The role of eukaryotic translation initiation factor 4E (eif4e) regulation during viral infection

Transcription

1 The role of eukaryotic translation initiation factor 4E (eif4e) regulation during viral infection Barbara Maria Herdy Department of Biochemistry McGill University Montreal, Quebec September, 2009 A thesis submitted to the faculty of Graduate Studies and Research in partial fulfilment of the requirements for the degree of Doctor of Philosophy Barbara M. Herdy, 2009

2 Abstract Translation of mrna into protein is a fundamental process and requires tight regulation. Primary control occurs at the initiation step. A critical protein for this regulation is the eukaryotic translation initiation factor 4E (eif4e), which binds to the 5 cap structure found on all nuclear transcribed mrnas. This interaction initiates translation by assembling the eif4f complex on the mrna, which subsequently recruits the ribosome. The function of eif4e is regulated in two ways, by the 4E binding proteins (4E-BPs), which disrupt the eif4f complex and secondly by eif4e phosphorylation on serine 209. However, the consequences of eif4e phosphorylation are not clearly understood. Cells continuously encounter pathogens including viruses. Lacking their own metabolic machinery, viruses rely on the translational apparatus of the host to produce their proteins. During many viral infections reprogramming of the cellular translation system occurs, favoring translation of viral mrnas. Altering the eif4f complex is one means by which reprogramming occurs. Picornaviruses initiate translation of their mrnas independent of eif4e through an internal ribosomal entry site (IRES). Work in this thesis demonstrates that eif4e still impacts picornavirus mrna translation, but only when viral mrnas compete with cellular RNAs. IRES mediated translation was strongly enhanced when eif4e abundance was decreased in an in vitro system by addition of 4E-BP or in vivo by sirna knock down. Decreased eif4e abundance reduced cap-dependent translation, but stimulated IRES mediated translation. Viral invasion is immediately recognized by the cell and accordingly, several defense mechanisms are activated. One of them entails secretion of type I interferon. This thesis shows that phosphorylation of eif4e on serine 209 impedes type I interferon production. Cells carrying an altered eif4e phosphorylation site, impairing eif4e phosphorylation, secreted elevated levels of type I interferon upon stimulation with synthetic RNA. Consequently, these cells were more protected against infection with interferon sensitive viruses. In conclusion, regulation of eif4e is important for picornavirus infection and plays a key role in generating a potent antiviral response. II

3 Résumé La traduction des ARN messagers (ARNm) en protéines est un processus essentiel qui requiert une étroite régulation. Le principal contrôle a lieu à l étape de l initiation. Une protéine essentielle pour cette régulation est le facteur d initiation de la traduction 4E (eif4e), qui se lie à la structure de la coiffe présente en 5 de tous les ARNm transcrits par le noyau. Cette interaction initie la traduction par l assemblage du complexe eif4f sur l ARNm, ce qui permet le recrutement des ribosomes. La fonction d eif4e est régulée d une part par les protéines se liant à 4E (4E-BPs) qui dissocient le complexe eif4f, et d autre part par la phosphorylation d eif4e sur la sérine 209. Toutefois, les conséquences de cette dernière ne sont pas clairement comprises. Les cellules rencontrent constamment des pathogènes, dont des virus. Ces derniers étant dépourvue de machinerie métabolique propre, ils se doivent d utiliser l appareil traductionnel de l hôte pour exprimer leurs protéines. De nombreuses infections virales résultent en un changement du système de traduction cellulaire favorisant la traduction de protéines à partir des ARNm viraux. L altération du complexe eif4f est l un des moyens utilisés lors de cette reprogrammation. Les Picornavirus initie la traduction de leurs ARNm indépendamment d eif4e par le biais d un site d entrée interne des ribosomes (IRES). Le travail effectué lors de cette thèse démontre qu eif4e peut toujours influencer la traduction des ARNm des picornavirus, mais uniquement lorsque ceux-ci entrent en compétition avec les ARN cellulaires. La traduction à partir de l IRES est accrue lorsque la quantité d eif4e disponible est réduite dans un système in vitro par addition de 4E-BP, ou in vivo par un ARN interférant. La diminution des niveaux d eif4e réduit la traduction dépendante de la coiffe, mais stimule la traduction par l IRES. Les infections virales sont reconnues immédiatement par les cellules et engendrent l activation de plusieurs mécanismes de défense. L un d entre eux implique la sécrétion d interféron de type I. Nous avons démontré que la phosphorylation d eif4e sur la sérine 209 entrave cette production d interféron de type I. Des cellules dont le site de phosphorylation d eif4e a été altéré pour empêcher sa phosphorylation, produisent des niveaux plus élevés d interféron de type I suite à une stimulation par des ARN synthétiques. En conséquence, elles sont mieux protégées contre les infections virales provenant de virus sensibles aux interférons. En conclusion, la régulation d eif4e est importante lors des infections par les picornavirus et elle joue un rôle essentiel dans la genèse d une réponse virale efficace. III

4 PREFACE This thesis is a compilation of one published manuscript, which I contributed as a second author and one manuscript in preparation, which I am the first author. Chapter 2 Svitkin YV, Herdy B, Costa-Mattioli M, Gingras AC, Raught B and Sonenberg N. (2005). Eukaryotic translation initiation factor 4E availability controls the switch between cap-dependent and internal ribosomal entry site-mediated translation. Mol Cell Biol. Chapter 3 Barbara Herdy, Yuri V. Svitkin, Liwei Rong, Luc Furic, Ryan J. Dowling, Amy B. Rosenfeld, Annie Silvestre, Nahum Sonenberg. (manuscript in preparation) Translation initiation factor eif4e phosphorylation impairs type I interferon production. Acknowledgements Chapter 2 Dr. Svitkin wrote the manuscript and prepared Figures 1, 2 and 3 of the manuscript. Dr Costa-Mattioli provided expert advice. Dr. Gingras and Dr. Raught provided constructs and their expert advice. I established Figure 4 and Supplemental figures 2 and 3 in this manuscript. Chapter 3 Dr Svitkin co-supervised me on the project as well as helped me to write and review the manuscript. Dr. Rosenfeld provided profound expert advice, guidance and was very helpful writing and reviewing the manuscript. Dr. Rong prepared the eif4e S209A knock in mouse and isolated mouse embryonic fibroblasts. Dr. Furic was responsible for backcrossing mice into a pure Balb/c and C57/black6 background. Ryan Dowling was helpful conducting plaque assay analysis. Annie Silvestre looked over the mouse housing and breeding. I performed all other experiments and wrote the manuscript. IV

5 Acknowledgements I want to thank my supervisor Dr. Nahum Sonenberg for allowing me to work in his laboratory. I am very grateful for this experience. Furthermore, I want to thank all past and present members of the Sonenberg laboratory for useful discussions and help over the years. Especially, I want to thank Colin, Pam, Sandra and Annie for providing technical experience and administrative help which alleviated my every day workload! I also want to take the opportunity to thank Dr. Yuri Svitkin who supported me and my work throughout the years. I also want to thank him for correcting this thesis. I want to thank Dr. Amy Rosenfeld, who took the time to help me improve my written and spoken English. More importantly, she was always interested in discussing science at any time, which helped me to grow as a scientist. Furthermore, I want to thank her for taking the time to correct this thesis. I want to thank Laurent for translating the abstract into French. I have to thank both of my RAC committee members Dr. Pelletier and Dr. Pause. It was always very helpful and entertaining to brain store with my RAC committee members and Colin over a beer, or two at Thomson House. Thank you! I want to thank Meena who always prepared great Indian food for me. She made me feel as if I was part of her family. Special thanks goes to all of my dear friends who supported me and got me out of the lab when I needed a break so desperately. I will always remember the trips we did together! Also exploring new restaurants in Montreal was a lot of fun! I will never forget the sports challenges: ice hockey (Knock Outs - Knock them out!), the softball battles on the reservoir and the volleyball games. Thank you for all these experiences! I am also very grateful that I had the opportunity to be invited to two weddings in Canada! I also need to thank Ryan and Kfir, who were helping me out in the lab uncounted times and made me laugh! A huge thank you note goes to Regina my current roommate and Jamie my first roommate in Canada. Thank you for being there for me when I needed you! Ich will mich auch besonders bei den Menschen bedanken, die absolut zu mir standen. Danke an meinen Vater, die Familie Süss und die Radlspäck Familie! I gained a lot of pleasant memories while I was living in Montreal and I will remember this time my entire life and look back with a smile. I have met so many great and interesting people and I hope I made a lot of friends. I will never forget any of you! V

6 Table of Contents ABSTRACT... II RÉSUMÉ...III PREFACE...IV ACKNOWLEDGEMENTS...V TABLE OF CONTENTS...VI LIST OF FIGURES AND TABLES...VIII CHAPTER 1: GENERAL INTRODUCTION...IX 1. TRANSLATIONAL CONTROL IN EUKARYOTES EUKARYOTIC TRANSLATION INITIATION mrna Properties The 5 Cap Structure and 3 Untranslated Regions (UTRs) The Start Codon The Poly(A) Tail Cap-dependent Translation Initiation Components of the eif4f Complex eif4e Three Dimensional Structure of eif4e eif4e Regulation eif4g eif4a TRANSLATIONAL CONTROL DURING VIRAL INFECTION VIRUS ORGANIZATION AND REPLICATION HOST CELL SHUTOFF STRATEGIES FOR VIRAL MRNA TRANSLATION Internal Ribosomal Entry Sites (IRES) Type I and II IRESs of Picornaviruses Specialized 5 -end Dependent Initiation Ribosome Shunting Leaky Scanning Reinitiation Ribosomal Frameshifting Suppression of termination MODIFICATION OF TRANSLATIONAL SYSTEM DURING VIRAL INFECTION Modifications on the eif4f Complex eif4e eif4g Inhibition of the Closed Loop Formation HOST DEFENSE MECHANISMS AGAINST VIRAL INFECTION TYPE I INTERFERON SYSTEM PATTERN RECOGNITION RECEPTORS (PRRS) Toll-like Receptors (TLRs) TLR2 and TLR TLR TLR7, 8 and VI

7 Cytoplasmic Pattern Recognition Receptors Type I Interferon Feedback Loop Interplay of Type I Interferon System and Translational Control Mechanisms P38/MAPK pathway and interferon signaling PI3K Pathway and its Role in Interferon Mediated Response CHAPTER 2.: EUKARYOTIC TRANSLATION INITIATION FACTOR 4E AVAILABILITY CONTROLS THE SWITCH BETWEEN CAP-DEPENDENT AND INTERNAL RIBOSOMAL ENTRY SITE-MEDIATED TRANSLATION ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS CHAPTER 3.: EUKARYOTIC TRANSLATION INITIATION FACTOR EIF4E PHOSPHORYLATION IS INHIBITORY TO INTERFERON TYPE I RESPONSE ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIAL AND METHODS CHAPTER 4.: DISCUSSION AND FUTURE DIRECTIONS REFERENCES APPENDIX VII

8 List of Figures and Tables Chapter 1 - Introduction Figure 1. mrna Features. 3 Figure 2. Cap dependent translation initiation. 7 Figure 3. Structural features of eif4e. 10 Figure 4. Signaling through the MAP kinase pathway and the mtor pathway. 12 Figure 5. Representation of mammalian eif4g homologues. 15 Figure 6. Overview of eukaryotic virus life cycle. 19 Figure 7. Unconventional strategies for translation of viral mrnas. 21 Figure 8. Bicistronic mrnas. 23 Figure 9. Cellular pathogen-recognition receptors (PRRs). 35 Figure 10. TLR mediated signaling in conventional dendritic cells. 37 Figure 11. Cytoplasmic pattern recognition receptors. 40 Chapter 2 - Eukaryotic Translation Initiation Factor 4E Availability Controls the Switch between Cap-Dependent and Internal Ribosomal Entry Site-Mediated Translation Figure 12. Regulation of cap-dependent and cap-independent translation by effectors of eif4f function in Krebs-2 cell extracts. 57 Figure 13. EMCV mrna translation, RNA replication, and virus yield in the untreated EMCV mrna-programmed S10 extracts. 60 Figure 14. 4E-BP1 and eif4e have no effect on EMCV protein synthesis and replication in nuclease-treated Krebs-2 cell extract. 62 Figure 15. The abundance of eif4f components and polyribosome distribution as affected by eif4e knockdown. 64 Figure 16. eif4e knockdown stimulates PV protein synthesis and replication. 65 Figure 17. eif4e knockdown stimulates translation and replication of EMCV in vivo. 67 Figure 18. A model explaining eif4f regulation of mrna competition in EMCV infected cells. 71 Table 1. Positions and sequences of the sirnas used to knockdown gene expression 76 Chapter 3 Eukaryotic Translation Initiation Factor eif4e Phosphorylation is Inhibitory to Interferon Type I response. Figure 19. Replacement of the eif4e phosphorylation site serine 209 to alanine in MEFs delays VSV replication. 82 Figure 20. KI MEFs from C57BL/6 background and Balb/c background reduced VSV infection. 84 Figure 21. MEFs deficient of the eif4e Kinases Mnk1 and 2 show a delay in VSV infection. 85 Figure 22. The Mnk inhibitor CGP57380 protects WT cells and KI cells from VSV infection. 87 Figure 23. EMCV, SV and Influenza A virus infection is reduced in KI MEFs. 89 Figure 24. KI MEFs show increased type I interferon production after stimulation with poly(i:c). 90 Figure 25. Neutralizing antibodies against IFN- render KI cells more susceptible to VSV infection. 92 Figure 26. Less recombinant interferon / is needed to protect KI cells against virus infection than WT MEFs. 94 VIII

9 Chapter 1: General Introduction IX

10 1. Translational Control in Eukaryotes Gene expression in eukaryotic cells underlies regulation at multiple levels, including transcription, splicing, nuclear export and translation. Once mrnas are transcribed in the nucleus they venture through nuclear processing events and are exported into the cytoplasm where they are engaged in translation. Briefly translation can be divided into four steps: Initiation, elongation, termination and ribosome recycling (M. B. Mathews 2007). Translation of a given mrna into the encoded protein is initiated by recruitment of the eukaryotic translation initiation factors (eifs), which assemble on the 5 end of the mrna. Once the eifs have placed themselves on the mrna the 40S ribosomal subunit is recruited (M. B. Mathews 2007). The complex scans along the 5 untranslated region (UTR) of the mrna until the first start codon (AUG) is found. When the 60S ribosomal subunit is recruited the eifs are released. Then the ribosome scans along the mrna and aminoacyl trnas are matched to the mrna, which results in establishment of a peptide bond between amino acids on the trnas. This step is called translation elongation. Translation is terminated when the processing ribosome encounters a stop codon. The grown polypeptide is released and the ribosomal subunits dissociate from the mrna so they engage in translation of a new mrna. This process is referred to as translation termination and recycling. Protein synthesis is a highly energy consuming biosynthetic process and needs to be tightly monitored. Although, all four steps of translation are subjected to regulation, translation is mainly regulated at the initiation step. Presumably, it is more efficient to control translation in its onset rather then interrupt the synthesis later on, when the cell has to deal with the consequences of aberrant protein synthesis. Furthermore, transcription, processing and export of mrnas are time consuming processes which cannot be implemented when the cell has to respond to environmental changes. Control of translation initiation on existing mrnas bypasses the need for nuclear mrna synthesis and therefore allows for more rapid adjustments to the environment. 1

11 1.1. Eukaryotic Translation Initiation Control of translation occurs mainly at the initiation step. This process employs several eukaryotic translation initiation factors, which organize the complex sequence of events (M. B. Mathews 2007). To accomplish regulatory variety and to introduce a connection to certain cellular control pathways, several initiation factors can be transiently modified. In addition to modulation of translation factors, mrnas contain structures in their 5 and/or 3 untranslated regions which determine their specific translational fate without influencing global protein synthesis mrna Properties Eukaryotic mrnas are transcribed from their corresponding genes in the nucleus. The mrnas are modified cotranscriptionally with the 5 cap structure and a poly(a) tail at the 3 end (Maniatis and Reed 2002; Proudfoot and O'Sullivan 2002). Once the introns are removed the mrna is exported into the cytoplasm where it is engaged in translation. Efficiently translated eukaryotic mrnas contain four features: the 5 cap structure, an open reading frame (ORF), a 5 and 3 untranslated region (UTR) and a poly(a) tail (Fig. 1. A.) The 5 Cap Structure The m 7 GTP cap structure (m 7 GpppN, where N is any nucleotide) is located at the 5 end of most cellular mrnas and plays an important role in the initiation of translation (Shatkin 1976). Although the cap is not essential for translation initiation, the efficiency of translation is significantly enhanced when the cap structure is present. An additional stimulation is seen when the 5 cap structure synergizes with the poly(a)tail (Gunnery et al. 1997; Kahvejian et al. 2001). The cap structure consists of a 7 methylated guanine. The modulated ribonucleotide is connected by a 5 -to-5 linkage consisting of three phosphates to the first ribonucleotide (N) on the RNA (Fig. 1.B). In most eukaryotes, with exception of plants and fungi, the first base of the mrna connecting to the cap structure (N) is also methylated at the 2 2

12 Figure 1. A. mrna Features. A) The 5 cap structure (m 7 GpppN) and the poly(a) tail ((A n ) are canonical motives which promote translation initiation. The 5 UTRs (untranslated regions) can contain structural features. While secondary structures like hairpins inhibit translation by increasing the need for extensive unwinding of the mrna, internal ribosomal entry sites (IRES), which encompass excessive secondary structures, actually mediate cap-independent translation initiation. Upstream open reading frames (uorfs) are another feature by which mrna translation can be regulated. uorfs are considered as negative regulators of translation as they reduce translation from the main ORF. Blue ovals represent binding sites for proteins or/and RNA regulators, which inhibit translation. B) The cap structure. The 5 cap structure is added cotranscriptionally in the nucleus and consists of guanine, which is methylated at the 7 position. The modulated base is linked to ribose which is connected to the first ribo-nucleotide in the mrna by a 5-5 bridge containing 3 phosphates. 3

13 A adapted from Gebauer and Hentze 2004 B adapted from von der Haar 2004 Figure 1

14 hydroxyl of the ribose (Banerjee 1980). The cap structure specifically binds the translation initiation factor 4E (eif4e) (Sonenberg et al. 1978; Sonenberg et al. 1980). eif4e is the cap binding protein and mediates the first events required for ribosome binding to mrnas. The presence of the cap also influences pre-mrna splicing, nucleocytoplasmic transport of mrnas, polyadenylation, and mrna stability (Izaurralde et al. 1992; Izaurralde et al. 1994; Cooke and Alwine 1996; Visa et al. 1996; Ishigaki et al. 2001; Lejeune et al. 2002) and 3 Untranslated Regions (UTRs) The translation efficiency of most mrnas is affected by modification of translationinitiation factors (Dever 2002). In contrast, regulation of specific mrnas is provided by several features in the UTRs of these mrnas (Gebauer and Hentze 2004). UTRs are non coding regions and can be found on both extremities on mrnas. These regions are specific to each mrna species and can determine its translational fate. The translational efficiency of eukaryotic mrna is limited by the rate of initiation (Palmiter 1972). Usually, 5 UTRs do not contain strong secondary structures, which can severely affect ribosomal recruitment, scanning and recognition of the initiation codon (Pelletier and Sonenberg 1985; Koromilas et al. 1992). An exception is a specialized secondary structure located in the 5 UTR of some mrnas, which is called internal ribosomal entry site (IRES). This structure was initially identified in the picornavirus genome (Jang et al. 1988; Pelletier and Sonenberg 1988; Hellen and Sarnow 2001b; Bushell and Sarnow 2002). Picornavirus mrnas lack the 5 cap structure but are able to recruit the translational apparatus via the IRES element, which directs the initiation complex directly to its internal sites on the RNA. Also cellular mrnas carry IRES elements, approximately 3-5% of the total mrna pool (Johannes et al. 1999a). These mrnas show a decreased dependence on cap mediated translation initiation, which is necessary to maintain production of their proteins when canonical cap-dependent translation is compromised, for example during mitosis (Qin and Sarnow 2004) or during stress conditions, as apoptosis (Bushell et al. 2006). Such factors are mostly involved in cell growth, proliferation, differentiation and survival (Holcik and Sonenberg 2005b). Another regulatory element found in 5 UTR are upstream open reading frames (uorfs). uorfs generally inhibit translation of the downstream ORFs. Only during stress 4

15 situations, post-terminal scanning can occur resulting in re-initiation at the main downstream ORF (Hinnebusch 1997; Vattem and Wek 2004). The 3 UTR of mrnas is also a region mediating translational control. The segment is rich in cis-acting elements that determine mrna stability, localization in the cytoplasm and translational efficiency. These cis-acting elements underlie regulation by trans-acting factors, which can be proteins or as recently described mirnas (Kuersten and Goodwin 2003; Sonenberg and Hinnebusch 2009) The Start Codon Initiation occurs predominantly at the first AUG triplet, the start codon, on the mrna. The optimal initiation codon selection in mammalian cells is GC(A/G)CCAUGG (Kozak 1991). If ribonucleotides at position -3 and +4 from the AUG are changed, initiation will occur downstream at the next AUG triplet The Poly(A) Tail Another critical cis-acting element on mrnas is the poly(a) tail. It is positioned at the 3 end of mrnas. The only cellular mrnas which do not carry a poly(a) tail are histone mrnas (Marzluff et al. 2008). Several lines of evidence suggest that mrna stability and even mrna export is influenced by the poly(a)tail (Munroe and Jacobson 1990; Kahvejian et al. 2001; Wilusz et al. 2001). Most importantly, the poly(a) tail stimulates translation as it forces the mrna into a closed loop conformation (Tarun and Sachs 1995). This is accomplished by the poly(a) binding proteins (PABPs), which are associated with the poly(a) tail (Mangus et al. 2003; Gorgoni and Gray 2004) and the eukaryotic translation initiation factor 4G (eif4g)-eif4e complex (section ), which is located at the 5 end of the mrna Cap-dependent Translation Initiation Translation initiation in eukaryotes can be defined as a process that leads to assembly of the 80S ribosome on the mrna. Translation initiation is very complex and requires several factors including the methionyl-transfer RNA (met-trna met i ), the 40S and the 60S ribosomal subunit and 12 eukaryotic initiation factors and energy in form of ATP and GTP (Holcik and Sonenberg 2005b; M. B. Mathews 2007). 5

16 A key step in initiating protein synthesis is the formation of the 43S pre-initiation complex (Fig. 2). It entails recruitment of the initiator trna (Met-tRNA met i ) to the 40S ribosome. This step requires the function of eif2. Together with GTP and the MettRNA met i, eif2 forms the ternary complex. eif2 is a heterotrimer composed of the, and subunits. Phosphorylation of eif2 on serine 51 inhibits release of GDP from the complex, which is mediated by the nucleotide exchange factor (NEF) eif2b (Nika 2000; Williams 2001). Therefore, when eif2 is phosphorylated translation initiation is inhibited because only when eif2 is bound to GTP, the protein is in the appropriate conformation to interact with the Met-tRNA met i (Kapp and Lorsch 2004a). Phosphorylation of eif2 is mediated by four stress inducible kinases, GCN2, PKR, PERK and HRI (Holcik and Pestova 2007). These kinases are mediators of the integrated stress response (ISR) which is activated upon heat shock, hypoxia, amino acid and serum deprivation and viral infection. PKR especially, has a key role in the antiviral host defense mediated by interferon. After the ternary complex is assembled it associates with the 40S ribosomal subunit, which is preassembled with eif1, eif1a, eif3 and eif5 forming the 43S pre-initiation complex (Trachsel et al. 1977; Benne et al. 1978; Peterson et al. 1979; Chaudhuri et al. 1999; Battiste et al. 2000; Majumdar et al. 2003). The 43S complex is now recruited to the 5 end of mrnas. This is facilitated by the eif4f complex. Assembly of the eif4f complex on the mrna is the rate limiting step of translation initiation and requires interaction of eif4e with the m 7 G cap. Once eif4e is bound to the cap structure eif4a, an RNA helicase and eif4g, a large scaffolding protein, are recruited forming the eif4f complex (Gingras et al. 1999b). Once eif4f complex and the 43S pre-initiation complex are assembled on the mrna, the entire complex scans in 5 to 3 direction along the mrna. eif4a in the eif4f complex is activated by eif4b and eif4h and unwinds cap proximal secondary structures in an ATP dependent manner until the first AUG start site for initiation is recognized (Ray et al. 1985; Rozen et al. 1990; Jaramillo et al. 1991). Now initiation factors must be removed from the 40S ribosomal unit and the 60S ribosomal unit needs to be recruited. 6

17 Figure 2. Cap dependent translation initiation. The small 40S ribosomal subunit loaded with eif3 and eif1a associates with the ternary complex, which is composed of the initiator trna (Met-tRNA i met ), eif2 and GTP, to form the 43S pre-initiation complex. eif2b, a nucleotide exchange factor, regulates ternary complex formation by activating of the complex through exchanging GDP for GTP. Simultaneously, another complex, the eif4f complex, is formed on the 5 cap structure of the mrna. The eif4f complex is composed of the eif4e subunit, which binds directly to the cap, eif4g, a large scaffolding protein and eif4a, a RNA helicase. When the eif4f complex is successfully formed on the mrna the 43S pre-initiation complex joins and the complex scans along the mrna until the first AUG is found. The scanning is promoted by the unwinding of the mrna by eif4a, in conjunction with eif4b and eif4h. After identification of the AUG by the anticodon of the initiator trna, initiation factors are released and the 60S ribosomal subunit joins to form an elongation competent 80S ribosomal initiation complex. 7

18 adapted from von der Holcik 2007 Figure 2

19 This is accomplished as soon as the correct base pairing between Met-tRNA Met i anticodon and the AUG codon is confirmed (Cigan et al. 1988). Subsequently hydrolysis of GTP bound to eif2 takes place with the assistance of the GTPase-activating protein eif5 (Huang et al. 1997). These events lead to dissociation of pre-assembled eifs and formation of the 40S initiation complex. The second guanine nucleotide-binding (G) protein, eif5b, then stimulates joining of the 60S subunit with the 40S initiation complex to form the 80S complex. After initiation is completed, elongation factors are recruited to facilitate elongation of the polypeptide chain Components of the eif4f Complex The eif4f complex is composed of three proteins, eif4e, the cap binding protein, eif4a, an RNA helicase and eif4g, a large scaffolding protein. Formation of this complex is crucial for efficient cap-dependent translation initiation (Gingras et al. 1999b; Holcik and Sonenberg 2005b; M. B. Mathews 2007; Sonenberg and Hinnebusch 2009) and it is thought to be the rate limiting step which determines the fidelity of translation initiation. Thus, this process is a frequent target for translational control mechanisms altering eif4f activity (Gingras et al. 1999b) eif4e eif4e is a 24 kda protein which specifically binds to the 5 cap structure present on most cellular mrnas (Sonenberg et al. 1978). The protein is evolutionary conserved. Structural analysis by NMR and the X-ray crystallography of eif4e in the presence of m 7 GDP cap analog revealed the conservation of the cap binding mechanism among eukaryotes (Marcotrigiano et al. 1997b; Matsuo et al. 1997). eif4e is indispensable for cap-dependent translation and is limiting in some initiation systems (Duncan and Hershey 1987; Gingras et al. 1999b). When eif4e and associated proteins were depleted from cell-free extracts translation of capped mrnas was drastically reduced, which was restored by addition of recombinant eif4e (Svitkin et al. 1996). This underscores the importance of eif4e as a crucial factor in translation initiation. Therefore, as the ratelimiting factor for cap-dependent translation initiation, eif4e is a target of extensive regulation. 8

20 Three Dimensional Structure of eif4e The three-dimensional structure of eif4e was solved by X-ray crystallography using mouse eif4e and by NMR, which examined yeast eif4e (Marcotrigiano et al. 1997a; Marcotrigiano et al. 1997b; Matsuo et al. 1997). Both proteins showed the same overall organization: a curved -sheet with eight anti-parallel strands backed by three long helices (Fig. 3.A and B). This structure resembles a cupped hand or baseball glove. Cocrystal structure of murine eif4e bound to m 7 GDP cap analog revealed that 7-methyl recognition takes place on the concave surface of eif4e and is facilitated by three types of interactions: - stacking, hydrogen bonds and Van der Waals interactions. The residues involved in these interactions are highly conserved from yeast to mammals (Fig. 3.C). - ring stacking is formed by stacking of the indole group of two tryptophan residues W56 and W102 (numbering is according to mouse eif4e) on the concave surface of eif4e with the guanidine in m 7 GDP. Hydrogen bonds form between E103 of eif4e and N-1 and N-2 hydrogens on the guanine base of the cap structure. Furthermore, W102 of eif4e is a hydrogen donor for O-6 of the guanine ring of the cap. Van der Waals interactions can be detected between the N-7 methyl group of guanidine in the cap structure and W166 in eif4e eif4e Regulation Two main regulation mechanisms modulate eif4e function. One entails regulation by a family of small proteins, the 4E binding proteins (4E-BPs), which specifically bind to eif4e (Pause et al. 1994a). The second regulatory mechanism is phosphorylation of eif4e by the Mnk kinases at serine 209 (Waskiewicz et al. 1999). eif4e Regulation by eif4e binding proteins (4E-BPs) One of the best understood translational control mechanism is the control of the eif4f complex formation by the 4E-BP family of translation repressors (Pause et al. 1994a; Gingras et al. 1999b; Gingras et al. 2001a). The 4E-BPs regulate eif4e activity by competing with eif4g for the same conserved binding site on eif4e (conserved binding site YxxxL, where denotes any hydrophobic amino acid and x any amino acid) (Mader et al. 1995). Once eif4e binds to the 4E-BPs, eif4e is sequestered from the eif4f complex and cap dependent translation initiation is inhibited. Mice and humans express three isoforms of 4E-BPs. 4E-BP1 is predominantly expressed but especially high 9

21 Figure 3. Structural features of eif4e. The three-dimensional structure of eif4e (blue) bound to the m 7 GDP cap analog (yellow) is shown. A) Ribbon diagram of the mouse eif4e structure determined by X-ray crystallography; B) Ribbon diagram of the yeast eif4e structure as determined by NMR; C) magnified view of the cap-binding area of mouse eif4e. The amino acid side chains involved in the interaction are highlighted. - ring stacking is formed between W56, W102 and the guanidine of the cap analog. Hydrogen bonds form between E103 on eif4e and N-1 and N-2 hydrogens on the guanidine base of the cap analog. Van der Waals interactions are detected between W166 on eif4e and N-7 of the methyl group of guanidine of the cap. 10

22 A B C W102 E103 W166 R157 K162 R157 D90 W56 adapted from Gingras 1999 Figure 3

23 levels are detected in adipose tissues. 4E-BP1 knock out mice show differences in metabolic rate and adipose tissue content (Tsukiyama-Kohara et al. 2001). 4E-BP2 is primarily expressed in the brain and analysis of the 4E-BP2 knock out mice showed that these mice have impaired long-term memory formation (Tsukiyama-Kohara et al. 2001; Banko et al. 2004). The third isoform 4E-BP3 is mainly expressed in the colon, liver and kidney (Poulin et al. 1998; Tsukiyama-Kohara et al. 2001). The 4E-BPs are themselves regulated. Their binding activity with eif4e is influenced by their phosphorylation status (Lin et al. 1994; Pause et al. 1994a; Gingras et al. 2001a). This modification was mainly studied using 4E-BP1. In their hypophosphorylated stated the 4E-BPs bind tightly to eif4e, whereas they do not bind to eif4e in their hyperphosphorylated state. Phosphorylation is controlled by the PI3K/AKT/mTOR pathway (Fig. 4). The mammalian target of rapamycin (mtor) in a complex with regulatory associated protein of mtor (Raptor), mammalian LST8/G-protein -subunit like protein (mlst8/g L) and the recently identified partner PRAS40 (the complex is called mtorc1) phosphorylates 4E-BP1 on Thr 37 and Thr 46 (Gingras et al. 2001a; Kim et al. 2002; Thomson et al. 2009). Phosphorylation on these sites is essential for subsequent phosphorylation at the serum-sensitive residues S65, Thr 70 and S83 by the PI3K/AKT pathway (Gingras et al. 2001a). As mentioned above, hypophosphorylated 4E-BPs exhibit increased affinity for eif4e and thereby sequesters the protein from the eif4f complex. Therefore, inhibition of the upstream kinase mtor by rapamycin has been explored as chemotherapeutic agent (Petroulakis et al. 2006). eif4e phosphorylation eif4e is a target of the mitogen-activated protein kinase (MAPK) signaling pathway and the MEK/ERK pathway (Fig. 4) (Roux and Blenis 2004). Activation of ERK or p38/mapk results in phosphorylation of the Mnk kinases (Fukunaga and Hunter 1997; Pyronnet et al. 1999; Waskiewicz et al. 1999), which then phosphorylate eif4e on serine 209 (Flynn and Proud 1995; Joshi et al. 1995; Waskiewicz et al. 1997; Waskiewicz et al. 1999). The Mnk kinases physically interact with eif4g, which localizes the kinases in close proximity to their target (Pyronnet et al. 1999). The phosphorylation status of eif4e is also controlled by the protein phosphatase 2A (PP2A) (Bu and Hagedorn 1992). This 11

24 Figure 4. Signaling through the MAP kinase pathway and the mtor pathway. The PI3K/AKT/mTOR pathway (in blue) is activated by growth factors, nutrients like amino acids and glucose, mitogens and hormones. Activation of phosphoinositide-3-kinase (PI3K) results in phosphorylation of downstream target AKT. AKT then phosphorylates tuberous sclerosis complex-2 (TSC2), which destabilizes the TSC1/2 complex and results in accumulation of Ras homologue enriched in brain (Rheb) bound to GTP. GTP-Rheb leads to activation of mtor. The mtor kinase in complex with RAPTOR and mlst8/g L is called the mtorc1 complex. This complex phosphorylates several targets such as S6 kinase and the 4E-BPs. Phosphorylated 4E-BPs cannot interact with eif4e, therefore cap-dependent translation initiation will occur. The second pathway regulating eif4e is the Ras/Raf/MAP kinase pathway (in green and yellow). This pathway is activated by growth factors, cytokines and stress. Activation of Ras and MEKK leads to a cascade of events such as JNKK and MAPK activation which culminates in Mnk activation. Mnk phosphorylates eif4e within the eif4f complex. Crosstalk between both pathways has been reported. Ras has several targets, for example MEKK, Raf and also PI3K. ERK has been shown to activate TSC1/2 complex and AKT inhibits Raf by phosphorylation. 12

25 stress, cytokines growth factors, hormones, mitogenes, cytokines RAS PI3K MEKK AKT RAF JNKK MEK TSC1/2 p38 ERK Rheb GTP mtorc1 MNK Rapamycin Raptor mtor GβL P eif4e 4E-BPs P P S6K adapted from Costa-Mattioli 2008 and Mamane 2005 Figure 4

26 phosphatase is a major Ser/Thr phosphatase and regulates effector proteins involved in numerous cellular processes. The biological significance of eif4e phosphorylation is not determined and its role in translation initiation is controversial (Scheper and Proud 2002; Scheper et al. 2002; Buxade et al. 2008). In some cases, eif4e phosphorylation and fidelity of translation go hand in hand. For instance, stimulation of cells with growth factors and mitogenes increases translation rates and also eif4e phosphorylation (Klein and Melton 1994; Morley 1997; Wang et al. 1998). Flies expressing Drosophila eif4e phosphorylation mutants show delays in development, reduced cell and organism size in adults (Lachance et al. 2002). Therefore, it was assumed that eif4e phosphorylation stimulates capdependent translation. However, the effects of eif4e phosphorylation on cap-dependent translation rates in vitro are insignificant. Also treatment with stress inducing agents as arsenite or anisomycin result in phosphorylation of eif4e while general translation is reduced (Wang et al. 1998). It was also shown that eif4e hyperphosphorylation by overexpression of Mnk kinases in mammalian cells or Drosophila leads to reduction of translation or/and growth rates (Knauf et al. 2001). Furthermore, mice lacking Mnk 1 and 2 kinases showed no change in global translation rates and developed normally (Ueda et al. 2004). Similarly, a Drosophila mutant containing a loss of function mutation in Lk6 (Drosophila Mnk homologue) developed normally under standard conditions (Arquier et al. 2005; Reiling et al. 2005). However, when the nutrients were limiting a reduced organism size was observed. Taken together, it appears that eif4e phosphorylation is not critical under normal conditions, when the 4E-BPs are phosphorylated by mtor and the eif4f complex can form normally. Mnk activity and thereby the phosphorylation state of eif4e becomes important only when nutrient supply is low or different environmental stresses occur which result in dephophorylation of the 4E-BPs and consequently in limited eif4f complex formation. Further discrepancies concern interpretation of structural analysis data. It was initially reported that the phosphorylated form of eif4e had greater affinity for the cap than the unphophorylated form (Joshi et al. 1995). Recent reports, based on new crystal structure studies indicated the opposite (Scheper et al. 2002; Zuberek et al. 2004). Therefore, the 13

27 role and function of eif4e phosphorylation remains divisive (Sonenberg and Dever 2003). Nevertheless, several lines of evidence suggest that phosphorylation of eif4e has a distinct role in eif4e activity. Overexpression of eif4e causes the development of malignancies (Lazaris-Karatzas et al. 1990; Ruggero et al. 2004) but only when eif4e contains a functional S209 site allowing phosphorylation (Wendel et al. 2004). eif4e phosphorylation was shown to impact nucleocytoplasmic transport of cyclin D1 mrna, which is promoting transformation (Topisirovic et al. 2004). Additionally, Wendel et al. used activated T332D or kinase-dead T2A2 mutations of Mnk 1 (Waskiewicz et al. 1997) to promote tumorigenesis in their lymphoma mouse system. Activated Mnk1 but not the kinase-dead mutant was able to accelerate lymphoma generation. The eif4e phosphorylation status seems to be important during several other cellular processes. eif4e dephosphorylation is monitored during mitosis, after heat shock and infection with viruses (Bonneau and Sonenberg 1987; Duncan 1987; Huang and Schneider 1991). Precisely, dephosphorylation of eif4e occurs during infection with several viruses, such as VSV, EMCV, adenovirus and influenza virus at the late stage of infection (Kleijn et al. 1996). The role of eif4e phosphorylation during viral infection is unknown and needs to be further analyzed eif4g eif4g is the large scaffolding protein in the eif4f complex. Two isoforms, eif4gi and eif4gii, were identified in mammals and in yeast (Goyer et al. 1993; Yan and Rhoads 1995; Gradi et al. 1998a). The mammalian proteins are 46% identical and have a predicted mass of 171 kda and 176 kda, respectively. They share similar biochemical activities and in mammals are functionally interchangeable; however, they differ in their level of expression among various tissues (Gradi et al. 1998a). Recently, two novel members of the eif4g family have been cloned: p97/novel apolipoprotein B mrna editing enzyme (APOBEC)/target no.1 (NAT1)/death-associated protein-5 (DAP-5) and PABP-interacting protein 1 (Paip-1) (Craig et al. 1998) (Fig. 5). As a component of eif4f, eif4g mediates several activities. It coordinates the capbinding activity of eif4e and the RNA helicase activity of eif4a. Additionally, it facilitates the formation of the closed loop of the mrna necessary for efficient 14

28 Figure 5. Representation of mammalian eif4g homologues. Conserved protein binding sites are displayed. The poliovirus 2A pro cleavage site in eif4gi is also shown. 15

29 2A pro eif4a eif4a Mnk eif4gi PABP eif4e eif eif4a eif4a Mnk eif4gii eif4a 907 p97 eif4a 480 Paip-1 PABP adapted from Marintchev 2009 and Gingras 1999 Figure 5

30 translation by interacting with the poly(a)binding protein (PABP) situated on the poly(a)tail of the mrna. eif4g also interacts with eif3. Via the latter interaction, eif4g recruits the 43S pre-initiation complex to the mrna. To accomplish these functions and attesting to its scaffolding function, eif4g possesses binding sites for eif4e, eif4a, PABP and eif3 (see schematic diagram of eif4g binding sites). At the very carboxyl terminus eif4g contains additional binding sites for MAP-kinase-interacting kinase (Mnk) (Tarun et al. 1997; Pyronnet et al. 1999), TRAF2 and Pak2 (Groft and Burley 2002; Kim et al. 2005; Ling et al. 2005). eif4g itself is also able to bind RNA through the RNA-recognition-motive (RRM)-like RNA-binding portion found within the middle of the protein, the so called HEAT domain (Goyer et al. 1993; Marcotrigiano et al. 2001). This interaction is not only utilized during 5 -end dependent initiation (Yanagiya et al. 2009) but also required for internal initiation mediated by type I and II IRES elements (Lamphear et al. 1995; Ohlmann et al. 1996; Pestova et al. 1996b; Marcotrigiano et al. 2001; Thoma et al. 2004) eif4a The eif4a family of proteins contains eif4ai, eif4aii and eif4aiii (Nielsen and Trachsel 1988; Weinstein et al. 1997). Human eif4ai and eif4aii are highly homologous and are functionally equivalent, but they are expressed differently in certain tissues (Yoder-Hill et al. 1993) and have distinct roles during development (Nielsen and Trachsel 1988). eif4a is a 46kDa polypeptide. The protein is a member of the DEAD-box protein family (Linder and Slonimski 1989) and exhibits RNA-dependent ATPase and bidirectional RNA helicase activities. eif4a functions in the eif4f complex to unwind structured mrnas at their 5 end to facilitate both ribosome binding and scanning of the mrna for the initiation codon (Ray et al. 1985; Rozen et al. 1990; Lorsch and Herschlag 1999). Free eif4a is a weak nonprocessive helicase but it is activated when it is incorporated into the eif4f complex (Rozen et al. 1990; Korneeva et al. 2000; Kapp and Lorsch 2004b). Both eif4b, and eif4h are cofactors of eif4a (Fleming et al. 2003). However they differ in their ability to stimulate eif4a helicase activity, with eif4b being a potent stimulator (Rogers et al. 2002). The third homologue is functionally different and cannot substitute for eif4ai in a ribosome binding assay (Li et al. 1999). This homologue plays an 16

31 important role in non-sensemediated decay (NMD) of mrnas, which degrades mrnas with pre-mature stop codons to prevent synthesis of retarded proteins (Chan et al. 2004; Ferraiuolo et al. 2004; Palacios et al. 2004; Shibuya et al. 2004; Conti and Izaurralde 2005; Lejeune and Maquat 2005). 17

32 2. Translational Control during Viral Infection Viral genomes encode for some components necessary to replicate. However, synthesis of viral proteins fully depends on the host translational apparatus. Therefore, it is not surprising that mechanisms have evolved leading to inhibition of cellular protein synthesis but at the same time allow translation of viral mrnas. This section will describe various translation strategies used by eukaryotic viruses to overcome the problems associated with translational dependence Virus organization and replication Viral genomes are exceptionally heterogeneous and accordingly, infection as well as replication strategies are highly diverse. Still, viruses exhibit similar stages during their life cycle (Fig. 6). All viruses consist of a nucleic acid genome (RNA or DNA), which is protected by a minimum of a protein coat, a more complex capsid or nucleocapsid. The first step of viral infection is attachment of the virus to the host cell. This is acquired by an interaction of the virus with its cognate receptor or other components on the surface of host cells (David M Knipe 2006). After host cell-binding occurred, the viral genome and associated proteins have to be exposed to the intracellular milieu of the host so the genetic material becomes available for genome replication and transcription. There are several strategies how this is accomplished, some viruses fuse with the plasma membrane, others get internalized via the endocytotic pathway and thereby release their genome. Replication of some viruses such as the poxvirus, the picornaviruses and flaviviruses is completely cytoplasmic; while others such as herpes and orthomyxoviruses including influenza virus replicate within the nucleus of the cell. Once the assembly of the mature virions occurred they are released from the host cell. This may result in membrane lysis and death of the host cell. Alternatively, viruses establish persistent infection during which enveloped viruses (e.g., HIV) viruses are released by budding. During this process the virus acquires its envelope, which is a modified piece of the host's plasma membrane Host Cell Shutoff During the viral replication cycle of several viruses a host cell shutoff is induced. This phenomenon is defined as the process in which all cellular macromolecular synthesis is suppressed due to domination of the host metabolic system by the viral replication cycle. 18

33 Figure 6. Overview of eukaryotic virus life cycle. Viral particles are shown in black. Viruses interact with their host via their receptor and get internalized. Uncoating occurs, genome replication and translation of the viral proteins is initiated. Once all the components for a mature virus particle are produced, viral assembly occurs and the virus is released. 19

34 adapted from Gale 2000 Figure 6

35 For instance, the intracellular ion concentration (Garry et al. 1979), the nucleotide metabolism (Inglis 1982; Kwong and Frenkel 1987), RNA stability, processing and export (Fortes et al. 1994; Lu et al. 1994; Zelus et al. 1996) and recruitment or degradation of specific host factors can be affected (Katze et al. 1986). Nonetheless, these events need to be precisely balanced as the success of viral propagation depends on the host to stay alive and functional until the viral replication cycle is completed. However, not all viral infections induce a host cell shutoff and it is not always a requirement for viral replication, but in general, induction of this phenomenon benefits viral propagation. Host cell shut off severely limits translation of cellular mrnas but viral mrnas can overcome this limitation by defined mechanisms which have evolved gradually Strategies for Viral mrna Translation The cell regulates translation of their proteins as a response to intra and extra cellular changes and thereby ensure that only the proteins required for survival are produced. Viral infection critically impedes this regulation. Viral translation depends on the cellular translational system, therefore modulatory events occurring during infection cannot be very dramatic but must be subtile enough to support the viral replication cycle without killing the cell. During viral infection, translation of cellular mrna can be suppressed without destroying the RNA (M. B. Mathews 2007), but at the same time translation of viral mrnas is allowed. This is based on specialized structures and organization found in viral mrnas, which allow translation under conditions when translation of cellular mrnas is inhibited (Fig. 7). One means by which this paradox is resolved is internal translation initiation which utilizes a 5 -end independent mechanism Internal Ribosomal Entry Sites (IRES) IRES elements were first described in viral mrnas, in particular in the 5 UTR of members of the picornavirus family (Jang et al. 1988; Pelletier and Sonenberg 1988). The highly structured, uncapped mrnas of picornaviruses facilitate binding of the 40S ribosomal subunit to the mrna without scanning or the requirement of a 5 cap structure. The crucial experiment proving the existence of an IRES element was performed by including the 5 -UTR of EMCV, a cardiovirus of the picornaviridae, in front of an ORF in a closed loop RNA. As the synthesis of the encoded protein was promoted it was 20

36 Figure 7. Unconventional strategies for translation of viral mrnas. Different mechanisms are indicated on the left of the diagram and correlating examples are shown on the right. The top diagram represents capped mrnas with a 5 UTR and a poly(a) tail. The first reading frame is indicated by the initiation codon AUG and the arrow points out translation initiation. The second and third diagram represents IRES containing mrnas and mrnas containing strong secondary structures. The arrow in the third figure describing the ribosome shunt mechanism represents the ribosome bypass or shunt. (Open reading frames are denoted by rectangles) 21

37 adapted from Gale 2000 Figure 7

38 proven that initiation happened independent of a free 5 -end (Hsieh et al. 1995). As a test for IRES activity the bicistronic mrna assay was proposed (Fig. 8.A). It consists of placing a 5 -UTR which is suspected to contain an IRES between a first cap-dependent open reading frame and second downstream open reading frame in a bicistronic mrna. The second open reading frame is only translated if the inserted 5 -UTR contained an IRES element. Numerous RNA viruses contain or produce mrnas with IRESs to overcome 5 cap dependent translation. Among these are picornaviruses, hepatitis C virus, herpes virus and HIV (Bieleski and Talbot 2001; Grundhoff and Ganem 2001; Hellen and Sarnow 2001b; Griffiths and Coen 2005; Jang et al. 2006). Additionally, some cellular mrnas containing IRES elements were identified (Hellen and Sarnow 2001b; Komar and Hatzoglou 2005; Pisarev et al. 2005). Among them are BiP (Macejak and Sarnow 1991), fibroblast growth factor-2 (FGF-2) (Vagner et al. 1995), myc proto-oncogene product (Nanbru et al. 1997; Chappell et al. 2000), vascular endothelial growth factor (VEGF) (Akiri et al. 1998), and others. These mrnas need to overcome regulation by the 5 cap structure. This function is especially required when cap-dependent translation is compromised for instance during mitosis (Pyronnet et al. 2000), apoptosis (Hernandez et al. 2004) or during the recovery from heat shock (Macejak and Luftig 1991). There are at least five different prototypes or classes of IRES elements. As there is little or no nucleotide sequence homology between IRESes, it is not possible to compare IRESes upon sequence conservation. Categorization occurs by comparing similar high order structures Type I and II IRESes of Picornaviruses Type I IRES elements can be found in genomes of enteroviruses (e.g. poliovirus) or rhinoviruses; and type II IRES elements reside in genomes of apthoviruses (e.g. foot-andmouth disease virus) and in cardioviruses (e.g. EMCV) (Fig. 8.B). Both IRES types show strong conservation of the predicted RNA structures among their members but only a few similarities exist between the two IRES types. For instance, conserved structural features are GNRA tetraloops, an oligopyrimidine tract, positioned approximately 25 nucleotides upstream of the AUG (Jang 2006), and an AUG codon at the 3 end of the IRES. All of 22

39 Figure 8. A. Bicistronic mrnas. A) In bicistronic constructs, the first cistron is translated in a cap-dependent manner, producing protein 1. The second cistron will only be translated if the intercistronic sequences can function as an IRES element because ribosome recruitment to the intercistronic spacer is independent of the 5'-cap structure. The second cistron will produce protein 2. For diagram simplicity, eifs and other proteins known to participate in these processes have been omitted. B) Type I and II IRES elements. Models of the type 2 EMCV IRES and the type I poliovirus IRES are represented. The proximity and orientation of the bound eif4g/eif4a complex relative to the Yn-Xm-AUG motif is shown. 23

40 A adapted from Lopez-Lastra 2005 B Poliovirus IRES (Type I) EMCV IRES (Type II) adapted from Breyne 2009 Figure 8

41 these features are essential for internal initiation, as alterations or deletion of these areas are detrimental (Kuhn et al. 1990; Muzychenko et al. 1991; Malnou et al. 2004). Type I IRES function depends on the oligopyrimidine tract but the translational start codon is not included in the region. The start codon is positioned nucleotides further downstream. Therefore, in type I IRES elements the 40S ribosomal subunit must travel downstream on the RNA to identify the authentic start codon. The oligopyrimidine tract of the type II IRESs does include the authentic start codon and binds directly to the 40S ribosomal subunit (Jang et al. 1988; Jang et al. 1989; Kaminski et al. 1994). Further characteristics to distinguish type I from type II IRESs are the internal binding of the 40S ribosomal subunit and the cellular factors needed for this interaction. The 40S ribosomal subunit binds directly to type II IRESs (Kaminski et al. 1990; Pilipenko et al. 1994). However, a minimal amount of canonical eifs are required for recruitment and positioning of the 40S ribosomal subunit to an EMCV IRES (Type II). These factors were identified in a reconstituted in vitro system and include eif2, eif3, eif4g, eif4a, eif4b and Met-tRNA met i (Pause et al. 1994b; Meyer et al. 1995; Pestova et al. 1996a; Kolupaeva et al. 1998; Lomakin et al. 2000a; Svitkin et al. 2001b). The cap binding protein eif4e is not needed. Numerous other cellular IRES interacting proteins have been identified. Type I IRES elements are influenced by the upstream of N-ras (Unr) factor (Hunt 1999), poly[rc]- binding protein (2PCBP-2) (Blyn et al. 1996; Blyn et al. 1997) and the IRES trans-acting factor, 45 kda (ITAF45) (Pilipenko et al. 2000). Furthermore, the Lupus erythematosus autoantigen (La autoantigen) and polypyrimidine tract binding protein (PTB) have been shown to influence type I IRESs. However these reports have been controversial (Meerovitch et al. 1993; Ali and Siddiqui 1995; Kaminski et al. 1995) and it is further suggested that these proteins execute a RNA chaperon like function. Type I IRES function may require additional cellular factors stabilizing the RNA as chaperones. These factors are the La autoantigen (Kim and Jang 1999), PTB (Kaminski et al. 1995; Niepmann et al. 1997) and the murine proliferating protein (mpp1) (Pilipenko et al. 2000). 24

42 Specialized 5 -end Dependent Initiation The number of proteins encoded by a viral genome is restricted by its size. Thus novel mechanisms have evolved to increase the number of viral proteins encoded in the viral genome Ribosome Shunting Translation initiation by discontinuous scanning or ribosome shunting is cap and eif4f dependent. It presents a specialized form of 5 -end dependent initiation. The 40S ribosomal subunit is recruited to the mrna via the eif4f complex. The translational machinery now scans along the mrna and encounters a secondary RNA structure, which is unable to be unwound. This unique RNA structure facilitates movement of the 40S ribosomal subunit about 100 nucleotides downstream without scanning. This process can be inhibited by cleavage of eif4g as it is an eif4f dependent mechanism. Still, it is believed to be less dependent on canonical 5 -end dependent initiation. This specialized form of 5 -end dependent translation initiation was first described to occur on the 35S RNA of the cauliflower mosaic virus (CaMV) (Futterer et al. 1993). Furthermore, the mechanism is used on the P/C mrna of Sendai virus and mrnas of adenovirus which contain the tripartite leader sequence (Yueh and Schneider 1996; Latorre et al. 1998; Remm et al. 1999) Leaky Scanning Leaky Scanning is found in a large number of viruses. Examples are HIV-1 (Schwartz et al. 1992), Simian virus 40 (SV40) and others. Multiple proteins can be encoded within the same mrna by overlapping reading frames. The Leaky scanning mechanism allows usage of downstream AUGs for initiation, in preference or in addition to the first AUG. The most favorable context for a start codon is GCCA/GCCAUGG (Kozak 1984; Kozak 1987; Kozak 1989). Sequences surrounding each of the initiation codons influence the efficiency at which the ribosome recognizes the initiation codon and therefore at which AUG translation begins. If the first encountered AUG is surrounded by a suboptimal context, the ribosome may bypass it and initiation will occur at the next AUG. This mechanism can create protein products in the same or different reading frame. 25

43 Reinitiation Some eukaryotic mrnas apply this mechanism when they have multiple non overlapping ORFs. Recent work on feline calicivirus suggests that the capsid proteins VP1 and VP2 are translated by reinitiation. The initiation at the second AUG was thought to depend upon eif4g. This has been proven wrong as termination-reinitiation is independent of eif4g. Precisely, ORFs of VP1 and VP2 overlap. The initiation codon of VP2 is approximately 30 nucleotides upstream of the VP1 termination codon. Reinitation on the VP2 AUG requires the 3 -terminal 87 nucleotides of the VP1 upstream open reading frame. These 87 nucleotides interact with the 40S ribosomal subunit and eif3. The complex of 40S ribosomal unit and eif3 forms during disassembly of the terminating 80S ribosome and facilitates initiation on the AUG of the VP2 ORF by acquisition of a second MettRNA met i (Poyry et al. 2007) Ribosomal Frameshifting Frameshifting occurs when the ribosome slips forward or backward by one nucleotide while translating on an mrna, thereby changing the reading frame. In other words, ribosomal frameshifiting is the slipping of the translating ribosome for 1 nucleotide backwards towards the 5 end (-1 frameshift) or 1 nucleotide forwards to the 3 end (+1 frameshift). This can be initiated by signals in the RNA itself. These signals include slippery homopolymeric sequences and downstream RNA pseudoknot or stem loops. Unlike other mechanisms of initiation, ribosomal frameshifting may be regulated by elongating factors eef1a and eef2 not by initiation factors. Frameshift events were first detected in Rous sarcoma virus (Jacks and Varmus 1985) and are more common in retroviruses Suppression of termination Suppression of the termination codons result in generation of a second protein with extended C-terminus. This is used by many viruses for example by alphaviruses (Myles et al. 2006) or the Moloney murine leukaemia virus (MoMuLV) (Orlova et al. 2003). The Gag and Pol genes of MoMuLV are encoded within the same mrna. Approximately 5% of the time the reading frame is translated, the termination codon which separates the two 26

44 proteins is misread by the translating ribosome as a Gln encoding codon and the Gag-Pol protein is produced as a consequence. This fusion protein is further cleaved into the mature Pol protein. The reverse transcriptase, the integrase and the RNA-dependent RNA polymerase nsp4 of alphaviruses is produced in the same manner Modification of Translational System during Viral Infection During viral infection several regulatory factors in the cell are modulated which affect cell metabolism. These responses by the cell are generally part of the antiviral defense and designed to limit virus production. Alterations of factors regulating the translation apparatus are no exception. The most prominent targets for changes to the translational system are those required for initiation. In particular, eif2 and the eif4f complex are modified during the course of infection. These modifications can result from direct interaction of viral and cellular proteins or can be the effect of perturbations in specific signaling cascades of the cell as a result of viral infection. Viral replication is fully dependent on the translational system of the host, thus viral genomes have evolved to encode for several proteins or nucleic acids that circumvent and neutralize this response Modifications on the eif4f Complex eif4e Phosphorylation of eif4e One key protein necessary for cap-dependent translation initiation is the cap binding protein eif4e (section ). This protein is phosphorylated on serine 209 by the Mnk 1 and 2 kinases, targets of the MAPK signaling cascade (Fukunaga and Hunter 1997; Waskiewicz et al. 1997; Pyronnet et al. 1999; Waskiewicz et al. 1999). The biological function of eif4e phosphorylation is not clearly understood (Scheper and Proud 2002) but a role in tumorigenesis and viral pathogenesis was suggested. Overexpression of eif4e and therefore increased phosphorylation on S209 correlates with tumorigenesis. On the contrary, dephosphorylation of S209 may be part of the host cell shut off responsible for inhibition of mrna translation during viral infection. A decrease of eif4e phosphorylation is detected during the infection with several viruses including 27

45 adenovirus, VSV, EMCV, avian reovirus and influenza virus and others (Huang and Schneider 1991; Kleijn et al. 1996; Feigenblum et al. 1998; Connor and Lyles 2002). The molecular mechanism by which eif4e phosphorylation is decreased during infection is not well understood, with the exception of adenovirus. The adenoviral L4 100-kD proteins bind the C-terminus of eif4g thereby blocking Mnk kinase binding (Cuesta et al. 2000). Adenovirus mrna escapes the modifications on the translational apparatus as the 5 UTR of adenovirus mrnas contain secondary RNA structures able to efficiently recruit eif4f. Viral mrnas translated by a cap-independent mechanism often negatively regulate eif4f to suppress cap-dependent translation. However, some viral RNAs, which utilize a cap-dependent mechanism to translate their mrnas can actually stimulate eif4f activity. eif4e phosphorylation is necessary in order to re-activate herpes simplex virus 1 (HSV- 1) and related viruses from latency (Walsh and Mohr 2004; Walsh et al. 2005). The intermediate early gene infected cell protein 0 (ICP0) is responsible for the altered eif4e phosphorylation status and it was shown that ICP0 mutant viruses are deficient in their ability to reactivate from the latent state (Cai et al. 1993; Halford and Schaffer 2001). Later on, the downstream target gene of ICP0, ICP6 was shown to associate with eif4g and thereby promote eif4f complex assembly and eif4e phosphorylation (Walsh and Mohr 2006). Furthermore, eif4e is unable to be phosphorylated during infection of HSV- 1 lacking ICP6 even when the Mnk kinases are activated. Additionally, inhibiting Mnk 1 prevents eif4e phosphorylation and reduces Kaposi s sarcoma associated herpesvirus (KSHV) reactivation from latency by causing reduction of the viral activator RTA. In order to enter the lytic cycle accumulation of RTA is required. Replication of the vaccinia virus is also compromised in cells deficient of both Mnk kinases. In these cells eif4e is not phosphorylated (Walsh et al. 2008). Regulation of eif4e Availability Modification of 4E-BPs A family of proteins that regulate eif4e activity are the 4E-BPs (section ). The ability of the 4E-BPs to interact with eif4e is dependent on their phosphorylation status. In their hypophosphorylated state these proteins are able to compete with eif4g for the 28

46 same binding site on eif4e. Thus, interaction of 4E-PBs with eif4e prohibits eif4f complex formation and hence cap-dependent translation initiation. It is believed that dephosphorylation of the 4E-BPs contributes to virus induced host cell shut off by inhibiting cap-dependent translation (Gingras et al. 1996a). Dephosphorylation of the 4E-BPs has been observed during the infection with several viruses including EMCV and poliovirus (Gingras et al. 1996a). The translation of viral mrnas is unaffected, as these mrnas possess an IRES element (see section ). Dephophorylation of 4E-BPs is the only known mechanism of how cellular mrna translation is inhibited during EMCV infection, as the viral proteases do not cleave eif4g. This finding supports the observation that in the presence of the chemical inhibitors wortmannin and rapamycin, which inhibit 4E-BP phosphorylation, viral protein production is positively affected (Beretta et al. 1996a). Infection with vesicular stomatitis virus (VSV), Simian virus 40 (SV40), poxviruses and herpesviruses results in dephosphorylation of the 4E-BPs (Connor and Lyles 2002; Walsh and Mohr 2004; Yu et al. 2005; Walsh et al. 2008). The mechanism by which dephosphorylation of these proteins occurs varies. The viral specific protein SV40 small antigen associates with the cellular protein phosphatase PP2A promoting dephosphorylation of the 4E-BPs (Yu et al. 2005). This reduces eif4e dependent translation. Paradoxically, VSV and SV40 mrnas also contain a cap structure; this suggests that these viruses can overcome limitation by another mechanism. VSV infection results in high levels of viral mrna in the cytoplasm clearly outnumbering cellular mrnas. This enhanced abundance of viral mrna may shift the equilibrium favoring synthesis of viral proteins (Lodish and Porter 1980; Schnitzlein et al. 1983). In some cases viral infection promotes 4E-BP phosphorylation. Two proteins of adenovirus, E4-ORF1 and E4-ORF4, activate mtor but through different mechanisms. E4-ORF1 mimics growth factor signaling resulting in the activation of PI(3)-kinase and subsequently mtor, whereas E4-ORF4 interacts with the subunit of PP2A thereby preventing dephosphorylation of 4E-BP (O'Shea et al. 2005). Inhibition of components of the PI(3)K/Akt/mTOR pathway are also means by which the 4E-BPs are actively regulated during viral infection. The pul38 protein of the human 29

47 cytomegalovirus virus (hcmv) inhibits the tuberous sclerosis tumor suppressor complex (TSC), which facilitates mtor activity and hence 4E-BP phosphorylation (Kudchodkar et al. 2007). Herpesviruses as HSV-1 and 2 (Walsh and Mohr 2004) as well as Epstein Barr virus (Moody et al. 2005) activate the PI(3)K/Akt/mTOR pathway resulting in phosphorylation of the 4E-BPs and in activation of cap-dependent translation. Additionally, activation of mtor and 4E-BP phosphorylation correlates with elevated levels of E7 encoded by the human pappiloma virus (HPV) (Oh et al. 2006). However, the mechanism by which this occurs is not known eif4g Cleavage of eif4gi and eif4gii occurs during the infection of several viruses. The 2A proteases (2A pro ) of poliovirus, rhinovirus and coxackievirus and the leader (L) protease together with the 3C protease from foot and mouth disease virus (FMDV) (Lloyd 2006) are well characterized as eif4gi and II proteases. The 3C of FMDV protease has also been shown to cleave eif4a (Kirchweger et al. 1994; Belsham et al. 2000; Gradi et al. 2004). Cleavage of eif4g separates the N-terminal domain responsible for eif4e and PABP binding from the C-terminal domain which coordinates eif3, eif4a and Mnk interaction. While in vitro assays demonstrated that the C-terminal fragment promotes 5 - end dependent initiation (Ali et al. 2001), cleavage of eif4g is believed to mediate translation inhibition during enterovirus and rhinovirus infection. Thought both isoforms of eif4g are cleaved during these infections, the slower kinetics of eif4gii cleavage correlates with the inhibition of host mrna translation (Gradi et al. 1998b; Svitkin et al. 1999). In addition, several retroviral proteases including MoMLV, mouse mammary tumor virus, human T-cell leukaemia virus type 1, HIV-2 and simian immunodeficiency virus cleave eif4gi. This cleavage is not complete as the viral mrnas contain a 5 cap structure and are polyadenylated. It is unclear how this alteration of the translation machinery favors viral replication, though it is possible that eif4gii somehow influences this preference. eif4g phosphorylation A second downstream target of mtor is eif4g. Surprisingly, phosphorylation of eif4g is sensitive to rapamycin, the classical mtor inhibitor molecule (Raught et al. 2000b). 30

48 Thus, it is not surprising that eif4g becomes phosphorylated during infection with viruses as human cytomegalovirus virus (HCMV) and SV40 (Kudchodkar et al. 2004; Yu et al. 2005). The role of eif4g phosphorylation is not well understood as mrnas of both viruses contain a cap structure and are polyadenylated. Interestingly the outcome of eif4g phosphorylation differs during the infection. For instance, late during SV40 infection reduction of translation rates are observed and eif4g is phosphorylated, while during HCMV infection enhanced eif4g phosphorylation correlates with increased eif4f complex formation. During infection with influenza virus eif4g phosphorylation is detected independent of mtor activity as mtor is not altered during the course of infection (Feigenblum and Schneider 1993). How eif4g phosphorylation influences viral protein production, mrna translation initiation or which proteins mediate the phosphorylation is currently unknown Inhibition of the Closed Loop Formation Efficient mrna translation occurs when the 5 -end and the 3 -end are brought in a closed loop configuration. This is mediated by the interaction of eif4g and PABP. When the integrity of the PABP-eIF4G interaction is challenged, the closed loop formation is inhibited and the mrna becomes less stable and the efficiency of initiation is dramatically reduced. Thus modulation of PABP and inhibition of the interaction between eif4g and PABP often occurs during viral infection. The rotavirus protein NSP3 inhibits translation of host mrna by binding to eif4gi and thereby competing with cellular PABP. This inhibits the closed loop formation. Rotavirus mrnas escape this mechanism. These mrnas possess a 5 cap structure but lack a poly(a)tail. This deficiency is compensated by binding of NSP3 to specific binding sites at the 3 end of the viral RNA. Therefore the interaction of NSP3 with both eif4g and the 3 -end of the viral mrna results in a closed loop formation (Wells et al. 1998; Piron et al. 1999; Kahvejian et al. 2001; Groft and Burley 2002). A second mechanism which disrupts the closed loop formation of the mrna is PABP cleavage. This event occurs during infection with enteroviruses mediated by two viral proteases the 2A pro and 3C (Kuyumcu-Martinez et al. 2004). Initially it was thought that the 2A pro preferentially targets PABP not associated with polysomes, whereas 3C pro 31

49 preferentially cleaves polysome-associated PABP (Kuyumcu-Martinez et al. 2002; Rivera and Lloyd 2008). New data demonstrate that PABP cleavage by 3C inhibits IRES mediated translation thereby allowing for the translation-replication switch. Viral translation and RNA replication cannot occur simultaneously; therefore IRES-mediated translation needs to decrease late in infection to allow genome replication. PABP cleavage occurs during infection with a large number of viruses, including poliovirus and coxsackievirus (Joachims et al. 1999; Kuyumcu-Martinez et al. 2002), feline and human caliciviruses (Kuyumcu-Martinez et al. 2004), and hepatitis A virus (Zhang et al. 2007) and human immunodeficiency virus (Alvarez et al. 2006). 32

50 3. Host Defense Mechanisms against Viral Infection The infected organism relies on several defense mechanisms to fight off virus infections. The first line of defense is the activation of the innate immune system. This rapid response results in unspecific protection of the organism and activation of the specific immune response. In vertebrates this non specific response to pathogens includes the production and secretion of type I interferons, complement, chemokines and proinflammatory cytokines Type I Interferon System The first line of defense of an organism against infection is the production of interferon. Interferons are critical signaling molecules and were first identified by Lindenmann and Isaacs who tried to understand why an inactivated virus reduced the ability of a normal virus to infect cells (Isaacs and Lindenmann 1987). Interferon, the molecule that interfered with viral infection, is now known to protect cells against all viruses. In contrast to specialized immune cells as macrophages, which internalize and destroy viruses (Janeway and Medzhitov 2002; Beutler 2004; Medzhitov 2007), non-immune cells do not exert this mechanism. In order to protect themselves, the cells activate the type I interferon system resulting in interferon secretion (Levy et al. 2003; Akira et al. 2006). As a crucial component of the innate immune response, the type I interferon system induces an antiviral response within the cell. Immediately after viral infection of a cell, type I interferons, interferon (IFN- ) and (IFN- ), are secreted. Depending on species, thirteen to fourteen different IFN- isoforms but only one IFN- subtype are produced. The first step in the antiviral response relies on recognition of the virus or viral protein by innate immune receptors expressed by immune and non-immune cells. The subsequent IFN- and secretion then activates the type I interferon receptor, which is present on every cell (Pestka et al. 1987; Pestka 1997; Pestka et al. 2004). Upon activation of the receptor, signals from the cell surface or the cytoplasm are amplified and transferred to the nucleus facilitating the transcription of more than 100 distinct interferon stimulated genes (ISGs). These gene products establish an antiviral state in the surrounding 33

51 uninfected cells and facilitate activation of an adaptive immune response (Le Bon and Tough 2002; Iwasaki and Medzhitov 2004; Theofilopoulos et al. 2005) Pattern Recognition Receptors (PRRs) Pattern recognition receptors (PRRs) are activated by molecules associated with cellular damage (DAMPs; danger associated molecular patterns) or invading pathogens (PAMPs; pathogen associated molecular patterns). PAMPs are composed of a wide and divergent group of molecules including proteins, lipids, carbohydrates and nucleic acids. Activation of these receptors ultimately leads to the production of molecules that drive the inflammatory response. These receptors include: the Toll-like receptors (TLRs), the RIG- I (retinoic acid inducible gene-i)-like receptors (RLRs) and the nucleotideoligomerization domain (Nod)-like receptors (NLRs) (Kawai and Akira 2007b; Kawai and Akira 2007a; Ting et al. 2008; Yoneyama et al. 2008; Kumar et al. 2009a; Nakhaei et al. 2009). The activation of PRRs results in the expression of cytokines, chemokines and costimulatory molecules which establish a strong immune response against the pathogen. During viral infections virus-specific molecules, including dsrna, ssrna, dsdna and proteins are recognized by different PRRs (Fig. 9) (Thompson and Locarnini 2007; Yoneyama et al. 2008) Toll-like Receptors (TLRs) The toll proteins were first described in Drosophila (Lemaitre and Hoffmann 2007). Subsequently, toll-like receptors (TLRs) have been identified in humans, mice and other higher eukaryotes (Medzhitov and Janeway 1997). Ten human and 13 mouse TLR members have been identified (Akira et al. 2006). These proteins are type I trans-membrane glyco-proteins and consist of an extracellular domain, which is responsible for interaction with different ligands, a transmembrane domain and a cytoplasmic TIR (Toll/interleukin-1 receptor) domain. TLRs are expressed on a variety of immune and non-immune cells and their expression profiles vary among tissues and cell types (Nishiya and DeFranco 2004). Unlike their counterparts in Drosophila, vertebrate TLRs directly bind their ligands (Akira et al. 2001). TLRs recognize very diverse molecules of pathogenic origin including flagellin, 34

52 Figure 9. Cellular pathogen-recognition receptors (PRRs). PRRs are activated by pathogen-associated molecular patterns (PAMPs). During viral infections viral proteins and nucleic acids are recognized by the PRRs including RIG-I (retinoic-acid-inducible gene I), MDA5 (melanoma differentiation-associated gene 5) and certain Toll-like receptors (TLRs). PRR PAMP interactions trigger signaling cascades that result in the activation of transcription factors, including interferon (IFN)-regulatory factor 3 (IRF3) and nuclear factor- B (NF- B), which induce the production of type I IFNs. The IFNs are secreted and upon binding activated the interferon receptor which results in expression of interferon-stimulated genes (ISGs) and pro-inflammatory cytokines and chemokines. 35

53 adapted from Katze 2008 Figure 9

54 porin, LPS and nucleic acid. TLRs 1-9 can be generally divided into two groups based upon their PAMP specificity. TLR1, 2, 4, 5 and 6 are expressed on the cell surface and recognize PAMPs of the cell wall and flagellin of bacterial, yeast and fungi. The remaining TLRs 3, 7/8 and 9 recognize both, viral and bacterial derived nucleic acids and are found intracellular with the endolysosomal compartment. Upon receptor activation the TIR domain induces TIR-domain-containing adaptor molecules resulting in a specific sequence of events. Activated TLRs induce several downstream pathways including the myeloid differentiation primary response gene 88 (MyD88)-dependent pathway and MyD88-independent pathways (Fig. 10). The MyD88 dependent pathway facilitates the association of TIR with IL-1 receptorassociated kinases (IRAKs) (Muzio et al. 1997). Furthermore, other proteins including TNF-receptor-associated factor 6 (TRAF6), transforming growth factor- -activated kinase (TAK1) and TAK-binding proteins (TABs) are involved in this signaling cascade (Akira and Takeda 2004a). Activation of TAK1 also stimulates the mitogen-activated protein kinase (MAPK) pathways, which induce the transcription factors AP-1 (heterodimer of ATF2 and c-jun) (Ninomiya-Tsuji et al. 1999; Platanias 2003; Schieven 2005). An important downstream event is phosphorylation and subsequent degradation of I B, the inhibitor of nuclear factor kappa B (NF B). I B is phosphorylated at its N- terminus. In the case of I B, this occurs at Ser-32 and Ser-36; the corresponding residues in I B are Ser-19 and Ser-23. Once I B is phosphorylated the protein is ubiquitinilated and degraded by the proteasome. This liberates NF- B (Karin and Ben- Neriah 2000) and the protein translocates into the nucleus to bind DNA (Chen et al. 1995), inducing expression of genes involved in the inflammatory response, such as those that encode TNF- and IL-6 (Akira and Hemmi 2003). The MyD88 independent pathway is so far only reported for TLR4 and TLR3 (Akira and Takeda 2004a) and includes signaling through the TIR-containing adaptor-inducing IFN- (TRIF), which is also referred to as TIR-domain containing adaptor molecule 1 (TICAM1) and the TRIF-related adaptor molecule (TRAM), also known as TICAM2. It activates interferon regulatory factor 3 (IRF3) and late-phase NF B (Akira 2003). Activation of these two factors facilitates production of IFN- and expression of IFNinducible genes. 36

55 Figure 10. TLR mediated signaling in conventional dendritic cells. Activation of TLRs by their respective ligands initiates signaling dependent on the two adaptor molecules MyD88 and TRIF via the TIR domains. MyD88 recruits the IRAK family of proteins and TRAF6, which activates TAK1. TAK1 then stimulates the IKK complex consisting of IKK, IKK and NEMO/IKK and phosphorylates I Bs. Upon phosphorylation I B is ubiquitinated and degraded. NF- B, consisting of p50 and p65, is released and translocates into the nucleus. TAK1 also activates the MAPK signaling pathway, which in conjunction with NF- B initiates the transcription of inflammatory cytokine. TRIF recruits RIP1 and TRAF6. These factors also activate NF- B and MAPK. TRIF interacts with TRAF3 and activates TBK1/IKKi, which phosphorylate IRF3 and IRF7. Activated IRF3 and IRF7 migrate to the nucleus to induce transcription of type I IFNs. 37

56 adapted from Kumar 2009 Figure 10

57 Both pathways, MyD88 dependent and independent, interconnect at several points but these events are not fully understood TLR2 and TLR4 TLR2 and 4 can be found on the outer cell membrane. TLR2 recognizes a very diverse range of proteins mostly of bacterial origin. Furthermore, TLR2 can recognize complete pathogens, including bacteria and viruses as herpes simplex virus and varicella-zoster virus (Kurt-Jones et al. 2004). Additionally, this receptor responds to components of measles, HCV and MCMV (Bieback et al. 2002). TLR4 has also a role in sensing viruses. For instance, detection of RSV, retrovirus and coxsackie B virus is mediated by TRL4 (Kurt-Jones et al. 2000; Rassa et al. 2002; Richer et al. 2006) TLR3 This receptor is expressed in endosomes of specialized immune cells including conventional dendritic cells (cdcs), which are antigen-presenting cells, macrophages, B- cells, natural killer cells, but also in non-immune cells, like epithelial cells. In fibroblasts TLR3 is expressed on the cell surface (Matsumoto et al. 2003). Surprisingly, plasmacytoid dendritic cells (pdcs), which are important mediators of the immune system as they produce high levels of type I interferon, do not express the receptor. TLR3 is activated by binding to dsrnas originating from dsrna viruses like reovirus or by dsrna produced during replication of ssrna viruses like West Nile virus, RSV and EMCV (Wang et al. 2004; Groskreutz et al. 2006; Hardarson et al. 2007). Conveniently, activation of the receptor can be initiated by the synthetic dsrna known as polyinosinic:polycytidylic acid (poly(i:c)), which allows to study TLR3 activation without side effects induced by viral infection. The role of TLR3 in viral infection is controversial. When TLR3 deficient mice were challenged with mouse cytomegalo virus (MCMV), VSV, lymphocytic chorimenigintis virus (LCMV), RSV or reovirus no difference in infection was detected when compared to WT control mice (Edelmann et al. 2004). However, these mice showed resistance to WNV infection, which shows TLR3 importance during infection with this virus. 38

58 Recently, it has been shown that TLR3 mediates the establishment of an antiviral state against Hepatitis C Virus in hepatoma cells (Wang et al. 2009). TLR3 senses hepatitis C virus (HCV) infection when expressed in permissive hepatoma cells and induces IRF-3 and synthesis of ISGs that restrict virus replication TLR7, 8 and 9 These three receptors are located in the endosomal compartment of cells and their expression can be induced by type I interferon. They recognize microbial RNA and DNA. TLR7 and TLR8 (not mouse TLR8) respond to nucleic acid homologues such as synthetic antiviral imidazoquinoline compounds, loxoribine and imiquimod and ssrna rich in guanosid or uridine derived from viruses (Heil et al. 2004; Hemmi and Akira 2005; Diebold 2008). TLR7 in pdcs recognizes replicating VSV and ssrna in the endosome. It is suggested that TLR7 is activated upon autophagosome formation due to VSV infection. The autophagosome, containing cytoplasmic molecules, fuses with the endosome and so TLR7 is exposed to viral RNA (Lee and Kim 2007). TLR9 is activated by unmetylated CpG motifs of ssdna, which are commonly present in genomes of bacterial and viruses (Hemmi et al. 2003; Lund et al. 2003; Hochrein et al. 2004; Krug et al. 2004b; Tabeta et al. 2004). Specifically, the response to DNA viruses such as MCMV, herpes simplex virus 1 (HSV-1) and HSV-2 entails TLR9 mediated IFN- production (Lund et al. 2003; Krug et al. 2004a; Krug et al. 2004b). To further understand TLR9 activation in the absence of viral infection synthetic ssdna containing CpG dinucleotides can be used Cytoplasmic Pattern Recognition Receptors TLR independent recognition of cytosolic microbial components and danger signals, such as ATP and toxins is mediated by several sensors in the cytoplasm (Fig. 11). Among these sensors is the nucleotide-binding domain leucine-rich repeat (LRR) containing NOD-like receptor (NLR) family. Currently this family contains 23 members of which only a few are studied (Franchi et al. 2009). NLRs are known to detect cytosolic microbial matter and danger signals. These receptors contain a characteristic C-terminal LRR and internal nucleotide-binding domain (NBD), leading to the activation of the inflammasome, a large multiprotein complex whose assembly activates caspase-1, 39

59 Figure 11. Cytoplasmic pattern recognition receptors. The presence of microbial or self cytoplasmic RNA and DNA is detected by specific sensors in the cytoplasm. Ligand binding induces helicases, such as RIG-I and MDA5, which signal to the mitochondrial adaptor IPS-1. IPS-1 activates ubiquitin ligases (TRAFs) and kinases (MEKK1, IKK, MAPKs, TBK1 and IKK ) which transfer the signal to transcription factors (NF- B, AP1 and IRF3), leading to the production of type I IFN. Similarly, cytosolic DNA activates DAI (and potentially other sensors) which then promotes the production of type I IFNs. STING, another ER-associated protein, also binds RIG-I and promotes type I IFN upon induction by both RNA and DNA. 40

60 adapted from Baccala 2009 Figure 11

61 promoting the maturation of pro-inflammatory cytokines interleukin-1 (IL-1) and IL-18 (Martinon 2007). In recent studies a function in the formation of an antiviral response was assigned to certain NLR family members. For instance, NLRP3 mediates stimulation of the inflammasome upon activation by adenovirus DNA (Muruve et al. 2008). A role for NLRP3 was also demonstrated for caspase 1 activation in cultured macrophages during viral infection (Kanneganti et al. 2006). Furthermore, a critical role for inflammasome signaling in the adaptive immune response against influenza A virus infection has been implicated (Allen et al. 2009; Ichinohe et al. 2009; Owen and Gale 2009). Recently, another cytoplasmic sensor was discovered: absent in melanoma 2 (AIM2). This protein is a member of the pyrin and HIN domain-containing protein (PYHIN) family and is believed to be the key sensor for cytoplasmic dsdna (Burckstummer et al. 2009; Fernandes-Alnemri et al. 2009; Hornung et al. 2009; Roberts et al. 2009). Cytoplasmic DNA triggers formation of the AIM2 inflammasome by inducing AIM2 oligomerization, leading to activation of the (apoptosis-associated speck-like protein containing a CARD) ASC pyroptosome and caspase-1 (Fernandes-Alnemri et al. 2009). Although both TLR7 and TLR9 are critical for recognition of viral nucleic acids in the endosomes of plasmacytoid dendritic cells (pdcs), most other cell types recognize viral RNA intermediates through the RLR arm of the innate immune response (Andrejeva et al. 2004; Yoneyama et al. 2004; Yoneyama et al. 2005). Early viral replicative intermediates are detected by two recently characterized cystolic viral RNA receptors of the RIG-I-like receptor family (RLRs); the retinoic acid-inducible gene-i (RIG-I) and melanoma differentiation-associated gene-5 (MDA-5). These two molecules are pivotal for the formation of an antiviral immunity in most cell types (Yoneyama et al. 2005). RIG-I and MDA5 are DExD/H box RNA helicases with two N-terminal CARD domains, a central ATPase and a helicase domain and a C-terminal regulatory domain (Yoneyama et al. 2004). Upon binding to viral RNA a conformational change is induced exposing the CARD domain. This results in dimerization and interaction with the CARD domain of the outer mitochondrial membrane protein MAVIs/IPS1/Cardif/VISA (mitochondrial antiviral signaling/ IFN promoter stimulator 1/CARD adaptor inducing IFN- / virusinduced signaling adaptor). This activation of the C-terminal domain of IPS1 initiates a signaling cascade inducing the transcription of several cytokines. Specifically in complex 41

62 with TNF receptor-associated factor (TRAF) 3 and NAK-associated protein 1 (NAP1), IPS1 facilitates activation of members of the IkB kinase (IKK) family, namely TANK binding kinase 1 (TBK-1) and IKK3 (Kawai et al. 2005; Seth et al. 2005; Kumar et al. 2006; Meylan and Tschopp 2006; Oganesyan et al. 2006; Sasai et al. 2006). The latter phosphorylates the transcription factors IRF3 and IRF7 (Fitzgerald et al. 2003; Sharma et al. 2003), which subsequently induces transcription of type I interferon (Honda et al. 2005a; Honda et al. 2005b). IPS-1 also associates with the adaptor Fas associated death domain (FADD), the kinases receptor interacting protein 1 (RIP1), transforming growth factor -activated kinase 1 (TAK1), IKK and IKK to mediate the activation of NF- B and the MAPK pathway (Balachandran et al. 2004; Yoneyama et al. 2004; Kawai et al. 2005; Seth et al. 2005; Xu et al. 2005; Yoneyama et al. 2005; Meylan and Tschopp 2006; Takahashi 2006). This ultimately results in the transcription of IFN- mrna. IPS-1 is regulated by two proteins, NLRX and the Polo-like kinase 1 (PLK1) (Moore et al. 2008; Vitour et al. 2009). NLRX is a member of the NLR family and is located at the outer membrane of mitochondria. NLRX was identified as an inhibitor of mitochondrial antiviral signaling via interaction with its highly conserved nucleotide-binding domain (NBD) and the CARD domain of IPS-1 (Moore et al. 2008). In contrast, NLRX1 has also been reported to be a positive regulator of the innate immune system by stimulating NF- B and c-jun n-terminal kinase (JNK) activity (Tattoli et al. 2008). The second inhibitory protein of IPS-1 is PLK1, which was identified in a yeast two-hybrid search for IPS-1 interacting proteins. The protein interacts with IPS-1 at two different domains and thereby strongly inhibits the ability of MAVS to activate the IRF3 and NF- B pathways (Vitour et al. 2009). RIG-I and MDA5 have emerged as the primary sensors for RNA viruses. These sensors detect diverse families of viruses such as paramyxoviruses, orthomyxoviruses, rhabdoviruses, flaviviruses, and picornaviruses and thereby induce the production of type I interferon (Yoneyama et al. 2004; Kato et al. 2005; Gitlin et al. 2006; Kato et al. 2006). These two molecules are not functionally redundant as each of them detects different families of viruses. RIG-I senses hepatitis C virus, Sendai virus, influenza virus, vesicular stomatitis virus (VSV), rabies virus, Japanese encephalitis virus and Newcastle disease virus (Yoneyama et al. 2004; Kato et al. 2005; Kato et al. 2006). MDA5 recognizes 42

63 picornaviruses and is stimulated by poly(i:c). However viruses such as Dengue and West Nile virus, both members of the Flaviviridae, induce type I interferon in mouse embryonic fibroblasts (MEFs) lacking either RIG-I or MDA5. This suggests that a combination of both sensors may detect infection of these viruses (Fredericksen et al. 2008; Loo et al. 2008). RIG-I distinguishes viral from cellular RNAs by recognizing a 5 triphosphate modification found on viral mrna, which cannot be found on capped or processed cellular mrnas (Hornung et al. 2006; Pichlmair et al. 2006; Plumet et al. 2007). RIG-I is also able to detect short dsrnas (Takahasi et al. 2008). Still, although RIG-I is able to bind to poly(i:c) and polyriboadneylic:polyribouridylic acid (poly A:U), it is not activated by these molecules (Rothenfusser et al. 2005; Yoneyama et al. 2005). Most importantly, RIG-I and MDA5 are not activated by cellular RNAs, mainly because trnas and rrnas are extensively modified with unusual bases which prevent activation of both molecules (Hornung et al. 2006). Furthermore, in vitro studies have revealed that short dsrnas of about 25 bp with a single phosphate 3 or 5 are able to trigger RIG-I (Takahasi et al. 2008). The activation of RIG-I needs to be tightly controlled to prevent an overactivation of the antiviral cascade, which can harm the host. One means by which RIG-I is regulated is by ubiquitination. Ubiquitination is an important mechanism regulating the innate immune response. Tripartite motive 25 (TRIM25 ), a member of the tripartite motif (TRIM) protein family, is pivotal for RIG-I regulation. This protein is mediating robust ubiquitination on the CARD domains of RIG- I on two different sites (Lys 172 and Lys 63) through its ubiquitin E3 ligase activity. Ubiquitination of Lys 172 on RIG-I is critical for efficient IPS-1 binding and hence, the ability of RIG-I to induce antiviral signal transduction (Gack et al. 2007). Recently, another ubiquitin ligase Riplet/RNF135 was reported to ubiquitinate RIG-I on Lys 63 resulting in enhanced IFN- production early during infection (Oshiumi et al. 2009). On the contrary, polyubiquitination on Lys 48 by the E3 ubiquitin ligase RNF125 can result in RIG-I proteasomal degradation (Arimoto et al. 2007). RIG-I is also negatively regulated by the deubiquitination enzyme CYLD. The enzyme associates with both RIG-I and IPS-1 and inhibits RIG-I signaling by disassociating the ubiquitin moieties of RIG-I of Lys 63 (Friedman et al. 2008; Zhang et al. 2008). 43

64 Interestingly, an additional molecule regulating downstream signaling of RIG-I is the autophagy regulator (Atg5 Atg12 conjugate) (Jounai et al. 2007). It negativly modulates the production of type I IFN by direct association with RIG-I and MAVS through CARD CARD Interactions. This suggests that the autophagy regulator appears to promote RNA virus replication by inhibiting innate antiviral immune responses. In addition to ubiquitination, ISGylation of RIG-I is also an important for limiting RIG-I signaling (Zhao et al. 2005). Conjugation of ubiquitin-like ISG15 to RIG-I reduced virusinduced IFN- promoter activity. To reveal the role of RIG-I in vivo, RIG-I deficient mice were generated. Interestingly a RIG-I-deficient mouse generated by Kato et al. was embryonic lethal due to liver degeneration (Kato et al. 2005). However, when Wang et al. used a different gene targeting strategy, RIG-I knockout mice were viable but developed a colitis-like phenotype associated with the downregulation of the G protein subunit Gi2 and subsequent abnormal T cell activation (Wang et al. 2007). RIG-I also stimulates the NF- B pathway, and recent work has shown that RIG-I signaling shares significant similarity with the signaling pathway down-stream of the TNF receptor (Mikkelsen et al. 2009). For instance, both receptors rely on TRADD, RIP1, and FADD to activate NF- B. However, many of the details in the regulation of signaling by TNF- and RIG-I to NF- B need to be clarified. The second sensor for detecting RNA viruses in the cytoplasm is MDA5. MDA5 is essential for the recognition of picornaviruses. The 5 end of the picornavirus genome is bound to the virus encoded protein Vpg, thus there is no free 5 triphosphate which can be sensed by RIG-I. MDA5 recognizes the RNA secondary structures present in the 5 end of the picornavirus mrna and is thereby activated (Gitlin et al. 2006; Kato et al. 2006). MDA5 is also required for Theiler s virus and Mengo virus recognition (Kato 2006). Furthermore, MDA5 is predominantly activated upon treatment with the synthetic dsrna p(i:c) (Pichlmair et al. 2006). Still, MDA5 signaling is less well understood than that of RIG-I. It was also reported that the RNF125 conjugates ubiquitin to MDA-5 and IPS-1, thus inhibiting the assembly of the downstream antiviral signaling complex (Arimoto et al. 2007). Another regulation mechanism of MDA5 is mediated by the deubiquitinating enzyme A (DUBA). It has been shown to negatively regulate IFN signaling following 44

65 RIG-I, MDA-5 or TLR3 stimulation (Kayagaki et al. 2007). Stimulation of these molecules results in the activation of TRAF3. DUBA specifically removes ubiquitin from Lys 63 of TRAF3, resulting in the disruption of the interaction between TRAF3 and the downstream kinases IKK and TBK1. This subsequently blocks IRF-3 and IRF-7 phosphorylation (Kayagaki et al. 2007). MDA-5 is also subject to specific regulation by the negative regulator dihydroxyacetone kinase (DAK) (Diao et al. 2007). DAK specifically interacts with MDA-5, not RIG-I, but the detailed mechanism on how this inhibition is mediated needs further clarification. The third member of the RRLs family is LGP2. Unlike MDA5 and RIG-I, LGP2 is a negative regulator of virus induced activation of type I interferon response (Rothenfusser et al. 2005; Yoneyama et al. 2005; Komuro and Horvath 2006; Saito and Gale 2007). LGP2 is also a sensor for nucleic acids, similar to MDA5 and RIG-I. The protein contains a DExH box ATPase domain and a C-terminal repressor domain but it lacks the two N- terminal CARD domains. Through its C-terminal repressor domain LGP2 disrupts homotypic CARD/helicase domain and/or C-terminus interactions and thereby inhibits RIG-I self association (Saito and Gale 2007). Moreover, LGP2 inhibits the RIG-I signaling at multiple steps, indirectly through sequestration of RNA substrates (Rothenfusser et al. 2005; Yoneyama et al. 2005) or by disrupting the IPS1 signaling complex (Komuro and Horvath 2006). Consistent with these observations mice lacking LGP2 showed enhanced production of IFN- and IFN- in response to poly(i:c) or VSV infection, but surprisingly EMCV infection impaired IFN- production (Venkataraman et al. 2007). This result suggests a positive role for LGP2 in antiviral signaling in response to picornaviruses augmenting MDA5 mediated signaling (Pippig et al. 2009). Lastly, dsdna is detected by the cytosolic DAI (Takaoka et al. 2007). Similar to MDA5 and RIG-I, DAI signals towards activation of IRF Type I Interferon Feedback Loop The initial activation of PPRs and TLRs by viruses results in the expression of IFN- followed by IFN-. The different kinetics of IFN- and expression underlies a complex regulation system which is crucially influenced by a positive-feedback loop. Transcription of the IFN- gene requires activation of an otherwise silent promoter by 45

66 nuclear localization of transcription factors, such as NF- B, ATF/JUN and IRF3 (Doly et al. 1998). The IFN- enhanceosome complex, consisting of IRF-3, IRF1, NF- B, CBP/p300, the architectural protein HMG I(Y) and ATF-2/c-Jun, is assembled on the IFN- promoter on positive regulatory domains (PRD) I-IV. In synergy these transcription factors trigger sustained transcription of the IFN- gene (Kim and Maniatis 1997; Merika et al. 1998). The transcription factor IRF3 is constitutively expressed in many tissues and cell types and neither infection with viruses nor interferon treatment enhances its transcription. When the protein is phosphorylated by members of the IkB kinase (IKK) family upon upstream stimulation, IRF3 migrates into the nucleus and there binds to the IFNstimulated response element of interferon stimulated genes (ISG) (Au et al. 1995), the promoter of the IFN- 4 gene and the positive regulatory domain (PRD) III of the IFN- gene. However, before the transcription factor NF- B can migrate into the nucleus it must be released from its inhibitor I B. This requires that I B be ubiquitinated and phosphorylated (Kopp and Ghosh 1995; Verma et al. 1995; Baldwin 2004). Besides transcription of the IFN- 4 gene, the other IFN- genes require IRF7 to be synthesized (Yeow et al. 2000; Levy 2002; Levy et al. 2002). Unlike IRF3, IRF7 is not constitutively expressed. Only small amounts of IRF7 are present in the cell (except in pdcs). Therefore it needs to be transcriptionally activated through the IFN receptor/janus kinase signal transducer and activator of transcription (JAK STAT) pathway, as do most IFN-induced genes (Sharma et al. 2003). IFN- and IFN- 4 provide the initial signal that allows IRF7 to be produced. Once these interferons are secreted they activate their interferon / receptor, which is expressed on every cell. This results in activation of receptors expressed on the infected cell as well as the receptors on the surrounding cells. Upon ligand binding the type I interferon receptor dimerizes and activates two members of the Janus-family tyrosine kinases (JAK) JAK1 and TYK1 (Darnell et al. 1994; Ihle 1995). This results in the phosphorylation of the transcription factors signal-traducing activators of transcript (STAT) which facilitates their dimerization. STAT molecules are found latent within the cytoplasm but upon phosphorylation their homo- or heterodimerization is triggered. Especially formation of the ISGF3 complex is important for the type I interferon response. This factor is composed of activated STAT1 and 46

67 STAT2, which interacts with IRF9. The activated ISGF3 complex migrates to the nucleus and binds to IFN-stimulated response element (ISREs) in promoters of more than 300 interferon-stimulated genes (ISGs) (Horvath et al. 1996; Li et al. 1998; Aaronson and Horvath 2002). Many of the gene products encode PRRs, including RIG-I and MDA5 that detect and modulate signaling pathways, or transcription factors, as IRF-7 (Lee and Kim 2007). This results in the amplification of the positive feed back loop increasing IFN production. Additionally, inhibitors of this response such as the SOCS proteins are induced to partly dampen this potent antiviral response (Alexander and Hilton 2004). However, only few of these ISGs have been directly implicated in instigating the antiviral state. Among these are the well studied antiviral effectors ISG15 (IFN-stimulated protein of 15 kda), ISG56, the GTPase Mx1 (myxovirus resistance 1), 2,5 -oligoadenlyate synthetase (2-5 OAS), ribonuclease L (RNaseL) and PKR Interplay of Type I Interferon System and Translational Control Mechanisms Key pathways which regulate cellular translation also modulate type I interferon production. These pathways include the MAPK/p38, the ERK-1 and the PI(3)K/AKT/mTOR pathway. Several studies demonstrated a role of these pathways in regulation of transcription and translation of RNAs encoding components of the type I interferon response P38/MAPK pathway and interferon signaling The mitogen activated protein (MAP) kinases are a family of four serine threonine kinases (,,, ) which regulate the cell s response to growth factors, proinflammatory cytokines (Lee et al. 1994; Raingeaud et al. 1995) and external stress signals such as lipopolysaccharide (LPS) (Freshney et al. 1994; Han et al. 1994), hyperosmolarity and heat shock (Rouse et al. 1994). Most importantly, the p38 kinase is stimulated upon IFN- treatment (Goh et al. 1999; Uddin et al. 1999) and inhibition of p38 activity blocks transcription of ISRE regulated genes upon IFN- stimulation, which is independent of STAT1 or STAT2, or the formation of the ISGF3 complex. Induction of the p38/mapk in response to activation of the interferon receptor is dependent on the small GTPases RAC1 (Uddin et al. 2000), a member of the Rho GTPase 47

68 family, the proto-oncogene VAV and possibly, Cdc42hs (Fig. 4) (Coso et al. 1995; Minden et al. 1995). Conjugation by the IFNAR through IFN- promotes the activation of the guanine-nucleotide-exchange activity of VAV by its associated kinase Tyk2 (Platanias and Sweet 1994; Uddin et al. 1997). Activated VAV then catalyses the exchange of GDP to GTP on RAC1 stimulating the GTPase (Crespo et al. 1997). RAC1 induces p21-activated kinase 1 (PAK1) resulting in the induction of MAPK kinases MKK3, 4 and 6 which phosphorylate p38 (Enslen et al. 1998; Goh et al. 1999). p38 activates several downstream targets, for instance the MAPK-activated protein kinase 2 and 3 (MAPKAPK2 and 3) (Uddin et al. 1999). Whereas the role of MAPKAPK3 has to be examined in interferon mediated signaling, MAPKAPK2 has been shown to be required to resolve infection with EMCV, an interferon sensitive virus, in mice (Pi 2004, Platanias 2003). It was shown that MAPKAPK2 is mediating the transcription of both IRF-9 and ISG15, thereby activating the type I interferon response (Li et al. 2005). Additional kinases activated by p38 in response to type I interferon include the mitogenand stress-activated kinases 1 and 2 (MSK1 and 2) (Deak et al. 1998; Soloaga et al. 2003) and the MAPK-interacting protein kinases 1 and 2 (Mnk 1 and 2) (Platanias 2003). Activity of both MSK kinases is required to phosphorylate histone H3, which is important for immediate early gene expression and chromatin remodeling of interferon stimulated genes. Mnk 1 and 2 are also phosphorylated in the presence of type I interferon. These kinases phosphorylated eif4e and eif4g, which modulates translation initiation (Waskiewicz et al. 1997). p38/mapk is also induced upon activation of several pattern recognition proteins. For instance, adaptor proteins downstream of the TLRs activate P38/MAPK. This, in conjunction with NF- B activation, induces the expression of proinflammatory cytokines (Jefferies et al. 2003). These adaptor proteins are members of the IL-1 receptor-associated kinase (IRAK) family and tumor necrosis factor (TNF) receptor associated factor (TRAF) 6 (Jefferies et al. 2003), with the exception of TRIF, which initiates signaling to NF- B via direct interaction with TRAF6 (Jiang et al. 2003; Sato et al. 2003). The cytoplasmic PRR RIG-I has also been shown to activate p38/mapk in a TRAF2- TAK1-dependent manner (Mikkelsen et al. 2009). 48

69 Furthermore, the NOD proteins initiate signaling to NF- B and MAP kinases, whereas IRF-3 is not activated (Inohara and Nunez 2003). Lastly, PKR is also capable of activating signal transduction to NF- B and MAP kinases (Chu et al. 1999; Gil and Esteban 2000; Zamanian-Daryoush et al. 2000) which induces cytokine expression (Deb et al. 2001). Upon activation and dimerization, PKR interacts with TRAF family proteins, which in turn trigger activation of IKK and NF- B (Gil et al. 2004). PKR also associates with MAP kinase kinase (MKK) 6, and activates this kinase through phosphorylation (Silva et al. 2004), hence stimulating the p38 MAP kinase pathway PI3K Pathway and its Role in Interferon Mediated Response The PI3K/AKT/mTOR pathway is a critical regulator of cell metabolism, growth and survival. It is also implicated in the control of cellular translation initiation and recently has been investigated for its role in type I interferon response (Platanias 2005). A link between the type I interferon response and the PI3K/AKT/mTOR pathway was first suggested when it was shown that IRS1 was phosphorylated upon treatment with interferon. This activation resulted in the association of the regulatory subunit p38 of PI3K with IRS1 (Uddin et al. 1995). Also IRS2 was shown to provide docking sites for PI3K upon interferon treatment (Platanias et al. 1996; Burfoot et al. 1997). Furthermore, PI3K can be directly activated during TLR/IL-1R signaling (Akira and Takeda 2004b). It was shown that LPS stimulation results in the tyrosine phosphorylation of MyD88 and the formation of a PI3K MyD88 complex. MyD88 also interacts directly with AKT, and a dominant-negative mutant of AKT causes a defect in MyD88-dependent NF- B transcriptional activity. Activated PI3K mediates signaling to its downstream target mtor, which then transfers the signal to its targets the 4E-BPs and S6K 1 and 2 (Brown et al. 1995; Brunn et al. 1997; Hara et al. 1997; Burnett et al. 1998). mtor mediated phosphorylation of the 4E- BPs prevents interaction with eif4e and thereby activates translation. Phosphorylation of S6K1/2 stimulates phosphorylation of the small ribosomal protein S6 which in turn results in enhanced translation (Gingras et al. 2001b; Bjornsti and Houghton 2004). 49

70 Incubation of cells in the presence of type I interferon enhanced phosphorylation of both target proteins. This activation was abrogated when a pharmacological inhibitor of PI3K or the inhibitor of mtor, rapamycin, was used (Platanias 2005). S6K1/2 and the 4E-BPs were believed to be regulators of general translation, but in fact several lines of evidence suggest that only a subset of distinct mrnas is regulated. This distinct regulation was also shown to be important for proteins mediating the antiviral and antiproliferation response of the cell. In fact, when mouse embryonic fibroblasts lacking the 4E-BP proteins 1 and 2 were examined for a type I interferon mediated antiviral response it was revealed that translation of the IRF7 mrna is regulated by the 4E-BPs (Colina et al. 2008). Precisely, IRF7 mrna was shown to carry extensive secondary structures in its 5 UTR, therefore the need for active eif4f complex is increased. Removal of 4E-BPs enhanced IRF7 translation and hence type I interferon production (Colina et al. 2008). MEFs deficient of the S6K proteins show an impaired IFN response. These cells are unable to synthesize type I interferon due to a defect in the signaling cascade induced by the TLRs (Cao et al. 2008). 50

71 Rational Remerging viral infectious diseases are on the rise. Therefore, it is increasingly important to understand the relationship between host and virus on the molecular level. This study was focused on exploring the role of the translation initiation factor eif4e and its regulation during viral infection. Previously it has been shown that picornaviruses carry an IRES element in their RNA (Pelletier and Sonenberg 1988) and lack a cap structure. Therefore, their mrna can recruit the ribosome in an eif4f complex independent manner. This suggests that these viruses have no requirement for eif4e. Still, it was shown that drugs such as wortmannin and rapamycin, which inhibit 4E-BP phosphorylation enhance EMCV replication (Beretta et al. 1996a). Interestingly, 4E-BP is dephosphorylated during host cell shut off mediated by EMCV (Gingras et al. 1999b). Prompted by these results the question arose if eif4e availability is important for the regulation of picornavirus mrna translation. Activation of the MAPK pathway has been shown to occur regularly during viral infection, thereby initiating the formation of an innate immune response. Additionally, MAPK signals to the translational machinery, and as a result eif4e is phosphorylated (Fukunaga and Hunter 1997; Waskiewicz et al. 1997; Pyronnet et al. 1999). It was shown for many viruses, including the picornaviruses, vesicular stomatitis virus and others that eif4e is dephosphorylated during infection (Feigenblum and Schneider 1993; Kleijn et al. 1996; Connor and Lyles 2002). This suggests that this mechanism is important for either viral replication or for the induction of an antiviral response mechanism by the cell. Therefore, the role of eif4e phosphorylation during viral infection was studied. 51

72 Chapter 2.: Eukaryotic Translation Initiation Factor 4E Availability Controls the Switch between Cap-Dependent and Internal Ribosomal Entry Site-Mediated Translation Yuri V. Svitkin, Barbara Herdy, Mauro Costa-Mattioli, Anne-Claude Gingras, Brian Raught, and Nahum Sonenberg 52

73 Abstract Translation of m 7 G-capped cellular mrnas is initiated by recruitment of ribosomes to the 5 end of mrnas via eif4f, a heterotrimeric complex comprised of a cap-binding subunit (eif4e) and an RNA helicase (eif4a) bridged by a scaffolding molecule (eif4g). Internal translation initiation bypasses the requirement for the cap and eif4e, and occurs on viral and cellular mrnas containing IRESs (Internal Ribosomal Entry Sites). Here we demonstrate that eif4e availability plays a critical role in the switch from cap-dependent to IRES-mediated translation in picornavirus-infected cells. When both capped and IRES-containing mrnas are present (as in intact cells or in vitro translation extracts), a decrease in the amount of eif4e associated with the eif4f complex elicits a striking increase in IRES-mediated viral mrna translation. This effect is not observed in translation extracts depleted of capped mrnas, indicating that capped mrnas compete with IRES-containing mrnas for translation. These data explain numerous reported observations where viral mrnas are preferentially translated during infection. Introduction Recruitment of ribosomes to mrna is the rate-limiting step in translation initiation and a frequent target for translational control (Mathews 2000). Two different mechanisms of ribosome binding exist in mammalian cells. Cap-dependent translation is mediated by the mrna 5 cap structure (m 7 GpppN, where N is any nucleotide), and represents the standard mode of translation used by most cellular mrnas. Cap-independent translation is utilized by some plus-stranded RNA viruses, including picornaviruses and hepatitis C virus, as well as by some cellular mrnas, and involves the binding of ribosomes to an mrna structural element termed an internal ribosome entry site, or IRES (Hentze 1997; Pestova et al. 2001). The eukaryotic translation initiation factor (eif) 4F mediates 40S ribosomal subunit binding to the 5 end of capped mrna. eif4f is a complex containing three proteins: eif4e, the cap-binding subunit; eif4a, an RNA-dependent ATPase/ATP-dependent RNA-helicase; and eif4g, a high-molecular-weight protein that acts as a scaffold for binding eif4e and eif4a. In addition, eif4g interacts with the 40S ribosome binding 53

74 factor eif3 and the poly(a)-binding protein (PABP), thereby establishing a critical link between mrna and the ribosome (Hentze 1997; Gingras et al. 1999b; Mathews 2000). The various eif4f subunits are expressed to remarkably different levels in most cell types, with the eif4e subunit being the least abundant subunit (Mathews 2000). Importantly, formation of the eif4f complex is dynamic, and tightly regulated (Richter and Sonenberg 2005). In particular, eif4e availability for participation in eif4f formation is modulated by a family of small translation repressor molecules, the eif4ebinding proteins (4E-BPs) (Lin et al. 1994; Pause et al. 1994a). While hypophosphorylated 4E-BPs interact strongly with eif4e, hyperphosphorylated 4E-BPs do not (Gingras et al. 2001b). 4E-BP phosphorylation levels are modulated by many types of extracellular stimuli. In particular, hormonal or nutritional stimulation tend to increase 4E-BP1 phosphorylation levels, while environmental or nutritional stress elicits 4E-BP dephosphorylation (Mathews 2000; Gingras et al. 2001b). Thus, a binary subcomplex consisting of eif4g and eif4a (eif4g/4a) appears to exist in a dynamic equilibrium with eif4f. This equilibrium may be shifted to increase or decrease eif4f formation in response to nutrients, hormonal stimulation or stress (Mathews 2000). Internal translation initiation on most IRES-containing mrnas, such as encephalomyocarditis virus (EMCV) mrna, requires the same canonical eifs that are required for translation of capped mrnas, except for eif4e (Anthony and Merrick 1991b; Pause et al. 1994b; Pestova et al. 1996a). In contrast to typical eif4e-mediated ribosomal recruitment, the initial step in recruitment of the ribosome to the EMCV IRES is the eif4a-dependent high-affinity binding of the central domain of eif4g to the J-K stem-loop of the IRES (Lamphear et al. 1993; Lomakin et al. 2000a). Subsequent addition of the 40S ribosomal subunit, presumably via the eif4g-eif3-40s interaction, and the 60S subunit complete the assembly of the initiation complex. EMCV and other picornavirus infections are accompanied by a shutoff of host cell protein synthesis (J. W. B. Hershey 1996). In cells infected with poliovirus (PV), human rhinovirus, and foot-and-mouth disease virus, the primary event responsible for this shutoff is the cleavage of the eif4g isoforms by virus-specific proteases (Etchison et al. 1982; Wimmer 2002). The C-terminal cleavage fragment of eif4g can efficiently support IRES-dependent, but not cap-dependent translation, as it retains the binding sites for 54

75 IRES, eif4a and eif3, but cannot bind eif4e. This situation is akin to a net increase of eif4g/4a at the expense of the eif4f complex. While infection of cells with EMCV also inhibits host cell protein synthesis, this inhibition develops more slowly than that caused by PV and is not mediated by cleavage of eif4g (Jen et al. 1980; Mosenkis et al. 1985). Although eif4g is not cleaved in EMCV-infected cells, it is highly likely that the ratio of eif4g/4a to eif4f is also increased during EMCV infection. We previously described the dephosphorylation and activation of 4E-BP1 following EMCV infection (Gingras et al. 1996b). Inasmuch as 4E-BP1 dephosphorylation coincides with the shutoff of host mrna translation in EMCV-infected cells, we hypothesized that these two events are causally related (Gingras et al. 1996b). Modulating eif4f levels may also favor viral protein synthesis, as suggested by experiments employing rapamycin and wortmannin, two inhibitors of 4E-BP1 phosphorylation (Beretta et al. 1996a; Svitkin et al. 1998). Upon forced dephosphorylation of 4E-BPs by treating cells with rapamycin and wortmannin at the beginning of EMCV infection, viral protein synthesis and viral titers were higher than in untreated control cells (Beretta et al. 1996a; Gingras et al. 2001b). However, because rapamycin and wortmannin also have other cellular targets (Gingras et al. 2001b), and because these in vivo studies were merely correlative, it was critical to directly assess the function of eif4f in the translation of EMCV mrna. We were recently able to reconstitute EMCV translation and replication in Krebs-2 cell extract (Svitkin and Sonenberg 2003). This system enabled us to address the importance of eif4f subunit composition in viral protein synthesis and replication. Here we report that when EMCV mrna is translated in competition with cellular mrnas (i.e., in extracts that are not treated with nuclease), addition of 4E-BPs significantly augments viral protein synthesis. In contrast, addition of eif4e dramatically inhibits viral protein synthesis. Furthermore, when eif4f is converted to the eif4g/4a subcomplex by eif4e knockdown, the onset of viral protein synthesis in EMCV or PV-infected cells is markedly accelerated and the viral yield is higher. These findings demonstrate that active eif4e functions as a negative modulator of IRES-mediated translation by increasing competition from capped mrnas for the eif4f complex. Results 55

76 Stimulation of EMCV IRES-mediated translation by 4E-BPs in an untreated cellfree translation system. We sought direct evidence of a role of the eif4g/4a subcomplex in the regulation of EMCV IRES-directed translation in vitro. Previously, we added recombinant 4E-BPs to a nuclease-treated rabbit reticulocyte lysate (RRL) to sequester eif4e and prevent its incorporation into eif4f; these conditions did not result in stimulation of EMCV IRES-mediated translation (Pause et al. 1994a; Svitkin et al. 1996). These data are in contrast to observations from EMCV-infected cells in vivo, where the prevention of eif4e incorporation into eif4f by rapamycin and wortmannin stimulated viral protein synthesis (Beretta et al. 1996a; Svitkina and Borisy 1998). We hypothesized that cellular mrnas present in virus-infected cells sequester eif4f, making this factor limiting for viral translation. Generation of eif4g/4a, which does not bind capped mrnas efficiently, would be expected to relieve this competition. We therefore mimicked in vivo conditions by performing assays under conditions of mrna competition, in extracts in which endogenous mrnas were not destroyed by pretreatment with a nuclease. We utilized a translation system derived from Krebs-2 cells to study the effects of 4E-BPs on the translation of capped [cap-luc(a+)] or EMCV IRES-containing [EMCV IRES-luc(A+)] polyadenylated luciferase mrnas. In this system, both 4E-BP1 and 4E-BP2 inhibited translation of cap-luc(a+) mrna by three- to fourfold, similar to the cap analog, m 7 GDP, which was used as a positive control (Fig. 12. A.; compare open bars 3, 5, and 15 with bar 1). Importantly, and in contrast to observations in nucleasetreated RRL, 4E-BPs and m 7 GDP stimulated translation of EMCV IRES-luc(A+) mrna (~threefold; compare black bars 3, 5, and 15 with bar 1). Both cap- and IRES-dependent translation were unaffected by 4E-BP1 4E, a 4E-BP1 mutant protein lacking the eif4e binding site (bars 4) (Gingras et al. 1999a; Marcotrigiano et al. 1999). Exogenous recombinant eif4e, either with or without a GST tag, stimulated cap-dependent translation (four- to fivefold), but inhibited EMCV IRES-directed translation (three- to fourfold; Fig. 12. A; compare black bars 6 and 8 with 1 for IRES-driven translation). The effects of eif4e on translation were exerted via the eif4e/4g complex, as they were negated by the W73A mutation in eif4e, which abolishes this complex formation (bars 7) (Gingras et al. 2001a). Neither cap-dependent nor IRES-driven translation was significantly affected by eif4a 56

77 Figure 12. Regulation of cap-dependent and cap-independent translation by effectors of eif4f function in Krebs-2 cell extracts. A) Translation in untreated extract. Cap-luc(A+) and EMCV IRES-luc(A+) mrnas (5 μg/ml) were translated in12.5 μl reactions at 32 C for 90 min in the presence of unlabeled methionine (Svitkin et al. 1998). Prior to the additions of mrna, the extracts were pre-incubated at 32 C for 2 min with either control buffer (control) or the following components: GST (20 μg/ml), GST-4E- BP1, GST-4E-BP1 4E, GST-4E-BP2, GST-eIF4E, GST-eIF4EW73A (40 μg/ml each), eif4e (16 μg/ml), eif4a (80 μg/ml), eif4g-ct (40 μg/ml), GDP, or m 7 GDP (0.5 mm), as indicated in the figure. Where indicated (2A), the reaction mixtures contained extract treated with 2Apro (25 μg/ml, 32 C and 5 min) (Svitkin et al. 2001b). B) Translation of cap-luc(a+) and EMCV IRES-luc(A+) mrnas in nuclease-treated extract. Protein additions and translation conditions were as described in panel A, except for S10, which was nuclease-treated. Luciferase activity (RLU) was determined as previously described (Svitkin and Sonenberg 2004) and is shown as percentage of that of the control sample. Data represent the average of three independent determinations. Error bars indicate the standard deviation from the mean. 57

78 A B Figure 12

79 (Fig. 12. A, bars 9). However, eif4g-ct, the C-terminal portion of eif4g that does not contain the eif4e-binding site and corresponds to the C-terminal picornavirus protease cleavage fragment (Lamphear et al. 1993), stimulated IRES-mediated translation approximately fourfold (Fig. 12. A.; compare black bar 10 with 1). An even more striking (>tenfold) enhancement of EMCV IRES-directed translation was observed upon cleavage of eif4g by rhinovirus protease 2A (2A pro ; Fig. 12.A.; compare black bar 11 with 1). Addition of 4E-BP1 did not further potentiate this effect. Also, eif4e did not influence IRES-mediated translation when eif4g was cleaved. Overall, there was an inverse correlation between the efficiencies of cap-dependent and IRES-dependent translation, suggesting that they are oppositely regulated by eif4f. These results also demonstrate that stimulation of EMCV IRES-directed translation by 4E-BPs can be reproduced in vitro. To prove that the relative excess of the eif4g/4a subcomplex as compared with the intact eif4f complex indirectly stimulates EMCV IRES-directed translation by decreasing competition from cellular mrnas, we performed assays similar to those above using an extract in which endogenous cellular mrnas were degraded by nuclease treatment (Fig. 12.B). EMCV IRES-directed translation was enhanced (~threefold) in the nuclease-treated extract (data not shown), demonstrating that competing cellular mrnas in the untreated extract indeed had an inhibitory effect on EMCV IRES activity. Our results were consistent with those reported for nuclease-treated RRL (Pause et al. 1994a; Svitkin et al. 1996). Although addition of 4E-BPs and m 7 GDP strongly inhibited capdependent translation (eight- to 20-fold), these components did not stimulate EMCV IRES activity in the nuclease-treated extract; in fact, 4E-BPs had a slightly adverse effect on EMCV IRES activity (Fig. 12.B; compare black bars 3 and 4 with 1). Nuclease treatment also abolished the ability of eif4e to inhibit translation from the EMCV IRES (Fig. 12.B; compare black bars 5 and 1). In contrast to observations in the untreated extract, 2A pro treatment or eif4g-ct addition did not substantially stimulate EMCV IRES activity in the nuclease-treated extract (Fig. 12.B.; compare black bars 8, 7 and 1). Taken together, these findings suggest that a relative excess of the free eif4g/4a subcomplex, compared with eif4f, upregulates EMCV IRES-driven translation only in the presence of competing cellular mrnas. 58

80 The eif4g/4a subcomplex is essential and limiting for EMCV replication in untreated extract. We next examined whether 4E-BP1 and eif4e modulate translation from the EMCV IRES when the full-length EMCV mrna is used in untreated extract. EMCV mrna was translated in the presence of increasing concentrations of 4E-BP1 or eif4e. Translation products were resolved by SDS-PAGE, transferred to a PVDF membrane, and detected by autoradiography. [ 35 S]methionine incorporation into polypeptides in the untreated extract primarily reflected elongation of preexisting polypeptide chains. However, consistent with the contribution from de novo translation initiation, incorporation of [ 35 S]methionine into cellular proteins was inhibited by 4E-BP1 and stimulated by eif4e (Fig. 13.A.; note the corresponding changes in the intensities of two prominent cellular proteins, p47 and p50). The latter observation indicates that eif4e is limiting for translation of endogenous cellular mrnas. It was difficult to discern virusspecific polypeptides on this autoradiograph due to the high degree of labelling of endogenous cellular proteins. We therefore assessed the efficiency of EMCV mrna translation by western blotting using an antibody against the nonstructural protein 3Dpol (an RNA-dependent RNA polymerase). The addition of 4E-BP1 stimulated (up to 3.2- fold) 3Dpol synthesis (Fig. 13.B., compare lanes 5 and 2). In contrast, eif4e dramatically inhibited 3Dpol synthesis by up to 14-fold (Fig. 13.B.; compare lanes 9 and 6). Thus, in the untreated extract, translation of full-length EMCV mrna was upregulated by 4E- BP1 and downregulated by eif4e. Thus, the effects of 4E-BP1 and eif4e on the translation of full-length EMCV mrna were similar to those measured using the surrogate template, EMCV IRES-luc(A+) mrna. To determine whether the stimulation of viral protein synthesis by 4E-BP1 is sufficient to affect EMCV RNA replication, we pulse-labelled the reaction mixtures with [ - 32 P]CTP 4 h after the beginning of incubation with 4E-BP1. The newly synthesized RNA was extracted and analyzed by agarose gel electrophoresis and autoradiography. EMCV RNA synthesis was stimulated approximately 15-fold in the presence of 4E-BP1 (Fig. 13.C.; compare lanes 2 and 1). Conversely, addition of eif4e reduced RNA synthesis to below detectable levels (Fig. 13.C.; compare lanes 3 and 1). Consistent with the importance of intact eif4g for the eif4e-mediated inhibition of translation, cleavage of eif4g by 2A pro restored viral RNA synthesis (Fig. 13.C.; compare lane 4 with lanes 3 and 1). 59

81 Figure 13. EMCV mrna translation, RNA replication, and virus yield in the untreated EMCV mrna-programmed S10 extracts. A) Effects of 4E-BP1 and eif4e concentration on protein synthesis in untreated EMCV mrna-programmed Krebs-2 cell extract. [ 35 S]methionine labeling of proteins was performed in a 20 μl total reaction volume in the absence (lane 1) or presence (lanes 2-9) of EMCV mrna (4 μg/ml). Prior to the additions of mrna, the extracts were pre-incubated with the indicated proteins, as described in the legend to Fig. 12. GST-4E-BP1 was used at 15, 30, and 60 μg/ml (lanes 3, 4, and 5, respectively). eif4e was used at 3, 6, and 12 μg/ml (lanes 7, 8, and 9, respectively). Translation products were separated by SDS-PAGE and transferred to a PVDF membrane. The autoradiograph of the membrane is shown. The positions of two abundant cellular proteins (p47 and p50, arrowheads), the EMCV-specific protein 3Dpol (arrow), and the [ 14 C]methylated protein molecular weight markers (GE Healthcare) are indicated. B) Western blotting analysis of EMCV-specific protein 3Dpol synthesis. The middle portion of the membrane from panel A was probed with anti-3dpol as described in Materials and Methods. 3Dpol band intensities in different lanes were compared using NIH Image Version 1.63 software. The values obtained from reactions performed in the absence of added proteins (lanes 2 and 6) were defined as 100%. C) EMCV RNA replication was assayed in 30-μl reactions containing untreated extract, EMCV mrna (4 μg/ml), and other components as described in Materials and Methods. Before mrna addition, reactions were pre-incubated with control buffer (lane 1), GST-4E-BP1 (lane 2), eif4e (lane 3), or a combination of eif4e and 2A pro (lane 4), as described for Fig. 12. A. The RNA products were pulse-labeled with [ - 32 P]CTP after 4-5 h of incubation at 32 C and analyzed by agarose gel electrophoresis and autoradiography. The position of the intact EMCV mrna is indicated (vrna). D) Reactions (30 μl) pre-incubated with either control buffer or the indicated components (as described for Fig. 12. A) and programmed with EMCV mrna (4 μg/ml), were incubated for 20 h at 32 C. The samples were then treated with a mixture of RNase A and T1 and assayed for infectivity after appropriate dilution, as described in Materials and Methods. Values represent the average of three independent titer determinations. Error bars indicate the standard deviation from the mean. 60

82 A 4E-BP1 EMCV mrna eif4e Mr 97 p50 p D 30 B D D band intensity (% of control) C control 4E-BP1 eif4e 2A+eIF4E vrna D PFU/ml (10 3 ) Figure

83 We then examined the effects of 4E-BP1, 4E-BP2, eif4e, eif4a, and eif4g-ct on EMCV yield. EMCV titers in reactions supplemented with different factors were determined after 20-h incubation (Fig. 13. D.). Strikingly, 4E-BP1 and 4E-BP2, as well as m 7 GDP, stimulated EMCV synthesis 24- to 35-fold (compare bars 3, 5 and 15 with 1), whereas 4E-BP1 4E, which cannot bind eif4e, had only a marginal effect. Conversely, eif4e, but not the eif4e W73A mutant, dramatically decreased the viral titer by 150- to 200-fold (Fig. 13. D.; compare bars 6 and 8 with 1). Addition of the C-terminal portion of eif4g or 2A pro potently stimulated infectivity (7 and 15-fold, respectively; compare bars 10 and 11 with 1), and this enhancement was not influenced by co-addition of eif4e or 4E-BP1. Overall, EMCV titers under different conditions co-varied with luciferase expression from EMCV IRES-luc(A+) mrna (compare Fig. 13. D. with Fig. 12. A). However, the magnitude of the changes in viral titer was greater than that measured for translation efficiency. Importantly, and in agreement with a role for mrna competition in the regulation of viral RNA translation, addition of eif4e to the EMCV mrna-programmed/nucleasetreated extract neither inhibited EMCV mrna translation (as judged by the accumulation of the viral protein 3D) nor changed the expression pattern of virus-specific polypeptides (Fig. 14. A., compare lane 4 with 2). Also, eif4e had no effect on EMCV synthesis (Fig. 14. B). These results rule out the possibility that contaminating bacterial proteins, which may be present in the eif4e preparation, adversely affected EMCV replication. Nuclease treatment also abolished the stimulatory effect of 4E-BP1 on EMCV translation and replication (Fig. 14. A. and B). Although these negative controls argued for the importance of mrna competition in the regulation of viral protein expression by eif4f, they did not rule out an alternative possibility. Specifically, the detrimental effect of nuclease treatment might be a consequence of destruction or inactivation of some labile regulatory components of the extract. To address this possibility, we restored mrna competition by adding saturating concentrations of capped mrnas (either total poly(a)+ mrna isolated from the cytoplasm of Krebs-2 cells or globin mrna) along with EMCV mrna to the nuclease-treated extract and examined the effects of 4E-BP1 and eif4e on viral protein expression. Translation of poly(a)+ mrna alone yielded heterogeneous polypeptides similar to the products of endogenous mrna translation in the untreated 61

84 Figure 14. 4E-BP1 and eif4e have no effect on EMCV protein synthesis and replication in nuclease-treated Krebs-2 cell extract. A) Products of EMCV mrna translation. Reactions contained the nuclease-treated Krebs-2 cell extract but otherwise were identical to that described in Fig. 13. A. Reactions were pre-incubated with control buffer, GST-4E-BP1 (60 μg/ml) or eif4e (12 μg/ml) where indicated. Translation was performed at 32 C for 3 h. Aliquots (5 μl) of the translation reactions were analyzed by SDS-PAGE. An autoradiogram of the dried gel is shown. [ 35 S]methionine incorporation into EMCV-specific protein 3Dpol was quantified using a Fuji BAS2000 phosphorimager. The value obtained from reaction performed in the absence of added protein (lane 2, control) was defined as 100%. B) EMCV yields. Reaction mixtures (30 μl, unlabelled) were pre-incubated with control buffer, GST-4E-BP1, or eif4e and programmed with EMCV mrna, as specified above. Plaques were scored following incubation for 20 h at 32 C and RNase treatment. The data presented in panels B and C represent the average of three determinations. Error bars indicate the standard deviation from the means. C) Co-addition of capped mrna competitors rescues the regulation of translation of EMCV mrna by 4E-BP1 and eif4e in nuclease-treated extract. EMCV mrna (4 μg/ml) was translated at 32 C for 90 min in the absence (lanes 2 and 9) or presence of either total Krebs-2 cell poly(a)+ mrna (40 μg/ml, lanes 3-6) or globin mrna (10 μg/ml, lanes 10-13). Reactions were pre-incubated with control buffer, GST- 4E-BP1 (60 μg/ml), eif4e (12 μg/ml) or m 7 GDP (0.5 mm) where indicated. Products of translation of Krebs-2 cell poly(a)+ mrna or globin mrna alone are shown in lanes 1 and 8, respectively. No mrna was added to the reaction analyzed in lane 7. Relative values for [ 35 S]methionine incorporation into EMCV-specific protein 3Dpol were determined as in panel A. On panels A and C, the assignment of EMCV polypeptides was as described previously (Svitkin and Sonenberg 2003). The positions of the [ 14 C]methylated protein molecular weight markers (GE Healthcare) are also shown. An asterisk on panel C indicates the position of globin. 62

85 A control 4E-BP1 eif4e EMCV mrna Mr P1-2A P1 P3 3CD 1ABC 3D 1AB/3ABC 2C 1D 1C 3C 2A 2B 3D (% of control) B C globin mrna EMCV mrna poly(a)+ mrna Mr control + 4E-BP1 + eif4e + m7gdp control + 4E-BP1 + eif4e + m7gdp L-P1-2A P1-2A P1 P3 3CD 1ABC 3D 2C 1D 3C * 2A 2B 3D (% of control) Figure 14

86 extract (Fig. 14. C., lane 1). Globin mrna translation yielded a 15-kDa polypeptide as expected (lane 8). When EMCV mrna was translated in the presence of poly(a)+ or globin mrna the expression of viral proteins was reduced two to threefold (compare lanes 3 with 2 and 10 with 9). In parallel, 18S ribosomal RNA was used as a negative control and found not to inhibit viral translation (data not shown). (It should be noted that a molar excess of capped mrnas over EMCV mrna was used in these experiments. This was to mimic the initial stage of infection when viral mrna constitutes a minor fraction of total mrna.) Addition of 4E-BP1 to the system programmed with a mixture of EMCV mrna and capped mrnas stimulated virus protein expression two to threefold (compare lanes 4 with 3 and 11 with 10), similar to the addition of m 7 GDP (compare lanes 6 with 3 and 13 with 10). In contrast, eif4e markedly inhibited viral protein synthesis ( fold inhibition, compare lanes 5 with 3 and 12 with 10). Thus, the addition of capped mrnas to the EMCV mrna-programmed nuclease-treated extract rescues the regulation of viral protein expression by eif4f. Specific stimulation of EMCV and PV translation in cells treated with sirna against eif4e. To provide evidence that eif4g/4a subcomplex concentration determines the rate of EMCV protein synthesis in vivo, we used RNA interference (RNAi) to specifically deplete eif4e (Zamore et al. 2000). Such depletion would be expected to decrease competition from cellular mrnas by increasing the amount of the eif4g/4a complex available for viral mrna translation. The selected sirna elicited strong (~85%) knockdown of eif4e (see Fig. 15. A, Fig.16. B and Fig. 17. B). eif4e knockdown did not lead to a decrease in the overall abundance of eif4gi or eif4ai, which account for the majority of total eif4g and eif4a (Conroy et al. 1990; Svitkin et al. 1999), but dramatically decreased the amount of eif4f, as determined by cap-column pull-down assays (Fig. 15. A and data not shown). To determine whether down-regulation of eif4e decreases the rate of translation initiation, polyribosomes isolated from the control and eif4e knockdown cells were fractionated by sucrose density gradient centrifugation. eif4e knockdown cells displayed a higher 80S monosome/polyribosome ratio, compared to control. In addition, a small shift of the polyribosome distribution in favor of lighter polysomes was evident (Fig. 15. B). The reduction in polyribosome loading after eif4e depletion is 63

87 Figure 15. The abundance of eif4f components and polyribosome distribution as affected by eif4e knockdown. A) Total HeLa cell extracts (50 μg protein) prepared from control and eif4e knockdown cells were resolved by SDS-15% PAGE and analyzed by western blotting using the indicated antibodies. B) Control and eif4e knockdown cells were lysed by incubating in a hypotonic buffer (5 mm Tris-HCl, ph 7.5, 1.5 mm KCl, 2.5 mm MgCl2, 2 mm DTT, 100 μg/ml cycloheximide, 0.5% Triton X-100, and 0.5% sodium deoxycholate). Nuclei were removed by centrifugation (16,000X g, 3 min, and 4 C). Aliquots of extracts containing equal amounts of RNA (OD260 units) were loaded on 12 ml 5-50% sucrose density gradients prepared in buffer containing 20 mm HEPES- KOH, ph 7.5, 100 mm KCl, and 5 mm MgCl2. Centrifugation was at 35,000 rpm, 4 C and 2 h in a Beckman SW40Ti rotor. Polyribosome profiles were traced at OD254 using an absorbance detecting system (ISCO UA-6). The positions of the 80S ribosomes are indicated. The results are representative of three independent experiments. 64

88 A control eif4e B control 80S eif4e polysomes eif4gi 80S polysomes eif4ai eif4e OD254 β-actin Sedimentation Figure 15

89 Figure 16. eif4e knockdown stimulates PV protein synthesis and replication. A) Time course of protein synthesis in PV-infected cells. sirnas directed against eif4e or a nonspecific target (control) were transfected into HeLa S3 cells. After transfection, cells were infected with PV type 1 (Mahoney) and protein synthesis was examined by pulse labeling with [ 35 S]methionine as described in Fig. 15. A. The labeled polypeptides were resolved by SDS-PAGE, transferred to a PVDF membrane and analyzed by autoradiography. M (lanes 6 and 12): mock-infected cells labeled at 6 h. The positions of the major PV-specific proteins are indicated on the right. B) eif4e levels in cells, as analyzed by western blotting. The membrane from panel A was probed with anti-eif4e, and signals were quantified. The average level of eif4e depletion for lanes 7-12 (versus lanes 1-6) was 88%. C) -actin detection by western blotting (a loading control). D) PV yield, as affected by eif4e sirna treatment. PV-infected cells (eif4e knockdown or control, unlabeled) were lysed 6 h post-infection. Virus titer was measured as described in Materials and Methods. Results are averages for three assays with standard deviations from the mean. 65

90 A sirna control eif4e Time (h) M M P1 P3 3CD 2BC 1AB 2C 1D 1C 2A 3AB B eif4e C β-actin D Figure 16

91 consistent with the inhibition of cellular mrna translation initiation under these conditions. To examine the effect of eif4e knockdown on viral protein synthesis, cells were pulselabelled with [ 35 S]methionine at various times after EMCV infection. At 4 h postinfection, only trace amounts of viral proteins could be detected in cells pretreated with the control sirna (Fig. 17. A, lane 8). In contrast, in cells treated with sirna against eif4e, viral protein synthesis was robust by 4 h post-infection (lane 3). The stimulatory effect of eif4e knockdown on viral protein synthesis was also observed after 5 h of infection (compare lane 4 to 9). After 6 h of infection, the eif4e knockdown cells, but not control cells, exhibited morphological changes, indicative of virus-induced cytopathic effect, accompanied by a general decline in protein-synthesizing capacity (Fig. 17. A., lane 5 and data not shown). These data suggest that knocking down eif4e expression accelerates EMCV protein synthesis. We also analyzed viral titers recovered from control and eif4e knockdown cells at 4 h post-infection. An approx tenfold increase in infectious virus production was associated with eif4e knockdown (Fig. 17. D. and E). It is worth mentioning that this elevation in the virus titer, although significant, is less than that exerted by the sequestration of eif4e by 4E-BPs in vitro (24-28-fold stimulation; see Fig. 13. D.). We attribute this to the fact that some residual eif4e (10-20%), and by inference some eif4f, is present in eif4e knockdown cells. On the contrary, sequestering of eif4e by 4E-BPs excess in vitro would completely disrupt eif4f. The abundance of the eif4g/4a subcomplex could also play a role at an early stage of enterovirus infection when cleavage of eif4g is not yet accomplished. To test this hypothesis, we examined the effect of eif4e knockdown on PV infection. HeLa cells transfected with either control sirna or sirna directed against eif4e were infected with PV, and the kinetics of viral protein synthesis were analyzed by [ 35 S]methionine pulse labelling. As with EMCV, eif4e knockdown significantly shortened the eclipse phase of infection, during which no virus proteins can be detected (see Fig. 16. A). Higher rates of PV protein synthesis were evident at 3, 4, and 5 h post-infection in eif4e knockdown cells as compared to control cells (Fig. 16. A). Consistent with these results, cells depleted of eif4e exhibited a 66

92 Figure 17. eif4e knockdown stimulates translation and replication of EMCV in vivo. A) Time course of protein synthesis in EMCV-infected cells. sirna against eif4e or a nonspecific sirna (control) was transfected into HeLa S3 cells. eif4e knockdown or control cells were infected with EMCV, and protein synthesis was examined by pulse labeling with [ 35 S]methionine at the indicated time points. After labeling, polypeptides were resolved by SDS-PAGE and transferred to a PVDF membrane. The autoradiograph of the membrane is shown. The positions of the major EMCV-specific proteins are indicated on the right. B) eif4e levels in cells, as analyzed by western blotting. The membrane from panel A was probed with anti-eif4e, and signals were quantified as described in Materials and Methods. The average level of eif4e depletion for lanes 1-5 (versus lanes 6-10) was 86%. C) -actin detection by western blotting (a loading control). D) Plaque assays of the indicated dilutions of the lysates from control and eif4e knockdown cells 4 h after infection. E) EMCV yield, as affected by eif4e sirna treatment. EMCV-infected cells (eif4e knockdown or control, unlabeled) were lysed at 4 h post-infection. Viral titer was measured as described in the legend to Fig. 13. D. 67

93 A sirna eif4e control Time (h) P1 3CD 3D 1AB 2C 1D 1C 3C 2A B eif4e C β-actin D sirna control eif4e E Figure 17

94 PV-induced cytopathic effect earlier than control cells, and produced more PV (Fig. 16. D, and data not shown). Thus, eif4e appears to be a general, rather than EMCV mrnaspecific, inhibitor of IRES-mediated translation. An unlikely possibility that cannot be rigorously excluded is that eif4e depletion primarily stimulates viral RNA replication and that the enhancement of viral protein accumulation is a secondary effect. Discussion Because eif4e is not required for translation by internal ribosome entry, it has been generally assumed that this factor does not play a role in the regulation of IRES activity. Here we demonstrate that eif4e is, in fact, a negative regulator of EMCV mrna translation under conditions of competition with cellular mrnas. Saturation of the eif4g/4a subcomplex with eif4e to generate eif4f dramatically decreases EMCV mrna translation and virus yield in untreated extracts. In contrast, sequestration of eif4e by 4E-BPs in vitro or depletion of eif4e in vivo stimulates EMCV mrna translation and increases viral titer. These results imply that the intact eif4f complex is unable to support efficient IRES function because it is sequestered by capped mrna. Freeing the eif4g/4a subcomplex relieves competition from cellular mrnas, thereby favoring viral mrna translation. Our findings support the idea that the cytoplasmic concentration of active eif4f is less than the concentration of total cellular mrnas (Anthony and Merrick 1991a; Thach 1992). The idea that a discriminatory initiation factor regulates competition between cellular and viral mrnas in EMCV infection was first proposed three decades ago (Lawrence and Thach 1974; Golini et al. 1976; Svitkin et al. 1978). However, the limiting step and limiting components in translation were not defined. Our results clearly demonstrate that eif4f plays this discriminatory role through the availability of its cap-binding subunit, eif4e. The abundance of functional eif4e, which is regulated by 4E-BPs, thus acts as a switch between cap-dependent and IRESmediated translation. How does eif4e dissociation from eif4f enhance virus-specific translation? eif4e dissociation is believed to cause a conformational change in eif4gi that can be detected by its slower rate of cleavage by picornavirus proteases (Haghighat et al. 1996; Ohlmann et al. 1997). However, several lines of evidence suggest that this conformational change 68

95 cannot account for the stimulatory effect of the eif4g/4a subcomplex on virus-specific translation. Firstly, UV cross-linking experiments suggest that eif4g binds efficiently to the EMCV IRES as a component of the eif4f complex (Pestova et al. 1996c). Secondly, the cap-analog m 7 GDP, which inhibits the cap-binding activity of eif4f but does not alter eif4f assembly, stimulates EMCV IRES activity and viral production in a manner similar to 4E-BPs (Fig. 12. A and 2D). Finally, and most importantly, 4E-BP1 and eif4e have no effect on viral RNA translation in a nuclease-treated extract (Fig. 14 A and B). Thus, in the reconstituted system or in nuclease-treated extract, EMCV IRES appears to interact with the eif4f or eif4g/4a complexes with comparable efficiency. We therefore conclude that competition from cellular mrnas for eif4f is required for the regulation of EMCV synthesis by eif4e and 4E-BPs. Luciferase translation from the EMCV IRES is enhanced in response to eif4g cleavage or upon addition of the eif4g C-terminal protein fragment to the extract. Similar results have been reported for PV IRES-mediated translation (Belsham and Jackson 2000; Svitkin et al. 2001a). If this effect were to be influenced by mrna competition, it should be more pronounced in the presence of competing cellular mrnas. Consistent with this prediction, we found that 2A pro treatment stimulated EMCV IRES activity much more potently in untreated than in nuclease-treated extracts [11-fold versus 1.4-fold, compare Fig. 12. A (black bar 11) and B (black bar 8)]. Likewise, cellular mrna competition was required for eif4g-ct or m 7 GDP-mediated stimulation of IRES activity, as this stimulation occurred exclusively in untreated extracts. RNA replication and virus yield correlated with EMCV mrna translation efficiency, indicating that translation is the limiting step in virus replication. Strikingly, the magnitude of modulation of RNA replication and virion formation by 4E-BP1 and eif4e was substantially higher than the magnitude of their effect on EMCV mrna translation. Thus, effects associated with competition for translation factors are amplified at subsequent steps of the infectious cycle. Interestingly, eif4f complex dissociation is also beneficial for PV gene expression, in as much as eif4e-depleted or rapamycin-treated cells supported viral protein synthesis to a higher level than the respective control cells [Fig. 17. A, also see (Beretta et al. 1996b)]. Presumably, this stimulation occurs early in infection prior to eif4g cleavage, when the viral RNA must compete with cellular 69

96 mrnas for the limiting pool of intact eif4f. It remains to be determined whether eif4f dissociation stimulates infectious processes induced by other picornaviruses. A model illustrating the regulation of EMCV replication by eif4f is shown in Fig. 18. Central to this model is the fact that EMCV IRES does not compete efficiently with capped cellular mrnas for eif4f unless the cap-binding subunit eif4e is sequestered in a complex with the 4E-BPs, and the relative abundance of the eif4g/4a subcomplex is increased. As there is no cap-binding subunit within the eif4g/4a subcomplex, one can assume that it is not recruited efficiently by cellular mrnas. Indeed, in the presence of eif4a, the binding affinity of eif4g for -globin mrna is lower than that for EMCV IRES by up to 100-fold (Lomakin et al. 2000b). However, the concentration of the eif4g/4a subcomplex in HeLa cells is limiting for translation of picornavirus RNAs, since eif4e knockdown significantly augments the expression of viral proteins in EMCV and PV-infected cells. Because eif4a is not tightly associated with eif4g and recycles during translation (Pause et al. 1994c), some eif4g may also exist outside the eif4f complex or the binary eif4g/4a subcomplex. However, this free eif4g is not expected to bind EMCV IRES with high affinity (Lomakin et al. 2000b) and is therefore not shown in the model. The cellular tropism and pathogenesis of picornaviruses may be determined in part by the availability of IRES transacting factors (ITAFs), which bind specifically to picornavirus IRESs and regulate their function (Belsham and Jackson 2000). An intriguing idea is that tropism and pathogenesis are also influenced by the concentration of free eif4g/4a, which is dependent on the expression of the eif4f components and regulation of the abundance and phosphorylation state of 4E-BPs. Host permissiveness for virus translation could be limited by exposure of cells to extracellular stimuli that activate mtor through PI3K signaling and increase the phosphorylation of 4E-BPs (Raught et al. 2000a). Hence, it is likely that both cell-type specific and environmental factors affect the ability of the virus to establish an efficient infection. It is also possible that pathological conditions that affect the concentration of eif4f components will affect the outcome of a viral infection. For example, several cancers are associated with increased levels of eif4f subunits, in particular eif4e and eif4g (Hershey and Miyamoto 2000; Avdulov et al. 2004a). Under 70

97 Figure 18. A model explaining eif4f regulation of mrna competition in EMCVinfected cells. It is presumed that there is equilibrium between eif4f (eif4e/4g/4a) and a binary subcomplex comprised of eif4g and eif4a (eif4g/4a) and that EMCV mrna competes with capped cellular mrna for the recruitment of eif4g shared by these complexes. Saturation of the eif4g/4a subcomplex with eif4e to generate eif4f increases its recruitment by capped cellular mrnas and dramatically inhibits EMCV translation and replication (Fig. 12. and 13.). Hence, EMCV mrna encounters a strong competition from cellular mrna when it binds to eif4g within the ternary eif4f complex. By default, EMCV mrna uses eif4g within the binary eif4g/4a subcomplex, which is recruited inefficiently by capped mrna (Novoa and Carrasco 1999; Lomakin et al. 2000a). Dephosphorylated (active) 4E-BP1 and 4E-BP2 (designated as 4E-BP) trigger the expulsion of eif4e from the eif4f ternary complex. Elevation of the concentration of the eif4g/4a subcomplex, resulting from either 4E-BP activation (Beretta et al. 1996a; Svitkin et al. 1998) or eif4e knockdown (Fig. 17.), stimulates EMCV IRES-directed translation and downstream virus-specific processes. Thick solid and thin dashed arrows designate efficient and inefficient pathways, respectively. m 7 G and AAA denote the cap structure and the poly(a) tail of the mrna, respectively. VPg denotes the genome-linked protein of EMCV. 71

98 4A 4G 4E 4A + 4G + 4E-BP 4E 4E-BP (5') m7g cellular mrna AAA (3') (5') VPg IRES EMCV mrna AAA (3') capped mrna translation EMCV IRES-mediated translation EMCV RNA replication and virus assembly Figure 18

99 conditions of eif4g excess, cellular mrnas may not compete with picornavirus mrnas for translation (or compete in another fashion for another limiting component), and these cells would be expected to be highly susceptible to infection. In this regard, it is noteworthy that picornaviruses preferentially kill malignant cells over normal cells (Gromeier et al. 2000; Shafren et al. 2004a). However, it is not known whether the tumor cells used in these studies contained higher levels of eif4g than normal cells. In addition to its role in the expression of virus genomes, the ratio between the different eif4g complexes may regulate cellular proliferation, survival and death, as IRES elements are often found in the mrnas of genes controlling these processes (Johannes et al. 1999b; Hellen and Sarnow 2001a; Stoneley and Willis 2004; Holcik and Sonenberg 2005a). Our results suggest that IRES-mediated translation of cellular mrnas should not only be resistant to eif4f dissociation, but stimulated by it. The following examples of selective translation conform to this notion. Despite a reduction in overall protein synthesis, the X-linked inhibitor of apoptosis (XIAP) mrna, which possesses an IRES, is translated more efficiently under serum starvation, which decreases 4E-BP phosphorylation (Holcik and Sonenberg 2005a). Elevated levels of XIAP are thought to delay the onset of apoptosis and allow the cell to survive under stress conditions. A rapid inhibition of translation as a consequence of 4E-BP1 dephosphorylation and eif2 phosphorylation by PERK develops during hypoxia, which is common in many human diseases such as stroke, heart disease and cancer (Holcik and Sonenberg 2005a). However, at least two proteins (HIF1 and VEGF) involved in cell survival are upregulated during hypoxia, presumably via their synthesis by IRES-dependent translation. It is also noteworthy that in neurons from mollusk Aplysia californica the switch from cap-dependent to IRES-dependent translation is believed to be triggered by dephosphorylation of eif4e (Dyer et al. 2003). Understanding how selective translation allows cells to adapt to environmental and physiological stresses, such as hypoxia, heat shock, toxins and drug exposure, is important for understanding many human disorders and may lead to the development of new therapeutic approaches for such conditions. Materials and methods Materials. mrnas encoding luciferase were transcribed from pt3luc(a)+ and 72

100 pt7emcvluc(a)+ using T3 or T7 RNA polymerase, respectively (Svitkin et al. 2001b). EMCV mrna was isolated from purified virus by extraction with a mixture of phenol and chloroform (Svitkin et al. 2001b). Total poly(a)-containing RNA [poly(a)+ mrna] was isolated from the cytoplasm of Krebs-2 cells and purified by two cycles of chromatography on oligo(dt)-cellulose (Aviv and Leder 1972; Svitkin et al. 1978). Globin mrna was purchased commercially (Gibco Invitrogen Corporation, discontinued). The components used to prepare Krebs-2 cell extracts and to supplement translation reactions were as specified previously (Svitkin and Sonenberg 2004). Recombinant GST-4E-BP1, GST-4E-BP1 4E (4E-BP ), GST-4E-BP2, GSTeIF4E, GST-eIF4EW73A, eif4e, eif4a, eif4g-ct were described previously (Svitkin et al. 1996; Gingras et al. 1999a; Marcotrigiano et al. 1999; Gingras et al. 2001a; Svitkin et al. 2001b; Svitkin et al. 2001c). Protein expression was performed in E. coli BL21 (DE3) cells according to the manufacturer s instructions (GE Healthcare). To assess purity, proteins were analyzed by SDS-PAGE and Coomassie blue staining. 2A pro was a kind gift of H.-D. Liebig (Svitkin et al. 2001b). Mouse monoclonal antibody 8D10 against recombinant Mengovirus protein 3Dpol (Duque and Palmenberg 1996) was kindly provided by Ann Palmenberg (University of Wisconsin, Madison, WI). Mouse monoclonal antibodies against eif4e and -actin were purchased from BD Biosciences and Sigma, respectively. Rabbit polyclonal antibodies against eif4gi and eif4ai were previously described (Li et al. 1999; Morino et al. 2000). Secondary horseradish peroxidase (HRP)-conjugated sheep anti-mouse and donkey anti-rabbit antibodies were obtained from GE Healthcare. HeLa S3 and BHK-21 cells were obtained from American Type Culture Collection (ATCC numbers CCL-2.2 and CCL-10, respectively). Dulbecco s modified minimal essential medium (DMEM), Lipofectamine 2000 and OPTIMEM were from Invitrogen. Krebs-2 cell extract preparation. Krebs-2 ascites cell propagation in mice and the preparation of extracts was as described (Svitkin and Sonenberg 2004). Before homogenization, cells were suspended in methionine-free DMEM and incubated at 37 C for 2 h with gentle agitation. The cells were broken with a Dounce homogenizer, and a post-mitochondrial supernatant (S10) was obtained by a high-speed centrifugation (18,000 X g, 4 C and 20 min). Where indicated, the extracts were treated with 73

101 micrococcal nuclease in the presence of CaCl 2 (Svitkin and Sonenberg 2004). In vitro assays for EMCV mrna translation, RNA replication and virion synthesis. EMCV mrna translation and replication reactions (30 μl) that contained either untreated or nuclease-treated Krebs-2 S10 extract were programmed with EMCV mrna (4 μg/ml), as described previously (Svitkin and Sonenberg 2003). For protein labeling, reactions were supplemented with [ 35 S]methionine. After incubation at 32 C for h, reactions were stopped with Laemmli sample buffer. Protein products were resolved by SDS- PAGE (15% gels), electrophoretically transferred to a polyvinylidene difluoride (PVDF) membrane and detected by autoradiography. Western blotting for 3Dpol was performed as described below. RNA replication and virion production were assayed in the reactions that contained unlabeled methionine as described previously (Svitkin and Sonenberg 2003). For RNA labeling, [ - 32 P]CTP was added to the reactions after 4 h of incubation. One hour later, RNA was extracted, and RNA products were analyzed by native 1% agarose gel electrophoresis and autoradiography (Svitkin and Sonenberg 2003). To assay for EMCV synthesis, reactions were incubated at 32 C for 20 h and treated with a mixture of RNase A and T1 (Svitkin and Sonenberg 2003). Plaque assays were performed using serial dilutions of samples as described below. sirna transfection. Target sequences for sirna were designed using the Dharmacon web-based criteria and were purchased from Dharmacon. The positions and sequences of the sirnas used in this study are listed in Table 1. HeLa S3 cells were seeded in a 24- well culture dish at a density of 7x10 5 cells per well. sirna transfection was performed using Lipofectamine 2000 as described (Costa-Mattioli et al. 2004). Virus infections and metabolic radiolabeling. Forty-eight hours after sirna transfection, HeLa S3 cells were infected with EMCV or PV at a multiplicity of infection of 5 PFU per cell. Virus adsorption was at room temperature for 30 min. The medium was then replaced with methionine-free DMEM and the incubation was continued at 37 C. At various times post-infection (see figure legends), the media were replaced with media containing 35 S-protein labeling mix (10 μci/ml). After 30 min of labeling, the cell 74

102 monolayers were washed with phosphate-buffered saline and lysed with Laemmli sample buffer. Radiolabeled proteins were resolved by SDS-PAGE (15% gels), transferred to a PVDF membrane, and detected by autoradiography. The same membrane was used for western blotting. Western blotting. PVDF membranes were blocked with Tris-buffered saline/0.1% (v/v) Tween-20 containing 5%-nonfat dry milk and probed with the indicated antibodies. The antibodies against eif4e, eif4gi, eif4ai, and -actin were used diluted 1:500, 1:1,000, 1:1,000, and 1:5,000, respectively. The antibody against Mengovirus protein 3Dpol was used at a dilution of 1:1,500. After washing, the membrane was incubated with HRPconjugated anti-mouse or anti-rabbit antibody, as appropriate (diluted 1:5,000). HRP was detected using Western Lightning chemiluminescence kit as recommended by the manufacturer (Perkin-Elmer Life Sciences, Inc.). Plaque assays. Plaque assays were performed as previously described using confluent monolayers of either BHK-21 cells (for EMCV) or HeLa R19 cells (for PV) in 60-mmdiameter plates (Rueckert and Pallansch 1981). Virus-infected cells from 24-well dishes were lysed in 500 μl DMEM per well by three cycles of freezing and thawing. Cell debris was removed by centrifugation (10,000 X g, 4 C and 5 min) and the supernatants were diluted with DMEM containing 2% fetal bovine serum. Cells were infected with 250 μl of serially diluted lysates. Plaques were allowed to develop under semisolid agar for 26 h (EMCV) or 36 h (PV) at 37 C and were detected by staining with 1% crystal violet. 75

103 TABLE 1. Positions and sequences of the sirnas used to knockdown gene expression Synthetic sirna Positions in the ORF sirna sequence a eif4e sirna GGACGAUGGCUAAUUACAUdTdT-3 3 -dtdtccugcuaccgauuaaugua -5 Control sirna b '- CGUACCGUGGAAUAGUUCC dtdt-3' 3'-dTdTGCAUGGCACCUUAUCAAGG -5' a dt, deoxyribosylthymine; b sirna against 4E-T (inverted). 76

104 Preface The translational apparatus is often modified during viral infection. As a part of the translational apparatus, the translation initiation factor eif4e is frequently targeted for modification. For instance, during infection with VSV, Influenza A virus, picornaviruses such as poliovirus and EMCV, eif4e is dephosphorylated. The consequences of this alteration are not well understood. The following chapter ascribes a biological function to this alteration. 77

105 Chapter 3.: Eukaryotic Translation Initiation Factor eif4e Phosphorylation is Inhibitory to Interferon Type I Response. Barbara Herdy, Yuri V. Svitkin, Amy Beth Rosenfeld, Liwei Rong, Luc Furic, Ryan J. O. Dowling, Annie Silvestre and Nahum Sonenberg 78

106 Abstract Organisms continuously encounter pathogens and depend upon the innate immune system to ward off infections. One component of the innate immune response is the production of type I interferon (IFN- and IFN- ). Their production is controlled both at the level of transcription and translation. In most cases translational control occurs at the initiation step, which entails binding of the eukaryotic translation initiation factor eif4e to the 5' cap structure of mrnas. The activity of eif4e is regulated by phosphorylation on serine 209 by Mnk 1 and 2 kinases, which are activated by p38/mapk and ERK signaling and by the translational repressor proteins, the eif4e binding proteins (4E-BPs). Here, we report that phosphorylation of eif4e on serine 209 regulates type-i IFN production. We previously generated an eif4e knock-in (KI) mouse, in which serine 209 was replaced with alanine. Mouse embryonic fibroblasts (MEFs) from both, wild type (WT) and KI mice, infected with vesicular stomatitis virus (VSV), Sindbis virus, encephalomyocarditis (EMCV) and Influenza A virus show less susceptibility to these RNA viruses. Similarly, MEFs lacking both Mnk kinases were less susceptible to VSV infection when compared to WT MEFs. This protective phenotype was ascribed to elevated type I interferon production in the KI MEFs. These data suggest that eif4e phosphorylation delays efficient production of type I interferon. Introduction Cells encounter a variety of pathogens including viruses. The first line of defense against pathogenic invasion is the innate immunity, which is a highly conserved, non specific defense mechanism (Katze et al. 2002; Kumar et al. 2009b). The type I interferon response is one component of the innate immune system. The rapid synthesis and secretion of type I interferons, interferon- (IFN- ) and interferon- (IFN- ), is central for a potent antiviral and immune modulatory response which ultimately leads to protection of host cells from the pathogen (Kawai and Akira 2006; Kawai and Akira 2007a). The cell regulates the type I interferon response through a well controlled signaling network feeding into both transcriptional and post transcriptional mechanisms. Recently, regulators of translational control mechanisms have presented themselves as potential modulators of type I interferon production. Several laboratories, including ours, 79

107 have demonstrated the role of translational control during type I interferon production (Lekmine et al. 2003; Kaur et al. 2007; Colina et al. 2008). Translational control enables the cell to rapidly react to changes in both the extracellular and intracellular environment. To avoid the production of truncated proteins in the cytoplasm, cells regulate protein synthesis mainly at the initiation step. Formation of the eif4f-mrna complex is considered as the rate limiting step of translation initiation. This complex connects the 5 end of capped mrnas to the ribosome. The eif4f complex consists of eif4e, the protein directly interacting with the cap structure on mrnas, eif4g, a scaffolding protein and eif4a, a DEAD-box RNA helicase (Gingras et al. 1999b). The eif4g subunit interacts with eif3 and thereby establishes the critical link between the mrna and the 40S ribosomal subunit. eif4e is the least abundant translation initiation factor (Duncan 1987; Gingras et al. 1999b) and its availability limits cap-dependent translation initiation. The activity of eif4e is highly regulated. Two regulatory mechanisms control eif4e function. Firstly, a family of small translational repressors, the eif4e binding proteins (4E-BPs), bind to eif4e competing with eif4g for eif4e binding. This inhibits eif4f complex formation (Pause et al. 1994a; Gingras et al. 1999b). Secondly, eif4e is controlled by phosphorylation on serine 209 (Flynn and Proud 1995; Joshi et al. 1995). Phosphorylation of eif4e is increased upon mitogenic stimuli and cytokines in vivo (Roux and Blenis 2004). Mitogens induce pathways which lead to the activation of p38 and ERK kinases (Morley 1997; Kleijn et al. 1998; Wang et al. 1998) resulting in activation of the eif4e kinases Mnk 1 and 2 (Fukunaga and Hunter 1997; Waskiewicz et al. 1997; Pyronnet et al. 1999; Waskiewicz et al. 1999). Mnk 1 and 2 bind to the C- terminus of eif4g and mediate the phosphorylation of eif4e (Pyronnet et al. 1999; Waskiewicz et al. 1999). Despite extensive in vitro studies, it is not clear how eif4e phosphorylation influences translation initiation. However, several recent in vivo studies showed that the overexpression of wild type eif4e was able to promote tumor development and progression, while overexpression of eif4e S209A mutant was not (Topisirovic et al. 2004; Wendel et al. 2004). Additionally, phosphorylation of eif4e is altered during the infection with several viruses (Kleijn et al. 1996; Connor and Lyles 2002; Walsh and 80

108 Mohr 2004), implying a role of this modification during viral replication. Based on these reports we wished to further understand how eif4e phosphorylation affects viral replication. We found that in mouse embryonic fibroblasts (MEFs), in which serine 209 on eif4e was replaced by an alanine, replication of several RNA viruses was repressed. This phenotype was based on a lower threshold for production of type I interferon in MEFs lacking the eif4e phosphorylation site. Hence, we propose that eif4e phosphorylation impedes type I interferon expression and thereby is crucial for the establishment of an interferon based antiviral response in MEFs. Results Viral infection dramatically alters the cell. Intracellular structures such as the endoplasmic reticulum are modified as well as metabolic pathways. These changes favor viral replication. No viral genome encodes all the components of the translational machinery; therefore proteins of the translational apparatus are altered to fit the needs for viral replication. Changes in eif4e phosphorylation is one alteration that is caused by infection with several viruses (Kleijn et al. 1996; Connor and Lyles 2002; Walsh and Mohr 2004). To further understand how eif4e phosphorylation regulates the cellular response to viral infection we studied the impact of the serine 209 to alanine alteration in eif4e on the replication of various viruses. VSV replication is delayed in eif4e knock-in (KI) MEFs. Mouse embryonic fibroblasts (MEFs) from transgenic mice carrying a serine to alanine alteration at position 209 in the eif4e (knock-in, KI) were isolated and infected with several viruses. This transgenic mouse line was previously established in our laboratory (Furic et al., submitted). Wild type (WT) and KI MEFs were infected with vesicular stomatitis virus (VSV), a negative strand RNA virus (rhabdovirus), at a multiplicity of infection (MOI) of 1 plaque 81

109 Figure 19. Alteration of the eif4e phosphorylation site serine 209 by alanine delays VSV replication. Wild type (WT) and eif4e knock-in (KI) MEFs were mock-infected or infected with VSV at MOI of 1 PFU/cell. A) At the indicated time points MEFs were pulse labeled with [ 35 S]methionine for 30 minutes. Cells were lysed in 1x Laemmli sample buffer and subjected to 15% SDS-PAGE. After transfer to PVDF membrane audioradiograms were taken. Viral proteins are indicated on the left. B) The membrane was blotted for -actin, eif4e, phosphorylated eif4e and VSV proteins. Cytopathic effect (C) and virus titer (D) were determined 10 hours post infection (h.p.i). 82

110 A h.p.i mock VSV WT KI C WT KI G N/P B h.p.i M mock VSV G N/P WT KI D VSV proteins PFU/ml WT KI β-actin eif4e S209 eif4e-p Figure 19

111 forming unit per cell (PFU/cell) (Fig. 19). Stunningly, metabolic [ 35 S]methionine pulse labeling revealed production of viral proteins after 6 hours post infection (h.p.i) in WT cells, while in KI MEFs no viral proteins were observed (Fig. 19.A). Western blot analysis confirmed this observation. Viral proteins were synthesized earlier in WT cells than in KI cells. Only the N/P protein was observed 10 h.p.i in the KI MEFs (Fig. 19.B). Levels for eif4e and -actin remained unchanged throughout the infection. As reported previously, eif4e was partially dephosphorylated late during VSV infection (compare eif4e phosphorylation levels 8 h.p.i with 10 h.p.i). Virus induced cell death, cytopathic effect (CPE), was seen 10 h.p.i. in WT cells while KI MEFs were barely damaged (Fig. 19.C). Plaque assays performed on KI and WT MEFs showed that the inability to phosphorylate eif4e resulted in an approximately 3-log decrease in viral yield. These data demonstrate that VSV infection was reduced in MEFs expressing eif4e in which serine 209 was replaced by an alanine. Furthermore, this phenotype is independent of the genetic background as a reduction of VSV replication was also observed in KI cells of C57BL/6 background (Fig. 20.A) and Balb/c background (Fig. 20.B) when compared to WT littermate control cells. Mnk double knock out (DKO) MEFs are less susceptible to VSV infection. The Mnk kinases regulate eif4e phosphorylation in response to changes in the environment of the cell. These proteins are the only known eif4e kinases (Ueda et al. 2004). To further understand how eif4e phosphorylation influences the outcome of viral infection, MEFs from WT mice and MEFs isolated from mice lacking both Mnk kinase genes (double knock out, DKO) were infected with VSV at various MOIs for 17 hours (Fig. 21). [ 35 S]methionine metabolic labeling and western blot analysis of infected WT and DKO MEFs clearly revealed a similar phenotype as described for the KI cells (Fig. 21. A, B). WT MEFs were more susceptible to VSV infection than the DKO MEFs. Western blot analysis detected only faint viral bands in the DKO MEFs, while the WT cells showed pronounced production of viral proteins when infected with MOI 1 PFU/cell (Fig. 21.C). A pronounced cytopathic effect was detected in the WT cells 17 h.p.i while DKO cells 83

112 Figure 20. KI MEFs of C57BL/6 background and Balb/c background revealed reduced efficiency of VSV infection. WT and KI MEFs of two different backgrounds were isolated from littermate mice embryos. WT and KI MEFs of C57BL/6 background (A) and Balb/c background (B) were infected at the indicated MOI for 17 hours. Western blot analysis was performed as specified for Fig

113 A WT KI MOI VSV VSV proteins β-actin eif4e S209 eif4e-p B MOI VSV 0 WT KI VSV proteins S209 eif4e-p eif4e β-actin Figure 20

114 Figure 21. MEFs deficient of the eif4e kinases Mnk 1 and 2 showed decreased VSV replication. Wild type (WT) and Mnk ½ double-knock out (DKO) MEFs were mockinfected or infected with VSV at the indicated MOIs. Seventeen hours post infection the cells were labeled with [ 35 S]methionine for 30 minutes (A). Cells were lysed in 1x Laemmli sample buffer and subjected to 15% SDS-PAGE. After transfer to PVDF membrane audioradiograms were taken. Viral proteins are indicated on the left. B) The membrane was blotted for -actin, eif4e, phosphorylated eif4e and VSV proteins, as indicated. Cytopathic effects are also shown (C). 85

115 A MOI VSV G 0 WT ,1 0,01 0,001 0 DKO Mnk ,1 0,01 0,001 B WT DKO Mnk N/P B M C G N/P 0 WT ,1 0,01 0,001 0 DKO Mnk ,1 0,01 0,001 VSV proteins β-actin S209 eif4e-p Figure 21

116 remained healthy (Fig. 21.B). This result strongly supports the previous data that dephosphorylating eif4e impedes VSV replication. Treatment with the pharmacological inhibitor of the Mnk kinases renders WT MEFs more resistant to VSV infection. Reducing or stimulating phosphorylation of eif4e and studding the effects on translation has been the subject of many researchers. A pharmalogical inhibitor of the Mnk kinases, CGP57380, is commercially available and has been used in studies addressing the role of eif4e phosphorylation in cap-dependent and independent translation as well as during viral infection (Walsh and Mohr 2004; Buxade et al. 2008; Walsh et al. 2008). To gain more evidence that dephosphorylating eif4e reduces the efficiency of VSV infection, WT and KI MEFs were grown in the presence of 10 μm CGP57380 for 12 hours. The cells were infected with VSV at an MOI of 0,001 PFU/ml for 17 hours in the presence of the Mnk inhibitor. Western blot and plaque assays were used to assess the effect of CPG57380 on VSV replication (Fig. 22). Western blot analysis clearly illustrates that in the presence of the Mnk inhibitor VSV protein production is reduced in WT cells while untreated or WT cells treated with the carrier, DMSO, were not able to limit viral infection (Fig. 22.A). As previously shown, eif4e was dephosphorylated in the presence of the inhibitor (Tschopp et al. 2000; Knauf et al. 2001). Plaque assay analysis revealed a 17-fold reduction of virus replication when WT MEFs were treated with the inhibitor as compared to untreated WT cells (Fig. 22.B). A less dramatic effect was observed in the KI cells which were grown and infected in the presence of the drug. These observations confirmed the previously presented data that the absence of eif4e phosphorylation retards VSV replication. However, when the Mnk inhibitor CGP57380 was added to the KI cells during infection additional inhibition of VSV replication was detected. The reason for this phenotype is unknown although this may be the result of the dephosphorylation of other Mnk targets. Replication of encephalomyocarditis virus (EMCV) and sindbis virus (SV) is also delayed in eif4e knock-in (KI) MEFs. Next, we wanted to determine if the absence of eif4e phosphorylation would protect cells from other viruses, 86

117 Figure 22. The Mnk inhibitor CGP57380 protects WT cells and KI cells from VSV infection. WT and KI MEFs were mock treated or treated with the 10 μm Mnk inhibitor for 12 hours. Post treatment WT and KI cells were infected with VSV at an MOI of PFU/cell for 17 hours in the presence of the inhibitor. (A) Samples were collected and subjected to SDS-PAGE and western blot analysis. The membrane was probed for actin, VSV proteins, total eif4e levels and phosphorylated eif4e levels. (B) Cytopathic effect and plaque assay analysis are shown. 87

118 x10 4 PFU/ml x10 4 PFU/ml x10 6 PFU/ml x10 6 PFU/ml A WT MEFs KI MEFs VSV DMSO Mnk Inh VSV proteins S209 eif4e-p eif4e B mock VSV VSV DMSO Mnk Inh VSV WT mock VSV KI 100 VSV DMSO Mnk Inh VSV Figure 22

119 which would suggest that this phenotype is an ubiquitous antiviral mechanism. We challenged the cells with encephalomyocarditis virus (EMCV), a positive strand RNA virus of the picornavirus family, and with sindbis virus (SV), an alphavirus also containing a positive strand RNA genome (Fig. 23). WT and KI MEFs were infected at various MOIs for 17 hours. Metabolic [ 35 S] pulse labeling was performed to visualize host cell shut off and dominant viral proteins. Again, the KI cells were less susceptible to infection with either virus. At an MOI of 0.01 PFU/cell the autoradiograph clearly shows that host cell shut off and the abundance of EMCV proteins was more pronounced in the WT cells while hardly any viral proteins were detected in the KI cells (Fig. 23.B). WT and KI MEFs challenged with SV showed similar results. Host cell shut off and viral proteins are detected at an MOI of 0.1 PFU/cell after 17 h.p.i. Hardly any viral proteins could be detected in the KI cells at this point (Fig. 23. A) Collectively, these results strongly suggest that infection of eif4e S209A mutant MEFs delays replication of different RNA viruses, irrespective of whether they use capdependent (VSV, Sindbis Virus) or IRES mediated translation initiation (EMCV). Type I Interferon production is upregulated in the eif4e KI cells. The replication of VSV, SV and EMCV has been reported to be highly sensitive to secretion of type I interferon (Belkowski and Sen 1987; Despres et al. 1995). One potential explanation for the protection of the KI cells from viral infection could be that a factor regulating the type I interferon system is altered resulting in an increased type I interferon secretion. To test this hypothesis, WT and KI MEFs were stimulated for 6 hours with poly(i:c), a synthetic dsrna, which is known to be a potent inducer of the type I interferon response. The presence of antiviral cytokines in the media from stimulated cells was tested by transferring the media to unstimulated WT cells and after 24 hours infecting these cells with VSV. Supernatants containing type I interferon 88

120 Figure 23. EMCV and SV infection is reduced in KI MEFs. WT and KI MEFs were infected with EMCV or SV at indicated MOI of PFU/cell. After 17 hours post infection the cells were processed as described for Fig. 19. Western blot analysis was conducted for infection with both viruses. The membranes were probed with antibodies directed against eif4e, phosphorylated eif4e and -actin. An autoradiographs and western blot analysis of WT and KI cells infected with the SV (A) and EMCV (B) are presented. Viral proteins are indicated on the left. 89

121 A SV MOI 0 WT KI B EMCV MOI WT KI pe2 E1 1P 3CD 3D C 1AB 2C 1D 1C 3C S209 eif4e-p eif4e β-actin S209 eif4e-p eif4e β-actin Figure 23

122 Figure 24. KI MEFs show increased type I interferon production after stimulation with poly(i:c). A) Experimental set up diagram. B) WT and KI MEFs were treated with indicated concentrations of poly(i:c) for 6 hours. Cultured media were transferred to WT cells and incubated over night. On the following day overlaid WT cells were infected with VSV at an MOI of 0.1 PFU/cell for 12 hours. CPE and virus titres are shown. C) Total RNA was isolated from poly(i:c) treated WT and KI MEFs and analyzed by RT- PCR for IFN-, MDA5, RIG-I, IRF7 and -actin. (D) IFN- levels in cell supernatants as determined by ELISA. 90

123 PFU/ml (10 6 ) PFU/ml (10 6 ) PFU/ml (10 6 ) PFU/ml (10 6 ) A pi:c 6hrs RNA RT-PCR IFN beta IRF7 B WT SN KI SN wt Supernatants ki Supernatants ELISA INF alpha INF beta 0 μg/ml p(i:c) wt VSV MOI 1 24h wt 0.5 μg/ml p(i:c) Plaque Assay C μg/ml p(i:c) WT KI D wt IFN-β 50 ELISA IFN-β IRF7 pg/ml RIG I 10 0 MDA 5 μg/ml p(i:c) β-actin Figure 24

124 are supposed to protect the cells from infection (Scheme Fig. 24.A). Naive WT cells treated with supernatants from unstimulated WT and KI cells were not protected from VSV infection. When WT cells were treated with 0.5 μg/ml p(i:c) and the supernatants transferred to naïve cells, the titer of VSV was reduced by half. Treatment of KI MEFs with the same dose of p(i:c) however, resulted in a complete block of virus infection in the overlaid cells (Fig. 24.B). These data suggest that when stimulated with p(i:c) a soluble factor is secreted into the supernatants of both WT and KI cells which is able to protect naive WT cells from viral infection. However, a much higher concentration of this factor is secreted from the KI cells. This factor is believed to be interferon. To further verify the presence of type I interferon in the media of stimulated WT and KI cells an ELISA for IFN- was performed (Fig. 24.D). Accordingly, three times more IFN- was secreted from stimulated KI cells when compared to WT cells. Total RNA was isolated to test for transcriptional upregulation of IFN- mrna (Fig. 24.C) (Colina et al. 2008). Reverse transcriptase-polymerase chain reaction (RT-PCR) for IFN- revealed that KI cells induce transcription of IFN- mrna after treatment with 0.25 μg/ml p(i:c). Stunningly, twice as much p(i:c) was needed to reach the same IFN- mrna levels in WT cells. The transcriptional upregulation of upstream components regulating IFN- mrna levels was also examined. The mrna levels of the cyoplasmatic RNA sensors MDA5 and RIG-I (Kato et al. 2006) as well as the master regulator of type I interferon, the transcription factor IRF7 (Honda et al. 2005b), which are all interferon inducible genes themselves, were analyzed. RT-PCR revealed no difference in the mrna levels of RIG-I while MDA-5 and IRF7 mrna was significantly upregulated in the KI cells. These data demonstrate that in KI MEFs the threshold for type I interferon production is significantly lowered. To eliminate the possibility that secretion of other components of the innate immune system than type I interferon are responsible for the antiviral effect observed in the KI cells, neutralizing antibodies against IFN- were added during infection with VSV (Fig. 25). WT and KI cells were incubated for 48 hours with neutralizing antibodies against IFN- or IFN- or in the presence of antibodies against both cytokines and infected with VSV at a MOI of 0.1 PFU/cell for 12 hours. Western blot analysis for viral proteins clearly illustrates that addition of the 91

125 Figure 25. Neutralizing antibodies against IFN- render KI cells more susceptible to VSV infection. MEFs were mock-treated or treated with anti IFN- antibodies or with anti IFN- and anti IFN- antibodies two days prior to infection. Medium was changed daily. Cells were infected with VSV at MOI of 0.1 PFU/cell in the presence of antibodies for 12 hours. MEFs were lysed and subjected to western blot analysis, probing for viral proteins, eif4e, phosphorylated eif4e and -actin (A). Plaque assays were performed (B). Titres are illustrated in (C). 92

126 PFU/ml x A anti-ifnα anti-ifnβ VSV MOI KI WT VSV proteins B S209 eif4e-p eif4e 0 anti-ifnα - anti-ifnβ - WT KI C WT - anti-ifnα anti-ifnβ KI dilutions Figure 25

127 neutralizing antibodies against IFN- rendered KI cells more susceptible to VSV infection (Fig. 25.A). Virus titres were determined to confirm the results from western blot analysis. A 5-fold increase in viral replication was detected when antibodies against IFN- were added to the infected cells. Although this confirmed that KI MEFs produce and secrete higher IFN- levels, the viral yield observed in the WT cells could not be fully restored. This rises the possibility that eif4e phosphorylation may modulate other factors of the innate immune system. The interferon signaling cascade depends on two phases. The initial phase is sensing the pathogen which results in stimulation of IFN- production and secretion. The second phase depends on the activation of the type I interferon receptor (IFNAR) by the secreted IFN-, which results in the production of interferon stimulated genes (ISGs) and more IFN-. To determine if eif4e phosphorylation also modulates the second phase of the type I interferon cascade WT and KI MEFs were incubated with indicated units of recombinant mouse IFN / for 12 hours and infected with a mutant of VSV, which is producing GFP, at an MOI of 0.1 PFU/cell for 48 hours (Fig. 26). KI cells were fully protected from the viral infection at 50 U/ml of recombinant interferon, while 100 U/ml were needed to protect the WT cells (Fig. 26.A). MTT assay was used to quantify the cytopathic effect comparing WT and KI MEFs (Fig. 26.B). KI cells were fully protected at 50 U/ml of recombinant interferon while WT cells were protected at 100 U/ml. This result demonstrates that the phosphorylation status of eif4e also modulates the events downstream of the interferon receptor. Taken together, removal of the eif4e phosphorylation site reduces the threshold for the establishment of an antiviral state and therefore protection is acquired upon fewer stimuli in the KI MEFs. Discussion Regulation of translation allows a cell to adjust rapidly to extracellular and intracellular changes. The cell controls translation mainly at the level of translation initiation. This regulation is challenged by viral infection. Dramatic changes can be inflicted upon the translation apparatus but they must be subtle enough to keep the cell viable until the viral replication cycle is completed. Often a shut off of cellular protein production is induced, 93

128 Figure 26. Less recombinant interferon / is needed to protect KI cells against virus infection than WT MEFs. WT and KI cells were incubated with indicated units of recombinant IFN- /. Cells were infected with VSV-GFP at a MOI of 0.1 PFU/cell for 48 hours (A). Fluorescent images are shown detecting VSV-GFP infected cells. Below, ligthmicroscope images of the same VSV-GFP infected cells are shown (B). The same samples were submitted to a MTT assay to measure cell viability. The percentage of viable cells is shown (C). 94

129 A NT 25 U/ml 50 U/ml 100 U/ml 150 U/ml VSV-GFP WT VSV-GFP KI B NT 50 U/ml 100 U/ml 150 U/ml 25 U/ml WT KI % cell survival NT 25 U 50 U 100 U 150 U WT KI Figure 26