the 80S ribosome (Pestova et al., 2000; Unbehaun et al., 2004). Finally, subunit joining triggers eIF5B to hydrolyze its bound GTP, producing a conformational rearrangement of the 80S initiation complex (IC) (Pestova et al., 2000). Subsequent release of eIF1A prompts the 80S IC to enter the elongation phase of translation (Acker et al., 2006; Acker et al., 2009).    3     Figure 1.1 Model of cap-dependent translation initiation in eukaryotes. 4  Legend to Figure 1.1. The pathway of canonical translation initiation in eukaryotes can be grouped into seven steps. Initiation starts with the formation of eIF2-GTP-initiator Met-tRNA ternary complex (TC) (1). The ternary complex then binds to the 40S subunit to form the 43S preinitiation complex (PIC) with the help of eIF1, eIF1A, eIF3, and eIF5 (2). Meanwhile, the mRNA will bind to the cap-binding complex, eIF4F and the PABP to form an activated mRNP (3a), which allows it to recruit the 43S PIC (3b). The PIC then scans the mRNA for the start codon (AUG) (4). Upon start codon recognition, release of eIF1 and hydrolysis of eIF2-bound GTP arrest the scanning process (5). This is followed by the dissociation of eIF2-GDP and eIF5 and then eIF5B mediated 60S subunit joining (6). Subunit joining is followed by eIF5B mediated GTP hydrolysis and factors dissociation to form the 80S initiation complex (IC) (7). Adapted from Aitken and Lorsch, 2012.   1.1.2 Eukaryotic translation elongation After translation initiation, the 80S ribosome is positioned on the mRNA with the anticodon of Met-tRNAi base paired with the start codon in the P site. At this step, the ribosomal acceptor site (A site) is vacant to accept delivery of the next aminoacyl-tRNA molecule (aa-tRNA) to begin the translation elongation cycle (Dever and Green, 2012).  The elongation cycle proceeds in three main steps (Figure 1.2). First, the eukaryotic elongation factor eEF1A-GTP recruits aa-tRNA to the A site of the ribosome (Hopfield, 1974; Ninio, 1975; Ruusala et al., 1982; Thompson and Stone, 1977). If a cognate aa-tRNA is delivered, a proper codon-anticodon match is formed. Codon recognition leads to GTP hydrolysis of the eEF1A-GTP-aa-tRNA ternary complex. The eEF1A-GDP is subsequently released from the ribosome thus leaving the aa-tRNA to be accommodated in the ribosomal A site.  5  In the second step, the 60S ribosome subunit spontaneously catalyzes peptide bond formation (Pape et al., 1998), transferring the nascent peptide chain onto the A-site tRNA. At this time point, ratcheting of the ribosomal subunits induces the tRNAs in the P and A sites to adopt intermediate conformational states, called the hybrid P/E and A/P states. Here, the acceptor ends of the tRNAs shift to the exit (E-site) and P sites but the anticodon loops remain in the P and A sites (Moazed and Noller, 1989), respectively. Finally, translocation is achieved by the translocase eEF2. The hybrid state is recognized and stabilized by the GTP-bound translocon eEF2 (Dorner et al., 2006; Semenkov et al., 2000). Conformational changes of eEF2, subsequent hydrolysis of GTP, and Pi release catalyze the movement of tRNAs and mRNA, moving the P/E and A/P hybrid tRNAs into the E and P sites (Frank and Agrawal, 2000; Frank et al., 2007; Taylor et al., 2007).  Consequently, the deacylated tRNA is in the E site and the peptidyl-tRNA is in the P-site. The ribosomal A site becomes vacant and is available for the next aa-tRNA.   6   Figure 1.2 Model of eukaryotic translation elongation cycle. Aminoacyl-tRNA is recruited by elongation factor eEF1A in the presence of GTP which delivers it to the empty A-site of the ribosome. When a cognate tRNA enters the A-site, peptidyl transfer is then catalyzed by the ribosome, thereby transferring the nascent peptide onto the A-site tRNA. The tRNAs present in the A- and P-sites fluctuate between the classical and hybrid states. The hybrid state is recognized by the GTP-bound translocon eEF2, and hydrolysis of GTP drives the peptidyl tRNA into the P-site and the deacylated tRNA into the E-site, freeing the A-site for another round of elongation. Adapted from Schneider-Poetsch et al., 2010. 7  1.2 Non-canonical translation 1.2.1 Cap-independent translation initiation by an internal ribosome entry site Although the canonical cap-dependent scanning mechanism accounts for translation initiation of the majority eukaryotic mRNAs, there are alternative strategies that mRNAs use to start translation, including leaky scanning, ribosome shunting, reinitiation and internal ribosome entry (Gale et al., 2000). This thesis will focus on the mechanism of cap-independent translation initiation through an internal ribosome entry site (IRES). An IRES is a sequence element within an mRNA which recruits ribosome translation initiation complexes independent of the 5’-end of the mRNA, thereby bypassing the commonly utilized 5' terminal m7G cap.  The first IRES elements were found in picornaviruses, poliovirus (PV) and encephalomyocarditis virus (EMCV) (Jang et al., 1988; Pelletier and Sonenberg, 1988). PV is a positive sense, single-stranded RNA virus. PV infection leads to an acute shut off of host translation whereas viral translation is active. The 5’ UTR of PV is highly structured and contains a number of AUG codons that are not utilized. Furthermore, instead of a 5`-7-methylguanosine cap, the PV genomic RNA contains a 5'- terminal pU residue linked to Vpg (Nomoto et al., 1976). These properties are not commonly observed in cap-dependent translation. Using bicistronic reporter assays, several authors showed that the 5’ UTR of PV or EMCV, which is inserted within the intercistronic region of the bicistronic construct, mediates translation initiation of the downstream cistron independently of the upstream cistron (Jang et al., 1988; Pelletier and Sonenberg, 1988). These results demonstrated that 8  the PV and EMVC 5’ UTRs utilize a cap-independent translation initiation mechanism in which ribosomes are recruited by an IRES.  To define an IRES, the bicistronic reporter assay is the hallmark test. As described above, the candidate IRES sequence is inserted between two reporter genes. To prevent readthrough of the stop codon of the upstream cistron or continual scanning of the ribosome, structured RNA sequences that are known to possess no IRES activity, like the delta (Δ) EMCV IRES, are inserted after the first cistron. Additional controls are also required to prove that there is no cryptic promoter activity or cryptic splice sites, which could result in false translation initiation start sites and expression of the downstream cistron through a cap-dependent manner. Northern blot analysis is commonly used to check the integrity of the mRNA transcribed from the bicistronic reporter gene. Reverse transcription polymerase chain reaction (RT-PCR) is also used to detect mRNA encoded by the bicistronic reporter. In addition, siRNA knockdown of individual cistrons, and direct transfection of bicistronic mRNAs are also approaches that are utilized (Jiang et al., 2007; Kim et al., 2007; Reboll et al., 2007; Van Eden et al., 2004). Using these methods, IRESs were discovered in other viruses, including hepatitis C virus (HCV), some DNA viruses and dicistroviruses (Cevallos and Sarnow, 2005; Domier et al., 2000; Griffiths and Coen, 2005; Sasaki and Nakashima, 1999; Tsukiyama-Kohara et al., 1992; Wilson et al., 2000b). In addition to viral IRESs, cellular IRESs which facilitate protein expression have also been discovered. Such genes include those encoding human immunoglobulin heavy-chain binding protein (BiP) (Macejak and Sarnow, 1991), human fibroblast growth factor 2 (FGF-2) (Vagner et al., 1995), the proto-oncogene c-myc and the vascular endothelial growth factor (VEGF)  (Stein et al., 1998). Cellular IRESs can continue or induce 9  translation under the conditions when cap-dependent translation is suppressed, including cellular stress and virus infection. However, mutations in cellular IRESs do not fully abolish translational activity (Coldwell et al., 2000; Stoneley et al., 1998), suggesting that cellular mRNAs can use both the canonical cap-dependent and IRES mediated cap-independent mechanisms. It is very difficult to separate the two, leading to debates whether cellular IRESs have been properly characterized (Kozak, 2003, 2005).  Though the primary sequences of viral IRESes differ, their RNA secondary structures are found to be conserved within each viral family. Thus, covariation is utilized to predict structure models for viral IRESes (Baird et al., 2006). However, unlike viral IRESes, the relatively low similarity of sequences among cellular IRESes does not allow us to define a structure bioinformatically.  The importance of the IRES in eukaryotic translation initiation and in other viruses has led to extensive studies on IRES mechanism and function. Chemical and enzymatic probing, cryo-electron microscopy (cryo-EM), X-ray crystallography and nuclear magnetic resonance spectroscopy (NMR) analysis and biochemical reconstitution studies have been widely used to study these IRES RNAs. As an RNA element, the IRES structure is a primary focus to define its function. The most studied viral IRESs have been categorized into four groups based on their structure and mechanism of translation initiation (Kieft, 2008) (Figure 1.3). Group 4 IRESs members include poliovirus and rhinovirus.  This type of IRESs require Met-tRNAi, IRES trans-activating factors (ITAFs) and all eIFs except for eIF4E to initiate translation (Kieft, 2008). A well studied example is the poliovirus IRES. The viral 3C protease cleaves eIF4G (Etchison et al., 1982) during viral infection. Thus it inhibits the 10  host translation by blocking 43S complex formation. However, the viral IRES can utilize the C-terminal fragment of eIF4G to continue the translation initiation process (de Breyne et al., 2009; Pestova et al., 1996).  Similarly, Group 3 IRESs also require ITAFs, Met-tRNAi and all canonical eIFs except for eIF4E. However, unlike the Group 4 IRESs, this group of IRESs can directly recruit the ribosome to the start codon without ribosome scanning. Members of this group include EMCV, foot and mouth disease virus (FMDV) and Theiler’s murine encephalomyelitis virus (TMEV) (Kieft, 2008).  Group 2 IRES RNAs bind to the 40S subunit directly and use a subset of initiation factors that includes eIF2, eIF3, and Met-tRNAi to position the ribosome at the AUG start codon (Kieft, 2008; Pestova et al., 1998). These IRESs are found in HCV, classical swine fever virus (CSFV) and porcine teschovirus 1 (PTV-1). The best studied IRES within this group is the HCV IRES which is located in the 5’ UTR. Studies showed that the HCV IRES can bind to 40S subunits without the need for any initiation factors, requiring only eIF2-GTP-Met-tRNAi and eIF3 for the assembly of functional 40S and 80S ribosomes (Pestova et al., 1998). Additional biochemical experiments suggest that the HCV IRES can initiate translation independently of eIF2-GTP-Met-tRNAi, allowing HCV to circumvent the cellular eIF2-mediated inhibition of translation that occurs as a protective mechanism during viral infection (Lancaster et al., 2006; Terenin et al., 2008). However, it is interesting to note that the HCV IRES can still recruit Met-tRNAi when eIF2 activity is compromised. New findings have found that eIF2D/ ligatin serves to substitute for eIF2 and is utilized by the HCV IRES to deliver Met-tRNAi to the ribosome (Dmitriev et al., 2010; Skabkin et al., 2010). 11  Group 1 IRESs utilize the simplest mechanism of translation initiation. They can bind directly to the ribosome and do not require any initiation factors (Wilson et al., 2000a; Jan et al., 2003; Pestova and Hellen, 2003). Moreover, these IRESs also do not require initiator Met-tRNAi and start translation at a non-AUG initiation codon (Wilson et al., 2000a; Jan et al., 2003; Pestova and Hellen, 2003; Costantino et al., 2008).  Examples of this group of IRESs are dicistroviruses such as cricket paralysis virus (CrPV), Plautia stali intestine virus (PSIV), and Taura syndrome virus (TSV). I will discuss the Group 1 IRESs in more detail in the following sections (Chapter 1.4 and 1.5).     12   Figure 1.3 A scheme of four viral IRESs groups. Group 4 IRESs require Met-tRNAMeti, some eIFs, and ITAFs. They function efficiently in rabbit reticulocyte lysates only when supplemented with extracts from other cell types. Additionally, these IRESs initiation require scanning. Similarly, Group 3 IRESs also require Met-tRNAMeti, some eIFs, and ITAFs. But they are able to recruit the ribosome directly to the initation codon. Group 2 IRES RNAs are able to bind to the 40S subunit by only using Met-tRNAMeti and a subset of eIFs like eIF3 and eIF2. Finally, Group 1 IRES RNAs bind directly to the ribosome and do not require any initiation factors. Moreover, these IRESs also do not require initiator Met-tRNAMeti. The canonical cap-dependent translation is shown at top for comparison. Adapted from Kieft, 2008.  13  1.2.2 Recoding mechanisms The elongation process is inherently error-prone. Fortunately, missense errors and frameshift errors occur at a relatively low frequency (5 × 10-4 and 3 × 10-5 per codon) due to proofreading and quality control during translation (Parker, 1989). Many examples have shown that certain mRNA sequences can alter reading frames of translating ribosomes to generate authentic proteins. These sequences are termed recoding sites and fall into two broad classes. A class of programmed frameshift signals that direct -1 ribosomal frameshifting and recoding sites from the Ty family of retrotransposons in the yeast Saccharmomyces cerevisae that lead to programmed +1 frameshifting will be discussed here. In addition, some of the most studied +1 frameshift mechanisms of suppressor tRNAs and antizymes will also be discussed. As my thesis focuses on alternative reading frame selection by a viral IRES, these mechanisms will be helpful to gain insights into the broad mechanisms of translational fidelity.  1.2.2.1 Programmed eukaryotic -1 ribosomal frameshifting  Programmed -1 ribosomal frameshifting has been found most commonly in the genomes of eukaryotic RNA viruses. It was first defined as underlying the mechanism of Gag-Pol protein expression by Rous sarcoma alpharetrovirus (Jacks and Varmus, 1985). Eukaryotic ribosomal frameshifting signals are comprised of two essential elements: a ‘slippery sequence’ and a downstream stimulatory RNA structure. The slippery sequence contains the consensus X XXY YYZ (Figure 1.4). In vitro frameshift assays revealed that X represents any nucleotide, Y represents A or U, and Z represents A, C or U (Brierley et al., 14  1992; Dinman et al., 1991; Jacks et al., 1988a; Weiss et al., 1989). It is thought that frameshifting at this consensus motif occurs by encounter of the stimulatory RNA followed by the simultaneous slippage of two ribosome-bound peptidyl- and aminoacyl-tRNAs backward one nucleotide from XXY to XXX and from YYZ to YYY, respectively. The stimulatory RNA in most cases is a stem-loop or pseudoknot. A spacer region (5-8nt) between the slippery sequence and the stimulatory RNA is also required for maximal frameshifting efficiency (Brierley et al., 1989; Brierley et al., 1992; Kollmus et al., 1994). In the simultaneous slippage model, frameshifting is proposed to happen prior to peptide-bond formation (Jacks et al., 1988a). However, a later model proposed that frameshifting could also occur following peptide-bond formation when the tRNAs occupy the hybrid E-P and P-A sites (Weiss et al., 1989). The precise point in the elongation cycle when frameshifting occurs remains unclear.  The structure of -1 frameshift sites was found to be evolutionarily conserved in retroviruses (Jacks et al., 1988a; Jacks et al., 1988b; Jacks et al., 1987; Jacks and Varmus, 1985), coronaviruses (Brierley et al., 1987; den Boon et al., 1991; Denison et al., 1991), toroviruses (Snijder et al., 1990) and in a virus-like element in S. cerevisiae (Dinman et al., 1991; Icho and Wickner, 1989). Using -1 frameshifting as an expression strategy brings numerous advantages. It can maximize the coding capacity and efficiently achieve the regulation of overlapping gene expression through an mRNA.   15   Figure 1.4 The simultaneous slippage models of programmed -1 ribosomal frameshifting.  16  Legend to Figure 1.4. (A) Jacks and colleagues propose a model that slippage of the two tRNAs occurs after aminoacyl-tRNA occupies the A site but prior to peptide-bond formation. (B) Weiss and colleagues point out that the slippage occurs following peptide-bond formation when the tRNAs occupy the hybrid E-P and P-A sites. N, L, and I are the one-letter codes for the amino acids asparagine, leucine, and isoleucine, respectively. Adapted from Farabaugh, 1996.    1.2.2.2 Programmed +1 translational frameshifting 1.2.2.2.1 The Ty1 and Ty3 retrotransposons in yeast S. cerevisiae   In contrast to −1 programmed ribosomal frameshifting, where there is only one well-understood type of frameshifting signal, +1 programmed ribosomal frameshifting signals in the yeast Saccharomyces cerevisiae seem to be case specific. Recoding sites of the yeast Saccharomyces cerevisiae Ty1 and Ty3 retrotransposons cause the ribosome to frameshift into the +1 frame to synthesize Gag-Pol (Belcourt and Farabaugh, 1990; Farabaugh et al., 1993). The Ty family of retrotransposons includes a GAG and a POL gene, which encode a structural protein and an enzyme, respectively. The POL ORF overlaps the last few nucleotides of the upstream GAG gene, shifted in the +1 frame. Translational frameshifting of the Gag-pol protein means that the ribosome shifts from the 0 frame, for GAG, to the +1 frame, for POL. Ty elements replicate via an RNA intermediate by a process of reverse transcription. Replication leads to reintegration of the element into the genome at a new chromosomal location. The Ty1 class of elements was the first characterized +1 Ty frameshifting site. A 7 nucleotide sequence is both necessary and sufficient to promote frameshifting: CUU-AGG-17  C (shown as codons of the upstream GAG gene). Extensive studies suggest that the mechanism of Ty1 frameshifting occurs during the pause caused by the low availability of the cognate tRNACCU Arg , specific for AGG, subsequently leading to tRNAUAG Leu slippage by +1 from CUU to UUA. The tRNAUAG Leu could pair in either the 0 frame (CUU) or +1 frame (UUA) implying that frameshifting occurs by tRNA slippage (Figure 1.5).      Figure 1.5 Models for programmed +1 frameshifting in yeast Ty1 elements.  The mechanism is descripted in the text. The tRNAs are depicted in complex with eEF-1A (oval) and GTP (star). GDP is cartooned as a black circle. Groups of three boxes indicate the three ribosomal decoding sites E (exit), P (peptidyl), and A (aminoacyl). Watson/Crick 18  pairing is indicated by a vertical line, wobble pairing by a dot. Adapted from Sundararajan et al., 1999.   The programmed frameshifting site of the Ty3 retrotransposon was found to utilize a different mechanism. Ty3 is a retrovirus-like element in the yeast Saccharomyces cerevisiae (Clark et al., 1988; Hansen et al., 1988). It has been shown that Ty3 and animal retroviruses require similar sets of host genes for their cellular propagation (Irwin et al., 2005). The Ty3 genome encodes two overlapping proteins, Gag3 and Pol3, which are functionally similar to the retroviral gag and pol polypeptides, respectively (Farabaugh, 1995). After proteolytic cleavage, new peptides will be generated from Gag3 to function as viral nucleocapsid proteins for the formation of Ty3 virus-like particles in yeast cells. Three enzymatic activities, including protease, integrase and reverse transcriptase/RNaseH, which are essential for Ty3 transposition are derived by proteolytic processing of Pol3 (Farabaugh, 1995; Kim et al., 1998). A programmed ribosomal frameshift in the +1 direction happens in translation of the Pol3 polypeptide (Farabaugh et al., 1993; Vimaladithan and Farabaugh, 1994). +1 slippage occurs over a seven-nucleotide sequence GCG-AGU-U site. But unlike the Ty1 element, frameshifting at Ty3 does not involve P-site tRNA slippage from the 0 frame to the +1 frame. Rather, it is thought that Ty3-directed +1 programmed ribosomal frameshifting skips the first A of the 0-frame P-site codon followed by out-of-frame recruitment of the next aa-tRNA in the +1 frame to recognize the GUU codon (Figure 1.6). It was shown that frameshifting occurred when GCG was decoded by peptidyl tRNAIGCAla (Sundararajan et al., 1999). It is proposed that the unusual 19  I-G wobble interaction in the P site promotes frameshifting at the next codon. A downstream stimulatory element has also been proposed to increase frameshifting about 7.5-fold (Farabaugh et al., 1993).     Figure 1.6 Models for programmed +1 frameshifting in yeast Ty3 elements. The tRNAs are shown in complex with eEF-1A (oval) and GTP (star). GDP is shown as a black circle. Groups of three boxes indicate the three ribosomal decoding sites E (exit), P (peptidyl), and A (aminoacyl). Vertical line represents Watson/Crick pairing. Dot represents wobble pairing. Letter X means a purine–purine clash. Adapted from Li et al., 2001.   1.2.2.2.2 Antizyme and +1 frameshift suppressor tRNA A unique ribosomal frameshift mechanism is also utilized by cellular genes encoding antizymes. All antizymes have two overlapping open reading frames, a short ORF1 with a translational start codon and a stop codon at the frameshift site and a longer ORF2, which encodes the rest of the functional protein (Ivanov and Atkins, 2007). The 20  ORF2 is in the +1 reading frame in relation to ORF1. Therefore, a +1 ribosomal frameshift is required in order to translate a functional, full-length antizyme. The nucleotide sequence surrounding the frameshift site is highly conserved, which provides a great basis in identifying antizyme orthologues among various organisms. The mechanism of how the +1 translational frameshift is induced by polyamines is not yet well characterized.  In another example, tRNAs with an extra nucleotide in the anticodon loop could suppress certain +1 frameshift mutations (Atkins and Bjork, 2009). In characterization of those altered tRNAs, it was inferred that tRNAs having four nucleotides in the anticodon loop can correct the frame error in the mRNA which has an inserted nucleotide. Such four nucleotides in the anticodon allow a quadruplet translocation resulting in ribosome movement into the zero frame (Roth, 1981). It was proposed as a yardstick model by which reading frame maintenance is monitored by basepairs within the anticodon-codon interactor. However, Bjork and colleagues showed that most of the frameshift suppressor tRNAs do not fit the yardstick model (Nasvall et al., 2009). They showed that for the classical frameshift mutations sufA6 and sufB2, which both have an extra G-nucleotide in the anticodon loop, aatRNAs are delivered to the A site inefficiently. After a three nucleotide translocation to the P site, the near-cognate tRNA slips into an overlapping reading frame (Qian et al., 1998).  1.3 The Dicistroviridae viral family The Dicistroviridae is a family of positive sense, single stranded, non-enveloped RNA viruses. Originally, they were classified as picorna-like viruses based on their 21  biophysical and biochemical properties (Moore et al., 1980). Following the molecular characterization of their genome sequences, they have been classified as a distinct family, first as the genus of cricket paralysis-like viruses (Christian et al., 2000), and later renamed as family Dicistroviridae (genus Cripavirus) (Mayo, 2002).  The Dicistroviridae family derives its name from the bicistronic genome, as there are two ORFs in the viral genome and each ORF is translated by an independent IRES (Wilson et al., 2000b). Natural infections of the twenty dicistroviruses have been described to date (Table 1.1), the majority leading to paralysis or death. Several members are harmful to commercially and agriculturally important organisms. For example, Taura syndrome virus (TSV) has had a negative impact in the shrimp industry (Yang et al., 2006), and the Israeli acute paralysis virus (IAPV) is associated with honey bee colony collapse disorder (CCD) (Cox-Foster et al., 2007), which raises major concerns for agriculture and the environment. The cricket paralysis virus (CrPV) and Drosophila C virus (DCV) can infect a broad range of insects including Drosophila melanogaster. Drosophila tissue culture cells, which have been studied extensively, serve as an excellent model to study Dicistroviruses. Overall, it is necessary to understand dicistrovirus biology as they play an important role in agriculture and economy. The Dicistroviridae family has been subdivided into two genera (Table 1.1), Cripavirus and Aparavirus, based on phylogenetic distance and the secondary structure of the IGR IRES.  22  Table 1.1  Dicistroviruses and their natural hosts Dicistrovirus  Acronym Orders of natural hosts Genus: Cripavirus Cricket paralysis virus  CrPV Diptera, Hemiptera, Hymenoptera, Lepidoptera, Orthoptera Drosophila C virus  DCV Diptera Aphid lethal paralysis virus  ALPV Hemiptera Black queen cell virus  BQCV Hymenoptera Himetobi P virus  HiPV Hemiptera Plautia stali intestine virus  PSIV Hemiptera Rhopalosiphum padi virus  RhPV Hemiptera Triatoma virus  TrV Hemiptera Homalodisca coagulata virus-1 HoCV-1 Hemiptera Genus: Aparavirus Acute bee paralysis virus  ABPV Hymenoptera Kashmir bee virus  KBV Hymenoptera Israeli acute paralysis virusb  IAPV Hymenoptera Solenopsis invicta virus-1  SINV-1 Hymenoptera Taura syndrome virus  TSV Decapoda Mud crab dicistrovirus MCDV Decapoda Tentative members of the family Cloudy wing virus CWV  Blackberry virus Z   Acheta domesticus virus   Ervivirus   Bombyx mori infectious flacherie virus BmIFV   23  1.3.1 Genomic organization  The genome of dicistroviruses is approximately 8–10 kb in length and contains a 3’-poly (A) tail (Bonning and Miller, 2010). The genome does not contain a 5’ 7-methylguanosine cap but instead has a viral protein genome-linked VPg covalently linked to the 5’ end of the genomic RNA (King and Moore, 1988). Radiolabeling of intracellular RNA in infected cells and Northern blot analysis reveal that no subgenomic RNA is produced (Eaton and Steacie, 1980; Wilson et al., 2000b).  Unlike viral families in the order Picornavirales containing one long open reading frame with the non-structural genes downstream of the structural genes, Dicistroviridae has two ORFs. The upstream ORF (ORF1) encodes the non-structural genes including a suppressor of host RNA-mediated silencing (in DCV and CrPV) (van Rij et al., 2006; Wang et al., 2006), helicase, protease, VPg, and RNA-dependent RNA polymerase (RdRP). The downstream ORF (ORF2) encodes the structural genes, VP1–4. Translation of ORF1 is mediated by an IRES within the 5’ UTR, whereas ORF2 translation is driven by the IGR IRES (Fig. 1.7) (Wilson et al., 2000b).      Figure 1.7 Genome organization of dicistroviruses. 24  Legend to Figure 1.7. The dicistrovirus genome contains two ORFs indicated by orange and green boxes, a poly (A) tail at the 3’ end and VPg at the 5’ end. The sphere on the left of genome represents the VPg. The ORF1 encodes the viral nonstructural proteins, which are currently known as a silencing suppressor (in CrPV and DCV) (SS), a superfamily 3 helicase (hel), genome-linked protein (VPg), a 3A-like protein, a chymotrypsin-like cysteine protease (except for 2A pro) (pro) and a RNA-dependent RNA polymerase (RdRp). The ORF2 encodes the structural proteins, VP2, VP4, VP3 and VP1. The translation of the two ORFs is driven by two different internal ribosome entry sites (IRESs), one in the 5’ untranslated region (UTR) and the other in the intergenic region (IGR). Adapted from Bonning and Miller, 2010.    Compared to the IGR IRES, the IRES present in the 5’ UTR has not been well studied. There is very low sequence and structural homology across Dicistroviridae. Translation of ORF1 initiates at a predicted AUG codon (Domier et al., 2000; Sasaki and Nakashima, 1999; Wilson et al., 2000a; Wilson et al., 2000b). The Rhopalosiphum padi virus (RhPV) 5’ UTR IRES has been characterized as a model. It functions efficiently in a number of systems including mammalian, plant, and insect cells and in vitro translation systems (Royall et al., 2004; Woolaway et al., 2001). Deletion analyses demonstrate that different regions or fragments (nt 425–579 and nt 300–429) of the RhPV 5' UTR display IRES activity (Groppelli et al., 2007). By using in vitro reconstitution assays, only the initiation factors eIF1, eIF2, and eIF3 are required for 48S complex formation on the 5’ IRES (Terenin et al., 2005). The factors eIF1A and eIF4F stimulate the assembly of 48S complexes on the RhPV RNA, but are not essential (Terenin et al., 2005).  25  In contrast, the IGR IRES folds into a compact structure and is largely conserved throughout the Dicistroviridae family viruses (Jan and Sarnow, 2002; Kieft, 2008; Nakashima and Uchiumi, 2008). The IGR IRES can directly recruit the ribosome without any eIFs or the initiator Met-tRNAi (Wilson et al., 2000a). Furthermore, the IGR IRES occupies the P site of the ribosome and initiates translation from the ribosomal A site at a non-AUG codon (Wilson et al., 2000a). Extensive biochemical and structural analyses on this IGR IRES will be described in following sections. Recently, additional features of the Dicistroviridae genome have been revealed using bioinformatics analyses. Interestingly, there is a short open reading frame, ORFx (or predicted overlapping gene, pog) located in the +1 reading frame which overlaps the 5’ proximal region of the ORF2 in members of Aparaviruses including honey bee viruses acute bee paralysis virus (ABPV), Israeli acute paralysis virus (IAPV), Kashmir bee virus (KBV) and the fire ant virus Solenopsis invicta virus (SINV-1) (Firth et al., 2009; Sabath et al., 2009) (Fig. 1.8).    26    Figure 1.8 The bioinformatic evidence for an overlapping ORFx in dicistroviruses. (A) The genomic scheme of KBV (NC_004807). (B2-B4) After alignment of 16 bee paralysis virus CDS2 sequences, positions of stop codons in the three ORFs are shown. The absence of stop codons in the +1 frame indicates the presence of ORFx. Adapted from Firth et al., 2009.   1.3.2 Modulation of host translation machinery As described in the previous section, the two ORFs of dicistrovirus are translated by distinct IRESs. This strategy overcomes a limitation of gene expression by translation of a single polyprotein in which all viral proteins are translated in equimolar quantities. During dicistrovirus infection, the expression of the structural proteins is in supramolar excess over the nonstructural proteins (Moore et al., 1980).  The mechanisms by which dicistroviruses evade the host antiviral response and hijack the host translational machinery for productive infection are still poorly understood. 27  CrPV infection of Drosophila S2 cells is used as a model system. Host translation is shut down early in infection and viral proteins are predominately synthesized during infection (Garrey et al., 2010; Moore et al., 1980; Wilson et al., 2000b). The mechanism by which this occurs was analyzed by initially assessing the activity of canonical translational initiation factors. CrPV infection led to the inhibition of the interaction of eIF4E (the cap binding protein) and eIF4G (the major scaffolding protein within the eIF4F mRNA binding complex) starting as early as 1 h.p.i., therefore promoting the synthesis of viral proteins and correlating with the shutoff of host translation (Garrey et al., 2010). CrPV infection is also concomitant with phosphorylation of eIF2α beginning at 3 h p.i.; however, this modification appears to be dispensable for viral production. It was initially proposed that phosphorylation of eIF2 α made mechanistic sense as host translation is shutoff while IGR IRES-mediated translation does not require eIF2 nor the initiator Met-tRNA. However, suppression of eIF2α phosphorylation does not prevent host translation shutdown, suggesting that other mechanisms, such as the impairment of the eIF4F complex, likely plays a more important role (Garrey et al., 2010; Khong and Jan, 2011). Thus, it is proposed that disruption of the interaction between eIF4E and eIF4G leads to host translation shutoff in CrPV infected cells which then leads to preferential IRES translation.  Other biological processes are also modulated during CrPV infection. CrPV and DCV infection of Drosophila S2 cells prevents the formation of stress granules (SG) even in the presence of SG inducers such as heat shock, Pateamine A and oxidative stress (Kedersha et al., 2002; Kedersha et al., 1999; Khong and Jan, 2011; Kimball et al., 2003). As the cellular machineries required for viral translation and replication are not sequestered in SGs, it allows virus infection to proceed (Khong and Jan, 2011). It was also 28  demonstrated that CrPV viral 3C protease is sequestered to SGs under cellular stress but not during virus infection (Khong and Jan, 2011).   1.4 The Dicistroviridae intergenic region internal ribosome entry site As described in the previous sections, the IGR IRES in dicistroviruses is remarkable because it can directly recruit ribosomes independently of initiation factors and Met-tRNAi. More specifically, the IRES can directly recruit 40S subunits followed by recruitment of 60S subunits to preformed IRES-40S complexes (Jan et al., 2003). The IGR IRES adopts a conformation by which one of its domains mimics a tRNA-like anticodon-codon interaction. 1.4.1 Structure of the IGR IRES The IGR IRES is about 200 nucleotides in length. Unlike the 5’ UTR IRES which has a high GC content, the dicistrovirus’ IGR IRESs are AU-rich. For example, the CrPV IGR IRES is 64 % AU (Plank and Kieft, 2012).  The IGR IRES was first identified in Plautia stali intestine virus (PSIV) (Nakashima et al., 1998; Sasaki and Nakashima, 1999, 2000). The PSIV IGR IRES mediates the translation of ORF2 at a CAA codon instead of an AUG codon. Also, computational analysis predicted that the PSIV IGR IRESs would fold into a complex structure with multiple stem loops (Sasaki and Nakashima, 1999) (Figure 1.9 PSIV). In combination with mutagenesis and compensatory mutagenesis, in vitro translation analyses identified that 29  stem loop (SL) VI interacts with downstream sequences to form a pseudoknot structure. This pseudoknot was shown to be essential for PSIV IGR IRES function (Sasaki and Nakashima, 2000). Using the mutational analysis, a tertiary structure model of the PSIV IGR IRES was proposed (Kanamori and Nakashima, 2001).  Similarly, an IGR IRES was found in CrPV (Wilson et al., 2000b). The CrPV IGR IRES mediates initiation at a GCU alanine codon. Mutagenesis and toeprinting analyses revealed that the CCU triplet, which precedes the GCU codon, occupies the ribosomal P site and the GCU triplet occupies the ribosomal A site. The CCU triplet is base paired with an upstream element to form a pseudoknot structure (Figure 1.10A CrPV). Disruption of base pairings in the psuedoknot inhibit IRES activity whereas compensatory mutations which restored the interactions rescue IRES activity. Thus, the integrity of the pseudoknot structure is important for IGR IRES activity (Wilson et al., 2000b).  Later, extensive biochemical studies using enzymatic and chemical probing confirmed that the IGR IRES adopts a complex overlapping triple pseudoknotted RNA structure (PKI, PKII and PKIII) (Jan and Sarnow, 2002; Nakashima et al., 2003; Nishiyama et al., 2003). Biochemical and structural analyses revealed that the three pseudoknots form a conserved two-domain architecture that is shared with all dicistrovirus IGR IRESs (Kieft, 2008, 2009; Nakashima and Uchiumi, 2009). Among the Dicistroviridae family members, most of the IGR IRES primary sequences are not conserved. However, there are exceptions for a few residues, such as residues in the loops of SL IV and V which are responsible for interactions with components of the 40S ribosomal subunit (Landry et al., 2009; Nishiyama et al., 2003; Nishiyama et al., 2007; Pfingsten et al., 2006; Schuler et al., 2006; Spahn et al., 2004). In 30  contrast, all the members adopt a similar tertiary structure (Figure 1.9). The secondary and tertiary structures of the IRESs play an important role in their function (Kieft, 2008, 2009; Nakashima and Uchiumi, 2009; Nishiyama et al., 2003).   31   Figure 1.9 The secondary structure of the IGR-IRES of 14 dicistroviruses. 32  Legend to Figure 1.9. Red asterisks and dots indicate base pairing in pseudoknots and stems, respectively. Corresponding regions between viruses are indicated by the same color. Adapted from Nakashima and Uchiumi, 2009.   The two domains of IGR IRES can fold independently (Costantino and Kieft, 2005; Costantino et al., 2008; Jan and Sarnow, 2002; Kanamori and Nakashima, 2001; Pfingsten et al., 2006; Schuler et al., 2006). PKII and PKIII form one domain which is primarily responsible for ribosome binding (Figure 1.10). Specifically, studies show that SL IV and V within this domain bind to rpS5 and rpS25 of the 40S ribosomal subunit, respectively (Nishiyama et al., 2003; Nishiyama et al., 2007; Pfingsten et al., 2006; Schuler et al., 2006; Spahn et al., 2004) (Figure 1.11). In order to recruit the 60S subunit, it appears that binding of the L1.1 region of the IRES to rpL1 is essential (Jang et al., 2009; Pfingsten et al., 2006) (Figure 1.11).  PKI forms the other domain which is responsible for ribosome positioning (Figure 1.10). Extensive biochemical and crystal structure analyses of the CrPV PKI domain reveal that it structurally mimics a tRNA (Figure 1.12) (Costantino et al., 2008), and has been modeled to occupy the P site of the ribosome (Costantino et al., 2008; Jan et al., 2003; Kamoshita et al., 2009; Pestova and Hellen, 2003; Wilson et al., 2000a; Zhu et al., 2011). This mimicry precisely mirrors the interaction between an initiator tRNA anticodon and the mRNA start codon. Docking of the PKI domain onto the ribosome suggest that this domain binds to the ribosome in a manner that mimics a P/E hybrid tRNA, the tRNA conformation that is present as an intermediate during the translation elongation cycle. Thus, the IRES is thought to set the ribosome into an elongation mode of translation. After positioning of the 33  ribosome by the tRNA mimicry domain in the P site, the cognate aa-tRNA is correctly delivered to the ribosomal A-site which leads to the ribosome to undergo a round of translocation without peptide bond formation called pseudotranslocation (Jan et al., 2003; Pestova and Hellen, 2003). Although adopting an overall similar secondary structure, the IGR IRESs can be grouped into two classes based on distinct features within the two domains, PKII/III and PKI (Jan, 2006; Nakashima and Uchiumi, 2009). The Type II IRESs, like the TSV IGR IRES, contain a longer L1.1 region and an extra SLIII within the PKI domain. In contrast, the Type I IRESs, such as the CrPV IGR IRES, have a smaller L1.1 region but no SLIII (Figure 1.10). Mutations within the L1.1 region of both CrPV and TSV IGR IRESs can inhibit IRES activity and disrupt 80S assembly, suggesting they might have common functions (Jang et al., 2009; Jang and Jan, 2010). The roles of SLIII are not fully understood. Deletion of SLIII within the TSV IGR IRES disrupts IRES translation, ribosome binding and positioning (Cevallos and Sarnow, 2005; Jang and Jan, 2010; Nakashima and Uchiumi, 2009).    34     Figure 1.10 Representative type I and II IGR IRES secondary structures. (A) Type I CrPV IGR IRES. (B) Type II TSV IGR IRES. Pseudoknots are labeled and a black horizontal line divides the two domains of the IRES. Numbering refers to the nucleotide position within the respective viral genome. Helical regions are indicated by a black dash between nucleotides. Conserved nucleotide positions are shown in uppercase and nonconserved nucleotides are in lowercase. Adapted from Jang and Jan, 2010.   35    Figure 1.11 The crystal structure of the type I IGR IRES at 3.1 Å resolution. PKI is shown in green, and PKII/PKIII are shown in red. The location of SL V, SL IV, and L1.1 are indicated with arrows and red boxes. Molecules that bind to the IGR IRES are coloured and indicated in boxes. Adapted from Pfingsten and Kieft, 2008.   36   Figure 1.12 Mimicry of a tRNA by PKI of the IGR IRES and a comparison with an authentic tRNA. The two views on the left, rotating 90° on a vertical axis, are the PKI domain of the CrPV IGR IRES. At right is the corresponding view of an authentic initiator tRNA anticodon–mRNA codon interaction in the P-site of a ribosome (PDB accession number 2J00). Corresponding bases in the structures are colored to match each other. Cyan in the tRNA–mRNA structure at right is the decoded mRNA. Adapted from Costantino et al., 2008.   37  1.4.2 Function of the IGR IRES In vitro reconstitution assays have revealed that the recruitment of 40S ribosomal subunits and the subsequent assembly of 80S ribosomes onto the CrPV IGR IRES can occur in the absence of any initiation factors and without GTP hydrolysis (Jan et al., 2003; Wilson et al., 2000a). The CrPV IGR IRES can bind to 40S ribosome subunits with high affinity (KD of 24 nM) (Costantino and Kieft, 2005; Jan and Sarnow, 2002). Preformed 80S ribosomes can directly bind to the IGR IRES without any initiation factors (Pestova et al., 2004). These IRES-80S ribosome complexes can start translation elongation in the presence of eEF1A, eEF2 and aa-tRNAs (Jan et al., 2003). In the case of the CrPV IGR IRES, the first codon is a GCU alanine codon. Moreover, it has been shown that the IGR IRES can initiate from all codons except stop codons (Shibuya et al., 2003). Biochemical studies showed that PKII and PKIII of the IGR IRES fold into a tightly packed core that can bind to 40S ribosomal subunits (Costantino and Kieft, 2005). To investigate the interaction of IGR IRES with the ribosome, cryo-EM reconstructions were used to solve the CrPV IGR IRES-40S and IGR IRES-80S complex structures (Spahn et al., 2004) (Figure 1.13). The IGR IRES occupies the E, P and partially the A site of the ribosome. It appears that the IGR IRES interacts with helices 18 and 34 of the small rRNA and ribosomal protein S5 (rpS5) (Figure 1.11 and 1.13). This is in agreement with later biochemical data that SL IV and SL V can bind to rpS25 (Landry et al., 2009; Nishiyama et al., 2003; Nishiyama et al., 2007). The conserved L1.1 loop region of the CrPV IGR IRES appears to interact with the 60 S ribosomal subunit and the L1 stalk, in particular rpL1 and H77 of the 28S rRNA (Figure 1.11).  38      Figure 1.13 Cryo-EM maps of the vacant 40S subunit and the 40S-CrPV IRES complex. (A) 40S subunit, (B) the 40S-CrPV IRES complex. The 40S subunit is shown from the intersubunit side in yellow and the IRES is pink. Landmarks for the 40S subunit are the following: b, body; bk, beak; h, head; pt, lf, left foot; rf, right foot; pt, platform; and sh, shoulder. The position of five 18S rRNA helices and of protein rpS5 are indicated, as identified by comparison with a cryo-EM map of the yeast 80S ribosome. (C) Secondary structure diagram of the CrPV IRES. Nucleotides belonging to pseudoknots PK I–PK III are in different colors. The triplets that are positioned in the P and A sites during the first A site occupation are designated P and A, respectively. Ala and Thr refer to the first two amino acids that are incorporated into the viral protein. The dashed arrows indicate the tentative location of PKI and PKII/III in the cryo-EM density map. Adapted from Spahn et al. 2004.   39  Conformational changes are also induced by the IGR IRES binding to the 40S subunit and the 80 ribosome. IRES binding induces a conformational change at the mRNA entry channel. Rotation of the head of the 40S subunit and a latch interaction between helix 18 and helix 34 of the rRNA is formed. This has been proposed to guide the mRNA into the channel of the ribosome (Spahn et al., 2004).  Moreover, the ribosomal P proteins in the stalk region of the 60S subunit become extended (Spahn et al., 2004). This region has been shown to be responsible for interacting with elongation factors (Agrawal et al., 1999; Gomez-Lorenzo et al., 2000). Thus, it suggests that the IGR IRES induces this conformational change to facilitate elongation factor binding (Jan, 2006; Spahn et al., 2004).  1.5 Thesis investigation Bioinformatics analyses revealed an overlapping +1 ORF (ORFx) within the structural protein ORF within a subset of dicistroviruses, including the IAPV, KBV, ABPV and SINV-1 (Firth et al., 2009; Sabath et al., 2009). Despite extensive structural and biochemical studies of CrPV and PSIV IGR IRESs, the honey bee viral IGR IRESs have not been  charaterized. Moreover, many key questions regarding whether the ORFx is expressed and how it is translated remain to be investigated. To address these questions, this thesis is mainly focused on the characterization of the IAPV IGR IRES and its mechanism of mediating alternative reading frame translation of ORFx. In Chapter 2, we hypothesize that the IGR IRES directs translation of ORFx and of the 0 frame ORF encoding viral structural polyprotein. Using IAPV as a model, we identify the existence of 40  ORFx and the start site of the alternative +1 reading frame. Chapter 3 examines how this reading frame is selected through the use of essential nucleotides within the IGR IRES tRNA-mimicry domain. The work shows that essential elements in the IGR IRES determine the structural differences that mediate alternative reading frame translation. In Chapter 4, the localization of ORFx is studied in Drosophila S2 cells to gain insight into the role of ORFx. Finally, Chapter 5 summarizes and discusses all chapters. Understanding the correlation between alternative reading frame selection and the IGR IRES will shed insight into better understanding of IRES-mediated translation initiation and the mechanism that this RNA virus uses to increase its coding capacity. 41  Chapter 2: Identification of the +1 Frame ORFx A version of this chapter has been published, and has been adapted with permission: Alternative reading frame selection mediated by a tRNA-like domain of an internal ribosome entry site. Ren, Q. *, Wang, Q.S. *, Firth, A.E., Chan, M.M.Y., Gouw, J.W., Guarna, M.M., Foster, L.J., Atkins, J.F., and Jan, E.  Proc Natl Acad Sci U S A. 2012 Mar 13; 109 (11):E630-9 (*shared first authorship).  2.1 Introduction As described in the previous sections, the Dicistroviruses possess a positive-sense, monopartite single-stranded RNA genome that encode nonstructural and structural polyproteins within ORF1 and ORF2 separated by an IGR IRES. The IGR IRES of the Dicistroviridae family is the simplest IRES to date. What makes this IRES unique is its ability to directly recruit the ribosome without the aid of translation initiation factors or the initiator Met-tRNAi. The IGR IRES adopts a conformation that interacts with and manipulates the ribosome. The end result is that the IGR IRES directs translation of the viral structural protein ORF in the 0 frame from the ribosomal A site.  Viruses have evolved elegant strategies to optimize the coding capacity of compact genomes through the utilization of overlapping genes whereby one transcript codes for two or more different proteins in different reading frames. Previous bioinformatic analysis revealed that a subset of dicistroviruses contained a hidden gene, called ORFx, in the +1 frame which overlapped with the 0 frame viral structural protein ORF just downstream of the IGR IRES (Firth et al., 2009; Sabath et al., 2009). This subset of dicistroviruses 42  includes honey bee viruses ABPV, IAPV and KBV and the fire ant virus SINV-1 (Firth et al., 2009; Sabath et al., 2009). Conserved tandem AUG codons, which provide one potential initiation site for ORFx translation, are located approximately 100 nucleotides downstream of the PKI domain. Interestingly, the honey bee viruses also contain a 14-18 bp stem-loop structure (SLVI) immediately 5’ to the IGR IRES. In these viruses, the ORF1 stop codon is positioned within SLVI (Firth et al., 2009) (Figure 2.1 a).  By using IAPV as a model, this chapter examines how ORFx is translated in vitro and determines whether ORFx is expressed during virus infection. Using extensive mutagenesis studies, we showed that IAPV IGR IRES direct translation of two overlapping open reading frames to produce the 0 frame viral structural polyprotein and an ORF in the +1 frame called ORFx. A U-G wobble base pair adjacent to the tRNA-mimicking domain of the IGR IRES mediates +1 frame ORFx translation. By mutagenesis and mass spectrometry analysis, we identified the start site of ORFx as GCG (3’ to the IGR IRES) alanine codon. Furthermore, an ORFx peptide was detected in virus-infected honey bees by multiple reaction monitoring mass spectrometry. In summary, this study not only reveals a novel viral strategy to increase the coding capacity of a viral genome but also uncovered a novel feature of the IGR IRES-mediated translation mechanism.   43  2.2 Material and methods 2.2.1 DNA constructs and reagents The cDNA plasmid containing the full-length IAPV genomic sequence (NC_009025) was synthesized by Operon Inc. and cloned into the pcDNA3.1 vector using restriction sites KpnI and XbaI. The bicistronic luciferase plasmid, pIAPV-WT, contains the IAPV IGR IRES and the downstream coding sequence of ORF2 (nucleotides 6399–6908). The AUG start codon of FLuc was deleted and the stop codon of ORFx UAG was mutated to UAU. A G was inserted at nucleotide 6909 to put ORFx in frame with FLuc (Appendix B1). All mutations of the full-length IAPV cDNA or bicistronic reporter were created using a site-directed mutagenesis protocol (Stratagene) and further confirmed by sequencing. Primers are listed in Appendix B2.  2.2.2 In vitro translation assay Plasmids containing the full-length IAPV genomic cDNA and the bicistronic reporter were linearized with XbaI. The linearized DNA (1 µg / µl) was incubated with 6.7 µl Sf21 extract (Promega), 0.5 mM Magnesium acetate and 40 mM potassium acetate (final concentration) to final volume of 10 µl for 3 h at 30°C. The expression of protein products was measured by incorporation of 0.4 µl of [35S]-methionine (> 1,000 Ci/ mmol; Perkin Elmer) per 10 µl reaction and then analyzed by SDS-PAGE gel followed by phosphor imaging using a Typhoon phosphorimager (Amersham).  44  2.2.3 Western Blot analysis Proteins separated by SDS-PAGE were transferred to a polyvinylidene difluoride Immobilon -FL membrane (Millipore). Blots were blocked for 1 h in 5% skim milk and Tris buffered saline (20 mM Tris, 150 mM NaCl, 0.1% Tween-20). Antibodies used were as follows: affinity-purified IAPV ORFx peptide rabbit polyclonal antibody (raised against the IAPV ORFx peptide sequence C-SLRPLVKTRLR PKKC-NH2; GenScript) (1∶500), affinity-purified IAPV VP2 peptide rabbit polyclonal antibody (raised against the IAPV VP2 peptide sequence C-NTMPGDSQQESNTPC-NH2; GenScript) (1∶500), anti-Luciferase (Promega) (1∶1,000). Primary antibodies were detected using either IRDye 800CW goat anti-rabbit IgG (LI-COR Biosciences) at 1∶10,000 at room temperature for 1 h, and subsequently scanned using an Odyssey imager (LI-COR Biosciences), or using a 1∶10,000 dilution of donkey anti-goat IgG-horseradish peroxidase (Promega) and subsequently detected by enhanced chemiluminescence (Millipore).  2.2.4 Northern Blot analysis RNA was isolated using TRIzol reagent (Invitrogen). Equal amounts of RNA were separated on a formaldehyde-containing agarose gel and transferred to Zeta-probe blotting membrane (Bio-Rad). Radiolabeled DNA hybridization probes were generated using a RadPrime kit (Invitrogen). Blots were analyzed by autoradiography (Typhoon, Amersham).   45  2.2.5 Mass spectrometry analysis  In vitro translated ORFx was isolated from an SDS PAGE gel. It was followed by Arg-C in-gel digestion. Samples were eluted from stop and go extraction tips with 20 µL of 80% acetonitrile in 0.5% acetic acid, vacuum-dried, reconstituted in sample buffer (1% trifluoroacetic acid and 3% acetonitrile in 0.5% acetic acid) and were injected (4 µL/ min, solvent A) onto a large capacity chip (II) consisting of a 160nL trap column and 150-mm analytical column (G4240-62010; Agilent Technologies). It was analyzed on a triple quadrupole mass spectrometer (6460; Agilent Technologies)  2.2.6 Virus infection in honey bee Honey bee pupae (approximately 17 d old) injected with a mixture of IAPV, KBV, and ABPV were dissected on ice to separate the head and body. Samples were bead-homogenized in 1% sodium deoxycholate and 50 mM NH4HCO3 at three pulses of 20 s and 6.5 m/ s, followed by incubation at 99 °C for 5 min. The lysate was centrifuged at 8,000 × g for 10 min, and the resulting supernatant was further centrifuged at 5,000 × g for 5 min, all at 4 °C. The final supernatant was subjected to ethanol precipitation. Proteins precipitated after 1.5 h incubation at RT were collected by centrifugation at 16,100 × g. The protein pellet was resolubilized in 6 M urea, 2 M thiourea, and 100 mM Tris-Cl (pH 8.0), and any insoluble material was subsequently removed by centrifugation at 16,100 × g. Resolubilized proteins were resolved on a 16% SDS gel, and a band corresponding to the molecular mass of ORFx was excised and in-gel digested with trypsin. The resulting peptides were stored on stop and go extraction tips until MS analysis (Chan et al., 2006). 46   2.2.7 RT-PCR detection assay RNA was extracted from mock-infected and virus-infected honey bees and subjected to RT-PCR using the following primers as described (de Miranda et al. 2010). IAPV:  5’-AGACACCAATCACGGACCTCAC-3’, 5’-GATTTGTCTGTCTCCCAGTGCACAT-3’; KBV: 5’-GTTTCTATGCAAATCGCA-3’, 5’-CCATCCAGGCACATTCTG-3’; ABPV: 5’-GGAACATGGAAGCATTATTG-3’, 5’-AATGTCTTCTCGAACCATAG-3’.  2.3 Results 2.3.1 ORFx is expressed from a Full-Length IAPV cDNA in in-vitro insect Sf21 extracts To determine whether ORFx can be produced, we obtained a full-length IAPV genomic cDNA (synthesized by Operon) and inserted an A at nucleotide 7348 to create a stop codon near the C terminus of VP2. From 5’ to 3’, ORF2 encodes the VP2, VP4, VP3, and VP1 structural proteins that have predicted molecular masses of 35.5, 7.2, 33.4, and 23.8 kDa, respectively. Insertion of the stop codon allows detection of 0 frame (VP2) and +1 frame (ORFx) translation without effects on the processing of ORF2 polyprotein.  The linearized IAPV cDNA plasmid was incubated in an in vitro transcription and translation system derived from Sf21 insect cell extracts (Promega) in the presence of [35S]-methionine. Translation products were analyzed by SDS-polyacrylamide gel electrophoresis and the radioactive products detected by autoradiography. The full-length 47  IAPV cDNA produced two major proteins (approximately 34 and approximately 11 kDa) (Figure 2.1 c, lane 2). To determine whether these proteins are expressed from the upstream nonstructural ORF1 or downstream structural ORF2, we disrupted the base pairing within PKI CC6615-6GG (ΔPKI) to inhibit expression of the downstream structural proteins. Disruption of PKI abolished expression of both proteins, which demonstrates that they are encoded within ORF2/ORFx (Figure 2.1 b and c, lane 3). Compensatory mutations (GG6564-5CC/CC6615-6GG) that restore PKI rescued expression, indicating that the expression of these proteins is IGR IRES-dependent (Figure 2.1 b and c, lanes 3 and 4). Furthermore, replacing the glycine GGC start codon of ORF2 with a UAG stop codon eliminated expression of both proteins, further confirming that these proteins depend on the IGR IRES are likely encoded downstream of the IGR IRES (Figure 2.1 c, lane 6).  To determine whether the approximately 34-kDa protein represents VP2, we raised a VP2 antibody against a peptide within the N terminus of VP2. The antibody detected the approximately 34-kDa protein in reactions that showed expression of the radiolabeled 34-kDa protein (Figure 2.1 c western blot, upper panel). From these findings, we conclude that the 34-kDa protein is encoded by the 5’ region of ORF2 and its presence is indicative of IAPV IGR IRES-mediated 0 frame translation. The lack of expression of the ORF1 proteins from the IAPV cDNA may be due to the weak 5’ UTR IAPV IRES activity in this extract.  The expression of an approximately 11-kDa radiolabeled protein may represent ORFx expression. Like VP2, the expression of the 11-kDa protein is dependent on the integrity of the IGR IRES (Figure 2.1 c, lanes 2–4). Importantly, introduction of stop codons within the +1 frame downstream of the IGR IRES abolished expression of the 11-48  kDa protein but not 0 frame VP2 production, demonstrating that the 11-kDa protein corresponds to the ORFx product (Figure 2.1 c, lanes 8–10). Furthermore, Western blotting using the anti-ORFx antibody confirmed the specific expression of ORFx (Figure 2.1 c western blot, lower panel).  To test the integrity of RNA, RNA was extracted from reaction lysate and Northern blot probing was performed. Northern blots showed that RNA extracted from lysate was substantially intact during the incubation (Figure 2.1 c northern blot). In summary, by using the IAPV full-length cDNA, both 0 frame VP2 and +1 frame ORFx proteins are expressed in Sf21 extracts in an IAPV IGR IRES-dependent manner.        49   50   Figure 2.1 Identification of the initiation codon of ORFx translation in IAPV genomic cDNA.  (a) (Top) Schematic organization of IAPV genome. Distinct IRESs direct translation of nonstructural (ORF1) and structural (ORF2) polyproteins. (Bottom) Secondary structure of the IAPV IGR IRES showing pseudoknots PKI, PKII, and PKIII, stem loops SLIII, SLIV, SLV, and SLVI (shaded gray), and loop L1.1. The UAA stop codon of ORF1 is shown in bold within the loop of SLVI. The overlapping +1 frame ORFx, within the 0 frame ORF2, 51  is shown. Conserved nucleotides among type II IGR IRESs are in capital letters. The CCU triplet mediates PKI base pairing and occupies the P site, whereas the first codon of the 0 frame ORF2 is the 3’-adjacent GGC codon in the A site. (b) A close-up view of the mutations introduced in the PKI region of the IAPV IGR IRES. The first amino acid of ORF2 is encoded by a glycine GGC codon in the 0 frame. The predicted first amino acid of ORFx - if ORFx is initiated at the IGR IRES - is encoded by an alanine GCG codon in the +1 frame. Tandem AUG codons within ORFx are shown. Mutations that replace codons with a stop codon in the 0 or +1 frame are denoted by S1-2 and S37. Mutations that disrupt PKI base pairing CC6615-6GG are shown (ΔPKI). Comp ΔPKI denotes compensatory mutations to restore PKI base pairing. (c) Translation of IAPV genomic cDNA mutants in Sf21 extracts. Full-length IAPV genomic cDNA was incubated in Sf21 extracts in the presence of [35S]-methionine, as described in section 2.2.2. A representative SDS-PAGE gel of radiolabeled protein products detected by autoradiography is shown. Mutations that affect 0 and +1 frame translation are indicated. The capsid protein VP2 represents 0 frame translation. In parallel, reactions were subjected to Western blotting using anti-VP2 and anti-ORFx antibodies and to Northern blotting using a probe that hybridizes to the IAPV IRES.  2.3.2 Mutagenesis studies reveal the start site of ORFx in a Full-Length IAPV cDNA clone The expression of ORFx is readily detected and dependent on the IAPV IGR IRES. We next determined the initiation site of ORFx. Several hypotheses for the mechanism of ORFx translation initiation can be proposed, including (i) a portion of the ribosomes recruited to the IGR IRES somehow start scanning and initiate at one of a conserved pair of AUG codons (nucleotides 6721-6) 103 nucleotides downstream of the IAPV IGR IRES (Figure 2.1 a); (ii) the IGR IRES places a portion of ribosomes directly into the +1 reading frame at or near to the normal 0 frame IGR IRES initiation site; or (iii) IGR IRES initiation 52  occurs only in the 0 frame, but a portion of ribosomes subsequently shift into the +1 reading frame at a putative slippery site. On the basis of the molecular mass of ORFx in reactions, we hypothesized that ORFx is initiated near the IAPV IGR IRES which would lead to an approximately 11.1-kDa ORFx protein.  First, we tested the role of AUG codons in +1 frame translation initiation. The IAPV genome contains an AUG codon in the +1 frame at positions 6592-6594 within the IGR IRES structure and tandem AUG codons at nucleotides 6721-6, 103 nucleotides 3’ of the IGR IRES (Figure 2.1 a). To check that ORFx initiation does not occur at these sites, we created mutations that substituted the AUG codons by mutating G6594 to A and AUGAUG6721-6 to ACGACG. Neither mutation abrogated expression of the 0 frame VP2 or +1 frame ORFx, which indicates that these AUG codons do not play a role in +1 frame translation initiation (Figure 2.1 c, lanes 5 and 11). Next, stop codons were systematically introduced at different positions in the +1 frame downstream of the IGR IRES to determine the initiation site of ORFx (Figure 2.1 b). Replacing the first, second, or 37th codon downstream from the IGR IRES with a UAG codon in the +1 frame abolished ORFx expression (Figure 2.1 c, lanes 8–10). Importantly, introducing these stop codons in the +1 frame did not significantly affect 0 frame ORF2 translation (Figure 2.1 c, lanes 8–10). These results suggest that +1 frame ORFx translation initiates either close to or upstream of the 0 frame IGR IRES initiation site. To determine whether IGR IRES-mediated 0 frame translation is required for +1 frame ORFx translation (e.g., if ORFx is translated by ribosomes frameshifting from the 0 frame), stop codons were introduced at different positions in the 0 frame (Figure 2.1 b). As expected, insertion of UAG codons in the 0 frame in any of the positions downstream of the 53  IGR IRES abolished 0 frame ORF2 translation (Figure 2.1 c, lanes 6 and 7). Insertion of a UAG codon as the second codon in the 0 frame did not significantly affect +1 frame ORFx translation, which suggests that +1 frame ORFx translation is independent of IGR IRES-mediated 0 frame ORF2 translation. Western blotting of the reactions using an anti-ORFx antibody confirmed the specific expression of the +1 frame ORFx protein (Figure 2.1 c Western blot, lower panel).  However, replacing the initial 0 frame codon, Gly GGC, with a UAG stop codon eliminated not only 0 frame translation but also +1 frame ORFx translation (Figure 2.1 c, lane 6 and Figure 2.2, lane 5). It is possible that the loss of +1 frame ORFx translation is due to disruption of the base pairing between nucleotides U6562 and G6618. To test this possibility, we created a compensatory mutation that restores the base pairing with the 0 frame stop codon. The mutation, U6562G in the 0 frame stop codon 1 mutant, restores the G-U base pairing between nucleotides 6562 and 6618 and also rescues +1 frame but not 0 frame translation (Figure 2.2, lane 6). Western blotting using the anti-VP2 and anti-ORFx antibodies confirmed the specific expression of VP2 and ORFx respectively (Figure 2.2, Western blot). Northern blots showed that the RNA is largerly intact during the in vitro process of transcription and translation (Figure 2.2, Northern blot). The results indicate that U6562 can potentially base pair with G6618. In summary, the base pairing between nucleotides 6562 and 6618 adjacent to the IGR IRES appear to be important for ORFx translation in vitro.  54   Figure 2.2 Compensatory mutations restore +1 frame ORFx translation. Full-length IAPV cDNA containing wild-type or the indicated mutant IAPV IGR IRESs were incubated in Sf21 extracts in the presence of [35S]-methionine, as described in section 2.2.2. A representative SDS-PAGE of radiolabeled protein products detected by autoradiography is shown. In parallel, reactions were subjected toWestern blotting using VP2 and ORFx peptide antibodies and to Northern blotting using a probe specific to IAPV IGR IRES. The 0 frame S1 stop denotes replacement of the 0 frame GGC start codon of the IAPV IGR IRES with a UAG stop codon.   55  2.3.3 U6562-G6618 base pairing directs +1 frame ORFx translation U6562 can potentially base pair with G6618 and perhaps set the ribosome into the +1 frame to initiate translation from an Ala codon, GCG6619-21 (Figure 2.1 b). To test this hypothesis, G6618 was first mutated to A, C, or U, and the translational products of 0 and +1 frame were determined. If the U6562-G6618 base pairing is important for +1 frame ORFx translation, mutant IRESs that allow for base pairing between nucleotides 6562 and 6618 should result in +1 frame ORFx translation. As expected, mutating G6618 to A, which is predicted to maintain a U6562∕A6618 base pair, retained +1 frame ORFx translation, albeit at a level lower than wild type (Figure 2.3, lane 5). In contrast, mutating G6618 to C or U abolished +1 frame ORFx expression, consistent with the idea that the U6562-G6618 base pair directs translation in the +1 frame (Figure 2.3, lane 6 and 7). Mutants in which the base pairing was restored by compensatory mutations: U6562G-G6618C, U6562G-G6618U, and U6562A-G6618U, again regained ORFx expression (Figure 2.3, lane 8-10). However, lower expression was observed for the A-U base pairing (Figure 2.3, lane 10) than for the G-C or G-U base pairings. In all cases, 0 frame VP2 expression was detected in these mutants. To confirm the expression of 0 frame VP2 and +1 frame ORFx, Western blotting detected the specific expression of VP2 and ORFx protein (Figure 2.3, Western blot). Furthermore, Northern blots showed that the integrity of RNA was largely intact in the in vitro translation extract (Figure 2.3, Northern blot).   56   Figure 2.3 U6562∕G6618 base pairing directs +1 frame ORFx translation in IAPV genomic cDNA.  Translation of IAPV genomic cDNA mutants in Sf21 extract, as described in section 2.2.2. A representative SDS-PAGE gel of radiolabeled protein products detected by autoradiography is shown. In parallel, reactions were subjected to Western blotting using anti-VP2 and anti-ORFx antibodies and to Northern blotting using a probe that hybridizes to the IAPV IRES.    57  To further characterize the role of U6562-G6618 base pairing for +1 frame translation, we utilized a bicistronic reporter plasmid, pIAPV-WT, as a template for in vitro transcription and translation. The pIAPV-WT contains the IAPV IGR IRES and the downstream region (NC_009025, nucleotides 6399-6908) within the intercistronic region (Figure 2.4 and Appendix B.1). The downstream +1 frame ORFx is fused in-frame with Firefly luciferase (FLuc) to produce an ORFx-FLuc fusion protein (~72 kDa). A shortened 0 frame ORF2 (Figure 2.4, sORF2) is also generated. Thus, the scanning-dependent Renilla luciferase (RLuc) and IGR IRES-mediated 0 and +1 frame translation can be monitored simultaneously. Using RLuc expression as an inner control, the 0 and +1 frame translation can be quantified and analyzed.     Figure 2.4 Diagram of a bicistronic reporter construct, pIAPV-WT. A bicistronic reporter construct, pIAPV-WT, contains the wild-type IAPV IGR IRES inserted in the intergenic region between two reporter genes: renilla luciferase (RLuc), which monitors scanning-mediated translation, and firefly luciferase (FLuc), which monitors IRES-mediated translation.   58  To confirm the finding that U6562-G6618 base pairing directs +1 frame translation in the full-length IAPV cDNA, we created the same mutations and compensatory mutations at nucleotides 6562 and 6618 in pIAPV-WT. Consistent with the in vitro data in Figure 2.3 using the full-length cDNA, disruption of the base pairing between these two nucleotides eliminated ORFx expression (Figure 2.5). While compensatory mutations which restore the base pairing rescued ORFx expression, a relatively lower level of +1 frame translation was observed compared to that of the wild-type construct (Figure 2.5, lane 9). The translation products were confirmed by Western blot analysis using anti-FLuc antibody (Figure 2.5 a, Western blot) and the RNA integrity was verifed by Northern blot analysis using a specific probe to the bicistronic reporter (Figure 2.5 a, Northern blot). Quantitation of +1 and 0 frame translation is shown in Figure 2.5 b. +1 frame ORFx translation was 20% and 5% of the 0 frame translational activity in reporter constructs containing the wild-type and the G6618A mutant IAPV IGR IRESs, respectively (Figure 2.5 b, lower). In contrast, the U6562G-G6618C mutations resulted in a large increase in the ORFx-FLuc/sORF2 ratio, which is likely due to a combination of higher +1 frame ORFx and decreased 0 frame sORF2 translation compared to the wild-type IRES (Figure 2.5 b, upper). Another possibility is that the G-C base pairing between nucleotides 6562 and 6618 is responsible for the higher +1 frame translation compared to that of the wild-type IRES where a weaker U-G base pair exists. In summary, these results clearly show that for both reporter constructs and the full-length IAPV genomic cDNA, ORFx translation is facilitated by base pairing between nucleotides 6562 and 6618 adjacent to the IGR IRES.  59   60  Figure 2.5 U6562∕G6618 base pairing directs +1 frame ORFx translation in bicistronic reporter. (a) Translational activities of bicistronic IAPV IGR IRESs mutants in Sf21 extracts. Experiments were performed as in section 2.2.2. In parallel, reactions were subjected to Western blotting or Northern blotting. (b) Quantitation of radiolabeled protein products normalized relative to the wild-type IAPV IGR IRES. The translation activity of ORFx-FLuc (+1 frame) and sORF2 (0 frame) also taking into account the number of methionines within each translated protein is shown. The data shown are the averages of at least three independent experiments ± s.d.   2.3.4 Mass spectrometry analysis confirms the start site of ORFx expressed in Sf21 extracts in-vitro Mutagenesis analyses indicate that base pairing of nucleotides 6562 and 6618 can direct +1 frame ORFx translation initiating at an alanine GCG6619-21 codon. To further confirm this result, in vitro expressed ORFx from Sf21 extracts was subjected to in-gel digestion with endopeptidase ArgC (which cleaves specifically after arginines), and the resulting peptides were analyzed by liquid chromatography-tandem mass spectrometry (LC-MS/MS). A peptide with the sequence AIHNKKAILPTYTIR was detected. This sequence corresponds to the predicted N terminus of ORFx on the basis of initiation at the GCG alanine codon (Figure 2.6). This result confirms that the in vitro translated ORFx initiates at GCG6619-21, 3’-adjacent to the IGR IRES.  61   Figure 2.6 Identification of the start site of ORFx by mass spectrometry.  Experiments were performed as in section 2.2.5. Fragment spectra from the 580.3498 Th, triply charged precursor ion of AIHNKKAILPTYTIR from ArgC-digested ORFx indicating that ORFx translation starts at this Ala residue. Individual fragment ions are annotated in the spectrum and in the sequence representation. Specifically, if the charge is maintained on the N-terminus of the peptide fragments, they are indicated by b (such as b2, b3 and b6++). If the charge is retained on the C-terminus, they are denoted by y (for example, y1, y3 and y4).  62  2.3.5 Detection of ORFx in virally infected honey bees by mass spectrometry We next examined whether ORFx is expressed in IAPV infected honey bees by first obtaining and harvesting IAPV from infected honey bees. IAPV virions were injected into approximately 17-day-old honey bee pupae, and after 96 h, RNA and protein were extracted. By using virus-specific primers, RT-PCR analysis detected KBV, ABPV, and IAPV RNAs in virus-infected samples (Figure 2.7), which indicates that the virus sample contained a mixed population of honey bee dicistroviruses.  Western blotting analysis detected the expression of IAPV VP2 but not IAPV ORFx, which suggests that ORFx may be expressed in low abundance during virus infection or may have more rapid turnover than VP2. However, using a more sensitive and specific technique, multiple reaction monitoring (MRM)-MS, we detected the expression of a peptide that is found in both IAPV and KBV ORFx, in virally infected bees at 6 and 24 h after injection (Figure 2.8 a). In MRM mode, all peptide samples were filtered on a triple quadrupole mass spectrometer (Q1/Q2/Q3). In the first step, the precursor is preselected in Q1 and fragmented in Q2. In the second stage, only sequence-specific fragment ions (transition ions) are analyzed in Q3. Thus, this will enhance the detection limit for peptides of interest. We monitored three different transitions, and the relative peak areas of those transitions in the infected samples were exactly the same as measured for the synthetic peptide (Figure 2.8 b). Importantly, no signal was seen in uninfected samples. These data demonstrate unequivocally that ORFx is expressed during virus infection.  63          Figure 2.7 Detection of KBV, ABPV, and IAPV from virus-infected honey bee pupae by RT-PCR using virus-specific primers. Experiments were performed as in section 2.2.6. and 2.2.7. RNA was extracted from honey bee pupae 96 h after injection with PBS (mock-infected) or virus particles. RT-PCR reactions were performed using virus-specific primers. PCR products were separated by electrophoresis on a 2% agarose gel and visualized by ethidium bromide staining.  64   Figure 2.8 Detection of ORFx in virus-infected honey bee pupae. (a) Fragment spectra from a shared peptide of KBV and IAPV ORFx with the three y ions (y6, y8 and y9) selected for transitions labeled in red. (b) MRM traces for the pure synthetic peptide and the same three transitions detected in virally infected bees at 6 h after injection. The relative percentages of area under the curve for each transition are labeled.   65  2.4 Discussion The dicistrovirus IGR IRES has the remarkable ability to recruit ribosomes directly and set them into a specific reading frame. Here, we describe a mechanism used by a subset of dicistrovirus IGR IRESs that can selectively place ribosomes and initiate translation in either of two open reading frames. The consequence of this mechanism is that, on top of initiating translation in the 0 frame to produce the viral ORF2-encoded structural proteins, an overlapping ORFx in the +1 frame is also translated. Translation of ORFx is directed by a U-G base pairing adjacent to the tRNA-like anticodon-codon interaction of the PKI domain. Several lines of evidence support this result: (i) Disruption of the PKI domain or the U6562-G6618 base pair abolishes ORFx translation, whereas compensatory mutations that restore base pairing rescue ORFx translation. The U-G base pairing directs the ribosome to initiate translation at a +1 frame GCG alanine codon occupying the ribosomal A site. (ii) Detection of a peptide from an in vitro synthesized ORFx by mass spectrometric analysis confirmed that ORFx is initiated at this GCG codon. (iii) Detection of an IAPV and KBV ORFx peptide in virally infected-honey bees by a sensitive selective MRM-MS approach demonstrated that ORFx is expressed during virus infection. (iv) Selective pressure to maintain the overlapping ORFx within a subset of dicis-troviruses strongly suggests that translation of ORFx benefits these viruses.  The tRNA-like anticodon-codon interaction is apparent from structural studies of the PKI domain of the type I IGR IRES (Costantino et al., 2008), and, although the structure of the type II IGR IRES has not yet been solved, it is presumed to be similar (Jang and Jan, 2010; Nakashima and Uchiumi, 2008). In our in vitro assays, +1 frame translation 66  occurs 20% of the time as compared to 0 frame translation (Figure 2.5 b). In general, disruption of +1 frame translation variably affects 0 frame translation (Figure 2.1), but there does not appear to be a strict correlation to suggest competition for ribosome site selection in the 0 or +1 frame (i.e., U-G base pairing does not appear to occlude the 0 frame A site such that the A-site tRNA can be delivered only into the +1 frame). However, mutations that disrupt +1 frame translation also change the 0 frame initiation codon and thus make it difficult to interpret whether alterations in 0 frame translation are due to effects of disruption of +1 frame translation or due to delivery of a different aminoacyl-tRNA to the ribosomal A site. Currently, it is not clear if different conformations exist that individually direct 0 and +1 frame translation. Alternatively initiation of 0 and +1 frame translation could rely on the same tRNA-like structure of the PKI domain and the U-G base pairing would direct +1 frame translation. Consistent with the latter hypothesis, similar to 0 frame translation, +1 frame translation requires base pairing of the PKI region (Figure 2.2, 2.3, 2.5). This scenario is reminiscent of the base pairing that has been postulated by some for a subset of tRNA frameshift suppressors that shift the reading frame by +1 nucleotide. The difference between the two mechanisms is that tRNA frameshift mutant suppressors shift the reading frame of elongating ribosomes, whereas the IAPV IGR IRES shifts the reading frame of an initiating ribosome, albeit a ribosome that is initiating directly into the elongation phase from the A site. Although the in vitro translation system allows relatively high +1 frame translation compared to 0 frame translation, we had to resort to a very sensitive MRM-MS approach to detect ORFx in virus-infected honey bees (Figure 2.8). It is possible that ORFx is translated only during a specific window of virus infection or that ORFx is inherently unstable (relative to the structural proteins) and is rapidly degraded. In 67  summary, because both 0 and +1 frame translation are IGR IRES-dependent, it is likely that both frames are regulated similarly during the viral life cycle, especially when certain initiation factors are compromised.                    68  Chapter 3: The mechanism of ORFx translation frame selection 3.1 Introduction Non-canonical translation mechanisms provide a number of ways to express viral proteins. To optimize viral coding capacity, the use of overlapping open reading frames is a common strategy among many viruses. Leaky scanning, IRESs-mediated initiation, ribosomal frameshifting and stop codon readthrough are also important strategies in regulating the expression level and the expression timing of different proteins. In Chapter 2, we discovered that the IAPV IGR IRES directs translation of +1 frame ORFx and identified the start site of ORFx as an GCG alanine codon, just one nucleotide downstream of the 0 frame start codon (Figure 2.1 a). This work revealed a novel recoding mechanism utilizing an IRES to initiate translation in the 0 and +1 frames to produce two overlapping gene products.  In this chapter, extensive mutagenesis and selective 2'-hydroxyl acylation analyzed by primer extension (SHAPE) analysis are performed to reveal that the IRES adopts different conformations to direct reading frame selection. Besides the adjacent U-G base pair, novel classes of mutants which can direct either only 0 or +1 frame translation are identified.  Ribosome positioning experiments indicate that the reading frame is selected at a downstream step of ribosome assembly. In summary, we propose a model that the IRES adopts different conformations to manipulate the ribosome into a conformation that at some frequency allows delivery of the +1 frame aminoacyl-tRNA to the A site to initiate 69  translation of ORFx. This work provides a new paradigm for understanding programmed recoding mechanisms in viral genomes that increase the coding capacity.  3.2 Materials and methods   3.2.1 DNA constructs and reagents The bicistronic and the monocistronic luciferase plasmids containing the IGR IRES of IAPV (NC_009025) (nucleotides 6399-6908), KBV (NC_004807) (nucleotides 6381-6908), ABPV (NC_002548) (nucleotides 6296–6814), and SINV-1 (NC_006559) (nucleotides 4189–4797) have been described (Ren et al., 2012). Mutants were generated using a site-directed mutagenesis protocol and confirmed by sequencing. Primers are listed in Appendix B2.  3.2.2 In vitro transcription and translation assay  For in vitro transcription, monocistronic luciferase plasmids were linearized with SpeI. Bicistronic luciferase plasmids were linearized with XbaI. The IGR IRES RNAs were transcribed in vitro using a bacteriophage T7 RNA polymerase and were purified with an RNeasy kit (Qiagen). The integrity and purity of the transcribed RNAs were confirmed by gel analysis. Bicistronic luciferase plasmids were linearized with XbaI.  The linearized DNA (1 µg / µl) was incubated with 6.7 µl Sf21 extract (Promega), 0.5 mM Magnesium acetate and 40 mM potassium acetate (final concentration) to final volume of 10 µl for 3 h at 30°C. The expression of protein products was measured by incorporation of 0.4 µl of [35S]-70  methionine (> 1,000 Ci/ mmol; Perkin Elmer) per 10 µl reaction. The expression of luciferase protein was analyzed by SDS-PAGE followed by phosphor imaging using a Typhoon phosphorimager (Amersham).  3.2.3 5’ End labeling of RNA and primers RNA was dephosphorylated in a reaction containing RNA, 10 × dephos buffer (400 mM Na-MES, pH6.0, 100 mM MgCl2, 50 mM DTT, New England Biolabs) and calf intestinal phosphatase (New England Biolabs). The reaction mix was incubated at 37°C for 30 min followed by RNA purification using RNeasy Kit (Qiagen). The 5’ end dephosphorylated RNAs or primers were radiolabeled as follows. 10 pmol dephosphorylated RNA or primers, 10 × polynucleotide kinase (PNK) buffer (70 mM Tris-HCl at pH 7.6, 10 mM MgCl2, 5 mM DTT, New England Biolabs), T4 PNK (10,000 U/mL, NEB) and 1.6 µM 32P-γ-ATP (6000 Ci/mmol, Perkin Elmer or MP Biomedicals) were combined in a reaction tube to a final volume of 20 µl, incubated at 37°C for 30 min. 7 M urea loading buffer (7 M urea in 1 × TBE buffer with 0.5 mg/ml bromophenol blue and xylene cyanol) was added and the reaction mix were loaded on a 10% acrylamide (29:1 bisacrylamide:acrylamide) 7 M urea-denaturing gel. After electrophoresis at 25 W for about 30 min – 1.5 h, dependending on the radiolabeled probe length, the gel was removed from the plates, wraped in plastic and then exposed for 5 sec. The bands were excised after alignment of the exposed film with the gel.  The cut bands were eluted in 0.5 M NaOAC at pH 5.2, 0.1% SDS overnight at 4°C and followed by ethanol precipitation. The radiolabeled probes were pelleted by centrifugation, washed with 70% ethanol and then dried. The dry pellet was then dissolved in RNAse-free water prior to use. 71    3.2.4 SHAPE analysis of IRES and IRES–ribosome complexes SHAPE probing was performed as described with some modification (Merino et al., 2005; Wilkinson et al., 2006). Briefly, 2 pmol RNA in 8.1 ul water was heated to 85 °C for 2 min, then 0.9 µl of folding solution (10 × E buffer: 200mM Tris, pH7.5, 1M KCl, pH 7.0, 25 mM MgOAc, 2.5 mM spermidine, 20 mM DTT) was added and the solutions were incubated for 20 min at 30 °C. To modify the RNA, 1 µl of 30 mM N-methyl-isatoic anhydride (NMIA) dissolved in DMSO was added to give a final reaction volume of 10 µl. The reaction was incubated for 1.5 h at 30 °C. The modified RNA was recovered by ethanol precipitation. Control reactions containing only DMSO (no NMIA) were also performed. Reverse transcription was performed using 5’ gacgaagcgagttccgta 3’ as primer and sequencing ladders generated as described (Wilkinson et al., 2006).  Reactions were resolved on 8% (w/v) denaturing polyacrylamide sequencing gels run at 60 W for various time periods. The gels were dried and analyzed by phosphorimager analysis. Quantitative analysis was performed using the SAFA (Semi-Automated Footprinting Analysis) program and normalized as described (Das et al., 2005). Salt-washed 40S and 60S subunits were purified from HeLa cells (see Section 3.2.5) (Jan and Sarnow, 2002) and SHAPE probing of IRES–80S ribosome complexes was done as described with some modification (Costantino et al., 2008; Pfingsten et al., 2006). Briefly, 20 pmol RNA was heated to 85 °C for 2 min, then folding solution (final  concentration is 1× E buffer, 20mM Tris, pH7.5, 0.1M KCl, pH 7.0, 2.5 mM MgOAc, 0.25 mM spermidine, 2 mM DTT) was added and the solutions were incubated for 20 min at 30 72  °C. Then 20 pmol of 80S ribosome were added and incubated for 10 min at 30 °C. To modify the RNA, 10 μl of 30 mM NMIA dissolved in DMSO was added in the final reaction volume of 100 µl. The reaction was incubated for 1.5 h at 30°C. IRES-ribosome complexes were separated by ultracentrifugation through 10–35% (w/v) sucrose gradients containing 20mM Tris, pH7.5, 0.1M KCl, pH 7.0, 5 mM MgOAc, 0.25 mM spermidine, 2 mM DTT. After fractionation, the IRES-80S complexes were ethanol-precipitated and resuspended in water. Reverse transcription reaction was performed using 5’ GACGAAGCGAGTTCCGTA 3’ as a primer. Reactions were resolved on 8% (w/v) denaturing polyacrylamide sequencing gels at 60 W. The gels were dried and analyzed by phosphorimager analysis. Individual band intensities were quantitated using SAFA and normalized as described (Das et al., 2005).  3.2.5 40S and 60S ribosomal subunit purification Ribosomal subunits were purified from HeLa cell pellets (National Cell Culture Centre) as described (Jan and Sarnow, 2002). Briefly, HeLa cells were first lysed in a Triton X-100 lysis buffer (15 mM Tris–HCl (pH 7.5), 300 mM NaCl, 6 mM MgCl2, 1% (v/v) Triton X-100, 1 mg/ml heparin). Lysates were centrifuged to remove debris and the supernatant layered on a 30% (w/w) 0.5 M KCl sucrose cushion and centrifuged at 100,000 g to pellet ribosomes. Ribosomes were resuspended in buffer B (20 mM Tris–HCl (pH 7.5), 6 mM magnesium acetate,150 mM KCl, 6.8% (w/v) sucrose, 1 mM DTT), treated with puromycin to release ribosomes from mRNA, and KCl was added to a final concentration of 0.5 M. The dissociated ribosomes were then separated on a 10–30% (w/w) sucrose gradient. The 40S and 60S peaks were detected at 260 nm, pooled, concentrated using 73  Amicon Ultra spin concentrators (Millipore) in buffer C (20 mM Tris–HCl (pH 7.5), 0.2 mM EDTA, 10 mM KCl, 1 mM MgCl2, 6.8 % sucrose). Western blot analysis verified the absence of eIF2. The purity of 40S and 60S was also examined by detecting 18S and 28S rRNA by ethidium bromide staining. The concentrations of 40S and 60S subunits were determined by spectrophotometry (Thermo Scientific), using the conversions 1 A260 nm = 50 nM for 40S and 1 A260 nm = 25 nM for 60S subunits.  3.2.6 Toeprinting analysis of ribosomal complexes using purified components Toeprinting analysis of ribosomal complexes using purified subunits was performed as described (Wilson et al., 2000a). 200 ng of dicistronic IGR IRES RNAs were first annealed with primer 5' GTAAAAGCAATTGTTCCAGGAACCAG 3' , which hybridizes 115 nucleotides downstream of the IGR IRES 0 frame start site in 40 mM Tris (pH 7.5) and 0.2 mM EDTA by slowly cooling from 65 °C to 30 °C. Annealed RNAs were incubated in buffer E with 40S and 60S subunits (final 40 nM). Ribosome-RNA complexes were analyzed by primer extension analysis using AMV reverse transcriptase in the presence of [α-32P]-dATP (3000 Ci/mmol, Perkin Elmer) (Jan and Sarnow, 2002). cDNA products were analyzed on 6% (w/v) polyacrylamide/8 M urea gel. Gels were dried and analyzed by phosphorimager analysis (Amersham).   74  3.3 Results 3.3.1 U6562-G6618 base pairing is not strictly necessary for +1 frame translation in specific mutant IRES contexts  In Chapter 2, we showed that the U6562-G6618 base pair adjacent to the PKI domain directs +1 frame ORFx translation (Ren et al., 2012). To further examine the role of the nucleotide identities of U6562 and G6618 in 0 or +1 frame translation, we generated mutant IRESs containing all 16 permutations of bases at positions 6562 and 6618 (Figure 3.1 b). The effect on IRES-mediated 0 and +1 frame translation was assessed in an insect Sf21 translation extract in the presence of [35S]-methionine, using the linearized bicistronic reporter construct that contains the IAPV IGR IRES and the downstream region (NC_009025, nucleotides 6399-6908) (Figure 3.1 a). In this construct, the downstream +1 frame ORFx is fused in-frame with Firefly luciferase (FLuc) to produce an ORFx-FLuc fusion protein (~72 kDa). A shortened 0 frame ORF2 (sORF2) is also generated (Figure 3.1 a). Thus, the scanning-dependent Renilla luciferase (RLuc) and IGR IRES-mediated 0 and +1 frame translation can be monitored simultaneously by SDS-PAGE analysis.  The translation activities of the 16 mutants are summarized in Figure 3.1 d in two graphs which are organized based on the base identities at positions 6562 and 6618. Both graphs show the relative ratio of +1 frame to 0 frame translation. G6618 mutated to C or U but not to A abolished +1 frame translation, consistent with the model that base pairing between nucleotides 6562 and 6618 is necessary for +1 frame translation (Figure 3.1 c, lanes 2,4-6, 3.1 d) (Ren et al., 2012). Consistent with this result, mutations that restored base pairing such as U6562A:G6618U, U6562C, U6562G:G6618C and U6562G:G6618U 75  (Figure 3.1 c, lanes 10, 11, 17 and 18), rescued +1 frame translation albeit to different extents. To further substantiate these results, mutating U6562 to A or C that should not restore base pairing with either G6618 to C or U (Figure 3.1 c, lanes 9, 13 and 14) inhibited +1 frame translation, thus reinforcing the importance of base pairing between nucleotides 6562 and 6618. However, in contrast to what was observed when U or C was in position 6618, different permutations with G or A at position 6618 showed that the adjacent base pairing was not absolutely necessary for +1 frame translation.  Mutating U6562 to A or G when G or A at position 6618, or U6562 to C and G6618 to A, all of which should not form a 6562-6618 base pair, resulted in a significant level of +1 frame translation (Figure 3.1 c, lanes 7-8, 12, 15-16).   Interestingly, specifically mutating U6562 to G regardless of the nucleotide identity of 6618 significantly increased +1 frame translation (Figure 3.1 d, left graph). This is most evident when the ratio of +1/0 frame translation is plotted (Figure 3.1 e). The most significant increase was afforded by U6562G. This mutation increased +1 frame translation by ~4.5-fold compared to the wild-type IRES and displayed a similar level of 0 frame translation (Figure 3.1 d). Changing the first +1 frame GCG codon to a stop UAG codon together with the U6618G mutation eliminated +1 frame translation, thus confirming that the translation product is in the +1 frame.  As shown previously, mutating CC6615-6 to GG which disrupts PKI basepairing eliminated both 0 and +1 frame translation (Figure 3.1 c, lane 3), whereas compensatory mutations that restore basepairing rescued translation in both frames, demonstrating that 0 and +1 frame translation are dependent on the integrity of PKI and are IGR IRES-dependent (Ren et al., 2012). In summary, these results demonstrate that the U6562/G6618 76  base pair adjacent to the anticodon-like PKI domain directs IAPV IRES-mediated +1 frame translation. However, in specific mutant IRES contexts, the adjacent base pairing was not strictly necessary for +1 frame translation, mutating U6562 to other bases can still support +1 frame translation.       77   Figure 3.1 Mutagenesis of the U6562/G6618 base pair adjacent to PKI. 78  Legend to Figure 3.1. (a) Diagram of a bicistronic reporter construct containing renilla luciferase (RLuc) and firefly luciferase (FLuc), the wild-type or mutant IAPV IGR IRES inserted in the intergenic region between two reporter genes. (b) PKI domain of the IAPV IGR IRES. Mutations within the PKI domain are shown. The U6562 and G6618 are bolded. (c) Translational activity of the WT and mutant IAPV IGR IRES in Sf21 extract, as described in section 3.2.2. WT and denote bicistronic reporter mutants of the U6562 or G6618 are shown. (d) Quantitation of radiolabeled protein products calculated by taking the ratio FLuc/RLuc and normalized by the containing numbers of methionine. The data are plotted by either grouping nucleotide changes to (left) U6562 or to (right) G6618. (e) Ratio of ORFx/ORF2 protein products produced from the indicated mutants. Data are presented as mean ± s.d. of three independent experiments.   The wild-type IAPV IGR IRES directs +1 frame translation at ~ 20-25% of 0 frame translation in vitro (Figure 3.1 c-e), which may be due to different conformations of the IRES that direct +1 frame translation. Alternatively, it is possible that the levels of or efficiency of delivery of the incoming aminoacyl-tRNAs to the 0 or +1 frame initiating codon affects +1 frame translation. To test the latter possibility, we mutated the +1 frame GCG alanine codon to a GGG glycine codon (C6620G). Both the 0 and +1 frame initiating codons thereby start with the GGG glycine codon which should utilize the same Gly-tRNA for delivery to the A site in both frames. Compared to the wild-type IRES, the C6620G mutant IRES displayed a lower level of translation in both frames which likely reflects the specific Gly-tRNA that is delivered to the GGG codons in the 0 and +1 frames (Figure 3.2). The ratio of +1/0 frame translation in the C6620G mutant IRES was 0.34 compared to 0.23 in the wild type (Figure 3.2, 10% vs 29%).  Given that this mutant IRES uses the same 79  incoming aminoacyl-tRNA for the initiating codon in the 0 and +1 frames, these results may reflect the true ratio of +1 to 0 frame translation mediated by the IAPV IGR IRES.    Figure 3.2 Translation of C6620G mutant IRES. (Top) Schematic of wild-type and C6620G mutant IRESs is shown. Mutating C6620 to G leads to an IRES-mediated GGG start codon in both 0 and +1 frames.  (Bottom) IRES-mediated 0 and +1 frame translation are plotted and normalized to the wild-type IRES-mediated 0 frame translation given as 100%. Averages are shown from at least three independent experiments ± s.d. (*P < 0.05 compared with the ratio of +1/0 frame translation) 80   3.3.2 Structure probing analysis of WT and mutants IAPV IGR IRESs in solution The mutagenesis analyses suggest that regions within the IGR IRES other than base pairing between nucleotides 6562 and 6618 may contribute to +1 frame translation. We hypothesize that the IRES may adopt two or more different conformations at some relative likelihood to favour either 0 or +1 frame translation. To test this, we probed the structure of wild-type and mutant IGR IRESs using selective 2'-hydroxyl acylation analyzed by primer extension (SHAPE) analysis, a chemical probing method that modifies the 2’ hydroxyl group of structurally dynamic nucleotides using N-methylisotoic anhydride (NMIA) (Merino et al., 2005). Modification of each nucleotide by NMIA is assessed by comparing band intensities from the modification reaction to those of a no NMIA control, thus revealing the flexibility of individual RNA nucleotide.  We focused our attention on the PKI domain as this region directs +1 frame translation. For the wild-type IRES, nucleotides that were reactive to NMIA were observed within single-stranded regions of the Variable Loop Region (VLR), the loop of SLIII, the unpaired A6554 bulge and U6562 (Figure 3.3, WT and Appendix Figure C.2).  In general, nucleotides within predicted helical stems and base pairs were not reactive to NMIA (Figure 3.3, WT and Appendix Figure C.2). The one exception is within PKI. Specifically, nucleotides G6563, GCA6565-7 and U6613 within the wild-type IRES were reactive to NMIA (Figure 3.3, WT and Appendix Figure C.2). These results are in agreement with previous reports that the PKI domain may be dynamic and is reactive toward chemicals and 81  enzymes that recognize both ssRNA and doublestranded RNA (Jan and Sarnow, 2002; Nishiyama et al., 2003). For comparison, SHAPE analysis of the mutant IRES ΔPKI, which disrupts PKI base pairing CC6615-6GG, resulted in changes in SHAPE reactivities within the PKI base pair region and VLR (Figure 3.3, ΔPKI and Appendix Figure C.2).  We hypothesize that the IRES can assume a range of conformations, a subset of which favors one or the other specific reading frames. To address this possibility, we analyzed the structures of two mutant IAPV IRESs, G6618U, which abolishes +1 frame translation and U6562G, which increases +1 frame translation. For the G6618U mutant IRES, SHAPE reactivities were obtained in the same regions as that observed with the wild-type IRES but to varying extents. Notably, we observed an increase of SHAPE reactivities at nucleotides A6605 to U6608 of the VLR and in nucleotides GGC6564-6 of the PKI base pair.  A decrease in SHAPE reactivity was observed at nucleotides A6612 and U6613 in the VLR and left side of PKI codon-anticodon base pairings and at nucleotide U6562 (Figure 3.3, G6618U and Appendix Figure C.2).  In contrast, the U6562G mutant IRES resulted in increased SHAPE reactivities of residues G6564, A6567 and G6568 in the anti-codon loop of PKI domain, CAU6611-6614 in the VLR and left side of PKI. A decrease in SHAPE reactivity was observed at nucleotides G6562 and G6563 in the anti-codon loop of PKI (Figure 3.3, U6562G and Appendix Figure C.2). The pattern of SHAPE reactivities within the VLR of the U6562G more closely resembled that of the wild-type IRES than of G6618U (Figure 3.3, U6562G and Appendix Figure C.2).  82  In summary, we observe modest changes in SHAPE reactivities between the mutant and wild-type IRESs, suggesting that a subset of conformations, albeit subtle, are associated with 0 or +1 frame translation.   83     Figure 3.3 SHAPE analysis of the PKI domain of the WT and mutant IAPV IGR IRES. (a) Quantitation data of SHAPE probing experiments. Experiments were performed as described in Section 3.2.4. SHAPE reactivity of each nucleotide is background-corrected and normalized and shown over the wild-type IRES. The gray bars represent wild-type IRES and the colored bars are mutant IRESs as indicated. The degree of modification is on 84  the y-axis, each nucleotide is on the x-axis, with the location of each domain indicated. The data shown are the averages of three independent experiments ± s.d. (b) [Continued on next page.] SHAPE data superimposed on the PKI domain of the IAPV IGR IRES secondary structure. The colour coding indicates the SHAPE reactivities of individual nucleotides (red, high reactivity (>0.8); orange, moderate reactivity (0.5-0.8); green, limited reactivity (0.2-0.5); black, relatively unreactive (<0.2).).       85  3.3.3 Mutations within PKI codon-anticodon base pairing region affect 0 and +1 frame translation The differences in SHAPE reactivities between the wild type and mutant IRESs may reflect a subset of conformations that are associated with 0 or +1 frame translation. To investigate this more closely, we engineered mutations within the IRES guided by the changes in SHAPE analysis. IRES-mediated 0 and +1 frame translation were monitored using the bicistronic reporter construct described in Figure 3.1 a. The translational activities of those mutants are summarized in Figure 3.4 by calculating the percentage of 0 and +1 frame translation of IRES mutants normalized to those of the wild-type IRES (given as both frame as 100%). We first focused on the anticodon-like PKI domain since the majority of changes in SHAPE reactivities occur in this region. The PKI domain may be dynamic and flexible given that some SHAPE reactivity occurred in predicted base-paired regions in agreement with previous reports (Jan and Sarnow, 2002; Jan, 2006; Costantino and Kieft, 2005; Kieft, 2008). We reasoned that strengthening or weakening PKI base pairing may differentially affect 0 or +1 frame translation. We investigated the base pairing on the left side of the PKI codon-anticodon base pairing region. Because the adjacent base pair U-G of PKI is important for +1 frame translation, one possibility is that the left most base pair (A6567-U6613) may not be required and that a realignment of the next five base pairs would still suffice to direct +1 frame translation (ie. U6562-G6618 would be base paired). Furthermore, increased SHAPE reactivities of mutant IRES G6618U, which eliminates +1 frame translation, were obtained for some of the nucleotides such as GGC6564-6 within the 86  PKI base pairing region, suggesting that the PKI basepairing is important for +1 frame translation. We chose to mutate the bottom strand of PKI in order to not disrupt the anticodon-like loop of PKI.  Mutating U6613A had a deleterious effect on +1 frame translation (~60% decrease) but did not significantly affect 0 frame translation (~10% decrease) (Figure 3.4). Mutating the two left most base pairs, U6613A and G6614C, abolished both 0 and +1 frame translation, indicating that a minimum of 4 base pairs is required for IRES activity.  We then investigated the right side of the PKI codon-anticodon base pairing region. A U6617G mutation on the bottom strand of PKI which disrupts base pairing, led to a decrease in translation of both reading frames (~85% decrease of 0 frame and ~60% decrease of +1 frame). In contrast, U6617A or U6617C mutations decreased +1 frame translation to ~20-30% but increased 0 frame translation (to 158% and 196%). Mutations were also introduced into G6563 in the anticodon loop of PKI base pairing region. Only G6563A displayed 0 frame translation (to 220%). G6563C or G6563U abolished both 0 and +1 frame translation (Figure 3.4). This is consistent with a previous finding that the three bases in the anticodon are conserved as purines (A and G) in order to be stacked in the same manner as in an authentic tRNA (Costantino et al., 2008; Jan, 2006; Kanamori and Nakashima, 2001).  In summary, these results suggest that five complete basepairs in PKI are required for +1 but not 0 frame translation.    87      Figure 3.4 Translational activities of mutant IGR IRESs within PKI codon-anticodon base pairing region Translational activities of mutants in PKI domain of IAPV IGR IRES in Sf21 extracts, as described in section 3.2.2. Quantitation of radiolabeled protein products was calculated by taking the ratio FLuc/RLuc and normalized by the relative methionine content. The percentage of 0 and +1 frame translation of IRES mutants normalized to those of the wild-type IRES (given for both frames as 100%). Data are presented as mean ± s.d. of three independent experiments.   88  3.3.4 Single mutant within IGR IRES that determines 0 or +1 reading frame translation  Phylogenetic analysis of the Type II IGR IRES secondary structures predicts that the PKI anticodon-like loop contains 4 unpaired nucleotides (Figure 3.6 a). The Type II IGR IRESs include IRESs of the members like the ABPV, IAPV, KBV, SINV-1 and TSV. Within the PKI domain of the IAPV IGR IRES, the unpaired nucleotides, C6561, U6562, G6568 and U6569, showed differential reactivities to NMIA by SHAPE probing.  In particular, C6561 and G6568 were not reactive to NMIA in the wild-type and mutant IRESs (Figure 3.3 WT). To examine these nucleotides further, we mutated them to other bases. Mutating G6568 to A, C, or U drastically inhibited +1 but not 0 frame translation. In contrast, mutating C6561 to A, U or G decreased 0 frame translation by 60-75% and increased +1 frame translation (Figure 3.5). In summary, these results reveal that the identities of C6561 and G6568 are important for selection of 0 or +1 frame translation.  U6569 is predicted to be unpaired within the anticodon-like loop of PKI and is reactive to NMIA by SHAPE (Figure 3.3 WT). Deletion of U6569 (ΔU6569) surprisingly resulted in complete loss of 0 frame translation and in a significant increase in +1 frame translation (~3.5 fold of wild-type). Interestingly, mutating U6569 to A, C, or G decreased 0 frame translation and increased +1 frame translation (Figure 3.5). This result indicates that this nucleotide bulge is important for 0 but not +1 frame translation. Given this result, we further examined other unpaired or bulged nucleotides within PKI. Most notably, A6554 is predicted to be unpaired and is highly sensitive to NMIA. Mutating A6554 to U, C, or G did not have an effect on +1 frame translation but did increase 0 frame translation 89  by 70% when mutated to C or G. Interestingly, deletion of A6554 (ΔA6554) abolished 0 frame but not +1 frame translation (Figure 3.5), suggesting that any unpaired residue at position 6554 will support 0 frame translation.  In summary, we have identified specific class of mutations within the PKI domain that affect either 0 or +1 frame translation. Specifically, deletion of U6559 or A6554 resulted in loss of 0 frame translation but increasing or maintaining level of +1 frame translation. In contrast, mutating G6568 specifically inhibited +1 frame translation.      90     Figure 3.5 Specific nucleotides direct IAPV IGR IRES-mediated 0 or +1 frame translation. Translational activities of mutants in PKI domain of IAPV IGR IRES in Sf21 extracts, as described in section 3.2.2. Quantitation of radiolabeled protein products was calculated by taking the ratio FLuc/RLuc and normalized by the relative methionine content. The percentage of 0 and +1 frame translation of IRES mutants normalized to those of the wild-type IRES (given for both frames as 100%). Data are presented as mean ± SD of three independent experiments.   91  We next asked if this trend is specific to IAPV IGR IRES translation or whether the other Type II IGR IRESs like KBV, SINV-1 and ABPV would also show the same nucleotide dependency.  We deleted unpaired nucleotides within the KBV, SINV-1 and ABPV IGR IRESs at equivalent positions to A6554 and U6569 of the IAPV IGR IRES (Figure 3.6 a).  As was observed with the IAPV IGR IRES, deletion of C6581 or U6582 of the KBV IGR IRES stimulated +1 frame translation by ~4-5 fold (Figure 3.6 b and c). In contrast, similar deletions within the SINV-1 IRES did not stimulate +1 frame translation but did lead to exclusive +1 frame translation (Figure 3.6 b and c). In general, the results revealed a similar trend where the deletion of these nucleotides results in an increase or exclusive IRES-mediated +1 frame translation, thus reinforcing the idea that a subset of conformations drives 0 or +1 frame translation.  Furthermore, the results suggest that the subset of Type II IGR IRESs use similar structural determinants to direct reading frame selection.      92   Figure 3.6 Translational activity of the Type II IGR IRESs. 93  Legend to Figure 3.6. (a) Secondary structures of the PKI domain of IAPV, KBV, ABPV and SINV-1 IGR IRES. (b) A representative gel of mutant IRESs mediated 0 and +1 frame translatin by bicistronic reporter constructs in Sf21 extracts, as described in section 3.2.2. (c) Quantitation of the radiolabeled +1 frame ORFx-Fluc (white bars) and 0 frame sORF2 (black bars) proteins normalized to RLuc. The ratios are normalized to the 0 and +1 frame translation of each wild-type IGR IRES given as 1.0. Averages are shown from at least three independent experiments ± s.d.   Given that +1 frame translation is ~20-25% of 0 frame translation, we propose that the wild-type IRES adopts different conformations in equilibrium prior to ribosome assembly and that mutations shift the equilibrium to favor an IRES conformation for 0 or +1 frame translation.  To determine if a class of mutations that favors 0 or +1 frame translation is dominant over the other, we created several double mutations. Combining G6618U mutation, which exclusively shows 0 frame translation, with either deletion of U6569 or A6554, both of which lead to only +1 frame translation, leads to an intermediate level of 0 and +1 frame translation as compared to that of the single mutation or deletion alone (Table 3.1). Combining mutation G6618U with G6568C, which eliminates +1 frame translation, results in an additive increase in 0 frame translation (~2.3 fold). To further explore this, combining mutation U6562G, which increases +1 frame translation ~4.5 fold as compared to the wild-type IRES, and mutation G6568C, which leads to a ~95% decrease in +1 but not 0 frame translation, resulted in an intermediate level of 0 and +1 frame translation (~80% translation in both frames as compared to the wild-type IRES, Table 3.1). However, combining mutation U6562G, which has ~60% 0 frame translation, with either deletion of U6569 or A6554, leads to complete abolition of 0 frame translation and an additive increase in +1 frame translation. These results indicate that 0 or +1 frame 94  translation directed by IRES mutants G6568C, ΔU6569 or ΔA6554 depends on the adjacent base pairing between U6562 and G6618.  In summary, these results suggest only in the combination of U6562G with ΔU6569 or ΔA6554 that ΔU6569 and ΔA6554 have a dominant effect over U6562G which selected only for +1 frame translation. In other cases, the combination of opposing classes of mutants likely shifts the equilibrium of different conformations of the IRES.   Table 3.1 IRES mutations that confer 0 or +1 frame translation. Quantitation of mutant IAPV IRES-mediated 0 and +1 frame translation in Sf21 extracts using bicistronic reporter constructs. The percent translation is normalized to the 0 and +1 frame translation (both as 100%) mediated by the wild-type IAPV IGR IRES. The results are from at least three independent experiments ± s.d.   95  3.3.5 Effects of SLIII on 0 and +1 frame translation SHAPE analysis shows that the loop of SLIII is subject to NMIA reactivity. Currently, the role of SLIII in Type II IRES translation remains elusive. Complete deletion of SLIII (nucleotides G6577-C6598) abolished both 0 and +1 frame translation activities (Figure 3.7). A similar effect was observed in other Type II IRES, like TSV (Jang and Jan, 2010). Based on this evidence, we reasoned that specific nucleotides within the loop or helical stem of SLIII may have a role in IRES translation. Mutating the nucleotides within the SLIII loop did not have a deleterious effect on 0 or +1 frame translation (Figure 3.7), which may not be surprising given that the nucleotides of the SLIII loop are not conserved among Type II IGR IRESs.  We next addressed whether the length of the helical stem has a role in IRES activity. Shortening the helical stem by deletion of one base pair at various positions (ΔA6582/ΔU6593, ΔU6581/ΔG6594, ΔG6580/ΔC6595, and ΔC6579/ΔG6596) resulted in a decrease in 0 frame translation by approximately 20-55% and in an increase in +1 frame translation by 10-20% as compared to that of the wild-type IRES (Figure 3.7). Deletion of ΔA6582/ΔU6593 resulted in a more drastic effect on 0 frame translation (55% decrease) suggesting that the identity of this base pair may be important for IRES activity. Furthermore, deletion of two base pairs within the stem exacerbated these effects, decreasing 0 frame translation by 50-70% and increasing +1 frame translation by 40-60% (Figure 3.7, ΔCG6579-80/ΔCG6595-6 and ΔUA6581-2/ΔUG6593-4). Thus, these results indicate that the full length of the helical stem is required for 0 frame translation and suggest that SLIII plays a structural role for 0 frame translation.   96     Figure 3.7 Effects of the SLIII and VLR on 0 and +1 frame translation Translational activities of mutant IAPV IGR IRES in Sf21 extract, as described in section 3.2.2. Quantitation of radiolabeled protein products was calculated by taking the ratio FLuc/RLuc and normalized by the relative methionine content. The percentage of 0 and +1 frame translation of IRES mutants is normalized to those of the wild-type IRES (given for both frames as 100%). Data are presented as mean ± s.d. of three independent experiments.  97  3.3.6 Nucleotide identity within the VLR affects IRES translation SHAPE analyses showed differential reactivity of the nucleotides in the VLR with NMIA. For instance, the mutant G6618U showed increased reactivities of nucleotides AA6606-7 as compared to the wild type (Figure 3.3 WT and G6618U). Altering AA6606-7 to UU or CC resulted in a general decrease in both 0 and +1 frame translation by approximately 25-40%. Similarly, mutating AUAC6607-10 also decreased 0 frame translation by 15-34% and did not have an effect on +1 frame translation (Figure 3.7). The overall decrease in translation resulting from some of the mutations in the VLR is in agreement with our previous results that the identities of specific nucleotides are important for IRES activity (Au and Jan, 2012).  It remains to be investigated how the VLR plays a role in IRES activity.  3.3.7 Structure probing analysis of mutant IRESs that directs 0 or +1 reading frame translation The mutational analyses revealed that 0 and +1 frame translation can be uncoupled and suggest that the IRES adopts conformations that are associated with specific reading frames. To address this possibility, we performed SHAPE analysis on three mutant IRESs: ΔU6569 and ΔA6554 which exclusively lead to +1 frame translation and G6568C, which directs only 0 frame translation. The SHAPE reactivities of the mutant IRESs differed from those of the wild-type IAPV IGR IRES depending on the mutated residues (Figure 3.8 and Appendix Figure C.2).  98  For the ΔU6569 mutant IRES, we observed an increase of SHAPE reactivities of G6564-A6567 in PKI. This profile differs from the SHAPE reactivities observed in the PKI of the U6562G mutant, which has increased +1 frame translation but maintains the same level of 0 frame translation as WT. In U6562G, the increased SHAPE reactivities were observed in the left of PKI, A6567-U6613, G6568 and A6612.  Interestingly, dramatic changes in shape reactivities of PKI were observed in the ΔA6554 mutant. Part of the stem regions of SL III and the helical stem of PKI were highly modified, including G6574 and A6576-G6584. It is likely that the two stems have assumed a rearranged structure based on the increased flexibilities of nucleotides positioned in only one side of the stems. Nevertherless, despite the altered PKI structure in the ΔA6554 mutant, the IRES was still able to direct +1 frame translation but not 0 frame translation (Figure 3.8 a and b Δ A6554).  As was observed with the G6618U mutant IRES, which also abolishes +1 frame translation, nucleotide U6562 showed decreased SHAPE reactivity in the G6568C mutant IRES, suggesting that this trend is associated with 0 frame translation. Also, similar changes in increased SHAPE reactivities of PKI anticodon-like loop were observed. The G6618U mutant IRES showed increased SHAPE reactivity of G6654-C6656 (Figure 3.3 a). Likewise, the G6568C mutant IRES showed similar increases in G6654 (Figure 3.8 a and b G6568C). To summarize, as was observed in Figure 3.3, subtle differences in SHAPE reactivities of nucleotides, especially within the PKI anticodon-like loop, are associated with IRES-mediated reading frame selection.   99     Figure 3.8 SHAPE analysis of more mutant IAPV IGR IRESs. (a) Quantitative data from SHAPE probing experiments. The SHAPE reactivity of each nucleotide is background-corrected and normalized and shown over the wild-type IRES. 100  The gray bars represent wild-type IRES and the colored bars are mutant IRESs as indicated. The degree of modification is on the y-axis, each nucleotide is on the x-axis, with the location of each domain indicated. The data shown are the averages of three independent experiments ± s.d. (b) SHAPE data superimposed on PKI domain of the IAPV IGR IRES secondary structure. The colour coding indicates the SHAPE reactivities of individual nucleotides (red, high reactivity (>0.8); orange, moderate reactivity (0.5-0.8); green, limited reactivity (0.2-0.5); black, relatively unreactive (<0.2).).      101  3.3.8 Structural changes in the PKI domain of mutant IRESs upon ribosome binding  IRES conformational changes, including the movement of the PKI domain, were observed when the IRES was docked into an 80S ribosome in cryo-EM reconstructions of IGR IRES–80S ribosome complexes (Costantino et al., 2008; Spahn et al., 2004; Zhu et al., 2011). SHAPE probing of the PKI domain of a related member, PSIV, revealed subtle structural changes within the PKI domain upon 80S binding (Costantino et al., 2008). The structural changes are thought to be part of the IGR IRES mechanism (Costantino et al., 2008). Thus, we hypothesize that the different conformations of the IAPV IGR IRESs mutants are maintained after ribosome binding and further subtle structural changes are associated with IGR IRES translation upon ribosome binding. To query the structure of IAPV IGR IRESs in the 80S bound states, we used SHAPE analysis to reveal RNA dynamics in the RNA-ribosome complex. We observed modification changes on nucleotides in the PKI domain between the free IRES and the 80S ribosome-bound state, suggesting there may be structural changes upon ribosome binding at these positions (Figure 3.9 a and b, Appendix Figure C.2). Nucleotides A6554 in the bulge region of the PKI helical stem, G6563-C6566 in the anticodon loop, unpaired nucleotides G6568-U6570 in the PKI anticodon loop, AA6587-8 and A6590 in the loop region of SL III and A6606-A6609 in the VLR region are modified less in the ribosome-bound state than the unbound state (Figure 3.9 b, blue colored nucleotides). Decreases in modification suggest that these nucleotides are protected upon forming an IRES-80S complex. Alternatively, these changes may reflect structural changes in the IRES upon 102  binding to the 80S ribosome. In contrast, nucleotides C6603 and CA6611-2 showed increased modification activity in the WT IRES 80S-bound complex compared to the unbound form (Figure 3.9 b, red colored nucleotides). The increases in modification suggest that these nucleotides are more flexible in the IRES-80S complex.    Figure 3.9 SHAPE analysis of WT IAPV IGR IRES in the unbound and 80S-bound forms. 103  Legend to Figure 3.9. (a) Graph of quantitative data from SHAPE probing of WT IAPV IGR IRES in the unbound and 80S ribosome bound forms. The changes in modification of 80S-bound WT IRESs over unbound WT IRES are shown. The degree of modification is on the y-axis, each nucleotide is on the x-axis, with the location of each domain indicated. (b) Changes of SHAPE reactivity in panel (a) are superimposed on the IAPV IGR IRES PKI domain secondary structure. Changes greater than one s.d. are colored. Red represents nucleotides that are modified more in the context of the 80S ribosome; blue represents those that are modified less. The dashed lines represent one s.d.   Like IAPV IGR IRES, similar changes were observed within the PKI domain of the PSIV IGR IRES from 80S ribosome binding, suggesting that the two IGR IRESs undergo similar conformational changes to facilitate translation (Costantino et al., 2008). However, there are also some unique changes in several nucleotides, notably G6568U6569U6570 and A6554, indicating that those nucleotides may be associated with translation at 0 or +1 reading frame selection by the IAPV IGR IRES (Figure 3.9 and Appendix Figure C.2). The additional element, SL III, of a Type II IGR IRES compared with the Type I IGR IRES members, may also play a role in 0 and +1 frame translation as nucleotides AA6587-8 and A6590 in the loop region of SL III were observed to be modified less upon 80S ribosome binding (Figure 3.9 and Appendix Figure C.2). To better assess the conformations that are associated with 0 and +1 frame translation, the SHAPE reactivity changes were compared between the mutant IRES-ribosome complexes and wild-type IRES-ribosome complexes. Generally, different modification patterns were generated among different IGR IRES mutants bound to 80S ribosome, suggesting different conformations are associated with the IGR IRESs mutants 104  which favour translation in only one or both reading frames translation (Figure 3.10 a and b). Specifically, compared to the wild-type IRES bound to the ribosome, the G6618U mutant IRES, which abolished +1 frame translation, exhibited higher reactivity at the bulge A6554, at nucleotides within VLR region (A6605-U6608), and at nucleotides U6569 and G6565 within the PKI anticodon loop region (Figure 3.10 a and b, G6618U). For the G6568C mutant IRES, which also directs only 0 frame translation, a similar relatively high reactivity was observed at nucleotides U6608-C6611 indicating the flexible nature of the VLR region which may be associated with 0 frame translation. In the G6568C mutant IRES, nucleotides U6562, GGC6564-6 and CU6568-9 in PKI were found to be less reactive to NMIA compared with those in WT-80S ribosome complex (Figure 3.10 a and b, G6568C). These results suggest that these nucleotides are less flexible in the G6568C mutant IRES-ribosome complex compared to the wild-type IRES-ribosome complex. The ΔU6569 mutant IRES, which only directs +1 frame translation, showed an increased reactivity of nucleotides U6608 and C6611 in the VLR region and U6613 on the left side of PKI (Figure 3.10 a and b, ΔU6569). No dramatic changes in reactivity were observed between the ΔU6569 mutant and WT IRES in the unbound and 80S bound forms (Figure 3.9, Figure 3.10 a and b, ΔU6569), suggesting that this region is protected in the IRES-80S ribosome complex. The ΔA6554 mutant IRES, which also only directs +1 frame translation activity, showed similar SHAPE reactivities as the unbound form. Relatively higher reactivites of nucleotides U6570, AG6573-4 and A6576-G6584 within part of the SL III stem region, and the PKI helical stem region were observed (Figure 3.10 a and b, Δ A6554), indicating that these nucleotides remain flexible upon binding to 80S ribosome. It is possible that these nucleotides have no interaction or are still accessible within IRES-80S 105  ribosome complexes. The different reactivity patterns observed between the ΔU6569 mutant IRES and the ΔA6554 mutant IRES, suggest that these two mutants, although both directing +1 frame translation, may utilize different conformations to achieve the same goal. To summarize, though subletly, different conformations of mutants IAPV IGR IRES are formed prior to binding to ribosomes. These conformations may be important for mediating specific reading frame translation.    106   107  Figure 3.10 SHAPE analysis of mutant IAPV IGR IRESs bound to 80S ribosomes. (a) Quantitative data for SHAPE probing of IGR IRESs mutants on 80S ribosome binding. The difference in modification of 80S-bound mutant IRESs over 80S-bound WT IRES is shown as indicated. The degree of modification is on the y-axis, each nucleotide is on the x-axis, with the location of each domain indicated. (b) Changes of SHAPE reactivity compared to WT shown in (a) are superimposed on the IAPV IGR IRES PKI domain secondary structure. Changes greater than two s.d. are colored. Red represents nucleotides that are modified more in the context of the 80S ribosome; blue represents those that are modified less.    108  3.3.9 Positioning of the 80S ribosomes on the mutant IRESs Previous studies have shown that upon 80S assembly, the ribosome is positioned on the CrPV IGR IRES such that the GCU alanine start codon and the preceding CCU triplet occupy the ribosomal A- and P-sites, respectively (Wilson et al., 2000a). To determine whether 0 or +1 reading frame is selected upon IRES binding to the ribosome, we performed toeprinting analysis of IRES-ribosome complexes. Toeprinting analysis is a primer extension -based approach in which the size of cDNA products can infer the nucleotides that occupy the ribosomal P and A sites and thus the translational reading frame of the ribosome on the IRES (Wilson et al., 2000a).  We monitored the toeprints of ribosomes assembled on wild-type and mutant IAPV IGR IRESs using purified salt-washed human ribosomes. Assembly of ribosomes on the wild-type IGR IRES resulted in a strong toeprint at A6628, which is 14 nucleotides downstream of the CCU triplet given that the first C of the CCU is +1. This result indicates that the CCU triplet and the adjacent GGC glycine codon occupy the ribosomal P and A sites, respectively (Wang et al., 2013) (Figure 3.11, lane 2 and 15). As expected, the ΔPKI mutant eliminated the A6628 toeprint (Figure 3.11, lane 4 and 16), in agreement with previous reports that the integrity of PKI is required for proper ribosome positioning on IGR IRESs (Jan and Sarnow, 2002).  To examine reading frame selection by the IRES, we analyzed toeprints of ribosomes assembled on mutant IRESs ΔU6569 and ΔA6554, which specifically mediate +1 frame translation, and on mutant IRES G6568C, which specifically mediates 0 frame translation. Ribosomes assembled on these IRESs produced a strong toeprint at A6628 similar to that observed with the wild-type IRES (Figure 3.11, lanes 6, 8, 10 and 17-19).   109  Furthermore, the toeprints of assembled ribosomes on mutant IRESs U6562G, which increases +1 frame translation, and G6618U, which abolishes +1 frame translation showed an identical strong toeprint at A6628 (Figure 3.11, lanes 12, 14 and 20-21). These results indicate that the position of the ribosome on the mutant IRESs is not altered and suggest that the reading frame may not be selected upon assembly of ribosomes on the IAPV IGR IRES.   110       Figure 3.11 Toeprinting analysis of 80S ribosomes on mutant IAPV IGR IRESs. 40S and 60S subunits were incubated with bicistronic RNAs containing wild-type or mutant IAPV IGR IRES and analyzed by primer extension analysis, as described in section 3.2.6. Reaction products were separated by electrophoresis on a denaturing polyacrylamide gel. The gels were dried and exposed by autoradiography. The location of the major toeprint at A6628 is shown at the right. The presence of toeprint A represents proper ribosome positioning on the IGR IRES such that the CCU triplet occupies the ribosomal P site and is base paired within PKI. Sequencing ladders for the wild-type IRESs are shown, with their respective nucleotide numbers as indicated.    111  3.4 Discussion Most recoding mechanisms act during the translation elongation process or alter reading of termination signals. In contrast, the IAPV IGR IRES is unique in that this IRES initiates translation in either the 0 or +1 reading frame. Because the IRES can directly recruit the ribosome, this system provides a simple yet powerful model for understanding how an RNA structure manipulates the ribosome and sets the reading frame. In this study, we use extensive mutagenesis and biochemical analysis to reveal how the tRNA-like anticodon PKI domain of the IAPV IGR IRES mediates 0 and +1 reading frame selection. Previously, we found that the U6562-G6618 base pair adjacent to the tRNA mimicry PKI domain directs IAPV IGR IRES-mediated +1 frame translation (Ren et al., 2012). This extra base pair in the P site anticodon-codon interaction of the IGR IRES may facilitate entry of a portion of incoming A site tRNAs into the +1 frame, while the remainder pair with the 0 frame start codon. In this study mutating U6562 and G6618 to other combinations can still support +1 frame translation (Figure 3.1), suggesting that base pairing adjacent to IGR IRES is not absolutely necessary for +1 frame translation. Specifically, mutating U6562 to G showed a dramatic increase in +1 frame translation activity and induced subtle conformational changes of the mutant IRES even prior binding to ribosomes. It seems that mutating U6562 to other bases likely alters the conformation of the PKI domain and affects reading frame selection. The wild-type IAPV IGR IRES which has a U at position 6562, adopts a native conformation that still permits basepairing at U6562 and G6618 for +1 frame translation. In summary, the U6562-G6618 base pair can promote +1 frame translation but is not absolutely necessary for +1 frame translation. 112  To assess the possibility that realignment of PKI base pairing or P-site slippage directs the alternative reading frame selection by the IAPV IGR IRES, we eliminated base pairing within PKI by mutating A6613 to U or U6617 to other bases. These mutants showed decreased +1 frame translation. This result suggests that realignment of PKI base pairing does not occur and argues against a P-site slippage event as the integrity of the full five base pairs of the PKI domain is required for +1 frame translation (Figure 3.4). Furthermore, P-site slippage is not likely given that five base pairs within the PKI domain would have to slip to allow realignment of the reading frame. Moreover, toeprinting analysis does not show a shift in the reading frame of ribosomes in complex with wild-type or mutant IRESs, suggesting that the reading frame has not been selected at this step (Figure 3.11). The related virus, CrPV which possesses a Type I IGR IRES, contains a PKI domain that mimics a tRNA anticodon-codon interaction to occupy the ribosomal P site and to initiate translation using a non-AUG codon in the A site (Costantino et al., 2008). On the other hand, the IAPV IGR IRES is a Type II IRES. Notable differences occur between the Type I and II IGR IRESs, especially in the PKI domain, including the presence of SLIII and a longer anticodon-like loop within the Type II IGR IRES. It is possible that these features contribute to the ability of the IAPV IGR IRES to initiate translation in 0 and +1 frames. Interestingly, our results revealed novel mutations within the PKI domain of the IGR IRES that can uncouple IRES-mediated 0 and +1 frame translation (Figure 3.5). One class of mutants including G6618U, G6618C and G6568 to other bases results in exclusive 0 frame translation while another class of mutants including ΔA6554 and ΔU6569 leads primarily to +1 frame translation (Figure 3.1 and 3.5). The role of SLIII is not clear, however. The 113  SLIII domain of the type II TSV IGR IRES is required for IRES-dependent translation (Hatakeyama et al., 2004; Nakashima and Uchiumi, 2008; Cevallos and Sarnow, 2005; Jan and Jang, 2010). Deletion of the TSV SLIII abolishes IRES activity. Moreover, ribosomes assembled on this mutant IRES are not positioned correctly (Jan and Jang, 2010). Similarily, deletion of SLIII within the IAPV IGR IRES disrupts IRES mediated translation in both the 0 and +1 frame. It will be interesting to determine whether this translational defect is due to a defect in ribosome positioning on the IRES. We propose a new model whereby the IRES adopts conformations that direct 0 or +1 frame translation by occluding delivery of the 0 frame aa-tRNA to the A site and thereby allowing the delivery of the +1 frame aa-tRNA.  These conformations are in equilibrium such that the conformations that direct +1 frame constitute 20% of the conformations that direct 0 frame translation. Mutations at key locations of the PKI domain shift the equilibrium of the IRES to adopt conformations that allow delivery of aminoacyl-tRNA in the 0 or +1 frame to the ribosomal A site. From the toeprinting analysis, we propose that the reading frame is not selected upon ribosome assembly on the IRES but instead at a later step of translation such as delivery of the incoming aminoacyl-tRNA which may lock the ribosome in the 0 or +1 frame for translation. Based on mutagenesis studies showing that shortening the SLIII helical stem inhibited 0 but not +1 frame translation, we propose that the SLIII facilitates the delivery of the 0 frame aminoacyl-tRNA to the A site.   We also propose that the U6562/G6618 basepair contributes to the overall structural conformation that directs +1 frame translation.  In support of this idea, we demonstrate that the U6562-G6618 basepair is necessary for exclusive +1 frame translation mediated by the 114  class of mutants such as ΔA6554 or ΔU6569 (Table 3.1). Mutant IRESs such as U6562G (that increases +1 frame translation) must force the IRES to adopt a conformation that allows +1 frame translation. Finally, the model proposes that the IRES adopts a conformation in solution prior to ribosome recruitment, which may prime the ribosome to initiate translation in either 0 or +1 frame. In support, SHAPE analysis reveals subtle conformational changes within the single-stranded regions of the IRES that correlate with 0 or +1 frame translation (Figure 3.3 and 3.8). Previous reports have shown that the IGR IRES adopts a pre-formed structure to recruit the ribosome (Costantino et al., 2008). Upon ribosome binding, the IRES may undergo further changes but maintains an overall similar conformation that mediates 0 or +1 frame translation.        115  Chapter 4: The localization of tagged-ORFx in Drosophila S2 cells 4.1 Introduction ORFx is a novel overlapping gene identified in a subset of dicistroviruses including the honey bee viruses ABPV, IAPV and KBV and the fire ant virus SINV-1 (Firth et al., 2009; Sabath et al., 2009). In Chapter 2, biochemical analyses were used to identify the presence and start site of ORFx. Furthermore, the key structural determinants that mediate the ORFx translation were identified. However, the structure and function of ORFx are unknown. Amino acid sequence alignment shows that the ORFx is highly conserved among this subset of dicistroviruses (Figure 4.1) (Firth et al., 2009; Sabath et al., 2009). Sabath and colleagues performed a protein motif search and identified matches to a protein kinase C phosphorylation site but with a very weak score (Sabath et al., 2009). In Chapter 2, an ORFx peptide of IAPV was detected in virally infected bees using multiple reaction monitoring-MS analysis strongly indicating a potential biological function. However, ORFx does not have any obvious domains that would hint at its function (Firth et al., 2009; Sabath et al., 2009). The ideal way to study the role of ORFx during virus infection is to create an inactivating mutation in ORFx in an infectious clone by introducing a stop codon into the +1 frame and leaving the polyprotein amino acid sequence unaltered. However, this method is compromised by a lack of an infectious cDNA and an established honey bee cell line.  116  As an alternative approach, in this chapter I have explored the effects of overexpression of ORFx in the Drosophila S2 cell line and have assayed its effects on cell viability. The subcellular localization of ORFx is also determined to gain insights into its function in Drosophila S2 cells. Finally, the effects of ORFx localization during dicistrovirus CrPV infection is monitored.    Figure 4.1 ORFx alignment in honey bee viruses. Amino acid sequence alignment of the +1 Frame ORFx of honey bee viruses. The color codes for the consistency of the amino acids. The warmer color (red) represents the more conserved amino acids, whereas the colder color (Blue) represents less conserved.     117  4.2 Materials and methods  4.2.1 DNA constructs and reagents The coding sequences of ORFx (NC_009025) (nucleotides 6619-6903) either in frame with eGFP or with a triple HA (3 × HA) tag at the N- or C- terminus were amplified by PCR and ligated into KpnI site of pAc5.1 vector, containing a Drosophila actin promoter. All the constructs were confirmed by sequencing.  4.2.2 Cell culture and transfection  Drosophila Schneider line 2 (S2) cells were passaged in Schneider’s insect medium supplemented with 10% fetal bovine serum. Transfections of 2 million cells were performed with Xtreme-gene (Roche) using 2 µg of plasmid DNA per 2-cm dish according to the manufacturer’s directions. Cells were analyzed after 2 days of transfection or at indicated time points.  4.2.3 Cell viability assay Two million transfected cells were incubated with a final concentration of 4 g/ml propidium iodide for 30 min. Cells were imaged with an ImageXpress Micro system (Molecular Devices). The nuclei of cells were stained blue with Hoechst 33342. The percentage of the dead cells among the transfected cells was calculated by counting the cells positive for propidium iodide and GFP against all of the GFP-positive cells.  118  4.2.4 Western Blot analysis  S2 cells were lysed in lysis buffer (20 mM HEPES, 150 mM NaCl, 1% Triton X-100, 10% glycerol, 1 mM EDTA, 10 mM tetrapyrophosphate, 100 mM NaF, 17.5 mM β-glycerophosphate, and protease inhibitor cocktail from Roche). Equal mass amounts of lysates were loaded and separated on a SDS-PAGE gel and then transferred to a polyvinylidene difluoride Immobilon-FL membrane (Millipore). Blots were blocked for 1 h in 5% skim milk and Tris-buffered saline (20 mM Tris, 150 mM NaCl, 0.1% Tween-20). Antibodies used were as follows: affinity-purified IAPV ORFx peptide rabbit polyclonal antibody (raised against the IAPV ORFx peptide sequence C-SLRPLVKTRLR PKKC-NH2; GenScript) (1∶500), affinity-purified IAPV VP2 peptide rabbit polyclonal antibody (raised against the IAPV VP2 peptide sequence C-NTMPGDSQQESNTPC-NH2; GenScript) (1∶500), anti-Luciferase (Promega) (1∶1,000), anti-3 X HA (1∶1,000) and anti-eGFP (Promega) (1∶1,000). Primary antibodies were detected using either IRDye 800CW goat anti-rabbit IgG (LI-COR Biosciences) at 1∶10,000 at room temperature for 1 h, and subsequently scanned using an Odyssey imager (LI-COR Biosciences).  4.2.5 Immunofluorenscence and subcellular localization analysis A total of 1 × 106 S2 cells were plated on coverslips precoated with 0.5 mg/ml concanavalin A (Calbiochem) for 1 h. Cells were subsequently fixed with 3% paraformaldehyde in PBS and permeabilized in 0.5% Tween in PBS. After fixation and permeabilization, the cells were washed with PBS and incubated with antibodies in PBS containing 5% BSA for overnight at 4°C. The antibodies include 119  1:1000 mouse-anti-3 X HA, 1:1000 rabbit anti-IAPV VP2, 1:500 rabbit anti-lamin Dm0 (nuclear lamina),  1:500 rabbit anti-gp120 (nucleoporin),  1:500 rabbit anti-KDEL (ER marker). Cells were subsequently washed with PBS three times and incubated for 1 h at 25°C with the secondary antibodies goat anti-rabbit Alexa Fluor 647 IgG (1:500; Invitrogen), goat anti-mouse Alexa Fluor 568 IgG (1:500; Invitrogen). Cells were then washed three times with PBS. Coverslips were mounted on slides with Prolong Gold Antifade Reagent (Invitrogen). Images were acquired by a confocal microscope (Olympus FV1000 using Olympus Fluoview, version 2.0a) with a 60 X oil immersion lens. Images shown are representative of a single Z-section.   4.2.6 Subcellular fractionation A total of 10 × 106 cells were rinsed with PBS, pelleted, resuspended in hypotonic lysis buffer (10 mM Tris-HCl pH 7.4, 10 mM NaCl, 3 mM MgCl2, 50 mM sucrose, 10 mM sodium pyrophosphate, 10 mM NaF, protease inhibitors from Roche Applied Science) and incubated on ice for 10 min. Samples were passed through a 27-gauge needle approximately 10 times. Intact cells were removed by centrifugation at 1,000x g. Samples were then centrifuged at 2,000x g for 10 min at 4 °C. The supernatant was collected as a nuclear-free fraction. The pellet containing nuclei was washed three times in hypotonic lysis buffer containing 250 mM sucrose and 0.1% Nonidet P-40.  120  4.2.7 CrPV propogation, infection and viral titer CrPV virus propogation in Drosophila S2 cells has been previously described (Garrey et al., 2010; Khong and Jan, 2011). A total of 3 × 109 S2 cells were infected with 1 ml of virus suspension for 30 min. S2 cells were subsequently incubated in 60 ml of conditioned medium for 14 h at 25°C. The cells were pelleted, resuspended in phosphate-buffered saline (PBS), and subjected to four freeze-thaw cycles.  To determine viral titer, S2 cells were infected with CrPV at indicated multiplicity of infection (MOI). The S2 cells were harvested and lysed. Viral titers were determined using the TCID50 assay (Reed and Muench, 1938).     121  4.3 Results 4.3.1 Overexpression of tagged-ORFx in Drosophila S2 cells has little effect on cell viability To gain insights into the biological function of ORFx, we overexpressed ORFx in Drosophila S2 cells. To determine the role of ORFx on cell viability, GFP-tagged ORFx (either N- or C- terminally tagged) was transiently transfected into Drosophila S2 cells and cells were subsequently monitered with propidium iodide (PI) staining to assess cell death. Propidium iodide is a fluorescent dye that binds to DNA. When excited with 488nm wavelength light, it fluoresces red. Only dead cells will be permeable to PI and thus will be exhibit red fluorescence after excitation with laser light. To calculate the total number of cells, the nuclei were stained with Hoechst 33342 (Figure 4.2 a). The percentage of the dead cells among the GFP-tagged, ORFx-expressing cells was calculated by counting the cells positive for propidium iodide and GFP against all of the GFP-positive cells. As a control, untransfected cells and cells transfected with eGFP were analyzed in parallel. For untransfected cells, the percentage of cell death was calculated by counting the cells positive for propidium iodide against all of the Hoechst staining positive cells. For eGFP transfected cells, the percentage of cell death was calculated similarly as ORFx-eGFP transfected cells (Figure 4.2 b). Western blot analysis was performed to confirm the expression of ORFx. A band of the expected molecular weight of the tagged ORFx protein was observed, indicating that the tagged ORFx is intact (Figure 4.2 c). Approximately 4% of untransfected S2 cells and 2% eGFP transfected cells displayed PI positive fluorescence indicative of cell death (Figure 4.2 b). Similarly, the 122  ORFx-eGFP and eGFP-ORFx transfected cells displayed about 3 % and 6 % PI positive fluorescence, respectively (Figure 4.2 b). These results indicate that overexpression of ORFx has little effect on cell viability in Drosophila S2 cells (Figure 4.2).  a)  b)    c)  Figure 4.2 The effect of ORFx on cell viability by propidium iodide. 123  Legend to Figure 4.2. (a) A representative picture of staining transfected cells of eGFP, eGFP-ORFx and ORFx-eGFP. (b) Quantitation of cell death rate of normal cells without transfection, eGFP, eGFP-ORFx and ORFx-eGFP transfected cells. Green fluorescence represents eGFP or eGFP tagged-ORFx. Blue fluorescence represents Hoechst staining with nuclear. Red fluorescence represents propidium iodide staining. Data are representative of two independent experiments. (c) In parallel, cell lysates were subjected to Western blotting using anti-GFP antibodies.  4.3.2 Subcellular localization of ORFx in Drosophila S2 cells  As a first step to study the subcellular localization of ORFx, confocal microscopy analysis was performed. In this study, eGFP-tagged-ORFx was transiently transfected into Drosophila S2 cells. The eGFP-tagged-ORFx localized in the peri-nuclear region and formed punctate dots in the cytoplasm in Drosophila S2 cells (Figure 4.3). As a control, the cells expressing only the eGFP tag were analyzed in parallel. As expected, in eGFP expressing cells, the eGFP was localized diffusely in the cytoplasm and nucleus (Figure 4.3 a, upper panel). As the ORFx product is a relatively small protein (~11 kDa) compared to the eGFP tag (~33 kDa), it is possible that the eGFP tag may have an effect on the localization of ORFx in Drosophila S2 cells. To address this, ORFx was cloned at its N- or C- terminus in frame with a triple HA tag. As was observed with eGFP-tagged ORFx, the 3 × HA-tagged ORFx also localized to the peri-nuclear region and punctate signals were observed in the cytoplasm (Figure 4.3 a, lower panel). Western blot analysis confirmed that the tagged-ORFx is intact in the transfected cells (Figure 4.3 b).  124  a)  b)   Figure 4.3 Subcellular localization of tagged-ORFx in Drosophila S2 cells. (a) Confocal scanning images of tagged-ORFx cells fixed after 2 days of transfection. HA tagged-ORFx slides were probed with anti-HA antibodies, as described in section 4.2.5. Green fluorescence (eGFP tagged-ORFx or HA tagged-ORFx) and blue fluorescence (Hoechst staining with nuclear) signals were recorded separately by using appropriate filters. Data are representative of cell images acquired from at least three independent experiments (cells number >100).  Bar, 10 µm. (b) In parallel, cell lysates were subjected to Western blotting using anti-HA antibodies.  125  To further determine where ORFx is localized, its localization was compared to established subcellular markers including nuclear lamin Dm0, nucleoporin gp210 and an ER marker (KDEL) (Figure 4.4). Previous studies have shown that Lamin Dm0, a Drosophila B-type lamin, defines the most peripheral part of the nucleoplasm (Klapper et al., 1997; Qi et al., 2004).  The nucleoporin marker gp210, a component of the nuclear pore complex, is found exclusively in the Drosophila nuclear periphery (Filson et al., 1985). In the eGFP-tagged ORFx expressing cells, the tagged-ORFx was observed to colocalize with nuclear lamina and nucleoporin along the nuclear periphery (Figure 4.4 b and c).  To determine whether the colocalization was significant, Pearson’s colocalization coefficients were calculated using ImageQuant. The GFP tagged-ORFx colocalization with nuclear lamin Dm0 and nucleoporin gp120 exhibited Pearson's coefficients of ~0.6 (a score of 1 indicates perfect correlation), whereas the ER marker KDEL and GFP tagged-ORFx had Pearson's coefficent of ~0.2.  Interestingly, both the N- and C- terminal GFP tagged-ORFx showed similar level of colocalization (Figure 4.4).  These data suggest that the localization of tagged-ORFx overlaps with the nuclear lamina and nucleoporin gp120 in Drosophila S2 cells.    126   127  Figure 4.4 Localization of tagged-ORFx in Drosophila S2 cells with subcellular markers. Localization of a) GFP, b) C- terminal GFP tagged ORFx (ORFx-GFP), c) N-terminal GFP tagged ORFx (GFP-ORFx) in Drosophila S2 cells against subcellular markers of nuclear lamin Dm0, nucleoporin gp210 and ER marker KDEL, as described in section 4.2.5. Graphs on the right represent Pearson’s colocalization coefficients between (b) ORFx-GFP or (c) GFP-ORFx and the indicated subcellular marker. Data are representative of cell images acquired from three independent experiments (cells number >100).  Bar, 10 µm.   To confirm the subcellular localization of ORFx obtained by immunofluorescence, subsequent biochemical cell fractionation analyses were performed. The tagged-ORFx expressing cells were fractionated into cytosolic and nuclear fractions. To determine the success of separation, Western blot analysis was performed using cytosolic and nuclear fractions markers, α-tubulin and the nuclear lamin Dm0, respectively. As expected, α-tubulin was found only in the cytoplasmic fractions whereas the nuclear lamin Dm0 was found exclusively in the nuclear fractions. Both the N- and C- terminal GFP tagged-ORFx were found in both cytosolic and nuclear fractions (Figure 4.5), in agreement with previous confocal analyses that the tagged-ORFx localized to the cytoplasm and nuclear periphery.    128   Figure 4.5 Subcellular fractionation of S2 cells expressing tagged-ORFx. Experiments were performed as in section 4.2.6. Fractions were subjected to Western blot analysis using the antibodies indicated on the left. Representative blots are shown from at least three independent experiments. T: total cell lysate, C: cytosolic fraction, N: nuclear fraction.   As colocalization of ORFx was observed with nucleoporin gp210 by confocal analyses, we hypothesized that ORFx may affect nuclear pore function. To address this, we cotransfected GFP-tagged ORFx with a mCherry-NLS plasmid, which contains a canonical nuclear localization signal (NLS). As a control, in cells expressing only mCherry-NLS plamids alone or with eGFP plasmid, mCherry fluoresence was observed primarily in the nucleus as expected (Figure 4.6).  In cells coexpressed with both the mCherry-NLS and the eGFP-tagged-ORFx, the mCherry-NLS maintained its localization to the nucleus after 48 hours after transfection (Figure 4.6). These results suggest that ORFx does not inhibit the classical NLS-mediated nuclear import (Figure 4.6).  129   Figure 4.6. Localization of mCherry-NLS in GFP-tagged-ORFx expressing S2 cells. Green fluorescence (eGFP or eGFP-tagged-ORFx), red fluorescence (mCherry-NLS), and blue fluorescence signals (Hoechst staining with nuclear) were recorded separately by using appropriate filters, as described in section 4.2.5. Overlay of the three fluorescent signals is shown in the merged column. Data are representative of cell images acquired from three independent experiments (cells number >100).  Bar, 10 µm.     130  4.3.3 Localization of ORFx in CrPV infected S2 cells We next determined whether ORFx localization is affected during virus infection. The ideal approach is to monitor ORFx expression and localization in a honey bee cell line under IAPV infection. However, an immortal honey bee cell line has yet to be established in our lab at this time point. As an alternative approach, we utilized a related model dicistrovirus, CrPV, which can infect Drosophila S2 cells.  In an effort to track whether the subcellular localization of ORFx is altered during CrPV infection, localization of GFP-tagged-ORFx was monitored during CrPV infection at MOI of 10. The cells were fixed, permeabilized and stained with antibodies to CrPV VP2 (Garrey et al., 2010) and to either nuclear lamin Dm0 (Figure 4.7) or the ER marker KDEL (Figure 4.8). Figure 4.7 shows images of GFP-tagged-ORFx colocalization with nuclear lamin Dm0 at 2, 4 and 6 h post infection in S2 cells. Representative images show GFP-tagged-ORFx (green), nuclear lamin Dm0 (red) and CrPV VP2 (white). Colocalization is shown in the merged column. To determine whether the colocalization was significant, Pearson’s colocalization coefficients were calculated using ImageQuant. Colocalization of both N- and C- terminal GFP tagged-ORFx with nuclear lamin Dm0 exhibited Pearson's coefficients of ~0.6 at 2, 4 and 6 h post CrPV infection in S2 cells. Similarly, Figure 4.8 shows images of GFP tagged-ORFx localization with the ER marker KDEL at 2, 4 and 6 h post infection. Pearson's coefficients of ~0.2 at 2, 4 and 6 h post infection were found. These results indicate that GFP-tagged-ORFx maintained a similar localization with and without CrPV infection in S2 cells.  131    132  Figure 4.7 Localization of GFP tagged-ORFx and nuclear lamin Dm0 during CrPV infection in Drosophila S2 cells. Localization of a) GFP, b) C-terminal tagged ORFx-GFP, c) N-terminal tagged GFP-ORFx during CrPV infection at an MOI of 10 at 2, 4 and 6 h post infection in Drosophila S2 cells. Green fluorescence signals indicated eGFP-tagged-ORFx. Cells were stained with anti-lamina antibody which was detected with anti-mouse IgG Alex fluor conjugate (red). CrPV was detected with anti-VP2 antibody followed by anti-rabbit IgG Alex fluor conjugate (far red, white color), as described in section 4.2.5. Overlay of the three fluorescent signals is shown in merge. d) Graph represents Pearson’s colocalization coefficients between ORFx-GFP or GFP-ORFx and subcellular marker nuclear lamin Dm0. Data are representative of cell images acquired from three independent experiments (cells number >100).  Bar, 10 µm.   133    134  Figure 4.8 Localization of GFP tagged-ORFx and ER marker KDEL during CrPV infection in Drosophila S2 cells. Localization of a) GFP, b) C-terminal tagged ORFx-GFP, c) N-terminal tagged GFP-ORFx during CrPV infection at an MOI of 10 at 2, 4 and 6 h post infection in Drosophila S2 cells. Green fluorescence signals indicated eGFP-tagged-ORFx. Cells were stained with anti-ER marker KDEL antibody which was detected with anti-mouse IgG Alex fluor conjugate (red). CrPV was detected with anti-VP2 antibody followed by anti-rabbit IgG Alex fluor conjugate (far red, white color), as described in section 4.2.5. Overlay of the three fluorescent signals is shown in merge. d) Graph represents Pearson’s colocalization coefficients between ORFx-GFP or GFP-ORFx and subcellular ER marker KDEL. Data are representative of cell images acquired from three independent experiments (cells number >100).  Bar, 10 µm.      135  4.3.4 ORFx expression has little effect on CrPV propagation in S2 cells To test whether ORFx expression affects CrPV virus yield in S2 cells, the viral titer of CrPV-infected ORFx-expressing cells harvested at 12 h post infection was determined using end point analysis. Briefly, tagged-ORFx expressing cells were infected with CrPV at an MOI of 10 or 0.1. The cell lysate was harvested at 12 h post infection.  S2 cells were then infected with serial dilutions of each virus stock, and the 50% tissue culture infectious dose was calculated according to the method of Reed and Muench (Reed and Muench, 1938). The cells transfected with N- or C- terminal GFP-tagged or HA-tagged ORFx produced similar viral titers as compared to control cells expressing GFP alone (Figure 4.9 and 4.10). The transfection efficiency of these experiments was optimized to approximately 80%. The results suggest that ORFx expression has little effect on CrPV propagation and viral yield.   Figure 4.9 Viral titer of CrPV infected tagged-ORFx transfected S2 cells. 136  Legend to Figure 4.9. S2 cells were transiently transfected with the indicated expression plasmids. Cells were then infected with CrPV at an MOI of 10 or 0.1 for 12 h. Samples were harvested and lysed, as described in section 4.2.7. Viral titers were measured by end point dilution analysis. Error bars indicate s.d. calculated from three independent experiments.   4.4 Discussion  Functional studies of this novel ORFx have been limited by the fact there are no infectious cDNAs of IAPV and no available established honey bee cell lines. Thus, the role of ORFx remains elusive. This chapter utilizes overexpression studies of ORFx in Drosophila S2 cells.  Amino acid sequence alignment shows that the ORFx is highly conserved among the honey bee dicistroviruses. The ORFx is highly basic, with an isoelectric point of about 11.7. Another viral protein, known as ARFP (alternative reading frame protein) or core+1 protein, is also synthesized by an open reading frame overlapping with the HCV core gene at nucleotide +1 (Varaklioti et al., 2002; Walewski et al., 2001; Xu et al., 2001). It shares similar biochemical properties with ORFx in that the ARFP/ core +1 protein is also highly basic with an isoelectric point of about 11.5 (Walewski et al., 2001). Moreover, ARFP/ core +1 protein is also in an overlapping ORF directed by the HCV IRES. Alternative translation initiation sites and different mechanisms for ARFP/ core+1 ORF expression have been described. But, controversy exists over the mechanism by which it initiates. The existing data suggest that there may be several translation initiation sites within the ARFP/ core+1 ORF. It remains unclear whether these sites are functional simultaneously or at different 137  times of infection in vivo (Baril and Brakier-Gingras, 2005; Boulant et al., 2003; Choi et al., 2003; Vassilaki and Mavromara, 2003). Approximately 10 years after its discovery, little is known about the biological role of the ARFP/ core +1 protein. Specific anti-core+1 antibodies were detected from patients providing strong evidence that it is produced in vivo (Walewski et al., 2001; Xu et al., 2001). However, subsequent studies still leave the function of the ARFP/core +1 protein as an enigma. The ARFP/core +1 protein is not required for viral replication using subgenomic HCV replicon systems lacking the entire structural region in cell culture or uPA-SCID mice (Blight et al., 2000; Lohmann et al., 1999). Moreover, recent studies using efficient cell culture HCV infection systems have shown that nonsense mutations in the ARFP/core +1 ORF have no effect on viral replication or yield (Vassilaki et al., 2008b).  To search for homologs of ORFx, a tool called Protein Homology/AnalogY Recognition Engine (Phyre) was used to predict the structure and function of ORFx (Kelley and Sternberg, 2009). The Phyre server uses a library of known protein structures taken from the Structural Classification of Proteins database and augmented with newer depositions in the Protein Data Bank (PDB). Using Phyre analysis, the IAPV ORFx is predicted to have two helices, possibly a transmembrane domain (Appendix Figure D.1). This is in agreement with previous study that similar proteins were predicted to contain two helical domains that may be a transmembranal segment (Sabath et al., 2009). In the Phyre structural prediction of IAPV ORFx, residues 34-94 of ORFx (63% coverage) show homology to n- acetylglucosaminidase, an enzyme involved in processing of free oligosaccharides in the cytosol; however, it is with very low confidence of 13.1% and 25% i.d (Figure B.2). 138  Although the structure prediction of ORFx does not provide a concrete hint for its function, in this chapter, an alternative approach of overexpressing ORFx in a Drosophila S2 cell line was utilized to study the IAPV ORFx. Overexpression of ORFx has little effect on Drosophila S2 cell viability (Figure 4.2). This result suggests that ORFx alone is not sufficient to induce apoptosis in S2 cells. This is unlike poliovirus 2A protease, which is sufficient to trigger apoptosis, possibly through a caspase-independent pathway (Goldstaub et al., 2000). The PV 2A protease cleaves many cellular proteins including eIF4GI, eIF4GII and PABP (Etchison et al., 1982; Goldstaub et al., 2000; Joachims et al., 1999). It may induce apoptosis by arresting cap-dependent translation of proteins essential for cell viability or by cleavage of other unidentified cellular substrates.  In addition to the cell viability test, the effect of ORFx on cellular gene expression was determined by cotransfecting a firefly luciferase gene reporter and S35 pulse labelling of overall protein synthesis (Appendix Figure E.1). Though no significant difference was observed, the possibility that minor or specific regulation effect towards individual genes could not be excluded in these experiments.  The tagged-ORFx was localized to the peri-nuclear region and forms cytoplasmic punctate dots in Drosophila S2 cells. Interestingly, ORFx colocalizes with nuclear lamin Dm0 and nucleoporin gp210. It is known that molecules of less than 20-40 kDa can cross the nuclear pore complex by passive diffusion (Gorlich and Kutay, 1999; Lyman et al., 2002; Nachury and Weis, 1999). Given the small size of tagged ORFx, the possibility that ORFx passively diffuses through nuclear pores cannot be excluded. That ORFx colocalized with the nuclear lamina is an interesting aspect that requires further investigation. The nuclear lamina not only serve as a structural framework for the nucleus, but are also 139  involved in many essential processes in nuclear function (Dechat et al., 2008). Studies have shown that cytomegalovirus (CMV) and herpes simplex virus type-1 (HSV-1) viral proteins could phosphorylate and disassemble lamins in locations where viruses are accessing the inner nuclear membrane (INM) in order to move to the cytoplasm (Mettenleiter et al., 2006). Both of these viruses are DNA viruses. They replicate their viral DNA and assemble their capsids in the nucleus. So the nucleocapsids exiting the nucleus need to disassemble the lamina (Mettenleiter et al., 2006; Radsak et al., 1991).  As a similar positive viral +1 frame protein, the HCV ARFP/core +1 protein was detected a perinuclear distribution, partially colocalized with the ER and associated with the cell periphery in subcellular localization studies (Vassilaki et al., 2007, 2008a; Vassilaki and Mavromara, 2009). It was also found to colocalize with microtubules during mitosis, suggesting possible functions of the core+1 protein in the regulation of microtubule dynamics and mitosis (Roussel et al., 2003; Vassilaki et al., 2007, 2008a; Xu et al., 2003). ORFx was observed forming cytoplasmic puntates. Colocalization studies showed that they may not be associated with the ER using the ER KDEL marker. It will be very interesting to further investigate where ORFx localized in the cytoplasm. Previous studies have shown that cytoplasmic granules such as stress granules (SGs), which are dense aggregations in the cytosol composed of proteins and RNAs form when the cells are under stress when protein synthesis is inhibited (Anderson and Kedersha, 2009). SGs are dynamic as they disassemble during recovery from the stress (Kimball et al., 2003). SGs do not aggregate in CrPV-infected cells even when the cells are challenged with potential SG inducers (Khong and Jan, 2011). Based on the fact that ORFx maintained similar cytoplasmic punctates during CrPV infection in S2 cells at 2, 4 and 6 hours post infection, 140  it suggests that those punctates are likely not SGs. The possibilities that ORFx may localize to other cellular granules like P bodies can not be excluded. Further investigations are required. Further end point analysis showed that the ORFx overexpression has little effect on CrPV viral yield. It is possible that ORFx overexpression has little effect on CrPV viral yield may not truely reflect ORFx function during IAPV infection. For instance, if any modification of ORFx occurs during natural infection, it will be missing in our overexpression system. Also, if ORFx functions by interaction with other host or viral factors which are possibly unavailable in S2 cells, it will lead to unintentional consequences or have little effect on biological processes that we observed in S2 cells.  141  Chapter 5: Summary and future directions A dicistrovirus has a compact RNA genome of approximately 9 kb in length. It has evolved several strategies to maximize virus production. For efficient virus translation after entry, dicistrovirus infection results in rapid shutdown of host translation through impairment of the eIF4F complex, specifically inhibition of the interaction of eIF4E and eIF4G as early as one hour post-infection (Garrey et al., 2010). To circumvent this block in translation, dicistroviruses emply two IRESs to recruit ribosomes. Remarkably, the expression level and timing of the viral enzymatic and structural proteins are regulated through these two IRESs, a 5’ UTR IRES and an IGR IRES. The two IRESs regulation mechanism is unique not only because the IRESs can use a non-canonical translation mechanism to express viral proteins as an alternative when canonical translation is compromised, but also it provides an efficient and economic way to produce a supermolar ratio of capsid proteins over enzymatic proteins. This thesis focuses on an additional unprecedented feature of IRESs. A subset of dicistroviruses can initiate translation of two overlapping reading frames, thus increasing the coding capacity of the viral genome which expands the list of strategies utilized by viruses. In chapter 2, we identified the existence and start site of the ORFx in the full length IAPV cDNA and dicistronic reporter using mutagenesis and MS analyses in an in vitro translation system. This is a novel feature of IGR IRES-mediated translation. Despite extensive biochemical and structural analyses that have been performed to study the IGR IRES (Costantino and Kieft, 2005; Costantino et al., 2008; Jan et al., 2003; Jan and Sarnow, 2002; Kamoshita et al., 2009; Kanamori and Nakashima, 2001; Pfingsten et al., 2006; 142  Schuler et al., 2006; Wilson et al., 2000a; Zhu et al., 2011), it is very hard to postulate how this RNA element manipulates the ribosome to set two overlapping reading frames. However, the remarkable features of IGR IRES serves as the simpliest model for understanding translation initiation. Clearly, elucidation of the structure of IAPV IGR IRES and its interactions with ribosomes are essential to understand the recoding mechanism.  In collaboration with Dr. Leonard Foster's lab, the expression of a conserved peptide within ORFx was found in IAPV, ABPV and KBV virally-infected honey bees by multiple reaction monitoring mass spectrometry, proving the expression of ORFx. Since ORFx is highly conserved within honey bee dicistroviruses, these data suggest that ORFx may have a role developed through evolution. At this time, the function of ORFx remains elusive. It will be very interesting to investigate the timing and level of ORFx expression during honey bee dicistrovirus infection. In addition, it also remains to be determined how ORFx translation is regulated. How ORFx is temporally and quantitively regulated during viral infection and how it interacts with host proteins might indicate its potential function.  In chapter 3, extensive mutageneses and SHAPE analyses are performed to reveal that the IRES likely adopts different conformations to direct reading frame selection. We propose a model that the IRES adopts conformations that direct 0 or +1 frame translation by occluding delivery of the 0 frame aa-tRNA to the A site and thereby allowing the delivery of the +1 frame aa-tRNA.  These conformations are in equilibrium such that the conformation that directs +1 frame comprises about 20% of the conformation that directs 0 frame translation. This model could be further tested using in vitro reconstitution assays showing the binding or translocation capability of certain aa-tRNA onto either the 0 or +1 reading frame. Previously, in a reconstituted system composed of purified elongation 143  factors eEF1A, eEF2 and aminoacylated tRNA, it has been shown that the CrPV IGR IRES-ribosome complexes displayed a 6nt downstream toeprints in the presence of cycloheximide, indicating that after the CrPV CCU triplet was positioned in P-site of the ribosome, it was followed by two cycles of translocation with the deacylated tRNA in the ribosomal E-site (Jan et al., 2003).  Novel classes of mutants which can direct either only 0 or +1 frame translation have been identified. Mutations at key locations of the PKI domain shift the equilibrium of the IRES to adopt conformations that prefer either the 0 or +1 frame translation. The fact that a single mutation can increase 0 or +1 frame translation suggests that the ratio of +1 to 0 frame translation is tightly regulated in the WT IRES. In most programmed recoding mechanisms in viral genomes, the ratio of frameshifted and in-frame viral protein expression is tightly regulated. Alterations by mutations or other regulation that change the ratio can affect virus infection. Thus, the ratio of +1 and 0 frame translation mediated by the IAPV IGR IRES is likely important for virus infection. Moreover, the regulation of reading frame selection by the IAPV IGR IRES may be important to control the timing of ORFx expression during virus infection. IAPV was found to be associated with honey bee colony collapse disorder in the last decade, which brings numerous environmental and economical losses (Bromenshenk et al., 2010; Cox-Foster et al., 2007). Finding compounds that affect the ratio of +1 and 0 frame translation might lead to a new strategy to control dicistrovirus production. Though there is no envidence that IAPV infection is a sole pathogen directly causing colony collapse disorder, this approach may be still helpful to control IAPV viral infection in honey bees. 144  Our data show that the IRES forms several conformations prior to binding to ribosomes. Upon 80S ribosome binding, the conformational features were maintained, suggesting that the IRES conformations that favour a particular reading frame are selected prior to binding to the ribosome. It will be interesting to determine whether trans-acting factors or specific pathways can modulate IAPV IGR IRES mediated 0 and +1 frame translation during virus infection, possibly by affecting the conformation of the IRES.  Additional studies on the conformations and interactions of the tRNA-like domain of the IAPV IGR IRES with the ribosome will shed light into determinants that direct reading frame selection. Recoding in the +1 frame has also been found in other systems. A subset of mutant tRNAs that contain mutations in the anticodon stem or elbow can suppress +1 frameshift mutations (Nasvall et al., 2009). It has been suggested that the major forces driving frameshifting include tRNA selection through stabilization of the A-site tRNA-mRNA interaction by stacking effects and the interaction between codon and the rRNA (Beier and Grimm, 2001; Buckingham, 1994).  The frameshift suppressor tRNA shifts the reading frame by +1 by permitting a near cognate tRNA in the ribosomal P site to slip (Qian et al., 1998). The IAPV IGR IRES, although it occupies the ribosomal P site, does not involve slippage. It directs +1 frame selection at the step of initiation, although the ribosome is initiating translation in the elongation phase, and adopts a structural conformation to start the delivery of the +1 frame aminoacyl-tRNA to the A site.  In chapter 4, it has been found that the tagged ORFx was colocalized to the peri-nuclear region in Drosophila S2 cell (Figure 4.3). Further subcellular fractionation analyses suggests that ORFx may localize to the nuclear membrane and cytoplasmic membrane-145  bound fractions (data not shown). It is possible that the positive charge of ORFx facilitates its localizing to membrane-rich regions. Whether the transmembrane domain of ORFx is involved in determing its localization or not remains an open question. Interestingly, my finding that ORFx may localize to the nuclear membrane, indicates that it may be involved in nuclear import or export. As no known NLS or NES was found in ORFx and no ORFx localization was observed in the cell nucleus, it is unlikely that ORFx itself is transported into the nucleus. Another possible function of ORFx is association with the nuclear pore complex (NPCs) to control the entry and exit of large molecules from the cell nucleus. Further analyses like pull down assays could be carried out to test this possibility.  Despite our efforts, the role of ORFx remains elusive. Thus, engineering a cDNA clone to knock out the ORFx in the +1 frame without altering the viral structural polyprotein is the most direct and natural way to test whether ORFx is essential in the viral life cycle. It remains to be investigated how ORFx is translated in a context with other viral proteins, viral genome and in the natural host, honey bees, which may modify or interact with ORFx to achieve its function. Considerable effort was made to generate an infectious cDNA of IAPV. We tried to construct different IAPV cDNAs using sequenced IAPV genomes as a guide. Different IAPV cDNA mutants according to published sequences were in vitro transcribed. These IAPV RNAs were transfected or electroporated into honeybee eggs primary cells and insect cells. Unforturnately, those cDNA clones were not found to be replicable by RT-PCR or Northern blot analysis.   146  We also used a baculovirus-expression system to generate an infectious IAPV virion. The idea was to use an insect cell line as a platform to propagate the infectious IAPV virus using the baculovirus system. Briefly, the IAPV cDNAs were cloned into a bacmid plasmid which was integrated into baculovirus genome. With the help of baculovirus, if an IAPV viron can be made, it will egress and enter the cells facilitated by the baculoviruses. However, no virus replication was detected by RT-PCR analyses. This suggests that the baculovirus system may not facilitate the replication of IAPV. Possible reasons for the failure of trials of IAPV infectious cDNA clones are as follows. First, the cDNA sequence of IAPV was synthesized and then engineered according to a published online database. RdRps are notoriously error-prone which results in the error-prone nature of RNA genome replication. It is possible that one or more deleterious mutations exist in the online published sequences that leads to a defect in virus replication. Second, the RNAs we used for transfection or electroporation were obtained from in vitro transcription. This product may differ from the natural viral RNA by lacking essential RNA modifications. Third, the insect cells, like S2 and Sf21 cells, in which we tried to do transfection may lack appropriate acceptors for virus entry, resulting in a low level of viral proteins and RNAs under the detection limit.  To obtain an infectious cDNA clone of IAPV, efforts should be put in the following aspects. First, a pure population of IAPV needs to be isolated. This is very difficult and requires tremendous work because honey bees are infected with a combination of pathogens. In particularly, the dicistroviruses which infect honey bees, like the IAPV, ABPV and KBV are very similar in size and physical properties. This leads to difficulty in isolating one population from another. Second, after acquiring the pure IAPV virons, the 147  genome of infectious IAPV virons can be amplified using RT-PCR and then cloned. By testing multiple clones directly derived from infectious virons, it is possible that an infectious cDNA will be found. Third, susceptible and permissive cells are needed. Those cells should have not only functional receptors but also should support viral replication. In order to promote the propagation and analyses of IAPV, the susceptible and permissive cell lines need to be immortal. Recently, an established insect cell line AmE-711 was developed and sustained growth in vitro, providing a potential tool for studying honey bee (Goblirsch et al., 2013). Though no success has yet been achieved in our laboratory, it will be very exciting if a platform of immortal honey bee cell lines can be created to speed up the attempt of building an infectious cDNA clone. These will serve as a basis to understand not only the function of ORFx but also the entire life cycle of IAPV. Summarizing, we have discovered a novel mechanism that increases the coding capacity of a virus through the IAPV IGR IRES. These studies of IAPV IGR-IRES will further our understanding of IRESs initiation translation. By extension, this research into the overlapping gene within the IAPV genome will also provide a wealth of insight into how IAPV and its host interact. The findings presented in this thesis provide an excellent foundation for such future prospects.      148  References Acker, M.G., Shin, B.S., Dever, T.E., and Lorsch, J.R. (2006). Interaction between eukaryotic initiation factors 1A and 5B is required for efficient ribosomal subunit joining. Journal of Biological Chemistry 281, 8469-8475. Acker, M.G., Shin, B.S., Nanda, J.S., Saini, A.K., Dever, T.E., and Lorsch, J.R. (2009). Kinetic Analysis of Late Steps of Eukaryotic Translation Initiation. J Mol Biol 385, 491-506. Agrawal, R.K., Heagle, A.B., Penczek, P., Grassucci, R.A., and Frank, J. (1999). EF-G-dependent GTP hydrolysis induces translocation accompanied by large conformational changes in the 70S ribosome. Nature Structural Biology 6, 643-647. Aitken, C.E., and Lorsch, J.R. (2012). A mechanistic overview of translation initiation in eukaryotes. Nat Struct Mol Biol 19, 568-576. Anderson, P., and Kedersha, N. (2009). Stress granules. Curr Biol 19, R397-398. Atkins, J.F., and Bjork, G.R. (2009). A gripping tale of ribosomal frameshifting: extragenic suppressors of frameshift mutations spotlight P-site realignment. Microbiol Mol Biol Rev 73, 178-210. Au, H.H.T., and Jan, E. (2012). Insights into Factorless Translational Initiation by the tRNA-Like Pseudoknot Domain of a Viral IRES. Plos One 7. Baird, S.D., Turcotte, M., Korneluk, R.G., and Holcik, M. (2006). Searching for IRES. Rna-a Publication of the Rna Society 12, 1755-1785. Baril, M., and Brakier-Gingras, L. (2005). Translation of the F protein of hepatitis C virus is initiated at a non-AUG codon in a +1 reading frame relative to the polyprotein. Nucleic Acids Res 33, 1474-1486. Beier, H., and Grimm, M. (2001). Misreading of termination codons in eukaryotes by natural nonsense suppressor tRNAs. Nucleic Acids Res 29, 4767-4782. Belcourt, M.F., and Farabaugh, P.J. (1990). Ribosomal frameshifting in the yeast retrotransposon Ty: tRNAs induce slippage on a 7 nucleotide minimal site. Cell 62, 339-352. 149  Blight, K.J., Kolykhalov, A.A., and Rice, C.M. (2000). Efficient initiation of HCV RNA replication in cell culture. Science 290, 1972-1974. Bonning, B.C., and Miller, W.A. (2010). Dicistroviruses. Annu Rev Entomol 55, 129-150. Boulant, S., Becchi, M., Penin, F., and Lavergne, J.P. (2003). Unusual multiple recoding events leading to alternative forms of hepatitis C virus core protein from genotype 1b. Journal of Biological Chemistry 278, 45785-45792. Brierley, I., Boursnell, M.E., Binns, M.M., Bilimoria, B., Blok, V.C., Brown, T.D., and Inglis, S.C. (1987). An efficient ribosomal frame-shifting signal in the polymerase-encoding region of the coronavirus IBV. EMBO J 6, 3779-3785. Brierley, I., Digard, P., and Inglis, S.C. (1989). Characterization of an efficient coronavirus ribosomal frameshifting signal: requirement for an RNA pseudoknot. Cell 57, 537-547. Brierley, I., Jenner, A.J., and Inglis, S.C. (1992). Mutational analysis of the "slippery-sequence" component of a coronavirus ribosomal frameshifting signal. J Mol Biol 227, 463-479. Bromenshenk, J.J., Henderson, C.B., Wick, C.H., Stanford, M.F., Zulich, A.W., Jabbour, R.E., Deshpande, S.V., McCubbin, P.E., Seccomb, R.A., Welch, P.M., et al. (2010). Iridovirus and microsporidian linked to honey bee colony decline. PLoS One 5, e13181. Buckingham, R.H. (1994). Codon context and protein synthesis: enhancements of the genetic code. Biochimie 76, 351-354. Cevallos, R.C., and Sarnow, P. (2005). Factor-independent assembly of elongation-competent ribosomes by an internal ribosome entry site located in an RNA virus that infects penaeid shrimp. J Virol 79, 677-683. Chan, Q.W.T., Howes, C.G., and Foster, L.J. (2006). Quantitative comparison of caste differences in honeybee hemolymph. Molecular & Cellular Proteomics 5, 2252-2262. Cheung, Y.N., Maag, D., Mitchell, S.F., Fekete, C.A., Algire, M.A., Takacs, J.E., Shirokikh, N., Pestova, T., Lorsch, J.R., and Hinnebusch, A.G. (2007). Dissociation 150  of eIF1 from the 40S ribosomal subunit is a key step in start codon selection in vivo. Genes Dev 21, 1217-1230. Choi, J., Xu, Z., and Ou, J.H. (2003). Triple decoding of hepatitis C virus RNA by programmed translational frameshifting. Mol Cell Biol 23, 1489-1497. Clark, D.J., Bilanchone, V.W., Haywood, L.J., Dildine, S.L., and Sandmeyer, S.B. (1988). A yeast sigma composite element, TY3, has properties of a retrotransposon. Journal of Biological Chemistry 263, 1413-1423. Coldwell, M.J., Mitchell, S.A., Stoneley, M., MacFarlane, M., and Willis, A.E. (2000). Initiation of Apaf-1 translation by internal ribosome entry. Oncogene 19, 899-905. Costantino, D., and Kieft, J.S. (2005). A preformed compact ribosome-binding domain in the cricket paralysis-like virus IRES RNAs. RNA 11, 332-343. Costantino, D.A., Pfingsten, J.S., Rambo, R.P., and Kieft, J.S. (2008). tRNA-mRNA mimicry drives translation initiation from a viral IRES. Nature Structural & Molecular Biology 15, 57-64. Cox-Foster, D.L., Conlan, S., Holmes, E.C., Palacios, G., Evans, J.D., Moran, N.A., Quan, P.L., Briese, T., Hornig, M., Geiser, D.M., et al. (2007). A metagenomic survey of microbes in honey bee colony collapse disorder. Science 318, 283-287. Das, R., Laederach, A., Pearlman, S.M., Herschlag, D., and Altman, R.B. (2005). SAFA: semi-automated footprinting analysis software for high-throughput quantification of nucleic acid footprinting experiments. RNA 11, 344-354. de Breyne, S., Yu, Y., Unbehaun, A., Pestova, T.V., and Hellen, C.U. (2009). Direct functional interaction of initiation factor eIF4G with type 1 internal ribosomal entry sites. Proc Natl Acad Sci U S A 106, 9197-9202. Dechat, T., Pfleghaar, K., Sengupta, K., Shimi, T., Shumaker, D.K., Solimando, L., and Goldman, R.D. (2008). Nuclear lamins: major factors in the structural organization and function of the nucleus and chromatin. Genes & Development 22, 832-853. den Boon, J.A., Snijder, E.J., Chirnside, E.D., de Vries, A.A., Horzinek, M.C., and Spaan, W.J. (1991). Equine arteritis virus is not a togavirus but belongs to the coronaviruslike superfamily. J Virol 65, 2910-2920. 151  Denison, M.R., Zoltick, P.W., Leibowitz, J.L., Pachuk, C.J., and Weiss, S.R. (1991). Identification of polypeptides encoded in open reading frame 1b of the putative polymerase gene of the murine coronavirus mouse hepatitis virus A59. J Virol 65, 3076-3082. Dever, T.E., and Green, R. (2012). The elongation, termination, and recycling phases of translation in eukaryotes. Cold Spring Harb Perspect Biol 4, a013706. Dinman, J.D., Icho, T., and Wickner, R.B. (1991). A -1 ribosomal frameshift in a double-stranded RNA virus of yeast forms a gag-pol fusion protein. Proc Natl Acad Sci U S A 88, 174-178. Dmitriev, S.E., Terenin, I.M., Andreev, D.E., Ivanov, P.A., Dunaevsky, J.E., Merrick, W.C., and Shatsky, I.N. (2010). GTP-independent tRNA delivery to the ribosomal P-site by a novel eukaryotic translation factor. Journal of Biological Chemistry 285, 26779-26787. Domier, L.L., McCoppin, N.K., and D'Arcy, C.J. (2000). Sequence requirements for translation initiation of Rhopalosiphum padi virus ORF2. Virology 268, 264-271. Dorner, S., Brunelle, J.L., Sharma, D., and Green, R. (2006). The hybrid state of tRNA binding is an authentic translation elongation intermediate. Nature Structural & Molecular Biology 13, 234-241. Eaton, B.T., and Steacie, A.D. (1980). Cricket Paralysis Virus-Rna Has a 3' Terminal Poly(a). Journal of General Virology 50, 167-171. Etchison, D., Milburn, S.C., Edery, I., Sonenberg, N., and Hershey, J.W. (1982). Inhibition of HeLa cell protein synthesis following poliovirus infection correlates with the proteolysis of a 220,000-dalton polypeptide associated with eucaryotic initiation factor 3 and a cap binding protein complex. Journal of Biological Chemistry 257, 14806-14810. Farabaugh, P.J. (1995). Post-transcriptional regulation of transposition by Ty retrotransposons of Saccharomyces cerevisiae. Journal of Biological Chemistry 270, 10361-10364. Farabaugh, P.J. (1996). Programmed translational frameshifting. Annu Rev Genet. 30, 507-28. 152  Farabaugh, P.J., Zhao, H., and Vimaladithan, A. (1993). A novel programed frameshift expresses the POL3 gene of retrotransposon Ty3 of yeast: frameshifting without tRNA slippage. Cell 74, 93-103. Filson, A.J., Lewis, A., Blobel, G., and Fisher, P.A. (1985). Monoclonal antibodies prepared against the major Drosophila nuclear Matrix-pore complex-lamina glycoprotein bind specifically to the nuclear envelope in situ. Journal of Biological Chemistry 260, 3164-3172. Firth, A.E., Wang, Q.S., Jan, E., and Atkins, J.F. (2009). Bioinformatic evidence for a stem-loop structure 5 '-adjacent to the IGR-IRES and for an overlapping gene in the bee paralysis dicistroviruses. Virology Journal 6. Frank, J., and Agrawal, R.K. (2000). A ratchet-like inter-subunit reorganization of the ribosome during translocation. Nature 406, 318-322. Frank, J., Gao, H.X., Sengupta, J., Gao, N., and Taylor, D.J. (2007). The process of mRNA-tRNA translocation. Proc Natl Acad Sci U S A 104, 19671-19678. Gale, M. Jr, Tan, S.L., Katze, M.G. (2000). Translational control of viral gene expression in eukaryotes. Microbiol. Mol. Biol. Rev. 64, 239–280 Garrey, J.L., Lee, Y.Y., Au, H.H.T., Bushell, M., and Jan, E. (2010). Host and Viral Translational Mechanisms during Cricket Paralysis Virus Infection. Journal of Virology 84, 1124-1138. Goblirsch, M.J., Spivak, M.S., and Kurtti, T.J. (2013). A cell line resource derived from honey bee (Apis mellifera) embryonic tissues. PLoS One 8, e69831. Goldstaub, D., Gradi, A., Bercovitch, Z., Grosmann, Z., Nophar, Y., Luria, S., Sonenberg, N., and Kahana, C. (2000). Poliovirus 2A protease induces apoptotic cell death. Mol Cell Biol 20, 1271-1277. Gomez-Lorenzo, M.G., Spahn, C.M.T., Agrawal, R.K., Grassucci, R.A., Penczek, P., Chakraburtty, K., Ballesta, J.P.G., Lavandera, J.L., Garcia-Bustos, J.F., and Frank, J. (2000). Three-dimensional cryo-electron microscopy localization of EF2 in the Saccharomyces cerevisiae 80S ribosome at 17.5 angstrom resolution. Embo Journal 19, 2710-2718. 153  Gorlich, D., and Kutay, U. (1999). Transport between the cell nucleus and the cytoplasm. Annual Review of Cell and Developmental Biology 15, 607-660. Griffiths, A., and Coen, D.M. (2005). An unusual internal ribosome entry site in the herpes simplex virus thymidine kinase gene. Proc Natl Acad Sci U S A 102, 9667-9672. Groppelli, E., Belsham, G.J., and Roberts, L.O. (2007). Identification of minimal sequences of the Rhopalosiphum padi virus 5 ' untranslated region required for internal initiation of protein synthesis in mammalian, plant and insect translation systems. Journal of General Virology 88, 1583-1588. Hansen, L.J., Chalker, D.L., and Sandmeyer, S.B. (1988). Ty3, a yeast retrotransposon associated with tRNA genes, has homology to animal retroviruses. Mol Cell Biol 8, 5245-5256. Hinnebusch, A.G. (2011). Molecular Mechanism of Scanning and Start Codon Selection in Eukaryotes. Microbiology and Molecular Biology Reviews 75, 434-467. Hinnebusch, A.G., and Lorsch, J.R. (2012). The Mechanism of Eukaryotic Translation Initiation: New Insights and Challenges. Cold Spring Harbor Perspectives in Biology 4. Hopfield, J.J. (1974). Kinetic proofreading: a new mechanism for reducing errors in biosynthetic processes requiring high specificity. Proc Natl Acad Sci U S A 71, 4135-4139. Icho, T., and Wickner, R.B. (1989). The double-stranded RNA genome of yeast virus L-A encodes its own putative RNA polymerase by fusing two open reading frames. Journal of Biological Chemistry 264, 6716-6723. Irwin, B., Aye, M., Baldi, P., Beliakova-Bethell, N., Cheng, H., Dou, Y., Liou, W., and Sandmeyer, S. (2005). Retroviruses and yeast retrotransposons use overlapping sets of host genes. Genome Res 15, 641-654. Ivanov, I.P., and Atkins, J.F. (2007). Ribosomal frameshifting in decoding antizyme mRNAs from yeast and protists to humans: close to 300 cases reveal remarkable diversity despite underlying conservation. Nucleic Acids Res 35, 1842-1858. Jacks, T., Madhani, H.D., Masiarz, F.R., and Varmus, H.E. (1988a). Signals for ribosomal frameshifting in the Rous sarcoma virus gag-pol region. Cell 55, 447-458. 154  Jacks, T., Power, M.D., Masiarz, F.R., Luciw, P.A., Barr, P.J., and Varmus, H.E. (1988b). Characterization of Ribosomal Frameshifting in Hiv-1 Gag-Pol Expression. Nature 331, 280-283. Jacks, T., Townsley, K., Varmus, H.E., and Majors, J. (1987). Two efficient ribosomal frameshifting events are required for synthesis of mouse mammary tumor virus gag-related polyproteins. Proc Natl Acad Sci U S A 84, 4298-4302. Jacks, T., and Varmus, H.E. (1985). Expression of the Rous sarcoma virus pol gene by ribosomal frameshifting. Science 230, 1237-1242. Jan, E. (2006). Divergent IRES elements in invertebrates. Virus Res 119, 16-28. Jan, E., Kinzy, T.G., and Sarnow, P. (2003). Divergent tRNA-like element supports initiation, elongation, and termination of protein biosynthesis. Proc Natl Acad Sci U S A 100, 15410-15415. Jan, E., and Sarnow, P. (2002). Factorless ribosome assembly on the internal ribosome entry site of cricket paralysis virus. J Mol Biol 324, 889-902. Jang, C.J., and Jan, E. (2010). Modular domains of the Dicistroviridae intergenic internal ribosome entry site. RNA 16, 1182-1195. Jang, C.J., Lo, M.C., and Jan, E. (2009). Conserved element of the dicistrovirus IGR IRES that mimics an E-site tRNA/ribosome interaction mediates multiple functions. J Mol Biol 387, 42-58. Jang, S.K., Krausslich, H.G., Nicklin, M.J., Duke, G.M., Palmenberg, A.C., and Wimmer, E. (1988). A segment of the 5' nontranslated region of encephalomyocarditis virus RNA directs internal entry of ribosomes during in vitro translation. Journal of Virology 62, 2636-2643. Jiang, H., Coleman, J., Miskimins, R., Srinivasan, R., and Miskimins, W.K. (2007). Cap-independent translation through the p27 5'-UTR. Nucleic Acids Res 35, 4767-4778. Joachims, M., Van Breugel, P.C., and Lloyd, R.E. (1999). Cleavage of poly(A)-binding protein by enterovirus proteases concurrent with inhibition of translation in vitro. J Virol 73, 718-727. 155  Kamoshita, N., Nomoto, A., and RajBhandary, U.L. (2009). Translation initiation from the ribosomal A site or the P site, dependent on the conformation of RNA pseudoknot I in dicistrovirus RNAs. Molecular Cell 35, 181-190. Kanamori, Y., and Nakashima, N. (2001). A tertiary structure model of the internal ribosome entry site (IRES) for methionine-independent initiation of translation. RNA 7, 266-274. Kedersha, N., Chen, S., Gilks, N., Li, W., Miller, I.J., Stahl, J., and Anderson, P. (2002). Evidence that ternary complex (eIF2-GTP-tRNA(i)(Met))-deficient preinitiation complexes are core constituents of mammalian stress granules. Mol Biol Cell 13, 195-210. Kedersha, N.L., Gupta, M., Li, W., Miller, I., and Anderson, P. (1999). RNA-binding proteins TIA-1 and TIAR link the phosphorylation of eIF-2 alpha to the assembly of mammalian stress granules. J Cell Biol 147, 1431-1442. Kelley, L.A., and Sternberg, M.J.E. (2009). Protein structure prediction on the Web: a case study using the Phyre server. Nature Protocols 4, 363-371. Khong, A., and Jan, E. (2011). Modulation of Stress Granules and P Bodies during Dicistrovirus Infection. Journal of Virology 85, 1439-1451. Kieft, J.S. (2008). Viral IRES RNA structures and ribosome interactions. Trends Biochem Sci 33, 274-283. Kieft, J.S. (2009). Comparing the three-dimensional structures of Dicistroviridae IGR IRES RNAs with other viral RNA structures. Virus Research 139, 148-156. Kim, J.M., Vanguri, S., Boeke, J.D., Gabriel, A., and Voytas, D.F. (1998). Transposable elements and genome organization: a comprehensive survey of retrotransposons revealed by the complete Saccharomyces cerevisiae genome sequence. Genome Res 8, 464-478. Kim, T.D., Woo, K.C., Cho, S., Ha, D.C., Jang, S.K., and Kim, K.T. (2007). Rhythmic control of AANAT translation by hnRNP Q in circadian melatonin production. Genes Dev 21, 797-810. 156  Kimball, S.R., Horetsky, R.L., Ron, D., Jefferson, L.S., and Harding, H.P. (2003). Mammalian stress granules represent sites of accumulation of stalled translation initiation complexes. Am J Physiol Cell Physiol 284, C273-284. King, L.A., and Moore, N.F. (1988). Evidence for the Presence of a Genome-Linked Protein in 2 Insect Picornaviruses, Cricket Paralysis and Drosophila-C Viruses. Fems Microbiology Letters 50, 41-44. Klapper, M., Exner, K., Kempf, A., Gehrig, C., Stuurman, N., Fisher, P.A., and Krohne, G. (1997). Assembly of A- and B-type lamins studied in vivo with the baculovirus system. J Cell Sci 110 ( Pt 20), 2519-2532. Kollmus, H., Honigman, A., Panet, A., and Hauser, H. (1994). The sequences of and distance between two cis-acting signals determine the efficiency of ribosomal frameshifting in human immunodeficiency virus type 1 and human T-cell leukemia virus type II in vivo. J Virol 68, 6087-6091. Kozak, M. (2003). Alternative ways to think about mRNA sequences and proteins that appear to promote internal initiation of translation. Gene 318, 1-23. Kozak, M. (2005). A second look at cellular mRNA sequences said to function as internal ribosome entry sites. Nucleic Acids Res 33, 6593-6602. Lancaster, A.M., Jan, E., and Sarnow, P. (2006). Initiation factor-independent translation mediated by the hepatitis C virus internal ribosome entry site. RNA 12, 894-902. Landry, D.M., Hertz, M.I., and Thompson, S.R. (2009). RPS25 is essential for translation initiation by the Dicistroviridae and hepatitis C viral IRESs. Genes & Development 23, 2753-2764. Li, Z., Stahl, G., and Farabaugh, P.J. (2001). Programmed +1 frameshifting stimulated by complementarity between a downstream mRNA sequence and an error-correcting region of rRNA. RNA 7(2), 275-84. Lohmann, V., Korner, F., Koch, J., Herian, U., Theilmann, L., and Bartenschlager, R. (1999). Replication of subgenomic hepatitis C virus RNAs in a hepatoma cell line. Science 285, 110-113. 157  Lorsch, J.R., and Dever, T.E. (2010). Molecular View of 43 S Complex Formation and Start Site Selection in Eukaryotic Translation Initiation. Journal of Biological Chemistry 285, 21203-21207. Lyman, S.K., Guan, T.L., Bednenko, J., Wodrich, H., and Gerace, L. (2002). Influence of cargo size on Ran and energy requirements for nuclear protein import. Journal of Cell Biology 159, 55-67. Maag, D., Algire, M.A., and Lorsch, J.R. (2006). Communication between eukaryotic translation initiation factors 5 and 1A within the ribosomal pre-initiation complex plays a role in start site selection. J Mol Biol 356, 724-737. Maag, D., Fekete, C.A., Gryczynski, Z., and Lorsch, J.R. (2005). A conformational change in the eukaryotic translation preinitiation complex and release of eIF1 signal recognition of the start codon. Molecular Cell 17, 265-275. Macejak, D.G., and Sarnow, P. (1991). Internal initiation of translation mediated by the 5' leader of a cellular mRNA. Nature 353, 90-94. Mayo, M.A. (2002). Virus taxonomy - Houston 2002. Arch Virol 147, 1071-1076. Merino, E.J., Wilkinson, K.A., Coughlan, J.L., and Weeks, K.M. (2005). RNA structure analysis at single nucleotide resolution by selective 2 '-hydroxyl acylation and primer extension (SHAPE). Journal of the American Chemical Society 127, 4223-4231. Mettenleiter, T.C., Klupp, B.G., and Granzow, H. (2006). Herpesvirus assembly: a tale of two membranes. Current Opinion in Microbiology 9, 423-429. Moazed, D., and Noller, H.F. (1989). Intermediate states in the movement of transfer RNA in the ribosome. Nature 342, 142-148. Moore, N.F., Kearns, A., and Pullin, J.S. (1980). Characterization of cricket paralysis virus-induced polypeptides in Drosophila cells. J Virol 33, 1-9. Nachury, M.V., and Weis, K. (1999). The direction of transport through the nuclear pore can be inverted. Proceedings of the National Academy of Sciences of the United States of America 96, 9622-9627. 158  Nakashima, N., Nagasaka, T., Murakami, S., Fukatsu, T., and Satake, T. (2003). Endometrial atypical hyperplasia with clear cell change spreading throughout the endometrium. Ann Diagn Pathol 7, 381-386. Nakashima, N., Sasaki, J., Tsuda, K., Yasunaga, C., and Noda, H. (1998). Properties of a new picorna-like virus of the brown-winged green bug, Plautia stali. Journal of Invertebrate Pathology 71, 151-158. Nakashima, N., and Uchiumi, T. (2009). Functional analysis of structural motifs in dicistroviruses. Virus Research 139, 137-147. Nasvall, S.J., Nilsson, K., and Bjork, G.R. (2009). The ribosomal grip of the peptidyl-tRNA is critical for reading frame maintenance. J Mol Biol 385, 350-367. Ninio, J. (1975). Letter: Specificity in initiation of mRNA translation. Nature 255, 429-430. Nishiyama, T., Yamamoto, H., Shibuya, N., Hatakeyama, Y., Hachimori, A., Uchiumi, T., and Nakashima, N. (2003). Structural elements in the internal ribosome entry site of Plautia stali intestine virus responsible for binding with ribosomes. Nucleic Acids Res 31, 2434-2442. Nishiyama, T., Yamamoto, H., Uchiumi, T., and Nakashima, N. (2007). Eukaryotic ribosomal protein RPS25 interacts with the conserved loop region in a dicistroviral intergenic internal ribosome entry site. Nucleic Acids Research 35, 1514-1521. Nomoto, A., Lee, Y.F., and Wimmer, E. (1976). The 5' end of poliovirus mRNA is not capped with m7G(5')ppp(5')Np. Proc Natl Acad Sci U S A 73, 375-380. Pape, T., Wintermeyer, W., and Rodnina, M.V. (1998). Complete kinetic mechanism of elongation factor Tu-dependent binding of aminoacyl-tRNA to the A site of the E. coli ribosome. EMBO J 17, 7490-7497. Parker, J. (1989). Errors and Alternatives in Reading the Universal Genetic-Code. Microbiological Reviews 53, 273-298. Parsyan, A., Svitkin, Y., Shahbazian, D., Gkogkas, C., Lasko, P., Merrick, W.C., and Sonenberg, N. (2011). mRNA helicases: the tacticians of translational control. Nat Rev Mol Cell Biol 12, 235-245. Pelletier, J., and Sonenberg, N. (1988). Internal initiation of translation of eukaryotic mRNA directed by a sequence derived from poliovirus RNA. Nature 334, 320-325. 159  Pestova, T.V., and Hellen, C.U. (2003). Translation elongation after assembly of ribosomes on the Cricket paralysis virus internal ribosomal entry site without initiation factors or initiator tRNA. Genes Dev 17, 181-186. Pestova, T.V., Lomakin, I.B., and Hellen, C.U. (2004). Position of the CrPV IRES on the 40S subunit and factor dependence of IRES/80S ribosome assembly. EMBO Rep 5, 906-913. Pestova, T.V., Lomakin, I.B., Lee, J.H., Choi, S.K., Dever, T.E., and Hellen, C.U. (2000). The joining of ribosomal subunits in eukaryotes requires eIF5B. Nature 403, 332-335. Pestova, T.V., Shatsky, I.N., Fletcher, S.P., Jackson, R.J., and Hellen, C.U. (1998). A prokaryotic-like mode of cytoplasmic eukaryotic ribosome binding to the initiation codon during internal translation initiation of hepatitis C and classical swine fever virus RNAs. Genes Dev 12, 67-83. Pestova, T.V., Shatsky, I.N., and Hellen, C.U. (1996). Functional dissection of eukaryotic initiation factor 4F: the 4A subunit and the central domain of the 4G subunit are sufficient to mediate internal entry of 43S preinitiation complexes. Mol Cell Biol 16, 6870-6878. Pfingsten, J.S., Costantino, D.A., and Kieft, J.S. (2006). Structural basis for ribosome recruitment and manipulation by a viral IRES RNA. Science 314, 1450-1454. Pfingsten, J.S., and Kieft, J.S. (2008). RNA structure-based ribosome recruitment: lessons from the Dicistroviridae intergenic region IRESes. RNA 14(7), 1255-63. Plank, T.D., and Kieft, J.S. (2012). The structures of nonprotein-coding RNAs that drive internal ribosome entry site function. Wiley Interdiscip Rev RNA 3(2), 195-212. Qi, H., Rath, U., Wang, D., Xu, Y.Z., Ding, Y., Zhang, W., Blacketer, M.J., Paddy, M.R., Girton, J., Johansen, J., et al. (2004). Megator, an essential coiled-coil protein that localizes to the putative spindle matrix during mitosis in Drosophila. Mol Biol Cell 15, 4854-4865. Qian, Q., Curran, J.F., and Bjork, G.R. (1998). The methyl group of the N6-methyl-N6-threonylcarbamoyladenosine in tRNA of Escherichia coli modestly improves the efficiency of the tRNA. J Bacteriol 180, 1808-1813. 160  Radsak, K.D., Brucher, K.H., and Georgatos, S.D. (1991). Focal Nuclear-Envelope Lesions and Specific Nuclear Lamin a/C Dephosphorylation during Infection with Human Cytomegalovirus. European Journal of Cell Biology 54, 299-304. Reboll, M.R., Oumard, A., Gazdag, A.C., Renger, I., Ritter, B., Schwarzer, M., Hauser, H., Wood, M., Yamada, M., Resch, K., et al. (2007). NRF IRES activity is mediated by RNA binding protein JKTBP1 and a 14-nt RNA element. RNA 13, 1328-1340. Reed, L.J., and Muench, H. (1938). A simple method of estimating fifty percent endpoints. Am. J. Hygiene 27, 493-497. Ren, Q., Wang, Q.S., Firth, A.E., Chan, M.M., Gouw, J.W., Guarna, M.M., Foster, L.J., Atkins, J.F., and Jan, E. (2012). Alternative reading frame selection mediated by a tRNA-like domain of an internal ribosome entry site. Proc Natl Acad Sci U S A 109, E630-639. Ren, Q., Au, H.H.T., Wang Q.S., Lee, S., and Jan, E. (2014) Structural determinants of an internal ribosome entry site that direct translational reading frame selection. Nucleic Acids Res. In press. Roth, J.R. (1981). Frameshift suppression. Cell 24, 601-602. Roussel, J., Pillez, A., Montpellier, C., Duverlie, G., Cahour, A., Dubuisson, J., and Wychowski, C. (2003). Characterization of the expression of the hepatitis C virus F protein. J Gen Virol 84, 1751-1759. Royall, E., Woolaway, K.E., Schacherl, J., Kubick, S., Belsham, G.J., and Roberts, L.O. (2004). The Rhopalosiphum padi virus 5 ' internal ribosome entry site is functional in Spodoptera frugiperda 21 cells and in their cell-free lysates: implications for the baculovirus expression system. Journal of General Virology 85, 1565-1569. Ruusala, T., Ehrenberg, M., and Kurland, C.G. (1982). Catalytic effects of elongation factor Ts on polypeptide synthesis. EMBO J 1, 75-78. Sabath, N., Price, N., and Graur, D. (2009). A potentially novel overlapping gene in the genomes of Israeli acute paralysis virus and its relatives. Virology Journal 6. Sasaki, J., and Nakashima, N. (1999). Translation initiation at the CUU codon is mediated by the internal ribosome entry site of an insect picorna-like virus in vitro. Journal of Virology 73, 1219-1226. 161  Sasaki, J., and Nakashima, N. (2000). Methionine-independent initiation of translation in the capsid protein of an insect RNA virus. Proc Natl Acad Sci U S A 97, 1512-1515. Schneider-Poetsch, T., Usui, T., Kaida, D., and Yoshida, M. (2010). Garbled messages and corrupted translations. Nat Chem Biol. 6(3), 189-198. Schuler, M., Connell, S.R., Lescoute, A., Giesebrecht, J., Dabrowski, M., Schroeer, B., Mielke, T., Penczek, P.A., Westhof, E., and Spahn, C.M.T. (2006). Structure of the ribosome-bound cricket paralysis virus IRES RNA. Nature Structural & Molecular Biology 13, 1092-1096. Semenkov, Y.P., Rodnina, M.V., and Wintermeyer, W. (2000). Energetic contribution of tRNA hybrid state formation to translocation catalysis on the ribosome. Nat Struct Biol 7, 1027-1031. Shibuya, N., Nishiyama, T., Kanamori, Y., Saito, H., and Nakashima, N. (2003). Conditional rather than absolute requirements of the capsid coding sequence for initiation of methionine-independent translation in Plautia stali intestine virus. Journal of Virology 77, 12002-12010. Skabkin, M.A., Skabkina, O.V., Dhote, V., Komar, A.A., Hellen, C.U., and Pestova, T.V. (2010). Activities of Ligatin and MCT-1/DENR in eukaryotic translation initiation and ribosomal recycling. Genes Dev 24, 1787-1801. Snijder, E.J., den Boon, J.A., Bredenbeek, P.J., Horzinek, M.C., Rijnbrand, R., and Spaan, W.J. (1990). The carboxyl-terminal part of the putative Berne virus polymerase is expressed by ribosomal frameshifting and contains sequence motifs which indicate that toro- and coronaviruses are evolutionarily related. Nucleic Acids Res 18, 4535-4542. Spahn, C.M., Jan, E., Mulder, A., Grassucci, R.A., Sarnow, P., and Frank, J. (2004). Cryo-EM visualization of a viral internal ribosome entry site bound to human ribosomes: the IRES functions as an RNA-based translation factor. Cell 118, 465-475. Stein, E.A., Pankiewicz, J., Harsch, H.H., Cho, J.K., Fuller, S.A., Hoffmann, R.G., Hawkins, M., Rao, S.M., Bandettini, P.A., and Bloom, A.S. (1998). Nicotine-162  induced limbic cortical activation in the human brain: A functional MRI study. American Journal of Psychiatry 155, 1009-1015. Stoneley, M., Paulin, F.E., Le Quesne, J.P., Chappell, S.A., and Willis, A.E. (1998). C-Myc 5' untranslated region contains an internal ribosome entry segment. Oncogene 16, 423-428. Sundararajan, A., Michaud, W.A., Qian, Q., Stahl, G., and Farabaugh, P.J. (1999) Near-cognate peptidyl-tRNAs promote +1 programmed translational frameshifting in yeast. Mol Cell 4(6), 1005-15. Taylor, D.J., Nilsson, J., Merrill, A.R., Andersen, G.R., Nissen, P., and Frank, J. (2007). Structures of modified eEF2 80S ribosome complexes reveal the role of GTP hydrolysis in translocation. EMBO J 26, 2421-2431. Terenin, I.M., Dmitriev, S.E., Andreev, D.E., Royall, E., Belsham, G.J., Roberts, L.O., and Shatsky, I.N. (2005). A cross-kingdom internal ribosome entry site reveals a simplified mode of internal ribosome entry. Molecular and Cellular Biology 25, 7879-7888. Terenin, I.M., Dmitriev, S.E., Andreev, D.E., and Shatsky, I.N. (2008). Eukaryotic translation initiation machinery can operate in a bacterial-like mode without eIF2. Nat Struct Mol Biol 15, 836-841. Thompson, R.C., and Stone, P.J. (1977). Proofreading of the codon-anticodon interaction on ribosomes. Proc Natl Acad Sci U S A 74, 198-202. Tsukiyama-Kohara, K., Iizuka, N., Kohara, M., and Nomoto, A. (1992). Internal ribosome entry site within hepatitis C virus RNA. J Virol 66, 1476-1483. Unbehaun, A., Borukhov, S.I., Hellen, C.U.T., and Pestova, T.V. (2004). Release of initiation factors from 48S complexes during ribosomal subunit joining and the link between establishment of codon-anticodon base-pairing and hydrolysis of eIF2-bound GTP. Genes & Development 18, 3078-3093. Vagner, S., Gensac, M.C., Maret, A., Bayard, F., Amalric, F., Prats, H., and Prats, A.C. (1995). Alternative translation of human fibroblast growth factor 2 mRNA occurs by internal entry of ribosomes. Mol Cell Biol 15, 35-44. 163  Van Eden, M.E., Byrd, M.P., Sherrill, K.W., and Lloyd, R.E. (2004). Demonstrating internal ribosome entry sites in eukaryotic mRNAs using stringent RNA test procedures. Rna-a Publication of the Rna Society 10, 720-730. van Rij, R.P., Saleh, M.C., Berry, B., Foo, C., Houk, A., Antoniewski, C., and Andino, R. (2006). The RNA silencing endonuclease Argonaute 2 mediates specific antiviral immunity in Drosophila melanogaster. Genes Dev 20, 2985-2995. Varaklioti, A., Vassilaki, N., Georgopoulou, U., and Mavromara, P. (2002). Alternate translation occurs within the core coding region of the hepatitis C viral genome. Journal of Biological Chemistry 277, 17713-17721. Vassilaki, N., Boleti, H., and Mavromara, P. (2007). Expression studies of the core+1 protein of the hepatitis C virus 1a in mammalian cells. The influence of the core protein and proteasomes on the intracellular levels of core+1. FEBS J 274, 4057-4074. Vassilaki, N., Boleti, H., and Mavromara, P. (2008a). Expression studies of the HCV-1a core+1 open reading frame in mammalian cells. Virus Res 133, 123-135. Vassilaki, N., Friebe, P., Meuleman, P., Kallis, S., Kaul, A., Paranhos-Baccala, G., Leroux-Roels, G., Mavromara, P., and Bartenschlager, R. (2008b). Role of the hepatitis C virus core+1 open reading frame and core cis-acting RNA elements in viral RNA translation and replication. J Virol 82, 11503-11515. Vassilaki, N., and Mavromara, P. (2003). Two alternative translation mechanisms are responsible for the expression of the HCV ARFP/F/core+1 coding open reading frame. Journal of Biological Chemistry 278, 40503-40513. Vassilaki, N., and Mavromara, P. (2009). The HCV ARFP/F/core+1 protein: production and functional analysis of an unconventional viral product. IUBMB Life 61, 739-752. Vimaladithan, A., and Farabaugh, P.J. (1994). Special peptidyl-tRNA molecules can promote translational frameshifting without slippage. Mol Cell Biol 14, 8107-8116. Walewski, J.L., Keller, T.R., Stump, D.D., and Branch, A.D. (2001). Evidence for a new hepatitis C virus antigen encoded in an overlapping reading frame. Rna-a Publication of the Rna Society 7, 710-721. 164  Wang, Q.S., Au, H.H.T., and Jan, E. (2013). Methods for studying IRES-mediated translation of positive-strand RNA viruses. Methods 59, 167-179. Wang, X.H., Aliyari, R., Li, W.X., Li, H.W., Kim, K., Carthew, R., Atkinson, P., and Ding, S.W. (2006). RNA interference directs innate immunity against viruses in adult Drosophila. Science 312, 452-454. Weiss, R.B., Dunn, D.M., Shuh, M., Atkins, J.F., and Gesteland, R.F. (1989). E. coli ribosomes re-phase on retroviral frameshift signals at rates ranging from 2 to 50 percent. New Biol 1, 159-169. Wilkinson, K.A., Merino, E.J., and Weeks, K.M. (2006). Selective 2 '-hydroxyl acylation analyzed by primer extension (SHAPE): quantitative RNA structure analysis at single nucleotide resolution. Nature Protocols 1, 1610-1616. Wilson, J.E., Pestova, T.V., Hellen, C.U., and Sarnow, P. (2000a). Initiation of protein synthesis from the A site of the ribosome. Cell 102, 511-520. Wilson, J.E., Powell, M.J., Hoover, S.E., and Sarnow, P. (2000b). Naturally occurring dicistronic cricket paralysis virus RNA is regulated by two internal ribosome entry sites. Mol Cell Biol 20, 4990-4999. Woolaway, K.E., Lazaridis, K., Belsham, G.J., Carter, M.J., and Roberts, L.O. (2001). The 5 ' untranslated region of Rhopalosiphum padi virus contains an internal ribosome entry site which functions efficiently in mammalian, plant, and insect translation systems. J Virol 75, 10244-10249. Xu, Z., Choi, J., Lu, W., and Ou, J.H. (2003). Hepatitis C virus f protein is a short-lived protein associated with the endoplasmic reticulum. J Virol 77, 1578-1583. Xu, Z., Choi, J., Yen, T.S., Lu, W., Strohecker, A., Govindarajan, S., Chien, D., Selby, M.J., and Ou, J. (2001). Synthesis of a novel hepatitis C virus protein by ribosomal frameshift. EMBO J 20, 3840-3848. Yang, B., Song, X.L., Huang, J., Shi, C.Y., Liu, Q.H., and Liu, L. (2006). A single-step multiplex PCR for simultaneous detection of white spot syndrome virus and infectious hypodermal and haematopoietic necrosis virus in penaeid shrimp. J Fish Dis 29, 301-305. 165  Zhu, J., Korostelev, A., Costantino, D.A., Donohue, J.P., Noller, H.F., and Kieft, J.S. (2011). Crystal structures of complexes containing domains from two viral internal ribosome entry site (IRES) RNAs bound to the 70S ribosome. Proc Natl Acad Sci U S A 108, 1839-1844. 166  Appendices Appendix A  Methods A.1 DNA constructs, cell culture and transfection  The coding sequences of ORFx in frame with eGFP or 3X HA tag at the N- or C- terminus were amplified by PCR reaction and ligated into pAc5.1 vector, containing a Drosophila actin promoter. The coding sequence of FFL was amplified by PCR reaction and ligated into pAc5.1 vector to generate pAc5.1-FFL plasmid. All the constructs were confirmed by sequencing. Drosophila Schneider line 2 (S2) cells were passaged in schineider's insect medium supplemented with 10% fetal bovine serum. Transfections were performed with Lipofectamine 2000 (Invitrogen) using 2 µg of plasmid DNA into S2 cells according to the manufacturer’s directions.  A.2 Translation assay  For translation assay, the tagged ORFx plasmid is cotransfected with a luciferase reporter expression plasmid, pAc5.1-FFL at indicated molar ratio. The FFL expression level was determined at indicated time points after cotransfection by using an enzymetic luciferase measuring kit (Promega).   A.3 Western Blot analysis S2 cells were lysed in lysis buffer (20 mM HEPES, 150 mM NaCl, 1% Triton X-100, 10% glycerol, 1 mM EDTA, 10 mM tetrapyrophosphate, 100 mM NaF, 17.5 mM β-167  glycerophosphate, and a protease inhibitor). Equal amounts of lysates were loaded and separated on a SDS-PAGE gel and then transferred to a polyvinylidene difluoride Immobilon-FL membrane (Millipore). Blots were blocked for 1 h in 5% skim milk and Tris buffered saline (20 mM Tris, 150 mM NaCl, 0.1% Tween-20). Antibodies used were as follows: affinity-purified IAPV ORFx peptide rabbit polyclonal antibody (raised against the IAPV ORFx peptide sequence C-SLRPLVKTRLR PKKC-NH2; GenScript) (1∶500), affinity-purified IAPV VP2 peptide rabbit polyclonal antibody (raised against the IAPV VP2 peptide sequence C-NTMPGDSQQESNTPC-NH2; GenScript) (1∶500), anti-Luciferase (Promega) (1∶1,000), anti-3 X HA (1∶1,000) and anti-eGFP (Promega) (1∶1,000). Primary antibodies were detected using either IRDye 800CW goat anti-rabbit IgG (LI-COR Biosciences) at 1∶10,000 at room temperature for 1 h, and subsequently scanned using an Odyssey imager (LI-COR Biosciences).  A.4 Mass spectrometry analysis All peptide samples were analyzed on a triple quadrupole mass spectrometer (6460; Agilent Technologies) equipped with an HPLC-Chip Cube system (Agilent Technologies) operated in selected ion and product ion mode [generation of the multiple reaction monitoring (MRM) assay] or MRM mode (honey bee samples). In all experiments, Q1 and/or Q3 operated at unit resolution (0.7 m/z full width at half height). For the generation of the MRM assay, a ProtID-Chip-43 (II) was used containing a 40-nL trap column and a 43-mm analytical column (G4240-62005; Agilent Technologies) and the trapping of the peptides was performed at 4 µL/ min using solvent A (3% acetonitrile in 0.1% formic acid) 168  only. Analysis was performed using a flow rate of 600 nL/ min with a gradient of 10–40% solvent B (90% acetonitrile in 0.1% formic acid) in 2.5 min. Honey bee samples were eluted from stop and go extraction tips with 20 µL of 80% acetonitrile in 0.5% acetic acid, vacuum-dried, and reconstituted in sample buffer (1% trifluoroacetic acid and 3% acetonitrile in 0.5% acetic acid) and were injected (4 µL/ min, solvent A) onto a large capacity chip (II) consisting of a 160nL trap column and 150-mm analytical column (G4240-62010; Agilent Technologies). Chromatographic separations were performed using a flow rate of 300 nL/ min with a gradient of 5–20% solvent B in 20 min and 20–35% in 14 min. Each transition was monitored for 333 ms leading to a total duty cycle of 1 s. Validation of the peptide identity was performed by comparing the relative peak area of each of the three transitions to the relative area of the transitions of the synthetic peptide.  A.5 MRM assay The synthesized peptide was diluted 100-fold by using 0.5% acetic acid, and multiple injections of 1 μL each were used to optimize the fragmentor voltage (0–180 V, 20-V increments) and subsequently the collision energy (5–25 V, 5-V increments; 13–18 V, 1-V increments). Fragment ions with a mass-to-charge ratio (m/ z) greater than the precursor ion m/ z were prioritized, leading to three transitions that were selected for the MRM assay. A fragmentor voltage of 40 V was selected for all transitions, and a collision energy of 15 V was selected for transitions 640.6 →1038.6 and 640.6 → 668.4 and of 14 V was selected for transition 640.6 → 925.5.  169   Appendix B  Sequences of plasmid and primers B.1 Sequence inserted in the IGR region of pIAPV-WT The following sequence of IAPV IGR IRES and the downstream coding sequence of ORF2 (nucleotides 6399–6908) were inserted into the IGR region of a bicistronic luciferase plasmid, pIAPV-WT.  5’- GAGCGGTTTCTGGAATACTATATGTAAGTATAGTGTTCTGGAGGCATCATTC TATGGTTACCCATCATTAGAGGAAATTTCCAATAAACTCTGGTGTAAGGCTTAGAGTGATGGTCGAGGTGCCCTATTTAGGGTGAGGAGCCTCGGTGGCAGCCCCACCAAATCCTCTATTGGATAGGAACAGCTGTACTGGGCAGTTACAGCAGTCGTATGGTAACACATGCGGCGTTCCGAAATACCATGCCTGGCGATTCACAACAAGAAAGCAATACTCCCAACGTACACAATACGGAACTCGCTTCGTCCACTAGTGAAAACTCGGTTGAGACCCAAGAAATCACAACCTTTCATGATGTGGAAACTCCAAATAGGATCGATACCCCCATGGCTCAGGATACTTCATCGGCTAGGAACATGGATGATACGCACAGTATTATTCAGTTTTTACAGCGCCCCGTTCTCATTGACAACATTGAGATCATTGCTGGAACAACGGCCGATGCAAACAAACCCCTTAGCCGATATGTGTTATATCAAGGAAGACGCCAAAAACATACAGAAAGGCCCGGCGCCATTCTATCCGCTGGAAGATGGAACCGCTGGAGAGCAACTGCATAA - 3’    170  B.2 List of primers used for site-directed mutagenesis Pr C6615G/C6616G: TGCGGCGTTCCGAAATACCATGGGTGGCGATTCACAACAAGAAAGCAA Pr G6564C/G6565C: TCTATTGGATAGGAACAGCTGTACTGCCCAGTTACAGCAGTCGTATGGT Pr G6618T/G6619A: GGCGTTCCGAAATACCATGCCTTAGGATTCACAACAAGAAAGCAATACTCC Pr G6621T: CGTTCCGAAATACCATGCCTGGCTATTCACAACAAGAAAGCAATACTCC Pr G6619T/C6620A: GGCGTTCCGAAATACCATGCCTGTAGATTCACAACAAGAAAGCAATACTCC Pr A6622T/T6623A/T6624G: GGCGTTCCGAAATACCATGCCTGGCGTAGCACAACAAGAAAGCAATACTCC Pr G6594A: TACAGCAGTCGTATGGTAACACATACGGCGTTCCGAAATACCATGCCTG Pr G6618A: ATGCGGCGTTCCGAAATACCATGCCTAGCGATTCACAACAAGAAAGCAA Pr G6618C: ATGCGGCGTTCCGAAATACCATGCCTCGCGATTCACAACAAGAAAGCAA Pr G6618T: ATGCGGCGTTCCGAAATACCATGCCTTGCGATTCACAACAAGAAAGCAA Pr T6562G: TCTATTGGATAGGAACAGCTGTACGGGGCAGTTACAGCAGTCGTATGGT 171  Pr T6562A: TCTATTGGATAGGAACAGCTGTACAGGGCAGTTACAGCAGTCGTATGGT Pr T6562C: TCTATTGGATAGGAACAGCTGTACCGGGCAGTTACAGCAGTCGTATGGT Pr C6620G: CGGCGTTCCGAAATACCATGCCTGGGGATTCACAACAAGAAAGCAA Pr T6613A: ATGCGGCGTTCCGAAATACCAAGCCTGGCGATTCACAACAAGAAAGCAA Pr T6613A/G6614C: ATGCGGCGTTCCGAAATACCAACCCTGGCGATTCACAACAAGAAAGCAA Pr G6563A: TCTATTGGATAGGAACAGCTGTACTAGGCAGTTACAGCAGTCGTATGGT Pr G6563C: TCTATTGGATAGGAACAGCTGTACTCGGCAGTTACAGCAGTCGTATGGT Pr G6563T: TCTATTGGATAGGAACAGCTGTACTTGGCAGTTACAGCAGTCGTATGGT Pr T6617A: ATGCGGCGTTCCGAAATACCATGCCAGGCGATTCACAACAAGAAAGCAA Pr T6617C: ATGCGGCGTTCCGAAATACCATGCCCGGCGATTCACAACAAGAAAGCAA Pr T6617G: ATGCGGCGTTCCGAAATACCATGCCGGGCGATTCACAACAAGAAAGCAA Pr G6568C: ATAGGAACAGCTGTACTGGGCACTTACAGCAGTCGTATGGTAACA 172  Pr G6568A:  ATAGGAACAGCTGTACTGGGCAATTACAGCAGTCGTATGGTAACA Pr G6568C:  ATAGGAACAGCTGTACTGGGCACTTACAGCAGTCGTATGGTAACA Pr ∆G6568:  ATAGGAACAGCTGTACTGGGCATTACAGCAGTCGTATGGTAACA Pr U6569A:  ATAGGAACAGCTGTACTGGGCAGATACAGCAGTCGTATGGTAACA Pr U6569C:  ATAGGAACAGCTGTACTGGGCAGCTACAGCAGTCGTATGGTAACA Pr U6569G:  ATAGGAACAGCTGTACTGGGCAGGTACAGCAGTCGTATGGTAACA Pr ∆ U6569:  ATAGGAACAGCTGTACTGGGCAGTACAGCAGTCGTATGGTAACA Pr C6561A: TCTATTGGATAGGAACAGCTGTAATGGGCAGTTACAGCAGTCGTATGGT Pr C6561G: TCTATTGGATAGGAACAGCTGTAGTGGGCAGTTACAGCAGTCGTATGGT Pr C6561T: TCTATTGGATAGGAACAGCTGTATTGGGCAGTTACAGCAGTCGTATGGT Pr A6554C: AAATCCTCTATTGGATAGGAACCGCTGTAATGGGCAGTTACAGCAGT 173  Pr A6554G: AAATCCTCTATTGGATAGGAACGGCTGTAATGGGCAGTTACAGCAGT Pr A6554T: AAATCCTCTATTGGATAGGAACTGCTGTAATGGGCAGTTACAGCAGT Pr ∆A6554:  AAATCCTCTATTGGATAGGAACGCTGTAATGGGCAGTTACAGCAGT ABPV  Pr ∆U6490:  GAACAGCTATATTGGGTAG TGTAGCAGTTGTATTCAAATG KBV  Pr ∆C6581:  GAAACCGCTATATCGGGTAG TATAGCAGTCGGATAGTAATATATCC Pr ∆U6582:  GAAACCGCTATATCGGGTAGC ATAGCAGTCGGATAGTAATATATCC SINV-1: Pr ∆C4376:  GGAACAGCTATATCGGGTTG TATAGCAGTCAGGATGTCATTCTG Pr ∆U4377:  GGAACAGCTATATCGGGTTGC ATAGCAGTCAGGATGTCATTCTG    174  Appendix C  SHAPE analysis of wild-type and mutant IAPV IGR IRESs  Figure C.1 Optimization of NMIA modification in SHAPE analysis of wide-type IAPV IGR IRES. (Left) An example gel of IAPV IGR IRES PKI domain modified with NMIA. The final concentration of NMIA is shown. Sequencing from reactions containing dideoxy nucleotides are shown in the left lanes with their respective nucleotides. The sequencing gel was dried and analyzed by phosphorimager analysis. (Right) Secondary structure model of IAPV IGR IRES. 175    Figure C.2 SHAPE analysis of wild-type and mutant IAPV IGR IRESs. 176  Legend to Figure C.2. Representative sequencing gels, which focus on the PKI domain of the IAPV IGR IRES, are shown. Sequencing from reactions containing dideoxy nucleotides are shown in the left lanes with their respective nucleotides. The sequencing gel was dried and analyzed by phosphorimager analysis.   Appendix D  Prediction of the structure and function of ORFx Phyre is used for predicting the three-dimensional (3D) structure of ORFx. It is based on the observation that the number of protein folds in nature appears to be limited and that many different remotely homologous protein sequences adopt remarkably similar structures (Kelley and Sternberg, 2009). If a homolog of known 3D structure can be found, an alignment of the two sequences can be generated and used directly to build a 3D model of the sequence of interest.    Figure D.1 Secondary structure and disorder prediction of ORFx. The amino acid sequence of ORFx is shown on the first line. The predicted secondary structure of ORFx is shown underneath. The green spiral represents the alpha helix. The blue arrow represents the beta strand. Question mark represents disorder. The confidence 177  key is shown in a rainbow color coded manner under indivisual prediction.  The warmer color means higher confidence score and the colder color means lower confidence score.     Figure D.2 The structure prediction of IAPV ORFx. (Left) The predicted structure of IAPV ORFx. The aligned residue 34-94 of ORFx (63% coverage) is shown in a rainbow color coded manner. The red represents the N-terminal and the purper represents the C-terminal of ORFx. (Right) The structure of the highest score target of ORFx homolog, endo-beta-n- acetylglucosaminidase, which is an enzyme involved in processing of free oligosaccharides in the cytosol. The endo-beta-n- acetylglucosaminidase and ORFx are with confidence of 13.1% and 25% i.d.   178  Appendix E  Overexpression of ORFx has little effect on cotransfected reporter translation or host translation As a viral protein, ORFx might be involved in the process of modulating the host or/and viral translation or/and transcription. To test the hypothesis that the expression of ORFx has an effect on host cell translation, I first performed a cotransfection assay. In this assay, the tagged ORFx plasmid is cotransfected with a luciferase reporter expression plasmid, pAc5.1-FFL at a 1:1 and 3:1 molar ratio. Cells were harvested at 12, 24, 48 and 72 h post transfection and firefly luciferase activities were monitored using an enzymetic luciferase measuring kit (Promega).  As a control, the firefly luciferase plasmids were transfected alone. Firefly luciferase activities increased over time (Figure E.1 a and b). However, there was no dramatic expression difference between the control and cotransfection with ORFx at the different molar ratios (1:1 and 3:1) (Figure E.1 a and b). These results suggest that overexpression of the HA tagged IAPV ORFx has little effect on the cotransfected reporter FFL gene expression level in Drosophila S2 cells. 179       Figure E.1 Cotransfection assays of ORFx with a reporter FFL. (a) The tagged ORFx plasmid is cotransfected with a luciferase reporter expression plasmid, pAc5.1-FFL at 1:1 molar ratio. (b) The tagged ORFx plasmid is cotransfected with pAc5.1-FFL at 3:1 molar ratio (ORFx : FFL). The FFL expression level was determined at indicated time points, 12, 24, 48 and 72 h after cotransfection by using an 180  enzymetic luciferase measuring kit (Promega). (c) Western blot analysis confirms the expression of HA-tagged-ORFx at indicated time points.    Figure E.2 The effect of ORFx on cell translation by pulse labelling. (a) A representative gel of the S35- Methionine pulse labelling of tagged-ORFx transfected cells at indicated time points. Quantitation of the intensities of radioactive bands within each lane were shown underneath. (b) Western blot analysis confirms the expression of HA-tagged-ORFx at indicated time points.   181   Appendix F  Positioning of the 80S ribosomes on the mutant IRESs    Figure F.1 Toeprinting analysis of 80S ribosomes on mutant IAPV IGR IRESs. (Experiment performed by Hilda Au) 40S and 60S subunits were incubated with dicistronic RNAs containing wild-type or mutant IAPV IGR IRES and analyzed by primer extension analysis. Reaction products were separated by denaturing polyacrylamide gel. The gels were dried and exposed by autoradiography.   182   Appendix G  Translational activities of mutant IRESs     Figure G.1 Translational activities of mutant IAPV IGR IRESs. (Experiment performed by Seonghoon Lee) Translational activities of mutants in PKI domain of IAPV IGR IRES in Sf21 extracts. Dicistronic reporter constructs were incubated at 30 °C for 3h in the presence of [35S]-methionine. Quantitation of radiolabeled protein products calculated by taking the ratio FLuc/RLuc and normalized by the containing numbers of methionine. The percentage of 0 and +1 frame translation of IRES mutants normalized to those of the wild-type IRES (given as both frame as 100%). Data are presented as mean ± SD of three independent experiments.   

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0167579/manifest

Comment

Related Items