Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) causes an outbreak of COVID-19 disease in humans with the aid of spike protein. It consists of a receptor-binding domain (RBD) that recognizes and binds to the host receptor angiotensin-converting enzyme 2 (ACE2). The aim of this study was to examine the mutational effect of spike protein on the sequence through an interaction study of the mutant spike protein and the human ACE2 protein at the structural level. A total of 17,227 spike proteins from Asia, Africa, Europe, Oceania, South America, and North America were compared to the Wuhan spike protein reference sequence (Wuhan-Hu-1). The structural and stability implications of D614G, N501Y, and S477N mutations were evaluated. The binding affinity between mutated RBD and human ACE2 protein was also studied. The D614G mutation may have originated in Germany, Europe based on the date of the first sample collection report. It is now widely circulated all over the world with most occurrences in North America. The mutations N501Y and S477N may have originated from Oceania based on the date of the first sample collection report and also have the highest occurrences in Oceania. Based on the computational analysis of mutational effects, the D614G, N501Y, and S477N mutations decreased stability and were tolerated. For disease propensity prediction, N501Y was more prone to disease compared to D614G, while S477N was not prone to disease. The mutation of D to G at position 614 and S to N at position 477 for secondary structure prediction shows no changes in secondary structure while remaining in the coil region, whereas the mutation of N to Y at position 501 changes from coil structure to extended strand. N501Y mutation has a higher affinity to human ACE2 protein compared to D614G and S477N based on a docking study. D614G spike mutation was identified to exist between the two hosts based on a comparison of SARS-CoV-2 derived between the mink and human. Further research is needed on the link between the mink mutation N501T and the mutation N501Y in humans, which has evolved as a separate variant.

Conservation history of D614 residue is valuable in predicting the consequences of D614G mutation in the SARS-CoV-2 spike glycoprotein (SGP). We report here that the D614 belonged to an extraordinarily conserved, densely hydrophobic eleven amino acid peptide-motif vavlyqdvnct (11-aa), in the Sarbecovirus group and the variant carrying vavlyqdvnct had appeared in Chinese samples and became predominant in several geographical hotspots in late March, 2020. Interestingly a 2009 annotation of SARS-CoV contained the same mutation.

We analyzed the SARS-CoV-2 spike (S) protein amino acid sequence extracted from 11,542 viral genomic sequences submitted to the Global Initiative on Sharing All Influenza Data (GISAID) database through April 27, 2020. Consistent with prior reports, we found a major S protein mutation, D614 to G614, that was represented in 56% of all the analyzed sequences. All other mutations combined were less than 10%. After parsing the data geographically, we found most of the Chinese patient samples showed D614 (97%). By contrast, most patient samples in many European countries showed G614 (51 to 88%). In the United States, the genotypic distribution in California and Washington was similar to Asian countries, while the distribution in other US states was comparable to Europe. We observed a dramatic increase in the frequency of G614 over time in multiple regions, surpassing D614 when both were present, suggesting G614 S protein virus outcompetes D614 S protein virus. To gain insight into the consequences of the D614G mutation, homology modeling using a multi-template threading mechanism with ab initio structural refinement was performed for a region of the S protein (S591 to N710) spanning the D614G mutation and the S1 furin cleavage site. Molecular models of this region containing D614 or G614 revealed a major difference in secondary structure at the furin domain (RRARS, R682 to S686). The D614 model predicted a random coil structure in the furin domain whereas the G614 model predicted an alpha helix. Critical residues in the cleavage domain of G614 model were found to better align with the PDB structure of a furin inhibitor. Thus, homology modeling studies suggest a potential mechanism whereby the D614G mutation may confer a competitive advantage at the furin binding domain that may contribute to the rise of the D614G SARS-CoV-2 mutant.

As reported by us and others previously (1, 2), the D614G mutation appeared in the spike glycoprotein (SPG) of the SARS-CoV-2 (the pathogen behind COVID-19) at the early stages of the pandemic and then G614 containing variant of SARS-CoV-2 became the predominant strain in most human populations across the world. However, one of the most recent reports from India (3) stated the incidence of G614 to be only 26% in the Indian population. This report is contradictory to the information available through the GenBank (4) SARS-CoV-2 sequence deposits made by various laboratories from India. The above stated report currently circulating in the Indian media is likely to create a public perception that the Indian strain is less contagious and such a notion could be harmful to people’s welfare. In view of this concern we have re-evaluated, updated and recalculated the incidence of the G614 variant in the Indian population by analyzing 395 Indian SARS-CoV-2 genomic sequences available in the GenBank as of June 26, 2020. In our analysis we have categorized the samples by the month in which the samples were collected. We have used an alignment-free software tool named Compare (5, 6), and the Basic Local Alignment Search Tool (BLAST) (7) in the present analysis. We finally inspected each of the 395 sequences physically for the presence of aspartic acid (D) or glycine (G) at the 614^th position of the spike glycoprotein. We analyzed an Australian cohort in parallel for comparison. We have found that the prevalence of G614 variant in the Indian samples for the month of June 2020 is 90.6%. The trends are similar with the Australian samples.