Fig 2: Knockout of DAZAP2 promotes SARS-CoV-2 infection in mouse models and human primary airway epithelial cells. (A) Schematic of the generation of Dazap2-knockout mice. Exons 2 and 3 were removed using CRISPR/Cas9, resulting in the deletion of 365 bp of coding sequence and disruption of protein function. (B, C) Female mice at 10–12 weeks old were intranasally inoculated with 1,000 focus-forming units (FFU) of mouse-adapted beta variant (B.1.351) of SARS-CoV-2 virus (MA17), and viral loads in the lungs (B) or nasal turbinates (C) at day 3 post-infection were titrated by focus-forming assay. The dashed line represents the limit of detection. (D) Mice infected with 50 FFU of MA17 virus, and viral loads in the lungs at day 1 post-infection were titrated. (E) Mice infected with 1,000 FFU (left) or 50 FFU (right) of non-adapted beta variant (B.1.351) of SARS-CoV-2, and viral loads in the lungs at day 1 post-infection were titrated. D-E, one independent experiment with a total of 4–5 mice was used. (F, G) Human ACE2 knock-in (hACE2) and hACE2 female mice with Dazap2 deletion (hACE2-Dazap2−/−) at 10–12 weeks old were intranasally inoculated with 50,000 FFU of original SH01 strain of SARS-CoV-2, and viral loads in the lungs (F) or nasal turbinates (G) were titrated at day 3 post-infection by qRT-PCR. Two independent experiments with a total of 8–11 mice were used. (H, I) mRNA detection of cytokines in the lungs harvested at day 3 post-infection from panel F. (J) Schematic of air-liquid interface cultures of human primary nasal and bronchial epithelial cells for SARS-CoV-2 infection. The figure was created with BioRender.com. (K) Editing efficiency of DAZAP2 in undifferentiated human nasal epithelial cells (HNEC) and human nasal epithelial cells (HNEC) was validated by western blotting. (L–O) SARS-CoV-2 infection in differentiated HNEC and HBEC. Viral RNA in cells was determined by qRT-PCR (L and N), and virus production in the supernatant was titrated by focus-forming assay (M and O). Unpaired two-tailed t test; n = 3; mean ± s.d. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001; ns, not significant.

Fig 3: DAZAP2 inhibits virion fusion with endolysosome membranes to release genomes into the cytosol. (A–C) Pseudovirus infection assay. The gene-edited A549-ACE2 cells were infected with murine leukemia retrovirus (MLV)-based pseudoviruses (30 µL, 24 h) bearing the spike protein of SARS-CoV-2 (A), the glycoprotein of vesicular stomatitis virus (VSV-G) (B), or the spike protein of SARS-CoV-1 (C), and the luciferase activity was measured and normalized to the control. (D, E) The inhibition of endosomal entry of SARS-CoV-2. Control and DAZAP2-knockout clonal cell line of A549-ACE2 (ΔDAZAP2) were infected with MLV-based pseudovirus bearing the spike protein of SARS-CoV-2 (30 µL, 14 h) (D) or single-cycle trVLP-NLuc (MOI 0.02, 24 h) (E), in the presence of 100 µM E-64d (aloxistatin), an inhibitor that blocks the cysteine protease activity of cathepsins B and L, which are required for the endosomal membrane fusion, and/or 100 µM camostat mesylate, a TMPRSS2 inhibitor that blocks viral fusion at the plasma membrane. (F) Virus binding and internalization assays. Cells were incubated with SARS-CoV-2 (MOI 5), and the bound or internalized virions were measured by qRT-PCR for genomic RNA. (G, H) Trafficking of SARS-CoV-2 trVLP-Nluc particles in the presence of cysteine protease inhibitor E-64d (100 µM). The representative confocal images (G) were obtained, and the quantification of spike and N protein double-positive particles per field (H) was analyzed. The endolysosome marker LAMP1 was stained. Scale bar, 20 or 5 µm. (I) Quantification of endosomal acidification. Control or DAZAP2-deficient A549-ACE2 cells were pre-treated with or without 20 µM chloroquine (CQ) followed by staining with LysoSensor Green dye. The fluorescence intensity was quantified. (J, L) The cleavage of the SARS-CoV-2 spike protein. Control or DAZAP2-deficient A549-ACE2 cells were incubated with MLV-based pseudoviruses bearing the spike protein for 0, 2, or 4 h, followed by western blotting analysis with anti-S2 antibody (J). Three independent experiments were performed, and the representative image is shown. The cysteine protease inhibitor E-64d (100 µM) was used as a control. The relative protein level of the cleaved S2’ domain was determined by normalizing the intensity of the S2’ band to the internal control GAPDH, and then to the control at 0 h (K). The ratio of S2’ to S2 was calculated similarly (L). (M) Split NanoLuc luciferase reporter-based virus-cell fusion assay. Cells expressing the LgBit were incubated with retrovirus particles encapsulated with CypA-HiBit to enable virion fusion in the endolysosomes. The re-complemented NanoLuc luciferase activity in the cytoplasm was determined and normalized to the control. (N, O) Quantification of virions in the endolysosomes. Control or DAZAP2-deficient A549-ACE2 cells were infected with SARS-CoV-2 for 4 h and fixed to stain N and endolysosome marker LAMP1. The colocalization of LAMP1 with N was visualized by confocal microscopy (N), and the number of colocalized foci per cell was counted (O). Three fields of view with a total of 27 to 42 cells were used for analysis. The representative confocal images (N) were shown. Scale bar, 20 or 5 µm. The cysteine protease inhibitor E-64d (100 µM) was used as a control. (P, Q) Quantification of double-stranded RNA (dsRNA). Control or DAZAP2-deficient A549-ACE2 cells were infected with SARS-CoV-2 for 4 h, then fixed to stain the dsRNA with J2 antibody. The percentage of dsRNA-positive cells per field was counted (P), and the dsRNA puncta were visualized by confocal microscopy (Q), and six fields of view with a total of 109 control cells and 79 DAZAP2-deficient cells were selected for analysis. The representative confocal images (Q) are shown. Scale bar, 50 µm. Data shown are from three independent experiments, and each independent experiment was performed in triplicate. (A–C, F, H, M, and P) Unpaired t test;( D, E, I, K, L, and O) one-way ANOVA with Dunnett’s test; n = 3; mean ± s.d.; *P < 0.05; **P < 0.01; ***, P < 0.001; ****P < 0.0001; ns, not significant.

Fig 4: DAZAP2 is a pan-coronavirus host restriction factor. (A) Genes identified from the CRISPR screen. ACE2-expressing A549 (A549-ACE2) cells transduced with a CRISPR knockout library were infected with SARS-CoV-2 transcription- and replication-competent virus-like particles where the N gene is replaced by the reporter GFP (trVLP-GFP) (MOI 0.5, 24 h). trVLP-GFP was packaged in cells expressing the N gene and only replicated for a single round in A549-ACE2 in the absence of N protein. GFP-positive cells were sorted for genomic extraction and sgRNA sequence analysis. The genes were analyzed by MAGeCK software and sorted based on the -log10 (MAGeCK score). (B) Experimental validation of the top 20 genes from the screen in A549-ACE2 cells. Two independent sgRNAs per gene were used, and cells were infected with SARS-CoV-2 transcription- and replication-competent virus-like particles where the N gene is replaced by the NanoLuc luciferase (trVLP-Nluc) (MOI 0.5, 24 h). The infection efficiency was quantified by measuring the luciferase activity. (C) High content imaging and quantification analysis of SARS-CoV-2 infection in DAZAP2-edited A549-ACE2 (MOI 0.1, 24 h). (D) Representative immunofluorescence images of SARS-CoV-2 infection in DAZAP2-edited A549-ACE2 cells (MOI 0.1, 24 h). Scale bar, 100 µm. (E) High content imaging and quantification analysis of SARS-CoV-2 infection in DAZAP2-edited HeLa-ACE2 (MOI 0.1, 24 h). (F) qRT-PCR was conducted to measure the N gene copies of SARS-CoV-2-infected Calu-3 cells (MOI 1, 24 h). GAPDH was used as an internal control. (G, H) High content imaging and quantification analysis (G) and immunofluorescence images (H) of SARS-CoV-2 infection in mouse Dazap2-edited MEFs expressing human ACE2 (MEF-ACE2) (MOI 0.1, 24 h). Scale bar, 100 µm. (I) Validation of DAZAP2 as a restriction factor during infection with other coronaviruses. Gene-edited cells were infected with alphacoronaviruses (HCoV-229E, MOI 1, 48 h; SADS-CoV, MOI 3, 24 h), betacoronaviruses (MHV, MOI 5, 24 h; HCoV-OC43, MOI 0.03, 12 h), gammacoronaviruses (IBV, MOI 0.5, 24 h), or deltacoronaviruses (PDCoV, MOI 0.3, 24 h). (J, K) Overexpression of DAZAP2 inhibits SARS-CoV-2 infection (MOI 1, 24 h). Human DAZAP2 or mouse Dazap2 cDNA was expressed in A549-ACE2 or MEF-ACE2, respectively. (L) Overexpression of human DAZAP2 in HeLa-ACE2 inhibits HCoV-OC43 (MOI 0.03, 24 h) and PEDV (MOI 1, 24 h) infection. The virus infection efficiency was determined by analyzing the percentage of viral N-positive cells using flow cytometry or Operetta High Content Imaging System. (M, N) Trans-complementation of DAZAP2 in DAZAP2-knockout cells recovered the antiviral effect. Clonal DAZAP2-knockout A549-ACE2 cells were generated and trans-completed with DAZAP2 cDNA. Cells were infected with SARS-CoV-2 (MOI 0.3, 24 h) for high content imaging (M) and quantification analysis (N). Data shown are from three independent experiments, and each independent experiment was performed in duplicate or triplicate. (B) Two-way ANOVA with Dunnett’s test; the mean of two sgRNAs was compared with the control sgRNA; (C, E, G, and N) one-way ANOVA with Dunnett’s test; (F and I–L) unpaired t test; n = 3; mean ± s.d.; *P < 0.05; **P < 0.01; ***, P < 0.001; ****P < 0.0001; ns, not significant.

Fig 5: DAZAP2 inhibits the plasma membrane entry of SARS-CoV-2. (A, B) Cell-cell fusion assay. Control and ΔDAZAP2 A549-ACE2 acceptor cells were co-cultured with 293T donor cells that express SARS-CoV-2 spike protein. Spike protein-induced syncytia were visualized under a brightfield microscope (A), and syncytial nuclei were counted after Giemsa staining (B). Scale bar, 100 µm. (C, D) Schematic (C) and the results (D) of the split NanoLuc luciferase reporter-based cell-cell fusion assay. A549-ACE2 acceptor cells expressing the LgBit were incubated with 293T donor cells expressing both HiBit and SARS-CoV-2 spike protein. The functional NanoLuc luciferase was re-complemented after cell-cell fusion, and the activity was measured and normalized to the control cells without spike expression. (E, F) The inhibition of plasma membrane entry of SARS-CoV-2. Control and ΔDAZAP2 A549-ACE2 cells ectopically expressing the TMPRSS2 (A549-ACE2-TMPRSS2) were infected with MLV-based pseudovirus bearing the spike protein of SARS-CoV-2 (30 µL, 14 h) (E) or single-cycle trVLP-NLuc (MOI 0.02, 24 h) (F), in the presence of cysteine protease inhibitor E-64d (100 µM), and/or TMPRSS2 inhibitor camostat mesylate (100 µM). Data shown are from three independent experiments, and each independent experiment was performed in triplicate. (B and C) Unpaired t test; (E and F) one-way ANOVA with Dunnett’s test; n = 3; mean ± s.d.; ***, P < 0.001; ****P < 0.0001; ns, not significant.