DAZAP2 functions as a pan-coronavirus restriction factor by inhibiting viral entry and genomic replication

ABSTRACT

The SARS-CoV-2 pandemic and the emergence of novel variants underscore the need to understand host-virus interactions and identify host factors that restrict viral infection. Here, we perform a genome-wide CRISPR knockout screen to identify host restriction factors for SARS-CoV-2, revealing DAZAP2 as a potent antiviral gene. DAZAP2, previously implicated in SARS-CoV-2 restriction, is found to inhibit viral entry by blocking virion fusion with both endolysosomal and plasma membranes. Additionally, DAZAP2 suppresses genomic RNA replication without affecting the primary translation of viral replicases. We demonstrate that DAZAP2 functions as a pan-coronavirus restriction factor across four genera of coronaviruses. Importantly, knockout of DAZAP2 enhances SARS-CoV-2 infection in mouse models and in human primary airway epithelial cells, confirming its physiological relevance. Mechanistically, the antiviral activity of DAZAP2 appears to be indirect, potentially through the regulation of host gene expression, as it primarily localizes to the nucleus. Our findings provide new insights into the host defense system against coronaviruses and highlight DAZAP2 as a potential target for host-directed antiviral therapies.

IMPORTANCE

During viral infection, the host defense response is mediated by a variety of host factors through distinct mechanisms that have yet to be fully elucidated. Although DAZAP2 was previously implicated in SARS-CoV-2 restriction, its mechanisms of action and in vivo relevance remain unclear. In this study, we identify DAZAP2 as a potent pan-coronavirus restriction factor that inhibits viral infection through dual mechanisms: blocking virion fusion with both endolysosomal and plasma membranes, and suppressing genomic RNA replication. We confirm its physiological relevance in host defense using mouse models and primary cell cultures. This study advances our understanding of host-pathogen interactions. Targeting DAZAP2 or its regulatory pathways could provide a new approach to enhance host defense against current and future coronavirus threats.

INTRODUCTION

Coronaviridae is a large family of enveloped, positive-sense, single-stranded RNA viruses and consists of four genera: Alphacoronavirus, Betacoronavirus, Gammacoronavirus, and Deltacoronavirus. Coronaviruses infect a variety of hosts, including humans, pigs, chickens, and other animals. One such virus, SARS-CoV-2, is the causative agent of the COVID-19 pandemic (1, 2). There are seven widely recognized human coronaviruses, and although four of them (HCoV-229E, HCoV-NL63, HCoV-OC43, and HCoV-HKU1) cause the common cold, three (MERS-CoV, SARS-CoV, and SARS-CoV-2) emerged as major public health concerns in this century by causing severe infection with high morbidity and mortality (3–5). In addition, some animal coronaviruses, such as porcine epidemic diarrhea virus (PEDV), swine acute diarrhea syndrome coronavirus (SADS-CoV), porcine deltacoronavirus (PDCoV), and infectious bronchitis virus (IBV), have led to significant economic loss in the agricultural industry (6). Closely related to a bat coronavirus, SADS-CoV was first identified in pigs and can infect primary human lung and intestinal cells (7). Furthermore, infections with PDCoV have been detected among young children in Haiti (8). Thus, coronaviruses pose a broad and considerable threat to public health, social stability, and the economy.

Coronaviruses have large RNA genomes of 28–31 kb, encoding for multiple structural and nonstructural proteins that have conserved functions in the infection cycles of different coronaviruses (4). As a result, coronaviruses usually employ and share the same host machinery to facilitate infection and, conversely, can be met with similar host defense responses to restrict the infection. A great effort has been made by scientists to understand the anti-coronavirus host response, particularly to identify restriction factors of SARS-CoV-2. Interferons (IFNs) play a key role during virus infection, and IFN-stimulated genes (ISGs) often possess antiviral activities (9, 10). ISGs, such as LY6E, BST2, DAXX, OAS1, and many others against SARS-CoV-2 infection, have been identified either by arrayed ISG cDNA overexpression (11–13) or by focused CRISPR knockout or activation library of ISGs (14, 15). In addition to ISGs, host restriction factors, like mucins, are identified for SARS-CoV-2 by genome-scale CRISPR activation strategy (16, 17). MHC class II transactivator CIITA-induced cell resistance to SARS-CoV-2 infection has been reported by using a transposon-mediated gene-activation screen (18). Moreover, genome-wide association studies as well as transcriptomic and proteomic analyses of SARS-CoV-2-infected cell lines and patient samples have uncovered additional host factors that may restrict or confer resistance to viral infection (19–28).

To comprehensively identify host restriction factors of SARS-CoV-2, we conducted a genome-wide CRISPR/Cas9 knockout screen coupled with fluorescence-activated cell sorting (FACS) to enrich for susceptible cells. This approach revealed a suite of host factors with antiviral activity, including DAZ-associated protein 2 (DAZAP2), which was previously identified as an antiviral factor in a CRISPR dropout screen by analyzing the depleted cells (29). Although DAZAP2 was shown to restrict SARS-CoV-2 infection, potentially through regulation of SERPINE1 expression (29), how it impacts the life cycle of virus infection and whether it functions in vivo remain to be elucidated. In this study, we demonstrate that DAZAP2 is a pan-coronavirus restriction factor that acts at the stages of viral entry and replication. Importantly, we validated its antiviral activity in mouse models and human primary airway epithelial cells, where knockout of DAZAP2 significantly enhanced SARS-CoV-2 infection. These findings provide insights into the host defense mechanisms against coronaviruses and highlight DAZAP2 as a potential target for broad-spectrum antiviral strategies.

RESULTS

DAZAP2 is a pan-coronavirus host restriction factor

To identify host factors that can restrict SARS-CoV-2 infection, we performed a genome-wide, cell sorting-based screen in ACE2-expressing A549 (A549-ACE2) transduced with a knockout library of single-guide RNAs (sgRNAs) targeting 19,114 human genes (30). The rationale is that the knockout of antiviral host genes will enhance the virus infection. The library-transduced cells were infected with transcription- and replication-competent SARS-CoV-2 virus-like particles where the N protein was replaced with GFP (SARS-CoV-2 trVLP-GFP), resulting in only single-cycle infection (31). Unlike previous studies that analyzed depleted cells to identify antiviral genes (29), we sorted virus-infected GFP-positive cells for genomic DNA extraction, sgRNA sequencing, and data analysis (Table S1). The genes identified from the screen were ranked based on the MAGeCK score (Fig. 1A). The top hit was programmed cell death 10 (PDCD10 or CCM3), a gene associated with apoptosis (32, 33), followed by DAZAP2, which encodes a multifunctional proline-rich protein involved in cell signaling (34, 35), transcription regulation (36, 37), and stress granule formation (38). Other highly ranked genes included calcium-binding protein 39 (CAB39 or MO25), phospholipid scramblase 1 (PLSCR1), vesicle trafficking 1 (VTA1 or LIP5), and lymphocyte antigen 6E (LY6E). DAZAP2 and VTA1 were previously identified as SARS-CoV-2 restriction factors in a CRISPR dropout screen, potentially through regulation of SERPINE1 expression (29). PLSCR1 is an antiviral gene strongly induced by virus infection or interferon treatment (39–41) and was previously identified as a host factor associated with critical COVID-19 (27). PLSCR1 is also reported to restrict the SARS-CoV-2 infection at a late entry step before viral RNA is released into host cells (42). CAB39, a component of a trimeric complex including serine/threonine kinase 11 (STK11) and STE20-related adaptor alpha (STRAD), has been studied in the context of cancer progression (43, 44), but with unknown function during SARS-CoV-2 infection. VTA1 is involved in the trafficking of endosomal multivesicular bodies and retrovirus budding (45).

Fig 1.

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.

To validate the candidate hits, we chose the top 20-ranked genes. For each specific gene target, A549-ACE2 cells were edited with two independent sgRNAs. We also modified the transcription- and replication-competent SARS-CoV-2 virus-like particle (SARS-CoV-2 trVLP) where the N gene was replaced by NanoLuc luciferase instead of GFP (SARS-CoV-2 trVLP-Nluc) (Fig. S1). Thus, infection efficiency could be assessed by quantifying luciferase activity over time. Knockout of the antiviral gene PLSCR1 had the greatest effect, resulting in an 8-fold increase of infectivity when compared with the control (Fig. 1B). DAZAP2 knockout increased infectivity by over 4-fold. Knockout of CAB39, PDCD10, VPS28, VPS39, VTA1, ZNF28, or UQCRH each resulted in up to 2-fold increases in infectivity compared with control.

Of the validated genes, DAZAP2 strongly restricted SARS-CoV-2 infection, consistent with previous findings (29). However, how DAZAP2 impacts the life cycle of coronavirus infection remains to be elucidated. We therefore focused the rest of our study on DAZAP2, attempting to understand its role in coronavirus infection. We first confirmed the antiviral activity of DAZAP2 in the context of infection with authentic SARS-CoV-2. Two independent sgRNAs were used to knock out DAZAP2 in A549-ACE2 cells. The infection efficiency was increased by approximately 10-fold, and the representative images of virus infection were indicated (Fig. 1C and D). Similarly, SARS-CoV-2 infection in ACE2-expressing HeLa cells (HeLa-ACE2) was increased by approximately 5-fold when DAZAP2 was edited (Fig. 1E). We further validated the role of DAZAP2 as a restriction factor in Calu-3 cells, a physiologically relevant lung epithelial cell line, by quantifying viral RNA levels in cells (Fig. 1F). Additionally, knockout of mouse orthologue Dazap2 in murine embryonic fibroblasts expressing human ACE2 (MEF-ACE2) significantly enhanced SARS-CoV-2 infection (Fig. 1G and H). Knockout efficiency of DAZAP2 was confirmed by western blotting in all cell lines (Fig. S2A).

To determine whether DAZAP2 acts as a pan-coronavirus restriction factor, we infected DAZAP2-edited A549-ACE2, HeLa-ACE2, and MEF-ACE2 with viruses from all four Coronaviridae genera: the alphacoronaviruses HCoV-229E and SADS-CoV; the betacoronaviruses HCoV-OC43 and mouse hepatitis virus (MHV); the gammacoronavirus IBV; and the deltacoronavirus PDCoV. DAZAP2 knockout significantly enhanced infection by all tested coronaviruses (Fig. 1I). Conversely, overexpression of either human or mouse DAZAP2 suppressed SARS-CoV-2 infection (Fig. 1J and K). Protein expression of exogenous human DAZAP2 cDNA was verified by western blotting (Fig. S2B). Similarly, overexpression of human DAZAP2 inhibited infection by the betacoronavirus HCoV-OC43 and the alphacoronavirus porcine epidemic diarrhea virus (PEDV) (Fig. 1L). To rule out the off-target effect, we generated and validated a DAZAP2-knockout clonal cell line (ΔDAZAP2) of A549-ACE2 (Fig. S2C to E). Trans-complementing the DAZAP2 cDNA in knockout cells recovered its antiviral effect (Fig. 1M and N). Together, these data demonstrate that DAZAP2 is a conserved, pan-coronavirus restriction factor that functions across cell types and species.

DAZAP2 inhibits the endosomal entry of SARS-CoV-2

Having established DAZAP2 as a common restriction factor for coronaviruses across four genera, we next sought to define its role in the coronavirus life cycle. To determine the stage of infection at which DAZAP2 acts, we infected DAZAP2-edited cells with murine leukemia retrovirus (MLV)-based pseudoviruses bearing the spike protein of SARS-CoV-2 or, as a control, the glycoprotein of vesicular stomatitis virus (VSV-G). Knockout of DAZAP2 resulted in an approximately 25-fold increase in SARS-CoV-2 pseudovirus infection, whereas no significant difference was observed for the VSV-G pseudovirus (Fig. 2A and B). Similarly, pseudoviruses packaged with the spike from SARS-CoV-1 exhibited about a 9-fold increase in infection, consistent with the results for SARS-CoV-2 pseudoviruses (Fig. 2C).

Fig 2.

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.

Since A549 cells express little or no TMPRSS2, a serine protease that facilitates viral entry via plasma membrane fusion, SARS-CoV-2 primarily enters A549-ACE2 cells through the endosomal pathway. To test whether DAZAP2 inhibits endosomal entry, control or ΔDAZAP2 cells were infected with SARS-CoV-2 pseudovirus in the presence of E-64d (aloxistatin), an inhibitor of the cysteine proteases cathepsins B and L (required for endosomal membrane fusion), and/or camostat mesylate, a TMPRSS2 inhibitor that blocks plasma membrane fusion (Fig. 2D). We found that the enhancement of pseudovirus infection in ΔDAZAP2 cells is significantly diminished in the presence of E-64d, whereas camostat mesylate had no effect. Similar results were observed when cells were infected with single-cycle SARS-CoV-2 trVLP-Nluc particles. These findings suggest that DAZAP2 primarily inhibits the endosomal entry pathway of SARS-CoV-2.

DAZAP2 inhibits virion fusion with endolysosome membranes to release genomes into the cytosol

To further dissect the SARS-CoV-2 entry process, we divided viral entry into four distinct steps: (i) binding, (ii) uptake (internalization), (iii) virion trafficking to late endosome/lysosomes, and (iv) virion fusion with the endolysosome membranes to release viral genomes into the cytosol. A previous study indicated that DAZAP2 does not affect the binding and internalization of SARS-CoV-2 (29). To confirm this, we incubated virions on ice for 45 min with control and ΔDAZAP2 cells. After washing, cells were lysed for qRT-PCR to assess the cell surface-bound virions. For internalization, cells were shifted to 37°C for 45 min after the binding and washing, to allow virion uptake. Uninternalized virions on the cell surface were removed by treating cells with protease K. After washing, cells were lysed for qRT-PCR to measure the internalized virions. Consistent with prior findings, neither binding to the cell surface nor internalization was affected in DAZAP2-deficient cells (Fig. 2F).

Next, we investigated whether DAZAP2 influences the trafficking of internalized virions to endolysosomes. After internalization at 37°C for 4 h in the presence of E-64d (to prevent virion fusion with endolysosome membranes), we quantified virions (spike-positive and nucleocapsid-positive) co-localized with the endolysosome marker LAMP1. No significant difference in virion counts was observed between control and ΔDAZAP2 cells (Fig. 2G and H).

We then examined whether DAZAP2 affects virion fusion with endolysosomes, a process requiring the activation of cysteine proteases (cathepsins B and L) in a low pH environment. Using LysoSensor Green dye, we found that endolysosomal acidification was unchanged in ΔDAZAP2 cells (Fig. 2I). Similarly, the cleavage of the spike protein into S2’ by cysteine proteases was unaffected in ΔDAZAP2 cells upon incubation with pseudovirus particles (Fig. 2J to L). These results demonstrate that DAZAP2 does not impact SARS-CoV-2 endosomal entry at the stages of binding, uptake, trafficking, or spike cleavage.

The final step in endosomal entry is the fusion of virions with endolysosome membranes to release viral genomes into the cytosol. To assess this, we employed an improved virus-cell fusion assay (42). Cyclophilin A (CypA), a Gag-interacting protein, was fused with HiBit and encapsulated into retrovirus particles bearing the SARS-CoV-2 spike protein. If internalized virions fuse with endolysosome membranes, the released HiBit into the cytosol complements LgBit to form functional NanoLuc luciferase. When ΔDAZAP2 cells expressing LgBit were incubated with these particles, luciferase activity was significantly higher compared with control cells (Fig. 2M), indicating enhanced fusion in ΔDAZAP2 cells.

To further validate these findings, control and ΔDAZAP2 cells were infected with authentic SARS-CoV-2 for 4 h, followed by detection of virion particles co-localized with endolysosome marker LAMP1 (Fig. 2N). As expected, fewer virions were observed in ΔDAZAP2 cells when compared with controls (Fig. 2N and O). Conversely, intracellular double-stranded RNA (dsRNA), an indicator of viral replication intermediates, was more abundant in ΔDAZAP2 cells (Fig. 2P and Q). As a control, the addition of E-64d resulted in similar numbers of virions in endolysosomes for both cell types (Fig. 2N and O). These findings collectively suggest that DAZAP2 inhibits the fusion of SARS-CoV-2 virions with endolysosome membranes, thereby preventing the release of viral genomes into the cytoplasm.

DAZAP2 restricts the plasma membrane entry of SARS-CoV-2

We next investigated whether DAZAP2 affects spike-protein-mediated fusion at the plasma membrane. Control and ΔDAZAP2 A549-ACE2 acceptor cells were co-cultured with 293T donor cells expressing the SARS-CoV-2 spike protein, and spike-mediated cell-cell syncytia formation was visualized. Notably, ΔDAZAP2 cells exhibited a larger fusion area and more syncytial nuclei compared with control cells (Fig. 3A and B).

Fig 3.

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.

To quantify the cell-cell fusion, we employed a split NanoLuc luciferase-based assay as previously described (42). In this system, A549 acceptor cells express the LgBit fragment, whereas the 293T donor cells express the HiBit fragment. Upon co-culture and spike-mediated cell-cell fusion, the two fragments complement each other to form functional NanoLuc luciferase, allowing for quantification of fusion activity (Fig. 3C). Consistent with the syncytia formation results (Fig. 3A and B), we observed a significant increase in luciferase activity in ΔDAZAP2 cells, indicating enhanced cell-cell fusion (Fig. 3D).

Because A549 cells express minimal or no TMPRSS2 protease that is required for spike-mediated plasma membrane entry, we generated A549-ACE2-TMPRSS2 cells (control and ΔDAZAP2) to further explore this pathway. These cells were infected with SARS-CoV-2 pseudovirus in the presence of E-64d (to block endosomal entry) and/or camostat mesylate (a TMPRSS2 inhibitor) (Fig. 2D). In TMPRSS2-expressing cells, E-64d failed to inhibit infection, whereas camostat mesylate significantly reduced the enhanced infection observed in ΔDAZAP2 cells (Fig. 3E). Similar results were obtained using single-cycle SARS-CoV-2 trVLP-Nluc particles (Fig. 3F). These results demonstrate that DAZAP2 inhibits the SARS-CoV-2 spike protein-mediated cell-cell fusion and plasma membrane entry, further underscoring its role as a broad-spectrum restriction factor.

DAZAP2 inhibits the genomic replication of SARS-CoV-2

The SARS-CoV-2 trVLP particles, which lack the N gene, recapitulate only the viral entry and replication stages of the life cycle, excluding virion assembly and release. Consequently, the CRISPR screen using SARS-CoV-2 trVLP is biased toward identifying host factors involved in entry and replication. In addition to its role in inhibiting viral entry, we sought to determine whether DAZAP2 also affects post-entry steps, particularly viral genomic replication.

To investigate this, we constructed a SARS-CoV-2 replicon system in which the portion of the genome encoding the spike protein all the way through ORF8 was replaced by NanoLuc luciferase (Fig. S1). The in vitro-transcribed replicon RNA was electroporated into cells, and viral replication was monitored by measuring the luciferase activity. Knockout of DAZAP2 led to significantly enhanced viral replication when compared with control cells (Fig. 4A). Treatment with remdesivir, an RNA-dependent RNA polymerase (RdRp) inhibitor, diminished the luciferase activity in both control and DAZAP2-edited cells, confirming the utility of this system to assess changes in replication (Fig. 4A). Conversely, overexpression of DAZAP2 could significantly inhibit the replication (Fig. 4B).

Fig 4.

DAZAP2 inhibits the genomic replication of SARS-CoV-2. (A) Replicon RNA assay in HeLa cells edited with control or DAZAP2 sgRNA. The SARS-CoV-2 replicon system was constructed by replacing the portion of the genome encoding the spike protein all the way through ORF8 with NanoLuc luciferase. The in vitro-transcribed replicon RNA was electroporated into cells. The RNA-dependent RNA polymerase (RdRp) inhibitor remdesivir (10 µM) was added as a control to verify the utility of the replicon system. One representative sgRNA was used to edit DAZAP2. The luciferase activity was determined and normalized to the control. (B) Replicon RNA assay in empty vector- and DAZAP2-overexpressing HeLa cells, and the results were normalized to the control. (C) Quantification of genomic RNA replication. The in vitro-transcribed replicon RNA was electroporated into cells, and the levels of genomic RNA replication were determined by qRT-PCR targeting the NSP10 gene at the indicated time points. The results were normalized to the control cells at 24 h. (D) Schematic of the construction of the inactivated replicon system to assess the primary translation of viral replicases. The NanoLuc luciferase reporter gene, flanked by a P2A cleavage site, was inserted between NSP1 and NSP2 of the SARS-CoV-2 replicon. The D760N and D761N double mutations were introduced into the NSP12 to inactivate the RdRp activity, ensuring that only translation could be assessed. (E) Detection of the primary translation of viral replicases as indicated by the luciferase activity. Control and DAZAP2-edited HeLa cells were electroporated with the modified replicon RNA, and the luciferase activity was monitored. (F) Confocal analysis of the localization of DAZAP2 and SARS-CoV-2 N protein. SARS-CoV-2-infected A549-ACE2 cells were fixed and stained with anti-DAZAP2 or anti-N antibody. The representative confocal images were shown. Scale Bar, 50 or 20 µm. (G, H) Validation of SERPINE1 gene. A549-ACE2 cells were edited with two independent sgRNAs targeting SERPINE1, followed by infection with authentic SARS-CoV-2 (G) or trVLP-NLuc particles (H). Data shown are from three independent experiments, and each independent experiment was performed in triplicate. As for the results shown as relative change, data are normalized to the control of the individual experiment. (A–C, G and H) Unpaired t test; (E) two-way ANOVA; n = 3; mean ± s.d.; ****P < 0.0001; ns, not significant.

To further validate these findings, we quantified genomic RNA levels by qRT-PCR targeting the NSP10 gene at different time points. Higher levels of genomic RNA were detected in DAZAP2-edited cells than in control (Fig. 4C). These results suggest that DAZAP2 inhibits viral genomic RNA replication, leading to reduced protein translation, as indicated by decreased NanoLuc luciferase activity.

We next explored whether the inhibition of genomic replication by DAZAP2 is due to its impact on the primary translation of non-structural replicases for incoming viral genomes released from endolysosomes. To test this, we modified the replicon system by inserting the NanoLuc luciferase reporter gene, flanked by a P2A cleavage site, between NSP1 and NSP2 of the SARS-CoV-2 replicon (Fig. 4D). Additionally, we introduced D760N and D761N double mutations into the NSP12 to inactivate the RdRp activity (46), ensuring that only translation could be assessed (Fig. 4D). After electroporating the modified replicon RNA into cells, the luciferase activity was monitored from 2 to 8 h, peaking at 4 h (Fig. 4E). However, no significant difference in luciferase activity was observed between control and DAZAP2-edited cells, indicating that DAZAP2 does not affect the primary translation of viral replicases.

DAZAP2 may regulate host gene expression to restrict viral infection

To understand how DAZAP2 inhibits coronavirus entry and replication, we first evaluated whether it is an interferon-stimulated gene (ISG). DAZAP2 gene expression was not upregulated upon treatment with 1,000 U/mL IFNα−2b (Fig. S3A), whereas the known ISGs IFITM3 and MX1 were increased over 100-fold and 1,000-fold, respectively (Fig. S3B and C). Furthermore, we performed SARS-CoV-2 infection in STAT1-, MAVS-, or IRF3-deficient A549-ACE2 cells, which lack critical innate immune signaling pathways. Despite these disruptions, infection efficiency was still significantly increased in DAZAP2-edited cells as compared to the controls (Fig. S4). These results suggest that the antiviral function of DAZAP2 is possibly independent of innate immune responses.

We next investigated whether DAZAP2 directly exerts its antiviral function at the endolysosome or cytoplasm to inhibit viral entry and replication, respectively. Localization studies revealed that endogenous DAZAP2 is predominantly located in the nucleus in both mock- and virus-infected cells, whereas viral N protein localizes to the cytoplasm (Fig. 4F). Additionally, DAZAP2 did not co-localize with the early endosome marker EEA1 or the endolysosome marker LAMP1 (Fig. S5), further supporting that DAZAP2 may not directly interact with viral entry machinery at these sites.

Given that DAZAP2 localizes to the nucleus and has been reported to interact with transcription factors (36, 37), we hypothesized that it may regulate the expression of specific host genes, contributing to its role as a restriction factor. Although Hou et al. previously suggested that DAZAP2, along with VTA1 and KFL5, regulates host cell responses to SARS-CoV-2 infection by controlling the SERPINE1 expression (associated with COVID-19 severity), we found that editing SERPINE1 had no effect on infection by authentic SARS-CoV-2 or trVLP-NLuc particles (Fig. 4G and H).

Since DAZAP2 inhibits viral entry at both the endosomal and plasma membranes, we also examined whether DAZAP2 affects the expression of SARS-CoV-2 entry factors. Using a validated antibody against the cell receptor ACE2 (Fig. S6A and B), the surface expression level of ACE2 was not changed when DAZAP2 was edited (Fig. S6C). Likewise, the expression of other known entry factors, such as AXL, heparan sulfate, TIM-1, SIGLEC1, DC-SIGN, CTSL, Furin, and TMPRSS2, was not affected by DAZAP2 (Fig. S6D to G).

Collectively, these findings suggest that DAZAP2 indirectly inhibits SARS-CoV-2 infection, potentially through the regulation of a specific subset of host genes. Although the precise mechanisms remain to be fully elucidated, our data highlight DAZAP2 as a multifaceted restriction factor that acts at both the entry and replication stages of the viral life cycle.

Knockout of Dazap2 promotes SARS-CoV-2 infection in mouse models

Given the functional conservation of human DAZAP2 and its mouse ortholog in cell-based assays, we extended our investigation to in vivo mouse models. To generate the Dazap2-knockout mouse (Dazap2^−/−), exons 2 and 3 of the Dazap2 locus were removed by CRISPR/Cas9 technology, resulting in the deletion of 365 bp coding sequence and disruption of protein function (Fig. 5A). The Dazap2^−/− mice are viable, are fertile, and do not exhibit any observable defects.

Fig 5.

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.

To assess viral replication, four female age-mapped wild-type (WT) and Dazap2^−/− mice were intranasally inoculated with the mouse-adapted Beta variant (B.1.351) of SARS-CoV-2. After 3 days, lungs and nasal turbinates were collected for infectious virus titration. Although the average titer in the lungs of WT mice reached 7 logs, the average titer in Dazap2^−/− mice was 7.6 logs (Fig. 5B). Viral titers in the nasal turbinates of both groups were below the limit of detection (Fig. 5C). To further characterize the virus infection in lung tissues, mice were intranasally inoculated with a low dose of 50 FFU of the adapted virus for one day. The average viral titer in Dazap2^−/− mice increased from 9.15 logs in WT mice to 9.9 logs (Fig. 5D).

The clinical isolate of the Beta variant (B.1.351) of SARS-CoV-2 can naturally infect WT mice, albeit less efficiently than in ACE2 transgenic mice (47–50). To further evaluate the restriction function of Dazap2, WT and Dazap2^−/− mice were intranasally challenged with 1,000 or 50 FFU of the non-adapted Beta variant for 1 day. At higher doses (1,000 FFU), viral replication increased from 5.95 logs in WT mice to 6.98 logs in knockout mice (Fig. 5E). A slight increase in viral titer was also observed at the low dose (50 FFU) (Fig. 5E).

To further investigate the restrictive function of DAZAP2, we cross-bred Dazap2^−/− mice with human ACE2 knock-in mice (hACE2) to enable infection by the original SARS-CoV-2 strain (51). Female age-matched hACE2 and Dazap2^−/− mice with ACE2 expression (hACE2-Dazap2^−/−) were intranasally challenged with 50,000 FFU of the SH01 strains isolated in early 2020 (52). Viral replication in tissues was quantified by qRT-PCR at day 3 post-infection. Viral titers in the lungs of hACE2-Dazap2^−/− mice were significantly higher than in hACE2 mice (Fig. 5F), whereas no difference was detected in the nasal turbinates (Fig. 5G). Additionally, the increased viral loads in the lungs of hACE2-Dazap2^−/− mice were associated with elevated cytokine expression. The levels of Il1b, Il6, Tnfα, Ifnar1, and Cxcl10 mRNA were significantly higher in hACE2-Dazap2^−/− mice compared with hACE2 mice (Fig. 5H and I). These results collectively demonstrate that knockout of Dazap2 enhances SARS-CoV-2 replication in mouse models.

Knockout of DAZAP2 promotes SARS-CoV-2 infection in human primary airway epithelial cells

To determine whether DAZAP2 exerts antiviral function in humans, we utilized air-liquid interface (ALI) culture of human primary nasal (HNEC) and bronchial epithelial cells (HBEC). These cultures represent physiologically relevant models for studying SARS-CoV-2 infection (Fig. 5J). Nasal and bronchial stem cells were isolated and edited with control or DAZAP2-specific sgRNA, and knockout efficiency was confirmed by western blotting (Fig. 5K).

After differentiation in transwells, the cells were infected with SARS-CoV-2, and viral production was assessed at 24 and 48 h post-infection. In nasal epithelial cells, we observed a modest 0.74-log increase in virus production in the supernatant of DAZAP2-edited nasal epithelial cells at 24 h (Fig. 5L and M). In contrast, bronchial epithelial cells exhibited significantly higher levels of viral RNA in cells and viral titers in the supernatant at both 24 and 48 h post-infection in DAZAP2-edited cells when compared with control cells (Fig. 5N and O). These findings demonstrate that DAZAP2 acts as a host restriction factor in human airway tissues, significantly inhibiting SARS-CoV-2 infection in physiologically relevant models. The stronger antiviral effect observed in bronchial epithelial cells may highlight the importance of DAZAP2 in protecting the lower respiratory tract from viral infection.

DISCUSSION

The global spread of SARS-CoV-2 has caused unprecedented devastation and mortality since the 1918 influenza pandemic. Identifying the host factors involved in SARS-CoV-2 infection is critical for understanding virus-host interactions, elucidating pathogenesis, and developing potential host-directed therapeutics. Although proviral host factors have been extensively studied using high-throughput strategies, such as genome-wide CRISPR screens (16, 17, 52–58), antiviral host factors remain less explored. Although several studies have focused on the screening of ISGs, a large family of antiviral genes (12–15, 18), a comprehensive genome-scale identification of host restriction factors is still needed.

Unlike genome-wide activation screens (16, 17) or CRISPR dropout screens analyzing depleted cells (29), we employed a FACS-based genome-wide CRISPR knockout screen to enrich the highly susceptible live cells. Through this work, we identified and validated a set of restriction factors, including PDCD10, DAZAP2, CAB39, and VTA1. Among these, the known IFN-related gene PLSCR1 exhibited the most potent antiviral activity (27, 40–42). We focused on the gene with the second greatest impact, DAZAP2, which, after being knocked out, led to an over 4-fold increase in SARS-CoV-2 infection.

Although Hou et al. previously identified DAZAP2 as an antiviral gene regulating SERPINE expression (29), its mechanisms of action and in vivo relevance remained unclear. In this study, we systematically dissected the stages of viral entry and replication, revealing that: (i) DAZAP2 inhibits virion fusion with both endolysosomal and plasma membranes, blocking viral entry; (ii) DAZAP2 inhibits genomic RNA replication but does not affect the primary translation of replicases from incoming genomes. Notably, the restriction of DAZAP2 against cell entry appears more pronounced than its effect on replication, as demonstrated by pseudovirus and replicon assays, respectively. We also established DAZAP2 as a pan-coronavirus restriction factor, functioning across four genera of coronaviruses, and validated its antiviral roles in mouse models and human primary airway epithelial cells.

DAZAP2 is an evolutionarily conserved protein ubiquitously expressed across tissues (59–61). It contains predicted Src homology 2 (SH2)/SH3 binding sites and a polyproline region and is reported to localize in both the cytoplasm and the nucleus (36–38, 59, 61). Its nuclear localization suggests a regulatory role in gene expression. For example, DAZAP2 interacts with the transcription factor SOX6 to regulate the expression of the L-type Ca²⁺ channel α_1c gene and cardiac differentiation (62). Similarly, it interacts with the transcription factor TCF-4 to regulate gene expression, possibly by modulating the affinity of its DNA-recognition motif (37). DAZAP2 is also implicated in diseases, with its downregulation reported in multiple myeloma (59, 63, 64). Intriguingly, DAZAP2 occupies p53 response elements to regulate the expression of a subset of genes upon DNA-damaging chemotherapeutic treatment and plays a protective role in cell survival (36).

Despite inhibiting virion fusion and genomic replication, DAZAP2 primarily localizes to the nucleus, with no detectable presence in the cytosol, plasma membranes, or endolysosomes. This suggests that the restrictive effects of DAZAP2 are likely indirect, mediated through the regulation of specific host genes. Although SERPINE1 was previously identified as a DAZAP2-regulated gene associated with COVID-19 severity (29), editing SERPINE1 had no impact on SARS-CoV-2 infection in our study. This implies that other, yet unidentified, genes regulated by DAZAP2 may underlie its antiviral effects. We hypothesize that DAZAP2 interacts with host transcription factors to modulate the expression of specific genes or pathways. Alternatively, DAZAP2 knockout may disrupt cellular homeostasis, rendering cells more susceptible to viral infection.

Stress granules (SGs) are host organelles that protect cells from harmful stress or virus infection by sequestering host and viral factors, including proteins and RNAs. In addition to its nuclear localization, DAZAP2 is found in the cytoplasm, where it participates in SG formation (38). DAZAP2 interacts with the RNA-binding protein deleted-in-azoospermia-like (DAZL), which is essential for SG formation and germ cell protection under heat stress (65). Whether the restrictive effect of DAZAP2 on coronavirus replication involves its role in SGs remains to be determined. Notably, coronaviruses encode proteins such as N protein, NSP1, and NSP15 to disrupt SG formation and evade antiviral responses (66–71). This may explain why DAZAP2 puncta were not observed in the cytoplasm during infection.

To validate the restriction effect of DAZAP2 on SARS-CoV-2 infection, we generated Dazap2-knockout mice. Consistent with previous reports, these mice exhibited no obvious developmental abnormalities (60). Upon viral challenge, viral loads in the lungs were significantly higher in knockout mice than the WT mice, confirming the protective role of DAZAP2 during infection. Similarly, in air-liquid interface cultures of human primary bronchial epithelial cells, SARS-CoV-2 infection was significantly enhanced in DAZAP2-edited cells when compared with the control cells. Interestingly, no difference in viral titers was observed in the nasal turbinates of mice infected with the beta variant (B.1.351) or the original strain. Similarly, only minor phenotypic differences were detected in primary nasal epithelial cells. This may reflect the inefficient replication of these viruses in the upper respiratory tract and nasal tissues (72).

In summary, we conducted a FACS-based genome-wide CRISPR knockout screen to identify SARS-CoV-2 host restriction factors, including the previously known DAZAP2. We thoroughly dissected its roles in viral entry and replication and validated its antiviral functions in mouse models and human primary airway epithelial cells. These findings provide new insights into the host defense system against coronavirus infection and highlight potential avenues for developing host-directed therapeutics.

MATERIALS AND METHODS

Cells and viruses

Vero E6 (Cell Bank of the Chinese Academy of Sciences, Shanghai, China), HEK 293T (ATCC #CRL-3216), A549 (ATCC #CCL-185), A549-ACE2 (52), HeLa (ATCC #CCL-2), HeLa-ACE2 (52), Calu-3 (Cell Bank of the Chinese Academy of Sciences, Shanghai, China), MEF expressing the human ACE2 (MEF-ACE2), BHK-21, Huh7, swine testicular (ST), LLC-MK2, HRT-18, and ΔMAVS A549 (73), all were cultured at 37°C in Dulbecco’s modified Eagle medium supplemented with 10% fetal bovine serum (FBS), 10 mM HEPES, 1 mM sodium pyruvate, 1× non-essential amino acids, and 100 U/mL of penicillin-streptomycin. The IRF3- and STAT1-knockout A549 clonal cell lines were generated by transduction of lentivirus expressing individual sgRNA and selected for 7 days with puromycin and blasticidin, respectively. Clonal cell lines were obtained by limiting dilution and verified by western blotting. All cell lines were tested routinely and were free of mycoplasma contamination.

The SARS-CoV-2 (nCoV-SH01-Sfull) stock (52), swine acute diarrhea syndrome coronavirus (SADS-CoV), porcine epidemic diarrhea virus (PEDV), and infectious bronchitis virus (IBV) were propagated in Vero E6 cells and titrated in Vero E6 by focus-forming assay (74). Other virus stocks of coronaviruses, porcine deltacoronavirus (PDCoV) (ST cells), HCoV-229E (Huh7 cells), and HCoV-OC43 (HRT-18 cells), were prepared and titrated similarly in their respective cell lines. All experiments involving SARS-CoV-2 live virus infection were performed in the biosafety level 3 (BSL-3) facility of Fudan University or Guangzhou Customs Technology Center, following the regulations.

Genome-wide CRISPR knockout screen

The human Brunello CRISPR knockout pooled library encompassing 76,441 different sgRNAs targeting 19,114 genes (30) was a gift from David Root and John Doench (Addgene #73178), and packaged in 293 FT cells after co-transfection with psPAX2 (Addgene #12260) and pMD2.G (Addgene #12259) at a ratio of 2:2:1 using Fugene^® HD (Promega). At 48 h post-transfection, supernatants were harvested, clarified by spinning at 3,000 rpm for 15 min, and aliquoted for storage at −80°C.

For the CRISPR sgRNA screen, A549-ACE2-Cas9 cells (52) were transduced with packaged sgRNA lentivirus library at a multiplicity of infection (MOI) of ~0.3 by spinoculation at 1,000 g and 32°C for 30 min in 12-well plates. After selection with puromycin for around 7 days, cells were inoculated with SARS-CoV-2 transcription- and replication-competent virus-like particles in which the N gene is replaced by the reporter GFP (trVLP-GFP) (31). 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. After infection at an MOI of 0.5 for 24 h, cells were harvested and sorted for the GFP-positive population. Genomic DNA from both sorted cells and uninfected cells was extracted for sgRNA amplification and next-generation sequencing using an Illumina NovaSeq 6000 platform. The sgRNA sequences targeting specific genes were trimmed using the FASTX-Toolkit (http://hannonlab.cshl.edu/fastx_toolkit/) and cutadapt 1.8.1 and further analyzed for sgRNA abundance and gene ranking by a published computational tool (MAGeCK) (see Table S1).

Gene validation

The top 20 genes from the MAGeCK analysis were selected for validation. Two independent sgRNAs per gene were chosen from the Brunello CRISPR knockout library and cloned into the plasmid lentiCRISPR v2 (Addgene #52961) and packaged with plasmids psPAX2 and pMD2.G. A549-ACE2 cells were transduced with lentiviruses expressing individual sgRNA and selected with puromycin for 7 days. The gene-edited mixed population of cells was used for validation using the SARS-CoV-2 transcription- and replication-competent virus-like particles, revealing that the N gene is replaced by the NanoLuc luciferase (trVLP-Nluc) at an MOI of 0.5 for 24 h. trVLP-Nluc was constructed in this study to measure virus replication for convenience. The luciferase activity was determined using the Nano-Glo Luciferase Assay kit (Promega #N1110), and the luminescence was recorded using a FlexStation 3 (Molecular Devices). The sgRNA sequences are listed in Table S2.

For authentic SARS-CoV-2 virus infection, gene-edited A549-ACE2, HeLa-ACE2, Calu-3, and MEF-ACE2 cells were inoculated for 24 h. Cells were fixed with 4% paraformaldehyde (PFA) diluted in PBS for 30 min at room temperature and permeabilized with 0.2% Triton x-100 in PBS for 1 h at room temperature. Cells were then subjected to immunofluorescence staining and high-content imaging. The gene-edited A549-ACE2, HeLa-ACE2, or MEF-ACE2 cells were also infected with members of the family Coronaviridae and then subjected to immunofluorescence staining, high-content imaging, or flow cytometry analysis.

Pseudotyped virus experiment

SARS-CoV-2 pseudoviruses were packaged as previously described (52). Shortly, pcDNA3.1 vector expressing the spike gene of SARS-CoV-2 lacking the C-terminal 21 amino acids, the full spike of SARS-CoV-1, or VSV-G (pMD2.G [Addgene #12259]), was co-transfected in HEK 293T cells with the murine leukemia retrovirus (MLV) expressing the NanoLuc luciferase gene and plasmid expressing the MLV Gag-Pol using Fugene HD transfection reagent (Promega). The virus entry was assessed by transduction of pseudoviruses in gene-edited cells in 96-well plates. After 48 h, the luciferase activity was determined using the Nano-Glo Luciferase Assay kit (Promega #N1110), and the luminescence was recorded by using a FlexStation 3 (Molecular Devices).

Generation of SARS-CoV-2 replicon system and trVLP-Nluc particles

To construct the replicon system, the full-length viral RNA of SARS-CoV-2 (nCoV-SH01, GenBank accession no. MT121215) was reverse-transcribed, PCR-amplified as six fragments, and cloned into the pSMART vector individually. The region encompassing the spike gene to ORF8 was replaced by the NanoLuc luciferase-P2A-puromycin cassette. The T7 promoter was inserted upstream of the viral genome to initiate transcription. The sequences of HDVr ribosome and transcription terminator were added downstream of the poly-A tail of the viral genome. Six pSMART plasmids and pBeloBAC11 vector were digested with type IIS restriction enzymes and assembled in vitro as one plasmid and then amplified in bacteria. Replicon RNA was transcribed from the single plasmid using the mMESSAGE mMACHINE T7 Transcription Kit (Invitrogen #AM1344) according to the manufacturer’s instructions. RNA was then transfected into target cells to assess the virus replication efficiency by measuring the luciferase activity using the Nano-Glo Luciferase Assay kit (Promega #N1110).

To detect the virus infection conveniently and safely, SARS-CoV-2 transcription- and replication-competent virus-like particles in which the N gene is replaced by the NanoLuc luciferase (trVLP-Nluc) were generated as described previously (31). The replicon plasmid constructed above was modified by maintaining the spike to ORF8 genes but replacing the N gene with NanoLuc luciferase. The single-round trVLP-Nluc particles were packaged in Vero E6 cells expressing the N gene. The virus infection was determined by measuring the luciferase activity using the Nano-Glo Luciferase Assay kit (Promega #N1110).

To modify the replicon system to assess the primary translation of genomic RNA, the NanoLuc luciferase-P2A-puromycin cassette in the replicon constructed above was removed, and a coding sequence of P2A-NanoLuc luciferase cassette was inserted at the NSP1/NSP2 junction. To make a replication-deficient replicon, the viral RdRp mutant was created by mutating NSP12 catalytic residues at positions 760 and 761 from aspartic acid (D) to asparagine (N) (46). As described above, modified replicon RNA was in vitro transcribed and then electroporated into target cells to assess the primary translation of genomic RNA by measuring the luciferase activity.

Plasmid constructs

The human DAZAP2 (Sino Biological #HG15906-G) or mouse Dazap2 (Sino Biological #MG52462-G) gene was PCR-amplified and cloned into the pLV-EF1α-IRES-puro (Addgene #85132). The PH-Halo-LgBiT fragment was synthesized (GENEWIZ), amplified, and cloned into pLV-EF1α-IRES-Hygro. The LgBiT fragment was cloned into the pCAGGS vector by using the synthesized PH-Halo-LgBiT as a template. The CypA-HiBiT fragment was synthesized (GENEWIZ), amplified, and cloned into pCAGGS. Lentiviruses were packaged by co-transfection with psPAX2 (Addgene #12260) and pMD2.G (Addgene #12259) and transduced into wild-type, DAZAP2-deficient A549-ACE2, or MEF-ACE2 cells.

Protein lysate preparation and western blotting

Cells in plates were washed twice with ice-cold PBS and lysed in RIPA buffer (Cell Signaling #9806S) with a cocktail of protease inhibitors (Sigma-Aldrich #S8830). Samples were prepared in reducing buffer (50 mM Tris, pH 6.8, 10% glycerol, 2% SDS, 0.02% [wt/vol] bromophenol blue, 100 mM DTT). After heating (95°C, 10 min), samples were electrophoresed in 10% SDS polyacrylamide gels, and proteins were transferred to PVDF membranes. Membranes were blocked with 5% non-fat dry powdered milk in TBST (100 mM NaCl, 10 mM Tris, pH 7.6, 0.1% Tween 20) for 1 h at room temperature and probed with the primary antibodies at 4°C overnight. After washing with TBST, blots were incubated with horseradish peroxidase (HRP)-conjugated secondary antibodies for 1 h at room temperature, washed again with TBST, and developed using SuperSignal West Pico or Femto chemiluminescent substrate according to the manufacturer’s instructions (Thermo Fisher).

The antibodies used are as follows: mouse anti-DAZAP2 (Santa Cruz #sc-515182, 1:1,000), mouse anti-FLAG (Sigma #F1804, 1:2,000), mouse anti-HA (Abmart #M20003M, 1:2,000), mouse anti-FURIN (Proteintech #67481-1-Ig, 1:2,000), rabbit anti-TMPRSS2 (Abcam #ab109131, 1:1,000), mouse anti-CTSL (Thermo Fisher #BMS1032, 1:1,000), rabbit anti-beta actin (Proteintech #20536-1-AP, 1:5,000), mouse anti-GAPDH (Proteintech #60004-1-Ig, 1:2,000), home-made mouse serum against N protein from SARS-CoV-2, HCoV-229E, PEDV, SADS-CoV or PDCoV (1:1,000), rabbit anti-HCoV-OC43 N (Sino Biological #40643-T62, 1:1,000), home-made mouse anti-MHV N mAb (1:1,000), rabbit anti-HCoV-NL63 N (Sino Biological #40641-T62, 1:1,000), home-made mouse anti-IBV N mAb (1:1,000). The HRP-conjugated secondary antibodies include goat anti-mouse (Sigma #A4416, 1:5,000) and goat anti-rabbit (Thermo Fisher #31460, 1:5,000).

Immunofluorescence staining and analysis

For high-content imaging analysis, virus-infected cells in plates were fixed with 4% paraformaldehyde in PBS for 30 min, permeabilized with 0.2% Triton X-100 for 1 h. Cells were then incubated with house-made mouse serum (1:1000) against nucleocapsid protein from different coronaviruses for 2 h at room temperature. After three washes, cells were incubated with the secondary goat anti-mouse IgG (H + L) conjugated with Alexa Fluor 555 (Thermo Fisher #A-21424, 2 µg/mL) for 1 h at room temperature, followed by staining with 4′,6-diamidino-2-phenylindole (DAPI). Images were collected using an Operetta High Content Imaging System (PerkinElmer) and processed using the PerkinElmer Harmony high-content analysis software v4.9 and ImageJ v2.0.0 (https://imagej.net/ij/).

For flow cytometry analysis, virus-infected cells were harvested with trypsin and fixed with 2% paraformaldehyde in PBS for 10 min. Cells were permeabilized with 0.1% saponin in PBS for 10 min and stained with house-made mouse serum (1:1,000) against nucleocapsid protein from different coronaviruses for 30 min at room temperature. After washing, cells were incubated with the secondary goat anti-mouse IgG (H + L) conjugated with Alexa Fluor 647 (Thermo Fisher #A21235, 2 µg/mL) for 30 min at room temperature. After two additional washes, cells were subjected to flow cytometry analysis (Thermo, Attune NxT) and data processing (FlowJo v10.0.7).

For surface staining of entry-related host factors and analyzed by flow cytometry, cells were collected with TrypLE (Thermo Fisher #12605010) and incubated with the rabbit anti-ACE2 (Sino Biological #10108-RP01, 1:500), mouse anti-AXL (R&D #MAB154-SP, 1 µg/mL), mouse anti-DC-SIGN (Biolegend #330102, 1 µg/mL), mouse anti-TIM-1 (Biolegend #354002, 1 µg/mL), mouse anti-SIGLEC1 (Abcam #ab199401, 1 µg/mL), or mouse anti-heparan sulfate (10E4) (USBiological #H1890, 1 µg/mL) primary antibody at 4°C for 30 min. After washing, cells were stained with goat anti-rabbit IgG (H + L) conjugated with Alexa Fluor 647 (Thermo Fisher #A21245, 2 µg/mL) or goat anti-mouse IgG (H + L) conjugated with Alexa Fluor 647 (Thermo fisher #A21235, 2 µg/mL) for 30 min at 4°C and subjected to flow cytometry analysis.

For confocal microscopy analysis, cells seeded on coverslips were fixed with 4% paraformaldehyde in PBS for 30 min, permeabilized with 0.1% saponin in PBS for 10 min. Cells were then incubated with primary antibody overnight at 4°C. After three washes, cells were incubated with the secondary antibody for 2 h at room temperature, followed by staining with 4’,6-diamidino-2-phenylindole (DAPI). Images were collected using a Leica Confocal Microscope (TCS SP8), processed using the Leica Application Suite X (LAS X, v3.7.0.20979), and ImageJ v2.0.0 (https://imagej.net/ij/). The primary antibodies used are as follows: house-made mouse anti-SARS-CoV-2 nucleocapsid protein serum (1:1,000), rabbit anti-SARS-CoV-2 spike (Sino Biological #40591-T62, 1:1,000), rabbit anti-LAMP1 (Abcam #ab24170, 1:1,000), mouse anti-DAZAP2 (Santa Cruz #sc-515182, 1:1,000), rabbit anti-HA (Abcam #ab9110, 1:1,000), mouse anti-dsRNA antibody (J2) (Scicons #10010200). The secondary antibodies used are as follows: goat anti-mouse IgG (H + L) conjugated with Alexa Fluor 555 (Thermo Fisher #A-21424, 2 µg/mL), goat anti-rabbit IgG (H + L) conjugated with Alexa Fluor 488 (Thermo Fisher #A-11034, 2 µg/mL), followed by staining with 4′,6-diamidino-2-phenylindole (DAPI).

Virus binding and internalization assay

For the binding assay, A549-ACE2 cells were pre-chilled on ice for 10 min followed by incubation with ice-cold virus (MOI 5) on ice for 45 min. After washing with ice-cold PBS three times, cells were lysed in TRIzol reagent (Thermo Fisher #15596018) for RNA extraction and qRT-PCR.

For the internalization assay, after virus binding as described above, cells were washed with ice-cold PBS three times, followed by incubation at 37°C for 45 min. Uninternalized virions on the cell surface were removed by treating cells with 400 µg/mL protease K on ice for 45 min. After washing with ice-cold PBS three times, cells were lysed in TRIzol reagent for RNA extraction and qRT-PCR. The relative amount of bound or internalized virions was normalized to the internal control GAPDH.

Virion trafficking assay

The experiments were conducted as described previously (42). Control or DAZAP2-deficient A549-ACE2 cells seeded on coverslips were pretreated with 25 µM of E-64d, a cathepsin B and L proteinase inhibitor. One hour later, the cells were inoculated with SARS-CoV-2 trVLP-Nluc. The inhibitor E-64d was maintained in the medium during the infection. At 4 h post-infection, cells were washed twice with PBS, fixed with 4% PFA for 10 min, and then permeabilized with 0.1% saponin for 10 min. Cells were blocked with 5% BSA in PBS for 1 h and incubated with primary antibodies (rabbit anti-SARS-CoV-2 spike protein, mouse anti-SARS-CoV-2 nucleocapsid protein, or rabbit anti-LAMP1) at 4°C overnight. After three washes, cells were incubated with the secondary goat anti-mouse or rabbit antibody conjugated with Alexa Fluor 555 or 488 for 2 h at room temperature, followed by staining with DAPI. Images were acquired using a Leica Confocal Microscope (TCS SP8) and processed using the Leica Application Suite X (LAS X, v3.7.0.20979). The number of spike and nucleocapsid double-positive particles per field was quantified.

Quantification of endosomal acidification

Control or DAZAP2-deficient A549-ACE2 cells seeded in a 96-well plate were pre-treated with or without chloroquine (CQ) (20 µM). One hour later, cells were incubated in phenol red-free DMEM containing 2 µM LysoSensor Green dye (Thermo #L7535) in the presence or absence of CQ (20 µM) for 30 min at 37°C. Images were obtained using an AMG microscope (EVOS M7000), and the fluorescence intensity of LysoSensor was analyzed using ImageJ v2.0.0 (https://imagej.net/ij/).

Virus-cell fusion assay

The experiments were conducted via an improved system as described previously (42). The gag-interacting protein cyclophilin A (CypA) was fused with HiBit and encapsulated into MLV retrovirus particles bearing the SARS-CoV-2 spike protein. Pseudoviruses containing CypA-HiBiT were packaged in HEK 293T cells by co-transfecting the retrovector pMIG, for which the target gene was replaced by mGreenLantern (52), plasmid expressing the MLV Gag-Pol, pCAGGS expressing SARS-CoV-2 spike protein with the deletion of the C-terminal 21 amino acids, and pCAGGS expressing CypA-HiBiT using Fugene HD transfection reagent (Promega). At 48 h post-transfection, the supernatant was harvested, clarified by spinning at 3500 rpm for 15 min, aliquoted, and stored at −80°C. Control and DAZAP2-deficient A549-ACE2 cells (target cells) were transduced with pLV-PH-Halo-LgBiT-hygro lentivirus to stably express the LgBiT fragment. Target cells were seeded in a black/clear bottom 96-well plate for 24 h, followed by spinfection with 50 µL of pseudoviruses per well at 1,000 × g, 4°C for 30 min. The luciferase activity was determined using the Nano-Glo Luciferase Assay kit at 8 h post-infection, and the luminescence was recorded by using a FlexStation 3 (Molecular Devices).

Cell-cell fusion assay

For visualization and quantification of the syncytia formed after cell–cell fusion, HEK293T cells (donor cells) were transfected with the pCAGGS vector expressing the spike protein of SARS-CoV-2 that lacks the C-terminal 21 amino acids. At 24 h post-transfection, control and DAZAP2-deficient A549-ACE2 cells (acceptor cells) and donor cells were trypsinized and seeded in a 24-well plate at a ratio of 1:1. After 6 h of co-culture, cells were washed with PBS and fixed with 4% PFA. Images were obtained using an AMG microscope (EVOS M7000). Cell-cell fusion was quantified by Wright-Giemsa staining according to the manufacturer’s instructions (Sangon #E607315). Images were obtained using an AMG microscope (EVOS M7000), and the number of syncytia or syncytial nuclei was analyzed using ImageJ v2.0.0 (https://imagej.net/ij/).

To quantify the cell-cell fusion based on luciferase activity as previously reported (42), control and DAZAP2-edited A549 cells (acceptor cells) were transfected with pCAGGS-LgBiT encoding the LgBiT fragment of the split-NanoLuc luciferase. HEK 293T cells (donor cells) were transfected with pCAGGS-HiBiT encoding the HiBiT fragment of split-NanoLuc luciferase, together with pCAGGS vector expressing the spike protein of SARS-CoV-2 that lacks the C-terminal 21 amino acids. At 24 h post-transfection, acceptor and donor cells were trypsinized and seeded together in a black/clear bottom 96-well plate at a ratio of 1:1. After 24 h of co-culture, the luciferase activity was determined using the Nano-Glo Luciferase Assay kit, and luminescence was recorded by using a FlexStation 3 (Molecular Devices).

Generation of Dazap2 knockout mice

The mouse Dazap2 gene is located on chromosome 15, and the sequence of the Dazap2 locus (NC_000081.6) was obtained from NCBI. It has four transcripts, with one encoding the protein. The CRISPR/Cas9 technology was employed to remove exons 2 and 3, resulting in the deletion of 365 bp coding sequences and disruption of protein function. sgRNA was transcribed in vitro. Cas9 and sgRNA were microinjected into the fertilized eggs of C57BL/6JGpt mice. Fertilized eggs were transplanted to obtain positive F0 mice, which were confirmed by PCR and sequencing. A stable F1 generation mouse model was obtained by mating positive F0 generation mice with C57BL/6 J mice. The Dazap2-knockout (Dazap2^−/−) mice are viable, fertile, and do not exhibit any observable defects. The generation of knockout mice was accomplished with the help of Cyagen Biosciences (Suzhou, China).

Mouse experiments

Wild-type and Dazap2-knockout female mice in the same background of C57BL/6J at 10–12 weeks old were used in the study. The clinical isolate of the Beta variant (B.1.351) of SARS-CoV-2 was passaged 17 times (MA17) in 6- to 8-week-old BALB/c mice to allow adaptation. Both BALB/c and C57BL/6 mice are vulnerable to the infection of mouse-adapted MA17 virus. The mice were inoculated intranasally with 1,000 or 50 focus-forming units (FFU) of mouse-adapted or wild-type SARS-CoV-2 Beta variant (B.1.351) in a volume of 50 µL. The Dazap2^−/− mice were also cross-bred with human ACE2 knock-in mice (hACE2) in the mouse ACE2-coding sequences, with the exception of the signal peptide region, which was replaced with human ACE2 (Cyagen Biosciences #C001191). The 10–12 weeks old hACE2 or Dazap2^−/− female mice with human ACE2 expression (hACE2-Dazap2^−/−) were challenged with 50,000 FFU of SH01 strain isolated in early 2020 (52). Mice were euthanized at days 1 or 3 post-infection, and the turbinate and lungs were harvested and homogenized in PBS for live virus titration by focus-forming assay. The cytokine production in the lungs at day 3 was detected by qRT-PCR after RNA extraction as described below.

Preparation of HNEC and HBEC in vitro models

Human primary nasal stem cells and bronchial stem cells were isolated from nasal swabs and bronchoscopic brushing biopsies, donated by lesion-free individuals. Following 20 min digestion with collagenase IV (Gibco #17104019, 1 mg/mL) in Ham’s F12 (Gibco #11765054), dissociated cells were washed thoroughly in cold wash buffer (Ham’s F12, 5% FBS, 100 µg/mL penicillin-streptomycin, 100 µg/mL gentamicin, and 0.25 µg/mL amphotericin B) and selectively expanded in a 3T3-J2 culture system, under the protocol as previously described (75). To knockout the DAZAP2 gene, DAZAP2-specific sgRNA or non-targeting sgRNA was cloned into plasmid lentiCRISPR v2, for which the puromycin resistance gene was replaced by turboGFP reporter. Cells were transduced with lentivirus packaged with helper plasmids psPAX2 and pMD2.G and sorted for GFP-positive cells. The knockout efficiency was confirmed by western blotting. Cells were then subjected to in vitro differentiation via air-liquid interface (ALI) modeling. Upon 20-day differentiation in PneumaCult-ALI medium (StemCell Technologies #05001), mature epithelial structures of HNEC and HBEC were histochemically confirmed and proceeded to subsequent infection studies.

For SARS-CoV-2 infection, differentiated cells on the insert of 6.5 mm transwells were incubated with 100 µL of culture medium containing 40,000 FFU of virus for 1 h and washed three times. At 24 or 48 h post-infection, 200 µL of culture medium was added to the cells. After 30 min of incubation, the medium was harvested for FFA, and the cells were lysed for qRT-PCR to determine the virus replication.

qRT-PCR

RNA from tissues or cells was extracted with the TRIzol reagent (Thermo Fisher #15596018). Host mRNAs were determined using the One Step PrimeScript RT-PCR Kit (TaKaRa #RR064B) on the CFX Connect Real-Time System (Bio-Rad) instrument. Host mRNAs were also reverse transcribed into cDNA with PrimeScript RT Reagent Kit (TaKaRa #RR047A) using RT primer mix of oligo dT primer and random 6mers. qPCR was performed using TB Green Premix Ex Taq II (TaKaRa #RR820A) on the CFX Connect Real-Time System (Bio-Rad) instrument. Relative gene expression was calculated relative to GAPDH. Primers used for qRT-PCR are listed in Table S3.

Statistical analysis

Statistical significance was assigned when P values were <0.05 using Prism Version 9 (GraphPad). Data analysis was determined by an ANOVA, unpaired t-test, or Mann-Whitney depending on data distribution and the number of comparison groups.

DATA AVAILABILITY

The authors declare that all relevant data supporting the findings of this study are available within the paper and its supplemental material. The supplemental material provides information for the CRISPR screen, qRT-PCR, and RNAseq analysis. Source data are provided with this paper. Any other data of this study are available upon request.

ETHICS APPROVAL

The mouse experiment protocol has been approved by the Animal Ethics Committee of the School of Basic Medical Sciences at Fudan University (No.20220228-122) with formal written consent. Experiment protocol was approved by the Institutional Review Boards of the First Affiliated Hospital of Guangzhou Medical University and conducted in the BSL-3 facility following the regulations. The protocol of human biopsy acquisition and culture was approved by the Ethics Committees of Shanghai Jiao Tong University Affiliated Sixth People’s Hospital (approval number: 2017-090).

SUPPLEMENTAL MATERIAL

The following material is available online at https://doi.org/10.1128/mbio.00385-25.

Supplemental figures. mbio.00385-25-s0001.pdf.

Fig. S1-S6.

Table S1. mbio.00385-25-s0002.xlsx.

List of genes and scores after MaGeck analysis.

Table S2. mbio.00385-25-s0003.xlsx.

sgRNA sequences of genes.

Table S3. mbio.00385-25-s0004.xlsx.

List of primers and probes used for qRT-PCR.

ASM does not own the copyrights to Supplemental Material that may be linked to, or accessed through, an article. The authors have granted ASM a non-exclusive, world-wide license to publish the Supplemental Material files. Please contact the corresponding author directly for reuse.