Effect of clinical isolate or cleavage site mutations in the SARS-CoV-2 spike protein on protein stability, cleavage, and cell-cell fusion

Abstract

The trimeric severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) spike protein (S) is the sole viral protein responsible for both viral binding to a host cell and the membrane fusion event needed for cell entry. In addition to facilitating fusion needed for viral entry, S can also drive cell-cell fusion, a pathogenic effect observed in the lungs of SARS-CoV-2-infected patients. While several studies have investigated S requirements involved in viral particle entry, examination of S stability and factors involved in S cell-cell fusion remain limited. A furin cleavage site at the border between the S1 and S2 subunits (S1/S2) has been identified, along with putative cathepsin L and transmembrane serine protease 2 cleavage sites within S2. We demonstrate that S must be processed at the S1/S2 border in order to mediate cell-cell fusion and that mutations at potential cleavage sites within the S2 subunit alter S processing at the S1/S2 border, thus preventing cell-cell fusion. We also identify residues within the internal fusion peptide and the cytoplasmic tail that modulate S-mediated cell-cell fusion. In addition, we examined S stability and protein cleavage kinetics in a variety of mammalian cell lines, including a bat cell line related to the likely reservoir species for SARS-CoV-2, and provide evidence that proteolytic processing alters the stability of the S trimer. This work therefore offers insight into S stability, proteolytic processing, and factors that mediate S cell-cell fusion, all of which help give a more comprehensive understanding of this high-profile therapeutic target.

Keywords: COVID-19; SARS-CoV-2; coronavirus; fusion protein; membrane fusion; viral protein; virology; virus entry.

Figure 1

SARS-CoV-2 spike is cleaved at the S1/S2 subunit border in a variety of cell lines.A, the indicated cell types transiently expressing S were metabolically labeled for 1 h and chased for times indicated (hours). Band densitometry was used to quantify bands representing full-length S or S cleaved at the S1/S2 border (S2). B, percent cleavage (S2 divided by S plus S2) and (C) overall protein stability (total S, S plus S2, for each time point, normalized to time point 0) were calculated for spike in each cell line (n = 3). D, Vero cells or A549 cells transiently expressing S alone or S with proteases were metabolically labeled and chased for the times indicated (hours). Percent cleavage was measured using band densitometry in both (E) Vero and (F) A549 cells (B, C, E, and F are represented as the average ± SD for three independent experiments). S, spike protein; SARS-CoV-2, severe acute respiratory syndrome coronavirus 2.

Figure 2

CoV-2 spike alone mediates cell–cell fusion. Vero cells expressing S and TMPRSS2, furin, or cathepsin L were imaged at 24 (A) and 48 (B) hpt for syncytia formation (black arrows). Magnification bar is 100 μM. C, a luciferase reporter gene assay was performed with target cells (BSR/T7 cells expressing hACE2 and additional proteases) overlaid onto effector cells (Vero or A549 cells expressing S) for 9 h. D, luciferase reporter gene experiment was performed with additional proteases coexpressed with S in Vero cells and overlaid with target cells expressing hACE2. E, the effect of neuropilin in both target and effector (Vero) cells was examined with a luciferase reporter gene assay. Expression of effector cells is listed along the x-axis. Target cell expression is listed in the legend to the graph. Results expressed as the percent fusion normalized to samples with S in the effector cells and hACE2 only in the target cells (C–E are average ± SD for three independent experiments, performed in duplicate). Significance was determined by two-way ANOVA. ∗p < 0.05, ∗∗∗∗p < 0.0001. CoV-2, coronavirus 2; hACE2, human angiotensin-converting enzyme 2; hpt, hours post transfection; S, spike protein; TMPRSS2, transmembrane serine protease 2.

Figure 3

Mutations at all three potential spike cleavage sites reduce cleavage at the S1/S2 subunit border.A, full or partial alanine substitution mutations were made at each of the three potential cleavage sites. B, plasmids expressing WT S or mutants were transfected into Vero and A549 cells, cells were metabolically labeled for 1 h, and chased for the times indicated. Percent cleavage was determined in (C) Vero cells and (D) A549 cells (average ± SD for three independent experiments). E, surface biotinylation was performed on cells expressing WT S and each mutant. Cells were radiolabeled for 6 h. Protein expression in (F) Vero and (G) A549 cells, results are normalized to WT S, and error bars represent the SD (average ± SD for three independent experiments). H, a luciferase reporter gene assay was performed using target cells expressing hACE2 and EV or TMPRSS2, and effector (Vero) cells with WT S or each mutant. I, luciferase reporter gene analysis with cells expressing hACE2 and effector (Vero) cells transfected with S or S mutants and EV or furin-expressing plasmids. Results of both reporter gene assays are shown normalized to samples with WT S in the effector with hACE2 in target cells (average ± SD for three independent experiments, performed in duplicate). EV, expression vector; hACE2; human angiotensin-converting enzyme 2; S, spike protein; TMPRSS2, transmembrane serine protease 2.

Figure 4

Spike S2 subunit mutations found in circulating variants variably affect spike-mediated cell–cell fusion.A, mutations in the S2 subunit of S identified in circulating SARS-CoV-2 variants. B, WT S or the mutants were transfected into Vero and A549 cells, metabolically labeled for 1 h, and chased for the times indicated. Percent cleavage was determined in (C) Vero and A549 cells (average ± SD for three independent experiments). D, surface biotinylation on cells expressing WT S or each mutant. E, total and surface protein expression normalized to WT S (average ± SD for three independent experiments). F, a luciferase reporter gene assay was performed using target cells expressing EV or hACE2, overlaid onto effector cells transfected with WT S or each mutant. Results are normalized to samples with WT S in the effector cells and hACE2 in target cells (average ± SD for three independent experiments, performed in duplicate). Significance was determined by two-way ANOVA: ∗p < 0.05 and ∗∗p < 0.01. EV, expression vector; hACE2, human angiotensin-converting enzyme 2; S, spike protein; SARS-CoV-2, severe acute respiratory syndrome coronavirus 2.

Figure 5

All circulating mutants form large syncytia, similar to WT spike. Immunofluorescence of WT S or the circulating mutants (S stained in green) transiently expressed in Vero cells. White arrows indicated S-positive cellular extensions. The magnification bar is 20 μM. S, spike protein.

Figure 6

Mutations at downstream potential cleavage sites render the S1/S2 border cleavage site less accessible to proteases.A, Vero or A549 cells expressing WT S or S cleavage mutants were metabolically labeled for 6 h. Surface proteins were biotinylated, and samples were either treated for 10 min with TPCK–trypsin or left as untreated controls (as indicated). B, Vero or A549 cells expressing indicated proteins were metabolically labeled for 6 h. Samples were treated at the indicated temperatures before separation on a nonreducing SDS-PAGE. Oligomers are labeled on the right based on size, and colored ∗ represents potential intermediate species (n = 3). Using band densitometry to quantify the bands in (A), percent cleavage was measured in (C) Vero and (D) A549 cells for both the surface (top graphs) and total (bottom graphs) populations (average ± SD for three independent experiments). Significance was determined by two-way ANOVA: ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.0005, and ∗∗∗∗p < 0.0001. S, spike protein.

Figure 7

Furin or furin-like proteases in pteropus bat cells can cleave the S1/S2 border site of SARS-CoV-2 spike.A, surface biotinylation was performed on pteropus lung and pteropus fetus cells transfected with plasmids for WT S or the del. PRRA mutant. B, surface or total protein expression levels were quantified using band densitometry and normalized to WT S levels. C, pt. lung and pt. fetus cells were transfected with WT S or del. PRRA mutant, metabolically labeled for 1 h, and chased for the times indicated. Using band densitometry to quantify the bands in (C), results are expressed as (D) protein cleavage and (E) protein stability over the times indicated. (B–E average ± SD for three independent experiments). SARS-CoV-2, severe acute respiratory syndrome coronavirus 2.