Cusack Group

Structural biology of RNA-protein complexes in gene expression and host-pathogen interactions

The Cusack group uses X-ray crystallography and cryo electron-microscopy (cryoEM) to study the structural biology of protein-RNA complexes involved in RNA virus replication, innate immunity and cellular RNA metabolism.


Background and previous research

Our goal is to understand the molecular mechanisms whereby the genomic RNA of influenza-like viruses is, on the one hand, the template for transcription and replication of the viral genome by its RNA-dependent RNA polymerase and, on the other hand, an Achilles’ heel, whose recognition as non-self can trigger an innate immune response to counter the viral infection. Molecular warfare between the virus and the host-cell occurs at many levels. Influenza polymerase has a unique mechanism of transcription priming called ‘cap-snatching’, which involves pirating short-capped oligomers from nascent cellular Pol II transcripts; this contributes to shutdown of host cell gene expression. The cell counters RNA viruses with innate immune pattern-recognition receptors, such as the RNA helicase RIG-I, which recognise particular viral RNA structural motifs (e.g. 5′ triphosphate-dsRNA) as non-self, thus activating a signalling pathway leading to interferon production and establishment of the anti-viral state. In response, viruses deploy proteins as counter-counter-measures to dampen the immune response, for instance, by supressing the RIG-I signalling pathway.

In 2014, we determined the first crystal structures of the complete heterotrimeric influenza polymerase (Pflug et al., Nature 2014) and proposed a mechanism of how cap-snatching is performed (Reich et al., Nature 2014). We have also elucidated how influenza polymerase interacts with the phosphorylated C-terminal domain (CTD) of Pol II during cap-snatching (Lukarska et al., Nature 2017). In 2011 we published the first structure-based mechanism of the activation of RIG-I, showing how RNA binding resulted in a major conformational change that liberated the N-terminal CARD domains for downstream signalling (Kowalinski et al., Cell 2011). Subsequently we determined structures of related innate immune receptors MDA5 and LGP2 (Uchikawa et al., Mol. Cell 2016). Previously, we worked on aminoacyl-tRNA synthetases, which play an essential role in protein synthesis by charging specifically their cognate tRNA(s) with the correct amino acid and editing mischarged amino acids if necessary. Our work led to the understanding of the mechanism of action of a new anti-fungal compound targeting the editing activity of leucyl-tRNA synthetase (Rock et al., Science 2007), and to the design of new antibiotics that target multi-resistant gram negative bacteria, tuberculosis and apicomplexan parasites (Palencia et al., Antimicrob. Agents Chemother. 2016).

Future projects and goals

Our current goal is to derive models explaining the detailed mechanisms of transcription and replication of the viral genome (vRNA) by influenza-like viral polymerases. To achieve this we use X-ray crystallography and single-particle cryoEM to determine structures after trapping successive states along the active transcription or replication pathways. For transcription by influenza polymerase, we have recently determined a series of high-resolution structures corresponding to the transcription initiation, elongation and poly-adenylation/termination and recycling states (Kouba et al., NSMB 2019, Wandzik et al., Cell 2020). In parallel, we are doing the same for viral replication, which is unprimed and occurs in two-steps via an intermediate complementary RNA (cRNA). These studies are being extended to viral RNPs (the physiological RNA synthesis units) to understand the behaviour of the viral nucleoprotein during replication and transcription and to include host factors important for viral replication. We complement structural studies with in vitro polymerase enzymology and in-cell studies using mini-replicon systems, and, in collaborations, recombinant viruses and live-cell imaging.

There are several other aspects of these projects of particular interest.

  • Our structural work on influenza polymerase has opened up the area of structure-based drug design of novel anti-virals targeting multiple functional sites on the polymerase. In collaboration with various pharmaceutical companies, we continue to work on cap-binding, endonuclease (Omoto et al. Sci Rep. 2018) and RNA synthesis inhibitors (e.g. nucleoside analogues). This work is now being extended to targeting the SARS-CoV-2 polymerase.
  • We have extended our work on viral polymerases to those of other segmented negative-strand RNA viruses such as the large order of Bunyavirales. These cytoplasmically-replicating viruses includes several emerging, vector-borne human pathogens e.g. Lassa fever virus, Crimean-Congo hemorrhagic fever virus, La Crosse virus, Severe Fever with Thrombocytopenia Syndrome virus. We have shown that these single-chain polymerases have an architecture similar to the heterotrimeric influenza polymerase (Gerlach et al., Cell 2015, Arragain et al., Nat. Comm. 2020)
  • We have for a long time been studying the nuclear cap-binding complex (CBC), which binds to the 5′ cap of nascent Pol II transcripts and mediates interaction with nuclear RNA processing, transport or degradative machineries (Mazza et al., EMBO J. 2002). One aim of this is to understand how influenza polymerase can compete with CBC for access to nascent, capped RNAs emerging from Pol II. Another is to understand how CBC complexes with proteins such as ARS2 and NCBP3 mediate sorting of Pol II transcripts (Schulze et al., Nat. Comm. 2017, 2018).
  • We have extended our interest in innate immune pattern recognition receptors to the NOD2 signalling pathway that is activated in response to cytoplasmic detection of bacterial wall fragments. Dysregulation of this pathway is implicated in Crohn’s disease. There are parallels bewteen the NOD2 and RIG-I signalling pathways. We have in particular worked on the RIP2 kinase which acts downstream of NOD2 and forms filaments with its CARD domain, like MAVS in the RIG-I pathway (Pelligrini et al., PloS One 2017, Nat. Comm. 2018).
Figure: Model of the activated state of RIG-I with bound dsRNA (centre) and ATP (top-right). The helicase domains (green and cyan), the insertion domain (yellow) and the C-terminal domain (gold) all contribute to RNA binding, which displaces the CARD domains thus allowing downstream signalling and interferon expression.