Targeting a key COVID protein with antivirals

SARS-CoV-2 and humans have distinct strategies for storing genes. The virus has an RNA genome, which is more prone to mutate through errors in replication than the DNA genome of humans. To reduce the risk of amassing mutations, RNA viruses keep their genomes small and compact. While the human genome has  genes interspersed with non-coding sequences, viral RNA genomes fuse genes together end to end, partly as a strategy to economise on space. So-called polyproteins are synthesised from these fused genes and are subsequently cleaved into their individual viral proteins. SARS-CoV-2 encodes an enzyme called the main protease, which serves as the carving knife for polyproteins: It cleaves them into 12 smaller sections to free up the individual proteins starting with the protease itself. Scientists have begun designing drugs that target this protease so that they can block this critical carving process, a necessary step early during virus infection, and disrupt the virus from propagating within cells. This protease has already proven to be an effective drug target, with nirmatrelvir (brand name Paxlovid) approved for use in China, Europe, and the USA since 2022. Now the COVID Moonshot Initiative, a collaboration between Diamond and other institutes, seeks to develop other drugs that inhibit this protease. As such, researchers need to determine how both licensed and potential drugs interact with the protease at a molecular level, which in turn requires an understanding of how the protease matures into an active enzyme. Currently, it is known that multiple steps have to take place during protease maturation: the protease cuts itself out of the polyprotein at two positions, folds into a specific 3D shape, and pairs up to form dimers. However, it is still unclear in what order these events occur.

It was impossible to detect the running order of these events in previous studies as the protease matures too quickly. In hopes of tracking the steps during maturation, Andre Schützer de Godoy and his colleagues used a mutant form of the protease that differs by only one atom and matured at a far slower rate of up to two days. But when they used mass spectroscopy to observe all the states in which the protease exists, they detected every possible combination of clipped and non-clipped proteins, suggesting the cuts, folding, and dimerisation occur in no particular order.

CryoEM at eBIC was used to visualise how two copies of the SARS-Cov-2 main protease pair up to form heart-shaped dimers in their active form (blue). A fragment of a viral polyprotein (red) was detected in the dimer’s active site, allowing the team to study how this protease cleaves proteins. Image Credit: Andre Schützer de Godoy

Having determined that the maturing protease exists in mixed states, the team sought to determine the structure of the complex. Although X-ray crystallography had been used to study this protease in the past, this technique struggles to resolve complexes with unfolded regions in multiple states. Cryo-electron microscopy was chosen as an alternative because different states could be separated and analysed individually with this approach. However, cryoEM posed new challenges because this technique has been optimised for protein complexes larger than the protease. Godoy says,

Using the state-of-the-art facilities at eBIC, the researchers successfully determined the structure of what is now the smallest protein to be resolved by cryoEM at Diamond. This dimer has a molecular weight of approximately 68 kiloDaltons, only 25 kiloDaltons larger than the current world-record for the smallest protein to have its structure solved with this technique. Godoy describes this work as a “landmark study” because although X-ray crystallography has been repeatedly used to study this protease, this is the only cryoEM structure available so far. By averaging together over 200,000 copies of the protein, they generated a refined structure of the protease dimer with a detailed view of its active site. What’s more, they found a protein fragment within the active site in multiple orientations, indicating that they had captured the protease processing its substrate. This allowed them to study how the active site contracts to cut the polyprotein and expands to release the divided products. Understanding the nature of this enzyme activity allowed the team to subsequently decipher the mechanisms by which antiviral drugs block the protease.

Dimerisation of the protease is required for its activity, and the researchers set out to determine how drugs affect this process. They used a technique called size exclusion chromatography coupled with multi-angle light scattering (SEC-MALS) that can measure the size of protease particles to determine whether they were formed from individual copies or were paired-up into dimers. Nirmatrelvir appeared to encourage dimerisation. By capturing how the protease interacts with its substrate using cryoEM, the researchers could model how the drug interacts with the enzyme, revealing that dimerization was triggered by covalent bonding of nirmatrelvir to the active site. Future work is needed to determine if this blockade prevents the protease from processing polyproteins, but this brings the researchers a step closer to understanding how nirmatrelvir staves off the virus. The effects of a second inhibitory compound were tested as well, a drug lead developed by the COVID Moonshot Initiative that Diamond is part of. This compound exerted the opposite effect on the protease: it inhibited dimerisation and forced pre-existing dimers to separate, which indicates the drug prevents the protease from assuming its active state. Godoy cautions that it’s too early to tell if one of these compounds has more clinical potential than the other and more experiments are needed. Moving forward, his team wants to determine the structure of the protease in its singular monomeric state. Godoy says,

Thus, his team are considering other techniques. By using X-ray crystallography at beamline I04-1, Godoy plans to investigate how the COVID Moonshot compound lead interferes with the monomer and prevents them from pairing into dimers. This structural work carried out at Diamond is integral to understanding how drugs interfere with viral proteins at a molecular level.

To find out more about eBIC or discuss potential applications, please contact Principal Electron Microscopist Daniel Clare: daniel.clare@diamond.ac.uk