Porcine deltacoronavirus nucleocapsid protein inhibits RIG-I signaling pathway by suppressing G3BP1-mediated stress granules formation

ABSTRACT

Porcine deltacoronavirus (PDCoV) is an emerging zoonotic pathogen causing severe diarrhea in suckling piglets. Stress granules (SGs), stress-induced cytoplasmic condensates, are pivotal to the antiviral innate immune response. However, the interaction between PDCoV and SGs remains unclear. Here, we found that PDCoV triggered the formation of SGs during early infection, but these structures gradually disassembled in the late stage, while exogenously induced SGs significantly inhibited PDCoV replication. Mechanistically, we identified the nucleocapsid (N) protein as a potent SGs antagonist. The N protein can interact with Ras-GTPase-activating protein SH3-domain-binding protein 1 (G3BP1, the core component of SGs), thereby disrupting SGs assembly. Furthermore, we found that G3BP1 promotes type I interferon (IFN-I) production by recruiting retinoic acid-inducible gene-I (RIG-I). The PDCoV N protein competitively blocks this G3BP1-RIG-I interaction, leading to impaired IFN-I signaling. In addition, we found that this SGs-inhibitory function is conserved across diverse coronaviruses and correlates with structural conservation of N proteins. Our findings reveal a novel and conserved immune evasion mechanism in which coronaviruses antagonize the SGs-mediated antiviral pathway via their N proteins to promote replication. Importantly, this mechanism is structurally conserved across diverse coronaviruses, thereby identifying the N protein as a promising target for broad-spectrum antiviral development.

Introduction

Coronaviruses (CoVs), the largest family of positive-sense single-stranded RNA viruses, are divided into four genera (α-CoV, β-CoV, γ-CoV, and δ-CoV). CoVs primarily induce respiratory, neurological, and gastrointestinal disorders in both humans and animals, and even leading to death [1–6]. Porcine deltacoronavirus (PDCoV), an emerging porcine enteric coronavirus, induces severe diarrhea, dehydration and high mortality in neonatal piglets [7,8]. PDCoV was first identified in 2014 from the intestinal sample of diarrheic piglets in the United States, and it has spread rapidly to numerous countries, including Canada, Vietnam, Mexico, South Korea and China [9–11]. Surveillance data from our laboratory on diarrheic piglets in China revealed a PDCoV-positive detection rate of 30%, with an increasing trend year by year [12–15]. Notably, PDCoV exhibits cross-species transmission potential. It can break species barriers to infect avian hosts such as chickens and sparrows, as well as mammalian hosts including calves and mice [11]. In 2021, PDCoV was detected in blood samples from three Haitian children with acute undifferentiated febrile illness, indicating that PDCoV has the ability to infect humans [16]. Moreover, compared to other porcine enteric coronaviruses, PDCoV has broad tissue tropism and can infect tissues such as the intestines, lungs and kidneys. However, there remain no effective preventive or therapeutic strategies to prevent PDCoV infection.

When cells are exposed to external stimuli such as viral infection, oxidative stress, or heat shock, distinct eukaryotic initiation factor 2α (eIF2α) kinases, including protein kinase R-like endoplasmic reticulum kinase (PERK), protein kinase R (PKR), heme-regulated inhibitor kinase (HRI), and general control nonderepressible 2 (GCN2), will be activated [17–21]. These activated kinases then induce the phosphorylation of eIF2α, which in turn triggers a global arrest of host protein translation [22]. Ultimately, mRNAs released into the cytoplasm aggregate with RNA-binding proteins (RBPs) to form stress granules (SGs) [20,23]. Among the key components of SGs, Ras-GTPase-activating protein-binding protein 1 (G3BP1) acts as a core scaffolding protein and serves as a molecular switch that mediates SGs formation [24]. Notably, the RBPs in SGs include not only host factors essential for viral replication but also viral proteins and antiviral factors. Consequently, SGs exert antiviral functions through two primary mechanisms: 1) sequestering viral RNA and host translation factors to reduce the viral proteins translation [21,22,25,26]; 2) serving as antiviral hubs that recruit pattern recognition receptors (PRRs) and host restriction factors to regulate the antiviral innate immune response [25–28].

Current studies indicate that some CoVs have evolved multiple strategies to suppress SGs formation. For instance, the nsp15 protein of infectious bronchitis virus (IBV, γ-CoV) inhibits SGs formation by reducing viral dsRNA accumulation and subsequent PKR activation. Porcine epidemic diarrhea virus (PEDV, α-CoV) utilizes host caspase-8 to cleave G3BP1, disrupting G3BP1-dependent SGs formation [29]. SARS-CoV-2 (β-CoV) suppresses PKR-induced SGs formation via two distinct pathways: hijacking host adenosine deaminase acting on RNA 1 (ADAR1) to edit viral dsRNA, and directly interacting with PKR [30]. However, the relationship between SGs and δ-CoVs replication remains largely unclear.

In this study, we demonstrated that SGs exhibit an inhibitory effect on PDCoV replication, while PDCoV infection suppresses SGs formation at the late stages of replication. Through screening viral proteins, we identified that the PDCoV N protein can bind to G3BP1 and disrupts G3BP1 mediated SGs formation. Furthermore, we found PDCoV N protein competitively impedes the recruitment of retinoic acid-inducible gene I (RIG-I) by G3BP1, leading to the suppression of type I interferon (IFN-І) production. Additionally, we demonstrated that multiple CoVs N proteins can also suppress both SGs formation and RIG-I signaling pathway. Our findings highlight the critical role of the CoVs N protein as a virulence factor and suggest it as a promising target for the development of novel antiviral therapies.

Materials and methods

PDCoV infection of piglets

Six 5-day-old healthy colostrum-deprived piglets (Deyuan Animal Husbandry, Henan, China) were randomly assigned to either a mock-infected group (n = 3, orally inoculated with 10 mL MEM) or a PDCoV-infected group (n = 3, challenged with 10 mL of 10⁷ TCID₅₀/mL PDCoV), with sample sizes determined by previous pathogenicity studies [13]. Animals were maintained on a milk replacer diet every 3 hours and monitored daily for clinical symptoms until 3 days post-infection (dpi), at which point all piglets were humanely euthanized for the collection of jejunum and ileum tissues (stored at −80°C). Euthanasia was performed via intramuscular injection of Zoletil to induce deep anesthesia and subsequent cessation of life, with an onset of action typically occurring within 1–2 minutes; death was verified by the permanent cessation of heartbeat and respiration. Following necropsy, all carcasses underwent high-pressure steam sterilization to ensure harmless disposal. While group allocations were known during the experiment, subsequent viral load measurements were conducted in a blinded manner to ensure objective analysis.

Cells and virus

LLC-PK1 cells were grown in MEM supplemented with 5% fetal bovine serum (FBS, Gibco, Carlsbad USA), 1% non-essential amino acids (NEAA, Gibco, Carlsbad USA), and 1% HEPES. Human embryonic kidney 293T (HEK 293T) cells were cultured in Dulbecco’s Modified Eagle Medium (DMEM, Gibco, Carlsbad USA) supplemented with 10% FBS. All cells were maintained in 5% CO₂ at 37°C. PDCoV HNZK-02 strain (GenBank: MH708123.1) was isolated and preserved in our laboratory [31] and propagated in LLC-PK1 cells with 5 µg/mL trypsin (Gibco, Carlsbad USA).

Antibodies and reagents

The PDCoV N-specific polyclonal antibody was preserved in our laboratory. Rabbit anti-G3BP1 (A3968), rabbit anti-PERK (A18196) and rabbit anti-GCN2(A7155) were purchased from ABclonal (Wuhan, China). Rabbit anti-eIF2α (11170–1-AP), mouse anti-β-actin (81115–5-RR) antibodies were obtained from Proteintech (Wuhan, China). Rabbit anti-RIG-I(AF7890), rabbit anti-phospho-PKR (AF1474) and rabbit anti-PKR (AF2125) were acquired from Beyotime (Shanghai, China). Mouse anti-HA tag (5017), mouse anti-FLAG tag (14793), rabbit anti-phospho-eIF2α (3398), rabbit anti-TBK1 (38066), and rabbit anti-IRF3 (11904), and rabbit anti-MYC tag (2278) were sourced from Cell Signaling Technology (Danvers, USA). Rabbit anti-GCN2 (phosphor T899, AB75836) was acquired from Abcam (Cambridge, United Kingdom). Rabbit anti-MAVS (PA5-84160) was obtained from Thermo Fisher Scientific (Waltham, MA, USA). Rabbit anti-phospho-PERK (BS66100) was procured from Bioworld Technology (Beijing, China). Alexa Fluor 488 goat anti-rabbit IgG, Alexa Fluor 594 goat anti-mouse IgG, Alexa Fluor 488 goat anti-mouse IgG, and Alexa Fluor 594 goat anti-rabbit IgG secondary antibodies were obtained from Solarbio (Beijing, China). Poly(I:C) was purchased from Invivogen (Toulouse, France).

Plasmids and transfection

The porcine G3BP1 gene (GenBank: NM_001205405) was amplified with cDNA isolate from LLC-PK1 cells and cloned into pCAGGS to construct pCAGGS-G3BP1 plasmid. The recombinant plasmids were successfully constructed to express of PDCoV nsp2, nsp5, nsp6, nsp7, nsp8, nsp9, nsp10, nsp12, nsp13, nsp14, nsp15, nsp16, S, E, M, N, NS6, NS7, NS7a, and N truncation mutants (N1-141, N1-176, N142-292, N177-292, N177-343), respectively. The N protein sequences of TGEV (GenBank: NC_038861.1), PEDV (GenBank: KX834131.1), Bovine coronavirus (BCoV) (GenBank: OP866727.1), Canine coronavirus (CCoV) (GenBank: MZ320954.1), SARS-CoV-2 (GenBank: NCO45512.2), IBV (GenBank: MZ367369.1), Sparrow coronavirus (SPCoV) (GenBank: NC016992.1) were synthesized by AuGCT Biotech (China) and cloned into pCAGGS-HA, respectively. To generate the recombinant plasmids used in this study, specific primers were designed and synthesized based on sequences available in GenBank. The full-length porcine G3BP1 gene (GenBank: NM_001205405) was amplified from cDNA of LLC-PK1 cells using primers F (5’-ccgggcatgcctcgagctagcATGGTTATGGAGAAGCCTA-3’) and R (5’-ttggcagagggaaaaagatctTCACTGCCTTGGAGTAA-3’) and subsequently cloned into the pCAGGS vector. To identify the functional domains of the PDCoV N protein, four truncated variants were constructed: N1–141 aa (F: 5’-gttccagattacgctgaattcATGGCCGCACCAGTAGTCC-3’ and R: 5’-attaagatctgctagctcgagTTAGTCAACTCTGAAACCTTGAGCTGG-3’); N1–176 aa (F: 5’-gttccagattacgctgaattcATGGCCGCACCAGTAGTCC-3’ and R: 5’-attaagatctgctagctcgagTTAAGATTGGTCGCGTTTCCTG-3’); N142–292 aa (F: 5’-gttccagattacgctgaattcATGCCCTTCAACGCTAGAGGA-3’ and R: 5’-attaagatctgctagctcgagTTAGGGTGGCTCATAGGTCTGG-3’); N177-292 aa (F: 5’-gttccagattacgctgaattcATGGCAACCGGTCTCGTACTG-3’ and R: 5’-attaagatctgctagctcgagTTAGGGTGGCTCATAGGTCTGG-3’); and N177–343 aa (F: 5’-gttccagattacgctgaattcATGGCAACCGGTCTCGTACTG-3’ and R: 5’-attagatctgctagctcgagCTACGCTGCTGATTCCTGCTTT-3’). All recombinant constructs were verified by DNA sequencing to ensure sequence accuracy before being utilized in transfection experiments. The cells were transfected with Lipofectamine 3000 (Thermo Fisher Scientific, Waltham, USA).

Real-time quantitative PCR (RT-qPCR)

Total RNA was extracted using Trizol reagent and reverse transcribed into cDNA with a reverse transcription kit (Vazyme, Nanjing, China). The cDNA was served as a template for the amplification reaction. The reaction program consisted of an initial denaturation at 98°C for 3 min, followed by 39 cycles of 98°C for 15 s, annealing at 54°C for 30 s, extension at 72°C for 1 min, and finally 72°C for 5 min. The specific primers used for the quantification of target genes were as follows: β-actin (F: 5’-GCGGCATCCACGAAACTAC-3,’ R: 5’-GATCTCCTTCTGCATCCTGTC-3’), PDCoV (F: 5’-CGTTAACCTTCTCACCACTT-3,’ R: 5’-GCTGAGAGTCTGGTTATT-3’), ISG15 (F: 5’-CCTGTTGATGGTGCAAAGCT-3,’ R: 5’-TGCACATAGGCTTGAGGTCA-3’), IFN-β (F: 5’-TCTTTCCATGAGCTACCAACTTGCT-3,’ R: 5’-GCAGTATTCAAGCCTCCCATTC-3’), and IFN-α (F: 5’-CATCCTGGCTGTGAGGAAATAC-3,’ R: 5’-GAGGTTTGTGGAGGAAGAGAAG-3’).β-actin was employed as an internal endogenous control to normalize mRNA levels, and the relative gene expression was calculated using the 2^−ΔΔT method.

G3BP1 expression knockdown

Small interfering RNAs (siRNA) targeting porcine G3BP1 were designed and synthesized by Gene Pharma (Shanghai, China). LLC-PK1 cells were seeded into 12-well plates and transfected with siRNA when they reached approximately 70% confluence. Cells were lysed to assess interference efficiency by Western blotting assay at 24 h post-transfection. The sequences of siG3BP1 are as follows, siG3BP1-1: 5’-CACCUCAUGUUGUUAAAGUTT-3,’ siG3BP1-2: 5’-CAGGCUUUGAGGAGAUUCATT-3,’ siG3BP1-3: 5’-GGACAAGUUAGAGCUUAAATT-3.’

Indirect immunofluorescence assay and confocal microscopy

Cells were fixed with anhydrous ethanol for 12 h at 4°C, and then permeabilized with 0.1% Triton X-100 at room temperature for 15 min. Subsequently, cells were blocked with 5% (w/v) bovine serum albumin (BSA, Sigma, USA) for 30 min at 37°C. Then, cells were incubated with primary antibodies and corresponding secondary antibodies and stained with 4,6-diamidino-2-phenylindole (DAPI, Sigma, USA). Images were captured using fluorescence microscope (Olympus, Tokyo, Japan) or laser scanning confocal microscope (ZEISS, Oberkochen, Germany).

Western blotting and immunoprecipitation

Cells were lysed by radio immunoprecipitation assay (RIPA) lysis buffer (Beyotime, Shanghai, China) containing 1 mM Phenylmethanesulfonyl fluoride (PMSF) and phosphatase inhibitors (Thermo Fisher Scientific, Waltham, USA). Samples were centrifuged at 12,000 rpm for 20 min, then the supernatant was purified by specific agarose beads (Thermo Fisher Scientific, Waltham, USA) on a rotating incubator for 8 h at 4°C. The beads were collected and washed three times with PBS. The immunoprecipitated complexes were detected by Western blotting with enhanced chemiluminescence (ECL) system. β-actin was used as the reliable marker of internal control.

LC-MS/MS analysis, protein-protein interaction (PPI) network, and Gene Ontology (GO) analysis

The supernatant of cell lysates from PDCoV-N-HA-transfected LLC-PK cells was immunoprecipitated with anti-HA agarose beads (Thermo Fisher Scientific, Waltham, USA) on a rotating incubator at 4°C for 8 h. The beads were collected and washed three times with PBS. The immunoprecipitated complexes were eluted and separated by SDS-PAGE. The protein bands of interest were excised and subjected to LC-MS/MS analysis at PTM BIO (Hangzhou, China). The raw MS data were processed and analyzed using Metascape and the STRING database (https://string-db.org/). Potential protein-protein interactions among the identified proteins were retrieved from the STRING database, and the PPI network was constructed using Cytoscape software. For Gene Ontology (GO) annotation, all identified proteins were classified into three categories (cellular component, biological process, and molecular function) using the DAVID database (https://david.ncifcrf.gov/home.jsp).

Luciferase reporter gene assays

HEK 293T cells were seed into 24-well plates and cultured until reaching 70% confluence. Then, the cells were transfected with luciferase reporter plasmids (pGL3-IFN-β-Luc or pGL3-ISRE-Luc), internal control plasmid, pTK-RL, and protein expression plasmids or empty plasmid as indicated for each experiment. After 24 h of transfection, the cells were stimulated with poly (I:C) for 18 h. Cells were lysed, and luciferase activity was measured using a dual-luciferase reporter gene assay kit (TransGen Biotech, Chengdu, China). Firefly luciferase activity is normalized to the Renilla luciferase activity. All reporter assays were done in triplicate (three wells) and then the whole experiment was repeated three times.

Statistical analysis

All data are presented as mean ± standard deviation (SD) from at least three independent experiments. Statistical analyses were performed using GraphPad Prism (GraphPad Inc., USA). For group comparisons, the following tests were applied as appropriate: Dunnett’s multiple comparison test, unpaired t-test, Sidak’s multiple comparison test, and Tukey’s multiple comparison test. A p-value of less than 0.05 was considered statistically significant.

Results

PDCoV inhibits SGs formation to ensure its replication

To investigate the potential relationship between PDCoV infection and the formation of host SGs, the PDCoV-infection model in piglet developed in our laboratory was applied in current study [13]. Briefly, 5-day-old piglets were orally inoculated with PDCoV at a dose of 10⁸ TCID₅₀ per animal. Piglets developed severe diarrhea at 3 dpi and were then euthanized, the intestinal tissues were collected and examined for SGs formation using G3BP1 as a marker protein. The results showed that SGs were observed in the intestinal cells of PDCoV-infected piglets, whereas no SGs were detected in the control group (Figure 1(A)). Furthermore, LLC-PK1 cells were infected with PDCoV at an MOI of 0.01 and the kinetics of SGs formation was monitored at 4-hour intervals, and arsenite (Ars) treatment (750 μM) was used as the SGs formation control. The results showed Ars treatment induced SGs formation in approximately 90% of uninfected cells, while the percentage of the PDCoV-infected cells containing SGs was markedly lower than Ars-treated cells. And it exhibited dynamic changes of SG-positive cell over time, 33% at 12 hpi, 40% at 16 hpi, peaking to 58% at 20 hpi, and then declining to 32% at 24 hpi (Figure 1(B), (p < 0.001)). These observations suggested that PDCoV may suppress SGs formation at the late stages of PDCoV replication. To further confirm this hypothesis, LLC-PK1 cells were treated with Ars following PDCoV infection. The results showed that SGs-positive cells and the number of per cell both markedly reduced in PDCoV-infected cells (p < 0.001), when comparing with the only Ars-treated group (Figure 1(C)), demonstrating that PDCoV can inhibit SGs formation.

Figure 1.

PDCoV inhibits SGs formation to ensure its replication. (A) The immunofluorescence analysis of SGs formation in jejunum and ileum of piglets. Infected cells were detected with anti-PDCoV antibody (red), SGs with anti-G3BP1 (green, pointed with white arrows) and cell nuclei with DAPI (blue). Scale bar: 50 μm. (B) The kinetics of SGs formation during PDCoV infection. LLC-PK1 cells were infected with PDCoV (MOI = 0.01) and subjected to immunostaining at the indicated time points (B₁). The scatter plot shows the percentage of SGs-positive cells in the total infected cells, which was calculated based on 6 randomly selected areas (B₂). Scale bar: 10 μm. (C) The effect of PDCoV infection on SGs induced by Ars. LLC-PK1 cells were infected with PDCoV (MOI = 0.01) or MEM, and treated with Ars for 50 min at 20 hpi, followed by immunostaining (C₁). The percentage of SG-positive cells among total PDCoV-infected cells was quantified and compared with Ars-treated control cells, and calculated based on 4 randomly selected areas (C₂). The percentage of the number of SGs in virus-positive cells compared with that in normal cells treated with Ars was quantified and calculated based on 10 randomly selected cells (C₃). Scale bar: 10 μm. (D) SGs formation induced by different kind of stresses. LLC-PK1 cells were treated with different stress (heat shock treatment 50°C for 20 min, 750 μM Ars for 50 min, 80 μM TG for 8 h, or 2 μM H₂O₂ for 6 h), and SGs formation was captured and confirmed by laser scanning confocal microscope. (E) LLC-PK1 cells were subjected to PDCoV infection (MOI = 0.1) after various stressors. The change of PDCoV N expression was detected by WB at 12 hpi. Statistical analysis: data are presented as mean ± SD (n ≥ 3). Dunnett’s multiple comparison test (df = 25) was used for Figure 1(B)₂, and unpaired t-test was used for Figure 1(C)₂ (df = 6) and 1C₃ (df = 18).

We next evaluated the effect of SGs formation on PDCoV replication. LLC-PK1 cells were treated with various stressors, including heat shock, Ars, thapsigargin (TG), and H₂O₂, to induce SGs formation. CCK8 assays were performed to find the non-cytotoxic concentrations/protocols in LLC-PK1 cells for each stressor (Supplementary Fig. S1). Compared with the control group, robust SGs formation was observed in treated cells (Figure 1(D)). Subsequently, cells exposed to these stimuli were infected with PDCoV for 12 h, and viral replication was assessed. The results showed that PDCoV N protein levels were significantly reduced in the stressed LLC-PK1 cells, suggesting a correlation between the induction of SGs and the restriction of PDCoV replication, implying that SGs may function as a cellular defense platform (Figure 1(E)). Together, these findings suggest that PDCoV inhibits SGs formation to facilitate its own replication.

PDCoV N protein inhibited SGs formation to promote PDCoV replication

To identify which PDCoV-encoding protein was essential for inhibiting SGs formation, the PDCoV viral structural proteins (S, E, M, and N), accessory proteins (NS6, NS7, and NS7a) and nonstructural protein (nsp2, nsp5, nsp6, nsp7, nsp8, nsp9, nsp10, nsp12, nsp13, nsp14, nsp15 and nsp16) were eukaryotic expressed and confirmed successfully expressed by using anti-HA antibody (Figure 2(A)). Among these, the PDCoV N protein exhibited the strongest inhibitory effect on SGs formation after Ars treatment (approximately 100%, p < 0.001), following with the nsp15 and NS6 (about 60–70%) (Figure 2(B)).

Figure 2.

PDCoV N protein inhibited SGs formation to promote PDCoV replication. (A) Screening viral proteins that suppress SGs formation. LLC-PK1 cells were transfected with plasmids encoding HA-tagged PDCoV proteins or with vector pCAGGS. At 24 h post-transfection, cells were treated with Ars (750 μM for 50 min). PDCoV proteins were detected with anti-HA antibody (red) and SGs were detected with anti-G3BP1 (green). Cell nuclei were stained with DAPI (blue). Scale bar: 10 μm. (B) The scatter plot shows the percentage of SGs-positive cells in the total transfected cells. (C) The scheme of functional division of PDCoV N protein and construction of N protein truncations. (D) The effects of PDCoV N truncation mutants on the formation of SGs. The HA-tagged N protein truncations were transfected into LLC-PK1 cells, and cells were treated with Ars at 24 h post-transfection. Cells were immunostained (D₁) and the percentage of SGs positive cells in transfected cells to total transfected cells were calculated based on 6 randomly selected areas (D₂). Scale bar: 10 μm. (E-F) PDCoV N protein and N142-292 promoted virus replication. Cells were transfected with expression vectors of N or N protein truncations and infected with PDCoV (MOI = 0.01), and the mRNA expression levels of PDCoV at 24 hpi were detected using RT-qPCR. Statistical analysis: data are represented as means ± SD (n ≥ 3). For Figure 2(B), Dunnett’s multiple comparison test was used (df = 38). For Figure 2(D) ₂, Sidak’s multiple comparison test was used (df = 35). Figure 2(E) was analyzed by unpaired t-test (df = 6). For Figure 2(F), Tukey’s multiple comparison test was used (df = 12).

Considering that coronavirus N protein is the most abundant structural protein and plays multiple roles in viral replication and pathogenesis, we focused on PDCoV N protein in our further study. To identify the region of the PDCoV N protein responsible for suppressing SGs formation, 5 truncated variants (N1–141 aa, N1–176 aa, N142–292 aa, N177-292, and N177–343 aa) were constructed (Figure 2(C)). Compared to the Ars positive control, the N142–292 aa significantly inhibited Ars-induced SGs formation (Figure 2, p < 0.001), indicating that the C-terminal domain (CTD) of PDCoV N protein is primarily responsible for this inhibitory function. We further assessed the impact of full-length N protein and its truncations on viral replication. Exogenous expression of the full-length N protein significantly increased PDCoV genomic copies (p = 0.002; Figure 2(E), p = 0.002). Among the truncations, N1–176aa, N142–292aa, N177-292, and N177–343aa all enhanced viral replication relative to the control, with N142–292aa exerting the strongest effect (Figure 2(F)). In summary, these results demonstrate that the PDCoV N protein inhibits SGs formation to facilitate viral replication (p < 0.001).

PDCoV N protein inhibits SGs formation independently of p-eIf2α

SGs are membraneless organelles that form in host cells when protein translation is stalled, enabling the temporary storage of translation-arrested mRNA–ribosome complexes. Phosphorylation of eIF2α represents a key mechanism through which host protein synthesis is inhibited, thus, we first examined the phosphorylation status of eIF2α and its upstream kinases during PDCoV infection. In intestinal tissues from PDCoV-infected piglets, levels of p-PERK and p-eIF2α were significantly elevated compared to the control group, whereas the phosphorylation levels of PKR and GCN2 remained unchanged (Figure 3(A)). We further assessed the phosphorylation dynamics in PDCoV-infected LLC-PK1 cells. The results revealed a time-dependent increase in N protein expression, along with increases in phosphorylated PERK and eIF2α levels following infection. These levels peaked at 16 hpi and remained until 24 hpi, reaching degrees of activation comparable to those observed in the Ars-treated group (Figure 3(B)). These findings collectively indicate that the PDCoV N protein does not suppress the activation of the host PERK-eIF2α pathway during PDCoV infection.

Figure 3.

PDCoV N protein inhibits SGs formation independently of p-eIf2α. (A) The phosphorylation levels of eIf2α and its kinases (PERK, PRK, and GCN2) in intestinal tissue samples of PDCoV-infected piglets. (B) The phosphorylation levels of eIf2α and its kinases in LLC-PK1 cells infected with PDCoV (0.01 MOI) at different timepoints post infection. Cells were treated with Ars (750 μM for 50 min) as a positive control. (C) The effect of PDCoV N protein on the activation of PERK-eIf2α pathway under the stimulation of Ars. LLC-PK1 cells were treated with 750 μM Ars for 50 min at 48 h after transfection with the pCAGGA-HA -N or pCAGGA-HA vector.

To further confirm that the PDCoV N protein indeed fails to inhibit PERK-eIF2α activation, we transfected LLC-PK1 cells with pCAGGS-PDCoV-N and subsequently treated them with Ars to evaluate PERK and eIF2α phosphorylation. Compared to the pCAGGS control, cells expressing the N protein showed no significant alterations in the levels of eIF2α, p-eIF2α, PERK, or p-PERK (Figure 3(C)). Collectively, these data indicate that PDCoV N-mediated inhibition of SGs formation occurs downstream of eIF2α phosphorylation, suggesting that the N protein targets the maturation or condensation phase of SGs assembly rather than the initiation of stress signaling.

The PDCoV N protein inhibits SGs formation by binding to G3BP1

To further elucidate the mechanism of inhibitory effect of PDCoV N protein on SGs formation, we transfected LLC-PK1 cells with pCAGGS-PDCoV-N and performed immunoprecipitation-mass spectrometry to identify host proteins that interact with the N protein. The gene ontology analysis showed that the proteins interacting with the N protein are primarily located in ribosomes and the endoplasmic reticulum, playing roles in RNA and protein synthesis (Figure 4(A)). A protein-protein interaction (PPI) network was constructed to further analyze these interactions. Notably, several key proteins essential for SGs formation, including G3BP1, G3BP2, DDX3X, and PABPC4, were identified among the interactors (Figure 4(B)).

Figure 4.

The PDCoV N protein inhibits SGs formation by binding to G3BP1. (A) Gene Ontology (GO) analysis of host proteins interacted with PDCoV N protein. Protein samples from LLC-PK1 cells transfected with PDCoV-N-HA protein were subjected to anti-HA agarose bead immunoprecipitation, and the proteins were identified by LC-MS/MS. (B) PPI network of proteins interacted with PDCoV N protein constructed using cytoscape. (C-D) The effects of G3BP1 overexpression (C) and knockdown (D) on the replication of PDCoV. After transfecting LLC-PK1 cells with G3BP1 or siG3BP1 24 h, the LLC-PK1 cells were infected with PDCoV (MOI = 0.01) and the replication of PDCoV was evaluated at 24 hpi. (E) Co-IP analysis of the interaction between Flag-G3BP1 and HA-N in HEK-293T cells. (F) Confocal microscopy showing the colocalization of N protein (red) and G3BP1 (green) in the cytoplasm. (G) Overexpression of N protein interferes with G3BP1-mediated SGs (750 μM Ars for 50 min) assembly. 3.0 μg per 1 × 10⁵ cells G3BP1 overexpression plasmid was transfected into LLC-PK1 cells. Scale bar: 10 μm. Statistical analysis: data are presented as mean ± SD (n ≥ 3). Statistical analyses were performed using unpaired t-test: df = 10 for Figure 4(C) (left); df = 4 for Figure 4(C) (right); df = 10 for Figure 4(D).

Given the central role of G3BP1 in SGs formation, we investigated whether G3BP1 influences PDCoV replication. To this end, we constructed a pCAGGS-G3BP1-Flag plasmid and transfected it into LLC-PK1 cells, followed by infection with PDCoV. Compared with cells transfected with the empty pCAGGS vector, cells overexpressing G3BP1 exhibited a significant reduction in both viral titers and PDCoV N protein expression (Figure 4, p < 0.001), suggesting that G3BP1 overexpression suppresses PDCoV replication. Conversely, when G3BP1 was knocked down using specific siRNA in LLC-PK1 cells prior to PDCoV infection, viral titers and PDCoV N protein levels were significantly increased relative to the control group (Figure 4, p < 0.001). These data suggested that G3BP1 can inhibit PDCoV replication, which might be associated with its interaction with PDCoV N protein.

We further validated the interaction between G3BP1 and PDCoV N protein using co-immunoprecipitation. HEK 293T cells were transfected with pCAGGS-PDCoV-N-HA, pCAGGS-G3BP1-Flag, or the pCAGGS vector. Immunoprecipitated proteins were analyzed by Western blot using anti-HA or anti-Flag antibodies, and the results confirmed that the PDCoV N protein interacts with G3BP1 (Figure 4(E)). Furthermore, confocal microscopy demonstrated that HA-tagged N protein colocalizes with endogenous G3BP1 in the cytoplasm of LLC-PK1 cells, further verify their interaction under physiological conditions (Figure 4(F)). The assembly of SGs relies on the mutual binding of translationally arrested mRNAs and RNA-binding proteins, with elevated RNA concentrations further promoting this process. Treatment of cells with RNase revealed that this interaction was diminished in the presence of RNase, indicating that RNA facilitates the N-G3BP1 binding (Figure 4(E)).

Previous studies have demonstrated that G3BP1 overexpression requires a threshold level to induce SGs formation, but this threshold can be reduced upon exogenous stress [23,32]. To determine the minimal dose required for spontaneous SGs formation in LLC‑PK1 cells, we transfected different amounts of pCAGGS‑G3BP1‑Flag plasmid (ranging from 0.5 to 3.0 μg per 1 × 10⁵ cells) and determined SGs formation in the absence of any exogenous stress (Supplementary Fig. S2). We found that G3BP1 expression can be detectable by confocal microscopy at 0.5 μg per 1 × 10⁵ cells, but SGs were not observed until the dose exceeded 2 μg. To investigate whether the interaction between the PDCoV N protein and G3BP1 affects SGs formation, we co-transfected G3BP1-Flag and N-HA expression plasmids into LLC-PK1 cell and assessed SGs formation. The results showed that G3BP1 over expression (3.0 μg per 1 × 10⁵ cells) alone induced robust SGs formation without Ars treatment, whereas co-expression of N protein significantly suppressed this G3BP1-driven SGs assembly (Figure 4(G)). Collectively, these findings demonstrate that the PDCoV N protein interacts with G3BP1, thereby inhibiting SGs formation to facilitate its replication.

G3BP1 promotes RIG-I-mediated antiviral signaling by recruiting it to SGs during PDCoV infection

SGs serve as hubs for antiviral factors, with G3BP1 playing a central role in recruiting these components. To determine whether SGs inhibit PDCoV replication via antiviral immune signaling, we first assessed their effect on IFN-І production. The expression level of G3BP1 was modulated in LLC-PK1 cells through overexpression or knockdown. Subsequently, these cells were stimulated with poly (I:C), and the resulting IFN-І response was quantified using a luciferase reporter assay. The results showed that G3BP1 overexpression significantly enhanced both IFN-β and ISRE promoter activities (Figure 5(A), p = 0.002), whereas G3BP1 knockdown by siRNA suppressed the poly(I:C)-induced activation of IFN-β (p = 0.016) and ISRE promoter (Figure 5(B), p = 0.003), indicating that G3BP1 is essential for IFN-І expression. To identify potential regulatory targets of G3BP1, we examined IFN-β promoter activity in response to key components of the RIG-I-like receptor (RLR) signaling pathway, we assessed IFN-β promoter activity upon stimulation with key pathway components, including RIG-I, mitochondrial antiviral signaling protein (MAVS), TNF-receptor-associated factor 3 (TRAF3) and TANK-binding kinase 1 (TBK1). LLC-PK1 cells were co-transfected individually with pCAGGS-G3BP1 along with RIG-I, MAVS, TBK1, or IRF3 for 48 h, followed by measurement of luciferase activity. G3BP1 overexpression significantly enhanced IFN-β luciferase activity induced by all four components, with the strongest enhancement observed in RIG-I-mediated activation (Figure 5(C); p = 0.005, p = 0.049, p = 0.025, p = 0.035, respectively).

Figure 5.

G3BP1 promotes RIG-I-mediated antiviral signaling by recruiting it to SGs. (A-B) The effects of G3BP1 overexpression (A) and knockdown (B) on the promoter activity of IFN-β and ISRE. LLC-PK1 cells were co-transfected with IFN-β and ISRE reporter constructs along with either siG3BP1 or pCAGGS-G3BP1 for 48 h, after which the cells were collected to measure luciferase activity. (C) The changes of promoter activity of IFN-β stimulated by RIG-I, MAVS, TBK1 and IRF3 and G3BP1 overexpression. RIG-I, MAVS, TBK1, and IRF3 were separately co-transfected with pCAGGS-G3BP1 into LLC-PK1 cells for 48 h, after which the cells were collected to measure luciferase activity. (D) The colocation of G3BP1 and RIG-I, MAVS, TBK1 and IRF3 with PDCoV infection after co-transfecting LLC-PK1 cells with G3BP1 (0.5 μg per 1 × 10⁵ cells) and either RIG-I, MAVS, TBK1 or IRF3 separately for 24 h, the cells were infected with PDCoV (MOI = 0.01) and then IFA assay was performed at 24 hpi. (E) Co-IP analysis of G3BP1 and RIG-I interaction with or without RNase a treatment. Statistical analysis: data are presented as mean ± SD (n ≥ 3) and analyzed by unpaired t-test. For panel A, df = 4. For panel B, df = 4 (left) and df = 8 (right). For protein interactions with G3BP1 (panel C), df = 4.

We next investigated whether G3BP1 enhances RLR signaling via its role in SGs formation. LLC-PK1 cells were co-transfected with low dose of G3BP1 and RIG-I, MAVS, TBK1, or IRF3 plasmid respectively, followed by PDCoV infection. Confocal microscopy revealed that RIG-I, MAVS, TBK1, IRF3, and G3BP1 were diffusely localized in the cytoplasm under control conditions, whereas RIG-I was recruited into G3BP1-positive SGs upon PDCoV infection (Figure 5(D)). Furthermore, Co-immunoprecipitation (Co-IP) assays in HEK-293T cells confirmed that G3BP1 interacts with RIG-I was associated with RNA, as RNase treatment diminished this interaction (Figure 5(E)). Together, these data demonstrate that G3BP1 mediated SGs promote RIG-I signaling activation through recruitment of RIG-I.

PDCoV N protein inhibits the expression of IFN-I by preventing G3BP1-SGs recruiting RIG-I

The PDCoV N protein has been reported to inhibit the RIG- I signaling pathway. Given its ability to suppress SGs formation, we hypothesized that the inhibitory effect of PDCoV N protein on RIG-I signaling was associated with its interaction with G3BP1. To validate this hypothesis, we first evaluated the impact of the PDCoV N protein on IFN-І expression. LLC-PK1 cells were transfected with the PDCoV N plasmid and subsequently stimulated with poly (I:C). Then, the expression levels of IFN-І and its downstream genes were measured by qRT-PCR. The result showed that compared with poly(I:C)-treated cells, PDCoV-N transfection significantly inhibited the mRNA levels of IFN-α, IFN-β, and ISG15 (Figure 6(A); p < 0.001, p = 0.009, p < 0.001, respectively).

Figure 6.

PDCoV N protein inhibits the expression of IFN-I by preventing G3BP1-SGs recruiting RIG-I. (A) The effects of PDCoV N on expression levels of IFN-α, ISG15, and IFN-β with poly(I:C) stimulation. HEK-293T cells were transfected with PDCoV N protein and stimulated with 1.5 µg of poly(I:C) for 24 h. Cells were collected, and total RNA was extracted for RT-qPCR analysis. (B) The effect of PDCoV N on the activation of RIG-I by G3BP1. HEK-293T cells were co-transfected with G3BP1, PDCoV N protein, a dual-luciferase reporter plasmid, and the RIG-I expression vector. Then cells were stimulated with poly(I:C), and promoter activity of IFN-β was detected. (C) PDCoV N protein outcompetes RIG-I for G3BP1 binding in a competitive Co-IP assay. After co-transfecting LLC-PK1 cells with PDCoV N, G3BP1, and RIG-I for 24 h, a Co-IP assay was performed. Western blot was used to detect whether there is a competitive interaction among the three proteins. Statistical analysis: data are presented as mean ± SD (n ≥ 3). Statistical significance was determined using Tukey’s multiple comparison test. For the mRNA expression assays in panel a (df = 6). For the luciferase reporter assay in panel B (df = 16).

Next, we investigated the role of the N protein in the G3BP1-mediated RIG-I signaling pathway by luciferase reporter assay. The results showed that upon poly (I:C) stimulation, G3BP1 remarkably enhanced RIG-I-induced activation of the IFN-β promoter (IFN-β-Luc). In contrast, the PDCoV N protein significantly abrogated this G3BP1-enhanced IFN-β promoter activation (Figure 6(B), p < 0.001). To elucidate the inhibitory mechanism, we performed a competitive Co-IP assay. The results revealed that the presence of PDCoV N protein dose-dependently impaired the association between G3BP1 and RIG-I, confirming that the N protein physically outcompetes RIG-I for G3BP1 binding (Figure 6(C)). Collectively, these findings indicate that the PDCoV N protein disrupts the G3BP1-SGs-RIG-I signaling axis by competitively blocking the interaction between G3BP1 and RIG-I.

Coronaviruses N protein inhibits SGs and RIG-I signaling

The N proteins of coronaviruses share a conserved architecture structure, comprising an N-terminal RNA-binding domain (NTD) and a CTD, which are connected by a flexible Ser/Arg-rich linker region. To explore the structural conservation of coronavirus N proteins, we modeled the 3D structure of the PDCoV N protein using AlphaFold (https://alphafold.com) and compared it with the N protein structures of six other coronaviruses via PyMOL. Structural alignment revealed that the PDCoV N protein exhibits high similarity to the N proteins of other coronaviruses, with particularly notable homology within the CTD (Figure 7(A)). This structural resemblance prompted us to hypothesize that N proteins from other coronaviruses may also possess the ability to suppress SGs formation and RIG-I signaling. To validate this hypothesis, we constructed plasmids expressing N proteins of multiple coronaviruses, including PEDV, TGEV, sparrow deltacoronavirus (SpDCoV), SARS-CoV-2, IBV, and canine coronavirus (CCoV). HEK-293T cells were transfected with plasmid of each coronavirus N, and SGs formation was observed under Ars treatment. The results showed that all of the heterologous N proteins effectively inhibited SGs formation (Figure 7(B), p < 0.01). Furthermore, both IFN-β–Luc (Figure 7, p < 0.001) and ISRE-Luc (Figure 7, p < 0.001) reporter assays demonstrated that these N proteins significantly suppressed IFN-І production by the RIG-I signaling, suggesting that the conservation of these N proteins might contribute to their ability of suppressing SGs formation and RIG-I signaling. Collectively, these findings indicated the N protein-mediated suppression of the SGs–RIG-I signaling axis constitutes a conserved mechanism across the coronavirus family.

Figure 7.

Coronaviruses N protein inhibits SGs and RIG-I signaling. (A) Structural modeling and identification of key amino acid residues in the N-CTD involved in G3BP1 interaction. (B) SGs formation induced by Ars stimulation were inhibited by N proteins of PDCoV, PEDV, TGEV, spCov, SARS-CoV-2, IBV, and CCoV. LLC-PK1 cells were transfected with PDCoV-N, PEDV-N, TGEV-N, spCoV-N, SARS-CoV-2-N, IBV-N, and CCoV-N, then cells were treated with Ars (750 μM for 50 min) at 24 h post transfection. Coronavirus N proteins were detected with anti-HA antibody (red) and SGs were detected with anti-G3BP1 (B₁). Scale bar: 10 μm. The scatter plot shows the percentage of SGs-positive cells in the total transfected cells (B₂). (C-D) The promoter activities of IFN-β and ISRE were inhibited by N protein of different coronaviruses. LLC-PK1 cells were co-transfected with an IFN-β reporter gene alongside expression constructs for PDCoV-N, PEDV-N, TGEV-N, spCoV-N, SARS-CoV-2-N, IBV-N, and CCoV-N. After 24 h, the cells were collected for luciferase activity measurement. Statistical analysis: data are presented as mean ± SD (n ≥ 3). For Figure 7(B)₂, Dunnett’s multiple comparison test was applied (df = 16). For Figure 7(C,D), Tukey’s multiple comparison test was used: Figure 7(C,D) (df = 18).

Discussion

SGs are membraneless cytoplasmic organelles that play a dual role in antiviral defense: SGs sequester the translation machinery to suppress viral protein synthesis or serve as platforms for initiating innate immune responses [22,33]. Among coronaviruses, α-, β-, and γ-CoVs has been reported to employ different strategies to inhibit SGs formation [25,26,34,35]. However, whether δ-CoVs have the ability to modulate SGs formation remains unclear. In this study, we found that PDCoV inhibits SGs formation to enhance its replication. Through systematic screening of PDCoV-encoded proteins, we identified the viral N protein as a key factor in this process, mediated by its interaction with G3BP1. Our data further revealed that G3BP1 recruits RIG-I into SGs to potentiate RIG-I signaling and restrict PDCoV infection. Conversely, the PDCoV N protein competitively blocks the G3BP1-RIG-I interaction, thereby suppressing RIG-I signaling and promoting viral replication. We further identified that N proteins from PEDV, TGEV, SpDCoV, SARS-CoV-2, IBV and CCoV also inhibited SGs formation and RIG-I signaling. Thus, we propose that coronavirus N proteins employ this conserved strategy to counteract SG-mediated RIG-I signaling and facilitate viral replications.

Several viruses have evolved diverse strategies to suppress the SGs formation, thereby facilitating viral replication. (ⅰ) Regulate upstream signaling of eIF2α, such as degrading viral RNA to avoid PKR-eIF2α activation, directly targeting eIF2α kinases to inhibit its phosphorylation, or modulating eIF2α dephosphorylation [36–38]. (ⅱ) Proteolytic cleavage or sequestration of core SGs components. PERK, a type I transmembrane protein, plays a key role in endoplasmic reticulum (ER) stress [39]. Under conditions of accumulated unfolded proteins in the ER, PERK dissociates from the GRP78–PERK complex and undergoes autophosphorylation, leading to its activation [40]. It was reported PDCoV infection can activate PERK pathway and ER stress [41]. In this study, we also observed the activation of PERK-eIF2α pathway both in PDCoV-infected LLC-PK1 cells and intestinal tissues, however, this phenomenon was not accompanied by expected sufficient number of SGs. This indicated PDCoV inhibited SGs formation independently of eIF2α phosphorylation. Notably, PKR, as another major eIF2α kinase activated by viral double-stranded RNA (dsRNA), was not activated during PDCoV infection. Previous studies have reported that SARS-CoV-2 hijacks host ADAR1 to perform A-I editing of viral dsRNA preventing the activation of PKR to inhibit SGs assembly [30]. Thus, how PDCoV blocks PKR activation needs further study. HRI, an eIF2α kinase, is primarily activated by heme deficiency and oxidative stress, not by viral infection, and is predominantly studied in erythrocytes [42,43]. Therefore, we focused on PERK, PKR, and GCN2 in this study. Moreover, besides the N protein, we found that PDCoV nsp15 and NS6 proteins could also suppress SGs formation. It was reported that IBV (γ-CoV) nsp15 suppresses SGs via its endoribonuclease activity [44]. Whether PDCoV nsp15 has a similar function remains unknown. We found that PDCoV NS6 also suppresses SGs formation, a novel observation that warrants further investigation. However, due to technical difficulties in cloning or protein expression, expression plasmids of nsp3 and nsp4 are not obtained, thus, their potential involvement in SGs regulation remains to be determined. In our experiments, we observed PDCoV nsp15 can suppress Ars-induced SGs formation.

As a core of SGs formation, G3BP1 is frequently targeted by viruses to suppress SGs formation [25,29,34,45]. For instance, SARS-CoV-2 employs its N protein to undergo phase separation with G3BP1, thereby disrupting G3BP1 interactions with other RNA-binding proteins and impairing SGs assembly [25,26,46,47]. PEDV exploits host caspase-8 to cleave G3BP1 at residues D168 and D169, thereby inhibiting G3BP1-dependent SGs formation to facilitate viral replication [29]. G3BP1 may also contribute to antiviral defense via SG‑independent mechanisms, as exemplified by its role in promoting cGAS-DNA binding and activation [48]. However, we found that both N protein overexpression and PDCoV infection suppressed G3BP1‑mediated SGs assembly. This effect was related to the competitive disruption of G3BP1 interactions with other core SGs proteins by the N protein, rather than G3BP1 degradation. Our competitive Co-IP data (Figure 6(C)) provide a molecular explanation for the observed reduction in IFN-β signaling. By acting as a molecular decoy, the PDCoV N protein “hijacks” G3BP1, preventing the essential recruitment of RIG-I into SGs. Coronaviral N proteins generally share a conserved architecture: an N-terminal RNA-binding domain involved in genome packaging, a C-terminal homodimerization domain responsible for oligomerization, and multiple intrinsically disordered regions (IDRs). We found that expression of the PDCoV N142-293 truncation in LLC-PK1 cells markedly reduced SGs formation upon Ars stimulation and enhanced viral replication. This observation aligns with reports showing that C-terminally truncated SARS-CoV-2 N protein fails to inhibit poly(I:C)-induced G3BP1-positive SGs formation [49]. Structural comparisons further revealed that N protein is highly conserved across multiple coronaviruses, including SARS-CoV-2, TGEV, PEDV, IBV, SPCoV, CCoV, and PDCoV. Consistent with this structural conservation, the N proteins from all these viruses effectively suppressed SGs formation.

SGs constitute a signaling hub that plays a critical role in type I interferon by recruiting PRRs and host restriction factors through interaction with RNA binding proteins including G3BP1, during viral infection [25,26,50,51]. Previous studies have shown that RIG-I colocalizes with G3BP1-positive SGs and recognizes uncapped viral RNA during Newcastle disease virus replication [52]. Additionally, during Sendai virus or respiratory syncytial virus infection, G3BP1 can recruit growth arrest and DNA damage-inducible beta (Gadd45β) protein to SGs [52,53]. This recruitment triggers a conformational change in G3BP1, enhances its RNA-binding capacity, and thereby promotes SGs assembly, ultimately upregulating type I interferon production. DEAH-Box Helicase 36 (DHX36) has also been recognized as a critical regulator of SGs [54]. Following infection by avian influenza virus or Newcastle disease virus, dsRNA promotes the formation of SGs containing PKR, RIG-I, and DHX36. Within this complex, DHX36 utilizes its ATPase activity to enhance PKR binding to dsRNA and promotes PKR activation, which in turn drives the rapid proliferation of stress granules. This process facilitates the recognition of viral RNA by RLRs. Here, we found that G3BP1 promotes the activation of RIG-I signaling during PDCoV infection. The interaction between G3BP1 and RIG-I was diminished in the presence of RNase, indicating that G3BP1-mediated activation of RIG-I signaling is RNA-associated. However, whether RNA is directly involved in the interactions of G3BP1 with RNA-binding proteins (PDCoV N or RIG‑I) cannot be determined by Co‑IP assay alone. Further biochemical assays, such as GST pull‑down or RNA-pull down assay, will be used to clarify whether these interactions are direct or RNA‑bridged in our future work. These results suggest that RIG-I is recruited into G3BP1-based SGs to activate RIG-I signaling during PDCoV infection. Our findings provide additional evidence that SGs function as an immunoregulatory platform during host defense against pathogenic infection.

Overall, our study delineates a sophisticated molecular mechanism by which PDCoV subverts the host innate immune response (Figure 8). Upon infection, although the host cell initiates stress signals and eIF2α phosphorylation, the PDCoV N protein acts as a critical molecular antagonist at the downstream assembly stage. By interacting with G3BP1, the N protein functions as a molecular decoy that competitively outcompetes RIG-I for binding to the G3BP1 scaffold. This competitive exclusion prevents the recruitment of RIG-I into antiviral SGs, effectively dismantling the G3BP1-RIG-I signaling platform. Consequently, the downstream RLR signaling axis – including the activation of IFN-I and subsequent ISGs – is paralyzed, creating a permissive environment for viral replication. Through structural modeling and cross-virus validation, we demonstrated that this N-mediated SGs suppression mechanism is structurally conserved across multiple coronavirus genera, including PEDV, TGEV, and SARS-CoV-2. This “hijacking” of the SG-mediated immune hub represents a universal and potent evasion strategy, identifying the N protein as a promising target for the development of broad-spectrum antiviral therapeutics.

Figure 8.

Model of PDCoV-mediated disruption of stress granule assembly and type I interferon signaling. porcine deltacoronavirus nucleocapsid protein inhibits RIG-I mediated antiviral signaling by suppressing G3BP1-mediated stress granules formation. The integrating stress response mediated by PERK-eIf2α signaling pathway activates after PDCoV infection. PDCoV N protein can interact with G3BP1 to inhibit the formation of SGs and the recruit of RIG-I by G3BP1, thereby inhibiting IFN-I expression and facilitating PDCoV replication (figdraw).

In conclusion, our findings demonstrate that SGs are indispensable for cellular defense against PDCoV. By identifying the N protein as a central virulence factor that antagonizes the SG-mediated RIG-I pathway, our study provides new insights into coronavirus pathogenesis and offers a structural basis for designing future interventions against emerging coronaviruses.

Funding Statement

This work was supported by grants from the National Natural Science Foundation of China [Grant Nos. 32402923 and 32302910]. This research was also supported by the Program for Innovative Research Team (in Science and Technology) in University of Henan Province [Grant No. 24IRTSTHN034].

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The raw data and supplementary materials supporting the findings of this study are openly available in Figshare under a Creative Commons Attribution 4.0 International (CC BY 4.0) license at https://doi.org/10.6084/m9.figshare.31259368. The deposited datasets include the raw LC-MS/MS-based proteomics data and the datasets used for Protein-Protein Interaction (PPI) network and Gene Ontology (GO) enrichment analyses.

Ethics statement

All animal experiments were performed in compliance with the ARRIVE guidelines and China’s Ministry of Health guidelines for laboratory animal care and use. The study protocol was approved by the Animal Care and Ethics Committee of Henan Agricultural University (Approval No. HNND2024030718). Every effort was made to alleviate animal suffering throughout the study, which was designed and conducted in accordance with the Guide for the Care and Use of Laboratory Animals.

Supplemental data

Supplemental data for this article can be accessed online at https://doi.org/10.1080/21505594.2026.2721825