The first point of contact between human cells and coronaviruses is mediated by the viral Spike (S) protein. The objective of this research is to identify antiviral target sites in S protein and predict which ones will maintain potency against future viral strains as the coronavirus S protein structure continues to evolve. The hypothesis is that based on the level of amino acid conservation in S protein, we can identify regions that are selectively potent toward the current CoV-2 strain or broadly targetable in a range of human- and animal-infecting coronaviruses. These conserved regions can potentially be targeted to prevent emergence of novel coronaviruses from zoonotic to human jumps. To test this hypothesis, two complementary in silico methods were used to analyze S protein sequence conservation among twenty, severe coronaviruses (representing MERS, SARS, and CoV-2) and ten animal-infecting coronaviruses. The first method aligned protein sequences and scored conservation at each position in the alignment. The second method used nucleic acid sequence input to calculate the ratio of non-synonymous over synonymous mutations as a measure of selective pressure on each residue. To identify drug targets in S protein, we performed molecular docking of the S protein receptor binding domain with the NCI diversity set 5 library. Four different pockets and six different compounds were identified using this approach. Analysis of human receptor binding data from available literature combined with the conservation results from our work identified a repertoire of key residues in pockets 2, 3, and 4 that possess several phylogenetic characteristics relevant to antiviral efficacy. Pocket 2 contains highly conserved residues implicated in zoonotic jumps, structural stability, and human receptor binding. Pocket 4 contains glycosylated residue N343, targeting which could potentially destabilize the S protein structure or uncover antigens that increase vulnerability to the host immune response. It also contains N501, a residue that is mutated in several new highly transmissible CoV-2 strains (e.g.,  UK, Brazil, and South Africa variants). Overall, pocket 2 is the most promising target for antivirals against a broad range of present and future coronaviruses, while pocket 4 is expected to be a potent antiviral target in current CoV-2 strains. Binding free energy calculations were used to assess the effect of mutations in the newly evolved CoV-2 strains (UK and South African). The results show enhanced binding energy for E484K and N501Y mutations supporting the independent evolution of these mutations in different strains. Future research aims involve tailoring the compounds to disrupt the key residues identified in pockets 2 and 4. These current results have advanced our long-term goal of developing antiviral drugs that are effective against CoV-2 today and have the longevity and targeting breadth needed to help prevent coronavirus pandemics in the future.

Presenter(s)

Sydney Newsome

Richard Van

Yihan Shao

Rakhi Rajan