Role of Q675H Mutation in Improving SARS-CoV-2 Spike Interaction with the Furin Binding Pocket

Abstract

Genotype screening was implemented in Italy and showed a significant prevalence of new SARS-CoV-2 mutants carrying Q675H mutation, near the furin cleavage site of spike protein. Currently, this mutation, which is expressed on different SARS-CoV-2 lineages circulating worldwide, has not been thoughtfully investigated. Therefore, we performed phylogenetic and biocomputational analysis to better understand SARS-CoV-2 Q675H mutants' evolutionary relationships with other circulating lineages and Q675H function in its molecular context. Our studies reveal that Q675H spike mutation is the result of parallel evolution because it arose independently in separate evolutionary clades. In silico data show that the Q675H mutation gives rise to a hydrogen-bonds network in the spike polar region. This results in an optimized directionality of arginine residues involved in interaction of spike with the furin binding pocket, thus improving proteolytic exposure of the viral protein. Furin was predicted to have a greater affinity for Q675H than Q675 substrate conformations. As a consequence, Q675H mutation could confer a fitness advantage to SARS-CoV-2 by promoting a more efficient viral entry. Interestingly, here we have shown that Q675H spike mutation is documented in all the VOCs. This finding highlights that VOCs are still evolving to enhance viral fitness and to adapt to the human host. At the same time, it may suggest Q675H spike mutation involvement in SARS-CoV-2 evolution.

Keywords: Q675H spike mutation; SARS-CoV-2; VOC; furin cleavage; molecular dynamics; phylogenesis.

Figure 1

Prevalence of SARS-CoV-2 Q675H mutants in Italy from October 2020 to October 2021. The number of SARS-CoV-2 Q675H mutants has been normalized to the total number of the Italian SARS-CoV-2 sequences retrieved from GISAID.

Figure 2

Maximum-likelihood tree of European SARS-CoV-2 sequences. The mid-point rooted, ML tree includes 1487 SARS-CoV-2 sequences retrieved in Europe from GISAID database from March 2020 to March 2021 and 11 SARS-CoV-2 spike Q675H sequences obtained in Brescia between the end of January 2021 and the end of February 2021. Branches are colored by lineages: B.1 + spike Q675H in red; B.1 in black; B.1.177 + spike Q675H in blue; B.1.177 in fuchsia; all of the other lineages carrying the Q675H mutation indicated as “others + spike Q675H” in green; all the other lineages not carrying the Q675H mutation indicated as “others” in brown. Red dots represent the 11 Brescia SARS-CoV-2 spike Q675H mutants.

Figure 3

Prevalence of Q675H spike mutation in SARS-CoV-2 VOC. For each lineage, solely the months in which the number of sequences of the concerned lineage is higher than 2000 are included. In each graph, the bars represent Q675H mutation percentage within each lineage, while the black line accounts for the fraction of each lineage compared to the global amount of SARS-CoV-2 sequences. (A) B.1.1.7; (B) B.1.351; (C) P.1; (D) B.1.617.2.

Figure 4

Time-scaled maximum likelihood trees including sequences representative of worldwide SARS-CoV-2 VOC (black circles) from November 2020 until July 2021. Genomes carrying the mutation Q675H are highlighted with red circles. (A) B.1.1.7 lineage includes 1500 genomes plus 278 carrying Q675H mutation; (B) B.1.351 lineage includes 1100 genomes plus 104 carrying Q675H mutation; (C) P.1 lineage includes 1400 genomes plus 59 carrying Q675H mutation; (D) B.1.617 + A.Y.x lineage includes 1776 genomes plus 664 carrying the Q675H mutation.

Figure 5

Root mean square deviation (RMSD) plot for the trimeric form of spike protein (chain A, chain B, and chain C) during the 200 ns of production runs. The X-axis indicates time in ns and the Y-axis represents RMSD values in Å.

Figure 6

(Panel A–F) H-bonds and conformational flexibility around furin cleavage site loop. Representative conformation of furin cleavage site loop for each chain of Q675 (left) and Q675H (right) obtained by cluster analysis. The residues involved in H-bonds network are shown in ribbon and sticks, respectively. Chains (A–C) of Q675 are shown in panels (A–C); chains A–C of Q675H are shown in panels (D–F). The H-bonds are shown as dotted lines. Plot of curvature (blue line) and torsion (red line) of the Q675 and Q675H of ensemble for each chain in function of sequence for residues 656 to 697 around furin cleavage site loop are shown. The curvature and torsion values of the residues of furin cleavage site affected by hydrogen bond network is highlighted (window).

Figure 7

(Top) The furin cleavage site sequence in Q675 and Q675H is shown. (panel A) The X-Ray structure of the inhibitor (m-guanidinomethyl-phenylacetyl-Arg-Val-Arg-(4-amidomethyl)- benzamidine) bound to active site of furin (PDB code 5JXH). The catalytic domain of furin is shown in yellow, the locations of the P1 (blue) and P4 (red) residues of inhibitor are shown in sticks, while their respective S1 and S4 interaction pockets are labelled and indicated by an arc. (panel B) The substrate conformations of the trimeric spike of Q675 and Q675H are shown. In the conformational ensemble, the side chain of the three-arginine residues of the furin recognition motif in their spatial arrangement is shown in line P1-R685 (blue), P3-R683 (green), and P4-R682 (red), and the arginine residues of the most representative conformation are highlighted in sticks. The arginine residues for each substrate conformation involved in the interaction with their respective S1 and S4 furin pockets are labelled.

Figure 8

(Upper panel A and B) Furin and Q675 (left) or Q675H (right) models are shown in Connolly surface gray and fuchsia, respectively. (Lower panel A and B) The binding pose of the arginines of the furin cleavage site motif in P1-S1/P3-S4 substrate conformation of Q675 (left) and P1–S1/P4–S4 substrate conformation of Q675H into the corresponding binding pockets S1 and S4 are shown. The catalytic domain of furin is shown in yellow; the locations of the P1 (blue), P3 (green), and P4 (red) of arginine involved in the interaction with enzyme are shown in sticks, while their respective S1 and S4 pockets are labelled and indicated by an arc.