UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Characterization of the Dicistroviridae intergenic region internal ribosome entry site Jang, Christopher 2011

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

Item Metadata

Download

Media
24-ubc_2011_fall_jang_christopher.pdf [ 4.85MB ]
Metadata
JSON: 24-1.0071870.json
JSON-LD: 24-1.0071870-ld.json
RDF/XML (Pretty): 24-1.0071870-rdf.xml
RDF/JSON: 24-1.0071870-rdf.json
Turtle: 24-1.0071870-turtle.txt
N-Triples: 24-1.0071870-rdf-ntriples.txt
Original Record: 24-1.0071870-source.json
Full Text
24-1.0071870-fulltext.txt
Citation
24-1.0071870.ris

Full Text

CHARACTERIZATION OF THE DICISTROVIRIDAE INTERGENIC REGION INTERNAL RIBOSOME ENTRY SITE by Christopher Jang B.Sc., Queen’s University, 2005 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Biochemistry & Molecular Biology) THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  June 2011 © Christopher Jang, 2011  Abstract The IRES found in the intergenic (IGR) region of viruses belonging to the Dicistroviridae family is remarkable for its ability to bind directly to the ribosome with high affinity and initiate translation without the requirement for any initiation factors by mimicking a P/E hybrid tRNA. Here, we have conducted an in-depth biochemical characterization of the CrPV IGR IRES. We have found that the L1.1 region of the IRES is responsible for 80S assembly and reading frame maintenance, and may play an additional role downstream of ribosome binding. Additional studies on the modularity of the IRES showed that the two domains of the IGR IRES work independently, but in concert with one another to manipulate the ribosome. We then addressed the question of how the IGR IRES recruits ribosomes during periods of cellular stress, when inactive 80S couples accumulate in the cell. Here, we found that the IRES is able to bind directly to eEF2-associated 80S couples, providing a rationale as to how the IRES remains translated during these periods. Finally, we developed a new in vitro translation system to assess the functionality of specialized ribosomes, and used this system and the IGR IRES in order to ask questions about the pathology of dyskeratosis congenita. Though divergent from other viral IRESs, the simplicity of this tRNA-like IRES serves as a powerful model for understanding IRES functions in general, the role of tRNA/ribosome interactions that occur normally during translation, and how these processes are linked to the greater context of the cell.  ii  Preface A version of chapter 2 has been published in the Journal of Molecular Biology. I conducted the majority of the experimental work, and wrote the manuscript with guidance from Dr. Eric Jan. Some of the activity assays were completed by Miranda C. Lo. The citation is as follows: Jang, C. J., Lo, M. C. and E. Jan. (2009). Conserved element of the dicistrovirus IGR IRES that mimics an E-site tRNA/ribosome interaction mediates multiple functions. Journal of Molecular Biology. 387, 42-58. A version of chapter 3 has also been published in RNA. All experimental work was completed by myself. I wrote the manuscript with guidance from Dr. Eric Jan. The citation is as follows: Jang, C. J. and E. Jan. (2010). Modular domains of the Dicistroviridae intergenic internal ribosome entry site. RNA. 16, 1182-1195. A version of chapter 4 is currently in preparation for publication alongside experimental work completed by Dr. Natalia Demeshkina, Dr. Shinya Kurata, and Dr. Go Hirokawa, in collaboration with Dr. Akira Kaji at the University of Pennsylvania. Figure 4.1 shows experimental work completed by Dr. Natalia Demeshkina, and has been used with permission by Dr. Akira Kaji. Human DC-positive cells that were used in the course of the experiments described in chapter 5 were obtained from Xi-Lei Zeng, in collaboration with Dr. Judy Wong from UBC Pharmaceutical Sciences. These cells were derived from participants in a National Cancer Institute Institutional Review Board approved study. The National Institutes of Health protocol number is 02-C-0052, and the clinical trials number is NCT00056121. Copyrighted materials presented herein have been reproduced or adapted with permission, as indicated. iii  Table of Contents Abstract.................................................................................................................................... ii Preface..................................................................................................................................... iii Table of Contents ................................................................................................................... iv List of Tables ......................................................................................................................... vii List of Figures....................................................................................................................... viii List of Abbreviations .............................................................................................................. x Acknowledgements ............................................................................................................... xii Dedication ............................................................................................................................. xiv 1 Introduction.......................................................................................................................... 1 1.1  Eukaryotic cap-dependent translation initiation.................................................................... 1  1.2  Translation elongation in eukaryotes .................................................................................... 4  1.3  Cap-independent translation initiation by internal ribosome entry ....................................... 6  1.3.1  Picornaviridae IRESs..................................................................................................... 11  1.3.2  Flaviviridae IRESs ......................................................................................................... 16  1.3.3  Group I IRESs ................................................................................................................ 20  1.3.4  Cellular IRESs ................................................................................................................ 20  1.4  The Dicistroviridae viral family ......................................................................................... 22  1.5  The Dicistroviridae intergenic region internal ribosome entry site .................................... 27  2 Characterization of the conserved L1.1 region of the IGR IRES ................................. 36 2.1  Summary ............................................................................................................................. 36  2.2  Introduction......................................................................................................................... 37  2.3  Materials and methods ........................................................................................................ 40  2.3.1  DNA constructs and reagents ......................................................................................... 40  2.3.2  In vitro transcription and translation............................................................................... 40  2.3.3  40S and 60S subunit purification.................................................................................... 40  2.3.4  Non-denaturing gel mobility shift assays ....................................................................... 41  2.3.5  40S and 80S competition assays..................................................................................... 42  2.3.6  RNase T1 enzymatic probing ......................................................................................... 43  iv  2.3.7  Toeprint analysis of ribosomal complexes using purified components.......................... 44  2.3.8  Ribosome assembly in rabbit reticulocyte lysates .......................................................... 44  2.3.9  Ribosome assembly using purified components............................................................. 45  2.3.10 2.4  GTPase assays ............................................................................................................ 45  Results................................................................................................................................. 46  2.4.1  Association of human 80S ribosomes with the CrPV IGR IRES ................................... 46  2.4.2  The conserved loop region of the CrPV IGR IRES is essential for IRES activity ......... 51  2.4.3  Mutations within L1.1 do not affect the secondary structure of the IGR IRES.............. 53  2.4.4  Function of mutations within L1.1 in 40S subunit recruitment ...................................... 55  2.4.5  Function of mutations within L1.1 in 80S subunit recruitment ...................................... 58  2.4.6  Positioning of the 40S and 80S ribosomes on the L1.1 mutant IRES ............................ 61  2.4.7  Assembly of 80S on the IGR IRES in RRL.................................................................... 66  2.4.8  Ribosomal positioning does not dictate 80S assembly on the IGR IRES....................... 69  2.4.9  L1.1 of the TSV and IAPV IGR IRES also mediates 80S ribosome assembly .............. 70  2.4.10  Strengthening of PKI compensates for L1.1 mediated translational defects.............. 72  2.4.11  L1.1 of the IGR IRES affects factor retention on the 40S subunit............................. 78  2.5  Discussion ........................................................................................................................... 83  3 Modularity of the Dicistroviridae intergenic internal ribosome entry site ................... 89 3.1  Summary ............................................................................................................................. 89  3.2  Introduction......................................................................................................................... 90  3.3  Materials and methods ........................................................................................................ 92  3.3.1  Purification of eukaryotic elongation factor 2 ................................................................ 92  3.3.2  80S filter binding competition assays............................................................................. 92  3.3.3  Toeprinting analysis of ribosomal complexes in RRL ................................................... 93  3.4  Results................................................................................................................................. 95  3.4.1  80S ribosome assembly on the chimeric IGR IRES ....................................................... 95  3.4.2  PKI of the chimeric IGR IRESs folds independently ..................................................... 99  3.4.3  Translational activity of the chimeric IRES ................................................................. 101  3.4.4  Positioning of the IGR IRES chimeras correlates with their translational activities .... 104  3.4.5  Modularity of the conserved L1.1A/B region of the IGR IRESs ................................. 108  3.4.6  The IGR IRES stimulates the ribosome-dependent GTPase activity of eEF2.............. 110  3.4.7  Deletion of SL III in the TSV IGR IRES abolishes 80S positioning............................ 113  3.5  Discussion ......................................................................................................................... 115  v  4 Novel pathways of 80S assembly on an IRES mediated by eukaryotic elongation factor 2 ................................................................................................................................ 120 4.1  Summary ........................................................................................................................... 120  4.2  Introduction....................................................................................................................... 121  4.3  Materials and methods ...................................................................................................... 125  4.3.1 4.4  Assembly and analysis of ribosomal complexes .......................................................... 125 Results............................................................................................................................... 126  4.4.1  eEF2 induces the association and stabilization of 80S ribosomes................................ 126  4.4.2  The IGR IRES is bound to 60S subunits following sucrose gradient centrifugation ... 128  4.4.3  80S ribosomes that are associated by eEF2 can bind to the IGR IRES ........................ 132  4.4.4  The IGR IRES itself induces 80S association .............................................................. 136  4.5  Discussion ......................................................................................................................... 138  5 The role of specialized ribosomes in human dyskeratosis congenita cells in IGR IRESmediated translation ........................................................................................................... 142 5.1  Summary ........................................................................................................................... 142  5.2  Introduction....................................................................................................................... 143  5.3  Materials and methods ...................................................................................................... 150  5.3.1  40S and 60S subunit purification from dyskerin mutant cells...................................... 150  5.3.2  Depletion and add back of ribosomes in rabbit reticulocyte lysates............................. 150  5.4  Results............................................................................................................................... 152  5.4.1  Ribosome depletion and add-back in rabbit reticulocyte lysates.................................. 152  5.4.2  The IGR IRES is active with ribosomes purified from mutant dyskerin cells ............. 157  5.5  Discussion ......................................................................................................................... 159  6 Conclusion ........................................................................................................................ 163 References............................................................................................................................ 167  vi  List of Tables Table 1.1. Dicistroviruses and their natural hosts................................................................... 26 Table 2.1. Summary of properties of L1.1 mutant CrPV IGR IRES ...................................... 50 Table 3.1. Properties of mutant IGR IRESs............................................................................ 98  vii  List of Figures Figure 1.1. A model of eukaryotic cap-dependent translation initiation .................................. 3 Figure 1.2. The eukaryotic translation elongation cycle........................................................... 5 Figure 1.3. A classification scheme of viral IRESs ................................................................ 10 Figure 1.4. The secondary structure of picornavirus IRESs ................................................... 15 Figure 1.5. The structure of the HCV IRES............................................................................ 19 Figure 1.6. The genomic architecture of a dicistrovirus ......................................................... 25 Figure 1.7. The secondary structure of the IGR IRES............................................................ 32 Figure 1.8. The crystal structure of the type I IGR IRES at 3.1 Å resolution ........................ 33 Figure 1.9. Mimicry of a tRNA by PKI of the IGR IRES ...................................................... 34 Figure 1.10. IRES-induced conformational changes in the human 80S ribosome ................. 35 Figure 2.1. Gelshifts to assess 80S assembly on the CrPV IGR IRES ................................... 48 Figure 2.2. 80S ribosome assembly on the IGR IRES............................................................ 49 Figure 2.3. Translational activities of mutant L1.1 IGR IRESs.............................................. 52 Figure 2.4. Enzymatic probing of the mutant L1.1 IGR IRESs.............................................. 54 Figure 2.5. Affinity measurements of 40S-IGR IRES complexes.......................................... 57 Figure 2.6. 80S assembly on mutant L1.1 IGR IRESs ........................................................... 60 Figure 2.7. Toeprint analysis of ribosomes on mutant L1.1 IGR IRESs ................................ 64 Figure 2.8. Ribosome translocation on the IGR IRES............................................................ 65 Figure 2.9. 80S assembly on mutant L1.1 IGR IRESs in RRL .............................................. 68 Figure 2.10. L1.1 of the TSV and IAPV IGR IRES mediates ribosome assembly ................ 71 Figure 2.11. Strengthening PKI results in a translationally stronger IRES ............................ 75 Figure 2.12. A strengthened PKI compensates for L1.1 mediated translational defects ........ 76 Figure 2.13. Strengthening PKI does not affect the GTPase activity of eEF2 ....................... 77 Figure 2.14. Addition of initiation factors results in a decrease in IRES-80S assembly........ 81 Figure 2.15. L1.1 mutants are defective in 80S assembly in the presence of eIF3................. 82  viii  Figure 3.1. 80S assembly of the chimeric IGR IRESs............................................................ 97 Figure 3.2. Enzymatic probing of the chimeric IGR IRESs ................................................. 100 Figure 3.3. Translational activities of the chimeric IGR IRESs ........................................... 103 Figure 3.4. Ribosome positioning on the chimeric IGR IRESs............................................ 107 Figure 3.5. The activity of L1.1 IGR IRES chimeras ........................................................... 109 Figure 3.6. GTPase stimulation of eEF2 by the chimeric IGR IRESs.................................. 112 Figure 3.7. Characterization of SL III within the TSV IGR IRES ....................................... 114 Figure 4.1. eEF2 induces the association of yeast 80S ribosomes........................................ 124 Figure 4.2. Human 80S ribosomes are associated by eEF2.................................................. 127 Figure 4.3. The IGR IRES is associated with 60S subunits after centrifugation.................. 130 Figure 4.4. The IGR IRES is bound to 60S ribosomal subunits ........................................... 131 Figure 4.5. Crosslinking does not affect the 80S assembly stimulated by eEF2 .................. 134 Figure 4.6. The IGR IRES binds to ribosomes associated by eEF2 ..................................... 135 Figure 4.7. The IRES itself induces 80S association in the presence of CTP ...................... 137 Figure 5.1. A schematic of the telomerase snoRNP complex............................................... 148 Figure 5.2. A schematic of what pseudouridylation is, and how it is generated .................. 149 Figure 5.3. A flowchart of the experimental procedure used ............................................... 154 Figure 5.4. A timecourse for ribosome depletion in rabbit reticulocyte lysates ................... 155 Figure 5.5. Optimization of ribosome add back in depleted rabbit reticulocyte lysates....... 156 Figure 5.6. Ribosomes from X-DC-positive cells can mediate IRES-driven translation ..... 158  ix  List of Abbreviations APS  ammonium persulfate  ATP  adenosine triphosphate  CrPV  cricket paralysis virus  CITE  cap-independent translation element  CSFV  classical swine fever virus  CTP  cytidine triphosphate  DC  dyskeratosis congenita  DMAPN  dimethylaminopropionitrile  DNA  deoxyribonucleic acid  EF  elongation factor  eEF  eukaryotic elongation factor  eIF  eukaryotic initiation factor  EMCV  encephalomyocarditis virus  FMDV  foot and mouth disease virus  GMP-PNP  guanosine 5′-[β,γ-imido]triphosphate  GTP  guanosine triphosphate  HAV  hepatitis A virus  HCV  hepatitis C virus  HIV  human immunodeficiency virus  hnRNP  heterogeneous nuclear ribonucleoprotein  HRV  human rhinovirus  IAPV  Israeli acute paralysis virus  IGR  intergenic region  IRES  internal ribosome entry site  ITAF  ires trans-acting factor  Kc  competitor dissociation constant  Kd  dissociation constant  Krel  relative dissociation constant  mRNA  messenger ribonucleic acid  PCBP  poly(rC)-binding protein x  PK  pseudoknot  PSIV  Plautia stali intestine virus  PTB  polypyrimidine tract-binding protein  RhPV  Rhopalosiphum padi virus  RNA  ribonucleic acid  PTV  porcine teschovirus  rp  ribosomal protein  RRF  ribosome release factor  RRL  rabbit reticulocyte lysate  SL  stem-loop  snoRNP  small nucleolar ribonucleoprotein  TERC  telomerase RNA component  TERT  telomerase reverse transcriptase  TMEV  Theiler's murine encephalomyelitis virus  tRNA  transfer ribonucleic acid  TSV  Taura syndrome virus  unr  upstream-of-N-ras protein  UTR  untranslated region  VPg  viral protein genome-linked  VSV  vesicular stomatitis virus  XIAP  X-linked inhibitor of apoptosis  xi  Acknowledgements I would first like to acknowledge my supervisor, Dr. Eric Jan, for the opportunity to work in his laboratory, and for the guidance that he has given me throughout my graduate career. I would also like to thank the past and present members of my supervisory committee for their encouragement and advice, which include Dr. Adam Frankel, Dr. George Mackie, Dr. Christopher Proud, and Dr. Filip Van Petegem. I would like to single out Dr. George Mackie in particular, who has been an amazing source of scientific and (interesting) historical knowledge. None of this work would have been possible without financial support, and I would like to thank the University of British Columbia, the Natural Sciences and Engineering Research Council of Canada, and the BC Proteomics Network for direct fellowship funding. I would also like to acknowledge the Canadian Institutes of Health Research, the Michael Smith Foundation for Health Research, and the Canada Foundation for Innovation for providing operational funds for the laboratory. I am also grateful to have met and worked with all of the members of the Jan Laboratory, both past and present. You all know who you are. The scientific discourse, interesting conversations, and emotional support that they have provided me with will forever be remembered and appreciated. I would also like to thank the love of my life, Katherine Yam. Thank you for dealing with me, and thank you for understanding everything. I love you.  xii  Lastly, I would like to mention my mother and father, who deserve special thanks. Over the course of my life, I’ve seen that my mother’s eloquence, patience, and understanding are infinite, and that my father’s integrity and honor are undying. I only dream to become a fraction of what you two epitomize. I am forever grateful for your unconditional love, hard work, and the immeasurable sacrifices that both of you have made throughout your lives for me and Angela. We both love you and hope we make you two happy. This is for you.  마지막으로 저의 어머니 그리고 아버지께 진심으로 감사드리고 싶습니다. 지금까지 살아오는 동안 저는 어머니의 무한한 현명하심, 인내하심과 이해하심을, 그리고 아버지의 의로우심과 강인함을 늘 지켜봐 왔습니다. 이 두 분이 살아오신 것에 절반이라도 따라가고 싶은 것이 제 바램입니다. 부모님께서 저와 은미에게 주신 무한한 사랑, 노력 그리고 희생을 저는 진심으로 감사드리고 언제나 기억할 것입니다. 저희 남매는 어머니 그리고 아버지를 정말 사랑하고, 두 분 보시기에 지금까지 자랑스러운 아들과 딸이 되었기를, 그리고 계속해서 되기를 바랍니다. 제가 지금까지 이뤄온 모든 것을 이 두 분께 바칩니다.  xiii  Dedication  For my mom and dad. 사랑하는 저의 어머니와 아버지께.  xiv  1 1.1  Introduction Eukaryotic cap-dependent translation initiation According to the central dogma of molecular biology, protein synthesis in living  organisms is the ultimate step of information transfer in life. Thus, this biological process is tightly controlled, and indeed, scientists have found that the initiation of protein synthesis is a complex, regulated process and is the target of numerous cellular signaling pathways. In the case of eukaryotes, the initiation of protein synthesis occurs through a stepwise, 5’ capdependent scanning mechanism (Figure 1.1) (Jackson et al., 2010). In this mechanism, we must first consider the cap-binding complex, eukaryotic initiation factor (eIF) 4F. This complex is comprised of three proteins: the cap binding protein eIF4E, the RNA helicase eIF4A, and the scaffold protein eIF4G. eIF4E promotes the attachment of this complex to the mRNA by direct binding to the 5`-7-methylguanosine cap of the transcript, and eIF4G serves as the point of recruitment for the ribosome, by binding the 43S preinitiation complex. The 43S preinitiation complex is composed of 40S ribosomal subunits that are bound to eIF3, eIF2-GTP-Met-tRNAiMet, eIF5, eIF1 and eIF1A. The interaction between eIF4G of the cap-binding complex and eIF3 mediates the recruitment of the 43S preinitiation complex to the 5’ end of the transcript. After recruitment of the 43S preinitiation complex to the transcript, the complex scans the 5’ untranslated region (UTR) for the initiating AUG codon. In order to facilitate this process, the eIF4A helicase unwinds local secondary structure in an ATP-dependent manner within the 5’ UTR, allowing the 43S complex to scan the transcript in search for the initiation codon. The factors eIF1 and eIF1A appear to assist in this process by inducing a scanning-competent “open” conformation in the ribosome, allowing the scanning process to  1  occur (Passmore et al., 2007). An AUG start codon is selected during scanning when the anticodon of initiator Met-tRNAiMet pairs with an AUG codon in the transcript. Once this recognition takes place, eIF1 is displaced from the P-site of the ribosome (Maag et al., 2005), which locks the ribosome into a “closed”, non-scanning conformation (Passmore et al., 2007). The ribosome further commits to the AUG initiation codon through the action of eIF5, which induces the hydrolysis of eIF2-bound GTP (Paulin et al., 2001). Hydrolysis of GTP leads to reduced affinity between Met-tRNAiMet and eIF2-GDP, resulting in the dissociation of eIF2-GDP from the positioned 40S subunit (Kapp and Lorsch, 2004; Pisarev et al., 2006). The subsequent dissociation of eIF1 and eIF5 on the preinitiation complex and joining of the 60S ribosomal subunit is faciliated by the action of eIF5B, a ribosome-dependent GTPase (Pestova et al., 2000; Unbehaun et al., 2004). The binding of eIF5B to the preinitiation complex is thought to bury solvent-exposed regions on the surface of the 40S subunit, which promotes 60S subunit joining (Allen et al., 2005). GTP bound to eIF5B is then hydrolyzed after 60S subunit joining, promoting the release of eIF5B-GDP from the complex along with eIF1A (Acker et al., 2006; Acker et al., 2009). What remains after this process is completed is an elongation-competent 80S ribosome, with the initiation codon and an initiator Met-tRNAiMet positioned within the P-site of the ribosome, ready to accept a new aminoacyl-tRNA in the A-site of the ribosome.  2  Figure 1.1. A model of eukaryotic cap-dependent translation initiation. The canonical pathway of eukaryotic translation initiation can be roughly divided into eight steps. These steps first start with the (1) formation of the eIF2-GTP-Met-tRNAiMet ternary complex, which (2) binds to the 40S subunit to form the 43S preinitiation complex with eIF3, eIF5, eIF1, and eIF1A. In parallel, the mRNA to be translated will (3) bind the cap-binding complex, eIF4F, which allows it to (4) recruit the 43S preinitiation complex. The 43S complex will then (5) scan the transcript for the initiation codon and upon initiation codon recognition, (6) eIF5mediated GTP hydrolysis will commit the ribosome to the start site. This is followed by the (7 and 8) displacement of initiation factors from the preinitiation complex and 60S subunit joining, mediated by eIF5B. Adapted with permission from Macmillan Publishers Limited, from Jackson, R. J., Hellen, C. U. T. and T. V. Pestova (2010). Nature Reviews Molecular Cell Biology. 11, 113-127. 3  1.2  Translation elongation in eukaryotes At the conclusion of the initiation process, the 80S ribosome is positioned at the AUG  codon on the transcript, with an initiator Met-tRNAiMet present in the P-site, and a vacant Asite. In order for the elongation cycle to begin (Figure 1.2), a cognate aminoacyl-tRNA is delivered to the A-site of the ribosome through the action of eukaryotic elongation factor (eEF) 1A. Correct anticodon-codon base pairing in the A-site leads to the hydrolysis of GTP on eEF1A (Hopfield, 1974; Ninio, 1975; Thompson and Stone, 1977; Ruusala et al., 1982) and its subsequent departure from the ribosome, leaving the aminoacylated tRNA in the Asite of the ribosome. After the tRNA acceptor arm is placed into the position that is required for peptide transfer in a process called accommodation, the ribosome then spontaneously catalyzes peptidyltransfer (Pape et al., 1998), transferring the growing peptide chain to the Asite tRNA. At this point, the tRNAs present in the A- and P-sites adopt intermediate conformational states (Blanchard et al., 2004; Kim et al., 2007; Fei et al., 2008), called the A/P and P/E hybrid states respectively, where the acceptor stem of the tRNA has shifted to its next position in the large subunit, but not in the small subunit (Moazed and Noller, 1989). This switch to the hybrid state is also accompanied by a “ratcheting” of the ribosomal subunits, relative to one another (Frank and Agrawal, 2000; Valle et al., 2003). It is thought that this pretranslocation conformation is recognized by the GTP-bound translocase eukaryotic elongation factor (eEF) 2 (Semenkov et al., 2000; Dorner et al., 2006). Upon eEF2 binding, GTP hydrolysis drives the translocation event, moving the A/P and P/E hybrid tRNAs into the P- and E-sites, respectively (Frank and Agrawal, 2000; Frank et al., 2007; Taylor et al., 2007). This leaves the peptidyl-tRNA in the P-site of the ribosome and the Asite vacant, ready to accept a new aminoacylated tRNA for another round of elongation.  4  Figure 1.2. The eukaryotic translation elongation cycle. A general overview of the elongation cycle. (a) The primed 80S ribsome will accept an aminoacylated tRNA in the Asite of the ribosome through the action of eEF1A. If the tRNA is a cognate tRNA, it is (b) accommodated by the ribosome and GTP is hydrolyzed, releasing eEF1A. Peptidyltransfer is then catalyzed spontaneously by the ribosome, and the tRNAs present in the A- and P-sites will (c and d) fluctuate between the classical and hybrid states, accompanied by subunit ratcheting. The hybrid state is recognized by the GTP-bound translocon eEF2, and GTP is hydrolyzed to drive the translocation event, yielding a (e) post-translocation ribosome. Adapted with permission from Agirrezabala, X. and J. Frank. (2009). Quarterly Reviews of Biophysics. 42: 159-200. 5  1.3  Cap-independent translation initiation by internal ribosome entry Although most mRNAs are translated via the cap-dependent scanning mechanism of  translation initiation, there are a few examples of mRNAs whose protein synthesis is initiated through alternative mechanisms. Several of these mechanisms include the use of small upstream open reading frames and cap-independent translation elements (CITE). In general, all of these mechanisms have evolved to allow translation during periods of time when overall translation is inhibited. Although interesting, details on these other mechanisms are beyond the scope of this thesis. Rather, this thesis is focused on cap-independent translation through internal ribosome entry, which will be discussed here. Historically, cap-independent translation was a curiosity first observed in picornaviruses. These viruses possess a positive sense, single stranded RNA genome which lacks a 5`-7-methylguanosine cap, precluding the possibility of eIF4E binding. Moreover, the picornaviral 5’ UTR is long with a number of non-initiating AUG codons, further suggesting that it is not translated by a scanning mechanism. Thus, it was curious how picornaviruses could recruit and activate the translational machinery through the conventional capdependent scanning mechanism. However, in the late 1980s, it was found independently by two groups (Jang et al., 1988; Pelletier and Sonenberg, 1988) that these viruses are translated in a cap-independent manner by the recruitment of the 43S preinitiation complex directly to the RNA by an element called an internal ribosome entry site (IRES). Since then, IRESs have been one of the best studied examples of cap-independent translation initiation in eukaryotes. These are often structured elements within a transcript, and as their name implies, IRESs possess the ability to recruit the ribosome to internal locations within the mRNA, allowing the ribosome to bypass binding to the 5`-7-  6  methylguanosine cap. However, the particular property that makes IRESs so peculiar is their ability to recruit and activate the ribosome for translation in the absence of some, or even all of the factors required for the initiation of cap-dependent translation. The ability to bypass this requirement often allows IRES-containing transcripts to remain translated during periods of cellular stress or viral infection when normal cap-dependent translation is rendered nonfunctional. One of the better known examples of this phenomenon is observed during poliovirus infection. After infection with poliovirus, the viral 3C protease cleaves the initiation factor eIF4G between the binding sites for eIF4E and eIF3 (Etchison et al., 1982; Bernstein et al., 1985). This ultimately results in the inhibition of host translation as it prevents the recruitment of the 43S preinitiation complex to host mRNAs. However, poliovirus is translated preferentially during this time. This begs the question; how does poliovirus translate its own genome when eIF4G has been compromised? In order to remain translated, poliovirus contains an IRES within its genome that possesses the ability to recruit the cleaved C-terminal fragment of eIF4G (Pestova et al., 1996; de Breyne et al., 2009). This allows the polioviral genome to ultimately recruit the ribosome and be translated under conditions when cap-dependent translation is inhibited by eIF4G cleavage. This ability to circumvent the requirement for initiation factors in part defines an IRES. Thus, the viral IRESs found to date have been roughly categorized into four groups according to their initiation factor requirements (Figure 1.3). Group 4 IRESs, such as the ones found in poliovirus and rhinovirus, require Met-tRNAiMet and all eIFs except for eIF4E. These IRESs directly bind to a portion of eIF4G and thus can still recruit the 43S preinitiation complex. The complex will then subsequently scan for the downstream initiation  7  codon and initiate translation in a manner similar to the canonical initiation mechanism. Additionally, these IRESs also have specific requirements for certain IRES trans-acting factors (ITAF), which are additional cellular factors that are required for IRES functionality. The identity and nature of these cellular factors will be discussed in later sections. In a similar manner, group 3 IRESs also do not require eIF4E and have the same factor requirements as the group 4 IRESs. However, these IRESs are able to recruit the ribosome directly to the initiation codon, and thus do not involve ribosome scanning. IRESs of this type are found in encephalomyocarditis virus (EMCV) and foot and mouth disease virus (FMDV). In contrast, group 2 IRESs, as is found in hepatitis C virus (HCV) and classical swine fever virus (CSFV), only use a subset of initiation factors that include eIF2, eIF3, and Met-tRNAi, and recruit the ribosome directly to the start site. Finally, group 1 IRESs are notable for the fact that they can bind directly to the ribosome and do not require any initiation factors or Met-tRNAiMet. Examples of these IRESs include the ones found in dicistrovirus family, which includes cricket paralysis virus (CrPV), Plautia stali intestine virus (PSIV), and Taura syndrome virus (TSV). This type of viral IRES classification scheme highlights the fact that viruses from distinct families have IRESs with different factor requirements. Since viruses often have diverse strategies to comandeer the cell upon infection, this may provide a rationale as to why each IRES utilizes a different mechanism during infection to capture the translational machinery. Accordingly, IRESs from different viral families have vastly different structures and varied mechanisms to accomplish this. For instance, as described above, the poliovirus IRES recruits C-terminal fragment of eIF4G to ultimately recruit the translational machinery. In contrast, the HCV IRES interacts directly with the 40S ribosomal subunit and eIF3,  8  eliminating the requirement for any proteins that are associated with the cap structure (Pestova et al., 1998; Hellen and Pestova, 1999; Pestova and Hellen, 1999; Kolupaeva et al., 2000b; Kolupaeva et al., 2000a; Hellen and Sarnow, 2001). Conversely, some IRESs forgo the need for any secondary structure at all such as the 5’ Rhopalosiphum padi virus (RhPV) IRES, which is an unstructured U-rich tract that appears to serve as a somewhat non-specific “landing pad” for the translational machinery (Terenin et al., 2005; Groppelli et al., 2007). However, despite their differences in molecular mechanism, there is a common theme that ties all these viral IRESs together. In all cases, these RNA elements, both structured or nonstructured, serve to recruit components of the translational machinery for the purpose of initiating translation. Although only viral IRESs have been listed above, this mechanism of capindependent translation initiation is not limited to viruses. Numerous IRESs which drive the translation of cellular genes such as the X-linked inhibitor of apoptosis (XIAP) and p27 have also been discovered. I will discuss these, and focus on the RNA structure and factor requirements of these different IRESs. Furthermore, I will discuss specifics of some of the other major viral IRES classes, and how viruses utilize distinct strategies for ribosome recruitment in the following sections.  9  Figure 1.3. A classification scheme of viral IRESs. Shown are the four groups in which currently known viral IRESs may be classified. Group 4 IRESs require Met-tRNAiMet, ITAFs, and all eIFs except for eIF4E, and function efficiently in RRL only when supplemented with extracts from other cell types. Additionally, these IRESs initiate at an AUG codon downstream of the IRES and thus require scanning. Similarly, group 3 IRESs have the same requirement for initiation factors as the group 4 IRESs, but are able to recruit the ribosome directly to the initation codon and thus do not require scanning to occur. Group 2 IRES RNAs are able to bind to the 40S subunit and also only use a subset of the canonical eIFs consisting of eIF3 and eIF2, as well as Met-tRNAiMet. Finally, group 1 IRES RNAs bind directly to the ribosome and do not require any protein factors to do so. These IRESs also do not require initiator Met-tRNAiMet. Shown at top for comparison’s sake are the factors required for cap-dependent translation. This figure has been adapted with permission from Kieft, J. S. (2008). Trends in Biochemical Sciences. 33, 274-283.  10  1.3.1  Picornaviridae IRESs The IRESs from groups 4 and 3 are comprised of the IRESs that are found in most  picornaviruses. These IRESs are found in the 5’ end of the viral RNA, and are responsible for driving the translation of the entire viral genome. A general theme upon picornavirus infection is the relatively quick inhibition of host translation, usually through the cleavage of eIF4G by viral proteases. In order to circumvent the need for eIF4G, the IRES can bind the central domain of eIF4G, and thus can still recruit the translational machinery (Pestova et al., 1996; de Breyne et al., 2009). Thus, for the most part, these IRESs require Met-tRNAiMet and all eIFs except for eIF4E. Despite being found in the members of a single viral family, the IRESs found in groups 4 and 3 are not conserved very well at the sequence level. However, similarities found at the structural level allow them to be classified into four types, I through IV (Figure 1.4). The type I IRESs are found in enteroviruses such as poliovirus and human rhinovirus (HRV). The type II grouping includes IRESs found in cardioviruses and apthoviruses, such as EMCV and FMDV. The atypical type III IRESs include hepatoviruses, such as hepatitis A virus (HAV). Finally, the unusual HCV-like IRESs found in the teschovirus genus belong in the type IV group. For the most part, the IRESs classified in types I through III adopt a similar secondary structure. In these IRESs, there is a central core, which contains a characteristic cloverleaf structure, similar to what is seen in a tRNA. In fact, this structure can be experimentally cleaved by RNase P, a tRNA processing enzyme, which may reflect the picornaviral IRES’s origins as an evolutionary relic of a tRNA (Lyons and Robertson, 2003; Serrano et al., 2007). Immediately downstream of this central domain are elements which are responsible for  11  eIF4G binding, allowing the IRES to recruit the translational machinery (Lopez de Quinto and Martinez-Salas, 2000; Saleh et al., 2001). Conversely, immediately upstream of the core are elements which bind ITAFs for the facilitation of translation (Witherell et al., 1993; Kolupaeva et al., 1996; Song et al., 2005). The enterovirus IRESs that constitute the type I IRESs were initially studied in the rabbit reticulocyte lysate (RRL) translation system. At the time, this translation system was considered to be attractive for its ability to recapitulate eukaryotic translation in a very robust manner. Curiously, poliovirus and HRV are poorly translated in RRL, if at all (Pelletier et al., 1988). Efficient IRES-driven translation in RRL was only observed if the lysate was supplemented with HeLa cytoplasmic extracts (Dorner et al., 1984). This implied that a cytoplasmic trans-acting factor present in HeLa cell extracts is required for efficient IRES translation, and was the first evidence for the existence of ITAFs. Since then, additional ITAFs have been identified by biochemical fractionation of cell lysates (Kaminski et al., 1995; Belsham and Sonenberg, 1996; Blyn et al., 1997; Hunt et al., 1999; Hunt and Jackson, 1999). These ancillary factors include the La human autoantigen, polypyrimidine tract-binding protein (PTB), poly(rC)-binding protein (PCBP), and the upstream-of-N-ras protein (unr). It is interesting to note that IRESs of the same type that share secondary structural elements often require different ITAFs. For instance, in the case of the type II IRESs, ITAF45 is required for FMDV IRES function, but not for Theiler's murine encephalomyelitis virus (TMEV) IRES function (Pilipenko et al., 2000; Pilipenko et al., 2001). It has been proposed that these observed differences in ITAF requirements may contribute to defining viral tissue tropism (Pilipenko et al., 2000; Pilipenko et al., 2001), but this has not yet been shown convincingly with these particular ITAFs, as they are present in  12  many cell types. In the case of poliovirus, however, ITAFs play a role in delineating the switch between viral translation and replication (Gamarnik and Andino, 1998; Gamarnik and Andino, 2000). Although it is agreed that these proteins may behave as RNA chaperones and stabilize IRES structures, or act as scaffolds to recruit other factors, their specific mechanistic role remains largely unknown. The type III IRESs, which are typified by the HAV IRES, are also oddly placed in this group, since this IRES has a requirement for the intact cap-binding complex, eIF4F. Since HAV is a non-lytic virus, there is no requirement for the inhibition of host translation upon HAV infection. Accordingly, no HAV protease cleaves eIF4G. Thus, this may explain why the HAV IRES still utilizes the entire eIF4F complex, including eIF4G (Ali et al., 2001). This observation also implies that the presence of a cap would stimulate HAV translation, but further experiments have found that the addition of a 5’-7-methylguanosine cap to the HAV RNA genome inhibited translation in vitro (Brown et al., 1994). To make matters more confusing, eIF4E has a stimulatory effect on the HAV IRES (Ali et al., 2001). To date, the basis for the requirement of eIF4E is still unknown, but it is thought that it either binds to the IRES directly through its cap binding pocket, or that it causes a conformational change in eIF4G upon binding to make it more amenable for IRES-driven translation. The latest type of IRES to be characterized in this viral family are the type IV IRESs found in the teschovirus genus. The prototype of this IRES was initially discovered in porcine teschovirus (PTV), another picornavirus. Preliminary studies found no obvious similarity in terms of structure and sequence with the previously discovered picornavirus IRESs (Figure 1.4). What made this IRES so peculiar was the observation that this IRES only had a requirement for eIF2-GTP-Met-tRNAiMet, with eIF3 as a stimulatory factor (Pisarev et  13  al., 2004). Biochemical evidence shows this IRES binds to 40S subunits and eIF3 directly to initiate translation (Pisarev et al., 2004), in stark contrast to the other picornaviral IRESs, which exclusively recruit initiation factors. In that sense, these IRESs are proposed to be HCV IRES-like as they can initiate translation with similar factor requirements as the HCV IRES, which will be discussed in the next section. In fact, sequence analysis shows that the PTV and HCV IRESs share ~50% sequence identity and are similar in secondary structure. It has been proposed by Richard Jackson of the University of Cambridge that all picornaviral IRESs have evolved from the same common ancestor (Jackson, 2005). Presumably, this common ancestor is similar to the ones exemplified by the type III IRES, which has been stuck in what he calls an “evolutionary backwater” and still requires all of the canonical translation initiation factors for functionality. From this common ancestor, the IRESs present in types I and II may have evolved the ability to function without the need for the eIF4E-eIF4G interaction, which allows these viruses the freedom to evolve mechanisms to inhibit cellular translation through the cleavage of eIF4G. The type IV IRESs may have taken an evolutionary step further, evolving to dispense with the cap-binding complex entirely, and may serve as the evolutionary link between the picornaviral IRESs and the flaviviral IRESs, of which HCV IRES is a member.  14  Figure 1.4. The secondary structure of picornavirus IRESs. The secondary structure of the four types of IRESs found in picornaviruses are shown. The tRNA-like four-way junction is indicated by an asterisk. The initiation factor binding domains in the type I, II, and III IRESs are indicated by broken lines. Adapted from M. Niepmann. (2009). Biochimica et Biophysica Acta - Gene Regulatory Mechanisms. 1789, 529-541.  15  1.3.2  Flaviviridae IRESs The best studied IRES to date from the entire Flaviviridae family is the HCV IRES.  Like picornaviruses, the entire flavivirus genome is translated as a single polyprotein from a positive sense, single-stranded genomic RNA. In the case of HCV, the translation of this polyprotein is driven by an IRES in the 5’ UTR. The flaviviral IRES is fundamentally different from the classical picornaviral IRESs, in terms of its structure, initiation factor requirement, and molecular mechanism. Using HCV IRES as a model, the flaviviral IRES possesses three domains, designated II through IV (Figure 1.5). These domains work in conjunction with one another to recruit the ribosome directly to the start codon, without the need for scanning. Domain III forms the largest domain and contains a conserved four way junction that can be cleaved by RNase P (Nadal et al., 2002), suggesting that the IRES has a tRNA-like structure, reminiscent of the core domain found in picornaviral IRESs. Portions of domain III were later found to be required for eIF3 binding, and domain II makes specific contacts with the 40S ribosomal subunit (Buratti et al., 1998; Sizova et al., 1998; Babaylova et al., 2009). This observation suggested that the mode of action of the flaviviral IRES is completely different from its picornaviral counterparts. Subsequent experiments confirmed these ideas, and showed that the HCV IRES can bind to 40S subunits without the need for any initiation factors, requiring only eIF2GTP-Met-tRNAiMet and eIF3 for the assembly of functional 80S ribosomes (Pestova et al., 1998). This property makes the flaviviral IRESs remarkable in the sense that they act as aptamers that bind directly to the 40S subunit, as opposed to the picornaviral IRESs, which serves as a scaffold to recruit the components of the translational machinery.  16  Additional biochemical observations suggested that the HCV IRES can initiate translation independently of eIF2-GTP-Met-tRNAiMet (Terenin et al., 2008), allowing HCV to circumvent the cellular eIF2-mediated inhibition of translation that occurs as a protective mechanism during viral infection. However, it was unknown how the HCV IRES could still recruit Met-tRNAiMet when eIF2 activity is compromised. This remained a point of confusion until recently. New findings have shown that ligatin, a protein that had been thought to be unrelated to translation, serves to substitute for eIF2 and is utilized by the HCV IRES to deliver Met-tRNAiMet to the ribosome (Dmitriev et al., 2010; Skabkin et al., 2010). Cryo-EM studies have illuminated the fashion in which the HCV IRES binds to the 40S ribosome (Figure 1.5) (Spahn et al., 2001b). The IRES binds on the solvent face of the 40S subunit, and domain IIb of the IRES reaches around to partially overlap with the E-site. Domain IIIb extends away from the ribosome surface, forming a platform for eIF3 binding. Remarkably, IRES binding causes a conformational change in the 40S subunit, suggesting that the IRES is actively manipulating the translational machinery (Spahn et al., 2001b). Like the other picornaviral IRESs described thus far, the flaviviral IRESs are regulated by other cellular factors. Canonical ITAFs, such as La autoantigen have been found to bind to the IRES, as well as others such as heterogeneous nuclear ribonucleoprotein (hnRNP) L and NSAP1 (Ali and Siddiqui, 1997; Hahm et al., 1998; Kim et al., 2004). However, the nature of how these ITAFs act positively upon the IRES and their biological importance remains a contentious issue (Gosert et al., 2000; Tischendorf et al., 2004). As mentioned in the previous section, the type IV picornaviral IRESs are similar in terms of sequence and structure to the flaviviral IRESs described in this section. This implies that there may have been some degree of horizontal gene transfer between these viral  17  families, although the direction in which this may have occurred is unknown. However, this hypothesis remains to be tested.  18  Figure 1.5. The structure of the HCV IRES. At left is the secondary structure of the HCV IRES. The tRNA-like four-way junction is indicated by an asterisk, and the target sites for microRNA-122 is indicated by boxes. At right is a cryo-EM reconstruction of the HCV IRES, shown in purple, bound to the 40S ribosomal subunit. The mRNA entry and exit channels are indicated. Adapted from M. Niepmann. (2009). Biochimica et Biophysica Acta - Gene Regulatory Mechanisms. 1789, 529-541, and Spahn, C. M., et al. (2001). Science. 291, 1959-1962.  19  1.3.3  Group I IRESs The Group I IRESs possess the simplest mechanism of translation initiation mediated  by an IRES found to date. These IRESs are remarkable for their ability to bind directly to the ribosome with high affinity and initiate translation without the requirement for any initiation factors. These IRESs are responsible for driving the translation of the structural proteins in the viruses present in the Dicistroviridae viral family, and are the major topic of interest in this thesis. These elements will be described in detail in following sections.  1.3.4  Cellular IRESs Cellular IRESs were discovered shortly after viral IRESs and were expected to be  similar, in terms of possessing structural conservation within closely related family members and clearly mediating factor-independent translation. Oddly enough, cellular IRESs appear to be vastly different from their viral counterparts in this regard. For the most part, viral IRESs have some degree of structural conservation, and small deletions that are made in conserved regions have a large repressive effect on their activity. However, these observations are not typical of cellular IRESs. Cellular IRESs do not have conserved structures, and even closely related genes that possess an IRES, such as c-myc and L-myc, share no structural conservation (Le Quesne et al., 2001; Jopling et al., 2004), illustrating this point. Furthermore, mutations in cellular IRESs generally do not fully abrogate translational activity (Yang and Sarnow, 1997; Stoneley et al., 1998; Coldwell et al., 2000), suggesting that these mRNAs can be either translated in a cap-dependent fashion, or that these cellular IRESs are composed of modular elements that act in combination to recruit the translational machinery effectively. In support of the latter hypothesis, mutations in both but not one of the two structural regions of the c-myc IRES disrupts IRES activity (Le 20  Quesne et al., 2001). Since cellular IRESs appear to be capable of mediating translation in a cap-dependent manner, this has led to some contention as to whether cellular IRESs have been properly characterized (Kozak, 2003; Kozak, 2005). Furthermore, some eukaryotic IRESs may not require secondary structural elements at all, as some data suggests that there is a correlation between high IRES translational activity and weak secondary structure, suggesting a shared mechanism in eukaryotic IRESs that rely on unstructured elements (Xia and Holcik, 2009). Thus, current evidence suggests that these cellular IRESs operate in a fundamentally different manner than currently known viral IRESs.  21  1.4  The Dicistroviridae viral family The dicistroviruses are a recently discovered family of RNA viruses that exclusively  infect a wide variety of arthropods. Recent metagenomic studies suggest that dicistro- and dicistro-like viruses are ubiquitous in the environment (Culley et al., 2006), and almost all dicistroviruses discovered to date are relevant in agriculture or medicine. For example, in the early 1900s, the collapse of shrimp stocks in the Taura River basin in Ecuador was considered a mystery, and was initially called “Taura syndrome”. This disease eventually spread to the majority of the Americas, and the causative agent of the disease was found to be a dicistrovirus which was later named Taura syndrome virus, or TSV (Hasson et al., 1995). TSV is still considered to be one of the most devastating diseases that have affected the shrimp farming industry. More recently, another dicistrovirus, Israeli acute paralysis virus (IAPV), has been associated with colony collapse disorder in honeybees, which currently threatens bee stocks worldwide (Cox-Foster et al., 2007). These and other dicistroviruses stress the need for further study into this family of viruses. The genome of a dicistrovirus consists of a positive sense, single-stranded RNA. The 5’ end is linked to a protein called viral protein genome-linked (VPg) (Figure 1.6) (Bonning and Miller). The name of the viral family is derived from the presence of two open reading frames, or cistrons, within the genome, which are separated by an untranslated intergenic region (IGR). Viral nonstructural proteins are encoded by the 5’ cistron, whereas the 3’ cistron encodes for the viral structural proteins. The translation of both cistrons is driven by two distinct IRESs directly upstream; one present in the 5’ UTR and the other in the IGR (Wilson et al., 2000b; Bonning and Miller).  22  To date, fourteen viruses that belong to the Dicistroviridae viral family have been discovered (Table 1.1), and these viruses have been putatively subdivided into two genera, Cripavirus and Aparavirus, based upon their evolutionary distance from one another, and the secondary structure of the IRES present within the IGR. One model system that has been used to study the dicistrovirus life cycle is CrPV. Upon infection, the CrPV virion is taken up by the cell and cellular translation is rapidly inhibited through the disruption of the eIF4E-eIF4G interaction (Wilson et al., 2000b; Garrey et al., 2010) through a mechanism that has yet to be elucidated. Both the 5’ UTR and IGR IRESs seem to require the distruption of host translation in order to be translationally active, presumably because this increases the pool of ribosomes available for recruitment (Fernandez et al., 2002; Garrey et al., 2010). It is interesting to note that the translational activity of the IGR IRES is stronger than that of the 5’ UTR IRES, and as a result, the structural proteins are produced in molar excess over the nonstructural proteins (Wilson et al., 2000b; Garrey et al., 2010). This is in stark contrast to the picornaviruses, which produce all viral proteins in equimolar amounts. Since a single RNA-dependent RNA polymerase (RdRp) molecule can generate multiple genomic RNAs, whereas many copies of packaging proteins are required for encapsidation, this expression strategy allows dicistroviruses to produce an appropriate number of catalytic proteins for a productive infection, and may provide a rationale as to why these viruses have evolved a two IRES expression system. In order to further study these two IRES, the RhPV IRES has been used as a model to study the 5’ UTR IRES. As discussed in previous sections, this IRES has been oddly characterized as an unconserved U-rich region with little or no secondary structure. In terms  23  of initiation factor requirements, the RhPV IRES only explicitly requires eIF1, eIF2, and eIF3, classifying it as a Group IV IRES (Terenin et al., 2005; Groppelli et al., 2007). It has been proposed that these factors have a large enough affinity for the unstructured U-rich regions of the IRES to allow for the sequential recruitment of the translational machinery in a non-specific manner. In contrast, the IGR IRES requires no factors for the purpose of translation initiation, and can recruit the ribosome independently. This thesis is concerned with the characterization of the IGR IRES in particular, and further information about this IRES will not be discussed here, but will be presented in the following section.  24  Figure 1.6. The genomic architecture of a dicistrovirus. The genome of a dicistrovirus is composed of a positive, single-stranded, poly(A)-tailed RNA that is capped with VPg at the 5’ end. The genome itself consists of two open reading frames whose translation is driven by two different IRESs immediately upstream; one in the 5’ UTR and the other in the IGR. The 5’ cistron encodes for the viral nonstructural proteins, which are currently known to encode for an RNA silencing suppressor (SS), a superfamily 3 helicase (hel), VPg, a 3A-like protein, a chymotrypsin-like cysteine protease (pro), and an RNA-dependent RNA polymerase (RdRp). Conversely, the 3’ cistron encodes for the viral structural proteins, VP1 through 4. Adapted with permission from Bonning B. C. and W. A. Miller. (2010). Annual Review of Entomology. 55, 129-15  25  Table 1.1. Dicistroviruses and their natural hosts. a  Virus Genus: Cripavirus Aphid lethal paralysis virus Black queen cell virus Cricket paralysis virus Drosophila C virus Himetobi P virus Homalodisca coagulata virus-1 Plautia stali intestine virus Rhopalosiphum padi virus Triatoma virus Genus: Aparavirus b Acute bee paralysis virus Israeli acute paralysis virus c Kashmir bee virus Solenopsis invicta virus-1 Taura syndrome virus  Year of original Acronym description  Year genome sequenced  ALPV BQCV  1988 1977  2002 2000  CrPV DCV HiPV  1970 1972 1992  2000 1998 1999  Hemiptera Hymenoptera Diptera, Hemiptera, Hymenoptera, Lepidoptera, Orthoptera Diptera Hemiptera  HoCV-1 PSIV RhPV TrV  2006 1998 1981 1987  2006 1998 1998 2000  Hemiptera Hemiptera Hemiptera Hemiptera  ABPV IAPV KBV SINV-1 TSV  1963 2007 1977 2004 1995  2000 2007 2004 2004 2002  Hymenoptera Hymenoptera Hymenoptera Hymenoptera Decapoda  Order of natural hosts  a  Adapted with permission from Bonning B. C. and W. A. Miller. (2010). Annual Review of Entomology. 55, 129-150. The genus has been approved, and is awaiting ratification from the International Committee on Taxonomy of Viruses. c This virus is currently an unclassified dicistrovirus, but is awaiting its inclusion into the Aparavirus genus, based on phylogenetic studies b  26  1.5  The Dicistroviridae intergenic region internal ribosome entry site As described in the previous section, the IGR IRES can recruit and activate the  translational machinery without the need of any initiation factors, which classifies it as a Group I IRES. More specifically, the IRES directly recruits both 40S and 60S subunits for 80S assembly. From extensive studies (Jan and Sarnow, 2002; Kieft, 2008a; Kieft, 2008b; Nakashima and Uchiumi, 2008), it has been shown that the IRES adopts a conserved twodomain architecture, composed of three pseudoknots (PK) (Figure 1.7, left). However, most of the IGR IRES sequence is unconserved across viral family members, except for a select few residues which reside mostly in single-stranded loop regions. Since RNAs generally interact with other macromolecules through single-stranded regions, this implies that these conserved regions of the IRES are important for interacting with components of the translational machinery. In agreement with this idea, the conserved single-stranded regions in stem-loop (SL) IV and V are responsible for binding to components of the 40S ribosomal subunit (Nishiyama et al., 2003; Spahn et al., 2004b; Pfingsten et al., 2006; Schuler et al., 2006; Nishiyama et al., 2007; Landry et al., 2009). The dicistrovirus IGR IRESs can be grouped into two types, according to their secondary structure. Type I IRESs are exemplified by the CrPV and PSIV IGR IRESs, whereas examples of type II IGR IRES are found in TSV and IAPV (Hatakeyama et al., 2004; Cevallos and Sarnow, 2005; Jan, 2006; Nakashima and Uchiumi, 2008). The main difference between the two classes is that the class II IRES possesses a larger L1.1 loop region, and an extra stem-loop (SL III) within PKI (Figure 1.7). The functional significance of these differences is not known. It has been shown, however, that both type I and II IRESs have related folding properties and function similarly in that neither require initiation factors  27  for ribosome recruitment and positioning (Hatakeyama et al., 2004; Cevallos and Sarnow, 2005; Pfingsten et al., 2007). A crystal structure (Figure 1.8) (Pfingsten et al., 2006) of the IGR IRES shows that one domain, consisting of PKII and PKIII can form a folded “core”, with stem-loops IV and V emerging from one surface. Analysis of the docked structure with cryo-EM reconstructions of the ribosome show that the conserved loops of SL IV and SL V appear to bind to rpS5 of the 40S ribosomal subunit. This is in agreement with cryo-EM reconstructions, which also show that portions of the IRES interact with rpS5 (Spahn et al., 2004b; Schuler et al., 2006). Biochemical data is also in agreement, as SL IV and SL V can bind to rpS25, which is a neighbor of rpS5 (Nishiyama et al., 2003; Nishiyama et al., 2007; Landry et al., 2009). Moreover, mutations in SL IV and SL V disrupt IRES activity (Jan and Sarnow, 2002; Nishiyama et al., 2003; Pfingsten et al., 2006). In the same domain, the docked structure and cryo-EM reconstructions both suggest that the conserved L1.1 loop binds to the L1 stalk of the 60S ribosomal subunit, in agreement with studies that show that mutations in this region abolish 60S subunit binding (Pfingsten et al., 2006; Jang et al., 2009). From another crystal structure from the Kieft group (Costantino et al., 2008), the PKI domain mimics a tRNA (Figure 1.9), and occupies the P-site of the ribosome upon ribosome binding (Wilson et al., 2000a; Jan et al., 2003; Pestova and Hellen, 2003; Costantino et al., 2008; Kamoshita et al., 2009; Zhu et al., 2011). What makes this mimicry striking is the observation that PKI mimics a P/E hybrid (Costantino et al., 2008; Zhu et al., 2011), which is a tRNA conformation that is only present as an intermediate during the translation elongation cycle. As a result, it appears that the IRES manipulates the ribosome and sets it into an elongation-competent mode. In agreement with this idea, cryo-EM reconstructions (Spahn et  28  al., 2004b) have shown that conformational changes that occur within the ribosome upon IRES binding are identical to ones observed during elongation factor 2 binding (Figure 1.10). IRES binding induces a rotation of the head of the 40S subunit and a latch interaction between helix 18 and helix 34 of the rRNA is formed which has been proposed to guide the mRNA into the channel of the ribosome (Spahn et al., 2004b). Most notably, the ribosomal P proteins in the stalk region of the 60S subunit become ordered (Agrawal et al., 1999; GomezLorenzo et al., 2000). Previous research has shown that this region is responsible for interacting with elongation factors (Agrawal et al., 1999; Gomez-Lorenzo et al., 2000), thus suggesting that the IRES is inducing this conformational change to facilitate elongation factor binding (Spahn et al., 2004b; Jan, 2006). Despite having different binding sites on the ribosome, the HCV IRES and IGR IRES induce similar conformational changes in the ribosome (Spahn et al., 2001b). This suggests that these two IRESs are inducing conformational changes that are part of the general pathway of translation initiation. After the tRNA mimic has positioned itself in the P-site of the ribosome, the cognate aminoacyl-tRNA is delivered to the A-site and the ribosome undergoes a round of translocation without peptide bond formation in a step called pseudotranslocation, mediated by eEF1A and eEF2 (Jan et al., 2003; Pestova and Hellen, 2003). This step is notable for the fact that this cognate aminoacyl-tRNA does not encode a methionine. In the case of the CrPV IGR IRES, the model IRES which is the focus of this thesis, the first codon is a GCU, which encodes an alanine. However, the IGR IRES can initiate from all codons except stop codons (Shibuya et al., 2003). Thus, translation of the second cistron in dicistroviruses is unorthodox in the sense that it is initiated at a non-AUG codon from the A-site of the ribosome.  29  The IGR IRES itself serves as a novel, streamlined form of translation initiation that functions without the need for any initiation factors. Due to this fact, we can use the IRES as a scientific tool and exploit it to study different aspects of translation. For instance, since the IRES resembles a tRNA to subvert the translational machinery, the study of the IGR IRES may provide insights into how the ribosome interacts with tRNAs during the canonical translation cycle, and how the ribosome functions mechanistically. Furthermore, the mechanistic simplicity of this IRES serves as a good model for understanding other IRESs that do not require initiation factors for 40S recruitment, such as the HCV IRES, the HIV-2 IRES (Locker et al.), and the c-src IRES (Allam and Ali, 2010). Additionally, within the context of dicistroviruses, the study of this IRES may help us understand how this family of viruses is translated, and may provide novel targets for combating these viruses. This thesis is primarily concerned with the characterization of the IGR IRES. We first hypothesized that certain conserved regions of the IGR IRES are responsible for separate roles in the process of translation initiation. In chapter 2, we focus on the biochemical characterization of one of these conserved regions, its roles, and how it mimics an E-site tRNA during translation. We then hypothesized that the IRES was functionally modular. The work presented in chapter 3 addresses the two-domain architecture of the IGR IRES, how it is indeed modular, and how this observation may tell us more about its evolutionary origins. We then considered how the IRES functions within the greater context of the the cell, and studied how it remains translated during periods of translational shutoff during infection. Here, we hypothesized that the IGR IRES is able to bind to eEF2-associated 80S ribosome couples which mimic what is seen during stress conditions when overall translation is  30  inhibited, thus providing a rationale as to how the IGR IRES remains translated during these periods. In the final chapter, we present our studies about the interplay between IRES translation and disease, and hypothesized that defective IRES translation is implicated in the pathology of a human disease, dyskeratosis congentia. Taken entirely, this body of work is a comprehensive study on the IGR IRES. The first two chapters pertain to its biochemical characterization, and its evolutionary origins. The subsequent chapters delve into how the IRES functions within the context of viral infection, and how IRES translation in general may be implicated in human disease states.  31  Figure 1.7. The secondary structure of the (A) type I CrPV IGR IRES and the (B) type II TSV IGR IRES. Conserved nucleotide positions are shown in uppercase and nonconserved nucleotides are in lowercase. Numbering refers to the nucleotide position within the respective viral genome. Helical regions are indicated by a black dash between nucleotides. Pseudoknots are labeled and colour coded as indicated. Underlined nucleotides represent the first two amino acid residues in the viral capsid protein. A black horizontal line divides the two domains of the IRES. Properly positioned 40S and 80S ribosomes on the IGR IRES produces a toeprint, denoted Toeprint A and B, shown by the arrows. The ΔSL III deletion within the type II TSV IGR IRES is marked by a black box. L1.1A and L1.1B are indicated by purple and brown letters, respectively.  32  Figure 1.8. 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. Macromolecules that bind to the IGR IRES are also indicated in boxes, and coloured accordingly. Adapted with permission from Pfingsten, J. S. and J. S. Kieft. (2008). RNA. 14, 1255-1263.  33  Figure 1.9. Mimicry of a tRNA by PKI of the IGR IRES, and a comparison with an authentic tRNA. At left are two views, one rotated 90° on a vertical axis, of PKI of the 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 two structures are colored to match each other. Cyan in the tRNA–mRNA structure at right is the decoded mRNA. Adapted with permission from Macmillan Publishers Limited, from Costantino, D. A., Pfingsten, J. S., Rambo, R. P., and J. S. Kieft. (2008). Nature Structural & Molecular Biology. 15, 57-64.  34  Figure 1.10. IRES-induced conformational changes in the human 80S ribosome. Shown at left is a cryo-EM reconstruction of a human 80S ribosome. Landmarks for the 40S subunit are the following: b, body; bk, beak; h, head; pt, platform; and sh, shoulder. The position of one 18S rRNA helix, h16, and the ribosomal proteins rpS3 and rpS5 are indicated, as identified by comparison with a cryo-EM map of the yeast 80S ribosome. At right is a CrPV IRES-80S ribosome complex. The conformational change in the complex is indicated with an asterisk, and the CrPV IRES-induced extended stalk is labeled St. All reconstructions are viewed from the 40S subunit solvent side with the 40S subunit painted yellow, the 60S subunit blue, and the CrPV IRES purple. Adapted with permission from Spahn C. M. T., et al. (2004). Cell. 118, 165-475.  35  2 2.1  Characterization of the conserved L1.1 region of the IGR IRES * Summary The IGR IRES from the Dicistroviridae family of viruses mimics a tRNA to directly  assemble 80S ribosomes and initiate translation at a non-AUG codon from the ribosomal Asite. A comparison of IGR IRESs within this viral family reveals structural but little sequence similarity. However, a few specific conserved elements exist which likely play important roles in IRES function. Here, we have generated a battery of mutations to characterize the role of a conserved loop (L1.1) region of the IGR IRES. Mutating specific nucleotides within the L1.1 region inhibited IGR IRES-mediated translation in rabbit reticulocyte lysates. By assaying different steps in IRES function, we found that the mutant L1.1 IRESs had reduced affinity for 80S ribosomes but not 40S subunits, indicating that the L1.1 region mediated either binding to preformed 80S or 60S joining. Furthermore, mutations in L1.1 altered the position of the ribosome on the mutant IRES, indicating that the tRNA-like anticodon/codon mimic within the ribosomal P-site is disrupted. Structural studies have revealed that the L1.1 region interacts with the L1 stalk of the 60S subunit, which is similar to the interactions between the T-loop of the E-site tRNA and ribosomal protein rpL1. Our results demonstrate that the conserved L1.1 region directs multiple steps in IGR IRES-mediated translation including ribosome binding and positioning, which are functions that the E-site tRNA may mediate during translation normally.  *  A version of this chapter has been published in the Journal of Molecular Biology, and has been adapted with permission. Jang, C. J., Lo, M. C. and E. Jan. (2009). Conserved element of the dicistrovirus IGR IRES that mimics an Esite tRNA/ribosome interaction mediates multiple functions. Journal of Molecular Biology. 387, 4258. 36  2.2  Introduction As stated in the previous sections, the positive single-stranded RNA virus of the  Dicistroviridae family possesses the simplest IRES found to date, which is exemplified by the prototypical IRES found within the IGR of CrPV. Briefly, this IRES can directly recruit 40S and 80S ribosomes without the requirement for initiation factors or the initiator MettRNA (Sasaki and Nakashima, 1999; Sasaki and Nakashima, 2000; Jan and Sarnow, 2002; Pestova and Hellen, 2003). Moreover, the IGR IRES sets the translational reading frame by occupying the P-site of the ribosome, and positioning the start codon of the IRES, a GCU alanine codon, in the ribosomal A-site. The ribosome is then ready for delivery of the first aminoacyl-tRNA, Ala-tRNAAla to the empty A-site (Jan et al., 2003; Pestova and Hellen, 2003). This is distinct from the translation initiation of most mRNAs in which an initiator Met-tRNA occupies the ribosomal P-site (Pestova et al., 2007). Though divergent from other viral IRESs, the simplicity of this tRNA-like IRES offers a powerful model for understanding IRES functions in general and the role of tRNA/ribosome interactions that occur normally during translation. The secondary and tertiary structures of the IGR IRES have also been well studied and are found to be fairly well conserved. The structure of the IGR IRES is composed of an overlapping triple pseudoknot RNA structure consisting of pseudoknots I, II and III (PKI, PKII, and PKIII), which can be divided into two independently folded domains, consisting of PKI and PKII/PKIII (Figure 1.7) (Kanamori and Nakashima, 2001; Jan and Sarnow, 2002; Costantino and Kieft, 2005; Pfingsten et al., 2006; Schuler et al., 2006; Costantino et al., 2008). The PKII/PKIII domain primarily mediates ribosome binding (Jan and Sarnow, 2002; Nishiyama et al., 2003), and PKI positions the ribosome such that the GCU start codon  37  occupies the ribosomal A-site (Jan and Sarnow, 2002). Moreover, the two domains fold independently prior to binding to the ribosome (Costantino and Kieft, 2005). The fact that these domains act independently suggests that different regions of the IRES mediate distinct aspects of its function, and work in concert to manipulate the ribosome. The dicistrovirus IGR IRESs can also be grouped into two classes, according to their secondary structure. The main difference between the two classes is that the class II IRES possesses a larger L1.1 loop region, and an extra stem-loop (SL III) within PKI (Figure 1.7). The functional significance of these differences is not known. Crystal structural analyses of the IGR IRES reveal that PKII and PKIII fold into a compact “core” and place the helical regions of SL IV and SL V adjacent with one another so that the loop regions interact with ribosomal protein S5 (rpS5) and possibly rpS25 of the 40S subunit, in agreement that these interactions are important for 40S binding (Jan and Sarnow, 2002; Costantino and Kieft, 2005; Pfingsten et al., 2006; Nishiyama et al., 2007). On the other hand, the conserved L1.1 loop region of the CrPV IGR IRES appears to interact with the 60S ribosomal subunit and the L1 stalk, in particular rpL1 and H77 of the 28S rRNA, which is the same region of the ribosome with which the T-loop of an E-site tRNA interacts (Nishiyama et al., 2003; Spahn et al., 2004b; Pfingsten et al., 2006; Schuler et al., 2006). Presumably, L1.1 is mimicking this portion of tRNA structure. In support of the idea that L1.1 interacts with the 60S subunit, footprinting analysis indicates that the L1.1 region becomes protected upon 80S binding to the IGR IRES (Nishiyama et al., 2003). Moreover, bulk substitution mutations within L1.1 disrupt 80S assembly, but not 40S binding on the IGR IRES in rabbit reticulocyte lysates (Pfingsten et al., 2006). In the crystal structure, the L1.1 region is disordered, probably due to its dynamic nature (Spahn et al., 2004b; Pfingsten  38  et al., 2006; Pfingsten et al., 2007). Current models propose that the L1.1 region changes from a disordered state to an ordered state upon interaction with the L1 stalk, which subsequently leads to a conformational change in the IRES to position PKI correctly in the Psite of the ribosome (Spahn et al., 2004b; Pfingsten et al., 2006; Costantino et al., 2008). In this study, we further explored the role of the conserved loop L1.1 region in IGR IRES function, demonstrating that mutations within L1.1 disrupt key functions for IRES activity. Since L1.1 is predicted to interact with the L1 stalk, these results likely point to roles of E-site tRNA interactions with the ribosome.  39  2.3 2.3.1  Materials and methods DNA constructs and reagents The dicistronic and the monocistronic luciferase plasmids containing the CrPV (gi:  21321708) and TSV (gi: 14701764) IGR IRESs have been described previously (Jan and Sarnow, 2002; Cevallos and Sarnow, 2005). NSC119889 was provided by the NCI/DTP Open Chemical Repository (http://dtp.nci.nih.gov). 2.3.2  In vitro transcription and translation Plasmids containing dicistronic and monocistronic luciferase were linearized with XbaI  and EheI, respectively. The EheI restriction site cleaves 33 nucleotides downstream of the start ATG codon of the firefly luciferase gene. Monocistronic and dicistronic RNAs were in vitro transcribed using a T7 RNA polymerase reaction, and the RNA was purified by RNeasy kit (Qiagen). The integrity and purity of the transcribed RNAs were confirmed by gel analysis. Uncapped dicistronic RNAs were incubated in RRL with 154 mM (final concentration) potassium acetate. The expression of luciferase protein was measured by incorporation of 35Smethionine/cysteine and analyzed by SDS-PAGE. Gels were dried and analyzed by phosphorimager analysis (Typhoon, GE Life Sciences). 2.3.3  40S and 60S subunit purification Ribosomal subunits were purified from HeLa cell pellets (National Cell Culture Centre)  as described (Jan and Sarnow, 2002). In brief, 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,  40  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 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 concentration of 40S and 60S subunits was determined by spectrophotometry, using the conversions 1 A260 nm=50 nM for 40S and 1 A260 nm=25 nM for 60S subunits. 2.3.4  Non-denaturing gel mobility shift assays Conditions for native 40S gel shift assays were adapted as described (Jan and Sarnow,  2002). The 5′ end-labeled RNAs (0.5 nM final concentration) were incubated in buffer E (20 mM Tris (pH 7.5), 100 mM potassium chloride, 2.5 mM magnesium acetate, 0.25 mM spermidine, 2 mM DTT) plus 50 ng/μl of non-competitor RNA at room temperature. Noncompetitor RNA represents nucleotides 880–948 from the pcDNA3 vector. 40S gel shifts were performed in 0.5% (w/v) agarose gels in THEM buffer (66 mM HEPES, 34 mM Tris-HCl, 2.5 mM MgCl2, 0.1 mM EDTA, final pH 7.5) at room temperature. RNAs were incubated for 20 minutes with increasing amounts of 40S subunits from 5 nM to 300 nM. Gels were dried and the radioactivity quantitated by phosphorimager analysis. All data was fit for a single binding site between the IGR IRES and 40S using Prism (GraphPad). 40S affinity measurements are based upon at least three independent experiments. Conditions for native 80S gel shift assays were adapted as described (Jan et al., 2003). In brief, 5′ end-labeled RNAs (0.5 nM final concentration) were incubated in buffer E with an  41  increasing amount of 40S subunits from 0.2 to 200 nM, and a two-fold excess of 60S subunits for 20 minutes at room temperature. Incubations were resolved on a composite gel containing 2.75% acrylamide/bis (19:1), 25 mM Tris acetate (pH 7.0), 6 mM KCl, 2 mM MgCl2, 1 mM DTT, 1% sucrose (w/v), 0.5% NuSieve GTG agarose, 0.45% 3-dimethylaminopropionitrile (DMAPN), and 0.045% ammonium persulfate (APS) at 100 V for 1.5 hours at room temperature. Gels were dried and quantitated by phosphorimager analysis. All 80S data were fit for a single binding site between the IGR IRES and 80S using Prism (GraphPad). All results from gel shift assays were confirmed by filter binding analysis, and 80S affinity measurements are based upon at least three independent experiments. For filter binding experiments, incubations were drawn through two successive membranes of nitrocellulose and nylon which were pre-wetted with buffer E in a Bio-Dot filtration apparatus (Bio-Rad). Membranes were then dried, and the radioactivity was quantitated by phosphorimager analysis (Typhoon, GE Life Sciences). Binding affinities of 40S- and 80SIGR IRES complexes were also determined in an alternate buffer (20 mM HEPES, pH 7.4, 2 mM MgCl2, 150 mM KCl, 2 mM DTT), which does not contain spermidine. The buffer conditions did not significantly alter binding affinity measurements. 2.3.5  40S and 80S competition assays 5′ end-labeled IRES RNAs (5 nM final concentration) were incubated in buffer E with  increasing amounts of IRES competitor from 2 to 250 nM and 50 ng/μl of non-competitor RNA. RNAs were then incubated with either purified 40S subunits or both 40S and 60S subunits at a final concentration of 25 nM for 20 minutes at room temperature. Incubations were resolved by gel shift or filter binding analysis, which produced similar results. Data were fit to the Linn-  42  Riggs equation, which describes the competitive binding of two ligands to a protein (Long and Dawid, 1980). The fraction bound, θ, is described as:    [ S ](1   ) K d (1  [C ]) K c   [ R](1   )  (1)  where [S], [R], and [C] are the concentrations of 40S subunits or 80S ribosomes [S], radiolabeled wild-type CrPV IGR IRES [R], and competitor IRES [C], respectively. The dissociation constants are denoted as Kd and Kc for the wild-type CrPV IGR IRES and competitor IRES interaction with either 40S subunits or 80S ribosomes, respectively. Solving for θ in equation 1:   Kd  1  [C ]  [ S ]  [ R ]    K d   2[ R]  K c     2    Kd   K d  [C ]  [ S ]  [ R ]   4[ R][ S ]  Kc     (2)  A competition curve, where θ=f[C], was fitted for the best value of Kc. Furthermore, the apparent relative dissociation constant, Krel, can be calculated, which is the ratio between Kd and Kc. Therefore, a higher Krel implies weaker mutant binding. All filter binding assays were confirmed by gel mobility shift assay as described above, and all Krel values are based upon at least three independent experiments. 2.3.6  RNase T1 enzymatic probing  For RNase T1 cleavage, 0.5 pmol of IGR IRES RNA was incubated with RNase T1 (Fermentas) at a final concentrations of 0.027 unit/μl, in buffer E for ten minutes at 30°C. RNAs were extracted with phenol/chloroform and ethanol precipitated. The 5′ end-labeled primer PrEJ94 5′-GCCTTTCTTTATGTTTTTGGCG-3′, which anneals 26 nucleotides 3′ to C6214, or primer PrEJ69 5′-GCCCTGGTTCCTGGAACAATTGCTT-3′, which anneals 120 nt 3′ to C6214, were annealed to the RNA and cDNA was synthesized by AMV reverse transcriptase (Promega)  43  as described (Stern et al., 1988). cDNA products were analyzed on a 6% (w/v) polyacrylamide/ 8 M urea gel. 2.3.7  Toeprint analysis of ribosomal complexes using purified components  Toeprinting analysis of ribosomal complexes using purified subunits was performed as described (Wilson et al., 2000a). 500 ng of dicistronic IGR IRES RNAs were first annealed with primer PrEJ69 in 40 mM Tris (pH 7.5) and 0.2 mM EDTA by slow cooling from 65 °C to 35 °C. Annealed RNAs were incubated in buffer E with 0.5 mg/mL cycloheximide containing 40S, 60S, or 40S and 60S subunits (final 40 nM). In translocation assays, assembled 80S-IRES complexes were incubated with ATP (1 mM), GTP (0.4 mM), eEF1A (0.05 mg/mL), eEF2 (0.01 mg/mL), and bulk aminoacyl-tRNAs. 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. 2.3.8  Ribosome assembly in rabbit reticulocyte lysates  5′ end-labeled RNAs (0.5 nM final concentration) were incubated with a final concentration of 154 mM potassium acetate, 0.1 mg/mL cycloheximide, 40 units of RiboLock (Fermentas), 40 μM of amino acids, and rabbit reticulocyte lysate (Promega) to a final volume of 10 μL. Samples were incubated for 20 minutes at 30°C, and then layered onto a 10-30% w/w sucrose gradient in buffer E with 5 mM Mg2+. Gradients were centrifuged for 3.5 hours in an SW41 rotor at 36,000 rpm, and fractionated. The radioactivity in each fraction was measured by scintillation counting (PerkinElmer).  44  2.3.9  Ribosome assembly using purified components  5′ end-labeled RNAs (0.5 nM final concentration) were combined in Buffer E with 40 units of RiboLock (Fermentas), 1 pmol of 40S subunits, and 2 pmol of 60S subunits. If required, the initiation factors eIF1, eIF1A, eIF3, eIF5, and eIF5B were added in 5-fold excess, when compared to 60S subunits. All initiation factors were graciously provided by Dr. Jon Lorsch (Johns Hopkins University) and Dr. Chris Fraser (University of California, Davis). Reactions were made up to 25 μL, incubated for 20 minutes at 30°C, and then layered onto a 10-30% w/w sucrose gradient in buffer E with 5 mM Mg2+. Gradients were centrifuged for 3.5 hours in an SW41 rotor at 36,000 rpm, and fractionated. The radioactivity in each fraction was measured by scintillation counting (PerkinElmer). 2.3.10 GTPase assays  11 pmol of IRES RNA was combined with 0.7 pmol of eEF2, 7 pmol of 40S, and 10 pmol of 60S in a buffer of 50 mM Tris, 50 mM KCl, 1 mM MgCl2, 0.2 mM DTT, pH 7.6. A mixture of 12.5 nmol unlabeled GTP and 0.415 pmol of γ-[32P] GTP (PerkinElmer) was added, and the volume was made up to 50 μL. The reaction was incubated at room temperature and 5 μL aliquots at each time point were quenched with 1 μL of 6 M formic acid. Samples were centrifuged at 13,000 rpm for 10 minutes, and 0.25 μL of the sample was spotted on a prewashed PEI-cellulose thin layer chromatography plate (Sigma-Aldrich). Plates were developed in a solution of 0.75 M KH2PO4, pH 3.5 and dried before phosphorimager analysis.  45  2.4 2.4.1  Results Association of human 80S ribosomes with the CrPV IGR IRES  To investigate the binding of 80S ribosomes to the CrPV IGR IRES, we used a composite agarose/acrylamide gel assay (Jan et al., 2003). Addition of increasing amounts of salt-washed human 40S and 60S subunits to radiolabeled CrPV IGR IRES resulted in a slower migrating 80S-IRES complex (Figure 2.1a, top). Analysis of a parallel incubation by sucrose gradient density centrifugation showed that the IRES associated with 80S ribosomes (Figure 2.2). Moreover, disruption of PKI and PKIII inhibited 80S assembly, in agreement with previous reports (Jan and Sarnow, 2002). The 80S ribosome associated with the CrPV IGR IRES with an apparent dissociation constant (Kd) value of 15 (±5) nM, which is consistent with previous measurements using filter binding analysis (Table 2.1) (Nishiyama et al., 2003). A similar Kd was observed when the incubation time for the binding mixtures was varied, indicating that measurements were taken under equilibrium conditions (data not shown). In contrast, the 80S ribosome did not bind to a previously characterized mutant of the CrPV IGR IRES (ΔPKI/PKIII), which contains mutations that disrupt PKI and PKIII (Figure 2.1a, bottom) (Jan and Sarnow, 2002). To confirm the specificity of the 80S-IGR IRES interactions, we monitored 80S assembly on wild-type or mutant ΔPKI/PKIII IRES in a competition assay. As expected, addition of increasing amounts of unlabeled wild-type but not mutant ΔPKI/PKIII IRES competed with radiolabled wild-type IGR IRES RNAs for binding to 80S ribosomes (Figure 2.1b). The ratio between the competitor and the wild-type dissociation constant determined by direct binding experiments (Kd=15 nM), here referred to as Krel (relative dissociation constant) was 0.3 (±0.5) (see section 2.3.5 for details on calculations). The Krel value indicates that the binding affinity observed through competition experiments was much higher than that obtained  46  from direct binding experiments, thus revealing that the competition binding assay is more sensitive than the direct binding approach. Taken together, these data show that the CrPV IGR IRES binds the 80S ribosome with high affinity.  47  Figure 2.1. Gelshifts to assess 80S assembly on the CrPV IGR IRES. (a) Gel mobility-shift assays of 80S/CrPV IGR IRES complexes. Radiolabeled wild-type or mutant ΔPKI/PKIII IRES RNAs were incubated with increasing amounts of purified salt-washed 40S and 60S subunits from 0 to 200 nM, as described in section 2.3.4. (b) Competition titrations by gel mobility-shift assays. Radiolabeled wild-type IRES RNAs and increasing amounts of unlabeled wild-type or mutant ΔPKI/PKIII IGR IRES RNAs from 0 to 250 nM were incubated with 80S ribosomes as described in section 2.3.5. All mixtures were incubated for 20 min at room temperature and then loaded onto the gel. Gels were dried and exposed for phosphorimaging. The free radiolabeled IRES and 80S-IRES complexes are indicated. The amount of free IRES and 80S-IRES complex was quantified by phosphorimager analysis. Representative gel mobility shifts from at least three independent experiments are shown.  48  Figure 2.2. 80S ribosome assembly on the IGR IRES. Radiolabeled wild-type CrPV or mutant ΔPKI/PKIII IGR IRESs were incubated with 40S and 60S subunits in Buffer E as described in section 2.3.7. Mixtures were loaded on a 10-30% sucrose gradient containing Buffer E with 5 mM Mg2+. Shown are the percent of total radioactive counts in each fraction. The top and bottom of the gradient is represented from left to right, respectively. Fractions containing free IGR IRES and 80S ribosomes are indicated.  49  Table 2.1. Summary of properties of L1.1 mutant CrPV IGR IRES. Mutant a  Activity b  WT GAUC6038-6041CUAG GAUC6038-6041DEL UGC6086-6088ACG G6038C A6039U U6040A C6041G G6087C  100 15 ± 3 3±1 21 ± 8 10 ± 2 32 ± 5 75 ± 16 18 ± 4 22 ± 3  40S Binding Kd c (nM) 15 ± 3 19 ± 5 69 ± 22 15 ± 3 -  80S Binding Kd c (nM) 15 ± 5 21 ± 8 19 ± 7 6±1 -  40S Krel d  80S Krel d  0.8 ± 0.5 2 ± 0.5 3 ± 0.5 4 ± 0.6 -  0.3 ± 0.5 n.m. n.m. n.m. 0.2 ± 0.5 n.m. -  40S Toeprint A e (% of WT) 100 75 57 77 44 84 109 63 50  80S Toeprint A e (% of WT) 100 52 31 38 28 66 99 34 36  a  WT indicates wild-type CrPV IGR IRES. Numbering refers the nucleotide position within the CrPV RNA genome. Translational activity of WT and mutant IGR IRES in RRL. Shown is the ratio of Firefly and Renilla luciferase of dicistronic RNAs containing mutant IGR IRESs, normalized to the ratio of WT CrPV IGR IRES. Average values (± s.d.) are shown from at least three independent experiments. c 40S and 80S binding affinities were obtained by direct-binding assays as described in section 2.3.4. Kd measurements were quantified by phosphorimager analysis and data was fitted to a one-site binding model using Prism. Average values (± s.d.) are shown from at least three independent experiments. d 40S and 80S Krel values were obtained by competition experiments as described in section 2.3.5. Data were fit to the Linn-Riggs equation to obtain the dissociation constant for the competitor IRES using a nonlinear least-squares method. The apparent relative dissociation constant, Krel, is shown, which is the ratio between the wild-type and competitor IRES dissociation constants. Therefore, a higher Krel indicates weaker competitor binding. Average values (± s.d.) are shown from at least three independent experiments. e The intensity of toeprint A (CA6226-7) of 40S and 80S ribosomes bound to the indicated IGR IRES. The intensity of toeprint A was measured as the fraction of radioactive counts in toeprint A within the total radioactive counts within each lane, normalized to the fraction measured for toeprint A of the WT CrPV IGR IRES. Shown are toeprint intensities measured in Figure 2.7. A similar trend in toeprint intensities was observed from a second independent experiment b  50  2.4.2  The conserved loop region of the CrPV IGR IRES is essential for IRES activity  Structural studies predict that the conserved L1.1 loop region of the CrPV IGR IRES interacts with rpL1 within the 60S subunit (Pfingsten et al., 2006; Schuler et al., 2006). From the secondary structure model, the L1.1 region is divided into a 5' L1.1A and a 3' L1.1B strand (Figure 2.3a). L1.1A consists of nucleotides 6035 to 6042 (5'-UGUGAUCU-3') and L1.1B consists of nucleotides 6086 to 6090 (5'-UGCUA-3'). A previous report showed that multiple mutations within L1.1A or L1.1B disrupted IRES activity (Pfingsten et al., 2006). To investigate in more detail whether specific conserved nucleotides play a role in IRES function, deletions and multiple or single nucleotide substitutions were introduced into the L1.1 region (Figure 2.3a). For substitution mutations, the nucleotides were mutated to their base complement. The translational activities of mutant IRESs were tested in rabbit reticulocyte lysates (RRL) using dicistronic RNAs that contained either wild-type or mutant IRESs within the intercistronic region. Deletion of nucleotides within L1.1A (GAUC6038-6041DEL) or multiple nucleotide substitutions within L1.1A and L1.1B (GAUC6038-6041CUAG and UGC6086-6088ACG) inhibited IRES activity by 80-97%, compared to the wild-type IRES (Figure 2.3b, Table 2.1). Single base substitutions within L1.1A and L1.1B inhibited IRES activity to varying degrees. Individual substitutions within residues G6038C, A6039U, C6041G, and G6087C substantially inhibited IRES activity to the same degree as the multiple substitution mutations (GAUC6038-6041CUAG and UGC6086-6088ACG) (>75%), indicating that these residues are essential for IRES activity. In contrast, a nucleotide substitution, U6040A, only moderately inhibited IRES activity (~25%, Figure 2.3b, Table 2.1), demonstrating that U6040 plays a minor role in IRES translation. Collectively, these data indicate that specific nucleotides within the conserved loop region are essential for IGR IRES activity.  51  Figure 2.3. Translational activities of mutant L1.1 IGR IRESs. (a) A close-up view of the mutations introduced in the L1.1 region of the IGR IRES. Underlined nucleotides indicate the mutations introduced. Deletions are denoted with a broken line. (b) Translational activities of mutant CrPV IGR IRES in RRL. Dicistronic RNAs containing wild type or mutant IGR IRESs were incubated in RRL as described in section 2.3.2. The first cistron of the RNA reporter shown in (b) encodes Renilla luciferase and measures scanning-mediated translation. The second cistron, firefly luciferase, measures IGR IRES-mediated translation. The ratio of firefly to Renilla luciferase, normalized to the ratio of a dicistronic RNA containing the wild type IGR IRES is shown. The data shown are the averages of at least three independent experiments ± s.d.  52  2.4.3  Mutations within L1.1 do not affect the secondary structure of the IGR IRES  The inhibition of IRES activity by the L1.1 mutations may be due to changes in the secondary structure of the IRES. To test this, the mutant IRESs studied in Figures 2.3 were examined by enzymatic probing using RNase T1 (Figure 2.4). RNase T1 cleaves 3' of unpaired G residues and can thus detect single-stranded regions within the IGR IRES. The positions of the cleavage sites within the IRES are detected by primer extension analysis using reverse transcriptase. The numbering in Figure 2.5 refers to the nucleotide whose 3' phosphodiester bond was enzymatically cleaved. Overall, the cleavage patterns observed for wild-type and mutant IRESs were similar (Figure 2.4). Specifically, cleavage sites were detected within the predicted single stranded regions of the IRES under equilibrium (Figure 2.4), in agreement with the cleavage patterns reported previously (Jan and Sarnow, 2002). The only differences were the expected cleavages that were introduced by mutations within the L1.1 region. For instance, the GAUC6038-6041CUAG mutations produced a cleavage site at G6041 instead at the G6038, which is normally observed with the wild-type IRES (Figure 2.4, G6038 versus *G6041). Similarly, the GAUC6038-6041DEL deletion mutant IRES did not produce a cleavage site at G6038 and UGC6086-6088ACG revealed a novel cleavage site at G6088, reflecting the mutation within this mutant IRES (*G6088). Therefore, these data support previous findings that mutations within L1.1 do not affect the overall structure of the IRES (Pfingsten et al., 2006), and it is likely that the defect in IRES activity can be attributed to another aspect of IRES function.  53  Figure 2.4. Enzymatic probing of the mutant L1.1 IGR IRESs. Dicistronic RNAs containing the IGR IRES were treated with RNase T1 as indicated. Primer extension was performed as described in section 2.3.6, using either oligonucleotide PrEJ94 (a), which anneals 26 nucleotides 3′ to C6214, or PrEJ69 (b), which anneals 120 nucleotides 3′ to C6214. The reaction products were separated by denaturing PAGE. The nucleotides that are cleaved 3′ by RNAse T1 are indicated to the right. A sequencing ladder of the dicistronic construct using the appropriate primer is shown on the left.  54  2.4.4  Function of mutations within L1.1 in 40S subunit recruitment  Structural and biochemical analyses indicate that mutations in L1.1 inhibit 80S but not 40S assembly on the IGR IRES (Nishiyama et al., 2003; Pfingsten et al., 2006; Schuler et al., 2006). In these studies, 40S and 80S assembly on the mutant IGR IRES was monitored by sucrose gradient analysis (Pfingsten et al., 2006). To determine more precisely which steps in IRES function were disrupted in the mutant L1.1 IRESs, we first tested whether 40S-IRES affinity was affected by using a previously described agarose gel shift assay (Jan and Sarnow, 2002). Incubation of increasing amounts of salt-washed 40S subunits with radiolabeled wild-type CrPV IGR IRES produced a slower migrating 40S-IRES complex (Figure 2.5a, top). The 40S bound to the wild-type IRES with an apparent Kd of 15 (±3) nM, which is consistent with previous reports (Jan and Sarnow, 2002; Nishiyama et al., 2003; Costantino and Kieft, 2005). In contrast, the Kd of the ΔPKI/PKIII mutant IRES was not measurable, indicating that this mutant IRES has low affinity for the 40S subunits (Figure 2.5a) (Jan and Sarnow, 2002). Overall, the mutant L1.1 IRESs bound to 40S subunits with similar affinity as the wild-type IRES (~15 nM) (Figure 2.5a, Table 2.1). The one exception was the GAUC6038-6041DEL mutant IRES which had approximately 4 fold lower affinity for the 40S subunit (Table 2.1). The deletion mutant IRES could not be saturated by 40S binding, indicating that a proportion of this IRES was not able to bind to 40S subunits (data not shown). To confirm these results, we monitored 40S binding by competition assays. Wild-type and mutant L1.1 IRESs competed with the radiolabeled wild-type IRES for 40S binding at a Krel ranging from 1-4 (Figure 2.5b, Table 2.1). The Krel values of GAUC6038-6041CUAG and UGC6086-6088ACG mutant IRESs were approximately 2-3 fold higher as compared to the wild-type IRES (Table 2.1). The relatively high affinity of 40S binding to the mutant IGR IRESs (Krel of 2-4) cannot account for the complete loss in translational activity (>5  55  fold decrease), thus supporting previous findings that mutations within L1.1 do not disrupt 40SIRES assembly and likely affect another aspect of IRES function (Nishiyama et al., 2003; Pfingsten et al., 2006).  56  Figure 2.5. Affinity measurements of 40S-IGR IRES complexes. Gel mobility-shift assays to assess 40S binding to wild-type or mutant IGR IRES RNAs. (a) Radiolabeled RNAs (0.5 nM, denoted below each gel) were incubated with increasing amounts of 40S subunits from 0 to 300 nM as described in section 2.3.4. (b) Quantification of the 40S-IGR IRES complex formation shown in (a). The fraction of IRES bound to 40S is shown plotted against the concentration of 40S. (c) Competition titration experiments. Radiolabeled wild-type IGR IRES and increasing amounts of either unlabeled wild-type or mutant IGR IRES RNAs (denoted below each gel) from 0 to 250 nM were incubated with 40S subunits as described in section 2.3.5. (d) The fraction of radiolabeled wild-type IGR IRES bound to 40S subunits in (c) is shown plotted against the log of IRES concentration. Free IGR IRES is indicated by the arrowhead and the 40S-IRES complex is denoted with an arrow. Representative gels and quantifications from at least three independent experiments are shown.  57  2.4.5  Function of mutations within L1.1 in 80S subunit recruitment  We next addressed whether 80S assembly was affected on these mutant L1.1 IRESs. Based on sucrose gradient analysis, 80S ribosomes cannot assemble on the L1.1 mutant IRES in RRL, indicating that the L1.1 mutations may be affecting either 80S affinity or specifically 60S joining (Pfingsten et al., 2006). To test this, we first monitored the affinity of 80S ribosomes on the IRESs using an electrophoretic mobility shift assay (Jan et al., 2003). Addition of increasing amounts of purified 80S ribosomes with radiolabeled L1.1 mutant IRES resulted in a slower migrating 80S-IRES complex (Figure 2.6a). 80S ribosomes assembled on the L1.1 mutant IRESs with similar affinity (~6-21 nM) to the wild-type IRES (Figure 2.6a, top and Table 2.1). To confirm these results, we assayed for affinity using competition assays. As shown in Figures 2.1b and 2.6c, incubation of increasing unlabeled wild-type but not mutant ΔPKI/PKIII IRES RNAs competed for 80S ribosomes from radiolabeled wild-type CrPV IGR IRES RNAs. Despite having similar affinities to the wild-type IGR IRES through direct binding experiments, the GAUC6038CUAG, GAUC6038DEL, and the AGC6086-6089UCG mutant L1.1 IRES RNAs did not compete for 80S ribosomes (Figure 2.6c, Table 2.1). To further confirm these results, we tested whether mutant IRESs containing single point mutations within L1.1, U6040A and C6041G, compete for 80S ribosomes. The L1.1 U6040A but not the C6041G mutant IRES competed for 80S ribosomes, which correlated with their IRES activities (Table 2.1). The U6040A L1.1 mutant IRES competed for the 80S ribosome with a Krel similar to the wild-type IRES (Table 2.1). Thus, mutations within L1.1 that significantly disrupted IRES activity also reduced 80S binding when monitored by competition experiments. As observed (Figure 2.6), the discrepancy in affinities between the binding assays support the idea that the competition assays are more sensitive in  58  distinguishing 80S-IRES binding affinities and that the binding affinities are likely underestimated using direct binding experiments.  59  Figure 2.6. 80S assembly on mutant L1.1 IGR IRESs. 80S-IGR IRES binding was monitored by gel mobility-shift assays. (a) Radiolabeled wild-type or mutant IGR IRES RNAs were incubated with increasing amounts of 40S and 60S subunits from 0 to 200 nM, as described in section 2.3.4. Phosphorimager analyses of the agarose gels are shown. (b) Quantification of the 80S-IGR IRES complex formation shown in (a). (c) Radiolabeled wild-type IGR IRES and increasing amounts of either unlabeled wild-type or mutant IGR IRES RNAs (denoted below each gel) from 0 to 250 nM were incubated with 80S ribosomes as described in section 2.3.5. (d) The fraction of radiolabeled wild-type IGR IRES bound to 80S subunits shown in (c) is plotted. Representative gels and quantification are shown from at least three independent experiments. 60  2.4.6  Positioning of the 40S and 80S ribosomes on the L1.1 mutant IRES  It has been proposed that the interaction between L1.1 of the IGR IRES and the L1 stalk of the 60S subunit leads to a conformational change in IRES structure to position the CCU triplet within PKI in the P-site of the ribosome (Spahn et al., 2004b; Pfingsten et al., 2006; Costantino et al., 2008). To test this directly, we asked whether mutations within L1.1 affected the position of the 40S and 80S on the IRES. In these experiments, a high concentration of 40S and 80S (100 nM) was used to ensure saturated binding to the mutant L1.1 IRESs as shown in Figures 2.5 and 2.6. Proper ribosome positioning on the IGR IRES is defined by an intact PKI such that the GCU alanine start codon and the preceding CCU triplet occupy the ribosomal A- and P-sites, respectively (Wilson et al., 2000a; Jan and Sarnow, 2002). The mixtures were subjected to “toeprinting” or primer extension analysis using oligonucleotides that hybridize approximately 100 nucleotides 3' to the IRES (see section 2.3.7 for details). By toeprint analysis, two toeprints, termed A and B, were observed at nucleotides CA6226-7 and AA6161-2, respectively, on wild-type IGR IRES bound to 40S or 80S ribosomes (Figure 1.7 and 2.7). The presence of toeprint A (CA6226-7) indicates that the ribosome is properly positioned on the IRES with the CCU triplet and GCU alanine triplet occupying the ribosomal P- and A- sites, respectively (Jan and Sarnow, 2002). In contrast, mutations that disrupt PKI base pairing (i.e. ΔPKI) reduced the intensity of toeprint A but not that of toeprint B, in accordance with the previous finding that disruption of PKI abolishes ribosome positioning and not recruitment (Figure 2.7) (Jan and Sarnow, 2002). Assembled 40S on all but one L1.1 mutant IRES reduced toeprint A intensity by 23-56% and did not significantly affect the intensity of toeprint B (Figure 2.7, Table 2.1, data not shown). The toeprint intensities were measured as a fraction of the total radioactivity in each lane. The presence of toeprint B, which is due to the arrest of reverse transcription 5’ of PKII (Figure  61  1.7A), indicated that the 40S subunit was bound to the mutant IRES and is consistent with the conclusion that these mutations do not affect 40S recruitment. In general, the intensity of toeprint A from assembled 40S-mutant IRES complexes did not correlate with their IRES activities (Figure 2.7, Table 2.1). In contrast, the intensities of toeprint A from 80S/mutant L1.1 IRES complexes correlated with their IRES activities (Figure 2.7, Table 2.1). L1.1 mutant IRESs with significantly disrupted IRES activities (>75% reduction) produced weaker toeprint A intensities of 80S-IRES complexes (Figure 2.7, Table 2.1). Assembled 80S ribosomes on GAUC60386041CUAG,  GAUC6038-6041DEL, UGC6086-6088ACG, G6038C, A6039U, C6041G, and G6087C mutant  L1.1 IRES RNAs reduced toeprint A intensities by 34%-72% as compared to wild-type (Table 2.1). In contrast, assembled 80S on mutant U6040A L1.1 IGR IRES did not reduce the intensity of toeprint A, which correlated with its minor inhibitory effect on IRES activity (Figure 2.3, Table 2.1). Thus, these results suggest that specific nucleotides within L1.1 mediate the position of PKI within the P-site of the ribosome. Given that a portion of 80S ribosomes is positioned properly on the mutant L1.1 IGR IRESs, we asked whether these 80S ribosomes are functional for translocation. Incubation of purified yeast elongation factors 1A, 2, and bulk aminoacyl-tRNAs with assembled 80S/wildtype IGR IRES complexes produced a shift in toeprint A by +6 nucleotides (AC6232-3), indicating that the ribosome translocated by two cycles of elongation (Figure 2.8) (Jan et al., 2003; Pestova and Hellen, 2003). The ribosome translocated by only 6 nucleotides because the mixture was incubated in the presence of cycloheximide, which allows only two cycles of elongation on the IGR IRES (Jan et al., 2003; Pestova and Hellen, 2003). Under these experimental conditions, L1.1 mutations that significantly repressed IRES activity in RRL did not produce Toeprint A or the +6 nucleotide translocation toeprint (Figure 2.8). In contrast, translocation was observed with  62  mutant A6039U and U6040A L1.1 IGR IRESs, which correlated with their IRES activities in RRL. Therefore, these results support the conclusion that particular L1.1 mutations disrupt ribosome positioning on the IGR IRES.  63  Figure 2.7. Toeprint analysis of assembled 40S and 80S ribosomes on mutant L1.1 IGR IRESs. 40S alone or 40S and 60S subunits were incubated with dicistronic RNAs containing wild-type or mutant IGR IRES and analyzed by primer extension analysis using oligo PrEJ69, as described in section 2.3.7. Reaction products were separated by denaturing PAGE. The gels were dried and exposed by autoradiography. The major toeprints, A (CA6226-7) and B (AA6161-2), are 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. The location of the GCU alanine codon that occupies the ribosomal A-site is shown at the left.  64  Figure 2.8. Ribosome translocation on the IGR IRES. Dicistronic RNAs containing wild-type or mutant L1.1 CrPV IGR IRESs were incubated with 40S and 60S subunits, purified elongation factors 1A, 2, and aminoacyl-tRNAs in the presence of 0.5 mg/mL of cycloheximide, as described in section 2.3.7. The reactions were subjected to primer extension analysis to locate the bound ribosome using oligo PrEJ69 and reverse transcriptase. Gels were dried and exposed by autoradiography. The location of toeprint A (CA6226-7) and toeprint (AC6232-3) are indicated to the right. Toeprint A represents the CCU triplet and the GCU triplet in the ribosomal P- and A-site of the ribosome, respectively. Toeprint (AC6232-3) represents a ribosome that has undergone two cycles of elongation.  65  2.4.7  Assembly of 80S on the IGR IRES in RRL  The direct binding experiments suggested that L1.1 mutant IRESs bound to 80S ribosomes with high affinity (Figure 2.6a). However, competition assays indicated that L1.1 mutant IRESs showed reduced affinity for 80S ribosomes (Figure 2.6b). To investigate this further, we tested whether 80S ribosomes could assemble on the mutant IRES in RRL. Using sucrose density gradient centrifugation, we monitored the assembly of 40S and 80S ribosomes on radiolabeled wild-type and mutant IRES RNAs. In the presence of cycloheximide, 80S ribosomes assembled on the wild-type but not on the ΔPKI/PKIII mutant IRES (Figure 2.9a). GAUC6038-6041CUAG, GAUC6038-6041DEL, and UGC6086-6088ACG mutant IGR IRES RNAs, like the ΔPKI/PKIII IRES, failed to bind the 80S ribosome (Figure 2.9a, data not shown). The GAUC6038-6041CUAG mutation is similar to a mutation in the PSIV IRES used in a previous report, which also showed a defect in 80S assembly in RRL (Pfingsten et al., 2006). Thus, 80S ribosomes cannot assemble on mutant L1.1 IGR IRESs in RRL. The defect in 80S assembly in RRL supports our finding that the mutant L1.1 IGR IRESs had reduced affinity for 80S ribosomes when measured by competition binding experiments (Figure 2.6b, Table 2.1). Alternatively, the defect may be due to a specific step in the 80S assembly pathway on the IGR IRES. To examine these possibilities, we used the chemical compound, NSC119889, which inhibits initiator Met-tRNAi binding to eIF2 and likely increases the pool of free 40S subunits (Robert et al., 2006). In this study, the authors showed that addition of this compound to Krebs-2 translation extracts enhanced 80S assembly on the wild-type CrPV IGR IRES, consistent with the notion that this compound increased the pool of available ribosomes (Robert et al., 2006). Incubation of RRL extracts with this compound allowed us to ask whether the increased pool of 80S ribosomes can assemble on the mutant L1.1 IGR IRES.  66  We predict that if there was a defect in affinity for 80S, addition of the compound will still not result in 80S assembly on the mutant IGR IRES. Alternatively, if there was a specific defect in the 80S assembly pathway on the IGR IRES, the increased available pool of 40S subunits may allow 80S assembly on the mutant IGR IRES. Similar to the previous report, addition of NSC119889 in RRL repressed scanning-dependent but not CrPV IRES-dependent translation and increased 80S assembly on the wild-type CrPV IRES in RRL (Figure 2.9b, data not shown) (Robert et al., 2006). Interestingly, addition of NSC119889 dramatically increased 40S binding but not 80S assembly on GAUC6038-6041CUAG and GAUC6038-6041DEL mutant IGR IRES RNAs (Figure 2.9c, data not shown). The increase in 40S binding is consistent with the idea that NSC119889 increased the pool of ribosomal subunits in RRL and agrees with our data that the L1.1 mutations do not affect affinity for 40S. However, even though more subunits were available, the defect in 80S assembly may be due to 60S joining. Alternatively, because we demonstrated that L1.1 mutations affected 80S affinity (Figure 2.6b), the defect may also be due to the reduced affinity of L1.1 mutant IGR IRESs with 80S ribosomes.  67  Figure 2.9. 80S assembly on mutant L1.1 IGR IRESs in RRL. (a) Radiolabeled wild-type or the indicated mutant IGR IRESs were incubated in RRL and then loaded on a 10%–30% (w/v) sucrose gradient, as described in section 2.3.8. (b–d) Radiolabeled wild-type or mutant IGR IRESs were incubated in RRL in the presence (+NSC) or in the absence (–NSC) of 25 μM NSC119889, as described in section 2.3.8. The percentage of total radioactive counts in each fraction is shown. The top and bottom of the gradient are represented from left to right. Fractions containing free IRES and the 40S and 80S ribosomes are indicated.  68  2.4.8  Ribosomal positioning does not dictate 80S assembly on the IGR IRES  During cap-dependent translation, base pairing of the initiator Met-tRNA codonanticodon is a prerequisite for 80S assembly (Pestova et al., 2007). Our results indicated that 40S and 80S ribosomes are not positioned properly on the mutant L1.1 IGR IRESs (Figure 2.7). One hypothesis is that proper ribosome positioning and thus an intact PKI within the ribosomal P-site is a prerequisite for 80S assembly on the IGR IRES. By extension, it is possible that the defect in 80S assembly in RRL on mutant L1.1 mutant IGR IRESs may be due to a defect in ribosome positioning. To address this possibility, we used the ΔPKI IGR IRES which possesses a mutation that disrupts the anticodon-codon base pairing of PKI and cannot position itself correctly within the ribosomal P-site (Jan et al., 2003). In the presence of NSC119889 in RRL, 80S ribosomes readily assembled on the ΔPKI IGR IRES RNA (Figure 2.9d), arguing that ribosome positioning on the IGR IRES is not necessary for 80S assembly.  69  2.4.9  L1.1 of the TSV and IAPV IGR IRES also mediates 80S ribosome assembly  The nucleotide sequences of the L1.1 region of class I and class II IGR IRESs are distinct, yet conserved within each class (compare IRESs in Figure 1.7). Class II IGR IRESs include the TSV and IAPV IGR IRESs (Jan, 2006; Nakashima and Uchiumi, 2008), whereas the CrPV IGR IRES is an example of a class I IRES. The TSV IGR IRES assumes a secondary structure similar to that of the CrPV IGR IRES and can mediate factorless ribosome recruitment and translation initiation (Hatakeyama et al., 2004; Cevallos and Sarnow, 2005; Pfingsten et al., 2007). Given the structural similarities between the two classes of IGR IRESs, we examined whether the L1.1 region of TSV and IAPV IGR IRES functions similarly to that of the CrPV IGR IRES. We mutated a core GUU within L1.1A of the TSV IGR IRES to CAA (TSV GUU6773-6775CAA) and a single G6472U within L1.1B of the IAPV IGR IRES (Figure 2.10a). As shown previously, the translation activity of the TSV IGR IRES was lower compared to the CrPV IGR IRES in RRL (Figure 2.10b) (Hatakeyama et al., 2004; Cevallos and Sarnow, 2005). The IAPV IGR IRES was also active in RRL, demonstrating for the first time that this IGR possesses IRES activity (Figure 2.10b). In contrast, mutant TSV GUU6773-6775CAA and IAPV G6472U IGR IRES activities were significantly inhibited (Figure 2.10b). The defect in translation activity was at 80S assembly. In RRL, 80S ribosomes assembled on wild-type but not on the mutant G6472U IAPV IGR IRES, even in the presence of NSC119889 (Figure 2.10c). Similarly, using competition binding assays, the mutant TSV GUU6773-6775CAA IGR IRES had a 4-fold lower affinity for purified 80S ribosomes when compared to the wild-type TSV IGR IRES (data not shown). Thus, despite nucleotide sequence differences, the L1.1 region of class I and II IGR IRESs function similarly.  70  Figure 2.10. The conserved L1.1 loop region of TSV and IAPV IGR IRES mediates ribosome assembly. (a) The secondary structure of the TSV IGR IRES. Conserved nucleotide positions are shown in uppercase, and nonconserved nucleotides are shown in lowercase. Numbering refers to the nucleotide position within the TSV RNA genome. Helical regions are indicated by a black dash between nucleotides. Underlined nucleotides represent the first amino acid in the viral capsid protein. The conserved loop region is denoted by L1.1A and L1.1B. The inset shows the analogous L1.1 loop region of IAPV IGR IRES. Mutations engineered within the L1.1 region within the TSV and IAPV IGR IRES are indicated by a black box. (b) Translational activity of the mutant TSV and IAPV IGR IRES in RRL, as described in section 2.3.2. The average ratio of radiolabeled firefly luciferase to Renilla luciferase protein normalized to the wild-type CrPV IGR IRES ratio is shown below each lane. The average values and standard deviations are from at least three independent experiments. (c) 80S assembly on mutant IAPV IGR IRES G6472U in RRL. Radiolabeled wild-type or G6472U IAPV IGR IRES was incubated in RRL in the presence (+NSC) or in the absence (–NSC) of 25 μM NSC119889 and loaded onto a 10%–30% (w/v) sucrose gradient, as described in section 2.3.8. The percentage of total radioactive counts in each fraction is shown. The top and bottom of the gradient are represented from left to right. Fractions containing free IRES and the 40S and 80S ribosomes are indicated.  71  2.4.10 Strengthening of PKI compensates for L1.1 mediated translational defects  The IGR IRES PKI domain contains five base pairs which are responsible for setting the ribosome within a specific reading frame by placing the non-AUG start codon within the ribosomal A-site (Figure 1.7A) (Wilson et al., 2000a; Jan et al., 2003; Pestova and Hellen, 2003). We refer to this as a properly positioned ribosome on the IGR IRES. Our results indicate that other parts of the IGR IRES, most notably the L.1.1 domain, may have a moderate role in ribosome positioning on the IGR IRES. How this occurs is not understood. To determine whether the PKI domain acts independently of the effects of the L1.1 domain, we tested whether strengthening PKI base pairing could compensate for translational defects seen with L1.1 mutations. Within the 5 pseudoknotted base pairs of the CrPV IGR IRES, there are 3 A-U base pairs and 2 G-C base pairs. However, in the TSV IGR IRES, there is only one A-U base pair, and 3 G-C base pairs, and 1 non-canonical G-U base pair. We modified PKI in these IRESs by mutating an A-U base pair to a G-C base pair (U6212C/A6191G in the CrPV IGR IRES, and A6069G/U6948C in the TSV IGR IRES), hence “strengthening” PKI by one hydrogen bond (Figure 2.11). Disrupting the base pairing within the CrPV IGR IRES PKI by mutating either U6212C or A6191G inhibited IRES activity (Figure 2.11). However, combining the two mutations and thus restoring base pairing (U6212C/A6191G) increased CrPV IGR IRES translational activity by approximately two-fold when compared to the wild-type CrPV IGR IRES (Figure 2.11). Similarly, distrupting the same A-U base pair by mutating A6069G or U6948C in the TSV IGR IRES PKI domain inhibited IRES activity. Generating the compensatory mutations with a G-C base pair (A6069G/U6948C) also enhanced activity above to that of the wild-type IRES (Figure 2.11). Thus, strengthening PKI leads to a translationally stronger IRES.  72  We then asked whether these “strengthened” mutants could compensate for the mutations within L1.1 which inhibited IRES activity. In agreement with previous data, the GAUC60386041DEL  mutation within the L1.1 region inhibited IRES translation (Figure 2.12). However,  when the PKI strengthening mutation (A6191G/U6212C) was combined with the GAUC60386041DEL  mutation, activity was restored to wild-type levels (Figure 2.12). This suggests that  strengthening the base pairing within PKI can compensate for defects in L1.1. Thus, this suggests that the L1.1 region likely interacts with the 60S subunit through rpL1 (Pfingsten et al., 2006) to stabilize PKI and mediate ribosome positioning. Future experiments will be needed to probe the structure of the PKI region containing the A6191G/U6212C mutation to test whether PKI is structured correctly. Previous results have shown that the related dicistrovirus IRES from PSIV can stimulate the GTPase activity of eEF2 when bound to the ribosome to mediate subsequent translation elongation steps (Yamamoto et al., 2007). This stimulation of GTPase activity is also observed when a deacylated tRNA occupies the ribosome in a P/E hybrid state (Lill et al., 1989; Valle et al., 2003; Zavialov and Ehrenberg, 2003; Sergiev et al., 2005), suggesting that the IGR IRES is able to mimic a P/E hybrid tRNA. We wished to confirm that mutant IGR IRES binding to the ribosome stimulated eEF2dependent GTP hydrolysis. To this end, we used a multiple-turnover assay where we observe the hydrolysis of γ-[32P]-GTP by incubating an excess of IRES with purified 80S ribosomes and eEF2. The fraction of released phosphate was resolved by thin layer chromatography and analyzed by phosphorimager analysis. Reactions containing the CrPV IGR IRES, ribosomes, eEF2, and GTP stimulated GTP hydrolysis (Figure 2.13). Reactions lacking the IRES and eEF2 (ribosomes and GTP alone) did not stimulate GTP hydrolysis, and reactions containing the non-  73  binding ΔPKI/PKIII mutant IRES showed lower amounts of hydrolysis (Figure 2.13). When we incubated the U6212C/A6191G mutant CrPV IGR IRES which increased IRES translation, GTP hydrolysis was stimulated to the same levels as that of the wild-type IGR IRES (Figure 2.13), thus indicating that strengthening PKI does not appear to perturb the IRES’s ability to mimic a P/E hybrid tRNA.  74  Figure 2.11. Strengthening PKI results in a translationally stronger IRES. At top, dicistronic RNAs containing wild-type or chimeric IGR IRESs were incubated in RRL in the presence of [35S]-methionine, as described in section 2.3.2. The first cistron, encoding Renilla luciferase (Rluc), measures scanning-mediated translation, and the second cistron, firefly luciferase (Fluc), measures IGR IRES-mediated translation. Shown is a representative gel of radiolabeled Fluc and Rluc protein products detected by autoradiography. (A) shows experiments carried out with the CrPV IGR IRES, and (B) shows experiments done with the TSV IGR IRES. At bottom are diagrams of PKI in both the CrPV and TSV IGR IRES. The numbering shown corresponds to the nucleotide numbering in the viral genome. The first amino acid encoded by the IRES is underlined. Regions in PKI where mutations were introduced have been denoted by a box.  75  Figure 2.12. A strengthened PKI compensates for L1.1 mediated translational defects. Dicistronic RNAs containing wild type or mutant IGR IRESs were incubated in RRL as described in section 2.3.2. The first cistron, encoding Renilla luciferase, measures scanning mediated translation and the second cistron, firefly luciferase, measures IGR IRES-mediated translation. The ratio of firefly to Renilla luciferase, normalized to the ratio of a dicistronic RNA containing the wild type IGR IRES is shown. The data shown are the averages of at least three independent experiments ± s.d.  76  Figure 2.13. Strengthening PKI does not affect the stimulation of the ribosome-dependent GTPase activity of eEF2. GTP hydrolysis was monitored by incubation of 80S, eEF2, and the indicated IRES with radiolabeled [γ-32P]-GTP, as described in section 2.3.10. Aliquots of the reaction were quenched over time and then resolved by thin-layer chromatography. The slope of the linear regression lines, which represents the mean percent hydrolysis of radiolabeled GTP per minute, is shown in a bar graph. These values were derived from three independent experiments, and the error bars correspond to a 95% confidence interval. Note that ribosomes alone (-IRES/eEF2) do not stimulate GTPase activity.  77  2.4.11 L1.1 of the IGR IRES affects factor retention on the 40S subunit  After start site selection, the pathway of translation initiation involves the concerted departure of initiation factors from the preinitiation complex to facilitate 60S subunit joining (Unbehaun et al., 2004). Our data showed that increasing the free pool of ribosomes by the addition of the NSC119889 in RRL, the mutant L.1.1 mutant IRESs were still unable to bind 80S ribosomes, but an increase in 40S binding was observed (Figure 2.9). This is in agreement with our data that the L1.1 mutations decreased 80S affinity. However, when we analyzed the binding of the L1.1 CrPV IGR IRES mutant UGC6086-6088ACG to purified 40S and 60S subunits by sucrose gradient analysis, 80S ribosomes were readily assembled on the mutant L1.1 IRESs, albeit to lower amounts when compared to wild-type IRES (Figure 2.14). From this, we hypothesize that the L1.1 region may also mediate another step in IRES translation, following 40S subunit joining. Because ribosomes in translation extracts are likely not free of initiation factors, we initially asked whether the addition of purified initiation factors may inhibit 80S formation on the IGR IRES in our assays. Purified initiation factors, eIF1, eIF1A, eIF3, eIF5, and eIF5B were generously provided by Dr. Jon Lorsch (Johns Hopkins University) and Dr. Chris Fraser (University of California, Davis). In this assay, we incubated purified 40S and 60S subunits with radiolabeled IRES, and then loaded the mixture on a sucrose gradient to fractionate free IRES and IRES-80S complexes. Radioactive IRES was then detected by scintillation counting. As expected, purified 80S ribosomes associate with the wild-type CrPV IGR IRES (Figure 2.14, top panel). Similarly, 80S ribosomes remained bound to the CrPV IRES when the eIFs were added to 80S ribosomes that had been prebound with IRES. However, when 40S subunits were preincubated with initiation factors prior to IRES and 60S addition, we observed a ~50%  78  decrease in 80S association (Figure 2.14, top panel), consistent with the previous report that some of these factors, namely eIF1, eIF1A and eIF3, inhibit subunit joining (Pestova et al., 2004). When we repeated this experiment using the UGC6086-6088ACG L1.1 mutant, we found that preincubating the 40S subunit with the eIFs decreased 80S assembly on the IRES by ~90% (Figure 2.14, bottom panel). However, 40S subunit binding was not affected, which is similar to what was observed in RRL with NSC119889 treatment (Figure 2.9). This suggests that the L1.1 region may have an effect on the pathway of 80S assembly. One possible explanation for these observations is the fact that 80S ribosomes have lower affinity for this mutant IRES as we have shown previously (Figure 2.6). However, purified 80S ribosomes can still bind to the L1.1 mutant IRES using sucrose gradient analysis. Furthermore, when the factors were added after prebinding 80S ribosomes with the IRES, the IRES still remained associated with 80S ribosomes (Figure 2.14, bottom panel). An alternate explanation is that an eIF may still be bound to the 40S subunit that prevents 60S subunit joining. It is well-established that eIF3 is associated with recycled 40S subunits to prevent reassociation with 60S subunits (Thompson and Stone, 1977; Trachsel and Staehelin, 1979; Thomas and Favre, 1980), and thus eIF3 was an attractive candidate for further study. Therefore, we asked whether 80S assembly on the L1.1 mutant IRES was affected by the addition of only recombinant eIF3. In our assays, 80S ribosomes assembled on the wild-type CrPV IGR IRES from purified subunits, in the presence or absence of eIF3 (Figure 2.15A). However, incubation of eIF3 with 40S subunits prior to the addition of the 60S subunit and either the GUAC60386041CUAG  (Figure 2.15B) or the GUAC6038-6041DEL L1.1 mutant IRES inhibited 80S assembly to  varying degrees, and an increase in 40S binding is observed (Figure 2.15C). This suggests that the L1.1 mutants have a defect in the pathway from 40S to 80S assembly, and that eIF3 is  79  inhibiting 80S assembly specifically on the L1.1 mutant IRES. Possible mechanisms for this phenomenon will be discussed later.  80  Figure 2.14. Addition of bulk initiation factors results in a decrease in IRES-80S assembly. 40S subunits were incubated with an excess of initiation factors, as described in section 2.3.9. Radiolabeled (A) wild-type or the (B) UGC6086-6088ACG mutant IGR IRES was then added, along with purified 60S ribosomal subunits. As controls, initiation factors were either not added to the mixture, or added after 60S addition, as indicated. Samples were then loaded onto a 10%– 30% (w/v) sucrose gradient. The fraction of total radioactive counts in each fraction is shown. The top and bottom of the gradient are represented from left to right. 81  Figure 2.15. CrPV IGR IRES L1.1 mutants are defective in 80S assembly in the presence of eIF3. 40S subunits were either incubated with or without an excess of eIF3, and radiolabeled wild-type or the indicated mutant IGR IRES was then added, along with purified 60S ribosomal subunits. Samples were then loaded onto a 10%–30% (w/v) sucrose gradient. The fraction of total radioactive counts in each fraction is shown. The top and bottom of the gradient are represented from left to right. 82  2.5  Discussion  The IGR IRESs of the Dicistroviruses possess several remarkable properties. They adopt two independently folded domains that carry out distinct functions, all of which work in concert to manipulate the ribosome. Structural conservation allows the IGR IRES to mimic a tRNA and occupy the mRNA channel of the ribosome, presumably to emulate authentic tRNA/ribosome interactions. The L1.1 loop region is one of a few conserved elements within the IRES and structural studies have predicted that this region interacts with rpL1, an interaction that mimics E-site tRNA/ribosome interactions. In this study, we have built on previous work on this region of the IRES and have elucidated several properties that are necessary for IRES function. Using mutagenesis, we showed that the conserved nucleotides within the L1.1 region are important for IGR IRES activity (Table 2.1). Mutations that disrupt IRES activity reduced affinity for 80S ribosomes and disrupted ribosome positioning on the IGR IRES. Furthermore, we demonstrated that the distinct L1.1 region of the class II IGR IRESs functions similarly as the class I IRES and also mediate affinity for 80S ribosomes. These results highlight the functional conservation of this L1.1 region of the IRES and by extension, provide insight into the role of E-site tRNA interactions within the ribosome. Our results indicate that the primary defect in the mutant L1.1 IGR IRES is 80S assembly. Using competition binding experiments, we showed that L1.1 mutations significantly reduced 80S affinity for the IGR IRES, which correlated with the loss in IRES activity (Figure 2.6c, Table 2.1). This was most evident with single point mutations within the L1.1A region, U6040A and C6041G. The C6041G mutation significantly inhibited IRES activity and affinity for 80S ribosomes, whereas the U6040A did not. In line with these results, 80S assembly was impaired on mutant L1.1 IGR IRESs in RRL, even in the presence of NSC119889, which  83  presumably increased the available pool of free ribosomes (Figure 2.9). This result supports previous findings that also showed that L1.1 mutations affected 80S binding in RRL (Pfingsten et al., 2006). Interestingly, addition of NSC119889 increased the amount of 40S subunits bound to the L1.1 mutant IRESs, implying that L1.1 has no role in 40S recruitment. This observation is consistent with structural data that the L1.1 region does not interact with the 40S subunit (Pfingsten et al., 2006; Schuler et al., 2006). The results also suggest that the L1.1 mutant IGR IRESs are defective in 60S joining, which assumes that the pathway of 80S assembly on the IGR IRES occurs via 40S binding then 60S joining (Jan et al., 2003). However, previous data have also shown that a preformed 80S ribosome can bind to the IGR IRES (Pestova et al., 2004). It remains to be further investigated whether the L1.1 region mediates a unique 80S assembly pathway on the IGR IRES. It is interesting to note that L1.1 mutants are still able to associate with 80S ribosomes from purified 40S and 60S subunits when analyzed by sucrose gradients (Figure 2.14). However, the L1.1 mutants appear to be unable to mediate 80S association in the presence of eIF3, when compared with the wild-type IGR IRES (Figure 2.15). This observation suggests that in addition to a direct effect on affinity for ribosomes, the L1.1 region may play a role in an event downstream of 40S subunit binding. One possibility is that the wild-type IRES is able to tolerate eIF3 binding to the 40S subunit, whereas the L1.1 mutants cannot. Another possibility may be that eIF3 dissociates from the IRES-40S complex in the presence of wild-type IRES to allow for 60S subunit joining, but is unable to do so in the presence of L1.1 mutant IRESs. From these observations, we propose that L1.1 plays a role in the modulation of factors that are bound to the 40S subunit, along with its role in 80S affinity.  84  The L1.1 region is disordered in the crystal structure of the IGR IRES, suggesting that this region is dynamic and becomes structured upon binding to the ribosome (Pfingsten et al., 2006; Kieft, 2008a; Kieft, 2008b). The L1.1 region is predicted to interact with rpL1 and helices H76 and H77 of the rRNA, which are regions of the ribosome that normally interact with the T loop of the E-site tRNA (Pfingsten et al., 2006; Schuler et al., 2006). Interestingly, nucleotide A6014 within the L1.1 region of the PSIV IGR IRES becomes protected from chemical modifications only upon binding to 80S ribosomes (Nishiyama et al., 2003). Moreover, substitution mutations within L1.1 of the PSIV IRES disrupted 80S assembly in RRL (Pfingsten et al., 2006). In our study, we showed that single point mutations within the L1.1 region (including the equivalent nucleotide A6014 of the PSIV IGR IRES) significantly inhibited IRES activity and reduced 80S affinity (Figure 2.6c, Table 2.1), suggesting that key conserved nucleotides of the L1.1 region may mediate specific interactions with rpL1 or H76 and H77 of the rRNA. These interactions may be important because rpL1 is universally conserved and is part of a highly dynamic structure of the 60S subunit called the L1 stalk, which includes H76-78 of the rRNA (Gomez-Lorenzo et al., 2000; Valle et al., 2003; Spahn et al., 2004a; Fei et al., 2008). Cryo-EM images of ribosomes in the unbound and elongation factor (EF) G-bound state reveal large 20 Å movements of the L1 stalk toward the intersubunit space, suggesting that the L1 stalk interacts with the E-site tRNA during translocation (Frank and Agrawal, 2000; Spahn et al., 2004a). Recent data indicates that after peptide bond formation, the L1 stalk interacts with the deacylated tRNA which spontaneously fluctuates between classic P-site and P/E hybrid states before locking into the P/E hybrid state when EF-G binds (Fei et al., 2008). Furthermore, it has been proposed that the L1 stalk actively assists in the translocation of the deacylated tRNA from the ribosomal P- to the E-site, and in the release of the E-site tRNA (Gomez-Lorenzo et al.,  85  2000; Valle et al., 2003; Spahn et al., 2004a; Fei et al., 2008). Given that the L1.1 region of the IGR IRES mediates 80S binding and is predicted to interact with the L1 stalk, the ribosomal L1 stalk may interact with L1.1 to 'lock' the IRES within the mRNA channel of the ribosome, thus contributing to the tight affinity of the IRES with the ribosome. Whether the L1 stalk may also facilitate movement of the IRES during translocation remains to be investigated. Unlike cap-dependent translation initiation, the first translocation step mediated by the IGR IRES is not accompanied by peptide bond formation (Sasaki and Nakashima, 1999; Sasaki and Nakashima, 2000; Wilson et al., 2000a; Wilson et al., 2000b; Jan et al., 2003; Nishiyama et al., 2003; Pestova and Hellen, 2003). This suggests that the IRES must set the ribosome into an elongation competent state. Crystal structure analysis revealed that PKI is tilted when bound to the ribosome, which resembles a P/E hybrid tRNA (Costantino et al., 2008). Also, IGR IRES binding to 80S ribosomes induces eEF2-dependent GTPase activity, which is also observed when a tRNA is bound in at P/E hybrid state in E. coli ribosomes (Lill et al., 1989; Valle et al., 2003; Zavialov and Ehrenberg, 2003; Sergiev et al., 2005; Yamamoto et al., 2007). Thus, it has been proposed that the IRES may adopt an elongation competent P/E tRNA-like hybrid conformation when bound to the ribosome, and possibly that the L1 stalk interacts with L1.1 to prime the ribosome into this translocation-ready state (Pfingsten et al., 2006; Yamamoto et al., 2007). The L1.1 mutations examined in this study did not have an obvious defect in the translocation step, but rather a defect in ribosome positioning and 80S affinity (Figure 2.8). It remains to be tested whether other nucleotides within L1.1 may affect other ribosome functions. During the course of this study, we found that measurements of binding affinity varied according to the experimental approach. In our hands, the competition experiments were more sensitive in distinguishing binding affinities and direct binding experiments underestimated the  86  affinity of 40S-IRES and 80S-IRES complexes. A similar scenario has been observed with previous reports, which proposed that differences between relatively tight associations can only be distinguished by competition binding assays (Long and Dawid, 1980; Elliott et al., 1999). In summary, this study is an example that the type of binding experiment has to be taken into consideration when analyzing IRES/ribosome associations. Our results indicated that the L1.1 mutations may affect the position of the ribosome on the IGR IRES. Normally, the IGR IRES sets the translational reading frame through a tRNA-like anticodon-codon mimic in the ribosomal P-site (Wilson et al., 2000a). Using a high concentration of ribosomes to saturate IRES binding, we found that mutations within L1.1 inhibited ribosome positioning on the IGR IRES and thus disrupted the translational reading frame (Figure 2.7). Mutations within L1.1 reduced the intensity of toeprint A of 80S-IRES complexes, suggesting that the integrity of PKI was compromised. Thus, the L1.1 region maintains the translational reading frame of the IRES likely by mediating the base pairing of the tRNA-like anticodon-codon mimic of PKI within the ribosomal P-site. Alternatively, the helical structure within the stem of PKI may be destabilized in 80S complexes bound with L1.1 mutant IRESs. In support of this, it has been shown that PKI undergoes conformational changes upon 80S binding on the IGR IRES (Pfingsten et al., 2006; Costantino et al., 2008). In addition, the observation that mutations in PKI which appear to strengthen its interaction, appear to compensate for L1.1 mutant defects (Figure 2.13). It is known that the base pairing in PKI is dynamic in nature (Jan and Sarnow, 2002), and strengthening this interaction may “lock” the IRES in a particular reading frame, thus supplanting the requirement for L1.1’s role in reading frame maintenance. Nevertheless, these observations suggest that there is some type of functional link between these two regions.  87  How can mutations within L1.1 affect ribosome positioning and the integrity of PKI? One possibility is that the L1.1 region in the ribosomal E-site interacts with the L1 stalk to induce conformational changes within the ribosome to affect the state of PKI in the P-site. In support of this, it has been shown that the presence of a deacylated tRNA in the E-site is important to maintain the translational reading frame of a P-site tRNA (Marquez et al., 2004). Moreover, the presence of a P/E hybrid tRNA can affect the stability of the codon/anticodon Psite stem (McGarry et al., 2005). Finally, it has been well documented that an E-site tRNA can have long-distance effects on aminoacyl-tRNA binding affinities in the A-site (Geigenmuller and Nierhaus, 1990; Blaha and Nierhaus, 2001; Nierhaus, 2006). These results are consistent with the idea that the L1.1 region may induce allosteric effects in the ribosome by mimicking properties of the E-site tRNA to mediate the integrity of PKI in the ribosomal P-site (Costantino et al., 2008). It remains to be investigated whether particular conformational changes within the ribosome contribute to this property of the L1.1 region of the IGR IRES. An interesting finding was that the L1.1 region of class I and II IGR IRESs, though distinct between the classes but conserved within each class, is functionally similar. This suggests a functional conservation of specific L1.1 nucleotides that interact with the L1 stalk. Given the predicted dynamic nature of L1.1, the conserved RNA structure of both classes of IGR IRES probably adopts a conformation that exposes the L1.1 region to the L1 stalk. It will be interesting to investigate whether the L1.1 regions are interchangeable between the two classes, thus providing insight into the functional conservation of this key region of the IGR IRES. Future investigations into the mechanisms of the IGR IRES will undoubtedly provide further insight into the roles of tRNA interactions with the ribosome during translation normally.  88  3 3.1  Modularity of the Dicistroviridae intergenic internal ribosome entry site † Summary  The IGR IRES of the Dicistroviridae viral family can directly assemble 80S ribosomes and initiate translation at a non-AUG codon in the ribosomal A-site. These functions are directed by two independently folded domains of the IGR IRES. One domain, composed of overlapping pseudoknots II and III (PKII/III), mediates ribosome recruitment. The second domain, composed of PKI, mimics a tRNA anticodon-codon interaction to position the ribosome at the ribosomal A-site. Although adopting a common secondary structure, the Dicistroviridae IGR IRESs can be grouped into two classes based on distinct features within each domain. In this study, we report on the modularity of the IGR IRESs and show that the ribosome binding domain and the tRNA anticodon mimicry domain are functionally interchangeable between the Type I and the Type II IGR IRESs. Using structural probing, ribosome binding assays, and ribosome positioning analysis by toeprinting assays, we show that the chimeric IRESs fold properly, assemble 80S ribosomes, and can mediate IRES translation in rabbit reticulocyte lysates. We also demonstrate that the chimeric IRESs can stimulate the ribosome-dependent GTPase activity of eEF2, which is suggestive that the ribosome is primed for a step downstream of IRES translation. Overall, the results show that the dicistrovirus IGR IRESs are composed of two modular domains that work in concert to manipulate the ribosome and direct translation initiation.  †  A version of this chapter has been published in RNA, and has been adapted with permission.  Jang, C. J. and E. Jan. (2010). Modular domains of the Dicistroviridae intergenic internal ribosome entry site. RNA. 16, 1182-1195. 89  3.2  Introduction  Phylogenetic analyses have revealed that all Dicistroviridae IGR IRESs adopt a similar secondary structure consisting of three overlapping pseudoknots, PKI, PKII, and PKIII (Figure 1.7) (Jan, 2006; Kieft, 2008a; Nakashima and Uchiumi, 2008). PKII and PKIII form one domain that folds into a compact core that is responsible for ribosome binding (Wilson et al., 2000a; Jan and Sarnow, 2002; Nishiyama et al., 2003; Pfingsten et al., 2006; Pfingsten et al., 2007), whereas PKI mimics an anticodon tRNA stem-loop to mediate ribosome positioning such that the start non-AUG codon of the IRES occupies the ribosomal A-site (Wilson et al., 2000a; Wilson et al., 2000b; Kanamori and Nakashima, 2001; Costantino et al., 2008). Previous reports indicate that the domains are functionally independent. First, disruption of PKI does not affect ribosome assembly on the IRES (Jan and Sarnow, 2002; Nishiyama et al., 2003). Second, the PKII/PKIII domain alone can fold independently and bind to ribosomes (Nishiyama et al., 2003; Costantino and Kieft, 2005). This has led to the hypothesis that the two domains of the IRES interact with different regions of the ribosome to direct IRES translation. Structural studies show that the IRES primarily occupies the P- and E-sites of the ribosome, confirming that the IRES mimics a tRNA (Spahn et al., 2004b; Pfingsten et al., 2006; Schuler et al., 2006). The CrPV PKI domain resembles that of a tRNA anticodon loop and appears tilted at an angle that is similar to that of a deacylated tRNA in a P/E hybrid state on the ribosome (Yamamoto et al., 2007; Costantino et al., 2008). It has been proposed that the P/E hybrid conformation of the IRES may help mediate the next steps of translation such as the delivery of aminoacyl-tRNA and translocation (Yamamoto et al., 2007; Costantino et al., 2008). In support of this, the IRES can stimulate ribosome-  90  dependent GTPase activity of eEF2 (Yamamoto et al., 2007), an effect that is also observed when a deacylated tRNA occupies the ribosome in a P/E hybrid state (Lill et al., 1989; Valle et al., 2003; Zavialov and Ehrenberg, 2003; Sergiev et al., 2005). It is unclear whether the proposed P/E hybrid conformation contributes to IRES activity. Although adopting an overall similar secondary structure, the IGR IRESs can be grouped into two classes based on distinct features within the PKII/III and PKI domains (Jan, 2006; Nakashima and Uchiumi, 2008). The type II IRESs, such as the Taura syndrome virus (TSV) IGR IRES, contain a longer L1.1 region and an extra stem-loop, SL III, within the PKI domain, when compared to the type I IRESs, such as the Cricket paralysis virus (CrPV) IGR IRES (Figure 1.7). We and others have shown that mutations within the L1.1 region of both CrPV and TSV IGR IRESs can inhibit IRES activity and disrupt 80S assembly, suggesting that some of the features that are different between the type I and type II IRESs possess common functions (Pfingsten et al., 2006; Jang et al., 2009; Pfingsten et al., 2010). Although both types of IRESs can mediate factorindependent translation initiation, the significance of the features that are non-conserved between the type I and II IRESs are poorly understood (Hatakeyama et al., 2004; Cevallos and Sarnow, 2005). In this study, we have used a chimeric mutagenesis approach to address whether the two distinct domains of the type I and II IRESs are modular. Furthermore, we have explored the biochemical properties of the different structural elements within each IRES type. Our results demonstrate that the IGR IRES is composed of functional modular domains that independently direct distinct ribosomal interactions and functions.  91  3.3 3.3.1  Materials and methods Purification of eukaryotic elongation factor 2  A yeast strain expressing a histidine-tagged version of eEF2 was generously provided by Dr. Terri Kinzy (UMDNJ). 4 L of yeast cells were grown in appropriate drop-out media to an optical density of 2.0. Cells were harvested by centrifugation at 5,000 rpm for 5 minutes, and resuspended in two volumes of lysis buffer (50 mM KPO4, 10 mM imidazole, 1 M KCl, 0.2 mM PMSF, and 1% v/v Tween-20, pH 8.0). Cells were then lysed by three passes through a French press (Thermo) and centrifuged at 20,000 g for 20 minutes. The supernatant was collected and centrifuged at 100,000 g for 1.5 hours. The supernatant was filtered through 3 MM Whatman paper and loaded onto a 1 mL Ni2+ chelating column which was pre-equilibrated with wash buffer (50 mM KPO4, 20 mM imidazole, 1 M KCl, 0.2 mM PMSF, 1% v/v Tween-20, pH 8.0). The column was washed with 10 column volumes of wash buffer and eluted with a step gradient using 20 column volumes of elution buffer (50 mM KPO4, 500 mM imidazole, 1 M KCl, 0.2 mM PMSF, 1% v/v Tween-20, pH 8.0). The fractions containing eEF2 were identified by SDS-PAGE, and these fractions were dialyzed against 4 L of dialysis buffer (20 mM Tris, 100 mM KCl, 0.1 mM EDTA, 10% glycerol, 1 mM DTT, 0.2 mM PMSF, pH 7.5) three times for four hours each time, before storage at -80°C. 3.3.2  80S filter binding competition assays  5′ end-labelled IRES RNAs (5 nM final concentration) were incubated in buffer E with increasing amounts of IRES competitor from 2 to 250 nM and 50 ng/μl of noncompetitor RNA (Jan and Sarnow, 2002). RNAs were then incubated with preformed 80S ribosomes at a final concentration of 12.5 nM for 20 minutes at room temperature.  92  Incubations were drawn through two successive membranes of nitrocellulose and nylon, which were pre-wetted with buffer E in a Bio-Dot filtration apparatus (Bio-Rad). Membranes were then dried, and the radioactivity was quantitated by phosphorimager analysis (Typhoon, GE Life Sciences). Data were fit to the Linn-Riggs equation, which describes the competitive binding of two ligands to a protein (Long and Dawid, 1980). The fraction bound, θ, is described as:    [ S ](1   ) K d (1  [C ]) K c   [ R](1   )  (1)  where [S], [R], and [C] are the concentrations of 80S ribosomes, radiolabeled wild-type CrPV IGR IRES, and competitor IRES, respectively. The dissociation constants are denoted as Kd and Kc for the wild-type CrPV IGR IRES and competitor IRES interaction with 80S ribosomes. Solving for θ in equation 1:   Kd 1    K d   2[ R]   Kc    [C ]  [ S ]  [ R ]    2    Kd   K d  [C ]  [ S ]  [ R ]   4[ R][ S ]  Kc     (2)  A competition curve, where θ=f[C], was fitted for the best value of Kc. All Kc values obtained are based upon at least three independent experiments. 3.3.3  Toeprinting analysis of ribosomal complexes in RRL  Toeprinting analysis of ribosomal complexes in RRL was done in the same manner as described in section 2.3.7, where annealed RNAs were incubated in 10 µL of RRL with 20 µM amino acid mix, 154 nM potassium acetate, and 10 µM edeine. Ribosomal complexes were analyzed by primer extension analysis using AMV reverse transcriptase in the presence of α-[32P] dATP (3000 Ci/mmol, PerkinElmer). (Jan and  93  Sarnow, 2002) cDNA products were analyzed on 6% (w/v) polyacrylamide/8 M urea gel. Gels were dried and analyzed by phosphorimager analysis.  94  3.4 3.4.1  Results 80S ribosome assembly on the chimeric IGR IRES  Previous studies have indicated that the PKII/III and PKI domains of the IGR IRES are functionally independent and mediate different ribosomal activities (Kanamori and Nakashima, 2001; Jan and Sarnow, 2002; Costantino and Kieft, 2005; Pfingsten et al., 2006; Costantino et al., 2008; Jang et al., 2009). To investigate this further, we constructed chimeric IRESs where the PKI domain was swapped between the type I (CrPV) and type II (TSV) IGR IRESs (Figure 1.7). The point at which the domains were swapped is a single-stranded variable linker, which is not predicted to be structured. For this study, we will refer to the wild-type IGR IRESs as CrPV and TSV IRES and the chimeric IRESs as CrPVII/III-TSVI (PKII/PKIII of CrPV IRES fused with PKI of TSV IRES) and TSVII/III-CrPVI (PKII/PKIII of TSV IRES fused with PKI of CrPV IRES). We have shown that the CrPV and TSV IRESs have similar affinities for purified 80S ribosomes (Jang et al., 2009). To investigate binding to ribosomes further, we used a competitive filter-binding assay to monitor 80S assembly on wild-type or chimeric IRESs (Figure 3.1). Specifically, we asked whether the addition of excess unlabeled IRES can compete with radiolabeled wild-type CrPV IRES for 80S ribosomes. As shown previously, increasing amounts of unlabeled wild-type CrPV or TSV IRES but not a nonbinding ΔPKI/PKIII CrPV IGR IRES competed with radiolabeled wild-type CrPV IGR IRES for 80S ribosome binding (Figure 3.1, Table 3.1) (Jan and Sarnow, 2002; Jang et al., 2009). Curve fitting analysis yielded an apparent dissociation constant (Kd) of 8 ± 1 nM for the wild-type CrPV IRES and 25 ± 3 nM for the wild-type TSV IRES, which is consistent with previously published results (Table 3.1) (Nishiyama et al., 2003; Jang et  95  al., 2009). The chimeric IGR IRESs also competed for 80S binding with an apparent dissociation constant of 14 ± 2 nM for the TSVII/III-CrPVI IRES and 20 ± 2 nM for CrPVII/III-TSVI IRES (Table 3.1). In general, these results demonstrated that the 80S binding to the wild-type and chimeric IRESs was relatively tight (Kd 8-25 nM range), thus suggesting that the PKII/III domain of the chimeric IRESs is folded properly for 80S assembly and supports previous findings that the PKII/PKIII domain primarily mediates ribosome recruitment.  96  Figure 3.1. 80S assembly of the chimeric IGR IRESs. 80S-IGR IRES binding was monitored by competition titration experiments as described in section 3.3.2. The fraction of radiolabeled wild-type CrPV IGR IRES bound to 80S ribosomes is plotted against the log of the indicated unlabeled IRES RNA. Representative curves and quantitations are shown from at least three independent experiments.  97  Table 3.1. Properties of mutant IGR IRESs. Mutant a Activity (%) 80S Binding Kd c (nM) Purified 40S Toeprint A d Purified 80S Toeprint A e RRL Toeprint A f WT CrPV 100 8±1 + + ++ WT TSV 45 ± 5 25 ± 3 + + + TSVII/III - CrPVI 29 ± 4 14 ± 2 + CrPVII/III - TSVI 93 ± 6 20 ± 2 + + ++ TSV ΔSL III 6±3 25 ± 2 CrPV + TSV L1.1A 23 ± 3 60 ± 3 NM NM NM TSV + CrPV L1.1A 3±1 256 ± 17 NM NM NM CrPV + TSV L1.1A/B 57 ± 1 11 ± 1 NM NM NM TSV + CrPV L1.1A/B 2±1 74 ± 7 NM NM NM a  WT indicates wild-type IGR IRES.  b  Translational activity of WT and mutant IGR IRESs in RRL. Shown is the ratio of firefly and Renilla luciferase of dicistronic RNAs containing mutant IGR IRESs, normalized to the ratio of WT CrPV IGR IRES. Average values ± standard deviations are shown from at least three independent experiments. c  80S binding affinities were obtained by competition experiments as described in section 3.3.2. Data was fit to the Linn–Riggs equation to obtain the dissociation constant for the competitor IRES using a nonlinear least-squares method. Average values ± standard deviations are shown from at least three independent experiments. d–f  The intensity of toeprint A using either purified ribosomal subunits or RRL with the indicated IRES mutant. The intensity of toeprint A was measured as the fraction of radioactive counts in toeprint A, and represented symbolically: “+” indicates that a toeprint is present; “++” indicates an enhanced toeprint; and “-” indicates no toeprint. Mutants that were not measured for toeprints are denoted as NM. Shown are toeprint intensities measured in Figure 4. A similar trend in toeprint intensities was observed from a second independent experiment.  98  3.4.2  PKI of the chimeric IGR IRESs folds independently  To determine whether the PKI domain of the chimeric IRESs is folded properly, we probed the structure of the IRESs by using RNase T1. RNase T1 cleaves 3' of unpaired G residues and can thus detect single-stranded regions within the IGR IRES. The positions of the cleavage sites within the IRES were detected by primer extension analysis using reverse transcriptase. The numbering in Figure 3.2 refers to the nucleotide whose 3' phosphodiester bond is enzymatically cleaved. Overall, the cleavage patterns observed for wild-type and chimeric IRESs were similar (Figure 3.2). Specifically, we observe cleavage sites at G6183, GG6188-6189, G6192, G6195 and G6206 within PKI of both the wild-type CrPV IRES and the chimeric TSVII/III-CrPVI IRES, suggesting that the PKI domain is folded similarly. Previously observed RNase T1 cleavage sites that are present within paired regions, such as GG6188-6189 and G6195, illustrate the dynamic nature of PKI (Jan and Sarnow, 2002). All RNase T1 cleavages observed were also consistent with previously published results (Jan and Sarnow, 2002). Furthermore, we also observed a cleavage site at GGG6905-6907 within PKI of both the wild-type TSV IRES and the chimeric CrPVII/III-TSVI IRES. If PKI was not folded properly within this chimera, we would have expected cleavages at GG6937-6936 and GG6932-33, which are normally in paired regions. In summary, the probing data of the chimeric IRES indicates that PKI folds independently and similarly to PKI of the wild-type IRESs, when compared to previously published data.  99  Figure 3.2. Enzymatic probing of the chimeric IGR IRESs. Dicistronic RNAs containing the wild-type and chimeric IGR IRESs were treated with RNase T1 as indicated. Primer extension was performed as described in section 2.3.6. The reaction products were separated on a denaturing polyacrylamide gel. The nucleotides that are cleaved 3′ by RNase T1 are indicated on the right. A sequencing ladder of the dicistronic construct using the appropriate primer is shown on the left.  100  3.4.3  Translational activity of the chimeric IRES  Since our data showed that the chimeric IRESs fold properly and can bind to 80S ribosomes with relatively similar affinities as the wild-type IRESs (Figures 3.1, 3.2, Table 3.1), we next investigated if IRES activity is affected. Dicistronic luciferase RNAs containing either the wild-type or the chimeric IRESs were incubated in rabbit reticulocyte lysates (RRL) in the presence of [35S]-methionine. Translation reactions were analyzed by SDS-PAGE and the amount of radioactivity incorporated in Renilla (scanning-dependent translation) and firefly (IRES translation) luciferase was quantified by phosphorimager analysis. As shown previously, translation directed by the TSV IRES was weaker (~45%) when compared to the CrPV IRES (Figures 3.3A, 3.3B, Table 3.1) (Hatakeyama et al., 2004; Cevallos and Sarnow, 2005; Jang et al., 2009). The hybrid CrPVII/III-TSV1 IRES activity was slightly less active (93%) compared to the wild-type CrPV IRES (Figures 3.3B, Table 3.1), indicating that the PKI domain of TSV can be swapped into the CrPV IRES without significantly affecting IRES translation. In contrast, the TSVII/III-CrPVI IRES activity was significantly reduced (29%) compared to the wild-type CrPV IRES. However, when compared to the weaker TSV IRES activity (here given as 100%), the TSVII/III-CrPVI IRES activity was only reduced to ~64% activity, demonstrating that swapping in the PKI domain of the CrPV IRES into the TSV IRES still retained IRES activity (Table 3.1). Interestingly, the CrPVII/III-TSVI IRES has ~2.5 fold lower affinity for the ribosome compared to the wild-type CrPV IRES, yet both IRESs displayed similar IRES activities (Table 3.1, Kd 8 nM vs 20 nM). Furthermore, the reduced translation of the TSVII/III-CrPVI IRES is not attributable to differences in 80S affinities as the chimeric TSVII/III-CrPV IRES had higher 80S binding affinity compared to that of the TSV IRES (Table 3.1, Kd 14 nM vs 25 nM).  101  Thus, the relatively small differences in 80S binding affinities of these IRESs do not correlate with the IRES translational activities. In summary, these results indicate that the PKI domains are functionally interchangeable between the CrPV and TSV IRESs. A previous report showed that CrPV IRES translation is resistant to the compound NSC119889, which disrupts the interaction between eIF2 and Met-tRNAi (Robert et al., 2006). Thus, this compound can be used to distinguish IRESs like the EMCV IRES that mediate internal initiation in an eIF2-dependent manner from ones that are eIF2-independent (Novac et al., 2004). We tested whether the chimeric IRES translational activities are resistant to the NSC119889 compound. Addition of NSC119889, as expected, reduced the translation of Renilla luciferase, indicating that scanning-dependent translation is inhibited (Figure 3.3C, lower panel). In contrast, the activities of the wild-type CrPV IRES, TSV IRES, and CrPVII/III-TSVI IRESs were relatively resistant to the effects of the compound at 10 and 20 µM (Figures 3.3A, 3.3B, upper panel). Interestingly, the chimeric TSVII/III-CrPVI IRES is not resistant to the effects of NSC119889 (Figure 3.3C). To examine this further, we tested the effects of the antibiotic edeine on IRES activity. It has been shown that the IGR IRES is also insensitive to edeine, which inhibits AUG codon recognition by the 40S-eIF2GTP/Met-tRNAi complex (Kozak and Shatkin, 1978; Wilson et al., 2000a; Dinos et al., 2004). As expected, addition of edeine inhibited scanning-dependent translation of Renilla luciferase (Figure 3.3D). In contrast, translation by both wild-type and chimeric IRESs was resistant to the effects of edeine (Figure 3.3D). In summary, these data indicate that the chimeric IRESs can direct internal initiation in an eIF2-independent manner similar to the wild-type IRESs.  102  Figure 3.3. Translational activities of the chimeric IGR IRESs. (A) Uncapped dicistronic RNAs containing wild-type or chimeric IGR IRESs were incubated in RRL in the presence of [35S]-methionine, as described in section 2.3.2. The first cistron, encoding Renilla luciferase (Rluc), measures scanning-mediated translation, and the second cistron, firefly luciferase (Fluc), measures IGR IRES-mediated translation. Shown is a representative gel of radiolabeled Fluc and Rluc protein products detected by autoradiography. Where applicable, the amount of the ternary complex inhibitor NSC119889 added to the reactions is shown above the gel. (B) Quantitations of translational activities of the chimeric IRESs. The ratios of firefly to Renilla luciferase are shown and are normalized to the ratio of the dicistronic RNA containing the wild-type CrPV IGR IRES. (C) Normalized quantitation of the chimeric IRES translation under NSC119889 treatment. Firefly (top) and Renilla (bottom) luciferase activities were normalized to the translational activity of each dicistronic RNA in the absence of NSC119889. The data shown are the averages of at least three independent experiments ± standard deviations. (D) Translational activities of the chimeric IRESs in the presence of edeine. Shown is a representative gel of radiolabeled Fluc and Rluc detected by autoradiography. The bottom panel shows quantitations of Rluc and Fluc, normalized to the amount of luciferase produced by each dicistronic RNA in the absence of edeine. The average value from three independent experiments ± standard deviations is shown. 103  3.4.4  Positioning of the IGR IRES chimeras correlates with their translational  activities  Upon 80S assembly on CrPV and TSV IRESs, the ribosome is positioned on the IGR IRES such that the GCU alanine start codon and the preceding CCU triplet occupy the ribosomal A- and P-sites, respectively (Figure 1.7) (Wilson et al., 2000a; Jan and Sarnow, 2002; Cevallos and Sarnow, 2005). To determine whether the 80S ribosomes are positioned properly, we performed toeprinting assays (i.e. primer extension assays) on dicistronic RNAs containing chimeric IRESs bound to purified 40S or 80S ribosomes. Purified ribosomes bound to the wild-type CrPV IRES produced two toeprints, toeprint A and B, that are observed at nucleotides CA6226-7 and AA6161-2, respectively (Figure 3.4A, left panel). Likewise, toeprints A and B were observed for ribosome-bound wild-type TSV IRES at UU6963-4 and A6876, respectively (Figure 3.4A, right panel). The presence of toeprint A is indicative of proper ribosome positioning, which is +13-14 nucleotides downstream of the CCU triplet of both IRESs, given that the first C is +1. Thus, the ribosome is properly positioned on the wild-type IRESs with the CCU triplet and GCU alanine triplet occupying the ribosomal P- and A- sites, respectively (Wilson et al., 2000a; Jan and Sarnow, 2002). Toeprint B represents a contact within the core PKII/PKIII domain of the IRES with the ribosome that impedes the reverse transcriptase in the primer extension analysis (Jan and Sarnow, 2002). Mutations that disrupt PKI base pairing (i.e. ΔPKI) eliminated toeprint A but not that of toeprint B (Figure 3.4A, left panel), in accordance with the previous finding that disruption of PKI abolishes ribosome positioning but not recruitment (Jan and Sarnow, 2002).  104  For the chimeric IRESs, the presence of toeprint A, in general, correlated with their translational activities in RRL. Specifically, the chimeric TSVII/III-CrPVI lacked toeprint A, whereas CrPVII/III-TSVI IRES produced toeprint A (Figure 3.4A). The presence of toeprint B was not significantly affected in both chimeras suggesting that the ribosome is bound to the chimeric IRES and supports the conclusion that these mutations do not affect ribosome recruitment (Figure 3.1). It is interesting to note that although toeprint A is not produced, the TSVII/III-CrPVI IRES translation is weakly active in RRL (Figures 3.4A, 3.3B). It is possible that toeprint A cannot be detected by the toeprinting assay when the translational activity is weak. To examine this further, we investigated the position of the ribosome on these IRESs in RRL under conditions identical to those in which the IRES activities were measured in Figure 3.3. Incubation of dicistronic RNAs containing the wild-type CrPV IRES in RRL in the presence of edeine produced a strong toeprint at CA6226-7 (toeprint A), which represents a properly positioned ribosome on the IRES (Figure 3.4B). The addition of edeine at this concentration (10 µM) inhibits the delivery of an aminoacyl-tRNA to the ribosomal Asite and can inhibit IGR IRES-dependent translocation (Carrasco et al., 1974; Wilson et al., 2000a). Incubation of dicistronic RNAs containing the TSV IRES produced a toeprint at UU6963-4 (toeprint A), which is similar to that observed when purified 80S ribosomes are bound (Figures 3.4A, 3.4B). However, the TSV IRES produced a weaker toeprint A (39%) compared to that of the CrPV IRES (100%), which correlates with the reduced translational activity of the TSV IRES (Figures 3.3B, 3.4B). As expected, disruption of the base pairing within PKI reduced toeprint A on the TSV and CrPV IRESs (Figure 3.4B, ΔPKI lanes). For the chimeric IRESs, incubation of these dicistronic RNAs in RRL produced the same toeprint A as their wild-type counterparts (Figure 3.4B). However, the intensity of toeprint A on the  105  chimeric TSVII/III-CrPVI IRES is reduced as compared to that on the CrPVII/III-TSVI IRES (38% vs 121%), which is the same trend observed with their translational activities (Figure 3.4B, Table 3.1). Thus, given the relatively similar 80S binding affinities between the chimeric IRESs, it appears that the intensity of the toeprint A reflects the translational activity of the IRESs. In support of this conclusion, similar intensities of toeprint A are observed on the CrPVII/III-TSVI and the wild-type CrPV IRESs (100 vs 121%), which display similar IRES activities (Figures 3.3B, 3.4B). In summary, our results show that the activity of the chimeric IRES correlates with the presence of toeprint A in RRL, and thus proper positioning of the ribosome.  106  Figure 3.4. Ribosome positioning on the chimeric IGR IRESs by toeprinting analysis. Dicistronic RNAs containing wild-type or mutant IGR IRES were incubated with (A) 40S alone or 40S and 60S subunits as described in section 2.3.7, or (B) in RRL in the presence of edeine as described in section 3.3.3. Mixtures were analyzed by primer extension analysis using oligo PrEJ69. Products were separated by denaturing polyacrylamide gels. The gels were dried and exposed by autoradiography. The location of major toeprint A and toeprint B, are shown on the right. Sequencing ladders for the wild-type IRESs are shown on the left, with their respective nucleotide numbers as indicated. The intensity of toeprint A in B is shown as a percentage of the total radioactivty in each lane, normalized to the wild-type CrPV IGR IRES.  107  3.4.5  Modularity of the conserved L1.1A/B region of the IGR IRESs  The conserved L1.1 region consisting of L1.1A and L1.1B of the IGR IRESs is conserved within each IRES type (Figure 1.7, color coded in purple and brown) (Jan, 2006; Nakashima and Uchiumi, 2008). Although L1.1 is different between the two classes, mutations within this region reduce IRES translation and can disrupt ribosome recruitment and positioning, suggesting that the L1.1 region of both IRES types mediate common functions (Pfingsten et al., 2006; Jang et al., 2009; Pfingsten et al., 2010). To test this directly, we asked whether the L1.1 region is interchangeable by swapping the L1.1A or both L1.1A/B regions between the CrPV and TSV IRESs. When the L1.1A or L1.1A/B of the TSV IRES was inserted into the CrPV IRES (CrPV + TSV L1.1A and CrPV + TSV L1.1A/B), IRES translational activity was preserved, albeit reduced when compared to the wild-type CrPV IRES (Figures 3.5A, 3.5B, compare lanes 1, 4, and 6, Table 3.1). In contrast, translational activity was abolished in the chimeric TSV IRES containing the L1.1A or both L1.1A/B of the CrPV IRES (Figures 3.5A, 3.5B, lanes 5 and 7, Table 3.1). When we assayed for 80S binding, the TSV IRES containing the swapped CrPV L1.1A or L1.1A/B had significantly lower affinity for 80S ribosomes (~4-6 fold lower) when compared to the CrPV IRES containing the swapped TSV L1.1A or L1.1A/B regions (Figure 3.5C, Table 3.1). Since the CrPV IRES with the TSV L1.1 region is still functional and can bind to 80S ribosomes, this chimeric CrPV + L1.1A/B IRES is likely folded properly to mediate IRES activity. In contrast, the TSV IRES cannot tolerate insertion of the CrPV L1.1 region suggesting that the TSV + CrPV L1.1 chimeric IRESs are defective. The nature of this defect remains to be elucidated.  108  Figure 3.5. The activity of L1.1 IGR IRES chimeras. (A) Dicistronic RNAs containing wild-type or chimeric IGR IRESs were incubated in RRL in the presence of [35S]methionine, as described in section 2.3.2. The first cistron, encoding Renilla luciferase (Rluc), measures cap-mediated translation, and the second cistron, firefly luciferase (Fluc), measures IGR IRES-mediated translation. Shown is a representative gel of radiolabeled Fluc and Rluc protein products detected by autoradiography. The IRES mutants tested are labeled at the top of the gel. (B) Amounts of Fluc and Rluc made are quantified below the gel, shown as a ratio between Fluc and Rluc. Results were normalized against wild-type CrPV IGR IRES. (C) The fraction of radiolabeled wild-type CrPV IGR IRES bound to 80S ribosomes is plotted against the log of the indicated unlabeled IRES concentration. Representative quantifications are shown from at least three independent experiments.  109  3.4.6  The IGR IRES stimulates the ribosome-dependent GTPase activity of eEF2  The GTPase activity of EF-G can be stimulated when a deacylated-tRNA occupies the ribosome in a P/E hybrid state (Lill et al., 1989; Zavialov and Ehrenberg, 2003; Sergiev et al., 2005). The IGR IRES of the dicistrovirus Plautia stali intestine virus (PSIV) can also stimulate the GTPase activity of eEF2 when bound to the ribosome, thus suggesting that the IGR IRES may mimic a P/E tRNA hybrid (Yamamoto et al., 2007; Costantino et al., 2008). Here, we examined whether the chimeric IGR IRESs can stimulate the ribosome-dependent GTPase activity of eEF2. Towards this, we monitored the multiple turnover of γ-[32P]-GTP by incubating an excess of IRES and eEF2 with purified 80S ribosomes. The fraction of released phosphate was resolved by thin layer chromatography and analyzed by phosphorimager analysis. Incubation of wild-type CrPV or TSV IRES with ribosomes and eEF2 stimulated GTP hydrolysis to similar levels (Figure 3.6). This stimulation was specific as incubations containing only ribosomes (-IRES/-eEF2) did not stimulate GTPase activity and reactions containing only ribosomes and eEF2 (-IRES) or ribosomes/eEF2 with a mutant ΔPKI/PKIII IRES, which does not bind to ribosomes (Jan and Sarnow, 2002), stimulated GTP hydrolysis to half the level of the wild-type IRESs (Figure 3.6). This trend was also observed with varying amounts of eEF2 indicating that the GTPase stimulation of eEF2 is dependent on IRES binding to the ribosome (data not shown). Similar to what was observed for the PSIV IRES (Yamamoto et al., 2007), the ΔPKI mutant of the CrPV IRES stimulated GTPase activity to the same extent as the wild-type IRESs, indicating that an anticodoncodon interaction within PKI is not required for GTPase activation (Figure 3.6). Collectively, these data point to PKII/III and the PKI stem of the IRES as the main determinants for stimulating ribosome-dependent eEF2 GTPase activity.  110  Similar to the wild-type IRESs, the chimeric IRESs also stimulated the GTPase activity of eEF2, but to varying degrees (Figure 3.6). Specifically, the chimeric CrPVII/IIITSVI IRES stimulated GTPase activity to the same extent as the wild-type CrPV and TSV IRESs. In contrast, the chimeric TSVII/III-CrPVI IRES only slightly stimulated GTPase activity over background (Figure 3.6). It is possible that this reduced GTPase activity is due to the TSVII/III-CrPVI IRES not being in the optimal P/E hybrid state on the ribosome. Alternatively, because the multiple turnover of γ-[32P]-GTP was monitored, the lower GTPase activity detected may be due to inhibition of another step in the eEF2 cycle (i.e. release of eEF2-GDP from the ribosome). Previous reports have indicated that the L1 stalk of the 60S subunit may interact with the deacylated tRNA in the P/E hybrid state and may assist in the translocation of the deacylated tRNA (Valle et al., 2003; Spahn et al., 2004b; Fei et al., 2008). Because the L1.1 region of the IGR IRES is predicted to interact with the L1 stalk, one hypothesis is that this region may direct the GTPase activity of eEF2 and thereby assist in the first translocation step. A mutant CrPV IRES containing a GACU6038-6041CUAG mutation, which has been shown to inhibit IRES translation, displayed similar GTPase stimulation as the wild-type IRESs (Figure 3.6) (Jang et al., 2009), thus suggesting that the L1.1 region is not required for stimulating the GTPase activity of eEF2.  111  Figure 3.6. Ribosome-dependent GTPase stimulation of eEF2 by the chimeric IGR IRESs. GTP hydrolysis was monitored by incubation of 80S, eEF2, and the indicated IRES with radiolabeled [γ-32P]-GTP, as described in section 2.3.10. Aliquots of the reaction were quenched over time and then resolved by thin-layer chromatography. The slope of the linear regression lines, which represents the mean percent hydrolysis of radiolabeled GTP per minute, is shown in a bar graph. These values were derived from three independent experiments, and the error bars correspond to a 95% confidence interval. Note that ribosomes alone (-IRES/-eEF2) do not stimulate GTPase activity.  112  3.4.7  Deletion of SL III in the TSV IGR IRES abolishes 80S positioning  One main difference between the two classes of IGR IRESs is an extra stem-loop, SL III, within PKI of the type II IRESs. Deletion of SL III or mutations within SL III inhibit IRES-driven translational activity, but can still assemble 80S ribosomes in RRL, suggesting that SL III may mediate a step downstream of 80S binding (Hatakeyama et al., 2004; Pfingsten et al., 2007). To explore this further, we assayed whether deletion of SL III within the TSV IRES affected ribosome binding, positioning, and ribosome-dependent eEF2 GTPase activity. As shown previously, translation of the ΔSL III TSV IRES was severely inhibited but this mutant IRES could still bind to 80S ribosomes in RRL (Figures 3.7A, 3.7B) (Pfingsten et al., 2007). Moreover, the addition of NSC119889 did not affect 80S binding on the mutant IRES (Figure 3.7B). To examine this more closely, we monitored 80S binding by using the competition binding assay. The ΔSL III IRES bound to purified 80S ribosomes with similar affinity as the wild-type TSV IRES, with a Kd of 25 ± 2 nM, which is consistent with the observation that this mutation does not affect 80S assembly (Table 3.1) (Pfingsten et al., 2007). By toeprinting analysis, 40S and 80S ribosomes assembled on the wild-type TSV IRES and produced toeprint A at UU6963-4, but not on the ΔSL III TSV IRES (Figure 3.7C), indicating that the presence of SL III is required for proper ribosome positioning. When we monitored the ribosome-dependent GTPase activity of eEF2, the stimulation of GTP hydrolysis by the ΔSL III TSV IRES was found to be similar to that of the wild-type TSV IRES (Figure 3.6). Therefore, despite a lack of proper ribosomal positioning, the ΔSL III mutant can still activate the GTPase activity of eEF2, further supporting the conclusion that GTPase activation is independent of the PKI domain of the IGR IRES.  113  Figure 3.7. Characterization of SL III within the TSV IGR IRES. (A) Dicistronic RNAs containing wild-type or ΔSL III TSV IGR IRESs were incubated in RRL in the presence of [35S]-methionine, as described in section 2.3.2. The first cistron, encoding Renilla luciferase (Rluc), measures scanning-mediated translation, and the second cistron, firefly luciferase (Fluc), measures IGR IRES-mediated translation. Shown are radiolabeled luciferase protein products detected by autoradiography. (B) 80S assembly on TSV and ΔSL III IGR IRESs was assessed in RRL by sucrose gradient analysis, as described in section 2.3.8. Radiolabeled wild-type or ΔSL III IGR IRES was incubated in RRL with cycloheximide in the presence (+NSC) or absence (−NSC) of 25 μM NSC119889. Incubations were loaded on a 10%–30% sucrose gradient and shown are the percent of total radioactive counts in each fraction. Fractions containing free IRES, 40S, and 80S ribosomes are indicated. (C) Toeprint analysis of assembled 40S and 80S ribosomes on TSV and ΔSL III IGR IRESs, as described in section 3.3.3. 40S alone or 40S and 60S subunits were incubated with dicistronic RNAs containing wild-type or mutant IGR IRES and analyzed by primer extension analysis using oligo PrEJ69. Products were separated in denaturing polyacrylamide gels, dried, and exposed by autoradiography. 114  3.5  Discussion  The type I and II IGR IRESs are comprised of two independently folded domains; a PKII/III ribosome binding domain and a PKI tRNA anticodon mimicry domain that recruit, position, and set the ribosome in an elongation-competent mode. While structurally similar, each domain of the type I and II IRESs possesses distinctive features when compared to one another, and their functions are poorly understood. This study demonstrates that the ribosome binding domain and the tRNA anticodon mimicry domain of the CrPV and TSV IGR IRESs are in general functionally interchangeable and thus modular. The results suggest that these modular domains mediate similar interactions and functions within the ribosome to direct factorless IRES translation. Many biological systems often use modularity to realize functionality (Tang and Breaker, 1998; Wierenga, 2001). Modularity also allows recombination of domains to evolve new RNAs or proteins with novel functions. An example of this is found in the universally conserved RNase P RNA, which processes precursor tRNAs to produce mature tRNA. RNase P RNA consists of two independently folded modular domains; the C- and S-domains which are responsible for catalysis and substrate recognition, respectively (Loria and Pan, 1996; Torres-Larios et al., 2006). It has been shown that mitochondrial RNase P RNAs from jakobid flagellates, which normally cannot process tRNAs on their own, can become active when the S-domain of E. coli RNase P RNA is swapped in (Seif et al., 2006). This chimeric approach has been informative as it suggests that over time the jakobid flagellates may have lost structural components within their S-domains and now rely on RNase P proteins for catalytic activity (Seif et al., 2006).  115  In the case of the dicistrovirus IGR IRES, it is attractive to envision that the IRES itself evolved or is “constructed” from the recombination of a ribosome binding RNA and a tRNA anticodon-like domain. Previous structural and biochemical studies are consistent with this idea. PKII/PKIII and PKI domains can fold independently and direct specific interactions with the ribosome (Jan and Sarnow, 2002; Nishiyama et al., 2003; Costantino and Kieft, 2005; Nishiyama et al., 2007; Pfingsten et al., 2007). Moreover, it has been shown that the PKII/PKIII domain alone can bind to the ribosome (Nishiyama et al., 2003; Costantino and Kieft, 2005). In this study, our results provide further support to this idea as the chimeric IRESs are translationally active and the PKII/PKIII and PKI domains are functionally interchangeable (Figure 3.3A, Table 3.1). The hybrid CrPVII/III-TSVI IRES is more translationally active than the TSVII/IIICrPVI IRES (Figure 3.3, Table 3.1). This is not attributed to significant differences in ribosome binding (Figure 3.1, Table 3.1) or defects in RNA folding (Figure 3.2). It is possible that the overall structure of the hybrid TSVII/III-CrPVI IRES may not allow for maximal IRES translation. Alternatively, because the CrPV IGR IRES and the chimeric IRES containing the CrPVII/III domain are translationally more active than the TSV IGR IRES and the chimeric IRES containing the TSVII/III domain, this suggests that the CrPV PKII/III domain may confer higher translational activity (Figure 3.3B). It has been proposed that an interaction of the PKII/III domain with the 60S subunit induces a structural change that propagates through the IRES to the PKI domain to properly position the ribosome (Yamamoto et al., 2007; Jang et al., 2009; Pfingsten et al., 2010). In support of this idea, mutations in the conserved L1.1 region of PKII/III can disrupt the anticodon-codon interaction within PKI, suggesting a functional linkage between the two domains (Jang et al.,  116  2009). Furthermore, our data indicate that 80S ribosome positioning is disrupted in the translationally weaker TSVII/III-CrPVI IRES chimera (Figure 3.4), which is consistent with the idea that a conformational structural change induced by an IRES/ribosome interaction is transmitted from PKII/PKIII to PKI. Thus, an interdomain signal within the CrPVII/III-TSVI chimera may activate IRES translation, whereas structural incompatibilities within the TSVII/III-CrPVI chimera may result in uncoupling of the two domains which prevents the successful transduction of this signal, thus resulting in weaker IRES activity. However, although RNase T1 probing analysis suggests that the secondary structure of the CrPV PKI is intact in the unbound state (Figure 3.2), we cannot rule out that the CrPV PKI domain does not fold properly when the hybrid TSVII/III-CrPVI IRES is bound to the ribosome. One domain that may be important for this allosteric interdomain signal is the L1.1 region, which is conserved within each type of IGR IRES and is predicted to interact with the L1 stalk of the 60S subunit (Jan and Sarnow, 2002; Pfingsten et al., 2006; Schuler et al., 2006; Pfingsten et al., 2007). We have previously demonstrated that mutations within the L1.1 domain of several IGR IRESs disrupt 80S binding and IRES translation, suggesting that all L1.1 domains of this viral family may function similarly (Jang et al., 2009). Interestingly, only the hybrid CrPV IRES containing the swapped L1.1.A/B region of the TSV IRES retained IRES activity (Figure 3.5A, 3.5B). Because the L1.1 region is flanked by two helices, the TSV L1.1A/B region inserted into the CrPV IRES likely interacts with the L1 stalk in the same manner as in the wild-type IRES. In contrast, swapping the L1.1A/B region of the CrPV IRES into the TSV IRES inhibited translation activity, suggesting that this smaller CrPV L1.1A/B region cannot be accommodated within the TSV IRES, possibly distrupting the overall structure of the PKII/III domain (Figure 3.5A, 3.5B). Indeed, the  117  affinity of this hybrid IRES (TSV + CrPV L1.1A/B) for 80S ribosomes was significantly (~4 fold) reduced, when compared to the wild-type IRES and the hybrid CrPV + L1.1A/B IRES (Figure 3.5C, Table 3.1). Our results are consistent with the idea that the IRES mimics a P/E hybrid tRNA to stimulate the ribosome-dependent GTPase activity of eEF2 (Yamamoto et al., 2007; Costantino et al., 2008). Specifically, wild-type and mutant IRESs that can bind to 80S ribosomes with high affinity stimulated GTPase activity (Figure 3.3, 3.6, Table 3.1). This supports the observation that disruption of PKII or PKIII, which inhibits ribosome binding, does not stimulate GTPase activity (Yamamoto et al., 2007). The PKII/PKIII domain is not the only determinant for GTPase stimulation as a PKII/PKIII domain alone partially stimulates GTPase activity (Yamamoto et al., 2007). Finally, our results are in agreement that the anticodon-codon interaction of PKI is not required for GTPase activity (Figure 3.6) (Yamamoto et al., 2007). In summary, it is likely that an IRES with an intact PKII/PKIII and helical stem of PKI, but not the anticodon-codon interaction of PKI, mimics a P/E tRNA hybrid, which can stimulate the GTPase activity of eEF2. How does the IRES stimulate the ribosome-dependent GTPase activity of eEF2? Cryo-EM studies have shown that the IGR IRES induces a conformational change in the ribosome where the stalk region of the 60S subunit consisting of the ribosomal P proteins becomes extended (Spahn et al., 2004b). This stalk region has been shown to be important for recruitment and stimulation of GTPase activation on elongation factors (Diaconu et al., 2005). Therefore, this IRES-induced structural change may be facilitating GTPase activation on eEF1A and/or eEF2 and thereby mediates the delivery of the next aminoacyl-tRNA and/or translocation. Previous reports point to the 3' end of a tRNA in a P/E hybrid state as being  118  responsible for ribosomal GTPase activation (Lill et al., 1989). Since the IRES occupies the ribosomal P- and E- sites, it will be interesting to see whether a specific element within the IRES may be coupling the structural conformation of the stalk region and the ribosomedependent GTPase activity of eEF2. In this study, our results show that mutations in the L1.1 region do not inhibit the GTPase activity of eEF2, suggesting that the interaction between L1.1 of the IRES and the L1 stalk of the 60S is not required for this activity (Figure 3.6). The functional role of SL III in the TSV IGR IRES is unknown, although it has been shown previously that deletion of SL III does not affect 80S assembly on the IRES (Pfingsten et al., 2007). We have expanded on this observation and showed that SL III is predominantly responsible for proper ribosome positioning, which explains why the ΔSL III mutation is translationally inactive (Figure 3.7). It remains to be investigated whether SL III is directly mediating anticodon-codon interactions within PKI or whether deletion of SL III simply disrupts proper folding of PKI. In summary, this report demonstrates that the IGR IRES is composed of two independent modular domains. Although we have only tested the CrPV and TSV IRESs, it is likely that the PKII/III and PKI domains of the other type I and II IGR IRESs are also interchangeable. The modularity of these domains within the Dicistroviridae family implies that these IGR IRESs may have evolved from the recombination of distinct RNA functional domains.  119  4 4.1  Novel pathways of 80S assembly on an IRES mediated by eukaryotic elongation factor 2 ‡ Summary  Eukaryotic elongation factor 2 (eEF2) is a conserved protein that drives the process of translocation during translation elongation. Here, we describe another role for eEF2 in the association and stabilization of 80S ribosomes. To determine whether these eEF2-stabilized 80S ribosomes are functional, we assessed whether the IGR IRES can bind to these ribosomes. Our experiments show that the IGR IRES associates with these eEF2-associated 80S ribosomes. During the course of these experiments, we also made the observation that the IRES binds to 60S subunits only when 80S ribosomes first bind to the IRES. The biological relevance of the ability of eEF2 to associate 80S ribosomes and the ability of the IRES to bind to both the eEF2-associated 80S ribosomes and 60S subunits are discussed.  ‡  A version of this chapter is in preparation for publication in collaboration with Dr. Natalia Demeshkina, Dr.  Shinya Kurata, Dr. Go Hirokawa, and Dr. Akira Kaji.  120  4.2  Introduction  Protein synthesis in all living organisms is a cyclical process that involves the recycling of ribosome couples after chain elongation and termination. In prokaryotes, after translation termination, the post-termination ribosomal complex consists of a 70S ribosome couple, bound to an mRNA with a tRNA present in the P-site. However, it was unclear how the post-termination complex is dissociated for a subsequent round of translation. Historically, it was initially thought that this process occurred spontaneously after translation termination, but later work found that this dissociation event involves the coordinated action of two protein factors: ribosome recycling factor (RRF) and EF-G (Hirashima and Kaji, 1973; Karimi et al., 1999; Gao et al., 2005; Hirokawa et al., 2005; Peske et al., 2005; Zavialov et al., 2005). The manner in which these two proteins mediate the dissociation of the post-termination complex is still under debate, but it is agreed that in prokaryotes, these two proteins are responsible for the recycling of ribosomal subunits for the next round of translation initiation. Since the translational machinery is very well conserved both mechanistically and structurally between prokaryotes and eukaryotes, it was a reasonable assumption that the eukaryotic homologue of EF-G, eEF2 would also have the same role in eukaryotic ribosome recycling. However, recent data obtained in Dr. Akira Kaji’s lab has unexpectedly suggested that eEF2 has a role in 80S ribosome association and stabilization using purified yeast ribosomes. His observations show that in the presence of eEF2 and the non-hydrolyzable GTP analogue GMP-PNP (guanosine 5′-[β,γ-imido]triphosphate), which mediates the stable binding of eEF2 to ribosomes (Nolan et al., 1975), purified 40S and 60S ribosomal subunits are assembled into 80S ribosome couples (Figure 4.1). The biological relevance of this  121  phenomenon is not known, but one possible hypothesis is centered on the events that occur during periods of cellular stress. When eukaryotic cells are subjected to stress conditions such as nutrient deprivation, viral infection, or are present in stationary phase, overall translation is often inhibited. This inhibition is accompanied by the dissociation of polysomes and an accumulation of translation-incompetent 80S couples in the cytosol (Kraig and Haber, 1980; Hensold et al., 1996; Ashe et al., 2000; Uesono and Toh, 2002). It has been hypothesized that eEF2 may mediate 80S association during these periods of time. We were initially interested in this phenomenon since it is known that the IGR IRES is translationally activated during periods of translational shutoff (Wilson et al., 2000b; Fernandez et al., 2002; Lee et al., 2009; Garrey et al., 2010). These observations imply that in addition to recruiting 40S and 60S sequentially, the IGR IRES may also directly recruit these translationally incompetent 80S ribosomes during periods of host translational shutoff. This may provide the rationale as to how the IGR IRES remains translationally active when overall translation is inhibited. However, this hypothesis has yet to be tested. Furthermore, it remains unclear whether the IGR IRES directly binds to 80S ribosomes in general. Past biochemical work has suggested that the IRES, in addition to binding 40S and 60S subunits sequentially, can also bind 80S ribosomes directly (Jan et al., 2003; Pestova et al., 2004). However, this remains an unresolved question. We hypothesize that the IGR IRES can bind to the eEF2-associated 80S ribosome. To test this idea, we have utilized the 80S association effect induced by eEF2 and directly asked whether the IGR IRES can recruit the eEF2-associated 80S ribosome. During the course of these experiments, we have also unexpectedly observed that the IRES itself can bind directly  122  to 60S subunits and that the IRES itself can induce 80S formation. We discuss the biological significance of the 80S/eEF2 complex, and how it may play a role in how the IGR IRES associates with 80S ribosomes in cells during translational shutoff.  123  Figure 4.1. eEF2 induces the association of 40S and 60S subunits into 80S ribosomes in the presence of a non-hydrolyzable GTP analogue. Shown is an OD254 trace obtained from a sucrose gradient separation of 40S and 60S assembly mixtures that were conducted as described in section 4.3.1. This figure was obtained from experiments done by Natalia Demeshkina, and is reproduced here with permission. Yeast ribosomal subunits were combined in buffer 2/150 with eEF2 and varying nucleotides or nucleotide analogues, as indicated. The position of the 40S, 60S, and 80S peaks are marked. Percentages given indicate the total percentage of the subunits found in the 80S peak.  124  4.3 4.3.1  Materials and methods Assembly and analysis of ribosomal complexes  In order to assay for 80S assembly, purified 40S and 60S subunits were combined at a final concentration of 75 nM in buffer 2/150 (20 mM Tris, 2 mM MgCl2, 150 mM KCl, pH 7.5). Recombinant eEF2 was then added to a final concentration of 0.6 µM, along with 0.4 mM of nucleotide or nucleotide analog, and 5 nM of radiolabeled CrPV IRES, as required. Mixtures were incubated at for 20 min at 20°C, and if required, were crosslinked by the addition of glutaraldehyde to a final concentration of 0.45%. Mixtures were then immediately layered onto a 10-30% sucrose gradient and centrifuged in a SW-41 rotor (Beckman) for 3.5 h at 36,000 rpm. The gradient was then fractionated, and the radioactivity in each fraction was measured by scintillation counting (PerkinElmer). eEF2 was precipitated from the fractions by the addition of 1 volume of 100% (w/v) trichloroacetic acid to every 4 volumes of sample. Mixtures were incubated at -80°C for 1 hour and centrifuged to pellet the protein. Pellets were washed twice with 200 μL of ice cold acetone, dried at 95°C, and suspended in SDS loading dye. eEF2 was identified by subsequent Western blotting, using an α-eEF2 antibody that was graciously provided by Dr. Akira Kaji (University of Pennsylvania). The CrPV IRES and 28S rRNA was isolated from the fractions by the addition of 3 mL of 8 M guanidine hydrochloride and 5 mL of 100% ethanol per every 0.75 mL of sample. Mixtures were left overnight at -80°C and then centrifuged at 10,000 rpm in a JA-20 rotor (Beckman). Pellets were suspended in 400 μL of H2O and combined with 1 mL of 100% ethanol and 40 µL of 3 M sodium acetate, pH 5.2. Mixtures were left at -80°C for 1 hour and centrifuged to pellet the RNA. The RNA was then detected by Northern blotting using standard procedures.  125  4.4 4.4.1  Results eEF2 induces the association and stabilization of 80S ribosomes  Past results have shown that the presence of a non-hydrolyzable GTP analogue can mediate the stable binding of eEF2 to the ribosome (Nolan et al., 1975). Recently, our collaborator, Dr. Akira Kaji has also shown that eEF2 and the non-hydrolyzable GTP analogue GMP-PNP can mediate the stabilization of purified yeast 40S and 60S ribosomal subunits into 80S ribosome couples (Figure 4.1). In order to first confirm this finding, we asked whether human ribosomes also have this property. In order to observe subunit association, we incubated purified human 40S and 60S ribosomal subunits with an excess of recombinant yeast eEF2 and GMP-PNP. The mixtures were analyzed by sucrose gradient ultracentrifugation to resolve the 40S, 60S, and 80S ribosomes. We found that the addition of eEF2 with GMP-PNP stimulated the association of 80S ribosomes, similar to what was observed using yeast ribosomes (Figure 4.2B). This association was specific as the addition of GTP did not stimulate 80S assembly (Figure 4.2A). As a control, loading purified 40S or 60S subunits alone on a sucrose gradient also confirmed their relative migration in the gradient (Figure 4.2C and 4.2D), verifying that the new peak that we had observed with eEF2 and GMP-PNP addition contains 80S ribosomes.  126  Figure 4.2. Human 80S ribosomes are associated by the addition of GMP-PNP and eEF2. 80S assembly was assessed by sucrose gradient analysis as described in section 4.3.1. 40S and 60S ribosomal subunits were combined in the presence of eEF2 and either (A) GTP or (B) GMP-PNP. Mixtures were loaded on a 10%–30% sucrose gradient and fractioned. Shown is the trace at an absorbance of 254 nm. (C) 40S and (D) 60S ribosomal subunits were also separated alone on gradients to confirm their migration. Peaks corresponding to 40S, 60S, and 80S ribosomes are indicated.  127  4.4.2  The IGR IRES is bound to 60S subunits following sucrose gradient  centrifugation  We next asked whether the eEF2-stabilized 80S ribosome is competent for IGR IRES binding. In order to do this, 80S ribosomes were assembled in the presence of eEF2 and GMP-PNP (Figure 4.2). Radiolabeled CrPV IGR IRES was then added to the mixture and subjected to sucrose gradient fractionation, where the CrPV IGR IRES was monitored by scintillation counting. Surprisingly, the majority of the IGR IRES was present in the fraction containing the 60S ribosomal subunit, even in the absence of eEF2 and GMP-PNP (Figure 4.3). To verify this finding, we conducted a battery of assays in which radiolabeled CrPV IGR IRES was incubated with purified ribosomal subunits alone (Figure 4.4). As expected and in agreement with previous reports (Jan et al., 2001; Nishiyama et al., 2003), the CrPV IGR IRES bound to 40S subunits but not to 60S subunits. Interestingly, when incubated with 40S and 60S subunits together, the CrPV IRES again associated with 60S subunits, in contrast to these reports. Upon careful analysis, we noted that the gradients used in prior association assays contained 5 mM Mg2+, in contrast to the gradients in Figure 4.3 which contained 2 mM Mg2+ (buffer 2/150). When we used conditions that were used in prior experiments, namely 5 mM Mg2+ in the gradients, the CrPV IGR IRES readily associated with 80S ribosomes (Figure 4.4). Thus, we conclude that the lower magnesium concentrations in our assay did not support IRES binding to the 80S ribosome. Thus, the IRES most likely dissociates from the 80S ribosome during sucrose gradient centrifugation. What is unusual about these results is when the CrPV IGR IRES is incubated with 40S, 60S, eEF2 and GMP-PNP under 2 mM Mg2+, the CrPV IGR IRES binds to the 40S and 60S subunit fractions, even though 80S ribosomes are still assembled (Figure 4.2 and 4.3).  128  This is in stark contrast to when the CrPV IRES is incubated with 60S subunits alone. Under these conditions, the CrPV IGR IRES does not have appreciable affinity for 60S subunits, similar to what has been reported previously (Jan et al., 2001; Nishiyama et al., 2003). Thus, the question remains; why does the CrPV IGR IRES bind to the 40S and 60S subunit fractions under conditions where 80S association is favored? The exciting outcome of these experiments is that the CrPV IGR IRES can bind to the 60S subunit, a property not observed before. This only appears to occur when the IRES is incubated with 40S and 60S subunits in buffers that contain 2 mM Mg2+ in both the initial incubation and gradients. In summary, under these experimental conditions which contain 2 mM Mg2+ in buffer 2/150, we find that the IRES does not bind to the 80S-eEF2 complex, but is rather associated with the 40S and 60S subunits.  129  Figure 4.3. The IGR IRES is found associated with 60S ribosomal subunits after sucrose gradient centrifugation. (A) Radiolabeled IGR IRES binding was assessed by sucrose gradient analysis, as described in section 4.3.1. 40S and 60S ribosomal subunits were combined in the presence of eEF2 and either GTP or GMP-PNP and incubated with wildtype CrPV IGR IRES. Mixtures were loaded on a 10%–30% sucrose gradient and fractionated. Shown is the proportion of total radioactive counts in each fraction. The trace at 254 nm is shown in (B). Peaks corresponding to free IRES, 40S, 60S, and 80S ribosomes are indicated. 130  Figure 4.4. The IGR IRES is bound to 60S ribosomal subunits. (A) Radiolabeled IGR IRES binding was assessed by sucrose gradient analysis. 40S subunits, 60S subunits, or both were combined with radiolabeled wild-type CrPV IGR IRES in buffer 2/150 as described in section 2.3.9. Mixtures were loaded on a 10%–30% sucrose gradient composed in buffer 2/150 and fractionated. Shown is the proportion of total radioactive counts in each fraction. As a control, one IRES binding mixture with 40S and 60S subunits was incubated and fractionated in Buffer E. The trace at 254 nm is shown in (B), and fraction numbers of notable fractions are indicated below the trace. Peaks corresponding to free IRES, 40S, 60S, and 80S ribosomes are also indicated.  131  4.4.3  80S ribosomes that are associated by eEF2 can bind to the IGR IRES  Since the CrPV IGR IRES dissociates from the eEF2-associated 80S complex during centrifugation under our gradient conditions with 2 mM Mg2+ in buffer 2/150, we asked whether the addition of the chemical crosslinker glutaraldehyde to the incubations prior to separation could be used to capture these complexes for analysis. To first determine if the addition of glutaraldehyde interferes with 80S assembly, 40S subunits, 60S subunits, and eEF2 were incubated with or without GMP-PNP in the presence of glutaraldehyde. In the absence of GMP-PNP, 80S assembly occurs, most likely due to glutaraldehyde’s crosslinking properties. This is also illustrated by the appearance of high molecular weight aggregates with glutaraldehyde treatment (Figure 4.5 and Figure 4.6). However, in the presence of GMP-PNP, a modest increase in the ratio of 80S to 60S from ~0.8 to ~1.3 was observed, compared to the incubations without GMP-PNP (Figure 4.5). This is consistent with the observation that GMP-PNP and eEF2 together stabilizes 80S ribosome couples (Figure 4.1 and 4.2). To confirm that 80S ribosomes were being assembled, we collected fractions across the gradient, isolated the RNA, and performed Northern blotting using a probe for the 28S rRNA. We also isolated the protein from the fractions and performed Western blotting using an anti-eEF2 antibody. We found that the 80S fraction contained 28S rRNA and eEF2 (Figure 4.5), in agreement with previous results obtained from similar experiments using yeast ribosomes in the presence of glutaraldehyde by Dr. Akira Kaji (data not shown). We next asked whether the CrPV IRES can bind to eEF2-stabilized 80S ribosomes in the presence of glutaraldehyde. We found that the IRES is present in the 80S fractions by Northern blotting using a probe specific to the IRES (Figure 4.6B), indicating that the IRES  132  can bind to the eEF2-associated 80S ribosome. The presence of the ribosome and eEF2 in the eEF2-associated 80S ribosome fraction was confirmed by Northern blotting using a probe to 28S rRNA and Western blotting using an anti-eEF2 antibody. However, in incubations where GMP-PNP was not added, the IRES was still found in the 80S ribosome fractions, indicating that the IRES associates with the 80S ribosome in the absence or presence of GMP-PNP (Figure 4.6A and 4.6B). To ensure that IRES association to the 80S was not an artifact caused by glutaraldehyde treatment, we utilized a previously characterized mutant IRES, the ΔPKI/PKIII double mutant, which contains mutations that disrupt PKI and PKIII and cannot bind to ribosomes (Jan, 2002). Northern blotting showed that the majority of the mutant IRES was found at the top of the gradient, indicating that glutaraldehyde treatment does not promote non-specific binding to the ribosome (Figure 4.6C). In summary, when we use glutaraldehyde to capture the 80S complex, we see that the wild-type IRES can associate with eEF2-associated 80S ribosomes, and this interaction is specific. These observations suggest that the IRES associates with 80S ribosomes that have been assembled through the action of eEF2.  133  Figure 4.5. Glutaraldehyde crosslinking does not affect the 80S assembly stimulated by eEF2 and GMP-PNP. 80S association in the presence of glutaraldehyde was assessed by sucrose gradient analysis as described in section 4.3.1. 40S and 60S subunits were combined with eEF2 and in the (A) absence or (B) presence of GMP-PNP, as indicated. Mixtures were loaded on a 10%–30% sucrose gradient and fractionated. Shown at top is the trace at 254 nm, and peaks corresponding to 60S and 80S ribosomes are indicated. The ratio of 80S to 60S is shown. Directly below the trace are representative gels from a Western and Northern blot, used to determine the levels of eEF2 and 28S rRNA, respectively. A graph displays the quantitation of eEF2 and 28S rRNA obtained in each fraction, shown at the bottom.  134  Figure 4.6. The IGR IRES binds to ribosomes associated by eEF2 and GMP-PNP. 80S association in the presence of glutaraldehyde was assessed by sucrose gradient analysis as described in section 4.3.1. 40S and 60S subunits were combined with eEF2 and in the (A) absence or (B) presence of GMP-PNP, as indicated. IRES was then added to the mixtures. Incubations were loaded on a 10%–30% sucrose gradient and fractionated. Shown at top is the trace at 254 nm, and peaks corresponding to 60S and 80S ribosomes are indicated. Directly below the trace are representative gels from Western and Northern blots, used to determine the levels of eEF2, 28S rRNA and IRES RNA. A graph displays the quantitation of the macromolecules obtained in each fraction, shown at the bottom. A control experiment with a non-binding mutant IRES is shown in (C). 135  4.4.4  The IGR IRES itself induces 80S association  We have found that the addition of eEF2 and GMP-PNP has a modest stimulatory effect on 80S association in the presence of glutaraldehyde, with an increase of the 80S/60S ratio from ~0.8 to ~1.3 (Figure 4.5). Interestingly, the IRES also stimulated 80S ribosome association even in the absence of GMP-PNP, and to the same extent as with GMP-PNP, with 80S/60S ratios of 3.25 and 3.4, respectively (Figure 4.6A and 4.6B). This suggested that the IRES itself may also play a role in subunit association. One possibility is that the IRES may have some effect on 80S association. In order to assess for this, we conducted 80S assembly assays in the presence of glutaraldehyde, eEF2, and CTP, which is a nucleotide that is not known for any effect on 80S assembly. In this set of experiments, we observed 80S association and an 80S/60S ratio of ~1.0 (Figure 4.7A). These observations are consistent with the previous observation that glutaraldehyde induces some degree of 80S association (Figure 4.5). However, when we add IRES into the binding incubations, 80S assembly was increased significantly with an 80S/60S ratio of ~6.5 (Figure 4.7B), suggesting that the IRES itself is stimulating 40S and 60S association.  136  Figure 4.7. The IRES itself induces 80S association in the presence of CTP. 80S association in the presence of glutaraldehyde was assessed by sucrose gradient analysis as described in section 4.3.1. 40S and 60S subunits were combined with eEF2 and CTP in buffer 2/150 in the (A) absence or (B) presence of IRES, as indicated. Mixtures were loaded on a 10%–30% sucrose gradient and fractionated. Shown at top is the trace at 254 nm, and peaks corresponding to 60S and 80S ribosomes are indicated. The ratio of 80S to 60S is shown. Directly below the trace are representative gels from Western and Northern blots, used to determine the levels of eEF2, 28S rRNA and IRES RNA (if present). A graph displays the quantitation of the macromolecules obtained in each fraction, shown at the bottom.  137  4.5  Discussion  Eukaryotic elongation factor 2 drives the process of translocation during translation elongation. Here, we have shown that eEF2 has another role in facilitating 80S association in the presence of the non-hydrolyzable GTP analogue, GMP-PNP, used in order to mediate the stable binding of eEF2 to 80S ribosomes (Figure 4.1 and 4.2) (Nolan et al., 1975). This effect may be directly related to eEF2’s role in translocation during protein synthesis. During the process of translation elongation, it is known that the 40S and 60S ribosomal subunits engage in a “ratchet-like” movement, relative to one another (Frank and Agrawal, 2000; Spahn et al., 2001a). The stabilization activity of eEF2 may serve to prevent the unexpected dissociation of the 40S and 60S subunits during the translation elongation cycle. During stationary phase or conditions of physiological stress, overall translation is inhibited (Ashe et al., 2000; Uesono and Toh, 2002). During this inhibition, 80S ribosomes dissociate from cellular mRNAs and accumulate as free 80S couples (Kraig and Haber, 1980; Hensold et al., 1996). Our results suggest that this phenomenon is mediated by eEF2, which may inhibit overall translation during cellular stress by sequestering 40S and 60S subunits within 80S ribosome couples. However, this hypothesis remains to be tested further. Previous cryo-EM reconstructions of eEF2 bound to yeast ribosomes provide a structural rationale for this association effect (Spahn et al., 2004a). eEF2 stretches out between the two subunits and serves as a clip that holds both the 40S and 60S subunit. In particular, domain II of eEF2 makes contacts with the 40S subunit, whereas domains I and V make contacts with the 60S subunit. Notably, domains III and IV make contacts with both subunits and the B2a intersubunit bridge (Spahn et al., 2001a; Yusupov et al., 2001).  138  80S ribosome association by eEF2 is only observed in the presence of GMP-PNP, a non-hydrolyzable GTP analogue (Figure 4.1 and 4.2), which is consistent with previous observations that stable binding of eEF2 to ribosomes is only observed in the presence of non-hydrolyzable GTP analogue (Nolan et al., 1975), or when GTP is continuously regenerated in the gradients (Nygard and Nilsson, 1984). We postulate that 80S assembly by eEF2 also occurs under biological conditions in the presence of GTP. In agreement with this idea, it has been shown by Dr. Kaji’s group that eEF2 and GTP induce 80S formation in the presence of a poly(U) mRNA message and N-acetylphenylalanyl-tRNAPhe, without glutaraldehyde (data not shown). These observations suggest that eEF2 plays a role in stabilizing the 80S complex during the elongation cycle. It is interesting to note that the CrPV IGR IRES may also bind to 80S ribosomes that are preassembled through the action of eEF2 (Figure 4.6). This suggests that the IRES has two pathways of ribosome binding; one through the sequential recruitment of 40S and 60S ribosomal subunits, and the second through the direct recruitment of 80S ribosomes. This second pathway may contribute to IGR IRES translation during cellular stress, and may explain how the IRES remains translated during these periods of time. However, it should be noted that this phenomenon has only been observed in the presence of glutaraldehyde, and these results should be interpreted conservatively. The presence of glutaraldehyde may be promoting the capture of spurious interactions, which is illustrated by the appearance of high molecular weight aggregates upon glutaraldehyde treatment (Figure 4.5 and Figure 4.6). An interesting observation is that the IGR IRES remains associated with the 60S ribosomal subunit after the centrifugation of incubations containing 40S and 60S subunits (Figure 4.3). This phenomenon was only observed when the gradients contained 2 mM Mg2+  139  (Figure 4.4). At higher concentrations of 5 mM Mg2+, the IRES remained bound to the 80S ribosomes, presumably because the higher magnesium stabilizes 80S ribosomes and/or keeps the IRES bound to the 80S ribosome during centrifugation. Intriguingly, the IRES binds weakly to 60S subunits when mixtures contained only IRES and purified 60S subunits (Figure 4.4). Thus, we hypothesize that in sucrose gradients with 2 mM Mg2+, the IRES binds to 80S ribosomes, but dissociates from 80S ribosomes during high speed centrifugation (Figure 4.3). This observation is the first to report that the IRES can bind to the 60S subunit. However, the biological significance, if any, of this interaction is not known. To some extent, the 60S binding that we observe under conditions where the 80S ribosome is dissociated may be an artifact, as harsh conditions such as high speed centrifugation are obviously not present in biological systems. However, these experiments may tell us something about the IGR IRES and how it interacts with the ribosome. For instance, our results may suggest that the IGR IRES possess a sequential type of binding mechanism with the ribosome, where its interactions with the 40S become relaxed and interactions with the 60S subunit become tighter upon 80S assembly. In agreement with this idea, previous studies have shown that the IRES itself appears to undergo structural changes during the process of ribosome recruitment. For instance, it is known that SL IV and SL V of the IGR IRES are the main determinants for 40S binding. Upon initial binding to the 40S subunit, conformational changes are seen in both these stem-loops and the 40S subunit (Spahn et al., 2004b; Pfingsten et al., 2010). Furthermore, after 80S ribosome binding, these stem-loops undergo another conformational change, reflecting an alteration in how the IRES interacts with the 40S and 60S subunits (Pfingsten et al., 2010).  140  Additionally, the L1.1 region is implicated in making contacts with the L1 stalk of the 60S ribosomal subunit, and is required for stable 80S association (Jang et al., 2009). Previous work has shown that this region remains unstructured when the IRES is unbound (Pfingsten et al., 2007), but becomes structured when bound to 80S ribosomes (Pfingsten et al., 2006), again reflecting a change in how the IRES may interact with the 60S ribosomal subunit upon 80S association. Taken as a whole, this work suggests that eEF2 plays a role in 80S association and stabilization. Furthermore, we have shown that the IGR IRES may be capable of binding to preformed 80S ribosome couples that have been stabilized by eEF2, providing a putative reason as to how the IGR IRES can remain associated with 80S ribosomes and translated under conditions of cellular stress and viral infection. Finally, we have also observed the binding of the IGR IRES to 60S subunits under certain conditions. We propose that the IRES possesses a sequential binding mechanism that modulates its interactions with the 40S and 60S subunits during the process of 80S ribosome recruitment.  141  5  The role of specialized ribosomes in human dyskeratosis congenita cells in IGR IRES-mediated translation  5.1  Summary  Dyskeratosis congenita (DC) is a rare genetic disease, characterized by symptoms that mirror premature aging. It is currently widely accepted that DC is primarily caused by defects in telomere maintenance. Previous reports have established that X-linked cases of DC are associated with mutations in dyskerin, a multifunctional protein that has roles in both telomere maintenance and rRNA pseudouridylation. However, a growing body of evidence has suggested that defects in rRNA pseudouridylation result in a defect in IRES-driven translation, implying that X-linked DC may be due to the misregulation of a subset of IREScontaining mRNAs. Thus, it has been difficult to delineate the contribution that dyskerin makes in its roles in telomere maintenance and rRNA processing to the pathology of Xlinked DC. Here, we take a direct approach in addressing this question by assaying the translational activity of purified ribosomes from X-linked DC-positive patients. Towards this, we have developed a new method by adding back purified ribosomes to a ribosome-depleted translation extract and then monitoring IRES translation using a dicistronic reporter RNA. Our results indicate that purified ribosomes from DC-positive patients can support IRES translation, suggesting that the defects in these specific DC-positive cells are due to cellular processes other than ribosome dysfunction.  142  5.2  Introduction  Dyskeratosis congenita (DC), also known as Zinsser-Cole-Engman syndrome, is a rare congenital human disorder associated with bone marrow failure (Walne and Dokal, 2008). It has been noted that the symptoms of this disease resemble progeria, or premature aging. Although the pathology of this disease is still unclear, the biochemical evidence and symptoms in humans initially implied that the cause of this disease was due to poor telomere maintenance. Data were later found to support this hypothesis. Mutations in genes that encode components of the telomerase complex (Figure 5.1) were found to be associated with the onset of DC. The first of these genes to be discovered was the X-linked dyskerin or DKC1 gene (Heiss et al., 1998; Knight et al., 1999), which encodes a protein associated with the core telomerase small nucleolar ribonucleoprotein (snoRNP) complex. Subsequently, autosomal dominant forms of DC were found to be associated with mutations in the telomerase RNA component (TERC) (Vulliamy et al., 2001) and telomerase reverse transcriptase (TERT) (Vulliamy et al., 2005; Yamaguchi et al., 2005), which are two components that respectively make up the template and catalytic component of telomerase. Autosomal recessive variants of the disease are associated with mutations in NOP10 (Walne et al., 2007) and NHP2 (Vulliamy et al., 2008), which are additional proteins that make up the core of the telomerase snoRNP. However, these genes do not account for all cases of DC, and approximately 50% of patients on the Dyskeratosis Congenita Registry (Knight et al., 1998) remain genetically uncharacterized. Nevertheless, from these observations, one can reasonably conclude that this disease is principally a disease of defective telomere maintenance.  143  Dyskerin contains two domains that are relevant to our discussion of its multifunctional nature. The first is the C-terminal PUA domain which plays a specific role in binding telomerase RNA, and is thus required for the activity of telomerase in cells (PerezArellano et al., 2007). The second domain is the N-terminal TruB domain, which possesses remarkable similarity to bacterial pseudouridine synthases, which catalyze the isomerization of uridine to pseudouridine in RNAs (Figure 5.2A) (Ofengand, 2002). Pseudouridylation is mediated by dyskerin by first associating with guide RNAs with an H/ACA motif (Figure 5.2B). These guide RNAs then target dyskerin to its RNA substrates through base complementarity, and allow it to carry out the isomerization reaction (Ofengand, 2002). Indeed, when dyskerin was initially discovered to be associated with DC, it was suggested that the disease is a result of defective rRNA pseudouridylation (Heiss et al., 1998). Subsequent studies in non-human model organisms support this idea. For instance, in dyskerin mutant mice which develop DC-like symptoms, telomere lengths are normal in early generations, suggesting that defects in rRNA pseudouridylation is the cause for these symptoms (Ruggero et al., 2003). Parallel studies in Drosophila found that partial loss-offunction mutations in mfl, the fly homologue of dyskerin, are associated with developmental delay, reduced body size, and other abnormalities (Giordano et al., 1999). Since it is known that the maintenance of telomeres in Drosophila is not mediated by telomerase, but through a DNA transposition mechanism (Levis et al., 1993), it is reasonable to assume that these phenotypes are a manifestation of mfl’s other roles in rRNA pseudouridylation. Furthermore, mutations in the yeast homologue of dyskerin, Cbf5p, lead to a loss of global pseudouridylation and are associated with a slow growth phenotype (Zebarjadian et al.,  144  1999). Taking this evidence into consideration, DC may in part be a result of defective rRNA pseudouridylation. As a result of these findings, the question still remains; what is the principal cause of DC in humans? Defects in telomere maintenance, rRNA pseudouridylation, or both? Although the link between rRNA pseudouridylation and DC has been shown in non-human models, the link between the two is much more unclear when we consider human cases. To date, there is evidence to suggest that both processes are affected in human cases of DC. The evidence that links DC with telomere maintenance in humans is abundant, and to date, most reports have shown that human dyskerin mutant cells lines are not associated with losses in global or rRNA pseudouridylation, nor do they display a gross accumulation in guide snoRNAs (Mitchell et al., 1999; Montanaro et al., 2002; Rashid et al., 2006; Wong and Collins, 2006). Rather, they are definitively associated with losses in telomerase activity. Furthermore, the majority of mutations in dyskerin found in clinical cases of DC are present in the PUA domain (Rashid et al., 2006), implicating dyskerin’s role in telomere maintenance as the primary cause of X-linked DC. On the other hand, there is an increasing amount of published data that suggests a linkage between dyskerin and rRNA pseudouridylation. Although the biological role of rRNA pseudouridylation is still largely unknown, recent work has suggested that mutations in dyskerin resulting in defective rRNA pseudouridylation leads to specific impairment of IRES-driven translation, while general translation remains unaffected (Yoon et al., 2006; Bellodi et al., 2010a; Bellodi et al., 2010b). Interestingly, both cellular and viral IRES translation are suppressed in human cells with dyskerin mutations (Yoon et al., 2006; Bellodi et al., 2010a; Bellodi et al., 2010b), and it has been postulated that this observation may form  145  the basis for the pathology of DC. However, a direct link between DC-positive status in humans and a decrease in rRNA pseudouridylation has not been shown to date, and the linkage between DC and rRNA pseudouridylation remains unclear in human cases. These observations illustrate a growing idea that ribosomes may not just be passive machines that mediate protein synthesis on any given mRNA, but are active regulatory bodies that are able to preferentially translate certain mRNAs. This idea, termed the ribosome filter hypothesis, was first formalized by Vincent Mauro and Gerald Edelman of the Scripps Research Institute in 2000 (Mauro and Edelman, 2002; Blanchard et al., 2004; Mauro and Edelman, 2007). This hypothesis states that ribosomes exist as a heterogeneous population, which may differentially affect their ability to translate subsets of mRNAs. In support of this idea, recent findings have shown that ribosomal protein heterogeneity plays a major role in modulating this “ribosome filter”, and one report has shown that the knockdown of rpS6 or rpL19 does not affect the replication of a non-IRES containing virus, vesicular stomatitis virus (VSV), while at the same time specifically decreasing Drosophila C virus (DCV) IRES translation (Cherry et al., 2005). Additional studies by Landry confirm the importance of ribosome heterogeneity on IRES translation, and have found that both the IGR and HCV IRESs require rpS25 for efficient translation (Landry et al., 2009). Taken together, these findings illustrate that ribosome heterogeneity could play a major role in regulating IRES translation. Thus, defects in rRNA pseudouridylation may affect a subset of ribosomes and thus translation of a subset of IRES-containing mRNAs, all of which may contribute to the pathology of X-linked DC. To date, all of the methodologies used to address these types of questions concerning DC have been conducted in vivo, and most in non-human models, which have complicated  146  interpretations. I have taken a direct approach to this question by developing a new in vitro method that depletes ribosomes from extracts and then adding back purified ribosomes from human subjects that are DC-positive and possess mutations in dyskerin. Using this system, we can directly address whether IRES-dependent translation is affected by dyskerin mutations.  147  Figure 5.1. A schematic of the telomerase snoRNP complex, bound to a telomere. The catalytic component of telomerase, TERT, is shown in orange. The RNA template of telomerase, or TERC, is shown as a blue strand. The components of the core snoRNP, which are NHP2, NOP10, GAR, and DKC1 (dyskerin), are shown as indicated. Adapted with permission from Calado, R. T. and N. S. Young. (2009). New England Journal of Medicine. 361, 2353-2365.  148  Figure 5.2. A schematic of what pseudouridylation is, and how it is generated through an H/ACA box guide RNA. (A) shows the isomerization reaction of uridine to pseudouridine, through the action of a pseudouridine synthase. Ring numbering in the nitrogenous base is given in red, as indicated. A simplified schematic of how a guide RNA with an H/ACA motif can mediate pseudouridylation is shown in (B). The guide RNA, shown in black, binds to a catalytic protein (in this case the pseudouridine synthase dyskerin), and recruits it to an RNA substrate. Dyskerin is shown in blue, and the RNA substrate is shown in green. The pseudouridylation modification that is made in the substrate is denoted as NΨ. Adapted with permission from Dr. Judy Wong (UBC Pharmaceutical Sciences).  149  5.3 5.3.1  Materials and methods 40S and 60S subunit purification from dyskerin mutant cells  Ribosomal subunits were purified from either HeLa cell pellets (National Cell Culture Centre) as described (Jan and Sarnow, 2002), or from human dyskerin mutant fibroblasts graciously provided by Dr. Judy Wong (UBC Pharmaceutical Sciences). In brief, 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 100,000g 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 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 concentration of 40S and 60S subunits was determined by spectrophotometry, using the conversions 1 A260 nm=50 nM for 40S and 1 A260 nm=25 nM for 60S subunits. 5.3.2  Depletion and add back of ribosomes in rabbit reticulocyte lysates  Nuclease treated rabbit reticulocyte lysates (Promega) were thawed on ice and centrifuged in a TLA-55 rotor (Beckman) for 2 hours at 50,000 rpm. The upper threequarters of the centrifuged lysate were immediately taken after centrifugation, aliquoted, and  150  stored at -80°C for storage. Ribosome depleted lysates were tested by an in vitro translation assay, where either mutant or wild-type ribosomes were added back to a final concentration of 140 mM. An uncapped dicistronic RNA was then incubated in this lysate with a final concentration of 154 mM potassium acetate for 1 hour at 30°C. Protein expression was measured by incorporation of [35S]-methionine and analyzed by SDS-PAGE. Gels were dried and analyzed by phosphorimager analysis (Typhoon, GE Life Sciences).  151  5.4 5.4.1  Results Ribosome depletion and add-back in rabbit reticulocyte lysates  To create an in vitro translation system suitable for testing ribosomes from DCpositive dyskerin mutant fibroblasts (Figure 5.3), we subjected nuclease-treated rabbit reticulocyte lysates to high speed ultracentrifugation to pellet ribosomes. We optimized this step by ultracentrifugation for different durations of time to determine the minimum amount of time required to pellet the ribosomes from the lysate. Then, to determine whether the lysates are devoid of ribosomes, we monitored translational activity by incubating in vitro transcribed dicistronic reporter RNAs containing the IGR IRES. This reporter contains two cistrons, and the translation of the cistron encoding Renilla luciferase is driven by scanningdependent translation, and the downstream synthesis of firefly luciferase is driven by the IGR IRES. The lysates were incubated in the presence of [35S]-methionine, loaded on an SDSPAGE gel, and the luciferase protein products were detected by autoradiography. We found that centrifugation for as little as 30 minutes could render the lysates incapable of mediating both scanning-dependent and IRES-dependent translation, suggesting that the lysates were depleted of ribosomes (Figure 5.4A). From this point onward, we used a centrifugation time of 2 hours in order to ensure complete ribosome depletion in our lysates. To restore translation in the ribosome-depleted lysates, we added back purified salt-washed human ribosomal subunits from HeLa cells to the lysate. We observed a recovery of both scanningdependent and IRES-driven translation. Thus, ribosome add back is able to restore translation in centrifuged RRL (Figure 5.4B), and further suggests that ribosomes are depleted from the lysate after ultracentrifugation. We also varied the concentration of ribosomes that we add back into the lysate to determine the optimal concentration of ribosomes to recover  152  translational activity, and found that the highest recovery of both scanning-dependent and IGR IRES translation was found at a final concentration of 140 nM (Figure 5.5). Thus, in all of our subsequent experiments, we added back ribosomes to a final concentration of 140 nM. Using this optimized ribosome-depleted translation system, we can specifically assay for the functionality of specialized ribosomes by adding them back to this extract.  153  Figure 5.3. A flowchart of the experimental procedure used for our in vitro translation assay. Ribosomal subunits from cell lysates obtained from wild-type or DC-positive patients will be isolated by high-speed centrifugation and fractionation. Subunits will then be pooled to generate wild-type or X-DC 80S ribosomes. In parallel, nuclease treated rabbit reticulocyte lysates will be centrifuged to pellet out ribosomes. The ribosomes isolated from human lysates will be added back to the ribosome depleted lysates, and the IGR IRES will be tested for activity in an in vitro translation assay.  154  Figure 5.4. A timecourse for ribosome depletion in rabbit reticulocyte lysates. Uncapped dicistronic RNAs containing the wild-type CrPV IGR IRES were incubated in uncentrifuged or centrifuged RRL in the presence of [35S]-methionine, as described in section 5.3.2. The first cistron, encoding Renilla luciferase (Rluc), measures scanning-mediated translation, and the second cistron, firefly luciferase (Fluc), measures IGR IRES-mediated translation. Shown are gels of radiolabeled Fluc and Rluc protein products detected by autoradiography. (A) A timecourse was done in order to determine the optimum centrifugation time for ribosome depletion. Centrifugation times are indicated above the gel. (B) Purified HeLa 80S ribosomes were added back to the ribosome depleted lysate to confirm that ribosomes were being removed during the centrifugation step.  155  Figure 5.5. Optimization of ribosome add back in depleted rabbit reticulocyte lysates. Uncapped dicistronic RNAs containing the wild-type CrPV IGR IRES were incubated in uncentrifuged or centrifuged RRL at 30°C for 60 min in the presence of [35S]-methionine. The first cistron, encoding Renilla luciferase (Rluc), measures scanning-mediated translation, and the second cistron, firefly luciferase (Fluc), measures IGR IRES-mediated translation. Shown is a gel of radiolabeled Fluc and Rluc protein products detected by autoradiography. Different amounts of 80S ribosomes from HeLa cells were added back into depleted lysates and assessed for the recovery of translation. The final concentration of HeLa 80S added back is indicated at the top of the gel, in nanomolar.  156  5.4.2  The IGR IRES is active with ribosomes purified from mutant dyskerin cells  We purified ribosomes from human dyskerin mutant fibroblasts obtained from DCpositive patients, which were collected and harvested by Xi-Lei Zeng, in collaboration with Dr. Judy Wong (Pharmaceutical Sciences, UBC). Since dyskerin has roles in both RNA processing and telomere maintenance, TERT and TERC were expressed in these cells using a retrovirus to rescue telomerase activity (Wong, J. M. Y., unpublished data). This ensures that any effects that we observe are specific to dyskerin’s functions related to rRNA pseudouridylation. Salt-washed ribosomes from these cells were purified using the same protocol for purifying HeLa ribosomal subunits. In this study, we purified ribosomes from cells obtained from two DC-positive patients that possess mutations in dyskerin. One patient possesses a mutation that has a deletion of leucine 37 (ΔL37), which is directly N-terminal of the pseudouridine synthase domain. The second mutation is derived from a patient that has a more severe DC phenotype, and is an alanine 353 substitution to valine (A353V) within the PUA domain of dyskerin. As a wild-type control, cells were obtained from a family match of the patient with the ΔL37 mutation. Using the ribosome-depleted lysates, we added back equimolar amounts of ribosomal subunits and monitored scanning-dependent and IGR IRES-dependent translation in our in vitro system. We found that ribosomes obtained from DC-positive cells could support both scanning- and IGR IRES-dependent translation. Furthermore, translation mediated by ribosomes obtained from wild-type and DC-positive cells was similar (Figure 5.6). These results indicate that ribosomes from DC-postive cells have no significant effect on IGR IRES translation.  157  Figure 5.6. Ribosomes from human, DC-positive dyskerin mutant cells are capable of mediating IRES-driven translation. Uncapped dicistronic RNAs containing the wild-type CrPV IGR IRES were incubated in uncentrifuged or centrifuged RRL at 30°C for 60 min in the presence of [35S]-methionine. The first cistron, encoding Renilla luciferase (Rluc), measures scanning-mediated translation, and the second cistron, firefly luciferase (Fluc), measures IGR IRES-mediated translation. Shown is a gel of radiolabeled Fluc and Rluc protein products detected by autoradiography. Ribosomes from different sources were added to a final concentration of 140 nM, as indicated above the gel.  158  5.5  Discussion  Dyskerin is a multifunctional protein that is implicated in both telomerase activity and rRNA processing. As a result, defects in dyskerin and how it contributes to the pathology of DC has been somewhat confounding. Our results presented here suggest that specific mutations in dyskerin found in DC-positive patients result in ribosomes that can support IRES-driven translation, at least in the case of the CrPV IGR IRES (Figure 5.3). Recent observations show that defects in dyskerin are associated with dysfunctions in IRES-driven translation. This has been observed with cellular IRESs that drive the translation of p27, p53, and XIAP, as well as viral IRESs such as the CrPV IGR IRES and the HCV IRES (Yoon et al., 2006; Bellodi et al., 2010a; Bellodi et al., 2010b). Nevertheless, most studies that link IRES translation to dyskerin and pseudouridylation defects in rRNA have been conducted in mouse embryonic fibroblasts, which may not reflect the true pathology of human X-linked DC. It must be stressed that in most human cases, no link has been found between dyskerin mutations and defects in rRNA modification (Mitchell et al., 1999; Montanaro et al., 2002; Rashid et al., 2006; Wong and Collins, 2006). Furthermore, there have only been two cases using primary human fibroblasts and lymphoblasts from a DCpositive patient that links dyskerin and defective IRES translation (Yoon et al., 2006; Bellodi et al., 2010a). However, it should be noted that the cells used by these groups exhibit telomere maintenance defects and have not been rescued for telomerase activity. This is in contrast to the primary cells that we use in our experimental system, which have telomerase activity rescued through retroviral infection. This ensures that any effects that we observe are specific to dyskerin’s functions related to rRNA pseudouridylation. Thus, differences in  159  rRNA modification or IRES translation that have been observed thus far in human cells by other groups may be due to processes related to cellular senescence. To date, although there has been no definitive link established between human DC cases and rRNA pseudouridylation defects, it has been difficult to delineate the contribution dyskerin has on rRNA pseudouridylation and IRES translation in human model systems and genuine cases of human DC. Because of these uncertainties in the field, we took the most direct approach and tested ribosomes obtained from DC-positive patients, in vitro. Here, we have developed a new experimental system to test the function of specialized ribosomes. There is a growing body of evidence suggesting that ribosome heterogeneity serves as a type of regulation upon both cap-dependent and IRES-dependent translation, and thus, this model system is an excellent method to test these types of ribosomes for translational activity (Mauro and Edelman, 2007). This ability allows us to distinguish between effects that are related to the ribosome, and other cellular effects that are unrelated to ribosome heterogeneity. Nevertheless, there are still several caveats to consider in this experimental system. Firstly, since we use an in vitro translation system that has been centrifuged to deplete cellular components, it can be argued that our assay does not reflect the true physiology of the cell. Furthermore, since we use a rabbit reticulocyte lysate as our translation extract, this may not mirror what occurs in human cells. However, it should be noted that this method may be used with different cell extracts as well, making this approach a powerful experimental tool to look at the role of specialized ribosomes in general. We initially expected that ribosomes from DC-positive cells would not be able to support IGR IRES-driven translation, based on previous in vivo observations in mouse embryonic fibroblasts and primary human fibroblasts and lymphoblasts (Yoon et al., 2006).  160  Thus, we were surprised to see no difference in IRES activity between the wild-type ribosomes and ribosomes obtained from dyskerin mutant cells. This observation could be due to a myriad of factors. First, it may be due to differences in dyskerin mutations. In previous in vivo studies, IRES activity was suppressed in cells that contained a threonine to alanine substitution (T66A) mutation in dyskerin, which is different than the ones utilized in our study. Second, as stated before, the cells used in all the previous studies on this topic have telomere maintenance defects, and do not have TERT and TERC rescued in them (Yoon et al., 2006; Bellodi et al., 2010a; Bellodi et al., 2010b). Thus, effects on IRES translation that have been observed may be due to processes related to cellular senescence. Third, most of the previous findings linking dyskerin mutations with rRNA processing and IRES dysfunction have been observed in model systems other than humans. As a result, it could simply be that these effects may not apply to human cases of X-linked DC. It should also be mentioned that the only IRES that we have utilized for our testing here is the CrPV IGR IRES. It is known that this IRES is activated under very specific periods of time during viral infection and host translational shutoff (Wilson et al., 2000b; Garrey et al., 2010). These stresses are lacking in our in vitro system, and thus may be the reason as to why we do not see any effect on IRES activity. This hypothesis remains to be tested, and it has not escaped our attention that we are able to use previously characterized chemical stressors that have been used previously in RRL, such as NSC119889, that mimic conditions of cellular stress and inhibit cap-dependent translation, while stimulating IGR IRES-driven translation (Robert et al., 2006). However, it is worth mentioning that our in vitro system is the most direct way to assess whether dyskerin-mediated defects in rRNA pseudouridylation lead to defects in IRES  161  activity. Effects on IRES activity that are observed in vivo, as seen in previous reports, may be inundated with other confounding factors that have been described above, which may have led to the observed differences in IRES activity. At the very least, our findings here suggest that this linkage between dyskerin and IRES-mediated translation should be examined more closely.  162  6  Conclusion The IGR IRES is a remarkable RNA machine that behaves as both a translation factor  and a tRNA in that it is able to recruit and activate the translational machinery and mimic a P/E hybrid tRNA for the purposes of translation. Here, we have conducted an in-depth characterization of the IGR IRES, starting with an analysis of the mechanistic role of the L1.1 region of the IRES, determining that this region may serve to mimic the role of an Esite tRNA. We then studied the modular nature of the IRES, determining that the two domains of the IGR IRES can work independently of one another, leading to implications about how the IRES evolved. We then addressed the outstanding question of how the IGR IRES can remain translated during periods of cellular stress when inactive 80S couples are associated, and found that the IGR IRES is able to bind to eEF2-associated 80S ribosomes directly. Finally, we used the IGR IRES in order to ask questions about the pathology of a human disease, dyskeratosis congenita. Taken as a whole, this body of work is a comprehensive study on the IGR IRES, as we have studied the IRES mechanistically, how it evolved, how it may function to remain active during viral infection, and how it is applicable to human disease states. Our biochemical characterization of the L1.1 region in chapter 2 has shown us that this region is responsible for multiple functions, including ribosome binding and positioning, which may be functions that an E-site tRNA would normally mediate during translation. Interestingly, we also found that the L1.1 region also may play a role in the assembly pathway from 40S to 80S, and found that the presence of eIF3 specifically inhibits 80S assembly on the L1.1 mutant IRES. Since one of the subunits of eIF3, eIF3j, is responsible for its interactions with the 40S subunit (Fraser et al., 2004), it would be interesting to see  163  whether eIF3j, or other subunits of eIF3 are modified during viral infection, and whether this ultimately affects the activity of the IGR IRES. Our subsequent finding that the IGR IRES is modular in nature implies that the IGR IRESs may have evolved from the recombination of distinct RNA functional domains. Although we have only tested the IGR IRES, it is likely that other IRESs have been constructed from distinct functional RNAs. For instance, both picornaviral (Lyons and Robertson, 2003; Serrano et al., 2007) and flaviviral IRESs (Nadal et al., 2002) can be cleaved by RNase P, a tRNA processing enzyme, which may reflect their origins as an evolutionary relic of a tRNA. It remains to be seen whether or not cellular IRESs have evolved in the same fashion. It is still unclear how the IGR IRES is translated during periods of viral infection, when inactive 80S ribosome couples accumulate, and overall translation is inhibited. Our results suggest that the IGR IRES may possess the ability to associate directly with these 80S ribosome couples. These data suggest that the IGR IRES may directly recruit these translationally incompetent 80S ribosomes during periods of host translational shutoff, thus providing the rationale as to how the IGR IRES remains translationally active when overall translation is inhibited. These results also suggest that in addition to binding 40S and 60S subunits sequentially, the IGR IRES is able to bind to preformed 80S ribosomes. Past evidence has only suggested that the IGR IRES is capable of this (Jan et al., 2003; Pestova et al., 2004), and further biochemical work, perhaps by chemically crosslinking the 40S and 60S subunits together, may definitively prove that the IGR IRES is capable of binding preformed 80S ribosomes.  164  Although the IGR IRES is one of the better studied viral IRES elements, recent findings have slowly been moving away from the direct biochemical characterization of the IGR IRES, such as what we have shown in the first two chapters, to how it functions within the greater context of the ribosome, the cell, and the pathology of disease. One such finding concerns the idea of the ribosome filter hypothesis, which states that the ribosome itself may behave as a regulatory element that is capable of differentially translating particular mRNAs (Mauro and Edelman, 2002; Mauro and Edelman, 2007). Recent findings from Dr. Sunnie Thompson’s lab illustrate that this appears to apply to the translation of the IGR IRES as well, as yeast ribosomes were found to require the non-essential ribosomal protein rpS25p, in order to mediate translation of the IGR IRES and the HCV IRES (Landry et al., 2009). Considering that these two IRESs operate through different mechanisms and binding sites on the ribosome, it would be interesting to determine if rpS25p is universally required for IRES translation in eukaryotic ribosomes. To this end, proteomic approaches that have been used in the past to identify proteins bound to HCV IRES-80S complexes (Locker et al., 2007) may provide interesting results with the IGR IRES. The mechanistic simplicity of the IGR IRES also serves as a powerful model and tool to understand other IRES elements and translation in general. Since the IRES does not require any initiation factors for translational activity, it can be used in in vitro assays under very controlled conditions with purified ribosomes to observe different steps of translation. Furthermore, this ability permits the identification of factor-independent processes during translation. One such example that utilized this capability of the IGR IRES was the determination of how selenocysteine was encoded in proteins. More specifically, it was unknown in the past whether or not the incorporation of selenocysteine, which is encoded by  165  a UGA codon, was dependent on the presence of initiation factors. One group found that the incorporation of selenocysteine in a reporter driven by the IGR IRES was normal, suggesting that initiation factors are not required for this process (Donovan and Copeland, 2010). This type of novel use of the IGR IRES is not limited to understanding biological systems, but is also starting to be used to create new synthetic expression systems. The EMCV IRES has long been used to drive the expression of genes in commercially available expression vectors, and recent work has demonstrated the creation of a novel, rationally designed riboswitch that is mediated by the IGR IRES (Ogawa, 2011). In the future, these IRESs may possibly serve as research tools with diverse applications. Nevertheless, understanding how the IGR IRES functions will help us to understand how IRESs function in general, as well as provide us with tools to understand how ribosomes, and translation at large, functions within the cell.  166  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. J Biol Chem. 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, 491506. Agrawal, R.K., Heagle, A.B., Penczek, P., Grassucci, R.A., and Frank, J. (1999). EF-Gdependent GTP hydrolysis induces translocation accompanied by large conformational changes in the 70S ribosome. Nat Struct Biol. 6, 643-647. Ali, I.K., McKendrick, L., Morley, S.J., and Jackson, R.J. (2001). Activity of the hepatitis A virus IRES requires association between the cap-binding translation initiation factor (eIF4E) and eIF4G. J Virol. 75, 7854-7863. Ali, N. and Siddiqui, A. (1997). The La antigen binds 5' noncoding region of the hepatitis C virus RNA in the context of the initiator AUG codon and stimulates internal ribosome entry site-mediated translation. Proc Natl Acad Sci U S A. 94, 2249-2254. Allam, H. and Ali, N. (2010). Initiation factor eIF2-independent mode of c-Src mRNA translation occurs via an internal ribosome entry site. J Biol Chem. 285, 5713-5725. Allen, G.S., Zavialov, A., Gursky, R., Ehrenberg, M., and Frank, J. (2005). The cryo-EM structure of a translation initiation complex from Escherichia coli. Cell. 121, 703-712. Ashe, M.P., De Long, S.K., and Sachs, A.B. (2000). Glucose depletion rapidly inhibits translation initiation in yeast. Mol Biol Cell. 11, 833-848. Babaylova, E., Graifer, D., Malygin, A., Stahl, J., Shatsky, I., and Karpova, G. (2009). Positioning of subdomain IIId and apical loop of domain II of the hepatitis C IRES on the human 40S ribosome. Nucleic Acids Res. 37, 1141-1151. Bellodi, C., Kopmar, N., and Ruggero, D. (2010a). Deregulation of oncogene-induced senescence and p53 translational control in X-linked dyskeratosis congenita. EMBO J. 29, 1865-1876. Bellodi, C., Krasnykh, O., Haynes, N., Theodoropoulou, M., Peng, G., Montanaro, L., and Ruggero, D. (2010b). Loss of function of the tumor suppressor DKC1 perturbs p27 translation control and contributes to pituitary tumorigenesis. Cancer Res. 70, 60266035. Belsham, G.J. and Sonenberg, N. (1996). RNA-protein interactions in regulation of picornavirus RNA translation. Microbiol Rev. 60, 499-511. Bernstein, H.D., Sonenberg, N., and Baltimore, D. (1985). Poliovirus mutant that does not selectively inhibit host cell protein synthesis. Mol Cell Biol. 5, 2913-2923. Blaha, G. and Nierhaus, K.H. (2001). Features and functions of the ribosomal E site. Cold Spring Harb Symp Quant Biol. 66, 135-146. Blanchard, S.C., Kim, H.D., Gonzalez, R.L., Jr., Puglisi, J.D., and Chu, S. (2004). tRNA dynamics on the ribosome during translation. Proc Natl Acad Sci U S A. 101, 1289312898. Blyn, L.B., Towner, J.S., Semler, B.L., and Ehrenfeld, E. (1997). Requirement of poly(rC) binding protein 2 for translation of poliovirus RNA. J Virol. 71, 6243-6246. Bonning, B.C. and Miller, W.A. (2010). Dicistroviruses. Annu Rev Entomol. 55, 129-150.  167  Brown, E.A., Zajac, A.J., and Lemon, S.M. (1994). In vitro characterization of an internal ribosomal entry site (IRES) present within the 5' nontranslated region of hepatitis A virus RNA: comparison with the IRES of encephalomyocarditis virus. J Virol. 68, 1066-1074. Buratti, E., Tisminetzky, S., Zotti, M., and Baralle, F.E. (1998). Functional analysis of the interaction between HCV 5'UTR and putative subunits of eukaryotic translation initiation factor eIF3. Nucleic Acids Res. 26, 3179-3187. Carrasco, L., Battaner, E., and Vazquez, D. (1974). The elongation steps in protein synthesis by eukaryotic ribosomes: effects of antibiotics. Methods Enzymol. 30, 282-289. 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. Cherry, S., Doukas, T., Armknecht, S., Whelan, S., Wang, H., Sarnow, P., and Perrimon, N. (2005). Genome-wide RNAi screen reveals a specific sensitivity of IRES-containing RNA viruses to host translation inhibition. Genes Dev. 19, 445-452. 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. Nat Struct Mol Biol. 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., Martinson, V., vanEngelsdorp, D., Kalkstein, A.L., Drysdale, A., Hui, J., Zhai, J., Cui, L., Hutchison, S.K., Simons, J.F., Egholm, M., Pettis, J.S., and Lipkin, W.I. (2007). A metagenomic survey of microbes in honey bee colony collapse disorder. Science. 318, 283. Culley, A.I., Lang, A.S., and Suttle, C.A. (2006). Metagenomic analysis of coastal RNA virus communities. Science. 312, 1795-1798. 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. Diaconu, M., Kothe, U., Schlunzen, F., Fischer, N., Harms, J.M., Tonevitsky, A.G., Stark, H., Rodnina, M.V., and Wahl, M.C. (2005). Structural Basis for the Function of the Ribosomal L7/12 Stalk in Factor Binding and GTPase Activation. Cell. 121, 9911004. Dinos, G., Wilson, D.N., Teraoka, Y., Szaflarski, W., Fucini, P., Kalpaxis, D., and Nierhaus, K.H. (2004). Dissecting the ribosomal inhibition mechanisms of edeine and pactamycin: the universally conserved residues G693 and C795 regulate P-site RNA binding. Mol Cell. 13, 113-124. 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. J Biol Chem. 285, 26779-26787. Donovan, J. and Copeland, P.R. (2010). The efficiency of selenocysteine incorporation is regulated by translation initiation factors. J Mol Biol. 400, 659-664.  168  Dorner, A.J., Semler, B.L., Jackson, R.J., Hanecak, R., Duprey, E., and Wimmer, E. (1984). In vitro translation of poliovirus RNA: utilization of internal initiation sites in reticulocyte lysate. J Virol. 50, 507-514. Dorner, S., Brunelle, J.L., Sharma, D., and Green, R. (2006). The hybrid state of tRNA binding is an authentic translation elongation intermediate. Nat Struct Mol Biol. 13, 234-241. Elliott, M.B., Gottlieb, P.A., and Gollnick, P. (1999). Probing the TRAP-RNA interaction with nucleoside analogs. RNA. 5, 1277-1289. 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. J Biol Chem. 257, 14806-14810. Fei, J., Kosuri, P., MacDougall, D.D., and Gonzalez, R.L., Jr. (2008). Coupling of ribosomal L1 stalk and tRNA dynamics during translation elongation. Mol Cell. 30, 348-359. Fernandez, J., Yaman, I., Sarnow, P., Snider, M.D., and Hatzoglou, M. (2002). Regulation of internal ribosomal entry site-mediated translation by phosphorylation of the translation initiation factor eIF2alpha. J Biol Chem. 277, 19198-19205. 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., Sengupta, J., Gao, N., and Taylor, D.J. (2007). The process of mRNAtRNA translocation. Proc Natl Acad Sci U S A. 104, 19671-19678. Fraser, C.S., Lee, J.Y., Mayeur, G.L., Bushell, M., Doudna, J.A., and Hershey, J.W. (2004). The j-subunit of human translation initiation factor eIF3 is required for the stable binding of eIF3 and its subcomplexes to 40 S ribosomal subunits in vitro. J Biol Chem. 279, 8946-8956. Gamarnik, A.V. and Andino, R. (1998). Switch from translation to RNA replication in a positive-stranded RNA virus. Genes Dev. 12, 2293-2304. Gamarnik, A.V. and Andino, R. (2000). Interactions of viral protein 3CD and poly(rC) binding protein with the 5' untranslated region of the poliovirus genome. J Virol. 74, 2219-2226. Gao, N., Zavialov, A.V., Li, W., Sengupta, J., Valle, M., Gursky, R.P., Ehrenberg, M., and Frank, J. (2005). Mechanism for the disassembly of the posttermination complex inferred from cryo-EM studies. Mol Cell. 18, 663-674. 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. Geigenmuller, U. and Nierhaus, K.H. (1990). Significance of the third tRNA binding site, the E site, on E. coli ribosomes for the accuracy of translation: an occupied E site prevents the binding of non-cognate aminoacyl-tRNA to the A site. EMBO J. 9, 45274533. Giordano, E., Peluso, I., Senger, S., and Furia, M. (1999). minifly, a Drosophila gene required for ribosome biogenesis. J Cell Biol. 144, 1123-1133.  169  Gomez-Lorenzo, M.G., Spahn, C.M., Agrawal, R.K., Grassucci, R.A., Penczek, P., Chakraburtty, K., Ballesta, J.P., 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 A resolution. EMBO J. 19, 27102718. Gosert, R., Chang, K.H., Rijnbrand, R., Yi, M., Sangar, D.V., and Lemon, S.M. (2000). Transient expression of cellular polypyrimidine-tract binding protein stimulates capindependent translation directed by both picornaviral and flaviviral internal ribosome entry sites In vivo. Mol Cell Biol. 20, 1583-1595. 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. J Gen Virol. 88, 1583-1588. Hahm, B., Kim, Y.K., Kim, J.H., Kim, T.Y., and Jang, S.K. (1998). Heterogeneous nuclear ribonucleoprotein L interacts with the 3' border of the internal ribosomal entry site of hepatitis C virus. J Virol. 72, 8782-8788. Hasson, K.W., Lightner, D.V., Poulos, B.T., Redman, R.M., White, B.L., Brock, J.A., and Bonami, J.R. (1995). Taura Syndrome in Penaeus-Vannamei - Demonstration of a Viral Etiology. Diseases of Aquatic Organisms. 23, 115-126. Hatakeyama, Y., Shibuya, N., Nishiyama, T., and Nakashima, N. (2004). Structural variant of the intergenic internal ribosome entry site elements in dicistroviruses and computational search for their counterparts. RNA. 10, 779-786. Heiss, N.S., Knight, S.W., Vulliamy, T.J., Klauck, S.M., Wiemann, S., Mason, P.J., Poustka, A., and Dokal, I. (1998). X-linked dyskeratosis congenita is caused by mutations in a highly conserved gene with putative nucleolar functions. Nat Genet. 19, 32-38. Hellen, C.U. and Pestova, T.V. (1999). Translation of hepatitis C virus RNA. J Viral Hepat. 6, 79-87. Hellen, C.U. and Sarnow, P. (2001). Internal ribosome entry sites in eukaryotic mRNA molecules. Genes Dev. 15, 1593-1612. Hensold, J.O., Barth-Baus, D., and Stratton, C.A. (1996). Inducers of erythroleukemic differentiation cause messenger RNAs that lack poly(A)-binding protein to accumulate in translationally inactive, salt-labile 80 S ribosomal complexes. J Biol Chem. 271, 23246-23254. Hirashima, A. and Kaji, A. (1973). Role of elongation factor G and a protein factor on the release of ribosomes from messenger ribonucleic acid. J Biol Chem. 248, 7580-7587. Hirokawa, G., Nijman, R.M., Raj, V.S., Kaji, H., Igarashi, K., and Kaji, A. (2005). The role of ribosome recycling factor in dissociation of 70S ribosomes into subunits. RNA. 11, 1317-1328. 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, 41354139. Hunt, S.L., Hsuan, J.J., Totty, N., and Jackson, R.J. (1999). unr, a cellular cytoplasmic RNAbinding protein with five cold-shock domains, is required for internal initiation of translation of human rhinovirus RNA. Genes Dev. 13, 437-448.  170  Hunt, S.L. and Jackson, R.J. (1999). Polypyrimidine-tract binding protein (PTB) is necessary, but not sufficient, for efficient internal initiation of translation of human rhinovirus-2 RNA. RNA. 5, 344-359. Jackson, R.J. (2005). Alternative mechanisms of initiating translation of mammalian mRNAs. Biochem Soc Trans. 33, 1231-1241. Jackson, R.J., Hellen, C.U., and Pestova, T.V. (2010). The mechanism of eukaryotic translation initiation and principles of its regulation. Nat Rev Mol Cell Biol. 11, 113127. 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. Jan, E., Thompson, S.R., Wilson, J.E., Pestova, T.V., Hellen, C.U., and Sarnow, P. (2001). Initiator Met-tRNA-independent translation mediated by an internal ribosome entry site element in cricket paralysis virus-like insect viruses. Cold Spring Harb Symp Quant Biol. 66, 285-292. 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. J Virol. 62, 2636-2643. Jopling, C.L., Spriggs, K.A., Mitchell, S.A., Stoneley, M., and Willis, A.E. (2004). L-Myc protein synthesis is initiated by internal ribosome entry. RNA. 10, 287-298. Kaminski, A., Hunt, S.L., Patton, J.G., and Jackson, R.J. (1995). Direct evidence that polypyrimidine tract binding protein (PTB) is essential for internal initiation of translation of encephalomyocarditis virus RNA. RNA. 1, 924-938. 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. Mol 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, 266274. Kapp, L.D. and Lorsch, J.R. (2004). GTP-dependent recognition of the methionine moiety on initiator tRNA by translation factor eIF2. J Mol Biol. 335, 923-936. Karimi, R., Pavlov, M.Y., Buckingham, R.H., and Ehrenberg, M. (1999). Novel roles for classical factors at the interface between translation termination and initiation. Mol Cell. 3, 601-609. Kieft, J.S. (2008a). Comparing the three-dimensional structures of Dicistroviridae IGR IRES RNAs with other viral RNA structures. Virus Res. Kieft, J.S. (2008b). Viral IRES RNA structures and ribosome interactions. Trends Biochem Sci. 33, 274-283. Kim, H.D., Puglisi, J.D., and Chu, S. (2007). Fluctuations of transfer RNAs between classical and hybrid states. Biophys J. 93, 3575-3582.  171  Kim, J.H., Paek, K.Y., Ha, S.H., Cho, S., Choi, K., Kim, C.S., Ryu, S.H., and Jang, S.K. (2004). A cellular RNA-binding protein enhances internal ribosomal entry sitedependent translation through an interaction downstream of the hepatitis C virus polyprotein initiation codon. Mol Cell Biol. 24, 7878-7890. Knight, S., Vulliamy, T., Copplestone, A., Gluckman, E., Mason, P., and Dokal, I. (1998). Dyskeratosis Congenita (DC) Registry: identification of new features of DC. Br J Haematol. 103, 990-996. Knight, S.W., Heiss, N.S., Vulliamy, T.J., Greschner, S., Stavrides, G., Pai, G.S., Lestringant, G., Varma, N., Mason, P.J., Dokal, I., and Poustka, A. (1999). X-linked dyskeratosis congenita is predominantly caused by missense mutations in the DKC1 gene. Am J Hum Genet. 65, 50-58. Kolupaeva, V.G., Hellen, C.U., and Shatsky, I.N. (1996). Structural analysis of the interaction of the pyrimidine tract-binding protein with the internal ribosomal entry site of encephalomyocarditis virus and foot-and-mouth disease virus RNAs. RNA. 2, 1199-1212. Kolupaeva, V.G., Pestova, T.V., and Hellen, C.U. (2000a). An enzymatic footprinting analysis of the interaction of 40S ribosomal subunits with the internal ribosomal entry site of hepatitis C virus. J Virol. 74, 6242-6250. Kolupaeva, V.G., Pestova, T.V., and Hellen, C.U. (2000b). Ribosomal binding to the internal ribosomal entry site of classical swine fever virus. RNA. 6, 1791-1807. 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. Kozak, M. and Shatkin, A.J. (1978). Migration of 40 S ribosomal subunits on messenger RNA in the presence of edeine. J Biol Chem. 253, 6568-6577. Kraig, E. and Haber, J.E. (1980). Messenger ribonucleic acid and protein metabolism during sporulation of Saccharomyces cerevisiae. J Bacteriol. 144, 1098-1112. 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 Dev. 23, 27532764. Le Quesne, J.P., Stoneley, M., Fraser, G.A., and Willis, A.E. (2001). Derivation of a structural model for the c-myc IRES. J Mol Biol. 310, 111-126. Lee, Y.Y., Cevallos, R.C., and Jan, E. (2009). An upstream open reading frame regulates translation of GADD34 during cellular stresses that induce eIF2alpha phosphorylation. J Biol Chem. 284, 6661-6673. Levis, R.W., Ganesan, R., Houtchens, K., Tolar, L.A., and Sheen, F.M. (1993). Transposons in place of telomeric repeats at a Drosophila telomere. Cell. 75, 1083-1093. Lill, R., Robertson, J.M., and Wintermeyer, W. (1989). Binding of the 3' terminus of tRNA to 23S rRNA in the ribosomal exit site actively promotes translocation. EMBO J. 8, 3933-3938. Locker, N., Chamond, N., and Sargueil, B. (2010). A conserved structure within the HIV gag open reading frame that controls translation initiation directly recruits the 40S subunit and eIF3. Nucleic Acids Res. Locker, N., Easton, L.E., and Lukavsky, P.J. (2007). HCV and CSFV IRES domain II mediate eIF2 release during 80S ribosome assembly. EMBO J. 26, 795-805.  172  Long, E.O. and Dawid, I.B. (1980). Alternative pathways in the processing of ribosomal RNA precursor in Drosophila melanogaster. J Mol Biol. 138, 873. Lopez de Quinto, S. and Martinez-Salas, E. (2000). Interaction of the eIF4G initiation factor with the aphthovirus IRES is essential for internal translation initiation in vivo. RNA. 6, 1380-1392. Loria, A. and Pan, T. (1996). Domain structure of the ribozyme from eubacterial ribonuclease P. RNA. 2, 551-563. Lyons, A.J. and Robertson, H.D. (2003). Detection of tRNA-like structure through RNase P cleavage of viral internal ribosome entry site RNAs near the AUG start triplet. J Biol Chem. 278, 26844-26850. 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. Mol Cell. 17, 265-275. Marquez, V., Wilson, D.N., Tate, W.P., Triana-Alonso, F., and Nierhaus, K.H. (2004). Maintaining the ribosomal reading frame: the influence of the E site during translational regulation of release factor 2. Cell. 118, 45-55. Mauro, V.P. and Edelman, G.M. (2002). The ribosome filter hypothesis. Proc Natl Acad Sci U S A. 99, 12031-12036. Mauro, V.P. and Edelman, G.M. (2007). The ribosome filter redux. Cell Cycle. 6, 2246-2251. McGarry, K.G., Walker, S.E., Wang, H., and Fredrick, K. (2005). Destabilization of the P site codon-anticodon helix results from movement of tRNA into the P/E hybrid state within the ribosome. Mol Cell. 20, 613-622. Mitchell, J.R., Wood, E., and Collins, K. (1999). A telomerase component is defective in the human disease dyskeratosis congenita. Nature. 402, 551-555. Moazed, D. and Noller, H.F. (1989). Interaction of tRNA with 23S rRNA in the ribosomal A, P, and E sites. Cell. 57, 585-597. Montanaro, L., Chilla, A., Trere, D., Pession, A., Govoni, M., Tazzari, P.L., and Derenzini, M. (2002). Increased mortality rate and not impaired ribosomal biogenesis is responsible for proliferative defect in dyskeratosis congenita cell lines. J Invest Dermatol. 118, 193-198. Nadal, A., Martell, M., Lytle, J.R., Lyons, A.J., Robertson, H.D., Cabot, B., Esteban, J.I., Esteban, R., Guardia, J., and Gomez, J. (2002). Specific cleavage of hepatitis C virus RNA genome by human RNase P. J Biol Chem. 277, 30606-30613. Nakashima, N. and Uchiumi, T. (2008). Functional analysis of structural motifs in dicistroviruses. Virus Res. Nierhaus, K.H. (2006). Decoding errors and the involvement of the E-site. Biochimie. 88, 1013-1019. 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 Res. 35, 1514-1521.  173  Nolan, R.D., Grasmuk, H., and Drews, J. (1975). The binding of tritiated elongation factors 1 and 2 to ribosomes from Krebs II mouse ascites tumor cells. Eur J Biochem. 50, 391402. Novac, O., Guenier, A.S., and Pelletier, J. (2004). Inhibitors of protein synthesis identified by a high throughput multiplexed translation screen. Nucleic Acids Res. 32, 902-915. Nygard, O. and Nilsson, L. (1984). Nucleotide-mediated interactions of eukaryotic elongation factor EF-2 with ribosomes. Eur J Biochem. 140, 93-96. Ofengand, J. (2002). Ribosomal RNA pseudouridines and pseudouridine synthases. FEBS Lett. 514, 17-25. Ogawa, A. (2011). Rational design of artificial riboswitches based on ligand-dependent modulation of internal ribosome entry in wheat germ extract and their applications as label-free biosensors. RNA. 17, 478-488. 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. Passmore, L.A., Schmeing, T.M., Maag, D., Applefield, D.J., Acker, M.G., Algire, M.A., Lorsch, J.R., and Ramakrishnan, V. (2007). The eukaryotic translation initiation factors eIF1 and eIF1A induce an open conformation of the 40S ribosome. Mol Cell. 26, 41-50. Paulin, F.E., Campbell, L.E., O'Brien, K., Loughlin, J., and Proud, C.G. (2001). Eukaryotic translation initiation factor 5 (eIF5) acts as a classical GTPase-activator protein. Curr Biol. 11, 55-59. Pelletier, J., Kaplan, G., Racaniello, V.R., and Sonenberg, N. (1988). Translational efficiency of poliovirus mRNA: mapping inhibitory cis-acting elements within the 5' noncoding region. J Virol. 62, 2219-2227. 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. Perez-Arellano, I., Gallego, J., and Cervera, J. (2007). The PUA domain - a structural and functional overview. FEBS J. 274, 4972-4984. Peske, F., Rodnina, M.V., and Wintermeyer, W. (2005). Sequence of steps in ribosome recycling as defined by kinetic analysis. Mol Cell. 18, 403-412. Pestova, T.V. and Hellen, C.U. (1999). Internal initiation of translation of bovine viral diarrhea virus RNA. Virology. 258, 249-256. 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, 332335. Pestova, T.V., Lorsch, J.R., and Hellen, C.U. 2007. The Mechanism of Translation Initiation in Eukaryotes. In Translational Control in Biology and Medicine, (ed. M.B. Mathews, N. Sonenberg, and J. Hershey), p. 87. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.  174  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., Castile, A.E., and Kieft, J.S. (2010). Mechanistic role of structurally dynamic regions in Dicistroviridae IGR IRESs. J Mol Biol. 395, 205-217. 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., Costantino, D.A., and Kieft, J.S. (2007). Conservation and diversity among the three-dimensional folds of the Dicistroviridae intergenic region IRESes. J Mol Biol. 370, 856-869. Pilipenko, E.V., Pestova, T.V., Kolupaeva, V.G., Khitrina, E.V., Poperechnaya, A.N., Agol, V.I., and Hellen, C.U. (2000). A cell cycle-dependent protein serves as a templatespecific translation initiation factor. Genes Dev. 14, 2028-2045. Pilipenko, E.V., Viktorova, E.G., Guest, S.T., Agol, V.I., and Roos, R.P. (2001). Cellspecific proteins regulate viral RNA translation and virus-induced disease. EMBO J. 20, 6899-6908. Pisarev, A.V., Chard, L.S., Kaku, Y., Johns, H.L., Shatsky, I.N., and Belsham, G.J. (2004). Functional and structural similarities between the internal ribosome entry sites of hepatitis C virus and porcine teschovirus, a picornavirus. J Virol. 78, 4487-4497. Pisarev, A.V., Kolupaeva, V.G., Pisareva, V.P., Merrick, W.C., Hellen, C.U., and Pestova, T.V. (2006). Specific functional interactions of nucleotides at key -3 and +4 positions flanking the initiation codon with components of the mammalian 48S translation initiation complex. Genes Dev. 20, 624-636. Rashid, R., Liang, B., Baker, D.L., Youssef, O.A., He, Y., Phipps, K., Terns, R.M., Terns, M.P., and Li, H. (2006). Crystal structure of a Cbf5-Nop10-Gar1 complex and implications in RNA-guided pseudouridylation and dyskeratosis congenita. Mol Cell. 21, 249-260. Robert, F., Kapp, L.D., Khan, S.N., Acker, M.G., Kolitz, S., Kazemi, S., Kaufman, R.J., Merrick, W.C., Koromilas, A.E., Lorsch, J.R., and Pelletier, J. (2006). Initiation of protein synthesis by hepatitis C virus is refractory to reduced eIF2.GTP.MettRNA(i)(Met) ternary complex availability. Mol Biol Cell. 17, 4632. Ruggero, D., Grisendi, S., Piazza, F., Rego, E., Mari, F., Rao, P.H., Cordon-Cardo, C., and Pandolfi, P.P. (2003). Dyskeratosis congenita and cancer in mice deficient in ribosomal RNA modification. Science. 299, 259-262. Ruusala, T., Ehrenberg, M., and Kurland, C.G. (1982). Catalytic effects of elongation factor Ts on polypeptide synthesis. EMBO J. 1, 75-78. Saleh, L., Rust, R.C., Fullkrug, R., Beck, E., Bassili, G., Ochs, K., and Niepmann, M. (2001). Functional interaction of translation initiation factor eIF4G with the foot-and-mouth disease virus internal ribosome entry site. J Gen Virol. 82, 757-763.  175  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. J Virol. 73, 1219-1226. 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. Schuler, M., Connell, S.R., Lescoute, A., Giesebrecht, J., Dabrowski, M., Schroeer, B., Mielke, T., Penczek, P.A., Westhof, E., and Spahn, C.M. (2006). Structure of the ribosome-bound cricket paralysis virus IRES RNA. Nature Structural and Molecular Biology. 13, 1092. Seif, E., Cadieux, A., and Lang, B.F. (2006). Hybrid E. coli--Mitochondrial ribonuclease P RNAs are catalytically active. RNA. 12, 1661-1670. 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. Sergiev, P.V., Lesnyak, D.V., Kiparisov, S.V., Burakovsky, D.E., Leonov, A.A., Bogdanov, A.A., Brimacombe, R., and Dontsova, O.A. (2005). Function of the ribosomal E-site: a mutagenesis study. Nucleic Acids Research. 33, 6048-6056. Serrano, P., Gomez, J., and Martinez-Salas, E. (2007). Characterization of a cyanobacterial RNase P ribozyme recognition motif in the IRES of foot-and-mouth disease virus reveals a unique structural element. RNA. 13, 849-859. 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. J Virol. 77, 1200212010. Sizova, D.V., Kolupaeva, V.G., Pestova, T.V., Shatsky, I.N., and Hellen, C.U. (1998). Specific interaction of eukaryotic translation initiation factor 3 with the 5' nontranslated regions of hepatitis C virus and classical swine fever virus RNAs. J Virol. 72, 4775-4782. 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. Song, Y., Tzima, E., Ochs, K., Bassili, G., Trusheim, H., Linder, M., Preissner, K.T., and Niepmann, M. (2005). Evidence for an RNA chaperone function of polypyrimidine tract-binding protein in picornavirus translation. RNA. 11, 1809-1824. Spahn, C.M., Beckmann, R., Eswar, N., Penczek, P.A., Sali, A., Blobel, G., and Frank, J. (2001a). Structure of the 80S ribosome from Saccharomyces cerevisiae--tRNAribosome and subunit-subunit interactions. Cell. 107, 373-386. Spahn, C.M., Gomez-Lorenzo, M.G., Grassucci, R.A., Jorgensen, R., Andersen, G.R., Beckmann, R., Penczek, P.A., Ballesta, J.P., and Frank, J. (2004a). Domain movements of elongation factor eEF2 and the eukaryotic 80S ribosome facilitate tRNA translocation. EMBO J. 23, 1008-1019. Spahn, C.M., Jan, E., Mulder, A., Grassucci, R.A., Sarnow, P., and Frank, J. (2004b). CryoEM 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.  176  Spahn, C.M., Kieft, J.S., Grassucci, R.A., Penczek, P.A., Zhou, K., Doudna, J.A., and Frank, J. (2001b). Hepatitis C virus IRES RNA-induced changes in the conformation of the 40s ribosomal subunit. Science. 291, 1959-1962. Stern, S., Moazed, D., and Noller, H.F. (1988). Structural analysis of RNA using chemical and enzymatic probing monitored by primer extension. Methods Enzymol. 164, 481489. 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. Tang, J. and Breaker, R.R. (1998). Mechanism for allosteric inhibition of an ATP-sensitive ribozyme. Nucleic Acids Res. 26, 4214-4221. 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. Mol Cell Biol. 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. Thomas, G. and Favre, A. (1980). 4-Thiouridine triggers both growth delay induced by nearultraviolet light and photoprotection. Eur J Biochem. 113, 67. 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. Tischendorf, J.J., Beger, C., Korf, M., Manns, M.P., and Kruger, M. (2004). Polypyrimidine tract-binding protein (PTB) inhibits Hepatitis C virus internal ribosome entry site (HCV IRES)-mediated translation, but does not affect HCV replication. Arch Virol. 149, 1955-1970. Torres-Larios, A., Swinger, K.K., Pan, T., and Mondragon, A. (2006). Structure of ribonuclease P--a universal ribozyme. Curr Opin Struct Biol. 16, 327-335. Trachsel, H. and Staehelin, T. (1979). Initiation of mammalian protein synthesis. The multiple functions of the initiation factor eIF-3. Biochim Biophys Acta. 565, 305-314. Uesono, Y. and Toh, E.A. (2002). Transient inhibition of translation initiation by osmotic stress. J Biol Chem. 277, 13848-13855. Unbehaun, A., Borukhov, S.I., Hellen, C.U., 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 Dev. 18, 3078-3093. Valle, M., Zavialov, A., Sengupta, J., Rawat, U., Ehrenberg, M., and Frank, J. (2003). Locking and unlocking of ribosomal motions. Cell. 114, 123-134. Vulliamy, T., Beswick, R., Kirwan, M., Marrone, A., Digweed, M., Walne, A., and Dokal, I. (2008). Mutations in the telomerase component NHP2 cause the premature ageing syndrome dyskeratosis congenita. Proc Natl Acad Sci U S A. 105, 8073-8078. Vulliamy, T., Marrone, A., Goldman, F., Dearlove, A., Bessler, M., Mason, P.J., and Dokal, I. (2001). The RNA component of telomerase is mutated in autosomal dominant dyskeratosis congenita. Nature. 413, 432-435.  177  Vulliamy, T.J., Walne, A., Baskaradas, A., Mason, P.J., Marrone, A., and Dokal, I. (2005). Mutations in the reverse transcriptase component of telomerase (TERT) in patients with bone marrow failure. Blood Cells Mol Dis. 34, 257-263. Walne, A.J. and Dokal, I. (2008). Dyskeratosis Congenita: a historical perspective. Mech Ageing Dev. 129, 48-59. Walne, A.J., Vulliamy, T., Marrone, A., Beswick, R., Kirwan, M., Masunari, Y., Al-Qurashi, F.H., Aljurf, M., and Dokal, I. (2007). Genetic heterogeneity in autosomal recessive dyskeratosis congenita with one subtype due to mutations in the telomeraseassociated protein NOP10. Hum Mol Genet. 16, 1619-1629. Wierenga, R.K. (2001). The TIM-barrel fold: a versatile framework for efficient enzymes. FEBS Lett. 492, 193-198. 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. Witherell, G.W., Gil, A., and Wimmer, E. (1993). Interaction of polypyrimidine tract binding protein with the encephalomyocarditis virus mRNA internal ribosomal entry site. Biochemistry. 32, 8268-8275. Wong, J.M. and Collins, K. (2006). Telomerase RNA level limits telomere maintenance in X-linked dyskeratosis congenita. Genes Dev. 20, 2848-2858. Xia, X. and Holcik, M. (2009). Strong eukaryotic IRESs have weak secondary structure. PLoS One. 4, e4136. Yamaguchi, H., Calado, R.T., Ly, H., Kajigaya, S., Baerlocher, G.M., Chanock, S.J., Lansdorp, P.M., and Young, N.S. (2005). Mutations in TERT, the gene for telomerase reverse transcriptase, in aplastic anemia. N Engl J Med. 352, 1413-1424. Yamamoto, H., Nakashima, N., Ikeda, Y., and Uchiumi, T. (2007). Binding mode of the first aminoacyl-tRNA in translation initiation mediated by Plautia stali intestine virus internal ribosome entry site. J Biol Chem. 282, 7770-7776. Yang, Q. and Sarnow, P. (1997). Location of the internal ribosome entry site in the 5' noncoding region of the immunoglobulin heavy-chain binding protein (BiP) mRNA: evidence for specific RNA-protein interactions. Nucleic Acids Res. 25, 2800-2807. Yoon, A., Peng, G., Brandenburger, Y., Zollo, O., Xu, W., Rego, E., and Ruggero, D. (2006). Impaired control of IRES-mediated translation in X-linked dyskeratosis congenita. Science. 312, 902-906. Yusupov, M.M., Yusupova, G.Z., Baucom, A., Lieberman, K., Earnest, T.N., Cate, J.H., and Noller, H.F. (2001). Crystal structure of the ribosome at 5.5 A resolution. Science. 292, 883-896. Zavialov, A.V. and Ehrenberg, M. (2003). Peptidyl-tRNA regulates the GTPase activity of translation factors. Cell. 114, 113-122. Zavialov, A.V., Hauryliuk, V.V., and Ehrenberg, M. (2005). Splitting of the posttermination ribosome into subunits by the concerted action of RRF and EF-G. Mol Cell. 18, 675686. Zebarjadian, Y., King, T., Fournier, M.J., Clarke, L., and Carbon, J. (1999). Point mutations in yeast CBF5 can abolish in vivo pseudouridylation of rRNA. Mol Cell Biol. 19, 7461-7472.  178  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.  179  

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.24.1-0071870/manifest

Comment

Related Items