Spike substitution T813S increases Sarbecovirus fusogenicity by enhancing the usage of TMPRSS2

Abstract

SARS-CoV Spike (S) protein shares considerable homology with SARS-CoV-2 S, especially in the conserved S2 subunit (S2). S protein mediates coronavirus receptor binding and membrane fusion, and the latter activity can greatly influence coronavirus infection. We observed that SARS-CoV S is less effective in inducing membrane fusion compared with SARS-CoV-2 S. We identify that S813T mutation is sufficient in S2 interfering with the cleavage of SARS-CoV-2 S by TMPRSS2, reducing spike fusogenicity and pseudoparticle entry. Conversely, the mutation of T813S in SARS-CoV S increased fusion ability and viral replication. Our data suggested that residue 813 in the S was critical for the proteolytic activation, and the change from threonine to serine at 813 position might be an evolutionary feature adopted by SARS-2-related viruses. This finding deepened the understanding of Spike fusogenicity and could provide a new perspective for exploring Sarbecovirus' evolution.

Fig 1. Replacement of S2 subunit affected the fusogenicity of S protein.

(A) Schematic diagram for the construction of the S2-chimeric Spike. The Saffron graph depicts SARS-CoV-2 S, which has the multibasic motif (PRRAR) at the S1/S2 Cleavage Site; the grey graph represents SARS-CoV S. The numbers in parentheses are identical to those in Fig 1B–1G. TM, transmembrane domain; CP, cytoplasmic domain. (B and C) Flow cytometry. After transfection, the expression of surface S proteins was detected using s309 antibody which binds RBD efficiently and mouse anti-human IgG-FITC, respectively (B) and the summarized results are shown (C). MFI, mean fluorescent intensity. (D) Western blot. Representative blots of cell lysates showing spike cleavage in parental and chimeric Spikes with or without trypsin treatment. Immunoblots were probed with the C-terminal part of the spike protein detected by using an anti-Flag tag antibody, the full-length (FL) S and cleaved (CL) S (i.e., S2) were marked as indicated; Beta-actin was probed as a loading control. The band intensity was densitometrically calculated using Image J, and the ratio of Cleaved-S/total S (%) was shown to quantify the difference in the S2 formation. (E-G) Spike-based cell-cell fusion assay. A schematic diagram showing the GFP-split system for Spike-ACE2 mediated cell fusion (F), and representative images at 2, 12, 24, 36 h post-treatment (E). The summarized results of the ratio of fusion were shown (G). Scale bar, 500 μm. (H) Microscopic observation of syncytia. Cells expressing S proteins were photographed under a regular phase contrast light microscope with a 10× lens, and representative images at 48 h after mixing with ACE2-293T cells. Scale bar, 200 μm. Results are means +/- SD from at least three fields per condition. Results are representative of at least three independent experiments. Statistically significant differences between parental S (Spike1 or Spike3) and chimeric S (Spike2 or Spike4) were determined by a two-sided Student’s t test (C and D, ns: non-significant), or two-sided paired t test (G, *: p<0.05, **: p<0.01).

Fig 2. Replacing IFP motif of parental Spike influenced the fusogenicity significantly.

(A and G) Schematic diagram of the S2-chimeric Spike bearing swapped motif. The S2 was divided into 3 parts according to the structure and function: F1 (686–833), F2 (834–984) and F3 (985–1213) of SARS-CoV-2 (A); F1 (668–815), F2 (816–966), F3 (967–1195) of SARS-CoV (F). Then replacing the corresponding parts with the others separately. Further, the F1 was divided into two parts: FP (686–787) and IFP (788–833) of SARS2 (A), or FP (668–769) and IFP (770–815) of SARS (F). The numbers in parentheses are identical to those in Fig 2B–2F and 2G–2L. (B and H) Flow cytometry. The summarized results of the surface S expression were shown. s309 antibody and mouse anti-human IgG-FITC were used respectively. (C, D and I, J) Western blot. A representative blot of cell lysates showing spike cleavage. FL-Spike and CL-Spike were marked as indicated. Beta-actin was used as a control. The band intensity was densitometrically calculated using Image J, and the ratio of Cleaved-S/total S (%) was shown. (E and K) Spike-based fusion assay. The fusion activity was quantified by measuring the ratio of GFP+ area to DAPI area by imaging at different times (2, 6, 12 and 24hpt). The results for SARS-CoV-2, Spike4, 10–14 were shown as Saffron lines and SARS-CoV, Spike2, 5–9 were shown as grey lines, respectively. (F and L) Representative images of cell-cell fusion. Scale bar: 500 μm. Results are means +/- SD from at least three fields per condition. Results are representative of at least three independent experiments. Statistically significant differences between parental S (Spike1 or Spike3) and chimeric Spikes were determined by two-sided paired t test (D and J, *: p<0.05, **:p<0.01), or Student’s test at each point (E and K, *: p<0.05, **: p<0.01).

Fig 3. S813T mutation affected the cell membrane fusion ability of IFP-chimeric Spike.

(A and F) Schematic diagram of the IFP-chimeric Spike mutants and the numbers in parentheses are identical to those in Fig 3B–3D and 3G–3I. Align the sequence of AA 788–833 in SARS-CoV-2 with SARS-CoV (AA 770–815). Residue numbering is shown according to SARS-CoV-2 S. The mutation sites were marked in red or blue, and the S2’ cleavage sites were in Green. (B and G) Flow cytometry. The summarized results of the surface S expression were shown. s309 antibody and mouse anti-human IgG-FITC were used respectively. (C and H) Western blot. Left panel: A representative blot of cell lysates from WT and mutant chimeric-Spike expressing 293T cells, FL-Spike and CL-Spike were marked as indicated. Beta-actin was used as a control. Right panel: quantified band intensity using Image J to analyze the protein expression and the ratio of Cleaved-S to the total S. (D and I) Spike-based fusion assay. The fusion activity was quantified by measuring the ratio of GFP+ area to DAPI area by imaging at different times (2, 6, 12 and 24hpt). The results for mutant Spike9 and 14 were shown as yellow-green, Spike25-30 as blue lines, and Spike31-36 as red lines, respectively. (E and J) Representative images of cell-cell fusion. Scale bar: 200 μm. Results are means +/- SD from at least three fields per condition. Results are representative of at least three independent experiments. Statistically significant differences between parental S (Spike9 or Spike14) and mutants were determined by Student’s test at each point (D and I, *: p<0.05, **: p<0.01).

Fig 4. Spike S813T mutation disturbed the membrane fusion and infection of Sarbecovirus significantly.

(A) Spike-based fusion assay. The fusion activity was quantified at different times (2, 6, 12 and 24hpt). ACE2-293T (Red) and Caco-2 (Green) were used, the results of S813 Spike were shown as full lines and T813 Spike as dotted lines. (B) Representative images of cell-cell fusion. Scale bar: 200 μm. (C) Flow cytometry. The summarized results of the surface S expression were shown. s309 antibody and mouse anti-human IgG-FITC were used. S813 S and T813 S were shown as saffron and magentas, respectively. (D) Western blot to monitor the S2 formation with or without trypsin treatment. Top panel: A representative blot of cell lysates from S813 and T813 Spike expressing 293T cells, FL-Spike and CL-Spike were marked as indicated. Beta-actin was used as a control. Bottom panel: quantified band intensity using Image J to analyze the protein expression and the ratio of Cleaved-S to the total S. (E) Pseudovirus assay. The replication of Pseudovirus with S813 or T813 Spike in ACE2-293T and Caco-2 cells was determined by RT-qPCR, and the infectivity percentage normalized with that of the virus pseudotyped with Spike1 was shown. S813 S and T813 S were shown as saffron and magentas, respectively. Results are means +/- SD from at least three fields per condition. Results are representative of at least three independent experiments. Statistically significant differences (*: p<0.05) between S813 S and T813 S were determined by Student’s test at each point (A), or two-sided paired t test (D and E).

Fig 5. The S813T mutation has no effect on S protein interactions with ACE2.

(A) SDS-polyacrylamide gel electrophoresis (PAGE) of SARS2-6P, SARS-2P and Residue 813 substitution S variants. Molecular weight standards are indicated at the left in KD. (B) Competitive ELSIA to detect the binding affinity between S proteins and ACE2. Red line indicated SARS-CoV-2, and blue line indicated SARS-CoV. (C) Neutralization curves for RBD representative antibodies, m396 and X65, with the VSVpp containing parent and mutant S proteins. Each point represents the mean and standard error of 2 independent measurements.

Fig 6. S813T mutation reduced the use of TMPRSS2 by S protein.

(A) Schematic illustration of SARS-CoV S and SARS-CoV-2 S including proteolytic cleavage sites: S1/S1, S2’ and CTSL cleavage sites in S1. Residue 813 was indicated as red. Arrow heads indicated the cleavage site. (B) Western blot. Top panel: A representative blot of VSVpp digested with TPCK-trypsin (2 μg/ml at 37°C for 30 min), FL-Spike (S) and CL-Spike (S2 and S2’) were marked as indicated. Bottom panel: quantified band intensity using Image J to analyze the protein expression and the ratio of S2 and S2’ to the total S. (C) Spike-based cell-cell fusion assay. With the overexpression of TMPRSS2 (purple) in ACE2-293T, the fusion activity was quantified at different times (2, 6, 12 and 24hpt). The results of S813 S were shown as full lines and T813 S as dotted lines. (D) Representative images of cell-cell fusion. The area of cell fusion was shown as green (up) and black (bottom). Scale bar: 500 μm. (E-G) Fusion inhibition of Camostat and E-64d in ACE2-293T+TMPRSS2. A schematic diagram showing the GFP-split system with the inhibitor Camostat and E-64d for Spike-ACE2/TMPRSS2 mediated cell fusion (E). After being pre-incubated with the indicated concentration (0, 10, 50, 100 μm) of Camostat (F) and E-64d (G), the fusion activity was quantified at different times (2, 6, 12 and 24hpt), and the results of Camostat were shown as purple and E-64d as blue. (H) Spike cleavage assay by TMPRSS2. A representative blot of SARS-CoV-2 (Up panel) or SARS-CoV (Down panel) Spikes digested with TMPRSS2 at 0, 0.1, 0.3, 0.5, 1.0, 2.0 μM in assay buffer, at 37°C for 30 min, FL-Spike (S) and CL-Spike (S2 and S2’) were marked as indicated. Results are means +/- SD from at least three fields per condition. Results are representative of at least three independent experiments. Statistically significant differences (*: p<0.05) between S813 Spike and T813 Spike were determined by Student’s test at each point (C, F and G).

Fig 7. Genomic description of AA 813 in Spike of Sarbecovirus.

(A) The phylogenetic tree based on Sarbecovirus S proteins (SARS-like strains, n = 24 genomes; SARS-CoV strains, n = 13 genomes; SARS-CoV-2 strains, n = 6 genomes). All strains invariantly containing serine at position 813 were marked red while containing threonine were marked mazarine. Color coding as indicated according to species. (B) Amino acid frequency of site 813. 114 complete spike protein sequences of SARS-CoV were collected from NCBI and 10,060,583 complete spike protein sequences of SARS-CoV-2 (2020–2022) were collected from GISAID. The method of analysis was performed as previously described [74]. Briefly, After removing redundant sequences with 100% sequence identity and multiple sequence alignment (MSA), site 813 based on the reference sequence of SARS-CoV-2 was derived and the amino acid frequency of site 813 can be calculated based on the un-redundant dataset of SARS-CoV and SARS-CoV-2, respectively. (C) Phylogenetic tree of Human coronavirus. The representative strains of 7 human coronaviruses were clustered by amino acid sequence phylogeny and observed the diversity of AA 813. In α-CoV (229E and NL63), there was only Alanine, while in β-CoV, Serine was in SARS-CoV-2, MERS-CoV, OC43-CoV and HKU1-CoV; threonine only in SARS-CoV. We used the WAG+F+I+G4 optimal model of the Iqtree software to construct a phylogenetic tree based on S proteins. The right-hand sequence mapping was based on texshade software for mapping. The secondary structure, i.e. the membrane fusion region, was predicted using the PSIPRED web page. (D) Schematic summary. By modifying the usage of TMPRSS2, Spike residue 813 affects the fusogenicity of Sarbecovirus. S813 S has a high utilization and complete the membrane fusion process with a small amount of TMPRSS2, but T813 S has a low ability and needs more TMPRSS2.Although both Serine and Threonine can be found at position 813, the evolutionary trend of Sarbecovirus maybe likely Serine.