The COVID-19 pandemic is caused by severe zoonotic coronavirus acute respiratory syndrome 2 (SARS-CoV-2). Like other pathogenic coronaviruses, SARS-CoV-2 crossed the species barrier to cause human disease. This fact prompted the current study, published on the printout server bioRxiv* in October 2020, examining the way coronaviruses engage with their host cell receptors in a range of animals.
The importance of this study lies in the close interaction between human beings and domestic animals of various kinds, where respiratory proliferation is very likely to occur. Not only is this a constant threat to the human race, but it is an economic challenge for livestock farmers and pet owners.
Studies on viral pathology and to quantify candidate therapeutic drugs also use virus-infected animals. Therefore, this study will assist public health experts, economic strategists, and those working with animal models of COVID-19.
Viral Binding Through Spike Protein
The virus binds to angiotensin-converting enzyme 2 (ACE2) on the host cells in the lungs and other high-affinity organs. The result is the fusion of the virus into the host cell membrane, which leads to cell entry and infection.
The SARS-CoV-2 spike-ACE2 complex was studied crystallographically and by electron microscopy. This illustrated how the amino acids interacted in the spike receptor binding domain (RBD) and the human ACE2.
Being able to measure potential spike RBD binding would help predict which animals are susceptible to this virus.
Variability in the ACE2 Sequence
The researchers in the current study examined the conserved and divergent sequences of the ACE2 receptor protein in several relevant species, especially in the receptor binding domain (RBD).
All vertebrae had a homologous ACE2 sequence to human ACE2. The sequences of macaque and rhesus cynomolgus monkeys were 95% and 94% identical to human sequences. For other mammals, it was about 80-87%.
With the horseshoe bat, suspected to be a reservoir for the virus, the identity for the entire ACE2 protein was only 81% and for the 25 interacting amino acids 76%. In fact, in many cases, the degree of identity was markedly different across the whole sequence and the interactive remains.
Corresponding to this finding, they found that, from cattle, one ACE2 protein (Taurus boss), although distinct from the human variation across the entire sequence (79% identity), showed high affinity binding to the SARS-CoV-2 pigmented RBD (84% identity). This suggests that there may be many species of mammals that allow infection with this virus.
Conversely, binding site residues are lower in identity compared to the whole sequence in dogs and rabbits. Future studies must establish the hypothesis that the residues interacting with the RBD are much more relevant to binding than those outside this site.
Variable residues at the ACE2-RBD interface of individual species. With human ACE2 360 as a reference, the various interactive residues for each species are colored red, and retain 361 cyan-colored residues. For the rest of the ACE2 protein, preserved residues are white and 362 variable residues are light red. Certain residues like M82, Q24, D30, and H34 are often mutated compared to 363 people. D30 is conservatively changed to glutamate, but more non-conservative changes can be seen for H34, for example.
ACE2 Sequence and Binding Connection
The researchers compared the importance of different regions and individual amino acids of the protein to its structure. Earlier this type of study has demonstrated the presence of complementary decision regions (CDRs) in antibodies, which interact with the corresponding antigen.
Shannon’s entropy is a powerful tool used for analyzing variability. Such tests showed that the ACE2 was highly conserved across the entire sequence. Of the 25 residues bound to or interacting with RBD, 21 were well conserved, and none had high entropy in Shannon.
There were 5 fully conserved residues, 4 close to the binding site, and only one as contact residue. Their conservation demonstrates the importance of these to cell entry through protease-mediated degradation or to the protein structure among species.
The most variable were positions 24 and 34, with the latter at the center of the ACE2-RBD interface and by far the highest Shannon entropy. Both of these are contact residues. Some researchers have suggested that these positions may help predict the infectivity of a particular SARS-CoV-2 variant.
Residue Matching and Binding Connection
The residues that failed to match human ACE2 were evaluated separately in the next step. They found, with the horseshoe bat, that only 5 of 8 residues in contact with the RBD were the same as in human ACE2, but 19/25 adjacent residues. This may mean that its ACE2 has lower affinity to the virus.
Another presumed intermediate animal host for the virus was the pangolin, and here only one link and three neighboring residues failed to match human ACE2. More data is needed to confirm its role in the transmission of SARS-CoV-2. Similar close games include dogs and cats.
In cattle, three interacting residues at the binding site are altered, and one is contact residue. Cows are known to harbor bovine coronavirus remotely associated with SARS-CoV-2, and two coronaviruses in this family have jumped across the species barrier that already infects humans with the common cold.
However, these bind to sialoglycans rather than ACE2. However, cows need to be evaluated for their ability to act as a reservoir host for SARS-CoV-2. In fact, interaction between bovine coronavirus and SARS-CoV-2 could lead to recombination emerging, with dangerous consequences.
Bovine ACE2 Highly Preserved at the Binding Site
Although the ACE2 of cows is about 78% identical to humans, the residues at the binding site show higher conservation (84%). The researchers examined whether this would allow high affinity binding to the target virus RBD despite the rest of the lower homology of the ACE2 sequence.
They found, when testing ELISA, that binding of bovine ACE2 appeared to be ten times lower than for human ACE2, mainly due to a faster disassembly rate of the complex after formation. Intriguingly, this lower affinity was comparable to human ACE2 for SARS-CoV-1 RBD, as it was in the low nanomolar range compared to the sub-nanomolar concentration of the latter.
Best Fit with Two Site Binding Model
Human and bovine ACE2 showed a much more consistent fit when a two-site model was used to simulate the interaction with RBD of the virus. This may result in RBD multimers being formed to bind vigorously to the surface of the ACE2. If so, the presence of a second site would explain the affinity ten times higher.
The researchers note, “Such interactions would be important to explore to understand the details of the interaction between RBD and ACE2 coronavirus spikes on the cell surface.” Bovine ACE2 binds to the high-affinity RBD SARS-CoV-2, although it is 5-10 times lower than with human ACE2, and appears to be able to allow the virus to infect cows.
The study also demonstrates the feasibility of using the biochemical characterization of the receptor to predict how various species will treat the virus, especially with rapid screening.
The researchers note, “Further study into the viral receptors of various species, including the recently discovered neuropilin-2 co-receptor for SARS-CoV-2, should enable a much more accurate and rapid development of predictive algorithms based on viral receptor sequence and structural modeling.. “
* Important Warning
bioRxiv publishes preliminary scientific reports that are not peer-reviewed and, therefore, should not be considered definitive, guide clinical practice / health-related behavior, or be treated as established knowledge.