IRES-mediated initiation of picornavirus translation

Viral infection activates innate immune responses that establish an antiviral state via translational inhibition. Viruses use various strategies to overcome immune responses and to usurp the translation apparatus. One such mechanism is internal ribosomal entry, in which internal ribosomal entry sites (IRESs), structural elements in the 5'UTR, mediate end-independent initiation using only a subset of canonical initiation factors. Viral IRESs form major groups based on common structures and mechanisms of action. Type 1 IRESs occur in picornaviruses, including important human pathogens such as human rhinovirus (HRV), poliovirus (PV) and enterovirus A71. They are ~450nt long and consist of 5 domains (II - VI). A Yn-Xm-AUG motif at their 3′-border is the ribosome attachment site, but the AUG in this motif has poor context, is sequestered in domain VI and is only weakly active. Translation of the viral polyprotein initiates at an AUG codon ~30-160 nt downstream that is reached by scanning. Initiation requires specific interaction of domain V with eIF4G, but domain V alone is not sufficient for IRES function, and interactions that are responsible for ribosomal recruitment of an eIF4G/eIF4A-bound IRES remain unknown. Initiation also requires poly(C) binding protein 2 (PCBP2) that binds to domain IV, and involves glycyl-tRNA synthetase (GARS). However, the mechanism of initiation on Type 1 IRESs remains obscure: the interactions that drive ribosomal recruitment of these IRESs and the mode of action of PCBP2 and GARS are unknown, and how ribosomal complexes unwind domain VI in e.g. HRV IRESs to reach the initiation codon sequestered in it is unresolved. Here, we propose to apply an integrated biochemical/structural approach to elucidate molecular mechanisms of all steps in initiation on Type 1 IRESs. Aim 1 will identify the factor(s) that enables ribosomal complexes to unwind domain VI to gain access to initiation codons sequestered in this domain because canonical factors together with PCBP2 are not sufficient for this process. We will use an activity-based approach to purify the factor(s) required for this step from a permissive HeLa cell extract and will then use recombinant versions to characterize its activity in in vitro reconstituted initiation reactions. We will also silence its expression in cells and use reporter assays and ribosome profiling to confirm its requirement for Type 1 IRES function and possible involvement in translation of cellular mRNAs. Aim 2 will establish the molecular interactions responsible for ribosomal attachment to Type 1 IRESs. It will include (i) determination of the cryo-EM structure of initiation complexes reconstituted from individual purified components on the PV IRES in collaboration with J. Frank (Columbia University) to identify and then to investigate interactions involving essential elements of the IRES for which molecular functions are unknown, (ii) elucidation of the mechanism by which PCBP2 promotes ribosomal attachment, and (iii) clarification of the role of eIF4B in ribosomal attachment. Aim 3 will establish the position of eIF4G/eIF4A bound to domain V in initiation complexes to gain insights into how their binding results in ribosomal entry on the Yn-Xm-AUG motif, and to clarify the role of GARS that also binds to domain V.