26 27 D I A M O N D L I G H T S O U R C E A N N U A L R E V I E W 2 0 1 9 / 2 0 D I A M O N D L I G H T S O U R C E A N N U A L R E V I E W 2 0 1 9 / 2 0 Structural studies offer details of flu virus replication Related publication: Fan H.,Walker A. P., Carrique L., Keown J. R., Serna Martin I., Karia D., Sharps J., Hengrung N., Pardon E., Steyaert J., Grimes J. M. & Fodor E. Structures of influenza A virus RNA polymerase offer insight into viral genome replication. Nature 573 , 287-290 (2019). DOI: doi.org/10.1038/s41586-019-1530-7 Publication keywords: Influenza virus; RNA polymerase; Virus replication I nfluenza (commonly known as the flu) is an acute respiratory infection responsible for over threemillion cases of severe illness each year. In order to stay one step ahead of flu outbreaks, scientists are trying to decipher the structure andmechanisms of influenza A viruses. When the virus infects host cells, it makes many copies of its RNA genome to produce new virus particles. The enzyme responsible for this genome replication is called RNA-dependent RNA polymerase, or FluPolA. A teamof researchers fromthe University of Oxford used Cryogenic ElectronMicroscopy (cryo-EM) imaging at the Electron Bio-Imaging Centre (eBIC) to analyse the influenza virus polymerase. They also used the Macromolecular Crystallography (MX) beamlines I03, I04, I24 to analyse crystals of the polymerase. Using X-ray crystallography, they were able to obtain the first, high-resolution structures of FluPolA of human and avian influenza viruses, revealing crucial details of how the structure of FluPolA is vital in initiating RNA synthesis. Using cryo-EM allowed them to study the protein binding to an RNA template. Their results suggest a target by which virus reproduction might be inhibited. In the future, this could lead to the development of antiviral drugs against the flu. This is a compelling example of integrative structural biology, with scientists using complementary techniques (including nanobody technology from Instruct-BE) to gain new insights into the replicationmechanisms of FluPolA. Influenza is an acute respiratory infection, caused by influenza A viruses, whichareresponsibleforseasonalpandemics. Themajorhostspeciesofinfluenza A virus are aquatic birds, but occasionally the virus can cause zoonotic infections in humans and other animals 1 . All viruses need to infect and replicate within cells for their continued existence. This involves replicating new copies of the viral genome, which for influenza is negative sense single-stranded ribonucleic acid. Like other similar viruses, influenza carries its own polymerase (FluPol), which catalyses the synthesis of this new viral RNA. The molecular mechanisms by which FluPol replicates the viral RNA (vRNA) through a complementary RNA (cRNA) intermediate remain largely unknown. In addition, FluPol is multi- functional and not only replicates the viral genome but transcribes capped viral messenger RNA (mRNA). It does this by binding to cellular PolII and cleaving off the cap structure from newly synthesised cellular mRNA, which it uses to prime transcription. How the polymerase switches between transcription and genome replication is not well understood and is an area of active study. As such, FluPol lies at the heart of the influenza virus lifecycle and has been the focus of a prolonged research effort to understand its structure and mechanisms. Ultimately, this may facilitate the development of antiviral compounds that target the polymerase. FluPol comprises a complex of three viral proteins: PA, PB1 and PB2, known together as a heterotrimer. PB1 is the RNA polymerase and has the canonical right-hand-likefold,withsubdomainsknownasfingers,palmandthumb. PB2 is involved in the binding of the cap of cellular mRNA, and PA has an endonuclease domain that cleaves the cap structure away from the cellular RNA. The FluPol trimer binds the conserved 5’and 3’ends of the viral RNA segment, known as the vRNA promotor, which form a partially dsRNA panhandle structure. Previous structural work on FluPol from different influenza viruses had revealedthatthemolecule ishighlydynamicandcantakeupstrikinglydifferent poses 2,3 . Though the fold of each of the domains within the FluPol trimer is Biological Cryo-Imaging Group eBIC, beamlines I03, I04 and I24 identical, the arrangement of domains within PB2 and PA are dramatically altered between the unbound (apo) formand promotor RNA-bound complexes. This is thought to reflect the balance between transcription and replication. Using X-ray crystallography, this paper describes the first, high-resolution structures of FluPol of human and avian influenza viruses. Intriguingly the structures showed that the heterotrimeric FluPol forms dimers in the crystals. FluPol was then solved by Single Particle Analysis (SPA) cryo-EM, using the microscopes at eBIC, which confirmed that this dimer interface of FluPol was retained in solution and revealed that the 5’-end of the RNA template was bound to FluPol in a hook conformation in an RNA binding pocket, whilst the 3’-end was disordered. To determine whether the dimeric FluPol structure was important to enzyme activity, mutations were introduced to destabilise the dimer interface. The viral transcription and replication by FluPol mutants were measured using a mini-replicon assay, showing that the dimeric structure of FluPolwas important for the initiation of vRNA synthesis from the cRNA template. It was clear from analysis of the protein in solution that FluPol is in dynamic equilibriumbetweenmonomeric and dimeric forms. To gain further insight into FluPol dimerisation and cRNA binding, a nanobody was used that reduces FluPol dimerisation. The use of nanobodies in structural analyses is well established but, in this work, they were of great use in probing function. Again, SPA cryo- EM was used to solve the structures of monomeric and dimeric FluPol bound to the nanobody and cRNA promoter. The structure of FluPol dimers with and without nanobody are essentially identical, with the same ordered 5’ and dynamic 3’cRNA. However, the monomeric FluPol bound to nanobody revealed the binding of both 5’ and 3’ cRNA. Intriguingly the 3’ cRNA bound at a site previously unobserved and in a groove close to the dimer interface and could represent a parking site for 3’RNA during replication and transcription. Further evidence comes from the observation that RNA binding is seen in a similar site of the RNA polymerase of La Crosse orthobunyavirus 4 . By comparing the structures of monomeric and dimeric FluPol-nanobody complexesitwasfoundthatdimerisationinducesamovementofahelicalbundle that is formed by the thumb subdomain of PB1 and the N1 subdomain of PB2. This movement results in an opening of the binding site for the 3' cRNA, which explains the absence of 3' cRNA at this site in the dimeric structure. Furthermore, dimerisation leads to rearrangements in the polymerase active site, specifically the retraction of the priming loop (part of the thumb subdomain) and could destabilise binding of the 3' cRNA in the active site. To show that the nanobody affected the function of FluPol, a cellular mini- replicon assay was used with co-expression of the nanobody, which showed the severe inhibition of replication and transcription, whereas another nanobody that does not affect FluPol dimerisation had no significant effect. Furthermore, viral infectivity assays with the nanobody caused a significant reduction in virus titre. In this paper, using an integrative structural and functional biology approach, itwasshownthatdimerisationofFluPol isrequiredforthe initiationof vRNA synthesis from the cRNA viral template. The dependency on dimerisation for replication of vRNA from cRNA template is consistent with previous observations that vRNA synthesis requires a trans-activating polymerase 5 . It is interesting to note that a requirement for trans-activation through polymerase dimerisation provides amechanism for tuning the amount of vRNA synthesised, where vRNA production is initiated only when a sufficient level of newly made free polymerase is available in the cell. This would help ensure that the virus does not trigger an antiviral response through recognition by pathogen- recognition receptors by producing vRNA that cannot be assembled into viral ribonucleoprotein complexes (vRNPs). References: 1. Mostafa A. et al. Zoonotic potential of influenza A viruses: a comprehensive overview. Viruses 10, 497 (2018). DOI: 10.3390/v10090497 2. Hengrung N. et al. Crystal structure of the RNA-dependent RNA polymerase from influenza C virus. Nature 527 , 114-117 (2015). DOI: 10.1038/nature15525 3. Pflug A et al. Structure of influenza A polymerase bound to the viral RNA promoter. Nature 516 , 355-360 (2014). DOI: 10.1038/nature14008 4. Gerlach P. et al. Structural insights into bunyavirus replication and its regulation by the vRNA promoter. Cell 161 , 1267-1279 (2015). DOI: 10.1016/j.cell.2015.05.006 5. York A. et al. Isolation and characterization of the positive-sense replicative intermediate of a negative-strand RNA virus. P. Natl. Acad. Sci. 110 , E4238–45 (2013). DOI: 10.1073/pnas.1315068110 Funding acknowledgement: This work was supported by the Medical Research Council, theWellcome Trust and by Instruct-ERIC, part of the European Strategy Forum on Research Infrastructures (ESFRI). Computation used the Oxford Biomedical Research Computing (BMRC) facility, a joint development between theWellcome Centre for Human Genetics and the Big Data Institute, supported by Health Data Research UK and the NIHR Oxford Biomedical Research Centre. Corresponding author: Dr Jonathan Grimes, University of Oxford,
[email protected] Figure 1: Single particle cryo-EM analysis of monomeric and dimeric cRNA-bound human H3N2 FluPolA heterotrimer in complex with Nb8205. (a) Representative micrograph of cRNA-bound FluPolA in complex with Nb8205, embedded in vitreous ice. (b) Representative 2D class averages. (c) Forward Scatter Curves (FCS) for the 3D reconstruction using gold-standard refinement in RELION, indicating an overall map resolution of 3.79 A and 4.15 A for the monomeric and dimeric FluPolA form, respectively, and the model-to-map FSC. Curves are shown for phase-randomisation, unmasked, masked and phaserandomisation- corrected masked maps. (d) and (f) The 3D reconstructions, locally filtered and coloured according to RELION local resolution, for the dimeric (d) and monomeric (f) form. (e) and (g) Angular distribution of particle projections for the dimeric (e) and monomeric (g) form, with the cryo-EMmap shown in grey. Figure 2: Structures of (a) dimeric and (b) monomeric cRNA-bound human H3N2 FluPol heterotrimer in complex with nanobody solved by single particle cryo-EM analysis. In the monomeric structure 3’ cRNA (coloured yellow) binds in a narrow groove, that in the dimeric form of FluPol is opened up and is not able to accommodate 3’ cRNA. The structures in panels (a) and (b) are in the same orientation but are not drawn to scale.