Enhanced sACE2-Fc with two single mutations exhibited broad-spectrum neutralization of SARS-CoV-2 variants.

a Schematic representation of sACE2-Fc structure (upper) and neutralization assay setup (lower). Key amino acid positions (90-92 and 374-378) involved in glycosylation and zinc binding are highlighted. Red stars mark the positions of mutations in the sACE2-Fc mutant B5-D3. SP, signal peptide; CLD, collectrin-like domain; hIgG1, human IgG1. b Comparative bar graph showing the half-maximal inhibitory concentration (IC50) values for neutralization of Wuhan-Hu-1 and D614G pseudovirues by WT sACE2-Fc and mutants (B2 to B6, A2, A3, D1 to D5, and B5-derivatives). The red arrow emphasizes the superior performance of the B5-D3 mutant. Enzymatic activity of each construct is plotted on the right axis. c List of pseudoviruses carrying spikes from different SARS-CoV-2 variants tested, categorized by the World Health Organization (WHO) into VOCs and VOIs. d Graph displaying IC50 values of WT sACE2-Fc, B5, and B5-D3/4/5 mutants against various SARS-CoV-2 VOCs and VOIs in neutralization assays. e Schematics of the plaque-reduction neutralization tests (PRNTs) process (upper) and the resulting IC50 values for B5-D3, Casirivimab, and hIgG1 against authentic SARS-CoV-2 (lower). f, g Dose-response curves depicting the neutralization efficacy of B5-D3 (orange), Casirivimab (purple), and hIgG1 (grey) in PRNTs against authentic SARS-CoV-2 Wuhan-Hu-1 and Delta strains (f), and Omicron sub-lineages (g). Data are presented as mean ± standard deviation (SD) from duplicate experiments.

Enhanced survival and reduced infection in K18-hACE2 mice through intranasal prophylaxis with B5-D3 against SARS-CoV-2.

a–e Female K18-hACE2 mice, aged 10 to 12 months, were inoculated with 1 × 104 PFU of SARS-CoV-2 (Wuhan-Hu-1 strain). Mice were treated with B5-D3 6 h prior (–6 h) via intranasal (IN, red) or intravenous (IV, green) routes, or 24 h post-infection (+24 h, blue) via IV (n = 4 + 1). IN PBS administered 6 h prior to viral challenge served as the vehicle control (black; n = 4 + 1), and PBS alone was used for mock control (grey; n = 3 + 1) (a). Body weight and survival (n = 3 or 4) were monitored over 14 days (b, c). One mouse from each group was sacrificed at 4 dpi for analysis of viral titers in lung homogenates using a median tissue culture infectious dose (TCID50) assay (d) and histological analysis of lung sections (upper, IHC staining for N protein; lower, H&E staining) (e). Black arrows indicate alveolar thickening, and yellow arrows show leukocyte infiltration. Scale bar = 100 μm. ND, not detected; LOD, limit of detection. f–i Young female K18-hACE2 mice, aged 2 to 3 months, were inoculated similarly and treated with B5-D3 via IN route at 24 h before (–24 h, pink), 6 h before (–6 h, red), or 24 h after (+24 h, orange) the viral challenge (n = 5). Mice receiving IN PBS 6 h before infection served as the vehicle control (black), with mock control mice receiving PBS alone (grey) (f). Body weight (g) and survival (h) were recorded for 14 days. Neutralizing antibody titers against Wuhan-Hu-1 in serum samples from surviving mice at 14 dpi were determined using Vero E6 cells (i). nAb, neutralizing antibody. Data are presented as the geometric mean ± geometric SD. Statistical significance was determined using Dunn’s multiple comparisons test.

Efficient viral clearance at early stages through intranasal prophylaxis with B5-D3 against SARS-CoV-2 challenge in K18-hACE2 mice.

a Workflow diagram showing timelines and treatments for different mouse groups. Young female K18-hACE2 mice aged 2 to 3 months received prophylactic administration of PBS (black), B5-D3 (red), or B5-D3-LALA (purple) via the IN route 6 h prior to inoculation with 1 × 104 PFU of Wuhan-Hu-1. Mice inoculated with PBS instead of the virus served as mock controls (grey). Mice from each treatment group were sacrificed for tissue collection at 1, 2, and 4 dpi (n = 3 per time point). b Quantitative PCR results showing relative amounts of S (upper) and N (lower) viral RNA in lung tissues collected from different groups at 1, 2, and 4 dpi, normalized to mouse Gapdh. c The titers of infectious viruses detected in lung homogenates, measured by TCID50 assays at 1, 2, and 4 dpi. d, e Fixed lung tissues were sectioned and stained; IHC for viral N protein (d) and H&E staining for tissue damage (e) are shown (scale bar = 100 μm). Data presented as mean ± standard error of the mean (SEM). Statistical significance was determined by Tukey’s multiple comparisons test.

Transcriptomic analysis of lungs revealed early immune activation in IN B5-D3-prophylaxis mouse group after SARS-CoV-2 challenge.

a–d DGE analysis comparing PBS (a), B5-D3 (c), and B5-D3-LALA (d) against the mock control at specific time points (n = 3). Volcano plots illustrate the gene expression changes (a, c, d), while red and blue dots represent significantly upregulated and downregulated genes, respectively, with |log2 fold change (log2FC)| ≥ 1 and a false discovery rate (FDR) < 0.05. Bar chart in b shows the enrichment of GOBP “response to virus” observed in PBS groups at 1, 2, 4 dpi, in which adjusted p values are indicated for individual comparisons. e–g Comparison between IN B5-D3 and PBS group at 1 dpi. Volcano plot illustrates the DGE analysis between IN B5-D3 to PBS group at 1 dpi (e), with red and blue dots representing significantly upregulated and downregulated genes, respectively, with |log2FC| ≥ 1 and FDR < 0.05. GSEA shows top 15 significantly activated GOBPs (f) and KEGG pathways (g) in IN B5-D3 compared to PBS group at 1 dpi. NES, normalized enrichment score; p.adj, adjusted p value. h–j GSEA plots of chemotaxis (h), Rap1 signaling pathway (i), and Th1 and Th2 cell differentiation (j) in B5-D3 vs PBS comparison at 1 dpi. k, l Heatmaps show NES of GSEA comparing various treatments to the mock control (k) and between B5-D3 to PBS (l), focusing on top 10 GOBPs in f and Supplementary Fig. 10c, d, respectively, and those related to immune cell chemotaxis. Significant NES values (p < 0.05, FDR < 0.25) are highlighted in yellow. Purple boxes indicate GOBPs where B5-D3 (1 dpi) group shows activation but PBS (1 dpi) group shows suppression. Benjamin–Hochberg method was used for FDR adjustment.

In vivo bio-distribution of B5-D3 after IN administration.

a Schematic workflow of in vivo and ex vivo imaging. Female K18-hACE2 mice aged 2 to 3 months received IN administration of fluorescently labeled B5-D3 (B5-D3-AF750) and was visualized at different time points. b Representative whole-body images of control and treated mice at 5 min, 1 h, and 24 h after B5-D3-AF750 administration, showing the signal captured by in vivo imaging (left). White circles indicate regions of interest (ROIs) for quantification of fluorescence signals in the nasal cavities. Average (Avg) Radiance measured at all time points are shown on the right. c Ex vivo images of tissues from control and treated mice sacrificed at indicated time points after B5-D3-AF750 administration. Blue circles indicate ROIs for signal quantification. Br, brain; NC, nasal cavity; T, trachea; Lu, lung; H, heart; Lv, liver; S, spleen; K, kidney; UB, urinary bladder; Bl, blood; Ur, urine. d Avg Radiance shows the fluorescence signals in excised tissues measured ex vivo. e Schematic workflow for BALF analysis. Female K18-hACE2 mice aged 2 to 3 months received IN administration of B5-D3-AF750 (n = 3) or PBS (n = 4) and were sacrificed at 6 h later for collection of BALF cells. f Percentage of CD45+ cells in live BALF cells. g Positive rates (left) and histograms (right) of B5-D3 binding/uptake in CD45+ BALF cells. Histograms show B5-D3-AF750 fluorescence intensities in CD45+ BALF cells from individual mice. h Frequency of individual immune cell types in CD45+B5-D3+ BALF cells. Red arrows point out AMs and mono-Macs with high abundance. AM, alveolar macrophage; Mono-Mac, monocyte-derived macrophage; cDC1/2, type 1 or 2 conventional dendritic cells. i, j Positive rates (left) and histograms (right) of B5-D3 binding/uptake in CD11c+Siglec-F+ AMs (i) and CD11b-F4/80+ mono-Macs (j). k Median fluorescence intensity (MFI) of AF750 indicates B5-D3 binding/uptake in different CD45+B5-D3+ populations. l Confocal images (scale bar = 50 μm) of BALF cells collected at 6 h and stained for sACE2-Fc (red, anti-Fc, Abcam #ab98596), Siglec-F (green, BD #564514), and nuclei (blue, Hoechst). Magnified views are shown in white rectangles. Data are presented as mean ± SEM, and statistical significance was determined by Tukey’s multiple comparisons test or Student’s t-test.

B5-D3 enhanced phagocytosis and degradation of SARS-CoV-2 pseudovirus in THP-1-derived macrophages.

a Immunostaining of p24 (Invitrogen #PA5-81773), sACE2-Fc, and LAMP1 (Abcam #ab25630) in THP-1-differentiated M0 macrophages showing phagocytosis of SARS-CoV-2 pseudovirus (PV, p24+) after 6 h of incubation with or without B5-D3 (scale bar = 50 µm). LAMP1 was stained to identify lysosomes. b Quantification of p24 signal intensity as shown in a. Intensity Density (IntDen) per cell number indicates the mean p24 signal per cell, calculated using ImageJ. Each dot represents one image. c Manders’ coefficient indicating the colocalization of p24 and LAMP1 in THP-1 M0 macrophages as shown in a. d Immunostaining of p24, sACE2-Fc, and LAMP1 in hACE2-Calu-3 cells after 6 h incubation with pseudovirus, with or without B5-D3 (scale bar = 50 µm). e Quantification of mean p24 signal intensity as shown in d. f Manders’ coefficient for the colocalization of p24 and LAMP1 in hACE2-Calu-3 cells, as shown in d. g Quantification of pseudovirus infection in THP-1, M0 macrophages, M1 macrophages, hACE2-Calu-3, and hACE2-293T cells, in the presence or absence of B5-D3. Results shown are luciferase activities measured at 2 days post-transduction. h Immunoblot staining of cell lysates to detect SARS-CoV-2 spike cleavage after cell entry. M0 macrophages, M1 macrophages, and hACE2-293T cells were incubated with pseudovirus for 6 h, with or without B5-D3, before protein extraction. Band locations of SARS-CoV-2 spike S2 and S2′ fragments are labeled in black and red respectively. Data are presented as mean ± SEM, and statistical significance was determined by Tukey’s multiple comparisons test.

Proposed mechanisms of action of IN sACE2-Fc decoy in preventing SARS-CoV-2 infection.

a, b Schematics illustrating the actions and outcomes of SARS-CoV-2 infection, in the absence (a) and presence (b) of IN delivered sACE2-Fc decoys. The figure was created in BioRender.com.

Establishment of the pseudoviral infection platform and generation of sACE2 decoys.

a Microscopic images show the expression of full length human ACE2 (hACE2) in stable hACE2-293T cells established by lentiviral transduction (Scale bar = 100 μm). Immunostaining was performed using antibody specific to hACE2 (Abcam # ab15348). b Representative fluorescence and phase contrast images showing GFP expression in hACE2-293T cells with and without infection by pseudovirus carrying the Wuhan-Hu-1 spike protein (Scale bar = 100 μm). c Schematic diagrams of hACE2 (top), sACE2 (middle), and sACE2-Fc (bottom) molecules indicating important amino acid positions (90-92 and 374-378) for glycosylation and zinc binding. aa, amino acid. d Line chart comparing the neutralization efficiencies of sACE2 (green) and sACE2-Fc (orange) against Wuhan-Hu-1 pseudovirus expressing luciferase, measured in relative luminescence units (RLU). Data are presented as mean ± SD from duplicate experiments.

Engineering and characterization of enhanced sACE2 decoys.

a 3D structural model of ACE2 (green) complexed with the SARS-CoV-2 spike receptor-binding domain (RBD, brown), highlighting mutations for spike-binding enhancement (magenta) and enzymatic inactivation (blue). Structures were adapted from Protein Data Bank (PDB ID: 6M0J). b List of mutations in sACE2 sequences tested for enhanced binding or enzymatic inactivation. c, d Neutralization assay results for WT sACE2-Fc and mutants (B2 to B6, A2, A3, and D1 to D5) against Wuhan-Hu-1 (c) and D614G (d) pseudoviruses. e Kinetic curves showing the ACE2 enzymatic activities of WT sACE2-Fc and B2 to B6, A2, A3, D1 to D5 mutants. f, g Neutralization results for WT sACE2-Fc and mutants (B5, B5-A2, B5-A3, B5-D1, B5-D3, B5-D4 and B5-D5) against Wuhan-Hu-1 (f) and D614G (g) pseudoviruses. h Kinetic curves showing the ACE2 enzymatic activities of WT sACE2-Fc and B5, B5-A2, B5-A3, B5-D1, B5-D3, B5-D4 and B5-D5 mutants. i Conformational comparison between WT and B5-D3 sACE2. The 3D structures of WT and B5-D3 sACE2 (aa 18–740) were predicted using AlphaFold 3 and superimposed for direct comparison. The WT sACE2 structure is coloured in grey, while the B5-D3 variant is highlighted in yellow. Mutations in the B5-D3 structure are specifically marked in red. Data are presented as mean ± SD from duplicate experiments.

Broad-spectrum neutralization against pseudoviruses of SARS-CoV-2 VOC/VOI strains by sACE2-Fc candidate mutants.

Graphical representation of dose-response curves in neutralization assays against pseudoviruses bearing spikes from various SARS-CoV-2 VOCs and VOIs. Data are presented as mean ± SD from duplicate experiments.

AAV-delivered prolonged overexpression of WT sACE2-Fc and candidate mutants in K18-hACE2 mice.

a Schematic of AAV administration. Male K18-hACE2 mice aged 2 months received tail vein injections of either PBS (black; n = 7) or AAV carrying WT sACE2-Fc (blue), B5-D3 (orange), B5-D4 (dark grey), or B5-D5 (light grey) (n = 4 each) at a dose of 1 × 1011 GC. Mice were observed for up to 15 weeks and then sacrificed for tissue analysis. wpi, weeks post-injection; M, month. b Serum concentrations of sACE2-Fc were quantified via ELISA using antibodies specific to hIgG1. c Quantification of AAV genomes in the mouse livers at the observation endpoint. Results shown are from qPCR of genome DNA normalized to mouse Gapdh. d–f Concentrations of renin (d), Ang II (e), and Ang (1-7) (f) in sera at 2 wpi, measured via metabolite-specific ELISA. ns, not significant with p value (p) ≥ 0.05. g Representative H&E staining of heart, lung, liver, kidney, spleen, ileum, and colon tissues from mice in the PBS, WT sACE2-Fc, and B5-D3 treatment groups (scale bar = 50 μm). Data are presented as mean ± SEM. Statistical analyses were performed using one-way analysis of variance (ANOVA) and Tukey’s multiple comparisons test following ANOVA.

Comparison between B5-D3 and B5-D3-LALA in in vitro neutralization against the SARS-CoV2 pseudovirus and Fc-mediated effector functions.

a Neutralization capability of B5-D3-LALA was compared to B5-D3 using an in vitro pseudovirus neutralization assay. IC50 values are indicated for each version. b, c Fc-mediated effector functions of B5-D3 and B5-D3-LALA were assessed using luciferase-based reporter cell lines. Dose-response curves and half maximal effective concentration (EC50) values for antibody-dependent cellular cytotoxicity (ADCC) (b) and antibody-dependent cellular phagocytosis (ADCP) (c) are displayed. Data are presented as mean ± SD from duplicate experiments.

Entire sections for histological examination at 1 dpi.

Fixed lung tissues were sectioned and stained for histological examination. IHC staining for viral N protein (left panels, SinoBiological # 40143-T62) and H&E staining for tissue damage (right panels) are displayed. Black arrows in H&E photos indicate areas of alveolar thickening. Scale bar = 500 μm.

Thickening of alveolar septum in K18-hACE2 mice after SARS-CoV-2 challenge.

Lung septum thickness of K18-hACE2 mice in Fig. 3 was measured from H&E staining. Data are presented as mean ± SEM, and statistical significance was determined by Tukey’s multiple comparisons test.

RNA-Seq analysis of K18-hACE2 mouse lungs with different pretreatments upon SARS-CoV-2 challenge.

a Bulk RNA-seq was conducted on lung homogenates from K18-hACE2 mice under mock treatment (n = 6) or post-viral challenge at 1, 2 and 4 dpi with each time point (n = 3). Pearson correlation analysis was executed on 27 samples using counts per million (CPM) data, with each cell displaying the Pearson correlation coefficient color-coded for visual ease. b–f GOBP enrichment analysis identifies biological processes enriched in up-regulated genes from comparisons at 1 (b), 2 (c), and 4 dpi (d) for PBS, 4 dpi for B5-D3 (e), and 4 dpi for B5-D3-LALA (f) versus the mock control, with top 15 significant terms displayed. p.adjust, adjusted p value; CAMKK, calmodulin-dependent protein kinase; AMPK, 5’ adenosine monophosphate-activated protein kinase. Benjamin–Hochberg method was used for FDR adjustment.

Leading-edge subsets in GSEA.

a–c Z-score plots show the relative level of gene expression in the leading edge subsets from GSEA comparing B5-D3 vs PBS at 1 dpi, corresponding to chemotaxis (a), Rap1 signaling pathway (b), and Th1 and Th2 differentiation (c).

Transcriptomic comparisons between B5-D3 and PBS pretreatments in K18-hACE2 mice upon SARS-CoV-2 challenge.

a, b DGE analysis between B5-D3 and PBS groups at 2 (a), and 4 dpi (b) showing up-regulated and down-regulated genes visualized in volcano plots. c, d GSEA of GOBPs significantly activated in B5-D3 groups compared to PBS groups at 2 dpi (c) and 4 dpi (d), with top 15 most significant terms displayed. e, f GSEA of KEGG pathways significantly activated in B5-D3 groups compared to PBS groups at 2 dpi (e) and 4 dpi (f), with all significant terms (e) and top 15 most significant terms (f) displayed. SFTS, Severe Fever with Thrombocytopenia Syndrome. Benjamin–Hochberg method was used for FDR adjustment.

Minimal transcriptomic alterations in lungs after IN B5-D3 administration without viral challenge.

a Pearson correlation analysis for lung tissues collected 4 days post-administration of IN B5-D3 or PBS in female K18-hACE2 mice (n = 3), depicting correlation coefficients. b Volcano plot showing differentially expressed genes between the B5-D3 treated and control groups, with significant changes marked. c Venn diagram showing overlaps among the upregulated genes in b and in B5-D3 (4 dpi) in Fig. 4c.

Flow cytometry analysis of mouse BALF cells.

a Flow cytometric gating strategy for BALF cells. b Percentages of individual cell types in CD45+ BALF cells collected from individual animals. Ctrl group received no treatment before sacrifice.

Binding/uptake rates of B5-D3-AF750 in BALF cells.

a–f Positive rates (left) and histograms (right) of B5-D3 binding/uptake as indicated by B5-D3-AF750 fluorescence intensities in CD11b+Ly6C+Ly6G+ neutrophils (a), CD11b+MHC-Ⅱ+ cDC2 (b), CD11b+MHC-Ⅱ- monocytes (c), CD11b-CD11c+MHC-Ⅱ+ cDC1 (d), CD11b-CD11c-MHC-Ⅱ+ B cells (e), and CD3+ T cells (f). B5-D3+ rates from individual mice are indicated on histograms. Data are presented as mean ± SEM, and statistical significance was determined by Student’s t-test.

Time-course analysis of sACE2-Fc-dependent pseudovirus entry in THP-1 cells.

a Representative image showing B5-D3-mediated phagocytosis of pseudovirus by THP-1 monocytes at various time points (1, 3, 6, and 18 h). Cells were incubated with pseudovirus and B5-D3, followed by immunostaining for p24 (red, Invitrogen # PA5-81773). b Quantification of mean p24 signal intensity per cell as shown in a. IntDen per cell number indicates the average p24 signal per cell, analyzed using ImageJ. Each dot represents one image.

Enhanced phagocytosis of SARS-CoV-2 pseudovirus by THP-1 and THP-1-derived macrophages facilitated by sACE2-Fc.

a, d Representative images illustrating phagocytosis of SARS-CoV-2 pseudovirus by THP-1 monocytes (a) and THP-1 differentiated M1 macrophages (d) after 6 hours of incubation with or without sACE2-Fc (scale bar = 50 µm). p24 (Invitrogen # PA5-81773) and LAMP1 (Abcam # ab25630) were used to identify the pseudovirus and lysosomes, respectively. b, e Quantification of mean p24 signal intensity per cell for THP-1 monocytes (b) and M1 macrophages (e). IntDen per cell number indicates the mean p24 signal per cell, analyzed using ImageJ. Each dot represents one image. c, f Manders’ coefficient demonstrating the colocalization of p24 and LAMP1 in THP-1 monocytes (c) and M1 macrophages (f). Data are presented as mean ± SEM, and statistical significance was determined by Tukey’s multiple comparisons test.

Generation of Calu-3 cell overexpressing hACE2 for enhanced pseudoviral infection.

a Immunostaining of hACE2 (Abcam # ab15348) in hACE2-Calu-3 and control Calu-3 cells. b Infection levels of SARS-CoV-2 pseudovirus in hACE2-Calu-3 versus control Calu-3 cells, quantified by luciferase assay 72 hours post-infection, performed in duplicate. Data are presented as mean ± SEM, and statistical significance was determined by Šídák’s multiple comparisons test.

Reduced uptake of SARS-CoV-2 pseudovirus in THP-1-derived macrophages due to malfunction or absence of Fc domain in B5-D3.

a Representative images illustrating uptake of SARS-CoV-2 pseudovirus (PV) by THP-1-derived M1 macrophages at 6 h after incubation with B5-D3, B5-D3-LALA, and hIgG1 isotype with or without PV. Scale bar = 50 µm. b Quantification of mean p24 signal intensity per cell. IntDen per cell number indicates the mean p24 signal per cell, analyzed using ImageJ. Each dot represents one image. Data are presented as mean ± SEM, and statistical significance was determined by Tukey’s multiple comparisons test.

Transcriptomic analysis revealed activation of THP-1-derived macrophages mediated by 6 h incubation with B5-D3-pseudovirus complex.

a DGE analysis between THP-1-derived M0 macrophages incubated with B5-D3 + pseudovirus (PV) and those incubated with PV only (incubation time = 6 h). b GSEA of GOBPs significantly altered in the B5-D3 + PV group compared to the PV group, with top 15 most significantly activated and top 15 most significantly suppressed terms displayed. c GSEA of KEGG pathways significantly altered in the B5-D3 + PV group compared to the PV group, with top 15 most significantly activated and top 15 most significantly suppressed terms displayed. d–i GSEA plots of response to virus (d), response to type I interferon (e), JAK-STAT signaling pathway (f), MAPK signaling pathway (g), PI3K-Akt signaling pathway (h), and Rap1 signaling pathway (i) in B5-D3 + PV vs PV comparison. Benjamin–Hochberg method was used for FDR adjustment.

Enhanced tolerance and stability of B5-derivatives compared to WT sACE2-Fc in AAV-administered K18-hACE2 mice.

Detailed post hoc comparisons among treatment groups shown in Supplementary Fig. 4b. Diff., difference; **, 0.001 ≤ p < 0.01; *, 0.01 ≤ p < 0.05. p values were determined by Tukey’s multiple comparisons test.

Primers used in quantitative PCR.