Functional analysis of polymorphisms at the S1/S2 site of SARS-CoV-2 spike protein

Abstract

Several SARS-CoV-2 variants emerged that harbor mutations in the surface unit of the viral spike (S) protein that enhance infectivity and transmissibility. Here, we analyzed whether ten naturally-occurring mutations found within the extended loop harboring the S1/S2 cleavage site of the S protein, a determinant of SARS-CoV-2 cell tropism and pathogenicity, impact S protein processing and function. None of the mutations increased but several decreased S protein cleavage at the S1/S2 site, including S686G and P681H, the latter of which is found in variants of concern B.1.1.7 (Alpha variant) and B.1.1.529 (Omicron variant). None of the mutations reduced ACE2 binding and cell-cell fusion although several modulated the efficiency of host cell entry. The effects of mutation S686G on viral entry were cell-type dependent and could be linked to the availability of cathepsin L for S protein activation. These results show that polymorphisms at the S1/S2 site can modulate S protein processing and host cell entry.

Fig 1. Mutations at S1/S2 site reduce SARS-CoV-2 S cleavage.

A) 3D reconstruction of the SARS-CoV-2 S protein trimer. The location of the mutations is highlighted and magnified. Colour code: light blue—S1 subunit with RBD in purple, grey—S2 subunit, orange—cleavage loop encompassing the S1/S2 cleavage site, red—S1/S2 cleavage site. Since the crystal structure PDB: 6XR8 lacks the structural information for residues Q677-A688 of the extended S1/S2 loop, the corresponding structure has been computationally reconstructed. B) Schematic illustration of S protein domain organization. RBD—receptor binding domain, TD—transmembrane domain. The S1/S2 and S2’ cleavage sites are indicated. Mutations located within or adjacent to the S1/S2 site are highlighted. Values in brackets indicate the scores by furin cleavage prediction (ProP 1.0; n.d., not determinable) C) Incorporation of SARS-CoV-2 S proteins into VSV particles. Pseudotyped particles harbouring the indicated S proteins equipped with a C-terminal HA antigenic tag were subjected to immunoblot analysis, using anti-HA antibody. Black and grey filled arrows indicate uncleaved precursor SARS-CoV-2 S (S0) and S2, respectively. Detection of VSV-M served as a loading control. Shown is a representative immunoblot from three independent experiments. Further, total S protein levels in particles were quantified. For this, S protein signals were first corrected with respect to the corresponding VSV-M signals and subsequently normalized (WT SARS-CoV-2 = 1). The average data from three experiments (+/- the standard error of the mean, SEM) are shown below the immunoblot. D) Quantification of cleavage efficiency. For each S protein total S protein signals (S0 + S2) were set as 100% and the relative proportion of the individual S0 and S2 signals were calculated. Displayed are the average data from three independent experiments. Error bars represent the SEM. Statistical significance was analysed by two-tailed Student’s t-test with Welch’s correction (p > 0.05, not significant [ns], p ≤ 0.05, *; p ≤ 0.01, **). WT = wildtype.

Fig 2. Mutations at the S1/S2 site have little impact on soluble ACE2-Fc binding.

293T cells expressing the indicated S protein mutants were incubated with soluble ACE2-Fc and binding was detected using an Alexa488-coupled secondary antibody and flow cytometry. Cells that did not express S protein were used as negative control. The average geometric mean channel fluorescence (GMCF) of three independent experiments is shown. Error bars represent the standard deviation (SD). Statistical significance was analysed by two-tailed Student’s t-test with Welch’s correction (p > 0.05, not significant [ns], p ≤ 0.05, *). WT = wildtype.

Fig 3. SARS-CoV-2 mutants mediate robust cell-cell fusion.

A) Effector cells cotransfected with Vp16-Gal4 transactivator plasmid and pCG1 empty vector (negative control) or expression plasmids for the indicated S proteins were co-cultured with target cells cotransfected with Gal4-TurboGFP-Luc reporter plasmid and either ACE2 expression plasmid alone or in combination with TMPRSS2 expression plasmid. At 24 h post transfection, luciferase activity in cell lysates was measured. Displayed are the average data from four independent experiments (each performed with biological triplicates) where cell-cell fusion was either normalized against the assay background (A; fold over background, set as 1) or against WT SARS-CoV-2 S (B; percentage, set as 100%). Error bars represent the SEM. Statistical significance was analysed by two-tailed Student’s t-test with Welch’s correction (p > 0.05, not significant [ns], p ≤ 0.05, *; p ≤ 0.01, **; p ≤ 0.001, ***). WT = wildtype.

Fig 4. S686G modulates SARS-CoV-2 S-driven entry in a cell line-dependent manner.

A) The indicated cell lines were inoculated with equal volumes of pseudotyped particles bearing the indicated S proteins or no S protein (negative control). Transduction efficiency was quantified by measuring virus encoded luciferase activity in cell lysates at 16–20 h post transduction. The average from six independent experiments is shown. Error bars represent SEM. B) Stable overexpression of cathepsin L (CTSL) in Calu-3 cells. Cell lysates of parental Calu-3 cells and Calu-3 cells stably expressing CTSL harbouring a C-terminal cMYC epitope tag (CTSL-cMYC) were subjected to immunoblot analysis using anti-cMYC antibody. Detection of beta-actin (ACTB) served as a loading control. Shown is a representative immunoblot from two independent experiments. C) The experiment was carried out as described for panel A but entry into Calu-3 WT and Calu-3 cells stably expressing CTSL was analysed. Statistical significance was analysed by two-tailed Student’s t-test with Welch’s correction (p > 0.05, not significant [ns], p ≤ 0.05, *; p ≤ 0.01, **; p ≤ 0.001, ***). WT = wildtype.