SARS-CoV-2 nucleocapsid induces hyperinflammation and vascular leakage through the Toll-like receptor signaling axis in macrophages

Zhenlan Yao; Pablo A Alvarez; Carolina Chavez; Yennifer Delgado; Prashant Kaushal; David W Buchholz; David Austin; Qian Li; Yanying Yu; Anne K Zaiss; Vaithilingaraja Arumugaswami; Qiang Ding; Jeffrey J Hsu; Robert Damoiseaux; Hector C Aguilar; Mehdi Bouhaddou; Alexander Hoffmann; Melody M H Li

doi:10.1126/sciadv.aea2780

. 2026 Aug 5;12(32):eaea2780. doi: 10.1126/sciadv.aea2780

SARS-CoV-2 nucleocapsid induces hyperinflammation and vascular leakage through the Toll-like receptor signaling axis in macrophages

Zhenlan Yao ^1,^†, Pablo A Alvarez ^1,^2,^†, Carolina Chavez ^1,³, Yennifer Delgado ^1,², Prashant Kaushal ^1,^‡, David W Buchholz ¹, David Austin ³, Qian Li ^4,⁵, Yanying Yu ⁶, Anne K Zaiss ³, Vaithilingaraja Arumugaswami ^3,⁷, Qiang Ding ⁶, Jeffrey J Hsu ^4,^5,⁸, Robert Damoiseaux ^3,^5,⁹, Hector C Aguilar ¹, Mehdi Bouhaddou ^1,^2,¹⁰, Alexander Hoffmann ^1,^2,¹⁰, Melody M H Li ^1,^2,^7,^*

^¹Department of Microbiology, Immunology and Molecular Genetics, University of California, Los Angeles, Los Angeles, CA 90095, USA.

^²Molecular Biology Institute, University of California, Los Angeles, Los Angeles, CA 90095, USA.

^³Department of Molecular and Medical Pharmacology, David Geffen School of Medicine at UCLA, Los Angeles, CA 90095, USA.

^⁴Department of Medicine, University of California, Los Angeles, Los Angeles, CA 90095, USA.

^⁵Department for Bioengineering, Samueli School of Engineering, University of California, Los Angeles, Los Angeles, CA 90095, USA.

^⁶Center for Infection Biology, School of Basic Medical Sciences, Tsinghua University, Beijing, China.

^⁷Eli and Edythe Broad Center of Regenerative Medicine and Stem Cell Research, University of California, Los Angeles, Los Angeles, CA 90095, USA.

^⁸Department of Medicine, Veterans Affairs Greater Los Angeles Health Care System, Los Angeles, CA 90073, USA.

^⁹California Nano Systems Institute, University of California, Los Angeles, Los Angeles, CA 90095, USA.

^¹⁰Institute for Quantitative and Computational Biosciences, University of California, Los Angeles, Los Angeles, CA 90095, USA.

Corresponding author. Email: ManHingLi@mednet.ucla.edu

^†

These authors contributed equally to this work.

^‡

Present address: Mass Spectrometry Technology Access Center at McDonnell Genome Institute, Washington University School of Medicine, St. Louis, MO 63108, USA.

Roles

Zhenlan Yao: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Software, Validation, Visualization, Writing - original draft, Writing - review & editing

Pablo A Alvarez: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Validation, Visualization, Writing - original draft, Writing - review & editing

Carolina Chavez: Formal analysis

Yennifer Delgado: Data curation, Formal analysis, Investigation, Methodology, Resources, Software, Validation, Visualization, Writing - review & editing

Prashant Kaushal: Conceptualization, Data curation, Formal analysis, Investigation, Methodology

David W Buchholz: Investigation, Methodology, Resources, Writing - review & editing

David Austin: Methodology, Resources

Qian Li: Investigation

Yanying Yu: Resources

Anne K Zaiss: Resources

Vaithilingaraja Arumugaswami: Conceptualization, Formal analysis, Methodology, Resources, Supervision, Visualization, Writing - review & editing

Qiang Ding: Methodology, Resources, Writing - review & editing

Jeffrey J Hsu: Funding acquisition, Investigation, Resources, Writing - review & editing

Robert Damoiseaux: Conceptualization, Methodology, Project administration, Resources, Supervision, Writing - review & editing

Hector C Aguilar: Funding acquisition, Investigation, Methodology, Resources, Supervision, Writing - review & editing

Mehdi Bouhaddou: Data curation, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Visualization, Writing - review & editing

Alexander Hoffmann: Conceptualization, Formal analysis, Funding acquisition, Methodology, Project administration, Supervision, Validation, Visualization, Writing - review & editing

Melody M H Li: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Writing - original draft, Writing - review & editing

PMCID: PMC13440407 PMID: 42555728

Abstract

A substantial proportion of hospitalized COVID-19 patients require ICU admission, often associated with an imbalance between antiviral responses and inflammatory signaling leading to uncontrolled cytokine secretion. The SARS-CoV-2 nucleocapsid (N) protein is a known immune antagonist, but its role in macrophage-driven cytokine storms is unclear. We demonstrate that N functions in a stimulus-specific manner, specifically amplifying extracellular and dampening intracellular RNA sensing. Moreover, we show that this is a conserved feature of pathogenic betacoronaviruses through distinct mechanisms. Our interaction networks with SARS-CoV-2 variant N proteins suggest that the Delta variant N drives inflammation through interactions with several proteins, most notably, cGAS. Profiling of secreted cytokines revealed that N disrupts the secretome in a variant-specific manner. Most notably, we found that supernatants from the Delta variant N–expressing macrophages dramatically disrupt heart endothelial barriers, implicating N in COVID-19–associated cardiac complications. Our findings highlight N-mediated immune imbalance as a driver of severe COVID-19 and identify N as a promising therapeutic target to mitigate hyperinflammation.

A coronavirus protein rewires immune signaling, driving inflammation and damaging blood vessels in severe COVID-19.

INTRODUCTION

The emergence of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has caused substantial global morbidity and mortality (1). Even with the deployment of safe and effective vaccines, thousands of patients are hospitalized worldwide each winter (2). Moreover, severe COVID-19 cases still surpass severe flu rates, further emphasizing the urgency for understanding this disease progression (3). Approximately one-third of these patients will develop acute respiratory distress syndrome (ARDS), for which there are currently no effective treatments (4, 5). It is well understood that ARDS is marked by unconstrained proinflammatory cytokine secretion (6), termed cytokine storm. An additional hallmark of severe disease in these patients is a dampened antiviral response mediated by type I interferons (IFNs) (7, 8). Moreover, post-infectious sequelae often result in neurological and cardiac complications mediated by a dysregulated immune response. Together, these complications emphasize the need to elucidate mechanisms of SARS-CoV-2 immune dysregulation, yet our current understanding of these mechanisms remains unclear.

Type I IFNs, including IFN-α/β, are a critical first line of defense against viral infections, including SARS-CoV-2. Their production is triggered by pattern recognition receptors (PRRs) that sense viral components within infected cells. In the case of SARS-CoV-2, retinoic acid–inducible gene I (RIG-I)–like receptors (RLRs) are considered the primary sensors of viral RNA (9), although other PRRs, such as Toll-like receptors (TLRs), also contribute. Endosomal TLRs are highly expressed in immune cells, especially phagocytes, like monocytes, macrophages, and dendritic cells, and are responsible for sensing foreign viral RNA (10–12). These TLRs have been implicated in amplifying inflammation during SARS-CoV-2 infection (12, 13). Upon PRR activation, key transcription factors—including IFN regulatory factors 3 and 7 (IRF3/7), nuclear factor κB (NF-κB), and related transcription factors AP-1, JUN, and CEBPD—translocate to the nucleus to induce antiviral and inflammatory gene expression. Secreted type I IFNs also engage the signal transducer and activator of transcription 1 (STAT1)/STAT2/IRF9 complex to drive the expression of a broad array of IFN-stimulated genes (ISGs), many of which have direct antiviral functions. Timely induction of ISGs is essential for containing viral spread, and a controlled inflammatory response is critical for recruiting the adaptive response that will ultimately clear the infection. Delayed and impaired IFN responses are associated with severe COVID-19, and sustained inflammatory signaling can lead to immunopathology. Dysregulated IFN and NF-κB responses are hallmarks of severe SARS-CoV-2 infection, underscoring the need to understand how the virus disrupts these pathways to evade immune control and promote disease.

Pathogenic coronaviruses, including Middle East respiratory syndrome coronavirus (MERS-CoV), SARS-CoV-1, and SARS-CoV-2, have several proteins known to interact with and dampen innate immune pathways (14–18). Among them, the nucleocapsid (N) protein is consistently the most abundant immune modulator. Several N molecules are packaged within the virion, and it is the most highly expressed protein during replication, with estimates reaching up to 10⁸ molecules per infected cell (19, 20). Moreover, N is reportedly found throughout the cell, including the nucleus, in addition to being presented on the cell surface (21, 22). Therefore, N proteins have several opportunities to interact with various cellular signaling components in infected and nearby uninfected cells. N is reported to associate with G3BP1/2 to inhibit the formation of stress granules, which are critical for promoting an antiviral state (23). N has also been shown to impede the signaling of RLRs by blocking RIG-I interactions with critical cofactors to prevent downstream signaling (9, 24). Newer SARS-CoV-2 variants have evolved mutations in N that better facilitate the manipulation of RLR signaling (25), emphasizing the need to monitor the effects of N evolution on immune signaling.

Macrophages are key players in innate immunity and are crucial antigen-presenting cells for activating the adaptive immune system. Macrophages are well-positioned to encounter SARS-CoV-2 N proteins, which are abundantly displayed on the surface of infected cells. Strikingly, macrophages are the predominant cell type responsible for the cytokine storm observed in severe COVID-19 (26–28). Specifically, monocyte-derived macrophages comprise a majority of the immune cells found in bronchoalveolar lavage fluid, further emphasizing their inflammatory role in SARS-CoV-2 infection (29); however, direct infection of macrophages by SARS-CoV-2 remains controversial. Earlier studies suggest that macrophages are refractory to SARS-CoV-2 infection (30); however, accumulating evidence so far reveals that SARS-CoV-2 infection depends on the macrophage subtype (27, 31). For example, interstitial macrophages allow productive SARS-CoV-2 replication, characterized by an altered transcriptome and dense viral RNA bodies (29, 31). While SARS-CoV-2 can also enter monocyte-derived macrophages and alveolar macrophages via phagocytosis, viral replication is restricted and there are low amounts of viral protein expression and virion production (31). Notably, limited infection can still enhance proinflammatory cytokine expression in these macrophages (31), highlighting macrophages as critical players in SARS-CoV-2 induced cytokine storm. To date, no studies have investigated the roles of individual SARS-CoV-2 proteins on macrophage immunity, underscoring an urgent need to understand how SARS-CoV-2 dysregulates macrophage functions to drive pathological hyperinflammation.

Here, we demonstrate the varying roles of the SARS-CoV-2 N protein during macrophage infection. Specifically, we found that de novo production of the N protein results in a hyperinflammatory phenotype during TLR7/8 stimulation. To further unravel stimulus specific responses in THP-1 macrophage-like cells, we systematically investigated the transcriptomic changes in innate immune signaling in response to endosomal (TLR7/8, TLR3) and cytoplasmic (RIG-I/MDA5) PRR stimulation in the presence and absence of the N protein. This revealed that the N protein specifically hyperactivates TLR-mediated inflammation while also dampening RIG-I–mediated IFN activation. Moreover, this was not pathway specific, as several transcription factors across various signaling pathways were predicted to be affected. When investigating whether this was a conserved mechanism of the N protein, we found that several SARS-CoV-2 variants, SARS-CoV-1, and MERS-CoV N proteins were also hyperinflammatory; however, SARS-related viruses primarily up-regulated CXCL10, while MERS-CoV up-regulated CCL4 instead, suggesting distinct mechanisms. In an effort to define the interactions driving hyperinflammation, we compared the interactomes of the Delta variant N and the Omicron BA.1 variant N, which exhibit differing levels of hyperinflammation despite similar expression levels in our system. We pinpointed cGAS as a critical interactor and found that cGAS inhibition results in dramatically amplified expression of inflammation-related transcripts during TLR7/8 stimulation. Finally, we investigated the consequences of this hyperactivation and found that the Wuhan-Hu-1 N protein globally amplifies the secretion of inflammatory cytokines, while the Omicron BA.1 N protein globally dampens this secretion. The Delta variant N has no global net change relative to parental cells yet significantly changes some cytokines. When we treated endothelial cells from the brain and heart with the supernatants of macrophages expressing these N proteins, we found that supernatant from TLR7/8-stimulated macrophages expressing the Delta variant N protein significantly disrupts the cardiac endothelial cell barrier, correlating with key COVID-19 complications such as cardiac injury. Overall, our study pinpoints the SARS-CoV-2 N protein as a major driver of macrophage dysregulation leading to severe COVID-19.

RESULTS

Incoming and de novo–expressed nucleocapsids elicit opposing immune responses in macrophages

Macrophages play a major role in SARS-CoV-2–associated complications and immunopathogenesis (32); however, whether this is driven by direct infection or indirect manipulation by free viral components remains unclear. Previous work demonstrated that monocyte-derived macrophages infiltrating the lungs of severe COVID-19 patients express high levels of inflammatory cytokines (27, 29). While SARS-CoV-2 can enter macrophages through receptors such as CD169, this often results in abortive infection, whereas additional ACE2 expression can enable productive expression (27, 30, 31, 33). In addition, macrophages may be activated through engulfment of viral particles or proteins independent of productive infection. Therefore, we sought to determine how different modes of viral exposure, ranging from uptake to abortive or productive infection, affect inflammatory gene expression in macrophages. We used phorbol-12-myristate-13-acetate (PMA)-differentiated THP-1 cells as a macrophage-like model (hereafter “THP-1 macrophage-like cells”) and the SARS-CoV-2 USA-WA1/2020 (WA1) strain for these infection comparisons. Specifically, we compared THP-1 macrophage-like cells, which are poorly infected, with CD169-expressing and CD169 + ACE2–expressing THP-1 macrophage-like cells, the latter supporting robust infection (31). We found that SARS-CoV-2 infection of THP-1 macrophage-like cells overexpressing CD169 + ACE2 up-regulates the expression of most inflammation-related transcripts tested, except for CXCL8, which is down-regulated (Fig. 1A). The expression levels of IFNB1 and CXCL10 transcripts increase dramatically, by approximately 200- and 5000-fold, respectively. These data suggest that productive infection induces the most amount of inflammatory-related gene expression, compared to exposure to virus particles, or abortive infection, which induces a weaker relative inflammatory response.

Fig. 1. — (A) Representative cytokine expression induced by SARS-CoV-2 infection in THP-1 macrophage-like cells and CD169- and CD169 + ACE2–overexpressing THP-1 macrophage-like cells. Cells were inoculated with SARS-CoV-2 at MOI 1 and infected for 24 hours. The expression of indicated genes is normalized and quantified by RT-qPCR. The relative fold changes are evaluated by comparisons to uninfected cells and plotted in log₁₀ scale. Data are representative of two independent experiments. Asterisks indicate statistically significant fold change differences of SARS-CoV-2–infected CD169- and CD169 + ACE2–expressing cells as compared to SARS-CoV-2–infected parental cells [two-way analysis of variance (ANOVA) and Dunnett’s multiple comparisons test: ****P < 0.0001]. (B to F) VLP regulation on TLR7/8 stimulation–induced transcript expression. The indicated cells were incubated with VLPs (Wuhan-Hu-1 strain structural proteins) for 24 hours. They were then stimulated with CL097 for 4 hours. The relative fold changes of the indicated gene are evaluated through RT-qPCR, by comparing to unstimulated cells. Data are representative of two independent experiments. Asterisks indicate statistically significant fold change difference of CL097-treated cells versus VLP + CL097–treated cells (two-way ANOVA and Šidák’s multiple comparisons test: *P < 0.05; ***P < 0.001; ****P < 0.0001). (G) Expression of *TNF*, *CCL4*, and *CXCL10* transcripts in parental THP-1 macrophage-like cells and cells expressing Wuhan-Hu-1 N protein. The cells were treated with doxycycline (2 μg/ml) for 48 hours to express N. They were then either untreated or stimulated with CL097 for 4 hours. The expression of the indicated genes is normalized and quantified by RT-qPCR. The relative fold changes are evaluated by the comparisons to unstimulated cells and plotted based on 2^−ΔΔCt. Data are representative of two independent experiments (multiple unpaired t tests, using two-stage step-up method of Benjamini, Krieger, and Yekutieli FDR: *P < 0.05; **P < 0.01).

We next interrogated the source of this immune manipulation. It is well known that the SARS-CoV-2 N protein is a master manipulator of the innate immune response across various cell types (17, 34). To this end, we also demonstrate that the N protein is the most inhibitory among other SARS-CoV-2 immune antagonists, as N-expressing human embryonic kidney (HEK) 293T cells have strongly dampened IFN activation and NF-κB signaling (fig. S1, A and B). Therefore, we sought to examine the role of the N protein in our THP-1 macrophage-like cells during the different stages of infection, namely, incoming/packaged N versus newly synthesized N during replication.

We focused on the packaged N protein by delivering it into THP-1 macrophage-like cells via nonreplicative SARS-CoV-2 virus-like particles (VLPs) (35). Since the VLPs still contain spike (S), we used the same cell lines as before, which express either neither receptor, CD169 only, or both CD169 + ACE2, to understand the differences between engulfment and effective entry/delivery of N into the cytosol. THP-1 macrophage-like cells were exposed to VLPs for 24 hours and then stimulated for 4 hours with CL097, a single-stranded RNA mimic that activates endosomal TLR7/8, which is a key sensor of viral RNA in macrophages (fig. S1C). We then quantified representative inflammation-related transcripts as was done for the infection (Fig. 1, B to F). CD169- or CD169 + ACE2–expressing THP-1 macrophage-like cells were more responsive to CL097-mediated TLR7/8 stimulation than parental cells, possibly due to increased phagocytic activity. In the parental THP-1 macrophage-like cells, VLP treatment resulted in decreased expression of all transcripts tested; however, only CCL4 and CXCL8 were statistically significant (Fig. 1, D and F). The same trend was observed for the other cell lines, where all transcripts tested were significantly suppressed with VLP treatment, with the exception of CXCL8 (Fig. 1, B to F). These data indicate that packaged N protein, potentially along with other structural proteins in the VLPs, acts as an antiviral antagonist, likely to set up a favorable environment for viral replication.

We next interrogated how de novo synthesis of the N protein alone would affect macrophages. Therefore, we constructed THP-1 macrophage-like cell lines with doxycycline-inducible expression of the Wuhan-Hu-1 strain N protein using lentivirus transduction. We then stimulated parental and N-expressing cells with doxycycline for 48 hours followed by CL097 for 4 hours and measured three inflammation-related transcripts via quantitative polymerase chain reaction (qPCR): tumor necrosis factor (TNF), IFNB1, and CCL4 (Fig. 1G). In this case, we observed the opposite trend, where newly synthesized N protein now promotes inflammation-related gene expression. Specifically, we saw a four- and threefold increase in TNF and CCL4 expression, respectively, and a twofold reduction in IFNB1 expression (Fig. 1G). Altogether, these data demonstrate distinct roles for the N protein that restructure the cellular environment to favor viral replication versus spread in a temporal manner, potentially contributing to the shift from early immune evasion to later hyperinflammation in COVID-19.

Nucleocapsid enhances TLR-driven inflammation and suppresses RLR-mediated antiviral signaling

We have shown that newly synthesized SARS-CoV-2 N protein is proinflammatory during TLR7/8 stimulation in THP-1 macrophage-like cells; however, given that SARS-CoV-2 is an RNA virus, it can activate multiple endosomal (TLR7/8, TLR3) or cytosolic (RIG-I/MDA5) RNA sensors. For a more comprehensive analysis, we treated our THP-1 macrophage-like cells with molecules to stimulate these PRRs during N protein expression before bulk RNA sequencing (RNA-seq). Specifically, we induced our cells with doxycycline for 48 hours and activated TLR7/8, TLR3, or RIG-I/MDA5 for 4 hours by treating cells with CL097 (single-stranded RNA mimic), poly(I:C) (polyinosinic:polycytidylic acid, double-stranded RNA mimic), or transfecting cells with poly(I:C), respectively. Multidimensional scaling (MDS) analysis reveals that untreated parental and N-expressing macrophages cluster together (fig. S2A). While RIG-I/MDA5–stimulated N-expressing cells cluster near RIG-I/MDA5–stimulated parental controls, TLR-stimulated N-expressing cells shift far from their respective parental controls (red and green arrows, fig. S2A). These data indicate that the N protein selectively drives large differential transcriptional changes during TLR stimulation.

For each PRR stimulation, our RNA-seq analysis identified differentially expressed genes [DEGs; false discovery rate (FDR) < 0.05] that fit into three categories: (i) DEGs unique to N-expressing cells, (ii) DEGs unique to parental cells, or (iii) DEGs shared by N-expressing and parental cells (Fig. 2A). As predicted by our MDS plots, the N protein dramatically alters the number of genes that are significantly regulated by TLR stimulation [TLR7/8 stimulation (left): 1422 DEGs unique to N-expressing cells versus 472 DEGs unique to parental cells; TLR3 stimulation (middle): 2827 DEGs unique to N-expressing cells versus 520 DEGs unique to parental cells]. RIG-I/MDA5 stimulation (right) triggers the largest gene expression changes in THP-1 macrophage-like cells, although a majority of these DEGs are shared, consistent with our MDS plots (1979 DEGs unique to N-expressing cells versus 1130 DEGs unique to parental cells; 3691 shared DEGs).

Fig. 2. — (A) Venn diagrams of DEGs in parental and N-expressing THP-1 macrophage-like cells after PRR stimulation. DEGs were defined as FDR < 0.05 (stimulated versus untreated), regardless of log₂ fold change. DEGs were classified as N-specific, parental-specific, or shared. CL097 (TLR7/8) altered 1974 shared and 1422 N-specific genes; poly(I:C) (TLR3) altered 1583 shared and 2827 N-specific genes; poly(I:C) transfection (RIG-I/MDA5) altered 3691 shared and 1979 N-specific genes. (B) Histograms showing N-induced changes in DEG expression, represented as delLog₂FC distributions for each PRR stimulation condition. delLog₂FC was calculated by comparing log₂ fold changes in N-expressing cells to parental cells under the same stimulation condition. (C) BP enrichment analysis of N-induced DEGs. Left: Top 10 GO terms from 69 N–up-regulated DEGs after TLR7/8 stimulation. Middle: Top 10 GO terms from 140 N–up-regulated DEGs after TLR3 stimulation. Right: Top 10 GO terms from 144 N–down-regulated DEGs after RIG-I/MDA5 stimulation. Bars above the −log₁₀(FDR) cutoff (red line, FDR = 0.05) indicate significant enrichment. (D to F) Fold change bar graphs of representative DEGs dramatically regulated by N in PRR-stimulated THP-1 macrophage-like cells. Log₂ fold changes of DEGs exhibiting enhanced expression in both parental and N-expressing cells to different extents, and those selectively up-regulated in N-expressing cells after TLR7/8 stimulation (D), TLR3 stimulation (E), or RIG-I/MDA5 stimulation (F). (G) Heatmap of innate immune response DEGs showing N-induced expression changes across PRR stimulations. DEGs from parental and N-expressing THP-1 macrophage-like cells were filtered using innate immune BP and Kyoto Encyclopedia of Genes and Genomes (KEGG) terms (GO:0140374, GO:0051607, GO:0009615, GO:0032728, GO:0070106, GO:0006954, GO:0045071, GO:0050729, GO:0045087, GO:0060337, GO:0032760, GO:0050728). Colors represent delLog₂FC [N versus WT log₂FC under the same PRR condition; see (B)]. Red, up-regulated by N; blue, down-regulated by N.

Next, we focused on the DEGs that fit into categories 1 and 3, which include DEGs unique to N-expressing cells and shared DEGs (annotated as N DEGs in Fig. 2A). Most notably, the primary effect of the N protein is modulated expression of these DEGs to varying degrees. To quantitatively determine the directionality and magnitude underlying N-mediated changes in gene expression, we calculated the difference in fold changes [delLog₂Fold Change (delLog₂FC)] for N DEGs between the N-expressing cells and parental cells for each PRR stimulation. We then plotted the delLog₂FC distribution of N DEGs for TLR or RLR stimulation (Fig. 2B). The histograms show that the delLog₂FC values slightly distributed to the right across all the PRR stimulation conditions, although there are several genes that are also further down-regulated.

To further unravel the pathways affected by the N protein, we performed gene ontology (GO) analysis on all hyperactivated genes (delLog₂FC > 1) and all suppressed genes (delLog₂FC < −1) across each receptor pathway (Fig. 2C and fig. S2B). However, due to the low number of suppressed genes for either TLR pathway, no GO terms were enriched and thus could not be plotted. Strikingly, the 69 and 140 genes highly up-regulated by N in TLR7/8 and TLR3 stimulation, respectively, are significantly enriched in inflammation-related GO terms (Fig. 2C, left and middle plots), highlighting the proinflammatory nature of the N protein on TLR pathways in THP-1 macrophage-like cells. In contrast, the 300 genes that are hyperactivated in RIG-I/MDA5 stimulation do not significantly overrepresent any GO terms (fig. S2B). Notably, the 144 genes down-regulated by N upon RIG-I/MDA5 stimulation are mostly related to the GO terms related to the antiviral response (Fig. 2C, right plot). This suggests that the N protein promotes the inflammatory response upon TLR stimulation and antagonizes the host IFN response upon RLR stimulation.

We next wanted to directly compare the expression fold changes of representative N DEGs related to innate immune response between the parental and N-expressing cells under TLR7/8 (Fig. 2D), TLR3 (Fig. 2E), and RIG-I/MDA5 stimulation (Fig. 2F) (table S1). During TLR7/8 and TLR3 stimulation, the presence of the N protein leads to enhanced expression of proinflammatory cytokine transcripts relative to parental cells. Specifically, genes involved in COVID-19 cytokine storm, including CCL4, CXCL10, TNF, and CCL8 (36), are dramatically elevated (log₂FC > 4) in N-expressing cells. IFNB1 is suppressed upon TLR3 stimulation in N-expressing cells. In contrast, the presence of the N protein generally suppresses expression of antiviral ISGs (including CH25H, OASL, and IFIT1), cytokines, and chemokines upon RIG-I/MDA5 stimulation. For example, both IFNB1 and CH25H are decreased by about 32-fold in the presence of the N protein relative to parental cells. These results demonstrate that N steers the TLR-mediated response to a more inflammatory direction and broadly dampens the RLR-mediated response.

Finally, we wanted to compare the effects of the N protein across all three stimulus conditions. Therefore, we took 729 DEGs shared among all six experimental groups (TLR7/8, TLR3, or RIG-I/MDA5 stimulated N-expressing or parental THP-1 macrophage-like cells) (table S2), and generated a heatmap with 106 genes that are enriched in inflammation and IFN pathways based on the GO terms in the BioMart database (37) (Fig. 2G). The innate immune genes clustered at the top and bottom of the heatmap demonstrate contrasting regulatory roles of the N protein in TLR7/8/3 and RIG-I/MDA5 pathways. For example, inflammatory innate immune genes are significantly up-regulated by N in the TLR pathways but down-regulated in the RIG-I/MDA5 pathways. These include inflammatory cytokines and chemokines such as TNF, CCL3, CCL4, CCL4L2, and CXCL10 (Fig. 2G, top) and components of the NF-κB pathway such as IRAK2, BCL3, RELB, TNFAIP3, and NFKBIA (Fig. 2G, top and bottom). In contrast, the IFN pathway components and negative regulators such as STAT1, IRF7, IRF9, and USP18 (Fig. 2G, bottom) and antiviral ISGs such as OASL, IFIT1, IFIT2, and IFIT3 (Fig. 2G, top) are dramatically suppressed in RIG-I/MDA5 pathway (Fig. 2G, bottom). CCL3L3, an inflammatory chemokine, and FFAR2, a fatty acid receptor involved in the inflammatory response, are both enhanced by N in all three PRR pathways (Fig. 2G, top). Together, these data emphasize the role of the N protein as a master manipulator of the cellular environment, which promotes the inflammatory response and dampens the antiviral response in a stimulus-specific manner.

Nucleocapsid-driven transcriptional changes are mediated by several shared and unique pathways

Given the strong proinflammatory gene signatures observed in TLR7/8- and TLR3-stimulated THP-1 macrophage-like cells expressing N, we leveraged our RNA-seq dataset to identify the pathways driving this phenotype. We first investigated transcriptional regulators by computationally predicting transcription factors most likely to govern the 69 and 140 up-regulated genes in TLR7/8 and TLR3 stimulation, respectively, using the TRRUST database (38, 39) (Fig. 3, A and B). The analysis identified several transcription factors with high statistical significance that may play a role in the hyperinflammatory phenotype we observed. Among the predicted transcription factors, we found RELA, NFKB1, the AP-1 factor JUN, and the immunoregulatory transcription factor FOXO1, to be shared across both the TLR7/8 and TLR3 N DEGs. STAT1 is also significantly enriched in both TLR stimulations; however, there are only a few genes (CXCL10, EDN1, and CCL3 in TLR7/8 stimulation; CXCL10, SOCS1, and IL27 in TLR3 stimulation) that map to the STAT1 network, while most of the genes are overlapped with the NFKB1 and RELA networks. TLR7/8-unique predicted transcription factors include the transcription factor for secondary inflammatory amplification, CEBPD, and the stress-responsive transcription factor NFIC. For the TLR3 axis, uniquely predicted transcription factors include IRF1 and the myeloid-associated transcription factor RUNX1. Overall, the fact that several transcription factors across the inflammation signaling axis are predicted to be affected by the N protein suggests that the N protein does not shift the PRR response in a pathway-specific manner.

Fig. 3. — (A and B) Transcription factor predictions of N-mediated gene up-regulation in TLR7/8 and TLR3 pathways. The 69 N–up-regulated genes after TLR7/8 stimulation and 140 N–up-regulated genes after TLR3 stimulation were analyzed for responsible transcription factors through TRRUST, a manually curated database of human and mouse transcriptional regulatory networks. The top 10 predicted transcription factors based on enrichment significance (FDR) are listed. The gradient colors correspond to combined enrichment scores. The size of the dots corresponds to the overlap ratio of the N–up-regulated genes in transcription factor–target genes in the database. The line segments exceeding −log₁₀(FDR) cutoff (red line, FDR = 0.05) shows significant enrichment. (C to E) Expression comparison of gene targets involved in multiple inflammatory pathways in parental THP-1 macrophage-like cells and Wuhan N–expressing THP-1 macrophage-like cells after different PRR stimulation. The log₂ fold changes of target genes of NF-κB and MAPK as well as proinflammatory transcription factors, AP-1 and EGR1, were extracted from the RNA-seq data and compared between parental THP-1 macrophage-like cells and Wuhan N–expressing THP-1 macrophage-like cells as bar graphs. The corresponding statistical significances of fold changes are indicated as asterisks: *P < 0.05; **P < 0.01; ***P < 0.001. Nonsignificant differences are not shown.

We wanted to confirm this hypothesis and thus also decided to take a closer look at the levels of target transcripts of some of these transcription factor pathways from our RNA-seq dataset. We hypothesized that if N expression favored any of these transcription factor pathways specifically, then their direct target transcript levels would be differentially affected. We plotted select target transcripts from parental and Wuhan-Hu-1 N–expressing THP-1 macrophage-like cells, focusing on the following transcription factor pathways: NF-κB, AP-1, EGR1, and MAPK (mitogen-activated protein kinase) (Fig. 3, C to E). We selected representative targets from the direct regulons of each transcription factor, including NF-κB targets (TNFAIP3, NFKBIA, NFKBIZ), genes co-regulated by NF-κB and AP-1 (RELB), genes co-regulated by NF-κB and MAPK signaling (TNF, IL6, IL1B), IRF3/ISGF targets (ISG15), and key transcription factors themselves (FOS, JUN, EGR1). We found that N increases almost all the target transcripts, except for EGR1, in the TLR3/7/8-stimulated THP-1 macrophage-like cells. Conversely, N decreases all the target transcripts, particularly IL6, in the RIG-I/MDA5–stimulated THP-1 macrophage-like cells. Together, these data suggest that the SARS-CoV-2 N protein shifts the cellular environment away from intracellular RNA sensing and potentiates extracellular RNA sensing. Moreover, since several transcription factors are affected, the mechanisms of the N protein may be taking place more upstream.

Hyperactivation of inflammation in macrophages is a conserved feature of pathogenic coronavirus nucleocapsids

We have thus far shown that expression of the SARS-CoV-2 Wuhan-Hu-1 N protein alone disrupts multiple signaling pathways in THP-1 macrophage-like cells, resulting in strong levels of inflammation compared to TLR stimulation alone. We were also curious as to whether this was a conserved phenotype across SARS-CoV-2 variants, as well as other pathogenic betacoronaviruses. Therefore, we decided to infect THP-1 macrophage-like cells expressing CD169 + ACE2 with SARS-CoV-2 WA1, Delta (B.1.617.2), Omicron BA.1 (B.1.1.529.1), Omicron BA.4 (B.1.1.529.4), or Omicron BA.5 (B.1.1.529.5) (Fig. 4A). Infection levels of these cells vary across the different SARS-CoV-2 variants and correlates with their reported severities. The Delta variant, for example, infects the THP-1 macrophage-like cells approximately 3 times better than the ancestral WA1 strain. Meanwhile, the Omicron BA.1 and BA.4 variants are 30 and 15 times worse, respectively, at infecting these cells relative to the WA1 strain. It is notable that as the Omicron lineage progresses, THP-1 macrophage-like cell infection level increases, with Omicron BA.5 infecting these cells at similar levels compared to WA1. This is in line with mouse pathogenesis data demonstrating more severe disease in mice with Omicron BA.5 compared to BA.1 (40). When we investigated levels of inflammatory transcript expression relative to WA1 infection, we found a few notable phenotypes (Fig. 4B). First, expression levels of several inflammatory transcripts in THP-1 macrophage-like cells infected with the Delta variant are similar to or lower than those from cells infected with WA1, despite significantly more virus, suggesting a potent immunosuppression mechanism by the virus. For Omicron BA.1 and BA.4, we found that IFNB1 and CXCL10 transcripts are generally up-regulated compared to those in WA1-infected cells, while other transcripts like TNF and CCL4 are trending lower, and CXCL8 is significantly lower. As we progressed along the Omicron lineage, we found that the BA.5 Omicron variant is consistently the most similar to the WA1 strain, with the exception of CXCL8, which is significantly higher in BA.5.

Fig. 4. — (A) Relative levels of N transcripts in SARS-CoV-2–infected THP-1 macrophage-like cells expressing CD169 + ACE2 receptors. The cells were infected with the indicated SARS-CoV-2 strain/variant at an MOI of 0.1 for 24 hours. The N transcripts were quantified with RT-qPCR and compared to uninfected cells. Results were plotted into bar graphs based on 2^−ΔΔCt. Data are representative of two independent experiments (ordinary one-way ANOVA; asterisks indicate statistically significant differences relative to WA1: **P < 0.01; ****P < 0.0001; nonsignificant differences are not shown). (B) Representative inflammation-related transcript expression profile induced by different SARS-CoV-2 variant infections in CD169 + ACE2–expressing THP-1 macrophage-like cells. The cells were inoculated with the indicated variant at an MOI of 0.1 for 24 hours. The expression of indicated transcripts was quantified by RT-qPCR. The relative fold changes were evaluated relative to uninfected and plotted relative to WA1. Data are representative of two independent experiments (two-way ANOVA, multiple comparisons; asterisks indicate statistically significant differences within each transcript analyzed relative to WA1: ****P < 0.0001; nonsignificant differences are not shown). (C to E) Expression of *TNF*, *CCL4*, and *CXCL10* in THP-1 macrophage-like cells expressing different betacoronavirus N proteins. The cell lines were treated with doxycycline (2 μg/ml) for 48 hours to express N from the indicated coronavirus. The cells were then either untreated or stimulated with CL097 for 4 hours. Transcripts were quantified by RT-qPCR. The relative fold changes are evaluated relative to unstimulated parental cells and plotted based on 2^−ΔΔCt. Data are representative of two independent experiments (two-way ANOVA, multiple comparisons; asterisks indicate statistically significant differences within each transcript analyzed relative to parental cells: *P < 0.05; **P < 0.01; ****P < 0.0001; nonsignificant differences are not shown).

To confirm this phenotype in a more physiologically relevant system, we isolated primary human monocytes from whole blood obtained from two independent donors through the UCLA (University of California, Los Angeles) Virology Core. We then differentiated the primary monocytes into macrophages with macrophage colony-stimulating factor (M-CSF) and infected them with the WA1 strain, Delta, and Omicron BA.1 variants (fig. S3A). While we observed low levels of infection compared to the CD169 + ACE2 THP-1 macrophage-like cells, we observed a similar trend where WA1 strain and Delta variant infection levels are high relative to Omicron BA.1, which show very low infection in primary monocyte-derived macrophages. As before, we also measured inflammation-related transcript levels and saw that these levels typically correlated with levels of infection, where more infection results in more inflammation-related gene expression (fig. S3, B and C). While almost no Omicron BA.1 RNA was detected in the infected samples, a nearly 2-fold up-regulation of inflammation-related gene expression was still observed, confirming that even exposure alone to the virion will activate the macrophages.

Overall, we observed differences in inflammation-related gene induction across various SARS-CoV-2 variants, with the Delta variant demonstrating strong suppression of the inflammatory and antiviral response. However, since infection is a complex process involving several steps and proteins, we moved toward isolating the various N proteins in macrophages to better investigate conserved mechanisms. Similarly to the Wuhan-Hu-1 N–expressing THP-1 macrophage-like cell line, we generated new cell lines that express either the Delta or Omicron BA.1 N proteins by introducing the following nonsynonymous mutations: D63G, R203M, and D377Y (for Delta), and P13L, deletion of residues 31 to 33, R203K, and G204R (for Omicron BA.1). Additionally, we generated cell lines expressing the SARS-CoV-1 N (95% identity to Wuhan-Hu-1) and MERS-CoV N proteins (65% identity to Wuhan-Hu-1) (41) (fig. S3, D to F). We chose to focus on the TLR7/8 signaling axis, since this is where we observed the hyperinflammatory phenotype. As done in Fig. 1G, we induced N expression with doxycycline for 48 hours, followed by a 4-hour stimulation with CL097. We then measured three representative inflammation-related transcripts informed by our RNA-seq data: TNF, CCL4, and CXCL10 (Fig. 4, C to E). There was no up-regulation of TNF or CCL4 under basal conditions for any coronavirus N protein, with the exception of the Delta N protein, which up-regulated basal CCL4 by 3-fold, although this was not statistically significant. After CL097 treatment, TNF expression was significantly down-regulated by both Delta and Omicron BA.1 yet significantly up-regulated by MERS-CoV (Fig. 4C). For CCL4, all SARS coronavirus N proteins significantly down-regulated expression to varying degrees, while MERS N protein up-regulated expression (Fig. 4D). Across all coronavirus N proteins tested, CXCL10 was already elevated under basal conditions, although only Delta, Omicron BA.1, and SARS-CoV-1 were statistically significant. Under CL097 treatment, all N proteins tested significantly induced more CXCL10 expression, with the exception of Wuhan-Hu-1 N, which is trending upward (Fig. 4E). This is especially striking as there are marked differences in the expression levels of these N proteins (fig. S3D); however, the Delta variant N seems to highly promote CXCL10 under basal and stimulated levels (approximately 130- and 50-fold, respectively), while the Omicron BA.1 N does not do this to the same degree (approximately 25- and 35-fold, respectively), despite similar levels of robust protein expression (fig. S3D). Altogether, these findings indicate that coronavirus N proteins differentially modulate TLR7/8-driven inflammatory responses in macrophages, with SARS-related N proteins promoting a CXCL10-biased transcriptional profile while suppressing TNF and CCL4, whereas MERS-CoV N exhibits a distinct regulatory pattern.

SARS-CoV-2 N engages cGAS to modulate macrophage inflammatory signaling

Our data thus far has demonstrated a proinflammatory role for the SARS-CoV-2 N protein that affects several pathways regulating inflammation-related gene expression, yet the protein-protein interactions underlying this phenotype are unclear. We specifically sought to determine whether differences in the inflammatory activity of distinct SARS-CoV-2 N strains and variants were associated with differential host protein interactions. We therefore compared THP-1 macrophage-like cell lines expressing the Delta or Omicron BA.1 N proteins, which exhibited stronger and more consistent expression than Wuhan-Hu-1 N (fig. S3D). Under basal conditions, Delta N induced a more pronounced inflammatory phenotype, most notably through CXCL10 and CCL4 up-regulation (Fig. 4, D and E). We reasoned that host factors preferentially interacting with Delta N could represent key mediators of N-driven inflammation. To identify such factors, we performed affinity purification–mass spectrometry (AP-MS) under unstimulated conditions in THP-1 macrophage-like cells expressing either Delta or Omicron BA.1 N (Fig. 5A). THP-1 macrophage-like stable cell lines without doxycycline treatment served as negative controls. We then analyzed interaction data through Significance Analysis of INTeractome (SAINT). High confidence interactors (Bayesian FDR ≤ 0.05 and average spectral count ≥ 11) were selected for fold change calculation. We then plotted the 39 host proteins that interact with at least one N protein into a protein-protein interaction network (PPI) (Fig. 5B) to visualize differences in interaction candidates.

Fig. 5. — (A) Schematic of AP-MS used to identify host factors interacting with Delta and Omicron BA.1 variant N proteins in macrophages. This figure was created using BioRender.com with the following credit: Created in BioRender. Li, M. (2026) https://BioRender.com/y4pvy47. (B) High-confidence host factors interacting with SARS-CoV-2 variant N proteins. Thirty-nine proteins passing SAINT BFDR ≤ 0.05 and average spectral count ≥ 11 were analyzed for induction fold changes in Delta N–expressing versus control cells and Omicron BA.1 N-expressing versus control cells. Colors in the PPI represent Delta Log₂ FC versus Omicron BA.1 Log₂ FC (red: stronger interaction with Delta N; blue: stronger interaction with Omicron BA.1 N). Dashed lines indicate interaction with at least one variant N, and solid lines indicate interaction with N proteins from both variants. (C) Functional annotation clustering of high-confidence N interactors. The 39 host proteins were analyzed using DAVID GO and pathway enrichment and grouped by enrichment score. Functional annotation terms are shown on the y axis and FDR on the x axis. Bars exceeding the −log₁₀(FDR) cutoff (red line, FDR = 0.05) indicate significant enrichment. (D). Orthogonal validation of Delta and Omicron BA.1 N protein interactors. HEK293T cells were cotransfected with myc-tagged Delta or Omicron BA.1 N and host factors (3xFLAG-cGAS, 3xFLAG-PABP1, V5-IGF2BP2) or empty vector control. Cells were lysed 24 hours post-transfection and immunoprecipitated with anti–c-myc agarose beads, and binding was detected by immunoblot. (E to G) cGAS inhibition amplifies inflammation-related transcript expression. Cells were pretreated with 10 μM G140 or DMSO for 1 hour before 4-hour CL097 stimulation. Indicated gene expression was measured by RT-qPCR and compared to each cell line’s respective DMSO-treated controls. Data are representative of two independent experiments (two-way ANOVA, multiple comparisons; asterisks indicate statistically significant differences between DMSO and G140 treatments: *P < 0.05; ***P < 0.001; ****P < 0.0001; nonsignificant differences are not shown).

To appreciate the host biological functions that N proteins interfere with, we clustered the GO and pathway annotations of the 39 host proteins through the DAVID database (Fig. 5C and table S3). Compared to the Omicron BA.1 variant, the Delta variant N protein interacts more with a group of proteins related to cytoplasmic stress granule formation, such as MBNL1, PABP1, IGF2BP2 (gene name: IF2B2), LAR1B, MOV10, and IGF2BP3 (gene name: IF2B3) (Fig. 5B and table S3). The Delta variant N protein shows high confidence interactions with several ribosomal proteins (RL21, RLA2, RL24, RL28, RS12), suggesting N-mediated interference of host translation (Fig. 5B and table S3). Nearly half of the N interactors are enriched in posttranslational modification pathways of isopeptide bond and ubiquitin-like modifier (Ubl) conjugation. It is known that isopeptide bonds are commonly formed during ubiquitination and SUMOylation, which implies that the N protein interactors may also be involved in these processes. Although not ranked among the top three annotations, both Delta and Omicron BA.1 variant N proteins significantly interfere with the host antiviral defense response (table S3, enrichment score: 1.42). Consistent with the PPI network analysis, the Omicron BA.1 N protein displays stronger interactions with antiviral effectors ISG15 and OASL (highlighted in Fig. 5B). These results suggest that the SARS-CoV-2 N protein potentially influences gene expression in THP-1 macrophage-like cells by modulating stress granule formation, host translation, and posttranslational modification.

Since stress granules are known to be associated with NF-κB signaling and inflammation (42–45), we decided to probe further into this potential mechanism. To provide orthogonal validation of the interaction identified by AP-MS, we performed confirmatory coimmunoprecipitation experiments in a HEK293T overexpression system. Specifically, we overexpressed myc-tagged Delta or Omicron BA.1 variant N in HEK293T cells together with different stress granule–related host factors (cGAS, PABP1, or IGF2BP2 tagged with 3xFLAG or V5). We pulled down the myc-tagged Delta or Omicron BA.1 variant N protein and detected host factors that coimmunoprecipitated with the N protein (Fig. 5D). Across our experiments, we found that IGF2BP2 did not pull down with either Delta or Omicron BA.1 N proteins, while both N proteins robustly pulled down PABP1 (Fig. 5D). The cGAS interaction was indeed unique to the Delta variant N protein, as cGAS only robustly immunoprecipitated with Delta variant N, while weakly precipitated with Omicron BA.1 N (Fig. 5D), suggesting that Delta N may interfere with cGAS-related signaling.

With the N-cGAS interaction validated, we decided to probe this further on both the virus and host sides. On the virus side, we chose to identify the N protein residues that mediate the binding to cGAS. We generated several myc-tagged Wuhan-Hu-1 N protein constructs, each carrying an individual point mutation from the Delta or Omicron BA.1 variant. We then transfected 3xFLAG-tagged cGAS constructs together with different N protein mutants into the HEK293T cells. We also included cotransfection of N proteins of the Wuhan-Hu-1 strain, Delta, or Omicron BA.1 variants and 3xFLAG-cGAS as controls. We found that N with the Delta-specific D63G mutation pulls down cGAS similarly to the Delta variant N, while other mutations at residues 203 and 204 have minor contributions (fig. S4A). This identifies glycine residue 63 as a critical viral genetic determinant for N interaction with cGAS, which likely confers increased inflammation. Residue 63 is located in the N-terminal domain (NTD), which is responsible for RNA binding.

On the host side, we were interested in further understanding how N interaction with cGAS may modulate inflammation-related signaling. Previous work has demonstrated an inhibitory role of the cGAS-STING axis on TLR9 signaling in human dendritic cells and TLR7 and TLR9 in murine macrophages (46, 47). To assess the role of cGAS in the proinflammatory phenotype we observed, we used the cGAS inhibitor G140, which binds the cGAS catalytic site and prevents cGAS from generating cyclic guanosine monophosphate-adenosine monophosphate (cGAMP) and activating the double-stranded DNA sensor STING (48). Specifically, we pretreated our THP-1 macrophage-like cells with 10 μM G140 or dimethyl sulfoxide (DMSO) for 1 hour before TLR7/8 stimulation with CL097. When cGAS is inhibited, we consistently observed further stimulation of key inflammation-related transcripts like TNF, CCL4, and CXCL10 in the parental cell line (Fig. 5, E to G). Notably, while CXCL10 and CCL4 were also up-regulated during G140 treatment in both the Delta and Omicron BA.1 N-expressing cell lines, the master regulator of inflammation, TNF, was not up-regulated in these cell lines (Fig. 5E). Because expression was normalized to each cell line’s DMSO control, this could suggest that TNF expression may already be near maximal levels in the presence of the N protein, limiting further induction by G140. Together, these data further establish a modulatory role for the cGAS-STING signaling axis during TLR activation, specifically TLR7/8. Moreover, these data support a model in which N binding to cGAS inhibits its activity, similar to the cGAS inhibitor G140, thereby dampening STING signaling and relieving modulatory constraints on the TLR7/8 axis to amplify inflammatory gene expression.

SARS-CoV-2 N variants differentially shape macrophage inflammatory responses and endothelial barrier integrity

Having characterized N protein–driven proinflammatory transcriptional and signaling changes, we next examined the downstream effects on secreted factors and vascular integrity. Using a multiplex immunoassay, we determined the concentration of 71 different cytokines in the medium of parental, Wuhan-Hu-1 N–, Delta variant N–, or Omicron BA.1 variant N–expressing THP-1 macrophage-like cells with or without TLR7/8 stimulation (table S4). To examine how SARS-CoV-2 N variants alter macrophage inflammatory responses, we first quantified cytokine secretion as log₂ fold changes relative to unstimulated controls. Cytokines were then ranked based on their magnitude of induction in parental THP-1 macrophage-like cells following TLR7/8 stimulation, and the 25 most strongly modulated cytokines were selected for further analysis (Fig. 6A). We next compared the stimulation responses of these parental-defined cytokines across cell lines expressing the Wuhan-Hu-1, Delta, or Omicron BA.1 N proteins (Fig. 6A). Visualization of these responses in a clustered heatmap revealed variant-specific remodeling of the TLR7/8-induced cytokine program. Statistical comparisons of log₂ fold-change values between parental and N-expressing cells identified multiple cytokines that were significantly altered by each strain/variant N expression, highlighting variant-dependent modulation of macrophage inflammatory signaling. Among these, we found that the protein counterparts to the transcripts we measured were indeed elevated for the Wuhan-Hu-1 N–expressing cell lines, namely, TNF-α (gene name: TNF), MIP-1β (gene name: CCL4), interleukin-10 (IL-10) (gene name: CXCL10), and IL-8 (gene name: CXCL8).

Fig. 6. — (A) Heatmap showing TLR7/8-induced cytokine responses across THP-1 macrophage-like cell lines expressing none, Wuhan-Hu-1, Delta, or Omicron BA.1 N proteins. Values represent mean log₂ fold change relative to untreated controls. Cytokines shown are the top 25 most modulated relative to parental cells (Wilcoxon test; asterisks indicate significant differences compared to parental cells: *P < 0.05; nonsignificant differences are not shown). (B and C) Global cytokine responses to TLR7 stimulation. Cytokines significantly induced (B) or suppressed (C) by TLR7/8 stimulation in parental THP-1 macrophage-like cells were used as fixed reference sets to assess how N variants modulate cytokine programs. Values represent log₂ fold change relative to untreated controls. Each point represents one cytokine. Box plots show the distribution of cytokine responses across parental and variant N-expressing cells (paired Wilcoxon test; asterisks indicate significant differences compared to parental cells: *P < 0.05; **P < 0.01; nonsignificant differences are not shown). (D) Schematic of THP-1 macrophage-like cell-conditioned medium experiment. This figure was created using BioRender.com with the following credit: Created in BioRender. Li, M. (2026) https://BioRender.com/0u1valk. (E and F) Fold change in TEER measurements in HCAECs (E) and BMECs (F) treated with conditioned medium from the indicated cell lines. Data shown are fold changes in TEER observed 24 hours post-treatment, relative to 0 hours post-treatment. TNF-α was included as a positive control for barrier disruption. All graphs show mean ± SD. Data shown are combined from two independent experiments (asterisks indicate statistically significant differences; two-way ANOVA and Šidák’s multiple comparisons: *P < 0.05; **P < 0.001; ****P < 0.0001; nonsignificant differences are not shown).

To better summarize the effects of each variant on cytokine secretion from a global perspective, we generated datasets of cytokines that were significantly up-regulated or down-regulated following stimulation in parental cells and used them to define a reference inflammatory program. For each cytokine, stimulation responses were expressed as log₂ fold changes relative to baseline of the respective cell line and compared across cell lines expressing the Wuhan-Hu-1, Delta, or Omicron BA.1 N proteins. Box plot visualization revealed that expression of the Wuhan-Hu-1 N protein broadly increased the magnitude of cytokine induction, with both up-regulated and down-regulated cytokines exhibiting significantly greater stimulation responses compared to parental cells (Fig. 6, B and C). In terms of global trends, Delta N expression produced responses largely comparable to parental macrophages in both sets; however, there were several cytokines that were significantly amplified and dampened; therefore, our observation is likely due to an overall net zero global change. Strikingly, cells expressing the Omicron BA.1 N protein displayed a global reduction in cytokine responses, with both up-regulated and down-regulated cytokine programs shifting toward lower stimulation magnitudes, although this was only significant for the up-regulated cytokine set (Fig. 6B). These results indicate that SARS-CoV-2 N variants differentially tune macrophage cytokine secretion, with Wuhan-Hu-1 N globally amplifying and Omicron BA.1 N globally dampening TLR7/8-driven cytokine programs.

Several studies have demonstrated that SARS-CoV-2 causes vascular damage, correlating to cardiac pathologies and, to a lesser extent, neuropathologies (49–51). To examine the functional impact of this cytokine-rich environment on vascular integrity, we transferred conditioned medium from unstimulated and stimulated THP-1 macrophage-like cells onto two previously established models of heart and brain endothelial barriers (52, 53) (Fig. 6D). Specifically, we used primary human coronary artery endothelial cells (HCAECs) and human pluripotent stem cell–derived brain microvascular endothelial cells (BMECs), which have been previously used to model virus-barrier interactions (54–57). We then measured transendothelial electrical resistance (TEER) as a proxy for barrier function. Because Wuhan-Hu-1 N enhanced macrophage cytokine secretion, whereas Delta N produced a response comparable to parental cells, these variants were selected to test whether differential inflammatory signaling alters endothelial barrier integrity. Omicron BA.1 N, however, suppressed macrophage cytokine secretion and was therefore not prioritized for endothelial barrier assays designed to test whether heightened inflammatory signaling promotes barrier dysfunction. As a positive control, we treated our barrier cells with TNF-α, which decreased barrier integrity to approximately 20% in both systems (Fig. 6, E and F). When we tested the barrier integrity of our HCAECs 1 day after treatment with the medium from our stimulated cells, we found that the parental cell-conditioned medium decreased barrier integrity down to 75%, while the medium from Wuhan-Hu-1 N– and Delta N–expressing cells decreases barrier integrity down to 61% and 47%, respectively, with the Delta N significantly promoting barrier leakiness compared to parental cells (Fig. 6E). BMECs seem to be more responsive to this conditioned medium, as the conditioned medium from all cell lines decreased BMEC barrier integrity down to approximately 60% (Fig. 6F). The effects of the N-expressing macrophage conditioned medium on the HCAECs suggest that the N protein could be mediating the cardiac damage seen in severe COVID-19. For BMECs, these results suggest that N alone does not disrupt blood-brain barrier integrity, and that blood-brain barrier dysfunction in severe COVID-19 is likely driven by other viral or host factors. Collectively, these data link SARS-CoV-2 N–driven macrophage inflammation to endothelial barrier dysfunction, revealing variant-specific effects that may contribute to vascular pathology during infection.

DISCUSSION

SARS-CoV-2, like many pathogenic viruses, has several immune antagonists that suppress host antiviral defenses to enable viral replication and spread. Multiple studies have identified several SARS-CoV-2 proteins—most notably nonstructural protein 1 (nsp1), nsp3, and ORF6—as potent IFN antagonists (16–18). The N protein, though primarily recognized for its role in genome packaging, has also been implicated in blunting type I IFN response, particularly in the context of RIG-I activation in immortalized cell lines like HEK293T (9, 16–18). Our initial work recapitulates these findings, showing that N expression in HEK293T cells suppresses antiviral signaling downstream of RIG-I. To better reflect the physiological interactions in an immune-relevant system during infection, we investigated the activation of several PRRs in THP-1 monocyte-derived macrophages. Through this work, we identified a proinflammatory role for the N protein in a TLR-stimulated immune cell context. Moreover, we demonstrated that the N protein from various SARS-CoV-2 variants and other pathogenic betacoronaviruses is hyperinflammatory to varying degrees and with distinct mechanisms. Through a comparative proteomic analysis, we identified cGAS as a critical interactor of the N protein that is likely driving the proinflammatory phenotype. We found that cGAS inhibition during TLR7/8 stimulation results in amplified expression of inflammation-related transcripts, suggesting a regulatory role of the cGAS-STING pathway during TLR7/8 signaling in macrophages. This promotes a model where N-cGAS interactions likely dampen cGAS-STING signaling, thus removing regulatory constraints on TLR7/8 signaling and amplifying inflammation. Finally, we demonstrated that the N protein leads to variant-specific global changes in cytokine production, which can result in barrier disruption of cardiovascular endothelial cells. Our work thus far establishes a previously unidentified and surprising function of the N protein, where it may interfere with cGAS to trigger a hyperactive inflammatory response that contributes to cytokine storm and severe COVID-19.

The N protein is the most abundantly expressed SARS-CoV-2 protein. In addition to its inhibitory role in innate immune responses, several clinical studies also revealed its proinflammatory role in facilitating disease severity (58, 59). Our RNA-seq data revealed that the N protein preferentially suppresses the expression of ISGs upon RIG-I/MDA5 stimulation, indicating that N-mediated IFN antagonism is unique to canonical RLR signaling. While this result supports existing literature, we also found that the N protein amplifies expression of inflammation-related transcripts upon TLR stimulation, suggesting that the N protein selectively manipulates immune signaling depending on the pathway engaged. These findings raise important questions regarding the mechanism of N-mediated immune modulation and the roles of the N protein in pathogenesis, specifically whether the N protein modulates similar pathways from HEK293T cells, such as TRIM25-mediated RIG-I ubiquitination, in macrophages, or if alternative macrophage-specific pathways are involved (9, 24). Moreover, while it is intuitive for pathogenic viruses to antagonize antiviral responses, it is less understood as to why pushing macrophages into a more proinflammatory state is evolutionarily advantageous. Future studies should explore whether this immune modulation is cell type dependent and how it ultimately promotes SARS-CoV-2 pathogenesis.

Macrophages not only serve as key innate immune sensors but also play a central role in the hyperinflammatory responses associated with severe COVID-19 (32). Mononuclear phagocytes (MNPs) comprise 80% of the cells in bronchoalveolar lavage fluid from COVID-19 patients, and several of these MNPs are inflammatory monocyte-derived macrophages (29, 32); yet, the mechanisms by which they are hyperactivated are unclear. Specifically, the field remains unsure as to whether indirect stimulation by peripheral SARS-CoV-2–infected cells, or if direct encounter with SARS-CoV-2 and interaction with viral immune modulators such as the N protein, leads to chronic activation. One hypothesis is that delayed or chronic type I IFN signaling in COVID-19 contributes to the persistent, hyperactivated state of macrophages. One study showed that SARS-CoV-2 infection triggers TLR7 in plasmacytoid dendritic cells, leading to IFN-α production that drives macrophages to a proinflammatory state (60). An alternative hypothesis is that internalized SARS-CoV-2 results in macrophage-mediated hyperinflammation; however, whether SARS-CoV-2 can productively infect macrophages remains controversial. Ultimately, SARS-CoV-2 infection of macrophages largely depends on the expression of cellular receptors used for virus entry by various macrophage subtypes (27, 31). ACE2 is lowly expressed in macrophages; therefore, SARS-CoV-2 harnesses lectin receptor–mediated phagocytosis to enter these cells. Siglec-1/CD169 mediates SARS-CoV-2 entry in monocyte-derived macrophages and alveolar macrophages, although with little to no production of infectious progeny (30), while CD209 mediates virus entry in interstitial macrophages (27). Therefore, we chose to investigate whether productive infection is required to induce transcriptional changes, or if packaged N protein in the virion is sufficient. Our data demonstrated that both mechanisms drive changes in macrophages, and with direct infection amplifying inflammation more than virion exposure. We also identify distinct roles of the N protein depending on the step of the virus life cycle, where incoming N protein dampens the antiviral response, and newly synthesized N protein amplifies inflammation. This suggests that the N protein has two roles in modulating the innate immune response during infection: (i) dampening the antiviral response to allow for effective replication while (ii) inducing a proinflammatory response to promote further dissemination of the virus. Ongoing work will investigate whether these roles are temporally and spatially distinct, and whether blockade of newly synthesized N protein will prevent its proinflammatory functions.

With the continuing spread of SARS-CoV-2, we are witnessing viral evolution and human adaptation in real time as newer variants emerge and gain fitness advantages. Clinical studies showed a higher disease severity in individuals infected with the ancestral SARS-CoV-2 strain and Delta variant compared to the Omicron BA.1 variant. This was reflected by a greater magnitude of inflammatory cytokines and chemokines in the plasma induced by infection with the Wuhan-Hu-1 strain and Delta variant (61). Moreover, during comparisons of Omicron BA.1 and BA.5 in mice, a previous study demonstrated significantly higher pathogenesis mediated by Omicron BA.5 relative to BA.1 (40). When infecting THP-1 macrophage-like cells with each of these variants, as well as Omicron BA.4, we found that infection levels correlate with disease severity, where variants with higher infection levels also have higher reported pathogenicity. Moreover, the infection levels are consistent in primary human monocyte-derived macrophages.

Introduction of the Delta and Omicron BA.1 variant N proteins into our THP-1 macrophage-like cell system leads to the discovery of several important differences. First, both the N proteins of the Delta and Omicron BA.1 variants express notably better compared to the Wuhan-Hu-1 strain N protein despite all constructs being codon optimized. This suggests that these variants have developed more efficient expression in human cells. In addition to introducing the variant-specific N proteins, we also generated THP-1 macrophage-like cells expressing SARS-CoV-1 and MERS-CoV N proteins. When investigating the induction levels of select inflammation-related transcripts, we found SARS-related virus N proteins to preferentially amplify CXCL10, while MERS amplifies TNF and CCL4 transcripts. Therefore, while all N proteins tested were hyperinflammatory, each family had distinct mechanisms. Even within the SARS-CoV-2 variant N proteins, there were striking differences in the inflammatory transcript induction levels. Specifically, we found that the Delta variant N protein can promote CXCL10 and CCL4 under basal conditions, while the Omicron BA.1 variant N protein weakly induces CXCL10, despite similar N expression levels. Ultimately, the mechanisms driving these different inflammatory profiles between variants remain elusive. Compared to the Wuhan-Hu-1 strain, the N protein of the Delta variant contains three mutations: D63G, R203M, and D377Y, while the Omicron BA.1 variant has four mutations: P13L, deletion of 31–33, R203K, and G204R. Both the N proteins of the Delta and Omicron BA.1 variants share a mutation at residue 203, but the Delta variant has a more drastic change than the Omicron BA.1 variant. Additionally, it was reported that both the R203M and D377Y mutations in the N protein can outcompete RIG-I to bind to viral RNA, preventing IFN activation (25). In the Omicron BA.1 variant, the R203K mutation, together with G204R, has been isolated as adaptive mutations associated with enhanced viral replication and infectivity (59, 62). Other studies suggested that post-translational modifications (PTMs) at K203 in the Omicron BA.1 variant N protein contribute to increased virulence by suppressing IFN expression (63). While past studies have focused on the role of SARS-CoV-2 variant N proteins in IFN antagonism, future studies should investigate the roles of each mutation in promoting inflammation in immune cells.

While several studies have identified mechanisms of N-mediated dampening of the IFN response, it is not understood how the N protein may promote inflammation. To begin unraveling this mechanism, we performed comparative proteomics between the hyperinflammatory Delta variant N protein and the less inflammatory Omicron BA.1 variant N protein. The proteomic results showed that the N protein of the Delta variant interacts more strongly with stress granule components, translation machinery, and, most notably, the DNA sensor cGAS. Stress granule formation is an important antiviral mechanism, where host translation stalling sequesters viral RNA transcripts and promotes the innate immune response through condensation of cytosolic sensors, such as cGAS (43, 44, 64). Current studies suggest that the N protein can be integrated into stress granules though liquid-liquid phase separation and interfere with stress granule–mediated antiviral activity by interacting with critical components such as PKR, G3BP1, and MOV10 (23, 65). Our immunoprecipitation results validated stronger interaction of the Delta variant N protein with cGAS, which indicates variant-dependent effects on immune sensors associated with stress granules. It is reported that cGAS binds to G3BP1 and DDX3X to undergo condensation into stress granules for prompt activation (43, 44). Another related function of cGAS seems to be regulatory modulation of TLR-related signaling. Specifically, cGAS-STING signaling seems to dampen TLR9 signaling in human dendritic cells and TLR7 and TLR9 in murine macrophages (46, 47). In our hands, inhibition of cGAS during TLR7/8 stimulation seems to amplify expression of downstream transcripts, suggesting a conserved modulatory role of the cGAS-STING pathway in macrophages. N protein binding to cGAS may also inhibit its signaling, thus removing constrains of TLR signaling and amplifying the inflammatory response. Overall, follow-up work will focus on rigorous dissection of this mechanism and whether this is conserved across pathogenic betacoronaviruses, since this could be a critical therapeutic target for mediating severe COVID-19 inflammation.

Finally, we sought to better understand the clinical relevance of the hyperactive phenotypes we observed. We characterized cytokines secreted from TLR7/8-stimulated macrophages expressing different N proteins and found up-regulation of several inflammatory factors in line with our transcriptomic analysis. Moreover, we found a variant-specific trend of global changes to these cytokine secretion profiles, where the Wuhan-Hu-1 N protein globally amplifies the cytokines secreted and the Omicron BA.1 N protein globally dampens the cytokines secreted. While the Delta N protein did not have a global effect, several cytokines were both amplified and dampened, suggesting a precise remodeling of the inflammatory response. Among the modulated cytokines, MIP-1β (gene name: CCL4) stood out as most strongly correlating with cytokine signatures from COVID-19 patients (61). MIP-1β is a macrophage-secreted chemokine that recruits monocytes, T cells, and natural killer cells, and is consistently elevated in severe SARS-CoV-2 infection (61, 66). It is also capable of activating microglia and astrocytes and has been implicated in HIV-associated neurocognitive disorders, stroke, and blood-brain barrier leakage in COVID-19 (66–69). In line with these pathologies, we chose to investigate the effects of the medium from these hyperactivated macrophages on endothelial cell barriers of the heart and brain. We observed HCAEC barrier disruption in our model, which has important implications for COVID-19–related cardiovascular pathology. HCAECs line the coronary arteries and are essential for maintaining vascular integrity and regulating immune cell trafficking (70). Disruption of this barrier can promote myocarditis, atherosclerotic plaque destabilization, microvascular dysfunction, and thrombosis, particularly under proinflammatory conditions. In the context of COVID-19, such vascular complications have been widely reported (71). These data suggest that the N protein can, at least in part, mediate the cardiovascular pathology seen in severe COVID-19. While we observed induced BMEC permeability in vitro, this was observed under all conditions, suggesting that the N protein alone does not mediate blood-brain barrier leakage in our system; however, it is possible that the N protein may have stronger effects on other BMEC models. Ultimately, endothelial dysfunction, in addition to hyperinflammation, has been shown to contribute to disease severity and death in patients infected with SARS-CoV-2 (72). Our data therefore support a mechanistic model in which the N protein alone is sufficient to drive pathogenic cytokine responses in macrophages that in turn compromise critical endothelial barriers in the heart. Together, these findings highlight N as a potent immune modulator with potential to exacerbate systemic complications of SARS-CoV-2 infection. Future studies should explore the feasibility of pharmacologically targeting N, or its downstream signaling pathways, to mitigate COVID-19–associated inflammatory damage.

Overall, this study reveals a previously unrecognized proinflammatory function of the SARS-CoV-2 N protein. We show that newly synthesized N protein, particularly from the Wuhan-Hu-1 strain and Delta variant, amplifies TLR-driven inflammation in macrophages. Proteomic analysis implicates N interactions with stress granule proteins and cGAS as a potential mechanism that promotes immune activation and inflammatory gene expression. These mechanisms drive excessive cytokine production in macrophages that disrupts cardiovascular integrity, pointing to a direct viral contribution to cytokine storm and vascular pathology. Together, these findings connect molecular events within infected cells to clinical outcomes in severe COVID-19. Future therapeutic strategies that target N-mediated immune modulation could reduce the inflammatory burden of SARS-CoV-2.

MATERIALS AND METHODS

Ethics statement

Whole blood samples used for monocyte purification were collected in accordance with ethical guidelines for human subject research and approved by the Institutional Review Board (IRB; approval no. IRB-11-0443). All human pluripotent stem cell experiments were approved by the UCLA Human Pluripotent Stem Cell Research Oversight (hPSCRO) Committee (protocol #2019-007-05B) and conducted in accordance with UCLA Institutional Biosafety Committee requirements.

Cell culture

THP-1 human monocytes were obtained from the American Type Culture Collection (ATCC). The CD169-expressing THP-1 and CD169 + ACE2–expressing THP-1 cell lines were provided by R. Gummuluru at Boston University. The THP-1, CD169-expressing THP-1, and CD169 + ACE2–expressing cell lines were cultured in RPMI 1640 (Gibco) supplemented with 10% fetal bovine serum (FBS), 1× penicillin/streptomycin (P/S; Fisher Scientific), 1× nonessential amino acids (NEAA; Gibco), and 0.05 mM β-mercaptoethanol (Sigma-Aldrich). Human embryonic stem cells (H9; WA09, WiCell) were cultured on T 75-cm² flasks coated with Matrigel (Corning) solution (0.2 mg/ml) in 1:1 Dulbecco’s modified Eagle’s medium (DMEM)/Ham’s F12 (Gibco). Stem cells were grown with mTeSR1 (STEMCELL Technologies) with daily medium changes. HEK293T cells were cultured in DMEM (VWR) supplemented with 10% FBS. HCAECs (Cell Applications) were cultured in Human Meso Endo Growth Medium (Cell Applications) on flasks coated with 0.1% gelatin solution (Millipore Sigma).

THP-1 and primary monocyte differentiation into macrophages

To differentiate THP-1, CD169-expressing THP-1, and CD169 + ACE2–expressing THP-1 cells into macrophages, the cells were subjected to a 24-hour stimulation with phorbol 12-myristate 13-acetate (PMA; 50 ng/ml; Sigma-Aldrich) in RPMI 1640 supplemented with 10% human AB serum (Sigma-Aldrich), 1× NEAA, and 1× P/S, which is followed by a 24-hour rest in human serum containing RPMI 1640.

To obtain primary monocyte-derived macrophages, we purified monocytes from human whole blood samples provided by UCLA/CFAR (Centers for AIDS Research) Virology Core Laboratory with RosetteSep kit (STEMCELL Technologies). The purified monocytes were then cultured in ImmunoCult-SF Macrophage Medium (STEMCELL Technologies) supplemented with Human Recombinant M-CSF (50 ng/ml) (STEMCELL Technologies) for 4 days.

Human pluripotent stem cell–derived BMEC differentiation and culture

Human pluripotent stem cells were cultured and differentiated into BMECs as previously described (56). Briefly, once human pluripotent stem cells reached 70% confluency, mTeSR1 was replaced with 1:1 DMEM/Ham’s F12 supplemented with 20% KnockOut Serum Replacer (Gibco), 1 mM l-glutamine (Sigma-Aldrich), 1× NEAA (Gibco), 0.1 mM β-mercaptoethanol (Sigma-Aldrich), and human basic fibroblast growth factor (50 ng/ml) (STEMCELL Technologies). Cells were maintained using this medium for a minimum of 5 days with daily medium changes. On the sixth day, the medium was changed to endothelial cell serum-free medium (EC SFM; Gibco) supplemented with 1× B-27 (Gibco), 10 μM retinoic acid (Sigma-Aldrich), and human basic fibroblast growth factor (20 ng/ml) (STEMCELL Technologies). Cells were maintained using this medium for a minimum of 48 hours. Differentiated cells were lifted off the culture vessel using StemPro Accutase (Gibco) and washed several times before seeding at a density of 1 × 10⁶ cells/ml on plates coated with type IV collagen (400 μg/ml) from human placenta (Sigma-Aldrich) and fibronectin (100 μg/ml) from human plasma (Millipore Sigma). BMECs were seeded in EC SFM supplemented with 1× B-27 and 10 μM ROCK inhibitor and maintained in EC SFM supplemented with 1× B-27 with daily medium changes.

SARS-CoV-2 infection and VLP inoculation

The SARS-CoV-2 viruses used for the infection experiments were obtained through the Biological and Emerging Infections Resources Program (BEI Resources) established by NIAID: ancestral WA1 strain (isolate USA-WA1/2020, BEI Resources catalog: NR-52281), Delta B.1.617.2 (isolate USA/CA-VRLC086/2021, BEI Resources catalog: NR-55691), Omicron BA.1 (BEI Resources catalog: NR-56461), Omicron BA.4 (BEI Resources catalog: NR-56806), and Omicron BA.5 (BEI Resources catalog: NR-58620). All the virus stocks were prepared and titrated with Vero cells in a biosafety level 2+ laboratory. To infect the macrophages differentiated from THP-1 parental cells and stable cell lines that express CD169 or CD169 + ACE2, virus stock was diluted in Dulbecco’s phosphate-buffered saline (DPBS) supplemented with 1% human AB serum (Sigma-Aldrich) and added to cells at a multiplicity of infection (MOI) of 2 plaque-forming units (PFU)/cell. The infection was carried out in triplicates in a 12-well plate with 5 × 10⁵ macrophages seeded per well. After 1-hour incubation with the virus, the freshly made medium (RPMI 1640 with 10% human serum) was added to cells. The total cellular RNAs were isolated for reverse transcription qPCR (RT-qPCR) detection.

To generate the SARS-CoV-2 VLPs, we followed the protocol established by Syed et al. (35). Briefly, we transfected HEK293T cells in a 15-cm plate with CoV2-N (0.67, Addgene #177937), CoV2-M-IRES-E (0.33, Addgene #177938), CoV2-Spike-D614G (0.25, Addgene #177960), and Luc-PS9 (1, Addgene #177942) at the indicated mass ratios for a total of 40 mg of DNA. Total DNA and PolyJet transfection reagent (120 ml) (SignaGen) were individually diluted in 1 ml of serum-free DMEM medium. Diluted PolyJet was added to the diluted DNA and mixed gently. Transfection mixture was incubated for 15 min at room temperature and then added dropwise while gently swirling the plate. The medium was replaced after 18 hours with 15 ml of fresh DMEM containing bovine serum and P/S. Forty-two hours after transfection, supernatant was collected and filtered with a 0.45-μm pore size syringe filter. The macrophages at the amounts of 5 × 10⁵ cells per well were incubated with 540 ml of VLP in 12-well plates for 24 hours, followed by a 4-hour stimulation of CL097 in RPMI 1640 with 10% human serum at the concentration of 1.6 mg/ml.

Construction of SARS-CoV-2 (Wuhan-Hu-1 strain, Delta and Omicron variants), SARS-CoV-1, MERS-CoV N–expressing plasmids, and tagged host factor (cGAS and PABP1) plasmids

The N protein sequences that are used to generate SARS-CoV-2 N–expressing THP-1 cells are SARS-CoV-2 isolate Wuhan-Hu-1 [National Center for Biotechnology Information (NCBI) accession: NC_045512], SARS-CoV-2 Delta B.1.617.2 (NCBI accession: PV361325), and SARS-CoV-2 Omicron BA.1 (NCBI accession: PP521471). All the primers that are used in gateway cloning, NEBuilder HiFi DNA Assembly, and site-directed mutagenesis are listed in table S5.

The lentivirus plasmid expressing myc-tagged SARS-CoV-2 N (Wuhan-Hu-1 strain) was constructed through Gateway cloning. Briefly, the myc-tagged Wuhan-Hu-1 strain N protein flanked with attB sequences was codon-optimized and synthesized by Codex DNA Inc. The myc-N (Wuhan-Hu-1 strain) fragment was first moved into pDONR221 (Invitrogen) vector through attB x attP (BP) reaction by using Gateway BP Clonase II Enzyme mix and then transferred from pDONR221 to pCW57.1 (Addgene #41393) tetracycline (tet)–inducible lentivirus vector through attL x attR (LR) reaction by using Gateway LR Clonase II Enzyme mix according to the manufacturer’s instructions to generate pCW57.1–SARS-CoV-2 N (Wuhan-Hu-1 strain).

To generate lentivirus plasmid expressing myc-tagged N from SARS-CoV-2 Delta variant, Delta variant–specific mutation sites D63G, R203M, and D377Y were included in the primers to amplify three fragments with overlapped ends from pCW57.1–SARS-CoV-2 N (Wuhan-Hu-1 strain). These three fragments were seamlessly ligated in one step through NEBuilder HiFi DNA Assembly Kit according to the manufacturer’s instructions. To generate lentivirus plasmid expressing myc-tagged N from SARS-CoV-2 Omicron variant, Omicron-specific mutation sites P13L, R203K, and G204R were introduced into pDONR221-N through NEBuilder HiFi DNA Assembly to make an intermediate mutant construct, pDONR221-N mutant. Del31–33 was introduced into pDONR221-N mutant, which has P13L, R203K, and G204R already, through amplification with a pair of primers missing Glu-Arg-Ser on each 5′ terminal end by Q5 Site-Directed Mutagenesis Kit. The Omicron variant N was finally cloned into Gateway-compatible pCW57.1 or pcDNA3.1 vectors by using Gateway LR Clonase II Enzyme mix.

We used site-directed mutagenesis approach on pDONR221-N to generate Delta single mutation mutants and Omicron BA.1 single-mutation mutants. We then transferred the N constructs containing individual variant mutations into pcDNA3.1 vector via Gateway LR reaction. To construct lentivirus plasmid expressing myc-tagged N of SARS-CoV-1 or MERS-CoV, the N proteins are cloned from the gateway entry plasmids pEntry-SARS-CoV-1-N (Addgene #168852) and pEntry-MERS-CoV-N (Addgene #168833), which were contributed by Weller et al. (73). The SARS-CoV-1 N sequence is derived from the isolate Tor 2 (NCBI accession: NC_019843), and the MERS-CoV N sequence is derived from the isolate HCoV-EMC/2012 (NCBI accession: NC_019843). The Kozak sequence and myc tag were inserted into the N-terminal ends of the N proteins through integration into the primers to amplify two fragments with overlapped ends for NEBuilder HiFi DNA Assembly. The myc-tagged SARS-CoV-1 and MERS-CoV N proteins were later cloned into pCW57.1 through Gateway LR Clonase II Enzyme mix.

To construct plasmid expressing cGAS tagged with 3xFLAG, the cGAS was cloned from the total RNA from THP-1 cells and inserted into pcDNA3.1-3xFLAG vector through NEBuilder HiFi DNA Assembly Kit. To construct plasmid expressing PABP1 tagged with 3xFLAG, the PABP1 was moved from an existing pcDNA3.1-myc vector–based construct to pcDNA3.1-3xFLAG vector through restriction sites Not I and Xba I. The pcDNA3.1 plasmid expressing V5-tagged IGF2BP3 was previously constructed by S. Huang in our laboratory.

Generation of stable HEK293T-inducible SARS-CoV-2 protein expression cell lines

To generate inducible HEK293T cell lines, we used the enhanced PiggyBac (ePB) transposable element system, provided by the Brivanlou laboratory at Rockefeller University, as previously described (74). Parental wild-type HEK293T cells were cotransfected with an equal ratio of the transposase plasmid and the ePB transposon vector encoding one of five SARS-CoV-2 proteins; N, membrane (M), nsp14, nsp15, and nsp16, previously identified to be innate immune antagonists. Forty-eight hours post-transfection, cells were subjected to selection with puromycin (1.5 μg/ml) to enrich for a population of HEK293T cells carrying stable, inducible expression of N, M, nsp14, nsp15, or nsp16.

Cloning of an NF-κB-responsive luciferase reporter plasmid

Previous work established the requirement of five tandem NF-κB binding sites before a promoter for the most efficient reporter activity when tracking NF-κB (75). Additionally, the NF-κB consensus binding site was determined to be GGRNNNNYCC (76). Therefore, we constructed a gene block containing five tandem NF-κB binding sites and the −56 to +20 region of an NF-κB–regulated gene, flanked by Eco RI (5′) and Nhe I (3′) cut sites along with four random additional nucleotides on either end (5′-AGGCGAATTCGGAATTTCCCGGGAATTCCCGGGATTACCCGGGATTTTCCGGGATTTCCCTCTGAATAGAGAGAGGACCATCTCATATAAATAGGCCATACCCATGGAGAAAGGACATTCTAACTGCAACCTTTCGCTAGCGGCC-3′). This gene block was initially cloned into the pCR-Blunt II-TOPO vector using the TOPO Blunt cloning kit (Thermo Fisher Scientific) for amplification. Both the TOPO vector containing the gene block and the pGL3-IFNB1-FLuc plasmid (Addgene #102597) were digested using Eco RI and Nhe I (New England Biolabs). The appropriate fragments were isolated from agarose gels and ligated together using the T4 ligation reaction kit from New England Biolabs. Chemically competent bacteria were then transformed with the ligation mixture followed by antibiotic selection.

IFN and NF-κB activation luciferase assays

Inducible HEK293T cells were seeded at a density of 1 × 10⁵ cells/ml in DMEM supplemented with 10% FBS and doxycycline (1 μg/ml). The following day, the cells were transfected using the TransIT-mRNA transfection kit (MirusBio) with a mix of 100 ng of plasmid expressing firefly luciferase under control of the IFNB1 promoter or plasmid expressing firefly luciferase under control of NF-κB binding sites, 10 ng of plasmid constitutively expressing Renilla luciferase under control of the CMV1 promoter, and 500 ng of poly(I:C) (for the IFNB1 reporter) or recombinant TNF-α (20 ng/ml) (for the NF-κB reporter). Twenty-four hours after transfection, cells were harvested by aspirating the medium and dispensing 100 μl of 1× passive lysis buffer (Promega). The dual luciferase kit was used to determine firefly and Renilla luciferase activities according to the manufacturer’s instructions (Promega). Cell lysates supplemented with luciferase substrates were analyzed for luciferase activity using a BioTek plate reader.

Lentivirus packaging and stable THP-1 cell line construction

To package lentivirus, Lenti-X HEK293T cells (Takara) were transfected with pMD2.G (Addgene, #12259), psPAX2 (Addgene, #12260), and pCW57.1-N plasmid of different coronaviruses. The supernatant samples containing lentivirus particles were collected 48 hours post-transfection and concentrated through Lenti-X Concentrator according to the manufacturer’s instruction.

Spinoculation was performed for lentivirus transduction in THP-1 cells to stably integrate the N protein in the genome. Briefly, 3 × 10⁵ THP-1 cells per well were seeded in 500 ml of transduction medium [RPMI 1640 with 5% FBS and polybrene (4 mg/ml)] with 100 ml of concentrated lentivirus in a 12-well plate. The THP-1 cells were spinoculated at 1500 rpm at 37°C for 1 hour and recovered in incubator for 4 hours before 500 ml of RPMI 1640 complete medium was added to each well. The transduced THP-1 cells were expanded for 48 hours and added with puromycin (2 mg/ml) for 5-day selection. The stable THP-1 cell lines were maintained in RPMI 1640 with 10% FBS, 1× NEAA, 1× P/S, and puromycin (0.2 mg/ml).

Macrophage stimulation with TLR and RLR agonists and supernatant cytokine and chemokine detection

For PRR stimulation, the macrophages derived from stable THP-1 cell lines or parental THP-1 cells were treated with doxycycline (2 mg/ml) in RPMI 1640 supplemented with 10% human AB serum, 1× NEAA, and 1× P/S for 48 hours to induce N expression or serve as controls. The doxycycline-treated macrophages were then stimulated with CL097 (1 mg/ml), an imidazoquinoline compound, to stimulate the TLR7/8 receptor, or poly(I:C) (10 mg/ml) to stimulate the TLR3 receptor in RPMI 1640 supplemented with 5% human AB serum and doxycycline (2 mg/ml). For RLR stimulation, the final concentration of poly(I:C) (2 mg/ml) was transfected into doxycycline-treated macrophages through Lipofectamine RNAiMAX (Invitrogen) according to the manufacturer’s instructions. Four hours after PRR stimulation, the cells were harvested for downstream processing.

For supernatant cytokine and chemokine detection, the culture media were collected after PRR stimulation, followed by a centrifugation at 200g at 4°C for 5 min to remove the cell debris. The clarified supernatants were then submitted to Eve Technologies (Canada) for proinflammatory/inflammatory cytokine and chemokine detection.

RNA isolation for RNA-seq analysis

The parental and N-expressing THP-1 cells were seeded in biological duplicates at the density of 5 × 10⁵ cells per well. After differentiation, doxycycline treatment, and PRR stimulation, the cells were lysed in TRizol reagent (Invitrogen), followed by total RNA extraction with Direct-zol (Zymo). The purified RNAs were submitted to BGI for transcriptome sequencing.

RNA-seq bioinformatic analysis

The RNA-seq data were first processed to remove adapter sequences and low-quality reads using Trimmomatic, followed by the pseudoalignment to the human genome GCHr38 using Salmon (v1.10.2) (77). The read counts of genes were generated by Tximeta (78), and the differential expression tests between the groups of interest were analyzed by edgeR (79) under the GLM framework. The similarities of the samples were explored by generating multi-dimensional scaling (MDS) plots through edgeR, and the visualization of MDS plot was further polished through ggplot2. For GO and pathway analysis, the selected DEGs were submitted to DAVID database (80) or Enrichr (39) to search for over-represented GO and pathway terms. For transcription factor prediction, the selected DEGs were submitted to the Enrichr to search most relevant regulatory transcription factors in the TRRUST (38) database. The volcano plots of gene expressions in N-expressing macrophages, the histograms of Delta variant fold changes of N DEGs, the bar charts of the GO analysis results, and the lollipop plots of transcription factor predictions were all visualized by corresponding functions in ggplot2 (81). The Venn diagrams were generated through ggVennDiagram (82), and the giant heatmap of overlapped DEGs in all treatment conditions was plotted through pheatmap (83).

Coronavirus N protein sequence alignment

The amino acid sequences of N from SARS-CoV-2 Wuhan-Hu-1 strain (NCBI: NC_045512), Delta B.1.617.2 (NCBI: PV361325), Omicron BA.1 (NCBI: PP521471), Omicron BA.4 (NCBI: OP093374), Omicron BA.5 (NCBI: OP984772), SARS-CoV-1 (NCBI: NC_004718.3), and MERS-CoV (NCBI: NC_NC_019843.3) were obtained from NCBI database and analyzed with EMBL-EBI MUSCLE. The alignment of the N proteins was visualized with Biostring and ggmsa packages through R.

Quantitative reverse transcription PCR (RT-qPCR)

To analyze intracellular cytokine and chemokine expressions after SARS-CoV-2 infection, VLP inoculation, and PRR stimulations, the total cellular RNAs were isolated as previously described and reverse-transcribed into cDNAs through the Protoscript II First Strand cDNA Synthesis Kit (NEB) according to the manufacturer’s instruction. The interested cytokines and chemokines were detected with NEB Luna qPCR dye with specific primers from PrimerBank (84). The qPCRs were run on CFX Opus system (Rio-Rad) with the cycling conditions suggested by NEB. The relative fold changes of the cytokines and chemokines of infected or stimulated samples to untreated controls were calculated by 2^−ΔΔCt method.

Immunoprecipitation and immunoblotting

To prepare for coimmunoprecipitation for AP-MS, 6 × 15-cm dishes of parental and myc-tagged SARS-CoV-2 N–expressing (Delta or Omicron variant) THP-1 cells were differentiated into macrophages with 2 × 10⁷ cells per dish. Half of the cells per dish were treated with doxycycline (2 mg/ml) to induce N expression or serve as drug-treated controls. The other half of the cells were maintained in RPMI 1640 with 10% human AB serum and 1× P/S without doxycycline treatment. The cell culture scale for AP-MS validation is smaller, and the THP-1 culture and differentiation for immunoprecipitation were carried out in six-well plates with 1.2 × 10⁶ cells seeded in each well. Forty-eight hours later, cells in each dish were lysed with 1% NP-40 lysis buffer (Thermo Fisher Scientific) supplemented with 1× Phosphatase Inhibitor Cocktail I (Abcam), 1 mM dithiothreitol (DTT), and Complete EDTA-free protease inhibitor mixture tablet (Roche). Cell lysate from each dish was centrifuged at 14,000 rcf for 15 min. The clarified supernatants were incubated with anti-myc agarose beads (EZview Red Anti-c-Myc Affinity Gel, Millipore) for 45 min at 4°C. After washing with NP-40 lysis buffer four times, proteins were eluted with urea buffer [8 M urea, 100 mM tris-HCl (pH 8)] for mass spectrometry analysis or immunoblot validation.

For immunoblot, proteins were resolved by SDS-PAGE (polyacrylamide gel electrophoresis) in 4 to 15% precast Mini-PROTEAN TGX Gels (Bio-Rad) in conventional tris/glycine/SDS buffer and later blotted to polyvinylidene difluoride (PVDF) membrane (Bio-Rad), followed by the detection with primary antibodies and horseradish peroxidase (HRP)-conjugated secondary antibodies. Immunoblots were imaged by chemiluminescence with the ProSignal Pico ECL Reagents (Genesee Scientific) on a ChemiDoc (Bio-Rad). The antibody information is in table S5.

AP-MS and proteomic data visualization

Following the affinity purification, samples were reduced with tris-(2-carboxyethyl) (TCEP) (10 mM final) for 30 min at room temperature and alkylated with 2-chloroacetamide (40 mM final) at room temperature for 30 min, sequentially, with shaking on a ThermoMixer (Eppendorf) at 1200 rpm. Next, tris-HCl buffer (0.1 M, pH 8) was added to the samples to dilute the urea to 1 M final concentration, and 2 μg of sequencing-grade trypsin (Promega) and 1 μg of LysC (Wako) were added to the samples and digested overnight on a ThermoMixer at 1200 rpm at room temperature. Peptides were acidified with trifluoroacetic acid (0.5% final concentration), desalted with spin desalting columns (Higgins Analytical), and eluted with 50% acetonitrile (ACN) and 0.1% formic acid. The eluted peptides were dried in a SpeedVac (Labconco) and resuspended in 50 μl of 0.1% formic acid. Dried peptides were resuspended in 0.1% formic acid (FA) in liquid chromatography–mass spectrometry (LC-MS) grade and analyzed on a timsTOF HT mass spectrometer (Bruker Daltonics) paired with a Vanish Neo UHPLC system (Thermo Fisher Scientific). Mobile phase A consisted of 0.1% FA in MS grade water, and mobile phase B consisted of 0.1% FA in 100% MS grade ACN. LC was performed in a trap-and-elute mode. First, peptides were trapped on a PepMap Neo trap column (5 mm, 100 Å pore size, 5 μm particle size). Then, they were separated by reversed-phase chromatography on an Aurora Elite C18 reverse phase column (15 cm length, 75 μm diameter, 1.7 μm particle size for captive spray, IonOptiks). The 45-min LC gradient was run with a flow rate of 300 nl/min set as follows: 5 to 35% B over 37 min, then to 45% B over 4 min, then to 60% B over 1 min, and then to 95% B for 3 min. The column was maintained at 50°C using a column oven for Bruker Captive Spray source (Sonation Lab Solutions). Ionization was performed using a CaptiveSpray source (Bruker Daltonics) at 1700 V. On the timsTOF HT, equal-size windows of 25 Da were designed with an overlap of 1 Da to maximize the precursor ion coverage for further MS/MS. The ion accumulation time and ramp times in the dual TIMS analyzer were set to 100 ms each. In the ion mobility (1/K0) range 0.6 to 1.6 Vs cm⁻², the collision energy was linearly decreased from 59 eV at 1/K0 = 1.46 Vs cm⁻² to 20 eV at 1/K0 = 0.62 Vs cm⁻² to collect the MS/MS spectra in the mass range 265.0 to 1370.0 Da. The estimated mean cycle time was 1.59 s.

The raw files were processed with Spectronaut (Biognosys, version 19.0) with the directDIA+ (Deep) search algorithm. Carbamidomethylation (cysteine) was set as a fixed modification for database search. Acetylation (protein N-term), oxidation (methionine), and phosphorylation (serine, threonine, tyrosine) were set as variable modifications. Reviewed human proteome and SARS-CoV-2 protein sequences (downloaded from UniProt) were used for spectral matching. The FDRs for the peptide-spectrum match (PSM), peptide, and protein groups were set to 0.01, and the minimum localization threshold for PTM was set to zero. For MS2-level area-based quantification, the cross-run normalization option was unchecked (normalization was performed later using MSstats), and the probability cutoff was set to zero for the PTM localization. Quantitative analysis was performed in the R statistical programming language (v.4.4.1). Initial quality control analyses, including inter-run clustering, correlations, principal components analysis (PCA), peptide and protein counts, and intensities, were completed using custom R code. Statistical analysis of phosphorylation and protein abundance changes between exposed and control samples was computed in MSstats (version 4.16.0) (85). MSstats was parameterized to perform normalization by median equalization, no imputation of missing values, and median smoothing (Tukey’s median polish) to combine intensities for multiple peptide ions or fragments into a single intensity for their protein group, and statistical tests of differences in intensity between conditions. Default settings for MSstats were used for adjusted P values. By default, MSstats uses the Student’s t test for P value calculation and the Benjamini-Hochberg method of FDR estimation to adjust P values. Identified proteins were subjected to protein-protein interaction scoring using SAINTexpress (http://apostl.moffitt.org/) (86). Protein interactions with a Bayesian false discovery rate (BFDR) ≤ 0.05 and average spectral count ≥ 11 were selected as high-confidence protein-protein interactions and visualized with Cytoscape (version 3.10.3) (87).

Measurement of TEER

TEER measurements of BMECs or HCAECs in a transwell system were conducted using EVOM3 (World Precision Instruments). STX2 Plus electrodes were positioned with one prong in the apical chamber and the other in the basolateral chamber, ensuring that both were submerged in the medium at an approximate depth of 0.5 cm in a 12-well plate. Readings were taken starting once daily after cell seeding, with resistance values recorded alongside a cell-free control well containing plain medium, which was used for background subtraction. The cell medium was changed daily, and for treatment with macrophage supernatant, the medium was replaced with the cells’ respective medium with 20% macrophage supernatant. TEER in Ωcm² was calculated by taking the corrected resistance values (Ω) and multiplying by the surface area (cm²) of the transwell insert, providing a measure of barrier integrity.

Statistical analysis

Bar graphs with appropriate statistical analyses were generated and performed by GraphPad Prism 10. Representative results from at least two independent experiments are shown as means ± SD. Statistical significance was indicated in figures as follows: *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001. Nonsignificant comparisons were not shown. Portions of the methods and conceptual framework described in this study were previously included in the doctoral dissertation of P.A.A., which is publicly available through ProQuest (88).

Acknowledgments

We thank R. Kan from A. Bhaduri’s laboratory and L. Ye from J. Hsu’s laboratory at UCLA for their important support and expertise on establishing the HEK293T SARS-CoV-2 protein cell lines and HCAEC maintenance. We thank R. Gummuluru from Boston University for kindly providing us with CD169- and CD169 + ACE2–expressing THP-1 cells. We thank L. Nguyen, S. Huang, E. Kim, M. Ruvalcaba, and M. V. Nieto for their support and insights that helped this work move forward. We thank the James B. Pendleton Charitable Trust and the McCarthy Family Foundation for their generous support, which provided critical equipment to the UCLA AIDS Institute. We also thank D. S. An and S. Venugopal from UCLA/CFAR Virology Core Lab (grant number 5P30 AI028697) for providing human blood samples. The following reagents were deposited by the Centers for Disease Control and Prevention and obtained through BEI Resources, NIAID, NIH: SARS-Related Coronavirus 2, Isolate hCoV-19/USA-WA1/2020, NR-52281.

Funding:

This work was supported by National Institute of Allergy and Infectious Diseases, grant F31AI179235 (P.A.A.), National Institute of Allergy and Infectious Diseases, grant R01AI158704 (M.M.H.L.), W.M. Keck Foundation, UCLA DGSOM Junior Faculty Award (M.M.H.L.), American Heart Association, Mechanisms Underlying Cardiovascular Consequences Associated with COVID-19 and Long COVID Award (M.M.H.L.), UCLA DGSOM Broad Stem Cell Research Center, COVID-19 Research Award (M.M.H.L.), Jonsson Comprehensive Cancer Center Broad Stem Cell Research Center, Ablon Scholars Award (M.M.H.L.), UCLA Microbiology, Immunology, and Molecular Genetics, Sydney Finegold Post-Doctoral Fellow Award (Z.Y.), National Institute of Allergy and Infectious Diseases, grant R01AI173214 (A.H.), National Institute of Allergy and Infectious Diseases, grant R01AI185026 (A.H.), National Institute of Allergy and Infectious Diseases, grant R01AI109022 (H.C.A.), Ruth L. Kirschstein National Research Service Award AI007323 (D.W.B.), and National Institute of Allergy and Infectious Diseases, grant 5P30 AI028697 (UCLA CFAR/Virology Core Lab).

Author contributions:

Conceptualization: Z.Y., P.A.A., A.H., and M.M.H.L. Methodology: Z.Y., P.A.A., C.C., Y.D., P.K., Y.Y., Q.D., R.D., M.B., A.H., and M.M.H.L. Software: C.C., Y.D., M.B., and A.H. Validation: Z.Y. and P.A.A. Formal analysis: Z.Y., P.A.A., C.C., Y.D., P.K., M.B., A.H., and M.M.H.L. Investigation: Z.Y., P.A.A., Y.D., and P.K. Resources: D.W.B., Y.Y., A.K.Z., V.A., Q.D., J.J.H., R.D., H.C.A., M.B., A.H., and M.M.H.L. Data curation: Z.Y., P.A.A., C.C., Y.D., M.B., A.H., and M.M.H.L. Writing—original draft: Z.Y., P.A.A., C.C., Y.D., J.J.H., and M.M.H.L. Writing—review and editing: Z.Y., P.A.A., C.C., Y.D., P.K., D.W.B., D.A., Q.L., Y.Y., A.K.Z., V.A., Q.D., J.J.H., R.D., H.C.A., M.B., A.H., and M.M.H.L. Visualization: Z.Y., P.A.A., Y.D., and M.B. Supervision: V.A., Q.D., J.J.H., R.D., H.C.A., M.B., A.H., and M.M.H.L. Project administration: R.D., A.H., and M.M.H.L. Funding acquisition: A.H. and M.M.H.L.

Competing interests:

The authors declare that they have no competing interests.

Data, code, and materials availability:

All data needed to evaluate and reproduce the results in the paper are present in the paper, Supplementary Materials, or in the Dryad repository (DOI: 10.5061/dryad.mpg4f4rg1). This paper does not include original code. The RNA sequencing data discussed in this publication have been deposited in NCBI’s Gene Expression Omnibus and are accessible through the GEO series accession number GSE327545. Mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE partner repository with the dataset identifier PXD067631 (https://www.ebi.ac.uk/pride/archive/projects/PXD067631). The following cell lines and constructs can be provided by M. Li pending scientific review and a completed material transfer agreement through University of California, Los Angeles: (i) inducible THP-1 cell lines expressing SARS-CoV-2 Wuhan/Delta/Omicron BA.1 N, SARS-CoV-1 N, and MERS-CoV-N; (ii) inducible HEK293T cell lines expressing SARS-CoV-2 N, M, nsp14, nsp15, and nsp16; (iii) the inducible lentivirus plasmids expressing the N proteins from SARS-CoV-2 (Wuhan, Delta, and Omicron BA.1 variants), SARS-CoV-1, and MERS-CoV; (iv) the pcDNA3.1 plasmids expressing cGAS, PABP1, IGF2BP2, and the N proteins from SARS-CoV-2 (Wuhan, Delta, and Omicron BA.1 variants) as well as those with individual Delta/Omicron mutations (D63G, R203M, D377Y, P13L, del31–33, R203K, and G204R). Request for these materials should be submitted to: ManHingLi@mednet.ucla.edu.

Supplementary Materials

The PDF file includes:

Figs. S1 to S4

Legends for tables S1 to S5

sciadv.aea2780_sm.pdf^{(4.2MB, pdf)}

Other Supplementary Material for this manuscript includes the following:

Tables S1 to S5

sciadv.aea2780_tables_s1_to_s5.zip^{(199.3KB, zip)}

REFERENCES

1.World Health Organization. WHO coronavirus (COVID-19) dashboard > Deaths. https://data.who.int/dashboards/covid19/deaths.
2.E. Mathieu, H. Ritchie, L. Rodés-Guirao, C. Appel, D. Gavrilov, C. Giattino, J. Hasell, B. Macdonald, S. Dattani, D. Beltekian, E. Ortiz-Ospina, M. Roser, Coronavirus (COVID-19) hospitalizations. Publ. Online OurWorldinDataorg (2020).
3.Xie Y., Choi T., Al-Aly Z., Mortality in patients hospitalized for COVID-19 vs influenza in fall-winter 2023-2024. JAMA 331, 1963–1965 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Rosas I. O., Bräu N., Waters M., Go R. C., Hunter B. D., Bhagani S., Skiest D., Aziz M. S., Cooper N., Douglas I. S., Savic S., Youngstein T., Del Sorbo L., Cubillo Gracian A., De La Zerda D. J., Ustianowski A., Bao M., Dimonaco S., Graham E., Matharu B., Spotswood H., Tsai L., Malhotra A., Tocilizumab in hospitalized patients with severe Covid-19 pneumonia. N. Engl. J. Med. 384, 1503–1516 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
5.The RECOVERY Collaborative Group , Dexamethasone in hospitalized patients with Covid-19. N. Engl. J. Med. 384, 693–704 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
6.Montazersaheb S., Hosseiniyan Khatibi S. M., Hejazi M. S., Tarhriz V., Farjami A., Ghasemian Sorbeni F., Farahzadi R., Ghasemnejad T., COVID-19 infection: An overview on cytokine storm and related interventions. Virol. J. 19, 92 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Ribero M. S., Jouvenet N., Dreux M., Nisole S., Interplay between SARS-CoV-2 and the type I interferon response. PLOS Pathog. 16, e1008737 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
8.Hadjadj J., Yatim N., Barnabei L., Corneau A., Boussier J., Smith N., Péré H., Charbit B., Bondet V., Chenevier-Gobeaux C., Breillat P., Carlier N., Gauzit R., Morbieu C., Pène F., Marin N., Roche N., Szwebel T.-A., Merkling S. H., Treluyer J.-M., Veyer D., Mouthon L., Blanc C., Tharaux P.-L., Rozenberg F., Fischer A., Duffy D., Rieux-Laucat F., Kernéis S., Terrier B., Impaired type I interferon activity and inflammatory responses in severe COVID-19 patients. Science 369, 718–724 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
9.Yang D., Geng T., Harrison A. G., Wang P., Differential roles of RIG-I like receptors in SARS-CoV-2 infection. Military Med. Res. 8, 49 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Kawasaki T., Kawai T., Toll-like receptor signaling pathways. Front. Immunol. 5, 461 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
11.Cervantes J. L., Weinerman B., Basole C., Salazar J. C., TLR8: The forgotten relative revindicated. Cell. Mol. Immunol. 9, 434–438 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
12.Kawai T., Ikegawa M., Ori D., Akira S., Decoding Toll-like receptors: Recent insights and perspectives in innate immunity. Immunity 57, 649–673 (2024). [DOI] [PubMed] [Google Scholar]
13.Mantovani S., Oliviero B., Varchetta S., Renieri A., Mondelli M. U., TLRs: Innate immune sentries against SARS-CoV-2 infection. Int. J. Mol. Sci. 24, 8065 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
14.Comar C. E., Otter C. J., Pfannenstiel J., Doerger E., Renner D. M., Tan L. H., Perlman S., Cohen N. A., Fehr A. R., Weiss S. R., MERS-CoV endoribonuclease and accessory proteins jointly evade host innate immunity during infection of lung and nasal epithelial cells. Proc. Natl. Acad. Sci. U.S.A. 119, e2123208119 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Totura A. L., Baric R. S., SARS coronavirus pathogenesis: Host innate immune responses and viral antagonism of interferon. Curr. Opin. Virol. 2, 264–275 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
16.Xia H., Cao Z., Xie X., Zhang X., Chen J. Y.-C., Wang H., Menachery V. D., Rajsbaum R., Shi P.-Y., Evasion of type I interferon by SARS-CoV-2. Cell Rep. 33, 108234 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Lei X., Dong X., Ma R., Wang W., Xiao X., Tian Z., Wang C., Wang Y., Li L., Ren L., Guo F., Zhao Z., Zhou Z., Xiang Z., Wang J., Activation and evasion of type I interferon responses by SARS-CoV-2. Nat. Commun. 11, 3810 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
18.Hayn M., Hirschenberger M., Koepke L., Nchioua R., Straub J. H., Klute S., Hunszinger V., Zech F., Prelli Bozzo C., Aftab W., Christensen M. H., Conzelmann C., Müller J. A., Srinivasachar Badarinarayan S., Stürzel C. M., Forne I., Stenger S., Conzelmann K.-K., Münch J., Schmidt F. I., Sauter D., Imhof A., Kirchhoff F., Sparrer K. M. J., Systematic functional analysis of SARS-CoV-2 proteins uncovers viral innate immune antagonists and remaining vulnerabilities. Cell Rep. 35, 109126 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Sender R., Bar-On Y. M., Gleizer S., Bernsthein B., Flamholz A., Phillips R., Milo R., The total number and mass of SARS-CoV-2 virions. Infectious Diseases (except HIV/AIDS). Proc. Natl. Acad. Sci. U.S.A. 118, e2024815118 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
20.Tugaeva K. V., Hawkins D. E. D. P., Smith J. L. R., Bayfield O. W., Ker D.-S., Sysoev A. A., Klychnikov O. I., Antson A. A., Sluchanko N. N., The mechanism of SARS-CoV-2 nucleocapsid protein recognition by the human 14-3-3 proteins. J. Mol. Biol. 433, 166875 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
21.López-Muñoz A. D., Kosik I., Holly J., Yewdell J. W., Cell surface SARS-CoV-2 nucleocapsid protein modulates innate and adaptive immunity. Sci. Adv. 8, eabp9770 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
22.Wang M., Valadez-Ingersoll M., Gilmore T. D., Control of nuclear localization of the nucleocapsid protein of SARS-CoV-2. Virology 600, 110232 (2024). [DOI] [PubMed] [Google Scholar]
23.Zheng Z.-Q., Wang S.-Y., Xu Z.-S., Fu Y.-Z., Wang Y.-Y., SARS-CoV-2 nucleocapsid protein impairs stress granule formation to promote viral replication. Cell Discov. 7, 38 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
24.Hu Y., Li W., Gao T., Cui Y., Jin Y., Li P., Ma Q., Liu X., Cao C., The severe acute respiratory syndrome coronavirus nucleocapsid inhibits type I interferon production by interfering with TRIM25-mediated RIG-I ubiquitination. J. Virol. 91, e02143-16 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
25.Li Y., Li M., Xiao H., Liao F., Shen M., Ge W., Ou J., Liu Y., Chen L., Zhao Y., Wan P., Liu J., Chen J., Lan X., Wu S., Ding Q., Li G., Zhang Q., Pan P., The R203M and D377Y mutations of the nucleocapsid protein promote SARS-CoV-2 infectivity by impairing RIG-I-mediated antiviral signaling. PLOS Pathog. 21, e1012886 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
26.Wang C., Xie J., Zhao L., Fei X., Zhang H., Tan Y., Nie X., Zhou L., Liu Z., Ren Y., Yuan L., Zhang Y., Zhang J., Liang L., Chen X., Liu X., Wang P., Han X., Weng X., Chen Y., Yu T., Zhang X., Cai J., Chen R., Shi Z.-L., Bian X.-W., Alveolar macrophage dysfunction and cytokine storm in the pathogenesis of two severe COVID-19 patients. EBioMedicine 57, 102833 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
27.Wu T. T.-H., Travaglini K. J., Rustagi A., Xu D., Zhang Y., Andronov L., Jang S., Gillich A., Dehghannasiri R., Martínez-Colón G. J., Beck A., Liu D. D., Wilk A. J., Morri M., Trope W. L., Bierman R., Weissman I. L., Shrager J. B., Quake S. R., Kuo C. S., Salzman J., Moerner W. E., Kim P. S., Blish C. A., Krasnow M. A., Interstitial macrophages are a focus of viral takeover and inflammation in COVID-19 initiation in human lung. J. Exp. Med. 221, e20232192 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
28.Zhou Z., Ren L., Zhang L., Zhong J., Xiao Y., Jia Z., Guo L., Yang J., Wang C., Jiang S., Yang D., Zhang G., Li H., Chen F., Xu Y., Chen M., Gao Z., Yang J., Dong J., Liu B., Zhang X., Wang W., He K., Jin Q., Li M., Wang J., Heightened innate immune responses in the respiratory tract of COVID-19 patients. Cell Host Microbe 27, 883–890.e2 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
29.Liao M., Liu Y., Yuan J., Wen Y., Xu G., Zhao J., Cheng L., Li J., Wang X., Wang F., Liu L., Amit I., Zhang S., Zhang Z., Single-cell landscape of bronchoalveolar immune cells in patients with COVID-19. Nat. Med. 26, 842–844 (2020). [DOI] [PubMed] [Google Scholar]
30.Sefik E., Qu R., Junqueira C., Kaffe E., Mirza H., Zhao J., Brewer J. R., Han A., Steach H. R., Israelow B., Blackburn H. N., Velazquez S. E., Chen Y. G., Halene S., Iwasaki A., Meffre E., Nussenzweig M., Lieberman J., Wilen C. B., Kluger Y., Flavell R. A., Inflammasome activation in infected macrophages drives COVID-19 pathology. Nature 606, 585–593 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
31.Jalloh S., Olejnik J., Berrigan J., Nisa A., Suder E. L., Akiyama H., Lei M., Ramaswamy S., Tyagi S., Bushkin Y., Mühlberger E., Gummuluru S., CD169-mediated restrictive SARS-CoV-2 infection of macrophages induces pro-inflammatory responses. PLOS Pathog. 18, e1010479 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
32.Merad M., Martin J. C., Pathological inflammation in patients with COVID-19: A key role for monocytes and macrophages. Nat. Rev. Immunol. 20, 355–362 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
33.Salina A. C., Dos-Santos D., Rodrigues T. S., Fortes-Rocha M., Freitas-Filho E. G., Alzamora-Terrel D. L., Castro I. M., da Silva T. F. F., de Lima M. H., Nascimento D. C., Silva C. M., Toller-Kawahisa J. E., Becerra A., Oliveira S., Caetité D. B., Almeida L., Ishimoto A. Y., Lima T. M., Martins R. B., Veras F., do Amaral N. B., Giannini M. C., Bonjorno L. P., Lopes M. I., Benatti M. N., Batah S. S., Santana R. C., Vilar F. C., Martins M. A., Assad R. L., de Almeida S. C., de Oliveira F. R., Neto E. A., Cunha T. M., Alves-Filho J. C., Bonato V. L., Cunha F. Q., Fabro A. T., Nakaya H. I., Zamboni D. S., Louzada-Junior P., Oliveira R. D. R., Cunha L. D., Efferocytosis of SARS-CoV-2-infected dying cells impairs macrophage anti-inflammatory functions and clearance of apoptotic cells. eLife 11, e74443 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
34.Bai Z., Cao Y., Liu W., Li J., The SARS-CoV-2 nucleocapsid protein and its role in viral structure, biological functions, and a potential target for drug or vaccine mitigation. Viruses 13, 1115 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
35.Syed A. M., Taha T. Y., Tabata T., Chen I. P., Ciling A., Khalid M. M., Sreekumar B., Chen P.-Y., Hayashi J. M., Soczek K. M., Ott M., Doudna J. A., Rapid assessment of SARS-CoV-2–Evolved variants using virus-like particles. Science 374, 1626–1632 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
36.Tang Y., Liu J., Zhang D., Xu Z., Ji J., Wen C., Cytokine storm in COVID-19: The current evidence and treatment strategies. Front. Immunol. 11, 1708 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
37.Durinck S., Spellman P. T., Birney E., Huber W., Mapping identifiers for the integration of genomic datasets with the R/Bioconductor package biomaRt. Nat. Protoc. 4, 1184–1191 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
38.Han H., Cho J.-W., Lee S., Yun A., Kim H., Bae D., Yang S., Kim C. Y., Lee M., Kim E., Lee S., Kang B., Jeong D., Kim Y., Jeon H.-N., Jung H., Nam S., Chung M., Kim J.-H., Lee I., TRRUST v2: An expanded reference database of human and mouse transcriptional regulatory interactions. Nucleic Acids Res. 46, D380–D386 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
39.Xie Z., Bailey A., Kuleshov M. V., Clarke D. J. B., Evangelista J. E., Jenkins S. L., Lachmann A., Wojciechowicz M. L., Kropiwnicki E., Jagodnik K. M., Jeon M., Ma’ayan A., Gene set knowledge discovery with Enrichr. Curr. Protoc. 1, e90 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
40.Imbiakha B., Ezzatpour S., Buchholz D. W., Sahler J., Ye C., Olarte-Castillo X. A., Zou A., Kwas C., O’Hare K., Choi A., Adeleke R. A., Khomandiak S., Goodman L., Jager M. C., Whittaker G. R., Martinez-Sobrido L., August A., Aguilar H. C., Age-dependent acquisition of pathogenicity by SARS-CoV-2 omicron BA.5. Sci. Adv. 9, eadj1736 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
41.Kannan S., Subbaram K., Ali S., Kannan H., Molecular Characterization and Amino Acid Homology of Nucleocapsid (N) Protein in SARS-CoV-1, SARS-CoV-2, MERS-CoV, and Bat Coronavirus. J. Pure Appl. Microbiol. 14, 757–763 (2020). [Google Scholar]
42.Du M., Chen Z. J., DNA-induced liquid phase condensation of cGAS activates innate immune signaling. Science 361, 704–709 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
43.Zhang X., Han L., Hou J., Yang H., Xu H., Li G., Shu Q., Zhu D., Zheng Y., Gao C., Stress granule-localized USP8 potentiates cGAS-mediated type I interferonopathies through deubiquitination of DDX3X. Cell Rep. 43, 114248 (2024). [DOI] [PubMed] [Google Scholar]
44.Zhao M., Xia T., Xing J.-Q., Yin L.-H., Li X.-W., Pan J., Liu J.-Y., Sun L.-M., Wang M., Li T., Mao J., Han Q.-Y., Xue W., Cai H., Wang K., Xu X., Li T., He K., Wang N., Li A.-L., Zhou T., Zhang X.-M., Li W.-H., Li T., The stress granule protein G3BP1 promotes pre-condensation of cGAS to allow rapid responses to DNA. EMBO Rep. 23, e53166 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
45.Huang H.-S., Chen L., Chi J.-X., Lai S.-Y., Pi J., Shao Y.-M., Xu J.-F., Stress granules and cell death: Crosstalk and potential therapeutic strategies in infectious diseases. Cell Death Dis. 16, 495 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
46.Deb P., Dai J., Singh S., Kalyoussef E., Fitzgerald-Bocarsly P., Triggering of the cGAS-STING pathway in human plasmacytoid dendritic cells inhibits TLR9-Mediated IFN production. J. Immunol. 205, 223–236 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
47.Sharma S., Campbell A. M., Chan J., Schattgen S. A., Orlowski G. M., Nayar R., Huyler A. H., Nündel K., Mohan C., Berg L. J., Shlomchik M. J., Marshak-Rothstein A., Fitzgerald K. A., Suppression of systemic autoimmunity by the innate immune adaptor STING. Proc. Natl. Acad. Sci. U.S.A. 112, E710–E717 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
48.Lama L., Adura C., Xie W., Tomita D., Kamei T., Kuryavyi V., Gogakos T., Steinberg J. I., Miller M., Ramos-Espiritu L., Asano Y., Hashizume S., Aida J., Imaeda T., Okamoto R., Jennings A. J., Michino M., Kuroita T., Stamford A., Gao P., Meinke P., Glickman J. F., Patel D. J., Tuschl T., Development of human cGAS-specific small-molecule inhibitors for repression of dsDNA-triggered interferon expression. Nat. Commun. 10, 2261 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
49.Xu S., Ilyas I., Weng J., Endothelial dysfunction in COVID-19: An overview of evidence, biomarkers, mechanisms and potential therapies. Acta Pharmacol. Sin. 44, 695–709 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
50.Eberhardt N., Noval M. G., Kaur R., Amadori L., Gildea M., Sajja S., Das D., Cilhoroz B., Stewart O., Fernandez D. M., Shamailova R., Vasquez Guillen A., Jangra S., Schotsaert M., Newman J. D., Faries P., Maldonado T., Rockman C., Rapkiewicz A., Stapleford K. A., Narula N., Moore K. J., Giannarelli C., SARS-CoV-2 infection triggers pro-atherogenic inflammatory responses in human coronary vessels. Nat. Cardiovasc. Res. 2, 899–916 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
51.Lee M.-H., Perl D. P., Nair G., Li W., Maric D., Murray H., Dodd S. J., Koretsky A. P., Watts J. A., Cheung V., Masliah E., Horkayne-Szakaly I., Jones R., Stram M. N., Moncur J., Hefti M., Folkerth R. D., Nath A., Microvascular injury in the brains of patients with Covid-19. N. Engl. J. Med. 384, 481–483 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
52.Li Q., Ye L., Talapaneni S., Meng Y., Wang C. R., Kalindjian K., Phan J., Chen Y., Demer L. L., Tintut Y., Hsu J. J., COL8A1 regulates endothelial phenotype in inflammatory endothelial-to-mesenchymal transition. Am. J. Physiol. Heart Circ. Physiol. 329, H1331–H1346 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
53.Qian T., Maguire S. E., Canfield S. G., Bao X., Olson W. R., Shusta E. V., Palecek S. P., Directed differentiation of human pluripotent stem cells to blood-brain barrier endothelial cells. Sci. Adv. 3, e1701679 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
54.Shen Y. H., Utama B., Wang J., Raveendran M., Senthil D., Waldman W. J., Belcher J. D., Vercellotti G., Martin D., Mitchelle B. M., Wang X. L., Human cytomegalovirus causes endothelial injury through the ataxia telangiectasia mutant and p53 DNA damage signaling pathways. Circ. Res. 94, 1310–1317 (2004). [DOI] [PMC free article] [PubMed] [Google Scholar]
55.Yamada S., Hashita T., Yanagida S., Sato H., Yasuhiko Y., Okabe K., Noda T., Nishida M., Matsunaga T., Kanda Y., SARS-CoV-2 causes dysfunction in human iPSC-derived brain microvascular endothelial cells potentially by modulating the Wnt signaling pathway. Fluids Barriers CNS 21, 32 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
56.Cheng Y., Medina A., Yao Z., Basu M., Natekar J. P., Lang J., Sanchez E., Nkembo M. B., Xu C., Qian X., Nguyen P. T. T., Wen Z., Song H., Ming G.-L., Kumar M., Brinton M. A., Li M. M. H., Tang H., Intrinsic antiviral immunity of barrier cells revealed by an iPSC-derived blood-brain barrier cellular model. Cell Rep. 39, 110885 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
57.Alvarez P. A., Tang A., Winters D. M., Kaushal P., Medina A., Nieto M. V., Kaczor-Urbanowicz K. E., St. Amant F., Reyes B. R., Kaake R. M., Fregoso O. I., Pyle A. D., Bouhaddou M., Tang H., Li M. M. H., Old World alphaviruses use distinct mechanisms to infect brain microvascular endothelial cells for neuroinvasion. Cell Rep. 44, 116305 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
58.Wick K. D., Leligdowicz A., Willmore A., Carrillo S. A., Ghale R., Jauregui A., Chak S. S., Nguyen V., Lee D., Jones C., Dewar R., Lane H. C., Kangelaris K. N., Hendrickson C. M., Liu K. D., Sinha P., Erle D. J., Langelier C. R., Krummell M. F., Woodruff P. G., Calfee C. S., Matthay M. A., COMET Consortium , Plasma SARS-CoV-2 nucleocapsid antigen levels are associated with progression to severe disease in hospitalized COVID-19. Crit. Care 26, 278 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
59.Shuaib M., Adroub S., Mourier T., Mfarrej S., Zhang H., Esau L., Alsomali A., Alofi F. S., Ahmad A. N., Shamsan A., Khogeer A., Hashem A. M., Almontashiri N. A. M., Hala S., Pain A., Impact of the SARS-CoV-2 nucleocapsid 203K/204R mutations on the inflammatory immune response in COVID-19 severity. Genome Med. 15, 54 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
60.Laurent P., Yang C., Rendeiro A. F., Nilsson-Payant B. E., Carrau L., Chandar V., Bram Y., tenOever B. R., Elemento O., Ivashkiv L. B., Schwartz R. E., Barrat F. J., Sensing of SARS-CoV-2 by pDCs and their subsequent production of IFN-I contribute to macrophage-induced cytokine storm during COVID-19. Sci. Immunol. 7, eadd4906 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
61.Shahbaz S., Bozorgmehr N., Lu J., Osman M., Sligl W., Tyrrell D. L., Elahi S., Analysis of SARS-CoV-2 isolates, namely the Wuhan strain, Delta variant, and Omicron variant, identifies differential immune profiles. Microbiol. Spectr. 11, e01256-23 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
62.Wu H., Xing N., Meng K., Fu B., Xue W., Dong P., Tang W., Xiao Y., Liu G., Luo H., Zhu W., Lin X., Meng G., Zhu Z., Nucleocapsid mutations R203K/G204R increase the infectivity, fitness, and virulence of SARS-CoV-2. Cell Host Microbe 29, 1788–1801.e6 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
63.Ren J., Wang S., Zong Z., Pan T., Liu S., Mao W., Huang H., Yan X., Yang B., He X., Zhou F., Zhang L., TRIM28-mediated nucleocapsid protein SUMOylation enhances SARS-CoV-2 virulence. Nat. Commun. 15, 244 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
64.Liu D., Yang J., Cristea I. M., Liquid-liquid phase separation in innate immunity. Trends Immunol. 45, 454–469 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
65.Wang L., Sola I., Enjuanes L., Zuñiga S., MOV10 helicase interacts with coronavirus nucleocapsid protein and has antiviral activity. mBio 12, e0131621 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
66.Menten P., Wuyts A., Van Damme J., Macrophage inflammatory protein-1. Cytokine Growth Factor Rev. 13, 455–481 (2002). [DOI] [PubMed] [Google Scholar]
67.Kitai R., Zhao M.-L., Zhang N., Hua L. L., Lee S. C., Role of MIP-1β and RANTES in HIV-1 infection of microglia: Inhibition of infection and induction by IFNβ. J. Neuroimmunol. 110, 230–239 (2000). [DOI] [PubMed] [Google Scholar]
68.Ciechanowska A., Popiolek-Barczyk K., Pawlik K., Ciapała K., Oggioni M., Mercurio D., De Simoni M.-G., Mika J., Changes in macrophage inflammatory protein-1 (MIP-1) Family Members expression induced by traumatic brain injury in mice. Immunobiology 225, 151911 (2020). [DOI] [PubMed] [Google Scholar]
69.Greene C., Connolly R., Brennan D., Laffan A., O’Keeffe E., Zaporojan L., O’Callaghan J., Thomson B., Connolly E., Argue R., Meaney J. F. M., Martin-Loeches I., Long A., Cheallaigh C. N., Conlon N., Doherty C. P., Campbell M., Blood–Brain barrier disruption and sustained systemic inflammation in individuals with long COVID-associated cognitive impairment. Nat. Neurosci. 27, 421–432 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
70.I. Ogobuiro, C. J. Wehrle, F. Tuma, “Anatomy, thorax, heart coronary arteries” in StatPearls (StatPearls Publishing, 2025). [PubMed] [Google Scholar]
71.Triantafyllis A. S., Sfantou D., Karapedi E., Peteinaki K., Kotoulas S. C., Saad R., Fountoulakis P. N., Tsamakis K., Tsiptsios D., Rallidis L., Tsoporis J. N., Varvarousis D., Hamodraka E., Giannakopoulos A., Poulimenos L. E., Ikonomidis I., Coronary implications of COVID-19. Med. Princ. Pract. 34, 1–12 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
72.Sagris M., Theofilis P., Antonopoulos A. S., Tsioufis C., Oikonomou E., Antoniades C., Crea F., Kaski J. C., Tousoulis D., Inflammatory mechanisms in COVID-19 and atherosclerosis: Current pharmaceutical perspectives. Int. J. Mol. Sci. 22, 6607 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
73.Weller B., Lin C.-W., Pogoutse O., Sauer M., Marin-dela Rosa N., Strobel A., Young V., Knapp J. J., Rayhan A., Falter C., Kim D.-K., Roth F. P., Falter-Braun P., A resource of human coronavirus protein-coding sequences in a flexible, multipurpose gateway Entry clone collection. G3 13, jkad105 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
74.Yang E., Huang S., Jami-Alahmadi Y., McInerney G. M., Wohlschlegel J. A., Li M. M. H., Elucidation of TRIM25 ubiquitination targets involved in diverse cellular and antiviral processes. PLOS Pathog. 18, e1010743 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
75.Badr C. E., Niers J. M., Tjon-Kon-Fat L.-A., Noske D. P., Wurdinger T., Tannous B. A., Real-time monitoring of nuclear factor κB activity in cultured cells and in animal models. Mol. Imaging 8, 278–290 (2009). [PMC free article] [PubMed] [Google Scholar]
76.Ngo K. A., Kishimoto K., Davis-Turak J., Pimplaskar A., Cheng Z., Spreafico R., Chen E. Y., Tam A., Ghosh G., Mitchell S., Hoffmann A., Dissecting the regulatory strategies of NF-κB RelA target genes in the inflammatory response reveals differential transactivation logics. Cell Rep. 30, 2758–2775.e6 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
77.Patro R., Duggal G., Love M. I., Irizarry R. A., Kingsford C., Salmon provides fast and bias-aware quantification of transcript expression. Nat. Methods 14, 417–419 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
78.Love M. I., Soneson C., Hickey P. F., Johnson L. K., Pierce N. T., Shepherd L., Morgan M., Patro R., Tximeta: Reference sequence checksums for provenance identification in RNA-seq. PLOS Comput. Biol. 16, e1007664 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
79.Chen Y., Chen L., Lun A. T. L., Baldoni P. L., Smyth G. K., edgeR v4: Powerful differential analysis of sequencing data with expanded functionality and improved support for small counts and larger datasets. Nucleic Acids Res. 53, gkaf018 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
80.Sherman B. T., Hao M., Qiu J., Jiao X., Baseler M. W., Lane H. C., Imamichi T., Chang W., DAVID: A web server for functional enrichment analysis and functional annotation of gene lists (2021 update). Nucleic Acids Res. 50, W216–W221 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
81.H. Wickham, Ggplot2: Elegant Graphics for Data Analysis (Springer International Publishing, ed. 2, 2016). [Google Scholar]
82.Gao C.-H., Chen C., Akyol T., Dusa A., Yu G., Cao B., Cai P., ggVennDiagram: Intuitive Venn diagram software extended. iMeta 3, e177 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
83.R. Kolde, pheatmap: Pretty heatmaps (2010). 10.32614/CRAN.package.pheatmap. [DOI]
84.Wang X., Spandidos A., Wang H., Seed B., PrimerBank: A PCR primer database for quantitative gene expression analysis, 2012 update. Nucleic Acids Res. 40, D1144–D1149 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
85.Choi M., Chang C.-Y., Clough T., Broudy D., Killeen T., MacLean B., Vitek O., MSstats: An R package for statistical analysis of quantitative mass spectrometry-based proteomic experiments. Bioinformatics 30, 2524–2526 (2014). [DOI] [PubMed] [Google Scholar]
86.Teo G., Liu G., Zhang J., Nesvizhskii A. I., Gingras A.-C., Choi H., SAINTexpress: Improvements and additional features in Significance Analysis of Interactome software. J. Proteomics 100, 37–43 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
87.Shannon P., Markiel A., Ozier O., Baliga N. S., Wang J. T., Ramage D., Amin N., Schwikowski B., Ideker T., Cytoscape: A software environment for integrated models of biomolecular interaction networks. Genome Res. 13, 2498–2504 (2003). [DOI] [PMC free article] [PubMed] [Google Scholar]
88.P. A. Alvarez, “The role of viral structural proteins in disrupting the physical and biochemical barriers to infection,” thesis, University of California, Los Angeles, Los Angeles, CA (2025). [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Figs. S1 to S4

Legends for tables S1 to S5

sciadv.aea2780_sm.pdf^{(4.2MB, pdf)}

Tables S1 to S5

sciadv.aea2780_tables_s1_to_s5.zip^{(199.3KB, zip)}

Data Availability Statement

[R1] 1.World Health Organization. WHO coronavirus (COVID-19) dashboard > Deaths. https://data.who.int/dashboards/covid19/deaths.

[R2] 2.E. Mathieu, H. Ritchie, L. Rodés-Guirao, C. Appel, D. Gavrilov, C. Giattino, J. Hasell, B. Macdonald, S. Dattani, D. Beltekian, E. Ortiz-Ospina, M. Roser, Coronavirus (COVID-19) hospitalizations. Publ. Online OurWorldinDataorg (2020).

[R3] 3.Xie Y., Choi T., Al-Aly Z., Mortality in patients hospitalized for COVID-19 vs influenza in fall-winter 2023-2024. JAMA 331, 1963–1965 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R4] 4.Rosas I. O., Bräu N., Waters M., Go R. C., Hunter B. D., Bhagani S., Skiest D., Aziz M. S., Cooper N., Douglas I. S., Savic S., Youngstein T., Del Sorbo L., Cubillo Gracian A., De La Zerda D. J., Ustianowski A., Bao M., Dimonaco S., Graham E., Matharu B., Spotswood H., Tsai L., Malhotra A., Tocilizumab in hospitalized patients with severe Covid-19 pneumonia. N. Engl. J. Med. 384, 1503–1516 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R5] 5.The RECOVERY Collaborative Group , Dexamethasone in hospitalized patients with Covid-19. N. Engl. J. Med. 384, 693–704 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R6] 6.Montazersaheb S., Hosseiniyan Khatibi S. M., Hejazi M. S., Tarhriz V., Farjami A., Ghasemian Sorbeni F., Farahzadi R., Ghasemnejad T., COVID-19 infection: An overview on cytokine storm and related interventions. Virol. J. 19, 92 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R7] 7.Ribero M. S., Jouvenet N., Dreux M., Nisole S., Interplay between SARS-CoV-2 and the type I interferon response. PLOS Pathog. 16, e1008737 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R8] 8.Hadjadj J., Yatim N., Barnabei L., Corneau A., Boussier J., Smith N., Péré H., Charbit B., Bondet V., Chenevier-Gobeaux C., Breillat P., Carlier N., Gauzit R., Morbieu C., Pène F., Marin N., Roche N., Szwebel T.-A., Merkling S. H., Treluyer J.-M., Veyer D., Mouthon L., Blanc C., Tharaux P.-L., Rozenberg F., Fischer A., Duffy D., Rieux-Laucat F., Kernéis S., Terrier B., Impaired type I interferon activity and inflammatory responses in severe COVID-19 patients. Science 369, 718–724 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R9] 9.Yang D., Geng T., Harrison A. G., Wang P., Differential roles of RIG-I like receptors in SARS-CoV-2 infection. Military Med. Res. 8, 49 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R10] 10.Kawasaki T., Kawai T., Toll-like receptor signaling pathways. Front. Immunol. 5, 461 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R11] 11.Cervantes J. L., Weinerman B., Basole C., Salazar J. C., TLR8: The forgotten relative revindicated. Cell. Mol. Immunol. 9, 434–438 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R12] 12.Kawai T., Ikegawa M., Ori D., Akira S., Decoding Toll-like receptors: Recent insights and perspectives in innate immunity. Immunity 57, 649–673 (2024). [DOI] [PubMed] [Google Scholar]

[R13] 13.Mantovani S., Oliviero B., Varchetta S., Renieri A., Mondelli M. U., TLRs: Innate immune sentries against SARS-CoV-2 infection. Int. J. Mol. Sci. 24, 8065 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R14] 14.Comar C. E., Otter C. J., Pfannenstiel J., Doerger E., Renner D. M., Tan L. H., Perlman S., Cohen N. A., Fehr A. R., Weiss S. R., MERS-CoV endoribonuclease and accessory proteins jointly evade host innate immunity during infection of lung and nasal epithelial cells. Proc. Natl. Acad. Sci. U.S.A. 119, e2123208119 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R15] 15.Totura A. L., Baric R. S., SARS coronavirus pathogenesis: Host innate immune responses and viral antagonism of interferon. Curr. Opin. Virol. 2, 264–275 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R16] 16.Xia H., Cao Z., Xie X., Zhang X., Chen J. Y.-C., Wang H., Menachery V. D., Rajsbaum R., Shi P.-Y., Evasion of type I interferon by SARS-CoV-2. Cell Rep. 33, 108234 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R17] 17.Lei X., Dong X., Ma R., Wang W., Xiao X., Tian Z., Wang C., Wang Y., Li L., Ren L., Guo F., Zhao Z., Zhou Z., Xiang Z., Wang J., Activation and evasion of type I interferon responses by SARS-CoV-2. Nat. Commun. 11, 3810 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R18] 18.Hayn M., Hirschenberger M., Koepke L., Nchioua R., Straub J. H., Klute S., Hunszinger V., Zech F., Prelli Bozzo C., Aftab W., Christensen M. H., Conzelmann C., Müller J. A., Srinivasachar Badarinarayan S., Stürzel C. M., Forne I., Stenger S., Conzelmann K.-K., Münch J., Schmidt F. I., Sauter D., Imhof A., Kirchhoff F., Sparrer K. M. J., Systematic functional analysis of SARS-CoV-2 proteins uncovers viral innate immune antagonists and remaining vulnerabilities. Cell Rep. 35, 109126 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R19] 19.Sender R., Bar-On Y. M., Gleizer S., Bernsthein B., Flamholz A., Phillips R., Milo R., The total number and mass of SARS-CoV-2 virions. Infectious Diseases (except HIV/AIDS). Proc. Natl. Acad. Sci. U.S.A. 118, e2024815118 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R20] 20.Tugaeva K. V., Hawkins D. E. D. P., Smith J. L. R., Bayfield O. W., Ker D.-S., Sysoev A. A., Klychnikov O. I., Antson A. A., Sluchanko N. N., The mechanism of SARS-CoV-2 nucleocapsid protein recognition by the human 14-3-3 proteins. J. Mol. Biol. 433, 166875 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R21] 21.López-Muñoz A. D., Kosik I., Holly J., Yewdell J. W., Cell surface SARS-CoV-2 nucleocapsid protein modulates innate and adaptive immunity. Sci. Adv. 8, eabp9770 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R22] 22.Wang M., Valadez-Ingersoll M., Gilmore T. D., Control of nuclear localization of the nucleocapsid protein of SARS-CoV-2. Virology 600, 110232 (2024). [DOI] [PubMed] [Google Scholar]

[R23] 23.Zheng Z.-Q., Wang S.-Y., Xu Z.-S., Fu Y.-Z., Wang Y.-Y., SARS-CoV-2 nucleocapsid protein impairs stress granule formation to promote viral replication. Cell Discov. 7, 38 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R24] 24.Hu Y., Li W., Gao T., Cui Y., Jin Y., Li P., Ma Q., Liu X., Cao C., The severe acute respiratory syndrome coronavirus nucleocapsid inhibits type I interferon production by interfering with TRIM25-mediated RIG-I ubiquitination. J. Virol. 91, e02143-16 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R25] 25.Li Y., Li M., Xiao H., Liao F., Shen M., Ge W., Ou J., Liu Y., Chen L., Zhao Y., Wan P., Liu J., Chen J., Lan X., Wu S., Ding Q., Li G., Zhang Q., Pan P., The R203M and D377Y mutations of the nucleocapsid protein promote SARS-CoV-2 infectivity by impairing RIG-I-mediated antiviral signaling. PLOS Pathog. 21, e1012886 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R26] 26.Wang C., Xie J., Zhao L., Fei X., Zhang H., Tan Y., Nie X., Zhou L., Liu Z., Ren Y., Yuan L., Zhang Y., Zhang J., Liang L., Chen X., Liu X., Wang P., Han X., Weng X., Chen Y., Yu T., Zhang X., Cai J., Chen R., Shi Z.-L., Bian X.-W., Alveolar macrophage dysfunction and cytokine storm in the pathogenesis of two severe COVID-19 patients. EBioMedicine 57, 102833 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R27] 27.Wu T. T.-H., Travaglini K. J., Rustagi A., Xu D., Zhang Y., Andronov L., Jang S., Gillich A., Dehghannasiri R., Martínez-Colón G. J., Beck A., Liu D. D., Wilk A. J., Morri M., Trope W. L., Bierman R., Weissman I. L., Shrager J. B., Quake S. R., Kuo C. S., Salzman J., Moerner W. E., Kim P. S., Blish C. A., Krasnow M. A., Interstitial macrophages are a focus of viral takeover and inflammation in COVID-19 initiation in human lung. J. Exp. Med. 221, e20232192 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R28] 28.Zhou Z., Ren L., Zhang L., Zhong J., Xiao Y., Jia Z., Guo L., Yang J., Wang C., Jiang S., Yang D., Zhang G., Li H., Chen F., Xu Y., Chen M., Gao Z., Yang J., Dong J., Liu B., Zhang X., Wang W., He K., Jin Q., Li M., Wang J., Heightened innate immune responses in the respiratory tract of COVID-19 patients. Cell Host Microbe 27, 883–890.e2 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R29] 29.Liao M., Liu Y., Yuan J., Wen Y., Xu G., Zhao J., Cheng L., Li J., Wang X., Wang F., Liu L., Amit I., Zhang S., Zhang Z., Single-cell landscape of bronchoalveolar immune cells in patients with COVID-19. Nat. Med. 26, 842–844 (2020). [DOI] [PubMed] [Google Scholar]

[R30] 30.Sefik E., Qu R., Junqueira C., Kaffe E., Mirza H., Zhao J., Brewer J. R., Han A., Steach H. R., Israelow B., Blackburn H. N., Velazquez S. E., Chen Y. G., Halene S., Iwasaki A., Meffre E., Nussenzweig M., Lieberman J., Wilen C. B., Kluger Y., Flavell R. A., Inflammasome activation in infected macrophages drives COVID-19 pathology. Nature 606, 585–593 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R31] 31.Jalloh S., Olejnik J., Berrigan J., Nisa A., Suder E. L., Akiyama H., Lei M., Ramaswamy S., Tyagi S., Bushkin Y., Mühlberger E., Gummuluru S., CD169-mediated restrictive SARS-CoV-2 infection of macrophages induces pro-inflammatory responses. PLOS Pathog. 18, e1010479 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R32] 32.Merad M., Martin J. C., Pathological inflammation in patients with COVID-19: A key role for monocytes and macrophages. Nat. Rev. Immunol. 20, 355–362 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R33] 33.Salina A. C., Dos-Santos D., Rodrigues T. S., Fortes-Rocha M., Freitas-Filho E. G., Alzamora-Terrel D. L., Castro I. M., da Silva T. F. F., de Lima M. H., Nascimento D. C., Silva C. M., Toller-Kawahisa J. E., Becerra A., Oliveira S., Caetité D. B., Almeida L., Ishimoto A. Y., Lima T. M., Martins R. B., Veras F., do Amaral N. B., Giannini M. C., Bonjorno L. P., Lopes M. I., Benatti M. N., Batah S. S., Santana R. C., Vilar F. C., Martins M. A., Assad R. L., de Almeida S. C., de Oliveira F. R., Neto E. A., Cunha T. M., Alves-Filho J. C., Bonato V. L., Cunha F. Q., Fabro A. T., Nakaya H. I., Zamboni D. S., Louzada-Junior P., Oliveira R. D. R., Cunha L. D., Efferocytosis of SARS-CoV-2-infected dying cells impairs macrophage anti-inflammatory functions and clearance of apoptotic cells. eLife 11, e74443 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R34] 34.Bai Z., Cao Y., Liu W., Li J., The SARS-CoV-2 nucleocapsid protein and its role in viral structure, biological functions, and a potential target for drug or vaccine mitigation. Viruses 13, 1115 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R35] 35.Syed A. M., Taha T. Y., Tabata T., Chen I. P., Ciling A., Khalid M. M., Sreekumar B., Chen P.-Y., Hayashi J. M., Soczek K. M., Ott M., Doudna J. A., Rapid assessment of SARS-CoV-2–Evolved variants using virus-like particles. Science 374, 1626–1632 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R36] 36.Tang Y., Liu J., Zhang D., Xu Z., Ji J., Wen C., Cytokine storm in COVID-19: The current evidence and treatment strategies. Front. Immunol. 11, 1708 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R37] 37.Durinck S., Spellman P. T., Birney E., Huber W., Mapping identifiers for the integration of genomic datasets with the R/Bioconductor package biomaRt. Nat. Protoc. 4, 1184–1191 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R38] 38.Han H., Cho J.-W., Lee S., Yun A., Kim H., Bae D., Yang S., Kim C. Y., Lee M., Kim E., Lee S., Kang B., Jeong D., Kim Y., Jeon H.-N., Jung H., Nam S., Chung M., Kim J.-H., Lee I., TRRUST v2: An expanded reference database of human and mouse transcriptional regulatory interactions. Nucleic Acids Res. 46, D380–D386 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R39] 39.Xie Z., Bailey A., Kuleshov M. V., Clarke D. J. B., Evangelista J. E., Jenkins S. L., Lachmann A., Wojciechowicz M. L., Kropiwnicki E., Jagodnik K. M., Jeon M., Ma’ayan A., Gene set knowledge discovery with Enrichr. Curr. Protoc. 1, e90 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R40] 40.Imbiakha B., Ezzatpour S., Buchholz D. W., Sahler J., Ye C., Olarte-Castillo X. A., Zou A., Kwas C., O’Hare K., Choi A., Adeleke R. A., Khomandiak S., Goodman L., Jager M. C., Whittaker G. R., Martinez-Sobrido L., August A., Aguilar H. C., Age-dependent acquisition of pathogenicity by SARS-CoV-2 omicron BA.5. Sci. Adv. 9, eadj1736 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R41] 41.Kannan S., Subbaram K., Ali S., Kannan H., Molecular Characterization and Amino Acid Homology of Nucleocapsid (N) Protein in SARS-CoV-1, SARS-CoV-2, MERS-CoV, and Bat Coronavirus. J. Pure Appl. Microbiol. 14, 757–763 (2020). [Google Scholar]

[R42] 42.Du M., Chen Z. J., DNA-induced liquid phase condensation of cGAS activates innate immune signaling. Science 361, 704–709 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R43] 43.Zhang X., Han L., Hou J., Yang H., Xu H., Li G., Shu Q., Zhu D., Zheng Y., Gao C., Stress granule-localized USP8 potentiates cGAS-mediated type I interferonopathies through deubiquitination of DDX3X. Cell Rep. 43, 114248 (2024). [DOI] [PubMed] [Google Scholar]

[R44] 44.Zhao M., Xia T., Xing J.-Q., Yin L.-H., Li X.-W., Pan J., Liu J.-Y., Sun L.-M., Wang M., Li T., Mao J., Han Q.-Y., Xue W., Cai H., Wang K., Xu X., Li T., He K., Wang N., Li A.-L., Zhou T., Zhang X.-M., Li W.-H., Li T., The stress granule protein G3BP1 promotes pre-condensation of cGAS to allow rapid responses to DNA. EMBO Rep. 23, e53166 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R45] 45.Huang H.-S., Chen L., Chi J.-X., Lai S.-Y., Pi J., Shao Y.-M., Xu J.-F., Stress granules and cell death: Crosstalk and potential therapeutic strategies in infectious diseases. Cell Death Dis. 16, 495 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R46] 46.Deb P., Dai J., Singh S., Kalyoussef E., Fitzgerald-Bocarsly P., Triggering of the cGAS-STING pathway in human plasmacytoid dendritic cells inhibits TLR9-Mediated IFN production. J. Immunol. 205, 223–236 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R47] 47.Sharma S., Campbell A. M., Chan J., Schattgen S. A., Orlowski G. M., Nayar R., Huyler A. H., Nündel K., Mohan C., Berg L. J., Shlomchik M. J., Marshak-Rothstein A., Fitzgerald K. A., Suppression of systemic autoimmunity by the innate immune adaptor STING. Proc. Natl. Acad. Sci. U.S.A. 112, E710–E717 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R48] 48.Lama L., Adura C., Xie W., Tomita D., Kamei T., Kuryavyi V., Gogakos T., Steinberg J. I., Miller M., Ramos-Espiritu L., Asano Y., Hashizume S., Aida J., Imaeda T., Okamoto R., Jennings A. J., Michino M., Kuroita T., Stamford A., Gao P., Meinke P., Glickman J. F., Patel D. J., Tuschl T., Development of human cGAS-specific small-molecule inhibitors for repression of dsDNA-triggered interferon expression. Nat. Commun. 10, 2261 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R49] 49.Xu S., Ilyas I., Weng J., Endothelial dysfunction in COVID-19: An overview of evidence, biomarkers, mechanisms and potential therapies. Acta Pharmacol. Sin. 44, 695–709 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R50] 50.Eberhardt N., Noval M. G., Kaur R., Amadori L., Gildea M., Sajja S., Das D., Cilhoroz B., Stewart O., Fernandez D. M., Shamailova R., Vasquez Guillen A., Jangra S., Schotsaert M., Newman J. D., Faries P., Maldonado T., Rockman C., Rapkiewicz A., Stapleford K. A., Narula N., Moore K. J., Giannarelli C., SARS-CoV-2 infection triggers pro-atherogenic inflammatory responses in human coronary vessels. Nat. Cardiovasc. Res. 2, 899–916 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R51] 51.Lee M.-H., Perl D. P., Nair G., Li W., Maric D., Murray H., Dodd S. J., Koretsky A. P., Watts J. A., Cheung V., Masliah E., Horkayne-Szakaly I., Jones R., Stram M. N., Moncur J., Hefti M., Folkerth R. D., Nath A., Microvascular injury in the brains of patients with Covid-19. N. Engl. J. Med. 384, 481–483 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R52] 52.Li Q., Ye L., Talapaneni S., Meng Y., Wang C. R., Kalindjian K., Phan J., Chen Y., Demer L. L., Tintut Y., Hsu J. J., COL8A1 regulates endothelial phenotype in inflammatory endothelial-to-mesenchymal transition. Am. J. Physiol. Heart Circ. Physiol. 329, H1331–H1346 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R53] 53.Qian T., Maguire S. E., Canfield S. G., Bao X., Olson W. R., Shusta E. V., Palecek S. P., Directed differentiation of human pluripotent stem cells to blood-brain barrier endothelial cells. Sci. Adv. 3, e1701679 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R54] 54.Shen Y. H., Utama B., Wang J., Raveendran M., Senthil D., Waldman W. J., Belcher J. D., Vercellotti G., Martin D., Mitchelle B. M., Wang X. L., Human cytomegalovirus causes endothelial injury through the ataxia telangiectasia mutant and p53 DNA damage signaling pathways. Circ. Res. 94, 1310–1317 (2004). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R55] 55.Yamada S., Hashita T., Yanagida S., Sato H., Yasuhiko Y., Okabe K., Noda T., Nishida M., Matsunaga T., Kanda Y., SARS-CoV-2 causes dysfunction in human iPSC-derived brain microvascular endothelial cells potentially by modulating the Wnt signaling pathway. Fluids Barriers CNS 21, 32 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R56] 56.Cheng Y., Medina A., Yao Z., Basu M., Natekar J. P., Lang J., Sanchez E., Nkembo M. B., Xu C., Qian X., Nguyen P. T. T., Wen Z., Song H., Ming G.-L., Kumar M., Brinton M. A., Li M. M. H., Tang H., Intrinsic antiviral immunity of barrier cells revealed by an iPSC-derived blood-brain barrier cellular model. Cell Rep. 39, 110885 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R57] 57.Alvarez P. A., Tang A., Winters D. M., Kaushal P., Medina A., Nieto M. V., Kaczor-Urbanowicz K. E., St. Amant F., Reyes B. R., Kaake R. M., Fregoso O. I., Pyle A. D., Bouhaddou M., Tang H., Li M. M. H., Old World alphaviruses use distinct mechanisms to infect brain microvascular endothelial cells for neuroinvasion. Cell Rep. 44, 116305 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R58] 58.Wick K. D., Leligdowicz A., Willmore A., Carrillo S. A., Ghale R., Jauregui A., Chak S. S., Nguyen V., Lee D., Jones C., Dewar R., Lane H. C., Kangelaris K. N., Hendrickson C. M., Liu K. D., Sinha P., Erle D. J., Langelier C. R., Krummell M. F., Woodruff P. G., Calfee C. S., Matthay M. A., COMET Consortium , Plasma SARS-CoV-2 nucleocapsid antigen levels are associated with progression to severe disease in hospitalized COVID-19. Crit. Care 26, 278 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R59] 59.Shuaib M., Adroub S., Mourier T., Mfarrej S., Zhang H., Esau L., Alsomali A., Alofi F. S., Ahmad A. N., Shamsan A., Khogeer A., Hashem A. M., Almontashiri N. A. M., Hala S., Pain A., Impact of the SARS-CoV-2 nucleocapsid 203K/204R mutations on the inflammatory immune response in COVID-19 severity. Genome Med. 15, 54 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R60] 60.Laurent P., Yang C., Rendeiro A. F., Nilsson-Payant B. E., Carrau L., Chandar V., Bram Y., tenOever B. R., Elemento O., Ivashkiv L. B., Schwartz R. E., Barrat F. J., Sensing of SARS-CoV-2 by pDCs and their subsequent production of IFN-I contribute to macrophage-induced cytokine storm during COVID-19. Sci. Immunol. 7, eadd4906 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R61] 61.Shahbaz S., Bozorgmehr N., Lu J., Osman M., Sligl W., Tyrrell D. L., Elahi S., Analysis of SARS-CoV-2 isolates, namely the Wuhan strain, Delta variant, and Omicron variant, identifies differential immune profiles. Microbiol. Spectr. 11, e01256-23 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R62] 62.Wu H., Xing N., Meng K., Fu B., Xue W., Dong P., Tang W., Xiao Y., Liu G., Luo H., Zhu W., Lin X., Meng G., Zhu Z., Nucleocapsid mutations R203K/G204R increase the infectivity, fitness, and virulence of SARS-CoV-2. Cell Host Microbe 29, 1788–1801.e6 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R63] 63.Ren J., Wang S., Zong Z., Pan T., Liu S., Mao W., Huang H., Yan X., Yang B., He X., Zhou F., Zhang L., TRIM28-mediated nucleocapsid protein SUMOylation enhances SARS-CoV-2 virulence. Nat. Commun. 15, 244 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R64] 64.Liu D., Yang J., Cristea I. M., Liquid-liquid phase separation in innate immunity. Trends Immunol. 45, 454–469 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R65] 65.Wang L., Sola I., Enjuanes L., Zuñiga S., MOV10 helicase interacts with coronavirus nucleocapsid protein and has antiviral activity. mBio 12, e0131621 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R66] 66.Menten P., Wuyts A., Van Damme J., Macrophage inflammatory protein-1. Cytokine Growth Factor Rev. 13, 455–481 (2002). [DOI] [PubMed] [Google Scholar]

[R67] 67.Kitai R., Zhao M.-L., Zhang N., Hua L. L., Lee S. C., Role of MIP-1β and RANTES in HIV-1 infection of microglia: Inhibition of infection and induction by IFNβ. J. Neuroimmunol. 110, 230–239 (2000). [DOI] [PubMed] [Google Scholar]

[R68] 68.Ciechanowska A., Popiolek-Barczyk K., Pawlik K., Ciapała K., Oggioni M., Mercurio D., De Simoni M.-G., Mika J., Changes in macrophage inflammatory protein-1 (MIP-1) Family Members expression induced by traumatic brain injury in mice. Immunobiology 225, 151911 (2020). [DOI] [PubMed] [Google Scholar]

[R69] 69.Greene C., Connolly R., Brennan D., Laffan A., O’Keeffe E., Zaporojan L., O’Callaghan J., Thomson B., Connolly E., Argue R., Meaney J. F. M., Martin-Loeches I., Long A., Cheallaigh C. N., Conlon N., Doherty C. P., Campbell M., Blood–Brain barrier disruption and sustained systemic inflammation in individuals with long COVID-associated cognitive impairment. Nat. Neurosci. 27, 421–432 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R70] 70.I. Ogobuiro, C. J. Wehrle, F. Tuma, “Anatomy, thorax, heart coronary arteries” in StatPearls (StatPearls Publishing, 2025). [PubMed] [Google Scholar]

[R71] 71.Triantafyllis A. S., Sfantou D., Karapedi E., Peteinaki K., Kotoulas S. C., Saad R., Fountoulakis P. N., Tsamakis K., Tsiptsios D., Rallidis L., Tsoporis J. N., Varvarousis D., Hamodraka E., Giannakopoulos A., Poulimenos L. E., Ikonomidis I., Coronary implications of COVID-19. Med. Princ. Pract. 34, 1–12 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R72] 72.Sagris M., Theofilis P., Antonopoulos A. S., Tsioufis C., Oikonomou E., Antoniades C., Crea F., Kaski J. C., Tousoulis D., Inflammatory mechanisms in COVID-19 and atherosclerosis: Current pharmaceutical perspectives. Int. J. Mol. Sci. 22, 6607 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R73] 73.Weller B., Lin C.-W., Pogoutse O., Sauer M., Marin-dela Rosa N., Strobel A., Young V., Knapp J. J., Rayhan A., Falter C., Kim D.-K., Roth F. P., Falter-Braun P., A resource of human coronavirus protein-coding sequences in a flexible, multipurpose gateway Entry clone collection. G3 13, jkad105 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R74] 74.Yang E., Huang S., Jami-Alahmadi Y., McInerney G. M., Wohlschlegel J. A., Li M. M. H., Elucidation of TRIM25 ubiquitination targets involved in diverse cellular and antiviral processes. PLOS Pathog. 18, e1010743 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R75] 75.Badr C. E., Niers J. M., Tjon-Kon-Fat L.-A., Noske D. P., Wurdinger T., Tannous B. A., Real-time monitoring of nuclear factor κB activity in cultured cells and in animal models. Mol. Imaging 8, 278–290 (2009). [PMC free article] [PubMed] [Google Scholar]

[R76] 76.Ngo K. A., Kishimoto K., Davis-Turak J., Pimplaskar A., Cheng Z., Spreafico R., Chen E. Y., Tam A., Ghosh G., Mitchell S., Hoffmann A., Dissecting the regulatory strategies of NF-κB RelA target genes in the inflammatory response reveals differential transactivation logics. Cell Rep. 30, 2758–2775.e6 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R77] 77.Patro R., Duggal G., Love M. I., Irizarry R. A., Kingsford C., Salmon provides fast and bias-aware quantification of transcript expression. Nat. Methods 14, 417–419 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R78] 78.Love M. I., Soneson C., Hickey P. F., Johnson L. K., Pierce N. T., Shepherd L., Morgan M., Patro R., Tximeta: Reference sequence checksums for provenance identification in RNA-seq. PLOS Comput. Biol. 16, e1007664 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R79] 79.Chen Y., Chen L., Lun A. T. L., Baldoni P. L., Smyth G. K., edgeR v4: Powerful differential analysis of sequencing data with expanded functionality and improved support for small counts and larger datasets. Nucleic Acids Res. 53, gkaf018 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R80] 80.Sherman B. T., Hao M., Qiu J., Jiao X., Baseler M. W., Lane H. C., Imamichi T., Chang W., DAVID: A web server for functional enrichment analysis and functional annotation of gene lists (2021 update). Nucleic Acids Res. 50, W216–W221 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R81] 81.H. Wickham, Ggplot2: Elegant Graphics for Data Analysis (Springer International Publishing, ed. 2, 2016). [Google Scholar]

[R82] 82.Gao C.-H., Chen C., Akyol T., Dusa A., Yu G., Cao B., Cai P., ggVennDiagram: Intuitive Venn diagram software extended. iMeta 3, e177 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R83] 83.R. Kolde, pheatmap: Pretty heatmaps (2010). 10.32614/CRAN.package.pheatmap. [DOI]

[R84] 84.Wang X., Spandidos A., Wang H., Seed B., PrimerBank: A PCR primer database for quantitative gene expression analysis, 2012 update. Nucleic Acids Res. 40, D1144–D1149 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R85] 85.Choi M., Chang C.-Y., Clough T., Broudy D., Killeen T., MacLean B., Vitek O., MSstats: An R package for statistical analysis of quantitative mass spectrometry-based proteomic experiments. Bioinformatics 30, 2524–2526 (2014). [DOI] [PubMed] [Google Scholar]

[R86] 86.Teo G., Liu G., Zhang J., Nesvizhskii A. I., Gingras A.-C., Choi H., SAINTexpress: Improvements and additional features in Significance Analysis of Interactome software. J. Proteomics 100, 37–43 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R87] 87.Shannon P., Markiel A., Ozier O., Baliga N. S., Wang J. T., Ramage D., Amin N., Schwikowski B., Ideker T., Cytoscape: A software environment for integrated models of biomolecular interaction networks. Genome Res. 13, 2498–2504 (2003). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R88] 88.P. A. Alvarez, “The role of viral structural proteins in disrupting the physical and biochemical barriers to infection,” thesis, University of California, Los Angeles, Los Angeles, CA (2025). [Google Scholar]

PERMALINK

SARS-CoV-2 nucleocapsid induces hyperinflammation and vascular leakage through the Toll-like receptor signaling axis in macrophages

Zhenlan Yao

Pablo A Alvarez

Carolina Chavez

Yennifer Delgado

Prashant Kaushal

David W Buchholz

David Austin

Qian Li

Yanying Yu

Anne K Zaiss

Vaithilingaraja Arumugaswami

Qiang Ding

Jeffrey J Hsu

Robert Damoiseaux

Hector C Aguilar

Mehdi Bouhaddou

Alexander Hoffmann

Melody M H Li

Roles

Abstract

INTRODUCTION

RESULTS

Incoming and de novo–expressed nucleocapsids elicit opposing immune responses in macrophages

Fig. 1. The effect of incoming or de novo–synthesized nucleocapsid on the induction of proinflammatory cytokines and chemokines in macrophages.

Nucleocapsid enhances TLR-driven inflammation and suppresses RLR-mediated antiviral signaling

Fig. 2. Nucleocapsid enhances TLR-driven inflammation and suppresses RLR-mediated antiviral signaling.

Nucleocapsid-driven transcriptional changes are mediated by several shared and unique pathways

Fig. 3. Nucleocapsid shifts THP-1 macrophage-like cells from intracellular sensing to extracellular sensing.

Hyperactivation of inflammation in macrophages is a conserved feature of pathogenic coronavirus nucleocapsids

Fig. 4. Differential infection and induction of inflammation-related transcripts in THP-1 macrophage-like cells across SARS-CoV-2 variants and pathogenic betacoronaviruses.

SARS-CoV-2 N engages cGAS to modulate macrophage inflammatory signaling

Fig. 5. N-cGAS interaction likely drives hyperinflammation in THP-1 macrophage-like cells.

SARS-CoV-2 N variants differentially shape macrophage inflammatory responses and endothelial barrier integrity

Fig. 6. SARS-CoV-2 N globally alters cytokine secretion and triggers endothelial barrier breakdown in a variant-specific manner.

DISCUSSION

MATERIALS AND METHODS

Ethics statement

Cell culture

THP-1 and primary monocyte differentiation into macrophages

Human pluripotent stem cell–derived BMEC differentiation and culture

SARS-CoV-2 infection and VLP inoculation

Construction of SARS-CoV-2 (Wuhan-Hu-1 strain, Delta and Omicron variants), SARS-CoV-1, MERS-CoV N–expressing plasmids, and tagged host factor (cGAS and PABP1) plasmids

Generation of stable HEK293T-inducible SARS-CoV-2 protein expression cell lines

Cloning of an NF-κB-responsive luciferase reporter plasmid

IFN and NF-κB activation luciferase assays

Lentivirus packaging and stable THP-1 cell line construction

Macrophage stimulation with TLR and RLR agonists and supernatant cytokine and chemokine detection

RNA isolation for RNA-seq analysis

RNA-seq bioinformatic analysis

Coronavirus N protein sequence alignment

Quantitative reverse transcription PCR (RT-qPCR)

Immunoprecipitation and immunoblotting

AP-MS and proteomic data visualization

Measurement of TEER

Statistical analysis

Acknowledgments

Funding:

Author contributions:

Competing interests:

Data, code, and materials availability:

Supplementary Materials

The PDF file includes:

Other Supplementary Material for this manuscript includes the following:

REFERENCES

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases