Inhibition of STING-mediated antiviral innate immunity activation by CD97 via modulation of ER-phagy

Abstract

Endoplasmic reticulum (ER) autophagy (ER-phagy) is a vital homeostatic process triggered by multiple signals and plays a crucial role in regulating innate immunity and viral replication. However, the mechanisms by which host proteins utilize ER-phagy to regulate innate immune response during viral infection remains largely unclear. Here, we uncover the regulatory crosstalk between innate immune adapter, ER retention protein Stimulator of Interferon Genes (STING), and the G protein-coupled receptor ADGRE5/CD97 (Cluster of Differentiation 97). Our results demonstrate that CD97 suppresses the STING-mediated type-I interferon (IFN-I) response against DNA virus and cytosolic DNA, thereby promoting herpes simplex virus type 1 (HSV-1) replication in both cells and mice. CD97 facilitates the recruitment of the ER-phagy receptor, FAM134B (family with sequence similarity 134, member B), to initiate ER-phagy, resulting in the degradation of STING subsequent to DNA virus infection. Furthermore, Cd97-deficient mice exhibit higher IFN-I response and greater resistance to HSV-1 infection. Additionally, our findings reveal that inhibiting CD97 with sanguinarine effectively disrupts HSV-1 replication. These findings shed light on the role of CD97 in the innate immune response against DNA virus infections and offer valuable checkpoint for anti-viral STING activation.

The authors show that the cell-surface membrane receptor CD97 inhibits the DNA-sensing STING signaling pathway in response to HSV-1 infection.

Introduction

The innate immune responses, initiated by pathogen-associated molecular patterns (PAMPs), serves as the first line of defense against invading viruses. During DNA virus infection, viral DNA is recognized by cyclic GMP-AMP synthase (cGAS), which catalyzes the production of the second messenger cyclic [G(2′,5′)pA(3′,5′)p] (2′3′-cGAMP)¹. This messenger activates the downstream adapter protein STING². STING, also known as MITA, is an endoplasmic reticulum protein that play a critical role in orchestrating innate immune signaling against viruses^3,4. Upon sensing cGAMP, STING undergoes a conformational change and forms a butterfly-shaped dimer via its ligand-binding domain^5,6. Activated STING then phosphorylates the adapters TBK1 and IRF3, triggering the production of IFN-I^7,8. These interferons are secreted extracellularly and bind to interferon receptors, leading to the activation of the JAK-STAT signaling pathway. This cascade induces the expression of various interferon-stimulated genes (ISGs), such as ISG56, IFITM3, and HSPE, which collectively enhance the antiviral response⁹.

Given its pivotal role in activating IFN-I responses, the activity of STING must be tightly controlled to prevent excessive immune responses that could lead to autoimmune diseases. A number of negative regulators of STING have been discovered that inhibit its activation, thereby suppressing IFN-I signaling. MYSM1 associates with STING and removes its K63-linked ubiquitination, thereby inhibiting STING signaling¹⁰. The HIV p6 protein suppresses STING-mediated antiviral signaling by disrupting STING’s interaction with AMFR and TRIM32, and by decreasing K27- and K63-linked ubiquitination at the K337 residue of STING¹¹. Additionally, STING has been shown to be degraded through autophagy, acting as a mechanism to downregulate IFN-I signaling¹². The Unc93 homolog B1 (UNC93B1) facilitates the trafficking of STING to the autophagy-lysosome pathway for degradation, thereby attenuating the activity of the cGAS-STING signaling cascade¹³. Recent studies have identified SESN1 as a key negative regulator of STING, facilitating its autophagic degradation by enhancing the interaction between SQSTM1/p62 and STING¹⁴. The newly identified autophagy receptor CCDC50 plays a key role in recognizing ubiquitinated STING and promoting its degradation through an autophagy-dependent mechanism¹⁵. Autophagic degradation of STING is generally mediated by specific autophagy receptors, such as p62 and Tollip^16,17. However, it has not been reported that STING can be degraded through selective autophagy processes (such as ER-phagy).

G protein-coupled receptors (GPCRs) constitute a large and diverse family of membrane proteins that are pivotal in cellular communication and signal transduction. In mammals, the roles of various GPCRs, such as mGluR2, FFAR2, GPR54, and Lgr4, in antiviral innate immune response are well established^18–21. However, the involvement of other GPCRs in innate immunity remains less understood. In this study, we identified CD97 as the negative regulator of STING-mediated IFN-I signaling. Our findings demonstrate that CD97 suppresses IFN-I signaling and enhances HSV-1 replication by triggering ER-phagy through the recruitment of FAM134B. Our study offers insights into the regulation of IFN-I signaling through its interaction with ER-phagy, uncovering a negative feedback mechanism that modulates STING-driven IFN-I activation.

Results

CD97 promotes HSV-1 replication in vitro

Our previous study demonstrated that CD97 inhibits innate immune response against RNA viruses by degrading RIG-I²². However, its functions in the context of the anti-DNA virus innate immune remain ambiguous. In this study, we investigated the involvement of CD97 in regulating HSV-1 replication. Stable overexpression and knockout of CD97 in HeLa cells were generated as described previously²². HeLa cells were infected with HSV-1, and viral replication was assessed by Immunoblotting, RT‒qPCR, and 50% tissue culture infectious doses (TCID₅₀) assays. At a multiplicity of infection (MOI) of 1 and 12 hours post-infection, CD97-overexpressing HeLa cells exhibited markedly higher levels of HSV-1 ICP0 expression, increased viral genome copies, and elevated virus titers compared to cells transfected with an empty vector (EV) (Fig. 1A–C). In contrast, CD97 knockout in HeLa cells significantly reduced HSV-1 replication (Fig. 1D–F). To further confirm that the observed effects were not specific to HeLa cells, we isolated the peritoneal derived macrophages (PMs) from CD97-deficient mice (Cd97^−/−)²³ and wild-type (WT) counterparts (Cd97^+/+) to examine viral replication. As expected, a substantial reduction in HSV-1 replication was observed in Cd97^−/− PMs compared to Cd97^+/+ PMs (Fig. 1G–I). Based on these findings, it can be concluded that CD97 facilitates HSV-1 replication in vitro.

Fig. 1. CD97 enhances the replication of HSV-1 in vitro.

HeLa cells overexpressing CD97 were infected with HSV-1 (MOI = 1) for 12 hours, and the following analyses were performed: (A) Immunoblotting of ICP0, (B) RT‒qPCR of HSV-1 gDNA copy number, (C) TCID₅₀ analysis of virus titers. HeLa cells with CD97 knockout were infected with HSV-1 (MOI = 1) for 12 hours, and the following analyses were performed: (D) Immunoblotting of the ICP0, (E) RT‒qPCR analysis of HSV-1 gDNA copy number, (F) TCID₅₀ analysis of virus titers. Cd97^+/+ and Cd97^-/- PMs were infected with HSV-1 (MOI = 1) for 12 hours, and the following analyses were performed: G RT‒qPCR analysis of HSV-1 UL30, (H) RT‒qPCR analysis of HSV-1 gDNA copy number, (I) TCID₅₀ analysis of virus titers. Data are presented as mean ± SD from n = 3 biologically independent experiments. Statistical significance was determined by two-way ANOVA (B, E, G, and H) and by t test (C, F, and I). Data in (A–I) are representative of three independent experiments. N.D., not detected; *, p < 0.05; **, p < 0.01; ***, p < 0.001.

CD97 suppresses IFN-I signaling during DNA virus infection

Given that viruses trigger innate immune signaling response, we investigated whether CD97 plays a role in regulating anti-DNA viral innate immune response. Overexpression of CD97 in HeLa cells were assessed for its impact on the immune response to HSV-1 infection. The results showed that CD97 attenuated the mRNA levels of IFN-β and IFN-α4, as well as the downstream mRNA of ISG15, ISG54, and ISG56 (Fig. 2A). In contrast, knockout of CD97 enhanced the transcripts of these genes (Fig. 2B). It has been reported that double-stranded DNA poly (deoxyadenylic-deoxythymidylic) [poly(dA:dT)] is a specific ligand for cGAS, leading to activation of STING-mediated IFN-I signaling pathway¹. We then analyzed the transcripts of interferon and ISGs in CD97-overexpressing cells and observed a significant reduction in their mRNA levels upon poly(dA:dT) transfection (Fig. 2C). In contrast, the mRNA levels of these genes were elevated in CD97-knockout cells (Fig. 2D). IFN-I production triggered by cGAS/STING activation requires TBK1 and IRF3 phosphorylation. In CD97-overexpressing HeLa cells, the levels of phosphorylated TBK1 and IRF3 were significantly reduced. In contrast, their phosphorylation was enhanced upon stimulation with HSV-1 or poly(dA:dT) (Fig. 2E–H), indicating that CD97 suppresses TBK1/IRF3 phosphorylation, leading to attenuation of IFN-I signaling. Moreover, IFN-I regulation by CD97 was also assessed using another DNA virus, human adenovirus 4 (HAdV-4), and HT-DNA to mimic cytosolic DNA stimuli. The results demonstrated that CD97 overexpression inhibits HAdV-4- and HT-DNA-induced mRNA expression of IFN-β and ISGs, as well as IRF3 phosphorylation, whereas CD97 knockout/knockdown had the opposite effect (Fig. S1A–H). Together, these findings suggest that CD97 suppresses IFN-I signaling induced by DNA virus or cytosolic DNA.

Fig. 2. CD97 attenuates the IFN-I signaling response induced by HSV-1 or Poly(dA:dT) in HeLa cells.

RT‒qPCR analysis was conducted to examine the mRNA levels of IFN-β, IFN-α4, ISG15, ISG54, and ISG56 in HeLa cells overexpressing CD97 after HSV-1 (MOI = 1) infection for 12 hours (A), HeLa cells with CD97 knockout after HSV-1 (MOI = 1) infection for 12 hours (B), HeLa cells overexpressing CD97 after Poly(dA:dT) (1 μg/mL) transfection for 12 hours (C), and HeLa cells with CD97 knockout after Poly(dA:dT) (1 μg/mL) transfection for 12 hours (D). The indicated proteins were examined by Immunoblotting in HeLa cells overexpressing CD97 after HSV-1 (MOI = 1) infection for 12 hours (E), HeLa cells with CD97 knockout after HSV-1 infection for 12 hours (F), HeLa cells overexpressing CD97 after Poly(dA:dT) transfection for 12 hours (G), and HeLa cells with CD97 knockout after Poly(dA:dT) transfection for 12 hours (H). Data are presented as mean ± SD from n = 3 biologically independent experiments. Statistical significance was determined by two-way ANOVA (A–D). Data in (A–H) are representative of three independent experiments. *, p < 0.05; **, p < 0.01; ***, p < 0.001.

We further evaluated the IFN-I response in primary PMs and found a significant increase in interferon and ISGs mRNA expression levels in Cd97^-/- PMs compared to Cd97^+/+ PMs following stimulation with HSV-1 or poly(dA:dT) (Fig. 3A, B). Consistent with the PM findings, we observed comparable enhancement of interferon and ISGs responses in bone marrow-derived macrophages (BMDMs) from Cd97^-/- mice. (Fig. 3C, D). Of importance, enhanced IFN-β protein levels were found in both Cd97^-/- PMs and BMDMs (Fig. 3E–H), along with elevated phosphorylation of TBK1 and IRF3 in Cd97^-/- PMs (Fig. 3I, J). These findings consistent with the observations made in cell lines. Taken together, these results further demonstrate that CD97 suppresses IFN-I signaling response triggered by DNA viruses in primary macrophages.

Fig. 3. CD97 diminishes the IFN-I signaling response in primary macrophage.

RT‒qPCR analysis was performed to evaluate the mRNA levels of Ifnb1, Ifna4, Isg15, Isg54, and Isg56 in PMs after HSV-1 (MOI = 1) infection for 12 hours (A), PMs following Poly(dA:dT) (1 μg/mL) transfection for 12 hours (B), BMDMs after HSV-1 (MOI = 1) infection for 12 hours (C), and BMDMs following Poly(dA:dT) (1 μg/mL) transfection for 12 hours (D). ELISA analysis was conducted to assess the expression of IFN-β in PMs after HSV-1 (MOI = 1) infection for 12 hours (E), PMs following Poly(dA:dT) (1 μg/mL) transfection for 12 hours (F), BMDMs after HSV-1 (MOI = 1) infection for 12 hours (G), and BMDMs following Poly(dA:dT) (1 μg/mL) transfection for 12 hours (H). Immunoblotting analysis of the indicated proteins in PMs after HSV-1 (MOI = 1) infection for 12 hours (I) and after Poly(dA:dT) (1 μg/mL) transfection for 12 hours (J). Data are presented as mean ± SD from n = 3 biologically independent experiments. Statistical significance was determined by two-way ANOVA (A–H). Data in (A–J) are representative of three independent experiments. *, p < 0.05; **, p < 0.01; ***, p < 0.001, ns, not significant.

CD97 interacts with STING and facilitates its degradation through autophagy

To identify the molecular target of CD97 in IFN-I signaling, we conducted a luciferase reporter assay. The results revealed that CD97 suppressed the activation of luciferase reporters triggered by cGAS/STING (Fig. 4A, B). Through co-immunoprecipitation (Co-IP) experiments using both transfected and native proteins, we consistently detected CD97-STING interaction, while no association was observed between CD97 and cGAS. (Fig. 4C–E). As we previously demonstrated, CD97 does not interact with TBK1 or IRF3²². Based on these results, we propose that CD97 suppresses IFN-I signaling through STING targeting. To validate this hypothesis, the expression of STING was evaluated by immunoblotting during HSV-1 or HAdV-4 infection, or stimulation with poly(dA:dT) or HT-DNA. The results demonstrated that the expression of STING was notably impeded by CD97 in both HeLa cells and PMs (Fig. 4F, G, I, J; Fig. S1G, H), as well as in HAdV-4-infected A549 cells (Fig. S1E, F). Moreover, CD97 blocked STING expression in a dose-dependent manner (Fig. 4H). To verify that CD97 inhibits STING-mediated IFN-I signaling, we employed previously generated STING knockout cells²⁴. Our results demonstrated that in STING-deficient cells, neither CD97 overexpression nor knockdown affected IFN-β and ISG mRNA levels or IRF3 phosphorylation (Fig. S2A–D). Viral titers and Cell viability also remained unchanged upon CD97 manipulation in STING knockout cells (Fig. S2E–H). These results indicate that CD97 specifically targets STING, thereby suppressing IFN-I signaling response.

Fig. 4. CD97 degrades STING via the autophagy pathway.

Dual-luciferase reporter assays were employed to investigate the promoter activity of IFN-β after transfection with pRL-TK, pIFNβ-Luc, and the indicated plasmids (A) or with pRL-TK, pISRE-TA-Gluc, and the indicated plasmids (B). C, D Co-IP analysis showing the interaction between CD97 and STING or cGAS in cotransfected HEK-293T cells. E Immunoprecipitation analysis of the interaction between endogenous CD97 and STING or cGAS in HeLa cells. F, G Immunoblotting analysis was performed to examine the indicated proteins in HeLa cells that overexpress CD97 or have CD97 knocked out, followed by HSV-1 (MOI = 1) infection for 12 hours. H Immunoblotting analysis of STING levels after transfection of different concentrations (0.6, 1.2, or 1.8 μg) of Flag-CD97 or empty plasmid into HeLa cells infected with HSV-1. (I and J) Immunoblotting analysis of indicated proteins in PMs after HSV-1 infection for 12 hours (I) and following Poly(dA:dT) transfection for 12 hours (J). K RT-qPCR analysis of STING mRNA levels in CD97-overexpressing HeLa cells infected with HSV-1. L Immunoblotting analysis of STING in CD97-overexpressing HeLa cells treated with DMSO, MG132 (10 µM), or CQ (20 µM) and infected with HSV-1 for 12 hours. M Immunoblotting analysis of the indicated proteins in CD97-knockout HeLa cells after treatment with cycloheximide (CHX) (100 μg/mL) for the indicated time. Data are presented as mean ± SD from n = 3 biologically independent experiments. Statistical significance was determined by two-way ANOVA (A, B, and K). Data in (A–M) are representative of three independent experiments. **, p < 0.01; ***, p < 0.001, ns, not significant.

Next, we investigated the mechanism through which CD97 regulates STING. Transcript levels of STING were analyzed in HeLa cells overexpressing CD97 after HSV-1 infection, and no significant change in STING mRNA abundance was observed (Fig. 4K). This suggests that CD97 may facilitate the degradation of STING protein levels. Given the essential roles of the proteasome and autophagy-lysosome pathways in cellular protein degradation, we determined the degradation mechanism of STING using specific pathway inhibitors (MG132 for proteasomal inhibition and chloroquine [CQ] for autophagy inhibition). The results showed that CD97-mediated inhibition of STING could be reversed by CQ but not MG132 (Fig. 4L). In addition, CD97-knockout HeLa cells impeded the degradation of STING following cycloheximide (CHX) treatment (Fig. 4M). Taken together, these findings suggest that CD97 promotes the degradation of STING via the autophagy‒lysosome pathway.

CD97 induces autophagic degradation of STING independently its ubiquitination pathway

Given that CD97 promotes STING degradation via the autolysosome pathway, we examined whether CD97 induces autophagy during HSV-1 infection. GFP-LC3B, an autophagic marker, was used to visualize autophagosomes. We observed that CD97 overexpression resulted in a markedly increased number of LC3B puncta compared to the EV group (Fig. 5A). Consistently, HeLa cells overexpressing CD97 exhibited increased LC3B-II expression, whereas this effect was reduced in CD97-knockout HeLa cells (Fig. 5B, C). LC3B, a membrane protein associated with autophagosomes, interacts with target molecules involved in autophagic degradation. Our results demonstrated that CD97 significantly enhanced the interaction between STING and LC3B (Fig. 5D), supporting the notion that CD97 promotes the autophagic degradation of STING.

Fig. 5. CD97 induces autophagic degradation of STING through an independent ubiquitin pathway.

A Immunofluorescence analysis of LC3B levels after cotransfection with the GFP-LC3B plasmid and Flag-CD97 plasmid or control plasmid for 24 hours, followed by HSV-1 infection for 12 hours. Scale bars: 10 µm. B, C Immunoblotting analysis of the indicated proteins in CD97-overexpressing and CD97-knockout HeLa cells infected with HSV-1 (MOI = 1) for 12 hours. D Co-IP analysis of the interaction between CD97, STING, and LC3B by transfecting the indicated plasmids into HeLa cells treated with CQ. E Co-IP and immunoblotting analysis were conducted to assess the ubiquitination of STING in HEK-293T cells co-transfected with the indicated plasmids and treated with CQ. F, G Immunoblotting analysis of indicated proteins in CD97-overexpressing HeLa cells with silenced p62 or NDP52. Data in (A–G) are representative of three independent experiments.

Emerging evidence indicates that autophagy receptor-mediated recognition of ubiquitinated proteins is essential for their selective degradation^14,25. To examine whether CD97-mediated STING degradation depends on its polyubiquitination, we transfected HSV-1-stimulated cells with plasmids encoding ubiquitin, STING, and CD97. Co-IP analysis demonstrated that STING polyubiquitination was unaffected by CD97 overexpression (Fig. 5E). Furthermore, we investigated the autophagy receptor involved CD97-mediated STING degradation. Consistent with previous reports²⁶, p62 knockdown upregulated the expression of STING after HSV-1 infection. However, the CD97-mediated STING inhibition persisted under p62 knockdown conditions (Fig. 5F). Parallel experiments with another selective autophagy receptor, NDP52, demonstrated that NDP52 knockdown similarly failed to prevent CD97-induced STING degradation (Fig. 5G). These findings suggest that CD97 promotes STING degradation via autophagy, independent of both STING ubiquitination and conventional autophagy receptors p62 or NDP52.

CD97 enhances degradation of STING by inducing ER-phagy via FAM134B

STING is an ER-resident protein, prompting us to hypothesize that CD97 induces selective ER-phagy, facilitating the degradation of STING in a manner analogous to the degradation of Mitochondrial Antiviral-Signaling Protein (MAVS) through Mitochondrial autophagy (mitophagy)^27,28. To test this hypothesis, we used the KDEL (RFP⁺GFP^-) marker to assess ER-phagy as previously described²⁵. The results demonstrated that CD97 overexpression promoted KDEL (RFP⁺GFP^-) expression following HSV-1 infection or EBSS treatment (Fig. 6A, with quantification in 6B), suggesting that CD97 enhances ER-phagy. Consistent with this finding, CD97-induced ER-phagy was further confirmed by SEC61B-GFP analysis (Fig. S3A), using a previously reported method²⁹. Whereas there is no interaction between CD97 and SEC61B (Fig. S3B). Based on the previous results, CD97 binds to STING but not to cGAS (Fig. 4C-E), indicating that CD97 specifically degrades STING through ER-phagy. More importantly, we isolated the ER fraction from whole-cell lysates and detected CD97 within the ER. CD97 overexpression was found to downregulate two key ER-phagy regulators - RTN4 and calnexin - providing mechanistic evidence for its role in promoting ER-phagy (Fig. 6C). In this result, we also observed a reduction in the expression of STING in the ER following CD97 overexpression (Fig. 6C). Moreover, in STING-deficient cells, although CD97 reduced expression of the ER-phagy marker Rtn4, it did not affect IFN-I signaling and viral replication (Fig. S2A–H). These findings altogether support the conclusion that CD97 induces ER-phagy to specifically facilitate the degradation of STING.

Fig. 6. CD97 enhances degradation of STING by inducing ER-phagy via FAM134B.

A HeLa cells stably expressing ssRFP-GFP-KDEL were transfected with either an EV or Flag-CD97. 24 hours post-transfection, the cells were infected with HSV-1 for 12 hours and then fixed for fluorescence detection. Scale bar: 10 μm. B The number of KDEL puncta per cell that red (RFP⁺GFP^-) was quantified from (A). Cell counts (n = 20 cells) from three independent experiments were analyzed by two-way ANOVA. *, p < 0.05; ***, p < 0.001. C ER-enriched analysis and immunoblotting analysis of the indicated proteins. D Co-IP analysis showing the interaction between STING and the ER-phagy receptor after 36 hours of transfection with the indicated plasmids. E, F Co-IP analysis of the interaction between FAM134B or TEX264 and STING during CD97 overexpression. G, H Immunoblotting analysis of STING in FAM134B or TEX264 knockout HeLa cells overexpressing CD97. I Co-IP analysis showing the interaction between CD97 and FAM134B after 36 hours of transfection with their plasmids. Data in (A–H) are representative of three independent experiments.

To identify the ER-phagy receptor involved in STING degradation, we examined the interaction between STING and previously reported ER-phagy receptors³⁰. The results showed that only FAM134B and TEX264 interacted with STING (Fig. 6D), suggesting that these two receptors may be involved in the STING degradation. To investigate which receptor mediates the CD97-induced degradation of STING, we examined the interaction between FAM134B, TEX264, and STING during CD97 overexpression. The results showed that CD97 enhanced the binding of FAM134B and STING, but had no effect on the interaction between TEX264 and STING (Fig. 6E, F). These data suggest that FAM134B may be involved in the degradation of STING by CD97. To verify this hypothesis, we generated FAM134B and TEX264 knockout cell lines. As expected, CD97 lost its ability to inhibit STING in FAM134B knockout cells, while it still suppressed the expression of STING in TEX264 knockout cells (Fig. 6G, H). These results indicate that CD97 promotes STING degradation through FAM134B. In Fig. 6G, we observed that knockout of FAM134B alone had no effect on STING expression. We therefore evaluated IFN-I signaling and viral replication by manipulating FAM134B. The results showed that neither overexpression nor knockout of FAM134B affected HSV-1 replication or HSV-1-induced IFN-β mRNA levels (Fig. S4A–D).

To investigate how CD97 regulates FAM134B-ER-phagy-STING signaling axis to respond IFN-I signaling, we first assessed the potential interactions between these molecules. Our co-immunoprecipitation assays revealed an interaction between CD97 and FAM134B (Fig. 6I). Given our previous finding that CD97 also interacts with STING (Fig. 4C, E), we hypothesized that these three proteins might form a functional complex. Consistent with this hypothesis, confocal microscopy demonstrated clear colocalization of CD97, STING, and FAM134B (Fig. S5A–C), suggesting the assembly of a ternary complex. Furthermore, to determine whether this complex is functionally relevant in antiviral defense, we evaluated these interactions under physiological stimulation by HSV-1 infection. Notably, HSV-1 infection markedly enhanced both the protein-protein interactions and their colocalization (Fig. S5A–C), indicating that the formation of this ternary complex is a regulated process in response to viral challenge.

To explore the interaction domains among them, we generated STING truncation mutants by deleting different domains (Fig. S5D). Co-IP analysis revealed that the CBD domain of STING is necessary for its interaction with FAM134B (Fig. S5E). Consistently, in STING knockout cells rescued with STING truncation mutants, CD97 was unable to degrade STING mutants lacking the CBD domain, while the degradation of other truncation mutants remained effective (Fig. S5F). This evidence confirms that the STING-FAM134B interaction, mediated by the CBD, is a prerequisite for CD97-mediated STING degradation. We next generated FAM134B truncation mutants to see which part connects to STING (Fig. S5G). Co-IP assays identified the amino acids 1-83 as the critical domain responsible for binding STING (Fig. S5H). The direct interaction was confirmed using purified proteins in a GST pulldown assay (Fig. S5I). We further determined that CD97 interacts with FAM134B via its C-terminus (Fig. S6A), while FAM134B binds CD97 through its amino acids 1-83 region (Fig. S6B). A direct interaction between the two proteins was established by GST pulldown assay (Fig. S6C).

Collectively, these data demonstrate that CD97 promotes STING degradation through FAM134B-dependent ER-phagy, revealing a regulatory mechanism controlling STING-mediated immune responses.

CD97-deficient mice are more resistant to HSV-1 infection

To investigate the physiological significance of CD97 during DNA virus infection in vivo, we subjected both Cd97^+/+ and Cd97^−/− littermate mice to HSV-1 challenge. Mice were intraperitoneally injected with 1 × 10⁸ PFU of HSV-1 analyzed 24 hours post-infection. The spleen, lung, and liver were isolated for assessment of IFN-I response and viral replication. Compared to Cd97^+/+ mice, Cd97^−/− mice exhibited a significantly increased mRNA levels of Ifnb1 and Ifna4, accompanied by a concurrent decrease in HSV-1 replication, as evidenced by the decreased HSV-1 UL31 gene transcripts (Fig. 7A). Additionally, histological examination using hematoxylin and eosin (H&E) staining revealed reduced thickening of alveolar walls in Cd97^−/− mice compared to Cd97^+/+ mice (Fig. 7B). However, no mortality was observed in either group following intraperitoneal injection of HSV-1. To evaluate the impact of CD97 on survival following HSV-1 infection, mice were subjected to intracranial injection with 1 × 10⁷ PFU/mouse. As expected, Cd97^+/+ mice exhibited rapid mortality from day 2 to day 5 post-infection, whereas Cd97^-/- mice showed a moderate reduction in mortality, with three mice surviving the infection (Fig. 7C). This indicates that Cd97^-/- mice exhibit have increased resistance to HSV-1 infection. Consistent with this, a notable reduction in HSV-1 copies was observed in the brain, spleen, lung, and liver of Cd97^−/− mice compared to Cd97^+/+ mice following intracranial HSV-1 infection (Fig. 7D). These data demonstrate that Cd97^-/- mice exhibit increased resistance to HSV-1 infection, confirming the crucial role of CD97 in regulating innate antiviral immunity against HSV-1.

Fig. 7. CD97 knockout weakens viral replication by increasing IFN-I production in vivo.

A RT‒qPCR analysis of Ifnb1, Ifna4, and HSV-1-UL30 in spleen, lung, and liver of Cd97^+/+ and Cd97^-/- mice administered 1 × 10⁸ PFU/mouse of HSV-1 via intraperitoneal injection for 24 hours. B H&E staining of lung sections of Cd97^+/+ and Cd97^-/- mice subjected to 1 × 10⁸ PFU/mouse of HSV-1 via intraperitoneal injection after 3 days. Scale bars: 50 µm. C Survival curve analysis of Cd97^+/+ and Cd97^-/- mice administered 1 × 10⁷ PFU/mouse of HSV-1 via Intracranial injection for 7 days. Kaplan-Meier survival analysis was performed for age- and sex-matched littermates, with statistical comparison by a two-sided log-rank test (n = 8 per group). *p < 0.05. D RT‒qPCR analysis of HSV-1 copy number in brain, spleen, lung, and liver of Cd97^+/+ and Cd97^-/- mice administered 1×10⁷ PFU/mouse of HSV-1 via Intracranial injection for 1 days. Data are presented as mean ± SD from n = 3 for (A), n = 5 for (D) biologically independent experiments. Statistical significance was determined by two-way ANOVA (A) and by t test (D). Data in (A, B) are representative of three independent experiments. *, p < 0.05; ***, p < 0.001.

To investigate how CD97 deficiency contributes to the ER-phagy-STING signaling pathway in antiviral responses, we evaluated STING and ER-phagy marker expression following HSV-1 infection across different tissues. The results showed that STING and Rtn4 expression levels in the brains of Cd97^-/- mice were significantly higher than in Cd97^+/+ mice after intracranial HSV-1 injection (Fig. S7A). We observed consistent results in spleen, lung, liver, and Peripheral Blood Mononuclear Cells (PBMCs) following intraperitoneal HSV-1 injection (Fig. 7B–E). These findings indicate that CD97 deficiency prevents STING degradation, maintaining a heightened innate immune state. Furthermore, we found that CD97 mRNA expression was upregulated by HSV-1 infection in brain, spleen, lung, and liver tissues (Fig. 7F). Cell-type-specific analysis of splenic populations²³ revealed selective CD97 upregulation in macrophages, while its expression in T and B lymphocyte populations showed no significant changes post-HSV-1 infection (Fig. S8). Taken together, these results demonstrate that CD97 inhibits STING expression through ER-phagy to negatively regulate IFN-I responses in vivo.

Sanguinarine impairs HSV-1 replication by inhibiting CD97

We previously reported that the small-molecule drug sanguinarine inhibits CD97 expression to suppress the replication of RNA viruses²². However, its impact on HSV-1 replication remains unclear. To investigate the effect of sanguinarine on HSV-1 replication, we treated HSV-1-infected HeLa cells with sanguinarine. The results demonstrated that sanguinarine inhibited CD97 expression, which in turn reduced HSV-1 ICP0 expression, as well as promoted STING and Rtn4 expression (Fig. 8A). Consistent with these findings, sanguinarine suppressed the expression of HSV-1 UL30 mRNA, viral copy numbers, and viral titers (Fig. 8B–D). Importantly, sanguinarine had no effect on viral titer and ICP0 expression in CD97-knockout cells infected with HSV-1 (Fig. 8D, E). These results indicate that sanguinarine inhibits HSV-1 replication by suppressing CD97 in vitro. To assess the efficacy of sanguinarine in reducing virus replication in vivo, we performed an intragastric experiment where sanguinarine was administered following HSV-1 infection. The administration of sanguinarine resulted in reduced lung damage, as evidenced by H&E staining (Fig. 8F). Moreover, sanguinarine treatment significantly improved the survival rate compared to the NC group (Fig. 8G). However, as expected, sanguinarine did not improve survival Cd97^-/- mice infected with HSV-1 (Fig. 8H). These results conclusively demonstrate that the inhibition of CD97 by sanguinarine effectively hampers the replication of HSV-1 both in vitro and in vivo.

Fig. 8. Sanguinarine attenuates viral replication by inhibiting CD97 in vitro and in vivo.

A Immunoblotting analysis of indicated proteins challenged with DMSO (NC) and sanguinarine (1 μM) after HSV-1 infection for 12 hours. B, C RT-qPCR analysis of HSV-1-UL30 mRNA and gDNA copy number treated with DMSO and sanguinarine (1 μM) after HSV-1 or mock infection for 12 hours. D Sanguinarine treatment in WT or CD97-knockout HeLa cells, and rescue of CD97 in CD97-KO cells, followed by TCID₅₀ analysis of virus titers. E Immunoblotting analysis of ICP0 expression under the same conditions as in (D). F H&E staining of lung sections from WT mice infected with HSV-1 via intraperitoneal injection and treated with intragastric sanguinarine. Scale bars: 50 µm. G, H Survival curve analysis of Cd97^+/+ and Cd97^-/- mice administered 1 × 10⁸ PFU/mouse of HSV-1 via Intracranial injection and intragastric administration of sanguinarine (30 mg/kg) for 7 days. Kaplan–Meier survival analysis was performed for age- and sex-matched littermates, with statistical comparison by a two-sided log-rank test (n = 8 per group). *p < 0.05; ns, not significant. Data are presented as mean ± SD from n = 3 biologically independent experiments. Statistical significance was determined by two-way ANOVA (B–D). Data in (A–F) are representative of three independent experiments. **, p < 0.01; ***, p < 0.001, ns, not significant; N.D. Not Detected.

Discussion

In this study, we elucidate the roles of CD97 in suppressing STING-mediated innate immune response to HSV-1 and pathogenic DNAs. Our findings highlight the important role of CD97 in promoting HSV-1 replication both in vitro and in vivo. We also identify FAM134B as the receptor responsible for recognizing STING and facilitating its transfer for ER-phagic degradation during CD97 overexpression. This process serves to restrict STING signaling and subsequent IFN-I expression, potentially forming an immune evasion mechanism during viral infections.

CD97, a member of the ADGRE family, is expressed in various cell types and has been implicated in multiple pathophysiological processes through interactions with cellular proteins. For instance, in hepatocellular carcinoma, CD97 collaborates with GPCR kinase 6 to stimulate the secretion of matrix metalloproteinases 2 and 9, promoting HCC metastasis³¹. In human fibrosarcoma cells, the RGD motif of CD97 enhances cell adhesion by upregulating αvβ5 and α2β1 integrins, while its anti-apoptotic effects accelerate tumorigenesis³². Our previous studies have revealed that CD97 promotes RNA virus replication by negatively regulating RIG-I-mediated innate immunity²². In the present study, we further explore the crucial role of CD97 in negatively regulating the IFN-I signaling response against HSV-1 infection. We observed that CD97 suppresses the expression of IFN-I upon HSV-1 or HAdV-4 infection and inhibits the IFN-I response triggered by poly(dA:dT) or HT-DNA. Combining previous findings²², we demonstrate that CD97 can modulate both RNA and DNA sensing pathways, indicating it may function as a multifunctional immune regulator rather than being restricted to a single PRR pathway. These results suggest CD97 acts as a pleiotropic immune modulator that employs distinct mechanisms to regulate different PRRs, potentially informing future therapeutic approaches for viral immune evasion.

Given cGAS/STING-mediated IFN-I signaling plays central role in antiviral immunity, it is essential to tightly regulate the activity of STING to prevent an excessive immune response that may cause harm. Previous studies have identified several mechanisms through which STING activity is regulated, including ubiquitination and deubiquitination. The E3 ligases RNF115 initiates STING ubiquitination to stabilize its expression, while the deubiquitinase OTUD5 cleaves K48-linked polyubiquitin chains on STING, preventing its degradation^33,34. Additionally, the phosphatase PPM1G dephosphorylates p-STING, leading to immune suppression³⁵. Autophagy is another key regulatory mechanism of STING activity. UNC93B1 directs STING towards autophagy-lysosome degradation, reducing cGAS/STING signaling¹³. Similarly, the autophagy receptor CCDC50 also influences IFN-I response by promoting the autophagic degradation of STING during viral infections and autoimmune diseases¹⁵. However, the role of selective autophagy in regulating STING, such as through ER-phagy, remains poorly understood. Studies have shown that innate immune adapters, like RIG-I, MDA5, and MAVS, are targeted to mitophagy, which suppresses IFN-I signaling^36,37. highlighting the crucial role of selective autophagy in regulating innate immunity. In this study, we demonstrate that CD97 promotes ER-phagy to attenuate STING expression and suppresses the IFN-I response. Our findings extend the role of CD97 in immune regulation and contributed to a better understanding of how CD97 restricts STING through ER-phagy.

To understand the molecular mechanisms by which CD97 promotes ER-phagic degradation of STING, we show that FAM134B, a key autophagy receptor, is critically required for the selective autophagic degradation of STING. FAM134B has been reported as a key autophagy receptor that mediates ER-phagy³⁸. To determine the ER-phagy receptor through which CD97 induces the degradation of STING, we examined the interaction between STING and several known ER-phagy receptor, including RTN3, ATL3, SEC62, CCPG1, FAM134B, and TEX264^38–43. Our results revealed that only FAM134B and TEX264 interact with STING. Moreover, CD97 fails to degrade STING only in the absence of FAM134B, suggesting that CD97 induces ER-phagy to degrade STING specifically via FAM134B. Previous studies have reported that FAM134B can be hijacked by SARS-CoV-2/coronaviruses to influence viral replication through ER-phagy^44,45. However, our results demonstrate that FAM134B has no effect on either HSV-1 replication or the HSV-1-induced IFN-I pathway. Although CD97 affects ER-phagy in STING knockout cells, it does not influence IFN-I signaling under these conditions. We further observed that the CBD domain of STING is responsible for binding FAM134B, and CD97 loses its ability to regulate STING when the CBD domain is deleted. These findings suggest that CD97 recruits FAM134B to induce ER-phagy-mediated degradation of STING, thereby negatively regulating the IFN-I signaling pathway. These findings provide valuable insights into the regulatory mechanisms of IFN-I signaling through crosstalk with ER-phagy. Additionally, a question is whether CD97 can mediate antiviral responses independently of its ligand and shear stress. Previous studies^46,47 have established that CD97’s functions are mechanosensitive, relying on either shear force-dependent binding to CD55 or force-induced phosphorylation of its PBM. To investigate this, we first treated cells with recombinant CD55 protein and found that it had no effect on viral replication (Fig. S9A). We then examined viral replication in the context of overexpressing CD97 deletion mutants. Both the CD97-Cterm and PBM deletion mutant (CD97-delPBM) enhanced viral replication, whereas the CD97-Nterm did not (Fig. S9B). Furthermore, we found that CD97-delPBM still interacted with FAM134B (Fig. S9C), suggesting that the CD97-FAM134B interaction is independent of shear force. These findings demonstrate that CD97 mediates antiviral responses in a manner independent from both ligand binding and shear stress.

In CD97-deficient mice, we observed enhanced innate immune responses following both intracranial⁴⁸ and intraperitoneal⁴⁹ HSV-1 infection. CD97 mRNA levels were upregulated upon HSV-1 infection across multiple tissues. We propose the existence of an uncharacterized mechanism that enhances CD97 expression to negatively regulate innate immunity during viral infection. Increased CD97 recruits FAM134B to induce ER-phagy-mediated STING degradation, as our results demonstrated that HSV-1 infection promotes CD97-FAM134B-STING interaction and colocalization. Furthermore, HSV-1 infection specifically increases CD97 expression in macrophages, confirming CD97’s role in innate immunity. However, the correlation between CD97 and STING expression levels across different cell types requires further investigation.

Increasing studies are focused on development of agonists or inhibitors targeting functional molecules based on theoretical findings. For example, the activation of GPER1 signaling by its agonist G1 could be proposed as a potential strategy to protect fetal health⁵⁰. The administration of the N-terminal 18-amino acid peptide of STING shows to enhance host outcomes in disseminated fungal infections⁵¹. In this study, our data demonstrate remarkable phenotypic concordance between sanguinarine treatment and CD97 knockout across multiple endpoints: similar reductions in HSV-1 viral titers, comparable enhancement of STING pathway activation, and identical ER-phagy. This multi-parameter consistency strongly suggests that CD97 inhibition represents the predominant mechanism underlying the observed suppression of innate immune responses. Through CD97 knockout and rescue experiments, we specifically validated sanguinarine’s action on CD97. Based on these findings, therapeutic inhibition of CD97 signaling may represent a promising approach for combating viral infections. However, while sanguinarine exhibits CD97-inhibitory activity, its incomplete protein suppression and potential off-target effects need to be considered and investigated.

In summary, our study identifies CD97 as a critical negative regulator of IFN-I activation during HSV-1 infection, facilitating viral replication in both in vitro and in vivo models. Mechanistically, CD97 suppresses STING-dependent IFN-I signaling by promoting FAM134B-mediated ER-phagic degradation of STING, revealing a regulatory pathway in innate immunity. Our study demonstrates that sanguinarine exhibits therapeutic potential by specifically inhibiting CD97, resulting in suppressed viral replication (Fig. 9). These findings establish CD97 as a promising therapeutic target for viral infections.

Fig. 9. The model depicting the mechanism of which CD97 enhances viral replication.

cGAS identifies viral DNA, triggering cGAMP production, which initiates STING translocation from ER to Golgi apparatus and the recruitment of TBK1 to phosphorylate IRF3. Consequently, this leads to the production of IFN-I, which inhibits viral replication. In this mechanism, CD97 recruits FAM134B to promote STING degradation via ER-phagy, thereby inhibiting IFN-I signaling pathway. Sanguinarine suppresses CD97 to inhibit HSV-1 replication. The illustration was created in BioRender. Chang, H. (2025) https://BioRender.com/yvn44v7.

Materials and methods

Cell lines, mice and viruses

The cell lines utilized in this study were as follows: HeLa (ATCC, CCL-2), HEK293T (ATCC, CRL-11268), and A549 (ATCC, CCL-185). The cells were cultured in a humidified 5% CO₂ incubator at 37°C using Dulbecco’s modified Eagle’s medium (DMEM; Shandong Sparkjade Biotechnology, CA0004-500ML). The medium was supplemented with 10% heat-inactivated fetal bovine serum (FBS; ExCell Bio, FSP500) and 100 μg/mL penicillin-streptomycin solution (Beijing Labgic Technology, Bl505A).

B6;129P2-Adgre5^tm1Dgen/J (Cd97^-/-) mice used in this study were obtained from the Jackson Laboratory (stock code: 005788). The genotype was identified using the JAX protocol (28954). All mouse experiments were conducted in accordance with protocols approved by the Institutional Animal Care and Use Committee of Shandong Normal University. Euthanasia was performed by placing mice in a sealed chamber and introducing CO₂ at a fill rate of 30–70% of the chamber volume per minute. Following this procedure, death was verified through cervical dislocation. Once confirmed, each carcass was bagged and transferred to a freezer for storage. Euthanasia was also carried out in cases where mice exhibited a body weight loss of 20% or more.

HSV-1 strains were kindly gifted from Pro. Y. Gao (Changchun Veterinary Research Institute). HAdV-4 strains were kindly gifted by Pro. J. Zhao (Henan Agricultural University).

Antibodies and chemicals

Rabbit anti-TBK1 antibody (clone D1B4; 3504, 1:1000 IB), Rabbit anti-pTBK1 antibody (Ser172 clone D52C2; 5483, 1:1000 IB), Rabbit anti-HA-Tag antibody (clone C29F4; 3724, 1:1000 IB, 1:200 IF), and Mouse anti-DYKDDDDK Tag antibody (clone 9A3; 8146, 1:1000 IB, 1:200 IF) were purchased from Cell Signaling Technology. Rabbit anti-β-actin antibody (AB0035, 1:1000 IB), Rabbit anti-IRF3 antibody (CY5779, 1:1000 IB), Rabbit anti-p-IRF3 antibody (CY6575, 1:1000 IB), Rabbit anti-GFP-Tag antibody (AB0045, 1:1000 IB), Rabbit anti-STING antibody (CY7204, 1:1000 IB), Rabbit anti-LC3B antibody (CY5992, 1:1000 IB), Rabbit anti-P62 antibody (CY9081, 1:1000 IB), Rabbit anti-Calnexin antibody (CY5839, 1:1000 IB), and Rabbit anti-Rtn4 antibody (CY7213, 1:1000 IB) were purchased from Abways Biotechnology Co.,Ltd. Rabbit anti-NDP52 antibody (A24021, 1:1000 IB) was purchased from ABclonal Biotech Co., Ltd. Rabbit anti-CD97 antibody (clone ERP4427; ab108368, 1:1000 IB) was purchased from Abcam. Mouse anti-cGAS antibody (clone D-9; sc-515777, 1:1000 IB), Mouse anti-ICP0 antibody (sc-53070, 1:1000 IB), and Mouse anti-TEX264 antibody (sc-100944, 1:1000 IB) were purchased from Santa Cruz Biotechnology. Rabbit anti-FAM134B antibody (21537-1-AP, 1:1000 IB) was purchased from Proteintech Group, Lnc. AffiniPure Goat Anti-Mouse IgG (H + L) (115-035-003, 1:5000 IB), and AffiniPure Goat Anti-Rabbit IgG (H + L) (111-035-003, 1:1000 IB) were purchased from Jackson Immuno Research Labs. Goat anti-Mouse IgG Highly Cross-Adsorbed Secondary Antibody (A32728, 1:200 IF) and Goat anti-Rabbit IgG Cross-Adsorbed Secondary Antibody (T-2726, 1:200 IF) were purchased from Thermo Fisher Scientific. Poly(dA:dT) (Tlrl-patn) were purchased from InvivoGen. CHX (HY-12320), MG132 (HY-13259), CQ (HY-17589A), and HT-DNA (D6898) were purchased from MedChemExpress. M-CSF (CB34) was purchased from Suzhou Novoprotein Scientific. Hematoxylin (C0107) and Eosin (C0109) were purchased from Beyotime Institute of Biotechnology. PBS (CR0014-500Ml) was purchased from Shandong Sparkjade Biotechnology. RIPA lysis buffer (WB3100) was purchased from NCM Biotech. The recombinant protein CD55 (2009-CD) was purchased from R&D Systems.

Plasmid and transfection

The CD97, cGAS, STING, FAM134B, TEX264, RTN3, ATL3, SEC62, and CCPG1 complementary DNA (cDNA) were amplified using a genomic cDNA generated from reverse transcription of total RNA extracted from HeLa cells. CD97 was cloned into the HA-, Flag-, or GFP-tagged vectors using the pLVX-IRES-puro vector and the PEGFP-N1 vector, respectively. cGAS STING, FAM134B, TEX264, RTN3, ATL3, SEC62, and CCPG1 were cloned into the PcDNA3.1 vector. The plasmids Pcw57-CMV-ssRFP-GFP-KDEL (P31327) was acquired from MiaoLing Biology.

Viral infection in vitro and in vivo

To assess the expression of genes and signaling molecules, cells were infected with HSV-1. Following the established protocol⁵². The virus was incubated in serum-free medium for one hour, after which the medium was replaced with a medium containing 2% fetal bovine serum (FBS). At specific time points, the total cells were collected for viral titration, RT‒qPCR, Immunoblotting, or immunofluorescence assays. For the in vivo infection, six- to eight-week-old mice were subjected to intraperitoneal or intracranial injections of HSV-1 following established protocols⁵³. Mouse survival was monitored for a period of 7 days after intracranial injection of HSV-1 at a dose of 1 × 10⁸ PFU/mouse. Viral replication and pathology were assessed by administering HSV-1 intraperitoneal injections at a dose of 1 × 10⁷ PFU/mouse, and after 24 h, RT-qPCR was conducted to quantify the expression of Ifnb1, Ifna4, and HSV-1-UL30 in different organs.

Macrophage cell culture and preparation

To isolate peritoneal-derived macrophages, peritoneal lavage was obtained from euthanized mice. The peritoneal macrophages were isolated and the sample volume was adjusted to 10 mL with medium. Next, 2 × 10⁶ cells were plated in 6-well culture plates and cultured in RPMI-1640 medium supplemented with 10% FBS. As described earlier, nonadherent cells were removed by thoroughly washing with RPMI-1640 medium⁵⁴. Macrophages were isolated from mouse bone marrow and cultured to differentiate into BMDMs. Briefly, femurs or tibias were harvested, and the bone marrow was flushed with sterile PBS. The isolated cells were cultured in RPMI-1640 medium supplemented with 10% FBS and 25 ng/mL macrophage colony-stimulating factor (M-CSF) to promote differentiation. The cultures were maintained at 37 °C in a 5% CO₂ incubator for 7–10 days, with medium changes every 2–3 days to ensure optimal macrophage differentiation. Differentiated macrophages were then harvested and used for subsequent experiments.

RT‒qPCR and siRNA silencing

siRNA was transfected into HeLa cells at a final concentration of 30pM using the Attractene transfection reagent (301007; QIAGEN). Each siRNA sequence was prepared by Beijing Tsingke Biotechnology Co., Ltd. The specific siRNA sequences can be found in Table S1. Following the methodology described by Hou et al.⁵⁵. RT‒qPCR was employed to measure mRNA levels. Briefly, following virus challenge, total RNA was obtained using the Total RNA Extraction Kit (AC0205-B, Shandong Sparkjade Biotechnology Co., Ltd). Reverse transcription was carried out using the RT Mix kit with gDNA Clean for qPCR Ver.2 (AG11728, Accurate Biology) to generate complementary DNA (cDNA) for subsequent RT-qPCR analysis. The relative fold change in mRNA expression was quantified using the ChamQ Universal SYBR qPCR Master Mix (Q711, Vazyme Biotech) and calculated utilizing the 2^-ΔΔCT method. Primer sequences for RT-qPCR were designed using Primer Premier 6.0 software and can be found in Table S1.

Co-Immunoprecipitation and Immunoblotting

HEK-293T cells were transfected with the specified plasmids and incubated for 36 hours. After washing with PBS, the cells were suspended in a RIPA lysis buffer (New Cell & Molecular Biotech, WB3100) and kept on ice for 20 minutes. Immunoprecipitation was carried out using either anti-DDDDK-tag mAb-magnetic beads (M185-11R, MBL Life science) or anti-HA-tag mAb-magnetic beads (M132-11, MBL Life science.) at 4 °C for 2 hours. The protein expressed in Escherichia coli was purified by ultrasonication and lysis, followed by enrichment with GST beads. The bead-bound proteins were then incubated with a lysate from HEK-293T cells for 2 hours at 4°C. For endogenous proteins immunoprecipitation, magnetic beads coupled with commercially available antibodies were used. Subsequently, the beads were rinsed three times with lysis buffer. The immunoprecipitates were eluted with SDS loading buffer (Beyotime, P0015) and then denatured for 10 minutes at 95 °C. To block nonspecific binding, the blots were incubated using Tris-buffered saline-Tween (TBST) containing 5% nonfat dry milk. Primary antibodies specific to the proteins of interest were applied, following which the membranes were washed. Enhanced chemiluminescence reagents (ED0015-C; Shandong Sparkjade Biotechnology) were utilized as per the manufacturer’s instructions to visualize the membranes⁵⁶.

Dual-luciferase reporter assay

In 24-well plates, HEK-293T cells (2 × 10⁵) were seeded and subsequently transfected with firefly luciferase (100 ng) and pRL-TK (a Renilla luciferase plasmid; 10 ng), along with different concentrations of either control or protein-expressing plasmids. The dual-luciferase reporter assay system kit (DD1205-01; Vazyme Biotech) and a SpectraMax M5 microplate reader were used to measure the firefly and Renilla luciferase activities.

Immunofluorescence assay

Following the established protocol⁵⁷, transfected cells were grown on coverslips and subsequently fixed with 4% paraformaldehyde after two washes with ice-cold PBS. After fixation, the cells were permeabilized using PBS containing 0.2% Triton X-100, followed by three washes with PBS. The cells were visualized using a Leica SP8 confocal microscope.

Enzyme-linked immunosorbent assay (ELISA)

The production of IFN-β in culture medium supernatants was measured using the Mouse IFN-β ELISA Kit (EK2236, MultiSciences Biotech Co., Ltd.) following the manufacturer’s instructions. Each well was treated with washing liquor for 30 seconds. Then, sample, control, or standard substances were added to each well along with diluted antibody. The cells were placed at room temperature for 2 hours. The medium in each well was aspirated and the wells were washed, with this process repeated five times for a total of six washes. Subsequently, horseradish peroxidase-labeled streptavidin frontalis was added to each well and incubated for 45 minutes. After an additional round of washing, substrate solution was added to each well. The optical density of each well was measured within 30 minutes utilizing a microplate reader set to 450 nm. Additionally, the absorbance at a wavelength of 570 nm was recorded to correct for any optical imperfections in the plate.

ER enrichment assay

ER enrichment was performed using the Minute™ ER Enrichment Kit (ER-036, Invent Biotechnologies, Inc.), following the manufacturer’s instructions. The collected cells were frozen for 10 minutes, then resuspended in buffer to precipitate the cells. The cell suspension was added to a specific centrifuge column, and the nucleus and debris were removed by high-speed centrifugation. The supernatant was then added to the ER extraction buffer and centrifuged multiple times to isolate the endoplasmic reticulum. Finally, protein lysate was added for cleavage, followed by western blot experiments.

Flow cytometry

Splenic single-cell suspensions were prepared by mechanically dissociating tissues through a 70 μm nylon mesh. Following erythrocyte depletion using a commercial lysis buffer, aliquots of 5 × 10⁵ cells were allocated for each immunostaining procedure. To minimize nonspecific antibody binding, samples were pre-incubated with 10% normal mouse serum prior to incubation with titrated monoclonal antibodies (ABflo^®594 Rabbit anti-Mouse CD97 mAb, A23823, 1:200 FC, ABclonal; ABflo^®594 Rabbit IgG isotype control, A23821, 1:200 FC, ABclonal; Anti-Mo F4/80, 17-4801-82, 1:200 FC, Invitrogen; PE-Cy^TM7 Anti-mouse CD3e, 552774, 1:200 FC; BD Biosciences; FITC Anti-Mouse CD19, 557398, 1:200 FC, BD Biosciences). Cellular analysis was conducted on a flow cytometer (LSR Fortessa, BD biosciences), with subsequent data processing performed using FlowJo analytical software.

Ethics statement

All animal experimentation strictly adhered to the regulations outlined by the State Council-approved People’s Republic of China Regulations on the Administration of Experimental Animal Affairs (1 November 1988). The experimental procedures involving mice complied with the guidelines of and were approved by the Institutional Animal Care and Use Committee of Shandong Normal University (permission number: AEECSDNU2024027).

Statistical analysis

GraphPad Prism software (version 8.0; GraphPad, San Diego, CA, USA) was employed for statistical analysis. The Student’s t test, one-way ANOVA, and two-way analysis of variance (ANOVA) were applied to determine statistical significance. The results, presented as means and standard deviations (SD), are representative of at least three independent experiments. A p value equal to or below 0.05 was considered statistically significant.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Author contributions

H.H. and H.W. concepted of the study. H.H., H.W., and H.C. designed the experiments. H.C., R.Y., W.Q., A.X., X.L., and R.K. performed most of the immunoblotting, immunofluorescence, q-PCR, and pathological experiment in vitro and in vivo. H.C. and R.Y. collected and analyzed the data. H.C. wrote the original draft. H.H. and P.H. edited and revised the original draft. All the authors reviewed and approved the final manuscript.

Peer review

Peer review information

Communications Biology thanks Gaopeng Hou and the other, anonymous, reviewers for their contribution to the peer review of this work. Primary Handling Editors: Si Ming Man and Mengtan Xing. A peer review file is available.

Data availability

All data supporting the findings of this study are available from the corresponding author on reasonable request. Source data behind the graphs and the original uncropped western blot images can be found in Supplementary Data 1.

Competing interests

The authors declare no competing interests.

Supplementary information

The online version contains supplementary material available at 10.1038/s42003-025-09380-2.