Viral entry at the human-animal interface; dissecting the pan-tropic nature of zoonotic viruses.

Anthropogenic shifts in patterns of land use, habitat infringement and climate change increase the probability of viruses such as SARS-CoV-2, the causative agent of Covid-19, spilling over into humans and/or animals. In turn, increased urbanisation and global travel, the absence of herd immunity and poor preparedness can contribute to turn a localised epidemic into a global pandemic. However, not all viruses in nature appear to share the same propensity to spill over; some are restricted to individual hosts while others have a broad host-range and represent a much greater risk to humans and/or animals (livestock, pets and wildlife). Developing a better understanding of the factors that determine the 'zoonotic potential' of viruses is especially prescient as we look to improve pandemic preparedness in a post-Covid-19 landscape.

One of the most important factors to understand in this context is how viral entry (the process whereby the virus attaches to and invades a host cell) correlates with zoonotic potential. In general viruses use specific receptors (proteins or sugars) to enter cells. These can vary between hosts, representing an important point of restriction that directly influences host-range and the potential for spill over. Developing an understanding of this relationship for whole taxonomic groups of viruses (genera or families) will help scientists and stakeholders to assess which viruses represent the greatest risk to humans and animals.

Characterising virus receptor usage and host-range at a broad level is technically challenging. For instance, choosing viruses which accurately represent the overall diversity of their family is prone to bias, favouring established pathogens over those isolated in their natural bat or rodent reservoirs. To address this, we have developed and optimised a pipeline which utilises bioinformatic algorithms to unbiasedly select representative viruses. Within this project we will use this approach to characterise the zoonotic potential of the coronavirus and arenavirus families. All human coronaviruses, as well as many of the coronaviruses which infect our pets and livestock, are thought to have a zoonotic origin (bats or rodents). The same is true for the arenaviruses, with viruses like Lassa continually spilling over from their rodent reservoirs. As proof-of-principle we have already gathered this dataset for the morbillivirus genus, uncovering a number of interesting restrictions which may explain the narrower host range of this smaller group (genus) of viruses.

Our experimental pipeline will proceed as follows: Once we have selected representative viruses, we will use a range of state-of-the-art techniques to quantify their receptor usage and host-range. Subsequently, we will use mutagenesis and protein-binding experiments to dissect the genetic determinants of this zoonotic potential. The assembled scientific research team has a longstanding interest in this area, with broad and overlapping interests in the morbillivirus, coronavirus and arenavirus families. Previously, we have identified amino acid changes in animal morbillivirus attachment proteins which convey tropism to human receptors, solved the structures of arenavirus attachment proteins and more recently examined the likely bat-origin of SARS-CoV-2.

The information we generate in this project will be used to improve our pandemic preparedness, helping us to identify high risk pathogens with broad host-ranges (based on entry). Ultimately this information could be used to design new drugs and vaccines, hopefully preventing future disease in humans and animals.

In order to forewarn future viral zoonotic spill-overs at the human-animal interface, or animal to animal spill-overs which jeopardise livestock, companion animal or wildlife health, it is essential to develop a broader understanding of the important contributory factors. Viral entry is particularly significant in this context, with efficient receptor usage and host-range being intrinsically linked co-variables. Developing the capacity to define the zoonotic potential of individual pathogens, or ideally groups of pathogens, will provide an evidence-base for risk assessment, improving preparedness in a post-Covid environment. Achieving these goals at a genus or family wide level is complicated by the availability and unbiased selection of representative isolates; however, using a 'greedy' maximum phylogenetic distance algorithm, in combination with gene synthesis and viral pseudotyping, we have developed a pipeline to successfully conduct such screens. Building on preliminary data gathered for the morbilliviruses we propose to expand our analysis to the coronaviruses and arenaviruses, leveraging our existing knowledge of their entry mechanisms to screen the receptor usage patterns of an unbiasedly selected library of viral glycoproteins. Using a diverse range of host receptors from human, livestock, companion animals and wildlife, we will then characterise the host-range of these viruses, mirroring the approach we successfully employed for SARS-CoV-2 (PLOS Biology, 2020). Subsequently, structure and sequence guided mutagenesis will be used to identify key residues, or motifs, within individual virus-host interactions that contribute to broad or narrow host range. Finally, we will compare our functional receptor usage data to in vitro assessment of receptor binding affinity, using this evidence to examine any correlation between host range and attachment. Combined, this research program can shed new light on the factors underpinning zoonotic potential.