Molecular epidemiology of SARS-CoV-2 genetic variants in Africa: The development of a guide for the selection of effective variant-specific vaccines

Abstract

Fears that newly emerging SARS-CoV-2 variants might resist antibody-based treatments and evade antibodies produced by vaccination or prior infection have led to various investigations on the effect of SARS-CoV-2 variants and mutations on neutralizing antibody activity. Since the immune response consists of only one component, the neutralizing antibody titers, and connection of these neutralizing antibody titers to protection continue to be determined, the findings from these studies cannot be utilized to make determinations on the efficacy or effectiveness of vaccines. The nature and quality of antibodies also serve as significant indicators of a strong neutralizing antibody response. In this investigation, the Molecular Dynamics simulations to test for vaccine efficacy and determine variant-specific vaccines were done. First determination of genetic diversity and spread dynamics of SARS-CoV-2 lineages circulating within African populations was done and only 465 of the 2610 Pango lineages identified worldwide circulated in Africa, with the following five Variants of Concern (VOCs) predominating at different stages: Alpha, Beta, Delta, Gamma and Omicron. We established that South Africa, Kenya, and Nigeria were significant drivers of viral transmissions between Sub Saharan African countries. These findings provided insight into the viral variants that are currently circulating throughout Africa's population. Secondly, identification of mutation hotspots of SARS-CoV-2 variants within the spike protein region using NextClade was done. The same five VOCs that predominated at different stages of the pandemic were identified as the most epidemiological important variants. These variants showed that there are three mutation hotspots within the spike protein region, at the NTD, RBD and CTD 2. Thirdly, we homology modelled different types of IgG antibodies, those that are infection-induced and vaccine-induced by retrieving antibody sequences from The Coronavirus Antibody Database (CoV-AbDab) and using RosettaAntibodyDesign (RAbD) server to homology model antibodies. Protein-protein docking of the antibodies which were homology modeled and spike protein structures of the five VOCs retrieved from PDB database was done using HADDOCK. Lastly; Molecular Dynamics simulations were done using GROMACS. XMGRACE was used to calculate RMSD and RMSF to determine the stability of the protein relative to its conformation. We then performed binding free energy calculations using MMPBSA to examine reduction of vaccine efficacy and determine variant-specific vaccines by noting the binding affinity of antibodies to variants. MD results implied that the AstraZeneca vaccine should be used as a multivalent vaccine on all VOCs or as v a bivalent vaccine on Alpha and Beta variants. Pfizer vaccine should be used as a bivalent vaccine on Delta and Omicron while Moderna vaccine should be used on Gamma variants