https://orcid.org/0000-0002-4005-6071
Figures
Abstract
The RIG-I-like receptor (RLR) signaling pathway plays a critical role in the host defense against RNA virus infection. Among the RLR family members, retinoic acid-inducible gene I (RIG-I) and melanoma differentiation-associated gene 5 (MDA5) are key cytosolic sensors that initiate type I interferon (IFN-I) responses. Their controllable expression, activation, and degradation are essential for maintaining immune homeostasis. However, the precise regulatory mechanisms governing RIG-I and MDA5 function during viral infection remain unclear. Here, we uncover that the E3 ubiquitin ligase RNF20 exerts dual regulatory roles in RLR signaling by modulating the expression and promoting the degradation of RIG-I and MDA5 in a nucleocytoplasmic translocation-dependent manner during viral infection. Under resting conditions, RNF20 resides in the nucleus, where it maintains immune readiness by regulating the basal and inducible transcription of RIG-I and MDA5. Upon RNA virus infection, RNF20 translocates to the cytoplasm via the export receptor CRM1. There, it recognizes the degron motifs of RIG-I and MDA5 through its coiled-coil domain and catalyzes their K27-linked ubiquitination and degradation, thereby preventing excessive antiviral signaling. These findings shed light on the significant and dual regulatory roles of RNF20 in maintaining innate immune homeostasis.
Author summary
The innate immune system uses RIG-I/MDA5 to fight RNA viruses, but their activity must be balanced—too weak and viruses escape, too strong and inflammation harms the body. We found E3 ubiquitin ligase RNF20 acts as a “switch”: in the nucleus, it maintains basal RIG-I/MDA5 levels via transcription to prime defense. Upon infection, RNF20 moves to the cytoplasm via CRM1, binds RIG-I/MDA5 through its coiled-coil domain, and promotes their K27-linked ubiquitination and degradation to curb overactivation. This dual role of RNF20 highlights a sophisticated mechanism that fine-tunes the immune response, offering insights into potential therapeutic strategies for managing viral infections and immune-related disorders.
Citation: Wang J, Liu Q, Zhao S, Shao Q, Fu F, Zhang K, et al. (2026) RNF20 dynamically regulates RIG-I and MDA5 transcription and degradation via nucleocytoplasmic translocation to balance antiviral signaling. PLoS Pathog 22(2): e1013890. https://doi.org/10.1371/journal.ppat.1013890
Editor: Ashley L. St. John, Duke-National University of Singapore, SINGAPORE
Received: July 23, 2025; Accepted: January 12, 2026; Published: February 19, 2026
Copyright: © 2026 Wang et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the manuscript and its Supporting information files.
Funding: This work was supported by the National Key Research and Development Program of China (2023YFD1802600 to J.M.), the Shanghai Municipal Agricultural and Rural Affairs Committee (K2023011 to J.S.), the National Natural Science Foundation of China (32373021 to J.S. and 32473047 to Y.C.), the Science and Technology Commission of Shanghai Municipality (25N42800300 to J.S.), the Zhaoqing Municipal Bureau of Science and Technology (202412008 to Y.C.), and the China Postdoctoral Science Foundation (2024M761988 to J.W.). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
The connections between pattern recognition receptors (PRRs) and pathogen-associated molecular patterns (PAMPs) constitute the crucial signaling pathways for triggering innate immune responses [1]. PRRs include Toll-like receptors (TLRs), NOD-like receptors (NLRs), and RIG-I-like receptors (RLR) [2]. RLR consist of three members: RIG-I, melanoma-differentiation-associated protein 5 (MDA5) and laboratory of genetics and physiology 2 (LGP2) [3]. Among them, RIG-I and MDA5 contain two caspase activation and recruitment domains (CARDs), which mediate oligomerization and downstream signal transduction. In contrast, LGP2 lacks CARD domains and thus cannot initiate signal transduction independently [4]. Therefore, RIG-I and MDA5 are responsible for recognizing cytosolic RNA, particularly virus-derived RNA [5]. These receptors remain inactivated and are expressed at low basal levels in most cells to facilitate the monitoring of PAMPs under normal conditions [6,7]. Studies have demonstrated that mice lacking RIG-I or MDA5 are highly susceptible to RNA virus infection, and RIG-I deficiency has been associated with hepatocellular carcinogenesis [8,9]. This indicates that RIG-I and MDA5 are crucial for maintaining immune defense.
Upon activation, both RIG-I and MDA5 undergo conformational changes that expose their caspase activation and recruitment domains (CARDs), which interact with the adaptor protein mitochondrial antiviral signaling (MAVS) to trigger a series of cascade reactions, ultimately inducing the release of type I interferon (IFNs) and proinflammatory cytokines [10,11]. This process involves ubiquitin-modifications, such as the E3 ubiquitin ligase tripartite interaction motifs 25 (TRIM25)-mediated K63-linked-polyubiquitination that is required for RIG-I activation [12], and the K63-linked ubiquitination of MDA5 at lysine 743 catalyzed by TRIM65, which is critical for MDA5 oligomerization and activation [13]. These findings suggest that the ubiquitination of RIG-I and MDA5 is essential for the activation of RLR signaling. However, sustained activation of RIG-I or MDA5 can lead to immune-mediated tissue damage and autoimmune diseases [14]. Therefore, precise regulation of the activation of RIG-I and MDA5 is essential to maintain innate immunity homeostasis.
The E3 ubiquitin ligase RNF20, a member of the RING finger protein family, possesses a RING domain in its C-terminus, which shares homology with the Saccharomyces cerevisiae protein Bre1 [15]. Research has shown that RNF20 mainly mediates the monoubiquitylation of histone H2B at lysine K119 or K120 (H2Bub1) [16], and plays a critical role in maintaining the stability of the chromatin structure and repairing DNA double-stranded breaks (DSBs) [17,18]. Our previous research found that chicken RNF20 dually regulates the MDA5-mediated immune response. However, there are notable differences in the cellular localization of chicken and human RNF20 [19]. These observations suggest that the roles of RNF20 in regulating innate immunity differ between chickens and humans, and that its regulatory functions in mammalian innate immunity remain to be elucidated.
Here, we discovered that RNF20 dually regulates both the activation and expression of RIG-I and MDA5. This dual regulation involves altering the spatiotemporal dynamics of RNF20, particularly its translocation from the nucleus to the cytoplasm. RNF20 is primarily localized in the nucleus, where it initiates the basal and inducible transcription of RIG-I and MDA5, to maintain immune defense. Upon RNA virus infection, RNF20 interacts with the nuclear export receptor CRM1, leading to its translocation from the nucleus to the cytoplasm. There, RNF20 catalyzes the ubiquitin-mediated degradation of RIG-I and MDA5 by specifically recognizing their degron motifs. This “braking” mechanism prevents the excessive activation of immunity mediated by RIG-I and MDA5. Overall, our findings demonstrate that RNF20 dynamically regulates RIG-I- and MDA5-mediated innate immune signaling to maintain immune homeostasis.
Results
RNF20 negatively regulates type I interferon responses induced by RNA virus infection
To identify the role of RNF20 in type I interferon (IFN) responses, we overexpressed RNF20 and observed a significant dose-dependent inhibition of VSV-GFP-induced IFN-β promoter activation (Fig 1A). Quantitative PCR (qPCR) analysis further revealed that RNF20 overexpression markedly suppressed VSV-GFP-induced expression of genes associated with innate immunity and inflammatory responses in A549, HeLa, and THP-1 cells (Figs 1B-1F and S1A-S1D). To validate these findings, cells overexpressing RNF20 were infected with either VSV-GFP or H1N1. After a 24-hour incubation, the culture supernatant was collected, and viral titers were determined using the 50% tissue culture infectious dose (TCID₅₀) assay or by measuring viral fluorescence. The results showed that virus titer and virus fluorescence increased in a concentration-dependent manner with RNF20 overexpression (Fig 1G-1J). The RIG-I-like receptors (RLR) signaling pathway is a key intracellular mechanism mediating antiviral defense. We found that overexpression of RNF20 significantly inhibited the expression of key proteins-including MDA5- and RIG-I-as well as the phosphorylation of TBK1, and IRF3 (Fig 1K-1N). These results collectively demonstrate that RNF20 negatively regulates innate immune responses.
(A) IFN-β luciferase reporter assay in HEK293T cells transiently transfected with increasing doses of RNF20-Flag plasmid (0, 125, 250, or 500 ng) and subsequently infected with VSV-GFP (MOI = 1.0) for 12h. (B-F) Quantitative real-time PCR (qPCR) analysis the mRNA levels of RNF20, type I interferons (IFN-α, IFN-β, IFN-ε, and IFN-κ), type II interferon (IFN-γ), type III interferon (IFN-λ1), interferon-stimulated genes (OAS1 and MX1), and proinflammatory cytokines (IL-1β and IL-6) in A549 cells transfected with RNF20 or control plasmid and infected with VSV-GFP for 24h. (G, H) TCID50 assay were performed to determine VSV-GFP and H1N1 viral titers in A549 cells transfected with the RNF20 expression plasmid and infected (MOI = 1.0) for 24h. (I) Fluorescence microscopy analysis of VSV-GFP replication in A549 cells transfected with empty vector or increasing doses of RNF20-Flag plasmid, followed by infection with VSV-GFP (MOI = 1.0) for the indicated times. Scale bars = 50 µm. (J) Quantitative mean fluorescence intensity. (K-N) Immunoblot and quantitative analysis of total and phosphorylated TBK1 and IRF3, total RNF20, RIG-I, MDA5, STING, and β-tubulin in A549 cells transiently transfected with increasing doses of RNF20-Flag plasmid (0, 250, or 500 ng) and left uninfected (UI) or infected with VSV-GFP or H1N1 (MOI = 1.0) for 24h. Data are expressed as mean ± SD; ns: no significance; *p < 0.05, **p < 0.01, ***p < 0.001; The results were repeated in three independent experiments with three replicas each. (Note: UI denotes uninfected virus; pcDNA3.1: empty vector control).
To further confirm that RNF20 negatively regulates the antiviral innate immune response mediated by the RLR signaling pathway, we knockdown the expression of endogenous RNF20 and infected cells with the VSV-GFP. The qPCR analysis revealed a significant upregulation of genes associated with innate immunity and inflammatory responses (Fig 2A-2D). Following gradient knockdown of RNF20, we observed that decreasing RNF20 protein levels led to a dose-dependent increase in RIG-I, MDA5, and phosphorylated TBK1, accompanied by a marked reduction in viral protein expression and fluorescence intensity (Fig 2E-2H). These indicate that knockdown of RNF20 expression promotes the activation of type I interferon responses mediated by the RLR signaling pathway and inhibits viral replication. Collectively, these results demonstrate that RNF20 functions as a negative regulator of RLR-mediated innate immune responses.
(A-D) qPCR analysis of RNF20, IFN-β, IFN-γ, IFN-λ1, MX1, OAS1, IL-1β, and IL-6 mRNA levels in A549 cells transiently transfected with RNF20 interfering plasmid and subsequently infection with VSV-GFP (MOI = 1.0) for 24h. (E, F) Immunoblot and quantitative analysis of RNF20, RIG-I, MDA5, GFP, β-tubulin and phosphorylated (p-) TBK1 in A549 cells were transiently transfected with increasing doses of RNF20 interfering plasmid (0, 500,1000, or 1500 ng) and subsequently infection VSV-GFP (MOI = 1.0) for 24h. (G) Fluorescence microscopy analysis of VSV-GFP replication in A549 cells transfected with RNF20-interfering or control plasmid and infected with VSV-GFP (MOI = 1.0) for 24h. Scale bars = 50 µm. (H) Quantitative mean fluorescence intensity. Data are expressed as mean ± SD; ns: no significance; *p < 0.05, **p < 0.01, ***p < 0.001; The results were repeated in three independent experiments with three replicas each. (Note: shNC: non-targeting shRNA control).
RNF20 targets RIG-I and MDA5
To demonstrate that RNF20 negatively regulates the innate immune response mediated by the RLR signaling pathway, we co-transfected RNF20 with key molecules involved in the RLR pathway. The results show that RNF20 significantly reduced the activation of IFN-β promoter induced by MDA5, RIG-I, STING, and TBK1 (Fig 3A). However, no significant effect was observed in MAVS-induced IFN-β promoter activity (Fig 3A), suggesting that RNF20 acts on molecules upstream of MAVS to suppress downstream signal transduction. Co-transfected of RNF20 with either MDA5 or RIG-I observed a concentration-dependent attenuation of IFN-β promoter activation (Fig 3B and 3C). Conversely, RNF20 knockdown significantly enhanced the expression of RIG-I and MDA5, and promoted activation of the IFN-β promoter (Figs 3D and S2A-S2D). Subsequently, we generated RIG-I/MDA5 double-knockdown cell lines using the CRISPR/Cas9 strategy and observed markedly enhanced viral replication in double-knockdown cells compared to wild-type controls (S2E-S2H Fig). Importantly, RNF20 overexpression in wild-type cells promoted viral replication and inhibited IFN-β mRNA expression. However, in RIG-I/MDA5 double-knockdown cells, RNF20 overexpression had no effect on IFN-β mRNA expression or other molecules within the RLR signaling pathway (S2I-S2K Fig). This suggests that RNF20 attenuate type I interferon signaling by targeting RIG-I and MDA5.
(A) IFN-β luciferase reporter assay of HEK293T cells transiently transfected with HA-tagged RIG-I, MDA5, MAVS, STING, or TBK1 (100 ng) together with either RNF20-Flag expression plasmid (100 ng) or an empty control vector, followed by infection with VSV-GFP (MOI = 1.0) for 12h. (B, C) HEK293T cells were transiently co-transfected with IFN-β-luc (120 ng), pRL-TK (60 ng), and HA-tagged RIG-I or MDA5, together with increasing doses of RNF20-Flag expression plasmid (0, 50, 100, 200, 300, or 400 ng). (D) HEK293T cells were transiently co-transfected with IFN-β-luc (120 ng), pRL-TK (60 ng), poly(I:C), and Flag-RIG-I (100 ng), together with RNF20-interfering plasmid. (E, F) HEK293T cells were co-transfected with Flag-RNF20 and either Myc-tagged RIG-I or Myc-tagged MDA5, followed by infection with VSV-GFP (MOI = 1.0) for 12h. Cell lysates were immunoprecipitated using an anti-Flag antibody and analyzed by immunoblotting with the indicated antibodies. (G, H) Confocal microscopy analysis of HeLa cells infected with NDV for 24 h, followed by immunostaining with anti-RNF20, anti-RIG-I, or anti-MDA5 primary antibodies, and Alexa Fluor 488-conjugated goat anti-mouse IgG (green) or Cy3-conjugated goat anti-rat IgG (red) secondary antibodies. (I, J) Schematic representation of RIG-I and MDA5 structures and their truncated mutants. (K) HEK293T cells were co-transfected with Myc-tagged MDA5 or its mutants together with Flag-RNF20, followed by infection with VSV-GFP (MOI = 1.0) for 12h. Cell lysates were immunoprecipitated using an anti-Myc antibody and analyzed by immunoblotting with the indicated antibodies. (L, M) HEK293T cells were co-transfected with Myc-tagged MDA5 or RIG-I (wild-type or dCARD mutants) together with Flag-RNF20, followed by infection with VSV-GFP (MOI = 1.0) for 12h. Cell lysates were immunoprecipitated using an anti-Myc antibody and analyzed by immunoblotting with the indicated antibodies. Data are presented as mean ± SD; ***p < 0.001; Scale bars = 10 µm. The results were repeated in three independent experiments with three replicas each. (Note: UI denotes uninfected virus; pcDNA3.1: empty vector control; shNC: non-targeting shRNA control).
To validate the interaction between RNF20 and either RIG-I or MDA5, we performed co-immunoprecipitation (co-IP) assays, which confirmed that both RIG-I and MDA5 physically interact with RNF20 (Fig 3E and 3F). Immunofluorescence staining further revealed that, following NDV infection for 24 h, endogenous RNF20 colocalized with RIG-I and MDA5 (Fig 3G and 3H). To identify the specific domains mediating the interaction between RNF20 and RIG-I/MDA5, we generated five truncated mutants of RIG-I and MDA5 (Fig 3I and 3J). Co-transfection experiments showed that RNF20-mediated degradation is mediated by the CARD domains (S2L-S2O Fig), and co-IP analysis further demonstrated that RNF20 specifically interacts with the CARD1 domain of MDA5 (Fig 3K). Moreover, deletion of the CARD domains from RIG-I or MDA5 abolished their interaction with RNF20 (Fig 3L and 3M). Collectively, these findings provide compelling evidence that RNF20 negatively modulates innate immune signaling by targeting RIG-I and MDA5.
RNF20 catalyzes K27-linked ubiquitination of RIG-I and MDA5
RNF20 is a RING-type E3 ubiquitin ligase that catalyzes the ubiquitination of substrate proteins. To determine whether RNF20 mediates the ubiquitination-dependent degradation of RIG-I and MDA5, RNF20 was co-transfected with RIG-I or MDA5. The co-IP results demonstrated that RNF20 overexpression enhanced the ubiquitination of RIG-I and MDA5. Furthermore, treatment with the proteasome inhibitor MG132 markedly increased RNF20-mediated ubiquitination of RIG-I and MDA5 (Fig 4A and 4B). These findings indicate that RNF20 promotes the ubiquitination and subsequent proteasomal degradation of RIG-I and MDA5.
(A, B) Co-immunoprecipitation (Co-IP) analysis of RIG-I or MDA5 ubiquitination in HEK293T cells transfected with plasmids encoding Myc-RIG-I or Myc-MDA5, HA-tagged ubiquitin (HA-Ub), and either control vector or RNF20-EGFP, followed by treatment with MG132 (10 µM) for 4h. (C, D) Co-IP analysis of RIG-I or MDA5 ubiquitination in HEK293T cells transfected with Myc-RIG-I or Myc-MDA5, RNF20-Flag, and HA-Ub (WT) or its lysine-only mutants (K6, K11, K27, K29, K33, K48, K63). (E) Quantification of the polyubiquitin linkage types on RIG-I and MDA5. (F, G) Co-IP analysis of RIG-I or MDA5 ubiquitination in HEK293T cells transfected with Myc-RIG-I or Myc-MDA5, RNF20-Flag, and either HA-Ub (WT) or HA-Ub-K27R mutant. (H, I) Quantification of RNF20-mediated ubiquitination of RIG-I and MDA5. Data are presented as mean ± SD from three independent experiments (n = 3). ns: no significance, *p < 0.05, **p < 0.01, ***p < 0.001.(Note: pcDNA3.1: empty vector control).
To identify the specific linkage type of RNF20-mediated polyubiquitination on RIG-I and MDA5, we employed a panel of ubiquitin mutants in which all lysine (K) residues were replaced by arginine (R), except for one remaining lysine at positions K6, K11, K27, K29, K33, K48, or K63. RNF20-mediated polyubiquitination of RIG-I and MDA5 was markedly enhanced in the presence of HA-Ub-K27 compared with other ubiquitin mutants (Fig 4C-4E). Further ubiquitination assays were performed using the ubiquitin K27R mutant, which contains a single lysine-to-arginine substitution at position 27. RIG-I and MDA5 were ubiquitinated by RNF20 in the presence of wild-type ubiquitin (WT), but not by the K27R mutant (Fig 4F-4I). Collectively, these findings demonstrate that RNF20 catalyzes K27-linked polyubiquitination of RIG-I and MDA5.
RNF20 coiled-coil (CC) domains mediate RIG-I and MDA5 ubiquitination via specific degron motifs
RNF20 contains internal coiled-coil (CC) domains and a C-terminal RING-finger domain (Fig 5A). To identify the functional domains of RNF20 involved in regulating the RLR signaling pathway, we constructed mutant plasmids with deletions of the CC domains (dCC1, dCC2, dCC3) or the RING domain (dRING) (Fig 5A). Notably, deletion of the first coiled-coil domain (dCC1) abolished RNF20’s inhibitory effect on IFN-β promoter activation (Fig 5B), suggesting that its active functional region resides within the N-terminal CC1 domain. To further delineate this region, we generated a series of N-terminal truncation mutants of RNF20. Reporter assays revealed that amino acids 40–90 at the N-terminus were essential for suppressing IFN-β promoter activation induced by RIG-I or MDA5 (Fig 5C and 5D). Western blot analysis further confirmed that this 40–90 aa segment mediates the ubiquitination and degradation of RIG-I and MDA5 (Figs 5C, 5D, S3A, and S3B). Interestingly, the CC1 domain is located precisely within residues 43–90 aa, supporting the conclusion that CC1 serves as the core functional domain responsible for the inhibitory activity of RNF20 on RLR signaling.
(A) Schematic diagram of RNF20 and its deletion mutants. (B) Luciferase reporter assay of HEK293T cells transiently transfected with IFN-β-luc (120 ng), pRL-TK (60 ng), and RNF20 (wild-type or mutant) expression plasmids or control vector. (C, D) HEK293T cells were transiently transfected with IFN-β-luc (120 ng), pRL-TK (60 ng), and Myc-RIG-I or Myc-MDA5, together with C-terminal truncation mutants of RNF20 (upper panels). Immunoblotting was performed with anti-Myc and anti-β-tubulin antibodies (lower panels). (E) HEK293T cells were co-transfected with Myc-MDA5 and RNF20-HA (wild-type or mutant) expression plasmids. Cell lysates were immunoprecipitated with anti-HA antibody and analyzed by immunoblotting with the indicated antibodies. (F) Co-immunoprecipitation analysis of the polyubiquitination of MDA5 in HEK293T cells transfected with Myc-MDA5, RNF20-Flag (wild-type or mutant), and HA-ubiquitin expression plasmids. (G) Quantification of MDA5 ubiquitination levels in immunoprecipitated samples. (H) Amino acid sequence alignment of the CARD domains of RIG-I and MDA5 showing the conserved “KENW” motif. (I, J) Co-immunoprecipitation analysis of the polyubiquitination of human RIG-I or MDA5 (wild-type or K-to-R mutants) in HEK293T cells transfected with Myc-RIG-I or Myc-MDA5, Flag-RNF20, and HA-ubiquitin expression plasmids. (K, L) Quantification of RIG-I and MDA5 ubiquitination levels in immunoprecipitated samples. (M-P) Immunoblot and quantification analysis of HEK293T cells transiently transfected with RNF20-HA and either RIG-I-Flag or RIG-I-dKENW-Flag, or MDA5-Flag or MDA5-dKENW-Flag, followed by VSV-GFP infection (MOI = 1.0) for 24h. Cell lysates were analyzed by immunoblotting with anti-Flag, anti-HA, and anti-β-tubulin antibodies. Data are shown as mean ± SD from three independent experiments (n = 3). ns: no significance, p < 0.05, *p < 0.01, **p < 0.001.(Note: pcDNA3.1: empty vector control).
To determine the structural domains of RNF20 responsible for interacting with RIG-I and MDA5, we performed co-immunoprecipitation following co-transfection of wild-type or deletion mutants of RNF20 with RIG-I or MDA5. Deletion of the CC3 domain abolished RNF20-MDA5 interaction, while deletions of CC1 or CC3 markedly reduced MDA5 polyubiquitination compared to wild-type RNF20 (Fig 5E-5G). These results indicate that the E3 ubiquitin ligase catalytic activity of RNF20 is localized to the CC1 domain, whereas the CC3 domain is required for physical interaction with the CARD domains of RIG-I and MDA5.
Because RNF20 targets the CARD domains of RIG-I and MDA5 for their ubiquitination and degradation, we hypothesized that this process is governed by a conserved mechanism. Sequence alignment of the CARD domains identified a conserved “KENW” motif that is present across diverse species (Fig 5H). We postulated that the lysine (K) residue within this motif functions as the acceptor site for RNF20-mediated polyubiquitination. To test this hypothesis, we introduced site-directed point mutations substituting lysine with arginine (R) in RIG-I (K164R) and MDA5 (K174R). Co-immunoprecipitation (Co-IP) analysis demonstrated that RNF20-mediated polyubiquitination was markedly reduced in these mutants compared with their wild-type counterparts (Fig 5I-5L). Similarly, RNF20-mediated ubiquitination was substantially diminished when the corresponding lysine residues were mutated in porcine RIG-I (K163R) or MDA5 (K174R) (S3C-S3F Fig), and in chicken MDA5 (K44R, K172R) (S3G and S3H Fig). However, mutation of chicken MDA5 K44 to R did not abolish RNF20-mediated degradation, whereas the K172R mutation completely eliminated this effect, indicating that RNF20 mediates K172 site-specific ubiquitination and degradation of chicken MDA5 (S3I and S3J Fig). Furthermore, deletion or mutation of the “KENW” motif completely abolished RNF20’s capacity to degrade RIG-I and MDA5, thereby eliminating its inhibitory effect on the activation of the IFN-β promoter by these receptors (Figs 5M-5P, S3K, and S3L). Collectively, these findings demonstrate that RNF20 acts as a negative regulator of the RLR signaling pathway by catalyzing K27-linked ubiquitination of RIG-I and MDA5, mediated by its CC1 catalytic domain and targeting a conserved “KENW” degron motif within their CARD domains.
RNA virus triggers RNF20 nuclear export to negatively regulate innate immunity
It has been shown that RNF20 regulatory functions occur mainly in the nucleus, while RLR is mainly localized in the cytoplasm and membrane [20,21]. To investigate whether the subcellular localization of RNF20 influences its regulatory role in antiviral immunity, we infected HeLa cells with H1N1, NDV, or VSV-GFP and examined RNF20 localization by immunofluorescence staining and nuclear-cytoplasmic fractionation. In uninfected cells or during the early phase of infection, RNF20 was mainly confined to the nucleus. However, as viral replication progressed, RNF20 gradually accumulated in the cytoplasm (Fig 6A-6D). Notably, neither RNA nor DNA virus infection altered the mRNA or total protein levels of RNF20 (S4A-S4E Fig). Consistently, single-cell transcriptomic data confirmed that RNF20 expression remained stable following SARS-CoV-1/2 infection in human cell lines (S4F Fig). These results suggest that viral infection induces the relocalization rather than transcriptional upregulation of RNF20.
(A-C) Confocal microscopy of HeLa cells that were uninfected (UI) or infected with H1N1, NDV, or VSV-GFP for the indicated time points (0, 3, 6, 9, 12, and 24h). Cells were stained with an anti-RNF20 primary antibody followed by Alexa Fluor 488-conjugated goat anti-rabbit IgG (green) or Cy3-conjugated goat anti-rat IgG (red). Scale bars = 10 μm. (D) Immunoblot analysis of RNF20 in cytoplasmic and nuclear fractions of A549 cells infected with VSV-GFP for the indicated time points. (E) Immunoblot analysis of RNF20 in cytoplasmic and nuclear fractions of A549 cells that were UI or infected with VSV-GFP for 24h and then treated with DMSO and KPT330. (F, G) qPCR analysis of RNF20 and IFN-β mRNA levels in A549 cells that were transiently transfected with RNF20 expression or control plasmid, infected with VSV-GFP for 24h and treated with DMSO or KPT330. (H) Co-immunoprecipitation analysis of CRM1-Flag and RNF20-HA in HEK293T cells. Cell lysates were immunoprecipitated with anti-Flag antibody and immunoblotted with the indicated antibodies. (I, J) Confocal microscopy of HeLa cells transfected for 24 h with RNF20-dNLS-EGFP or RNF20-dNES-EGFP expression plasmids, followed by infection with NDV for 24 h or left uninfected. Scale bars = 10 μm. (K, L) Luciferase reporter assay of HEK293T cells transiently transfected with IFN-β-luc (120 ng), pRL-TK (60 ng), and RNF20 (wild-type or mutant) expression plasmids or control vector. (M, N) qPCR analysis of IFN-β, OAS1, IL-1β, IL-6, and IL-8 mRNA levels in A549 cells that were transiently transfected with RNF20 (WT) or RNF20-dNLS-expressing plasmid or control plasmid and infected with VSV-GFP for 24h. n = 3; Data were represented by the mean ± SD with corresponding significance. *p < 0.05, **p < 0.01, ***p < 0.001. (Note: UI denotes uninfected virus; pcDNA3.1: empty vector control; shNC: non-targeting shRNA control).
To determine whether cytoplasmic RNF20 arises from de novo synthesis or nuclear export, we treated cells with KPT330, a selective inhibitor of the nuclear export receptor CRM1. Strikingly, KPT330 treatment markedly reduced the cytoplasmic accumulation of RNF20 following viral infection (Fig 6E). Furthermore, inhibition of RNF20 nuclear export by KPT330 significantly enhanced IFN-β mRNA expression, even in the presence of RNF20 overexpression (Fig 6F and 6G). These findings indicate that RNA virus infection promotes RNF20 nuclear export, which in turn suppresses type I interferon responses.
Nucleocytoplasmic translocation of RNF20 is expected to depend on transporter proteins. Mass spectrometry analysis of RNF20-interacting partners identified several nucleocytoplasmic transport proteins, among which IPO5 and CRM1 drew our attention. IPO5 mediates nuclear import, whereas CRM1 is responsible for nuclear export [22,23]. Co-immunoprecipitation assays confirmed the interaction between RNF20 and CRM1 (Fig 6H). RNF20 contains both a nuclear localization signal (NLS) and a nuclear export signal (NES), which are essential for its dynamic shuttling between the nucleus and cytoplasm. To further assess whether cytoplasmic RNF20 is responsible for suppressing antiviral responses, we constructed RNF20 mutants lacking either the NLS or NES (RNF20-dNLS and RNF20-dNES, respectively). As expected, RNF20-dNLS-EGFP was predominantly localized in the cytoplasm, while RNF20-dNES-EGFP was retained in the nucleus before and after NDV infection (Fig 6I and 6J). Functional assays revealed that RNF20-dNLS markedly enhanced its inhibitory effect on IFN-β promoter activation (Fig 6K), whereas RNF20-dNES had no impact on IFN-β promoter activation (Fig 6L). In addition, RNF20-dNLS substantially suppressed the expression of multiple innate immune genes (Fig 6M and 6N). Collectively, these results demonstrate that RNA virus infection induces CRM1-dependent nuclear export of RNF20, enabling its accumulation in the cytoplasm where it acts as a negative regulator of RLR-mediated type I interferon signaling.
RNF20 is essential for RLR-mediated activation of innate immunity in vivo and in vitro
We established RNF20-deficient cell lines using the CRISPR/Cas9 strategy to further investigate the function of RNF20 in antiviral innate immunity. To our surprise, after infecting VSV-GFP in RNF20-/- and wild-type A549 cells, a stronger fluorescence intensity of vesicular stomatitis virus (VSV-GFP) was observed in the RNF20-/- cells (Fig 7A and 7B). And the VSV-GFP induced mRNA expression of type I interferons (IFN-α, IFN-β, IFN-ε, and IFN-κ), type II interferons (IFN-γ), and type III interferons (IFN-λ1), as well as genes associated with innate immunity (OAS1 and MX1), and inflammation factors (IL-1β and IL-6) was significantly reduced in the RNF20-/- cells (Fig 7C-7F). Subsequently, we generated Rnf20-deficient mice (Rnf20-/-) to explore the function of RNF20 in innate immunity. It is regrettable that, Rnf20-/- mice exhibited embryonic lethality, while the heterozygous Rnf20+/- mice reproduced normally. Therefore, we utilized Rnf20+/- mice for subsequent investigations (S5A and S5B Fig). By infecting wild-type and Rnf20+/- mice with H1N1 or VSV-GFP viruses, it was found that Rnf20+/- mice exhibited significantly higher mortality rates compared to wild-type mice (Fig 7G). We also found that the expression of genes related to innate immunity (IFN-β, OASL, and MX1), and inflammatory factor (IL-1β, TNF-α, and IL-6) were significantly reduced in Rnf20+/- mice spleen and lung compared to wild-type mice (WT) following H1N1 or VSV-GFP infection (Figs 7H, 7I and S5C-S5F). These findings indicate that RNF20 is essential for maintaining the innate immune response.
(A) Microscopy imaging showing VSV-GFP in A549 WT or RNF20-/- cells after being infected with VSV-GFP for 12h or 24h. Scale bars = 50 µm. (B) Quantitative mean fluorescence intensity. (C-F) qPCR analysis of type I interferons (IFN-α, IFN-β, IFN-ε, and IFN-κ), type II interferons (IFN-γ), type III interferons (IFN-λ1), interferon stimulation-related genes (OAS1 and MX1) and inflammatory factor related genes (IL-1β and IL-6) mRNA levels in A549 WT or RNF20-/- cells infected with VSV-GFP for 24h. (G) Survival data for the mice infected with H1N1 (1.0 × 106 TCID50) (n = 5). (H, I) qPCR analysis of IFN-β, OASL, MX1, IL-1β, and IL-6 mRNA levels in spleen and lung from wild type (WT) or Rnf20+/- mice after infecting of H1N1 for 18h. (J) Transcriptome sequencing analysis of gene expression profiles in RNF20-/- and wild-type cells infected with VSV for 24h. KEGG pathway enrichment analysis of downregulated genes in RNF20-/- A549 cells infected with VSV-GFP for 24h. Bars represent -log₁₀(q-value); all shown pathways have q < 0.05. (K) Heatmap analysis of downregulated genes involved in innate immune signaling pathways. (L) qPCR analysis of RIG-I, MDA5, MAVS, and TBK1 mRNA levels in A549 (WT) and A549 (RNF20-/-) cells infected with VSV-GFP for 24h. (M,N) Immunoblot and quantitative analysis of RNF20, β-tubulin, and RLR signaling pathway proteins MDA5, RIG-I, MAVS and STING in A549 WT or RNF20-/- cells uninfected or infected with VSV-GFP for 24h. (O, P) Immunoblot and quantitative analysis of RNF20, β-tubulin, and RLR signaling pathway proteins MDA5, RIG-I, phosphorylated (p-) TBK1 and IRF3 in wild type or Rnf20+/- mice uninfected or infected with VSV-GFP for 24h. (Q) H&E staining of lung tissue sections is shown. n = 3; Data were represented by the mean ± SD with corresponding significance. ns: no significance, *p < 0.05, **p < 0.01, ***p < 0.001. (Note: RNF20-/- denotes RNF20 knockout cells, whereas Rnf20+/- denotes RNF20 heterozygous knockout mice.).
To further explore the role of RNF20 in innate immunity, we performed transcriptome sequencing on RNF20-/- and wild-type cells after 24h of VSV-GFP infection. The results demonstrated that RNF20 deficiency significantly alters gene expression profiles (S5G and S5H Fig). Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis showed that the down-regulated genes were associated with innate immune signaling pathways and inflammatory factor signaling pathways, including the RLR-like receptors pathway, TNF signaling pathway, and NF-κB signaling pathway (Fig 7J). Notably, heat map analysis revealed that the expression of MDA5 and RIG-I, two important members of the RLR signaling pathway, was significantly reduced in RNF20-/- cells (Fig 7K). We hypothesized that RNF20 may regulate RLR-mediated innate immune responses.
Therefore, we examined the expression of key signaling molecules in the RLR signaling pathway and further demonstrated significantly decreased RIG-I and MDA5 mRNA and protein expression without affecting the expression of other genes in RNF20 knockout cells and Rnf20+/- mice (Fig 7L-7P). This results in the inability to induce IFN-β expression after RNA virus infection (S5I Fig). In addition, RNF20 deficiency greatly reduced VSV-GFP-induced phosphorylation of TBK1, total STAT1, IRF3 and IRF7 protein expression (S5J and S5K Fig). Conversely, the viral protein GFP of VSV-GFP was markedly increased in the RNF20-/- cells (S5J and S5K Fig), and hematoxylin-eosin staining revealed greater infiltration of immune cells and injury in the lungs of Rnf20+/- mice after VSV-GFP infection (Fig 7Q). Overall, these results strongly suggest that RNF20 is essential for RLR signaling-induced IFN-β activation.
RNF20 controls both the basal and inducible expression of RIG-I and MDA5 in the nucleus
To demonstrate that RNF20 regulates the expression of RIG-I and MDA5, we first analyzed RNF20-interacting proteins by mass spectrometry (S1 Table). KEGG and GO enrichment analysis revealed that RNF20 involved mRNA transcriptional regulatory and protein translation pathways (Fig 8A and 8B). Western blot analysis showed that in RNF20-/- cells, overexpression of RNF20 restored and upregulated RIG-I and MDA5 protein levels in a concentration-dependent manner (Fig 8C and 8D), suggesting that RNF20 is essential for regulating the expression of RIG-I and MDA5.
(A) Mass spectrometry analysis of RNF20-interacting proteins. Circle plots show KEGG pathway enrichment of RNF20-associated proteins (only pathways with P < 0.05 are displayed). (B) Bubble plots showing the top 30 Gene Ontology (GO) terms enriched among RNF20-interacting proteins. (C, D) Immunoblot and quantitative analysis of total and phosphorylated (p-) TBK1, RNF20, MDA5, RIG-I, IRF3, and β-tubulin in RNF20-/- A549cells transiently transfected with an RNF20 expression plasmid and infected with VSV-GFP for 24h. (E, F) Immunoblot and quantitative analysis of total and phosphorylated (p-) TBK1 and RNF20, MDA5, RIG-I, IRF3, and β-tubulin in RNF20-/- A549 cells transiently transfected with RNF20 (WT), RNF20-dNLS, or RNF20-dNES expression plasmids, followed by VSV-GFP infection for 24h. (G. H) Luciferase reporter assays of HEK293T cells transiently transfected with RIG-I or MDA5 promoter reporter plasmids together with RNF20 expression or control plasmids. (I, J) ChIP-qPCR analysis of RNF20 enrichment at the RIG-I and MDA5 promoter regions in cells under uninfected or NDV-infected cells. Data are expressed as mean ± SD, n = 3; ns: no significance, *p < 0.05, **p < 0.01, ***p < 0.001.
Previous studies have demonstrated that RNF20 primarily exerts its regulatory functions within the nucleus [24]. Because nuclear localization signals (NLS) and nuclear export signals (NES) determine the subcellular localization and functional distribution of proteins [25]. We investigated whether RNF20 regulates RIG-I and MDA5 expression in the nucleus. Overexpression of wild-type or mutant RNF20 in RNF20-/- cells revealed that deletion of the NLS (RNF20-dNLS) abolished the ability of RNF20 to restore RIG-I and MDA5 expression, whereas deletion of the NES (RNF20-dNES) markedly enhanced their expression (Fig 8E and 8F). Consistently, qPCR analysis confirmed that RNF20 positively regulates the mRNA expression of RIG-I and MDA5 (S6A and S6B Fig), indicating that RNF20 modulates their transcriptional activity.
To further verify that RNF20 directly regulates the basal transcription of RIG-I and MDA5, we constructed luciferase reporter plasmids containing the promoters of RIG-I and MDA5. Overexpression of RNF20 significantly enhanced the activity of both promoters (Fig 8G and 8H). To determine whether RNF20 directly binds to these promoter regions, we performed chromatin immunoprecipitation (ChIP) using RNF20-specific antibodies in uninfected and NDV-infected cells. Promoter mapping analysis indicated that RNF20 interacts with the + 100 to -500 bp regions of the RIG-I and MDA5 promoters. Accordingly, three pairs of primers were designed to amplify the RIG-I and MDA5 promoter fragments interacting with RNF20, with GAPDH serving as a control. The results showed that the RIG-I (0 to -200 bp) and MDA5 (-130 to -291 bp) promoter regions were successfully amplified from ChIP samples (S6C and S6D Fig). ChIP-qPCR further revealed a significant enrichment of endogenous RNF20 at these promoter regions compared with the IgG control. Notably, NDV infection markedly reduced RNF20 enrichment at these loci (Fig 8I and 8J), consistent with our earlier findings that viral infection triggers RNF20 nuclear export, leading to its reduced nuclear abundance.
Discussion
The RLR family members RIG-I and MDA5 play critical roles in anti-RNA virus immunity [26,27]. In uninfected cells, RIG-I and MDA5 exhibit an autoinhibition state [5]. Upon RNA virus infection, RIG-I and MDA5 undergo conformational changes, initiating a cascade of responses that involve the release of type I interferons, pro-inflammatory cytokines, and chemokines, ultimately controlling and eliminating the virus [28,29]. However, sustained activation of RIG-I or MDA5 can lead to immune overstimulation [30]. Hence, the expression and activation of RIG-I and MDA5 must be strictly controlled. In this study, we introduce a novel mechanism for delicately regulating RLR-mediated innate immune homeostasis, which is conserved across species. This mechanism involves E3 ubiquitin ligases RNF20 spatiotemporal dynamic change from the nucleus to the cytosol. Specifically, RNF20 regulates the basal transcription of RIG-I and MDA5 in the nucleus to maintain immune defense. Upon RNA virus infection, RNF20 interacts with CRM1 to undergo nucleocytoplasmic translocation. In the cytoplasm, RNF20 recognizes the degron motifs of RIG-I and MDA5 and catalyzes their ubiquitinated degradation to prevent overstimulation of immunity.
Ubiquitination of the CARD domains of RIG-I and MDA5 is crucial for activating the RLR signaling pathway. For instance, TRIM25 mediates K63-linked polyubiquitination of the RIG-I CARD domain, which is essential for its activation [12], whereas TRIM40 promotes K27- and K29-linked ubiquitination, leading to the degradation of the CARD domains of RIG-I and MDA5 [31]. Furthermore, RNF123 interacts with the CARD domains of RIG-I and MDA5 to negatively regulate innate immune signaling [32]. The ability of multiple E3 ubiquitin ligases to interact with the CARD domains of RIG-I and MDA5, suggests a conservation mechanism controlling their activation and inactivation. We by analysis of the amino acid sequences, conserved “KENW” lysine motifs were discovered both in the RIG-I and MDA5 CARD domains. RNF20 recognizes this motif and mediates the K27-linked polyubiquitination of RIG-I and MDA5, thereby preventing immune overstimulation. These “KENW” motifs can act as a “degron” sequence for RNF20 to specifically recognize and degrade RIG-I and MDA5. Degrons play a central role in protein quality control and intracellular signaling pathways by targeting unneeded or damaged proteins for degradation, thereby preventing their potential dysfunction. Dysregulation of degron activity has been linked to various diseases, including cancer, immune disorders, and neurodegenerative conditions [33,34]. Although the human genome encodes approximately 600 E3 ubiquitin ligases, only a small fraction of these enzymes have well-defined degron motifs, largely due to methodological limitations [35]. In this study, we identified a novel degron motif, “KENW,” through functional validation and sequence analysis. This degron may serve as a potential therapeutic target for antiviral intervention and clinical applications.
Research indicates that most E3 ubiquitin ligases mediate substrate protein ubiquitination through their RING domain [36]. Consequently, it has been widely assumed that RNF20 also exerts its catalytic activity via the RING domain [37]. However, our deletion studies on RNF20 functional domains reveal that the E3 ubiquitin ligase active site is located in its first coiled-coil domain, while the domain interacting with RIG-I and MDA5 resides in its third domain. The latest research has also found that the coiled - coil domain of E3 ubiquitin ligase regulates the master switch of enzyme activity and functional specificity [38]. It shows that the coiled domain of RNF20 not only mediates its interaction with RIG-I and MDA5, but also participates in the regulation of RNF20 enzyme activity.
Previous studies have shown that RNF20 is mainly localized in the nucleus regulation of histone ubiquitination [39], DNA damage repair [35], chromosome segregation and homologous recombination repair [17,24] cell division [40], and differentiation [41]. However, RIG-I and MDA5 belong to the cytoplasmic receptors. The subcellular localization of proteins is closely related to the function [42], implying that viral infection can trigger nucleoplasmic relocalization of RNF20. To investigate this, we tracked the RNF20 changes during viral infection and found that the subcellular localization and morphology of RNF20 underwent a dynamic change. RNF20 was translocated out of the nucleus by transporter proteins CRM1. Therefore, we removed the nuclear localization signal (NLS) of RNF20 and found that RNF20-dNLS significantly increased the inhibitory effect of RNF20 on innate immune-related genes and inflammatory factor-related genes. These data suggest that the spatiotemporal dynamics of RNF20 are critical for the maintenance of innate immune homeostasis. Given RNF20’s dual role in priming basal RIG-I/MDA5 expression while preventing excessive activation upon infection, targeted modulation of its nucleocytoplasmic shuttling or substrate recognition could offer therapeutic opportunities. For example, selective inhibition of CRM1-mediated nuclear export (e.g., using clinically approved inhibitors such as selinexor) might enhance antiviral responses during acute infection, whereas degron-mimetic or coiled-coil-targeted strategies could dampen overactivation in chronic inflammatory or autoimmune settings.
To further insight into the mechanism of RNF20 regulating innate immunity, we generated an RNF20-deficient cell line and mice. Unfortunately, Rnf20-/- mice are embryonic lethal, but heterozygous mice (Rnf20+/-) survive normally. This observation is consistent with previous studies, which have demonstrated that RNF20 is essential for embryonic development and that homozygous deletion results in embryonic lethality [43]. Therefore, current in vivo studies of RNF20 predominantly utilize heterozygous knockout mice [44]. This limitation greatly restricts our ability to precisely investigate the regulatory functions of RNF20. However, we also observed impaired basal and inducible expression of RIG-I and MDA5 in Rnf20+/- mice, which was consistent with the findings obtained from RNF20-deficient cells. Moreover, Rnf20+/- mice were more susceptible to infection with VSV-GFP and H1N1 viruses, exhibiting higher mortality rates. These findings further indicate that RNF20 is indispensable for maintaining effective antiviral innate immunity.
To elucidate the potential mechanism underlying the defective expression of RIG-I and MDA5 caused by RNF20 knockout, we constructed cellular models in which RNF20 was restricted to either the nucleus or the cytoplasm. The results showed that nuclear localization of RNF20 alone was sufficient to restore RIG-I and MDA5 expression. Furthermore, mass spectrometry-based identification of RNF20-interacting proteins revealed that RNF20 participates in biological processes related to gene transcription and translation. Previous studies have demonstrated that RNF20 primarily regulates gene expression through H2B ubiquitination (H2Bub) [45]; however, other studies have also reported that RNF20 can bind to and repair damaged DNA [17,44], indicating that RNF20 has the ability to directly associate with genomic DNA. Thus, by constructing the promoters of RIG-I and MDA5, we found that RNF20 could indeed activate the promoters of RIG-I and MDA5, and chromatin immunoprecipitation further confirmed that RNF20 binds to their promoter regions. These results suggest that RNF20 directly participates in the transcriptional regulation of RIG-I and MDA5 expression.
Previous studies have elucidated the nuclear functions of RNF20 in inhibiting the expression of cancer-promoting genes by disrupting the interaction between transcription elongation factor A protein 1 (TFIIS) and RNA polymerase II-associated factor 1 homolog (PAF1) complexes, thereby hindering transcriptional elongation [46]. Therefore, we speculate that RNF20 regulates the expression of RIG-I and MDA5 through at least two steps. First, it mediates histone ubiquitination modification to expose the genomic locations of RIG-I and MDA5. Subsequently, RNF20 forms a transcription complex with transcription factors to initiate the expression of RIG-I and MDA5. However, the precise regulatory mechanism remains to be elucidated.
In conclusion, our study provides compelling evidence that RNF20 is an important regulator for the maintenance of innate immune homeostasis. Moreover, the regulatory mechanism of RNF20 is highly conserved among species, which means that RNF20 may represent a new target for regulating immune homeostasis. In addition, the cytoplasmic localization of RNF20 may also serve as a potential target for cancer prediction and immunotherapy.
Materials and methods
Ethics statement
All experiments involving animals were conducted according to the ethical policies and procedures approved by the Shanghai Shengchang Biotechnology Co, Shanghai, China (Approval no. 2024TEEA0200251).
Animal study
C57BL/6J WT (SM-001) and Rnf20+/- (Cat # NM-KO-2102810) mice were purchased from the Shanghai Model Organisms and maintained in our facilities. All the mice were on the C57BL/6 background and were maintained under specific pathogen-free conditions. Experiments were conducted with randomly chosen littermates of the same sex and matched by age and body weight. We conducted all animal care and experimentation according to the Association for Assessment and Accreditation of Laboratory Animal Care guidelines.
Cell Culture and Virus
HEK293T, A549, HeLa, and THP-1 cells were obtained from the American Type Culture Collection (ATCC, Manassas, VA, USA). THP-1 cells were cultured in RPMI-1640 (Gibco, Carlsbad, CA) medium, and HEK293T, A549, and HeLa cells were cultured in Dulbecco’s modified Eagle’s (DMEM), (Gibco, Carlsbad, CA) containing 10% FBS supplemented with 1% penicillin/streptomycin. Cells were maintained at 37 °C and 5% CO2, were propagated as described by ATCC guidelines. Multiple human cell lines were employed to ensure robustness and broad applicability of the findings: HEK293T cells for high-efficiency transfection in luciferase reporter and overexpression/ubiquitination assays; A549 lung epithelial cells as a physiologically relevant model for RNA virus infection and antiviral responses; HeLa cells for confocal microscopy and subcellular localization studies due to their favorable imaging properties; and THP-1 monocytic cells to extend observations to immune cells.
The A/WSN/1933 (H1N1), Herpes simplex virus (HSV), NDV, a low virulent strain of LaSota, AIV, was an A/Chicken/Shanghai/010/2008 (H9N2) (SH010) were stored in our Laboratory as described [47]. The GFP-tagged vesicular stomatitis virus (VSV) VSV-GFP was a gift by prof. Tao Sun, Shanghai Jiao Tong University, China (Shanghai, China). VSV-GFP (a negative-sense single-stranded RNA virus recognized primarily by RIG-I), H1N1 influenza A virus (an orthomyxovirus detected by RIG-I), and NDV (a paramyxovirus preferentially sensed by MDA5) were selected to represent a range of RNA viruses that differentially engage RIG-I and/or MDA5, thereby allowing evaluation of RNF20’s regulatory role across diverse RLR-mediated antiviral responses
Reagents and antibodies
Poly (I: C) was purchased from InvivoGen. MG132 and KPT330 were purchased from Selleck. The antibodies IgG were purchased from the indicated manufacturer’s, including HRP-conjugated anti-mouse or rabbit IgG (Cell Signaling Technologies, 7076 and 7074), Alexa Fluor 488-conjugated Goat anti-rabbit IgG (Sangon Biotech, D110061), Cy3-conjugated Goat anti-mouse IgG (Sangon Biotech, D110088), Anti-Flag tag mouse monoclonal antibody (Sangon Biotech, D191041), Anti-HA tag mouse monoclonal antibody (Abmart, 26D11), Anti-Myc tag mouse monoclonal antibody (Abmart, 19C2), Anti-GFP tag mouse monoclonal antibody (ABclonal,AE012), anti-β-tubulin (Abmart, 2H4), anti-Lamin B1 (Beyotime, AF1408), anti-RNF20 (proteintech, 21625–1-AP), anti-RNF20 (Santa Cruz biotechnology, sc-517358), anti-RIG-I (Cell Signaling Technologies, 3743), anti-MDA5 (Cell Signaling Technologies, 5321), anti-MAVS (Cell Signaling Technologies, 3993), anti-TBK1 (Cell Signaling Technologies, 3504), anti-p-TBK1 (Cell Signaling Technologies, 5483), anti-IRF3 (Cell Signaling Technologies, 11904), anti-p-IRF3 (Cell Signaling Technologies, 4947), anti-STING (Cell Signaling Technologies, 13647S), anti-STAT1 (Sangon Biotech, D120084).
Cell treatment
To inhibit proteasomal degradation, cells were treated with 10 mM MG132. To inhibit nucleocytoplasmic transport, cells were treated with 10 μM KPT330.
CRISPR-Cas9-mediated genome editing
All the specific gene knockout A549 cell lines were generated using CRISPR-Cas9 gene-editing technology. Guide sequences for RNF20, RIG-I, and MDA5 were cloned into the PX459 vector (obtained from Addgene). A549 cells were transfected with a PX459 vector and selected with puromycin (10 μg/mL) for 7 days. Cell clones were isolated, and the expression of the target gene was analyzed by Western blots analysis. The target sequences were as follows: RNF20, ACAGTGGAAACAATTAAGCT; RIG-I, CTTCTCAGGTCCCAAGTC; MDA5, CTTGGACATAACAGCAACAT.
Viral infection and tittering
The A549 cells were infected with VSV-GFP (MOI = 1.0), H1N1 (MOI = 1.0), NDV (MOI = 1.0), and HSV (MOI = 10). HeLa cells were infected with VSV-GFP (MOI = 1.0) or H1N1 (MOI = 1.0) for the indicated times. HEK293T cells were infected with VSV-GFP (MOI = 1.0) for the indicated times. For the TCID50 assay, 1 × 104 HEK293T cells were seeded in 96-well plates 1 day before measurement; cell culture supernatants with VSV-GFP or H1N1 infection were then serially diluted on a monolayer of HEK293T cells and cultured for 3–7 days, and the TCID50 was measured.
Plasmid construction and transfection
Standard molecular biology procedures were performed for all plasmid constructions. cDNA sequences were generated for RNF20 tagged with Flag, HA and EGFP; RIG-I and MDA5 tagged with Flag, HA and Myc; and MAVS, TBK1, STING tagged with HA; CRM1 tagged with Flag. The sequences were cloned by PCR from the cDNA of A549 cells and inserted into the pcDNA3.1 backbone. RIG-I and MDA5 mutants expression plasmid was cloned from pcDNA3.1-Myc-RIG-I and Myc-MDA5; RNF20 mutants expression plasmid was cloned from pcDNA3.1-RNF20-HA and pcDNA4.0-RNF20-EGFP. The nuclear localization signal (NLS, amino acids 961–975) and nuclear export signal (NES, amino acids 945–954) of human RNF20 were identified using NovoPro and LocNES prediction tools, respectively. RNF20-dNLS was generated by deleting residues 961–975, and RNF20-dNES by deleting residues 945–954. All the above-described constructs were verified by sequencing. The plasmids were transfected into HEK293T, A549, or HeLa cells with Gulen Transfection Reagent according to the manufacturer’s protocol. Poly (I: C) was transfected into HEK293T in a reagent according to a standard protocol.
RNA interference
The shRNAs targeting RNF20 were constructed by plasmid pGPU6/Neo transferred into cells followed by qPCR or immunoblot analysis. The shRNA sequences used in this study are as follows: shRNF20-1#: 5′-CTGCACGGGCCTTGGAAAA-3′; shRNF20-2#: 5′-GCTTTCTCTTTCCTTGCTACT-3
Nuclear and cytoplasmic fractionation
Cells were lysed in ice-cold cytoplasmic lysis buffer (Beyotime, P0027) supplemented with a protease inhibitor cocktail. After centrifugation at 1700g for 5 min, the supernatants were collected as cytoplasmic extracts. The pellets were washed four times with cytoplasmic lysis buffer. The washed cell pellets were then lysed in ice-cold nuclear lysis buffer supplemented with a protease inhibitor cocktail. After centrifugation at 14,000g for 10 min, the supernatants were collected as nuclear extracts and subjected to SDS-PAGE.
Ubiquitination assay
For analysis of the ubiquitination of RIG-I and MDA5 in HEK293T cells, HEK293T cells were transfected with plasmids expressing Myc-RIG-I or its mutants, or Myc-MDA5 or its mutants, HA-ubiquitin (WT), HA-ubiquitin (K6), HA-ubiquitin (K11), HA-ubiquitin (K27), HA-ubiquitin (K29), HA-ubiquitin (K33), HA-ubiquitin (K48), HA-ubiquitin (K63) and RNF20-Flag (WT) or its mutants, and then whole-cell extracts were immunoprecipitated with the Myc-specific antibody and analyzed by immunoblot with anti-HA.
RT-qPCR analysis
Total RNA was extracted with TROZL Reagent (Accurate Biology, Hunan, China). A reverse transcription system (Vazyme Biotech co.,ltd) was used to synthesize cDNA. ChamQ Blue Universal SYBR qPCR Master Mix (Vazyme Biotech co.,ltd) and ABI 7500 detection system were used for qRT-PCR. The mRNA results were normalized to β-actin expression. The sequences of primers used in this study were included in S2 Table.
RNA sequencing
Total RNA was extracted using the Total RNA Extractor kit (B511311, Sangon, China). The libraries were then quantified and pooled. Paired-end sequencing of the library was performed on the NovaSeq sequencers (Illumina, San Diego, CA) by Sangon Biotech (Shanghai, China).
Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis of differentially expressed genes
Functional enrichment analyses of KEGG were performed to identify which DEGs. The Kyoto Encyclopedia of Genes and Genomes (KEGG) database is a public database of pathway data, KEGG pathway analysis identifies significantly enriched metabolic pathways or signal transduction pathways enriched in DEGs compared to a reference gene background, using the hypergeometric test. GO terms and KEGG pathway with false discovery rate (q-value) < 0.05 were considered as significantly altered.
Immunofluorescence and confocal microscopy
HeLa cells were seeded in confocal dishes. After transfection or stimulation, the cells were fixed with 4% paraformaldehyde for 30 min, permeabilized for 10 min with 0.1% Triton X-100, blocked for 1.5 h with 5% bovine serum albumin (BSA) in PBS, and stained with specific antibodies. HeLa cells were stained with anti-RNF20, anti-RIG-I, anti-MDA5, or anti-NP antibodies followed by fluorescent-dye-conjugated secondary antibodies. The nuclei were stained with DAPI. Fluorescence in cells was visualized with an Olympus IX83 confocal microscope (Olympus Corp, Tokyo, Japan).
Immunoprecipitation and immunoblot analysis
For immunoblot, cells were collected and lysed with RIPA Lysis Buffer (Beyotime, Shanghai, China) supplemented with a protease inhibitor ‘cocktail’ (Yeasen, Shanghai, China). For immunoprecipitation (IP), whole-cell extracts were collected 36 h after transfection and were lysed in IP buffer (Beyotime) supplemented with a protease inhibitor cocktail (Yeasen). After centrifugation for 10 min at 14,000 × g. The lysate was supernatants collected and incubated with anti-Flag or anti-Myc affinity Magnetic Beads at 4 °C with pre-cooled rotors. After 6 h of incubation, beads were washed five times with TBST buffer. The sample was boiled with 1% (wt/vol) SDS sample buffer. For immunoblot analysis, immunoprecipitates or whole-cell lysates were loaded and subjected to SDS-PAGE, transferred into nitrocellulose membranes, and then blotted with indicated Antibodies.
Image quantification methods
Fluorescence intensity and colocalization were quantified using Fiji/ImageJ software (version 2.1.0). For mean fluorescence intensity of VSV-GFP (e.g., Figs 1J and 2H), ≥ 50 cells per condition from at least 5 random fields were analyzed across three independent experiments using the ‘Measure’ function after background subtraction. Nuclear/cytoplasmic distribution of RNF20 was quantified by measuring integrated density in manually outlined nuclear and cytoplasmic regions (defined by DAPI staining).
Chromatin immunoprecipitation (ChIP)
Chromatin immunoprecipitation (ChIP) assays were performed using the ChIP Assay Kit (P2083S, Beyotime) according to the manufacturer’s instructions. DNA fragments were amplified using 2 × Phanta Flash Master Mix (Dye Plus) (Vazyme), and DNA enrichment was evaluated by real-time PCR (RT-PCR) with 2 × Taq Master Mix (Vazyme). The primer sequences used for the ChIP assay are listed in the S2 Table.
Luciferase assay
HEK293T cells were seeded in 24-well plates and co-transfected with IFN-β luc, pRL-TK, and other expression vectors where indicated. Luciferase activity was measured with the Dual-Luciferase Reporter Assay system according to the manufacturer‘s instructions (Promega). Data were normalized for transfection efficiency by calculating the ratio between firefly luciferase activity and Renilla luciferase activity.
Statistical analysis
All data are shown as mean ± SD, all experiments were repeated at least of three or more times. Statistical significance between different groups was calculated by two-tailed unpaired Student’s t-test or two-way ANOVA test. GraphPad Prism 8.0 was utilized to graph the results. Probability (p) values of < 0.05 were considered significant: *p < 0.05, **p < 0.01, ***p < 0.001; ns, not significant.
Supporting information
S1 Fig. RNF20 regulates antiviral innate immunity.
(A and B) THP-1 cells (A) or Hela cells (B) were infected with VSV-GFP for 24h, qPCR analysis of IFN-β mRNA expression. (C and D) qPCR analysis of (C) RNF20 (D) IFN-β mRNA levels in A549 cells transiently transfected with RNF20 expression plasmid or control plasmid, uninfected (UI) or infected with VSV-GFP for 24h. n = 3; Data are expressed as mean ± SD; *p < 0.05, **p < 0.01, ***p < 0.001; The results were repeated in three independent experiments with three replicas each.
https://doi.org/10.1371/journal.ppat.1013890.s001
(TIF)
S2 Fig. RNF20 negatively regulates RLR-mediated innate immune responses.
(A) HEK293T cells were transiently transfected with IFN-β-luc, pRL-TK, poly(I:C) and MDA5 along with RNF20-interfering plasmid. (B, C) Immunoblot analysis of anti-Flag and β-tubulin in HEK293T cells that were transfected with (B) RIG-I-Flag or (C) MDA5-Flag together with shNC or shRNF20–1# or shRNF20–2# for 24h. (D) Quantification of RIG-I and MDA5 protein expression levels shown in (B) and (C). (E, F) A549 cells were transiently transfected with RIG-I and MDA5 knockout plasmids (PX459), immunoblot and quantification analysis of RIG-I, MDA5. (G) Fluorescence microscopy imaging of the VSV-GFP in the RIG-I and MDA5 double knockout A549 cells transiently transfected with RNF20 expression plasmid infection with VSV-GFP (MOI = 1.0) for 24h. (H) Quantitative mean fluorescence intensity from (G). (I, J) RIG-I/MDA5 double-knockout A549 cells were transiently transfected with RNF20 expression plasmid infected VSV-GFP (MOI = 1.0) for 24h. Immunoblot and quantification analysis of total and phosphorylated (p-) TBK1, and RIG-I, MDA5, and β-tubulin. (K) qPCR analysis of IFN-β mRNA levels in RIG-I/MDA5 double-knockdown A549 cells transiently transfected with the RNF20 expression plasmid and infected with VSV-GFP (MOI = 1.0) for 24 h. (L-O) Immunoblot and quantitative analysis of anti-Myc and β-tubulin in HEK293T cells that were transfected with (L) Myc-RIG-I or its mutants, (N) Myc-MDA5 or its mutants, together with RNF20 or control plasmid. Data are expressed as mean ± SD; ns: no significance; *p < 0.05, **p < 0.01, ***p < 0.001; Scale bars = 50 µm. The results were repeated in three independent experiments with three replicas each.
https://doi.org/10.1371/journal.ppat.1013890.s002
(TIF)
S3 Fig. RNF20 mediates the ubiquitination of conserved lysine residues in RLR.
(A, B) Quantification of RIG-I and MDA5 protein levels corresponding to Fig 5C and 5D. (C, D) Co-immunoprecipitation analysis of polyubiquitination of porcine RIG-I or MDA5 (wild-type or K-to-R mutants) in HEK293T cells co-transfected with Myc-RIG-I or Myc-MDA5, His-RNF20, and HA-ubiquitin expression plasmids. (E, F) Quantification of ubiquitination levels of porcine RIG-I and MDA5 in immunoprecipitated samples. (G, H) Co-immunoprecipitation and quantification analysis of polyubiquitination of chicken MDA5 (wild-type or K44R, K172R mutants) in HEK293T cells co-transfected with Myc-MDA5, RNF20-Flag, and HA-ubiquitin expression plasmids. (I, J) Immunoblot and quantitative analysis of anti-Myc in HEK293T cells transiently co-transfected with RNF20-Flag and MDA5-K44R or MDA5-K172R expression plasmids, followed by infection with VSV-GFP (MOI = 1.0) for 24 h. (K, L) IFN-β luciferase reporter assay in HEK293T cells transiently co-transfected with human RNF20-Flag and Myc-RIG-I or MDA5 (wild-type or mutant) expression plasmids, followed by VSV-GFP infection (MOI = 1.0) for 12 h. Data are presented as mean ± SD. ns, not significant; *p < 0.05, **p < 0.01, ***p < 0.001. The results were repeated in three independent experiments with three replicas each.
https://doi.org/10.1371/journal.ppat.1013890.s003
(TIF)
S4 Fig. Viral replication does not affect RNF20 expression.
(A) A549 cells were infected with NDV, H9N2, VSV-GFP, or HSV for 24 h, followed by immunoblot analysis of total and phosphorylated (p-) TBK1, RNF20, RIG-I, MDA5, MAVS, STING, and β-tubulin expression. (B, C) A549 cells were either uninfected (UI) or infected with VSV-GFP or NDV for 12 h, followed by qPCR analysis of (B) RNF20 and (C) IFN-β mRNA expression. (D, E) A549 cells were either uninfected (UI) or infected with VSV-GFP or NDV for 24 h, followed by qPCR analysis of (D) RNF20 and (E) IFN-β mRNA expression. (F) Online analysis of RNF20 expression based on single-cell transcriptomic data from SARS-CoV-1/2-infected human cell lines (https://www.ebi.ac.uk/gxa/sc/home). n = 3. Data are presented as mean ± SD. ns, not significant; *p < 0.05, **p < 0.01, ***p < 0.001. The results were repeated in three independent experiments with three replicas each.
https://doi.org/10.1371/journal.ppat.1013890.s004
(TIF)
S5 Fig. RNF20 knockout leads to immune suppression in vivo and in vitro.
(A) Genotyping analysis of Rnf20 knockout mice. (B-F) qPCR analysis of IFN-β, MX1, TNF-α, and IL-6 mRNA levels in the spleen and lung of wild-type (WT) and Rnf20+/- mice following intraperitoneal injection of VSV-GFP for 18 h. (G) RNA-seq analysis of wild-type and RNF20-/- A549 cells. Heatmap showing transcriptome-wide effects of RNF20 deletion, with significantly (padj ≤ 0.05) upregulated (log2FC ≥ 1, red) or downregulated (log2FC ≤ −1, blue) genes. (H) Volcano plot showing transcriptome-wide effects of RNF20 deletion, with significantly (padj ≤ 0.05) upregulated (log2FC ≥ 1, red) or downregulated (log2FC ≤ −1, blue) genes. (I) qPCR analysis of IFN-β, MDA5, and RIG-I mRNA levels in wild-type and RNF20-/- A549 cells, either uninfected or infected with VSV-GFP for 24 h. (J, K) Wild-type and RNF20-/- A549 cells were either uninfected (UI) or infected with VSV-GFP for 24 h, followed by immunoblot and quantitative analysis of STAT1, phosphorylated TBK1 (p-TBK1), IRF3, IRF7, and β-tubulin. n = 3; Data are presented as means ± SD. ns, not significant; *p < 0.05, **p < 0.01, ***p < 0.001. The results were repeated in three independent experiments with three replicas each.
https://doi.org/10.1371/journal.ppat.1013890.s005
(TIF)
S6 Fig. RNF20 regulates the transcription of RIG-I and MDA5.
(A, B) qPCR analysis of RIG-I and MDA5 mRNA levels in RNF20-/- A549 cells transiently transfected with RNF20 wild-type (WT), RNF20-dNLS, or RNF20-dNES expression plasmids, followed by infection with VSV-GFP (MOI = 1.0) for 24 h. (C) Schematic representation of primer design used for the amplification of RIG-I and MDA5 promoter regions. (D) Chromatin immunoprecipitation (ChIP) was performed using RNF20-specific antibodies, and the precipitated DNA was used as a template for PCR amplification of RIG-I and MDA5 promoter regions, with GAPDH serving as an internal control. n = 3; Data are presented as means ± SD. ns, not significant; *p < 0.05, **p < 0.01, ***p < 0.001. The results were repeated in three independent experiments with three replicas each.
https://doi.org/10.1371/journal.ppat.1013890.s006
(TIF)
S1 Table. Listing of proteins interacting with RNF20 identified by mass spectrometry analysis and KEGG and GO analyses of RNF20-interacting proteins.
https://doi.org/10.1371/journal.ppat.1013890.s007
(XLSX)
S2 Table. Listing of PCR and q-PCR primer sequences used in this study.
https://doi.org/10.1371/journal.ppat.1013890.s008
(DOCX)
Acknowledgments
We thank Professor Tao Sun from Shanghai Jiao Tong University for the gift of the VSV-GFP virus.
References
- 1. Onomoto K, Onoguchi K, Yoneyama M. Regulation of RIG-I-like receptor-mediated signaling: interaction between host and viral factors. Cell Mol Immunol. 2021;18(3):539–55. pmid:33462384
- 2. Taguchi T, Mukai K. Innate immunity signalling and membrane trafficking. Curr Opin Cell Biol. 2019;59:1–7. pmid:30875551
- 3. He Q-Q, Huang Y, Nie L, Ren S, Xu G, Deng F, et al. MAVS integrates glucose metabolism and RIG-I-like receptor signaling. Nat Commun. 2023;14(1):5343. pmid:37660168
- 4. van Huizen M, Gack MU. The RIG-I-like receptor family of immune proteins. Mol Cell. 2025;85(20):3793–806. pmid:41106369
- 5. Thoresen D, Wang W, Galls D, Guo R, Xu L, Pyle AM. The molecular mechanism of RIG-I activation and signaling. Immunol Rev. 2021;304(1):154–68. pmid:34514601
- 6. Lu H, Lu N, Weng L, Yuan B, Liu Y-J, Zhang Z. DHX15 senses double-stranded RNA in myeloid dendritic cells. J Immunol. 2014;193(3):1364–72. pmid:24990078
- 7. van Gent M, Chiang JJ, Muppala S, Chiang C, Azab W, Kattenhorn L, et al. The US3 Kinase of Herpes Simplex Virus Phosphorylates the RNA Sensor RIG-I To Suppress Innate Immunity. J Virol. 2022;96(4):e0151021. pmid:34935440
- 8. Kato H, Takeuchi O, Sato S, Yoneyama M, Yamamoto M, Matsui K, et al. Differential roles of MDA5 and RIG-I helicases in the recognition of RNA viruses. Nature. 2006;441(7089):101–5. pmid:16625202
- 9. Hou J, Zhou Y, Zheng Y, Fan J, Zhou W, Ng IOL, et al. Hepatic RIG-I predicts survival and interferon-α therapeutic response in hepatocellular carcinoma. Cancer Cell. 2014;25(1):49–63. pmid:24360797
- 10. Rehwinkel J, Gack MU. RIG-I-like receptors: their regulation and roles in RNA sensing. Nat Rev Immunol. 2020;20(9):537–51. pmid:32203325
- 11. Jiang L-Q, Xia T, Hu Y-H, Sun M-S, Yan S, Lei C-Q, et al. IFITM3 inhibits virus-triggered induction of type I interferon by mediating autophagosome-dependent degradation of IRF3. Cell Mol Immunol. 2018;15(9):858–67. pmid:28435159
- 12. Gack MU, Shin YC, Joo C-H, Urano T, Liang C, Sun L, et al. TRIM25 RING-finger E3 ubiquitin ligase is essential for RIG-I-mediated antiviral activity. Nature. 2007;446(7138):916–20. pmid:17392790
- 13. Lang X, Tang T, Jin T, Ding C, Zhou R, Jiang W. TRIM65-catalized ubiquitination is essential for MDA5-mediated antiviral innate immunity. J Exp Med. 2017;214(2):459–73. pmid:28031478
- 14. Chauhan S, Jena KK, Mehto S, Chauhan NR, Sahu R, Dhar K, et al. Innate immunity and inflammophagy: balancing the defence and immune homeostasis. FEBS J. 2022;289(14):4112–31. pmid:34826185
- 15. Foglizzo M, Middleton AJ, Day CL. Structure and Function of the RING Domains of RNF20 and RNF40, Dimeric E3 Ligases that Monoubiquitylate Histone H2B. J Mol Biol. 2016;428(20):4073–86. pmid:27569044
- 16. Wang E, Kawaoka S, Yu M, Shi J, Ni T, Yang W, et al. Histone H2B ubiquitin ligase RNF20 is required for MLL-rearranged leukemia. Proc Natl Acad Sci U S A. 2013;110(10):3901–6. pmid:23412334
- 17. Li J, Zhao J, Gan X, Wang Y, Jiang D, Chen L, et al. The RPA-RNF20-SNF2H cascade promotes proper chromosome segregation and homologous recombination repair. Proc Natl Acad Sci U S A. 2023;120(20):e2303479120. pmid:37155876
- 18. So CC, Ramachandran S, Martin A. E3 Ubiquitin Ligases RNF20 and RNF40 Are Required for Double-Stranded Break (DSB) Repair: Evidence for Monoubiquitination of Histone H2B Lysine 120 as a Novel Axis of DSB Signaling and Repair. Mol Cell Biol. 2019;39(8):e00488-18. pmid:30692271
- 19. Wang J, Liu Q, Zhang K, Zhao S, Shao Q, Fu F, et al. RNF20 dual regulation of MDA5 signaling to maintain immune homeostasis in chickens. J Virol. 2025;99(3):e0200824. pmid:39998124
- 20. Dong B, Wang X, Song X, Wang J, Liu X, Yu Z, et al. RNF20 contributes to epigenetic immunosuppression through CDK9-dependent LSD1 stabilization. Proc Natl Acad Sci U S A. 2024;121(7):e2307150121. pmid:38315842
- 21. Kuniyoshi K, Takeuchi O, Pandey S, Satoh T, Iwasaki H, Akira S, et al. Pivotal role of RNA-binding E3 ubiquitin ligase MEX3C in RIG-I-mediated antiviral innate immunity. Proc Natl Acad Sci U S A. 2014;111(15):5646–51. pmid:24706898
- 22. Ishizawa J, Kojima K, Hail N Jr, Tabe Y, Andreeff M. Expression, function, and targeting of the nuclear exporter chromosome region maintenance 1 (CRM1) protein. Pharmacol Ther. 2015;153:25–35. pmid:26048327
- 23. Li D, Fu S, Wu Z, Yang W, Ru Y, Shu H, et al. DDX56 inhibits type I interferon by disrupting assembly of IRF3-IPO5 to inhibit IRF3 nucleus import. J Cell Sci. 2019;133(5):jcs230409. pmid:31340999
- 24. Nakamura K, Kato A, Kobayashi J, Yanagihara H, Sakamoto S, Oliveira DVNP, et al. Regulation of homologous recombination by RNF20-dependent H2B ubiquitination. Mol Cell. 2011;41(5):515–28. pmid:21362548
- 25. Shrikondawar AN, Chennoju K, Ghosh DK, Ranjan A. Identification and functional characterization of the nuclear and nucleolar localization signals in the intrinsically disordered region of nucleomethylin. J Cell Physiol. 2024;239(12):e31433. pmid:39245872
- 26. Loo Y-M, Gale M Jr. Immune signaling by RIG-I-like receptors. Immunity. 2011;34(5):680–92. pmid:21616437
- 27. Iurescia S, Fioretti D, Rinaldi M. The Innate Immune Signalling Pathways: Turning RIG-I Sensor Activation Against Cancer. Cancers (Basel). 2020;12(11):3158. pmid:33121210
- 28. Quicke KM, Diamond MS, Suthar MS. Negative regulators of the RIG-I-like receptor signaling pathway. Eur J Immunol. 2017;47(4):615–28. pmid:28295214
- 29. Esser-Nobis K, Hatfield LD, Gale M Jr. Spatiotemporal dynamics of innate immune signaling via RIG-I-like receptors. Proc Natl Acad Sci U S A. 2020;117(27):15778–88. pmid:32571931
- 30. Rex V, Stempel M, Halle S, Brinkmann MM. The two faces of oligoadenylate synthetase-like: effective antiviral protein and negative regulator of innate immunity. Curr Opin Virol. 2023;60:101329. pmid:37079941
- 31. Zhao C, Jia M, Song H, Yu Z, Wang W, Li Q, et al. The E3 Ubiquitin Ligase TRIM40 Attenuates Antiviral Immune Responses by Targeting MDA5 and RIG-I. Cell Rep. 2017; 21: 1613–23. pmid:29117565
- 32. Wang S, Yang Y-K, Chen T, Zhang H, Yang W-W, Song S-S, et al. RNF123 has an E3 ligase-independent function in RIG-I-like receptor-mediated antiviral signaling. EMBO Rep. 2016;17(8):1155–68. pmid:27312109
- 33. Zhang Z, Mena EL, Timms RT, Koren I, Elledge SJ. Degrons: defining the rules of protein degradation. Nat Rev Mol Cell Biol. 2025;26(11):868–83. pmid:40659789
- 34. Varshavsky A. N-degron and C-degron pathways of protein degradation. Proc Natl Acad Sci U S A. 2019;116(2):358–66. pmid:30622213
- 35. Zhang F, Paramasivam M, Cai Q, Dai X, Wang P, Lin K, et al. Arsenite binds to the RING finger domains of RNF20-RNF40 histone E3 ubiquitin ligase and inhibits DNA double-strand break repair. J Am Chem Soc. 2014;136(37):12884–7. pmid:25170678
- 36. Zhang X, Simon GM, Cravatt BF. Implications of frequent hitter E3 ligases in targeted protein degradation screens. Nat Chem Biol. 2025;21(4):474–81. pmid:39870762
- 37. Liu G, Li J, He B, Yan J, Zhao J, Wang X, et al. Bre1/RNF20 promotes Rad51-mediated strand exchange and antagonizes the Srs2/FBH1 helicases. Nat Commun. 2023;14(1):3024. pmid:37230987
- 38. Chen Q, Yang P. Coiled-coil-domain-mediated TRIM E3 protein condensation modulates enzymatic activity and functional specificity. Cell Rep. 2025;44(6):115766. pmid:40450692
- 39. Tarcic O, Pateras IS, Cooks T, Shema E, Kanterman J, Ashkenazi H, et al. RNF20 Links Histone H2B Ubiquitylation with Inflammation and Inflammation-Associated Cancer. Cell Rep. 2016;14(6):1462–76. pmid:26854224
- 40. Ishino Y, Hayashi Y, Naruse M, Tomita K, Sanbo M, Fuchigami T, et al. Bre1a, a histone H2B ubiquitin ligase, regulates the cell cycle and differentiation of neural precursor cells. J Neurosci. 2014;34(8):3067–78. pmid:24553946
- 41. Liang X, Tao C, Pan J, Zhang L, Liu L, Zhao Y, et al. Rnf20 deficiency in adipocyte impairs adipose tissue development and thermogenesis. Protein Cell. 2021;12(6):475–92. pmid:32797353
- 42. Wang G, Zhai Y-J, Xue Z-Z, Xu Y-Y. Improving Protein Subcellular Location Classification by Incorporating Three-Dimensional Structure Information. Biomolecules. 2021;11(11):1607. pmid:34827605
- 43. Tetik-Elsherbiny N, Elsherbiny A, Setya A, Gahn J, Tang Y, Gupta P, et al. RNF20-mediated transcriptional pausing and VEGFA splicing orchestrate vessel growth. Nat Cardiovasc Res. 2024;3(10):1199–216. pmid:39322771
- 44. Liu H, Tang Y, Singh A, Vong J, Cordero J, Mathes A, et al. RNF20 links the DNA damage response and metabolic rewiring in lung cancer through HIF1α. Nat Commun. 2025;16(1):4929. pmid:40436847
- 45. Bhattacharya D, Dwivedi HK, Nagaraju G. RNF20-mediated H2B monoubiquitination protects stalled forks from degradation and promotes fork restart. EMBO Rep. 2025;26(15):3773–803. pmid:40495033
- 46. Shema E, Kim J, Roeder RG, Oren M. RNF20 inhibits TFIIS-facilitated transcriptional elongation to suppress pro-oncogenic gene expression. Mol Cell. 2011;42(4):477–88. pmid:21596312
- 47. Cheng Y, Huang Q, Ji W, Du B, Fu Q, An H, et al. Muscovy duck retinoic acid-induced gene I (MdRIG-I) functions in innate immunity against H9N2 avian influenza viruses (AIV) infections. Vet Immunol Immunopathol. 2015;163(3–4):183–93. pmid:25604129