Abstract
RIG-I is a pivotal cytoplasmic sensor that recognizes different species of viral RNAs. This recognition leads to activation of the transcription factors NF-κB and IRF3, which collaborate to induce type I interferons (IFNs) and innate antiviral response. In this study, we identified the TRIM family protein TRIM4 as a positive regulator of RIG-I-mediated IFN induction. Overexpression of TRIM4 potentiated virus-triggered activation of IRF3 and NF-κB, as well as IFN-β induction, whereas knockdown of TRIM4 had opposite effects. Mechanistically, TRIM4 associates with RIG-I and targets it for K63-linked polyubiquitination. Our findings demonstrate that TRIM4 is an important regulator of the virus-induced IFN induction pathways by mediating RIG-I for K63-linked ubiquitination.
Introduction
The innate antiviral immunity represents the first line of host defense against viral infection. Upon detection of viral infection, the host cells activate a series of signaling events that lead to induction of type I interferons (IFNs), including IFN-β and IFN-α family cytokines, and proinflammatory cytokines (Akira et al., 2006; Takeuchi and Akira, 2010). Type I IFNs further induce the expression of downstream proteins, which mediate innate immune responses, such as suppression of viral replication, clearance of virus-infected cells, and facilitation of adaptive immune response (Akira et al., 2006; Honda et al., 2006; Hiscott, 2007).
Structurally conserved microbial components called pathogen-associated molecular patterns (PAMPs), including nucleic acids, proteins, and replicative intermediates of the microbials, are recognized by pattern recognition receptors (PRRs) (Ronald and Beutler, 2010; Takeuchi and Akira, 2010; Barbalat et al., 2011). Among the PRRs, Toll-like receptor 3 (TLR3), LSm14A, and RIG-I-like receptors (RLRs), including RIG-I and MDA5, play important roles in sensing viral RNAs (Akira et al., 2006; Li et al., 2012). TLR3 acts at the endosomes and is highly expressed in certain immune cells. LSm14A is localized at the P-bodies and translocated to the peroxisomes upon viral infection. It has been demonstrated that LSm14A functions as a sensor for viral RNA and DNA at the early phase of viral infection (Li et al., 2012). So far, the mostly studied viral RNA sensors are RIG-I and MDA5. Both RIG-I and MDA5 are lowly expressed in the cytoplasm of most cell types and induced by viral infection. Cells lacking RIG-I and MDA5 are deficient in producing proper type I IFNs in response to various RNA viruses (Kato et al., 2005, 2006). RIG-I and MDA5 are structurally related. Both proteins contain two N-terminal CARD modules and a C-terminal DexD/H-box RNA helicase domain (Andrejeva et al., 2004). Upon recognition of viral RNAs through their C-terminal RNA helicase domains, RIG-I and MDA5 undergo conformational changes and are recruited to the downstream mitochondria-associated adaptor protein VISA (also called MAVS, IPS-1, or Cardif) through their respective CARD domains (Kawai et al., 2005; Meylan et al., 2005; Seth et al., 2005; Xu et al., 2005). VISA acts as a central platform and activates downstream TAK1-IKKβ and TBK1/IKKε kinases, leading to activation of NF-κB and IRF3 and induction of type I IFNs (Fitzgerald et al., 2003). Although RIG-I and MDA5 share similar signaling features and structural homology (Yoneyama et al., 2004), various studies have demonstrated that the two helicases may discriminate among different ligands to trigger the innate immune response to RNA viruses. It has been suggested that RIG-I may preferably recognize viral single-strand (ss) RNA bearing 5′-triphosphate and short double-strand (ds) RNA, whereas MDA5 mostly detect long viral dsRNA (Gitlin et al., 2006; Kato et al., 2006, 2008).
Several studies have suggested critical roles for post-translational modifications, particularly ubiquitination, in regulation of RIG-I activity. It has been shown that the E3 ubiquitin ligases TRIM25 and Riplet/REUL catalyze K63-linked ubiquitination of RIG-I and positively regulate RIG-I-mediated signaling (Gack, 2007; Gao et al., 2009; Oshiumi et al., 2009). In contrast, the E3 ubiquitin ligase RNF125 mediates Lys-48-linked ubiquitination and destabilization of RIG-I and negatively regulates RIG-I-mediated signaling (Arimoto et al., 2007).
The tripartite motif containing (TRIM) proteins are a family of proteins that have been implicated in many biological processes including cell differentiation, apoptosis, and transcriptional regulation. The E3 ligase activity of many TRIMs has been linked to their antiviral function as well as to their ability to regulate immune signaling (McNab et al., 2011). For example, TRIM8 upregulates TNFα- and IL-1β-triggered NF-κB activation by targeting TAK1 for K63-linked polyubiquitination (Li et al., 2011), whereas TRIM21 has been shown to regulate IRF3 ubiquitination and type I IFN induction (Yang et al., 2009; Li et al., 2011). In the present study, we identified TRIM4 as a positive regulator of virus-induced IFN induction. Furthermore, TRIM4 specifically interacted with and targeted RIG-I for K63-linked polyubiquitination, which was essential for RIG-I-triggered IFN induction. These findings suggest that TRIM4 regulates virus-induced IFN induction and cellular antiviral innate immunity by targeting RIG-I for K63-linked polyubiquitination.
Results
Identification of TRIM4 as a positive regulator of virus-induced signaling
Ubiquitination has emerged as a critical regulatory mechanism for virus-triggered innate immune response. To identify additional regulators involved in virus-induced type I IFN induction pathways, we screened a human cDNA array containing 352 expression clones of ubiquitin-related enzymes by IFN-β promoter reporter assays. These experiments led to the identification of TRIM4, a member of the TRIM family. In reporter assays, overexpression of TRIM4 activated the IFN-β promoter in a dose-dependent manner in HEK293 cells, and potentiated Sendai virus (SeV)-induced activation of the IFN-β promoter (Figure 1A). Moreover, quantitative real-time PCR (qPCR) experiments also indicated that overexpression of TRIM4 could induce as well as potentiate SeV-induced transcription of downstream genes, including IFNB1, CCL5, and ISG56 (Figure 1B). The induction of type I IFNs requires coordinative and cooperative activation of the transcription factors IRF3 and NF-κB. Consistently, overexpression of TRIM4 activated ISRE in a dose-dependent manner in HEK293 cells and potentiated SeV-triggered activation of the ISRE (Figure 1C). Similarly, overexpression of TRIM4 also activated NF-κB and enhanced SeV-induced NF-κB activation (Figure 1D). However, TRIM4 had no marked effects on IFN-γ-induced activation of the IRF1 promoter (Figure 1E). These results suggest that TRIM4 positively regulates virus-induced type I IFN induction and transcription of downstream genes.
TRIM4 activates virus-induced signaling. (A) Overexpression of TRIM4 activates the IFN-β promoter and potentiates SeV-induced activation of the IFN-β promoter in a dose-dependent manner. HEK293 cells (1 × 105) were transfected with the IFN-β promoter reporter plasmids (0.1 μg) and the indicated amounts of TRIM4 expression plasmid. Eighteen hours after transfection, cells were left uninfected (left histograph) or infected with SeV (right histograph) for 12 h before luciferase assays were performed. (B) Overexpression of TRIM4 enhances SeV-induced transcription of endogenous IFNB1, CCL5, and ISG56 genes. HEK293 cells (2 × 105) were transfected with an empty vector or TRIM4 expression plasmid (0.1 μg). Twenty hours after transfection, cells were left uninfected or infected with SeV for 10 h before qPCR was performed. Data shown are presented as mean ± SD, n = 3. **P < 0.01, *P < 0.05. (C) Overexpression of TRIM4 activates ISRE and potentiates SeV-induced ISRE activation in a dose-dependent manner. Reporter assays were performed similarly as in A except that the ISRE reporter plasmid was used. (D) Overexpression of TRIM4 activates NF-κB and enhances SeV-induced NF-κB activation in a dose-dependent manner. Reporter assays were performed similarly as in A except that the NF-κB reporter plasmid was used. (E) Overexpression of TRIM4 does not affect IFN-γ-induced IRF1 promoter activation. HEK293 cells (1 × 105) were transfected with IRF1 promoter reporter plasmid (0.05 μg) and a control or TRIM4 expression plasmid (0.1 μg each). Twenty hours later, cells were treated with IFN-γ (100 ng/ml) for 10 h before luciferase assays were performed.
Knockdown of TRIM4 inhibits activation of virus-triggered IFN signaling
To determine the roles of endogenous TRIM4, we investigated the effects of knockdown of TRIM4 on virus-triggered IFN signaling. To do this, we made three TRIM4-RNAi constructs that target different sites of TRIM4 mRNA. Immunoblot analysis indicated that all the three RNAi plasmids could efficiently inhibit the expression of both transfected and endogenous TRIM4 in HEK293 cells (Figure 2A). Reporter assays showed that knockdown of TRIM4 markedly inhibited SeV-triggered activation of the IFN-β promoter (Figure 2B). We selected the #2 TRIM4-RNAi plasmid for additional experiments described below, and similar results were obtained with the #1 and #3 plasmids. Knockdown of TRIM4 also inhibited SeV-induced activation of the IFN-β promoter in HeLa and A549 cells (Figure 2C), suggesting that the function of TRIM4 in virus-triggered IFN induction is not cell specific. Consistently, qPCR experiments indicated that knockdown of TRIM4 inhibited SeV-induced transcription of downstream genes such as IFNB1 and CCL5 in HEK293 cells (Figure 2D). Collectively, these data suggest that TRIM4 is required for efficient induction of type I IFNs and downstream genes upon viral infection. Furthermore, knockdown of TRIM4 also inhibited SeV-triggered ISRE (Figure 2E) and NF-κB (Figure 2F) activation, suggesting that TRIM4 is required for virus-triggered activation of both IRF3 and NF-κB pathways. However, knockdown of TRIM4 could not inhibit TNFα- and IL-1β-induced NF-κB activation (Figure 2G) or IFN-γ-induced activation of the IRF1 promoter (Figure 2H). These results suggest that TRIM4 specifically regulates SeV-triggered signaling.
Knockdown of TRIM4 inhibits SeV-induced signaling. (A) Effects of TRIM4-RNAi plasmids on the expression of transfected and endogenous TRIM4. In the upper panels, HEK293 cells (2 × 105) were transfected with expression plasmids for Flag-TRIM4 and Flag-V59 (0.1 μg each) and the indicated RNAi plasmids (2 μg). Twenty-four hours after transfection, cell lysates were analyzed by immunoblot with anti-Flag. In the bottom panels, HEK293 cells (2 × 105) were transfected with control or the indicated TRIM4-RNAi plasmids (2 μg each) for 24 h. Cell lysates were then analyzed by immunoblots with the indicated antibodies. (B) Effects of TRIM4-RNAi plasmids on SeV-induced activation of the IFN-β promoter. HEK293 cells (1 × 105) were transfected with the indicated TRIM4-RNAi (0.5 μg) and the reporter (0.1 μg) plasmids. Twenty-four hours after transfection, cells were left uninfected or infected with SeV for 12 h before luciferase assays were performed. (C) Effects of TRIM4 knockdown on SeV-induced activation of the IFN-β promoter in HeLa and A549 cells. HeLa or A549 cells (1 × 105) were transfected with the indicated plasmids (0.5 μg each). Twenty hours after transfection, cells were infected with SeV or left uninfected for 16 h before reporter assays were performed. Data shown are presented as mean ± SD, n = 3. **P < 0.01, *P < 0.05. (D) Effects of TRIM4-RNAi on SeV-induced transcription of endogenous IFNB1 and CCL5 genes. HEK293 cells (2 × 105) were transfected with the indicated RNAi plasmids (2 μg). Thirty-six hours after transfection, cells were left uninfected or infected with SeV for 10 h before qPCR experiments were performed. Data shown are presented as mean ± SD, n = 3. **P < 0.01. (E and F) Effects of TRIM4-RNAi plasmids on SeV-induced activation of ISRE (E) and NF-κB (F). The experiments were similarly performed as in B. Data shown are presented as mean ± SD, n = 3. **P < 0.01. (G and H) Effects of TRIM4-RNAi on TNFα- and IL-1β-induced activation of the NF-κB (G) or IFN-γ-induced IRF1 promoter activation (H). The experiments were similarly performed as in B. (I) Overexpression of TRIM4 increases VSV replication. HEK293 cells (1 × 105) were transfected with the indicated expression plasmids (0.5 μg each). Eighteen hours later, cells were further transfected with poly(I:C) (1 μg) or left untransfected. Twenty-four hours after transfection, cells were infected with VSV (multiplicity of infection (MOI) at 0.1), and the supernatants were harvested at 24 h post-infection. Supernatants were analyzed for VSV titers with standard plaque assays. Data shown are presented as mean ± SD, n = 3. **P < 0.01. (J) Knockdown of TRIM4 inhibits VSV replication. Plaque assays were performed as in H except that a control or TRIM4 RNAi plasmid (1 μg) was transfected. Data shown are presented as mean ± SD, n = 3. *P < 0.05.
TRIM4 positively regulates cellular antiviral response
Because TRIM4 is critically involved in virus-triggered IFN-β induction, we investigated whether TRIM4 plays a role in cellular antiviral response. In plaque assays, we found that overexpression of TRIM4 mildly inhibited vesicular stomatitis virus (VSV) replication and markedly enhanced the inhibitory effect triggered by cytoplasmic poly(I:C) (Figure 2I). Conversely, knockdown of TRIM4 enhanced VSV replication and significantly reversed cytoplasmic poly(I:C)-mediated inhibition of VSV replication (Figure 2J). These data suggest that TRIM4 plays an important role in efficient cellular antiviral response.
Knockdown of TRIM4 inhibits RIG-I-mediated signaling
Various components are involved in virus-induced signaling. We next examined by which mediator that TRIM4 regulates virus-triggered signaling pathways. In reporter assays, knockdown of TRIM4 inhibited activation of the IFN-β promoter induced by overexpression of RIG-I, but not by MDA5, VISA, TBK1, or IRF3 (Figure 3A). These data suggest that TRIM4 regulates virus-induced signaling mediated by RIG-I.
TRIM4 targets and interacts with RIG-I. (A) TRIM4 inhibits RIG-I-CARD- but not MDA5-, VISA-, TBK1-, or IRF3-mediated activation of the IFN-β promoter. Control- or TRIM4-RNAi-transduced stable HEK293 cells (1 × 105) were transfected with an IFN-β promoter reporter (0.05 μg) and the indicated expression plasmids (0.1 μg each) for 20 h before luciferase assays were performed. Data shown are presented as mean ± SD, n = 3. **P < 0.01. (B) TRIM4 interacts with RIG-I in mammalian overexpression system. HEK293 cells (2 × 106) were transfected with the indicated plasmids (4 μg each). Coimmunoprecipitation and immunoblot analysis were performed with the indicated antibodies (upper panels). Expression of the transfected proteins was analyzed by immunoblots with anti-HA and anti-Flag (lower panels). (C) TRIM4 is colocalized with RIG-I in the cytosol. HEK293 (1 × 105) cells were transfected with cherry-TRIM4 and RIG-I-GFP for 18 h. The cells were left uninfected (upper panels) or infected with SeV (bottom panels) for 12 h before examined by immunofluorescent microscopy. (D) Endogenous association between TRIM4 and RIG-I. HEK293 cells were left uninfected or infected with SeV for the indicated periods. Immunoprecipitation and immunoblot analyses were performed with the indicated antibodies.
TRIM4 interacts with RIG-I
Since our results suggest that TRIM4 regulates virus-triggered signaling via RIG-I, we examined whether TRIM4 interacts with RIG-I. In transient transfection and coimmunoprecipitation experiments, TRIM4 interacted with full-length RIG-I as well as its CARD and helicase domains (Figure 3B). However, overexpression of TRIM4 did not interact with MDA5, TBK1, or IRF3 (Figure 3B). In confocal microscopy experiments, TRIM4 was colocalized with RIG-I in the cytosol before and after SeV infection (Figure 3C). We further performed endogenous coimmunoprecipitation experiments, and the results indicated that TRIM4 was constitutively associated with RIG-I in uninfected cells and this association was increased after infection with SeV (Figure 3D). These results suggest that SeV infection promotes endogenous association of TRIM4 with RIG-I.
TRIM4 triggers K63-linked polyubiquitination of RIG-I
It has been demonstrated that the TRIM family proteins function as E3 ubiquitin ligases for various cellular substrates. To examine whether TRIM4 is an E3 ubiquitin ligase for RIG-I, expression plasmid for HA-RIG-I was transfected together with Flag-TRIM4. Coimmunoprecipitation experiments showed that overexpression of TRIM4 markedly enhanced polyubiquitination of RIG-I (Figure 4A). In contrast, TRIM4 did not increase polyubiquitination of VISA or IRF3 (Figure 4A). In these experiments, an enzymatic inactive mutant TRIM4(C27S) failed to cause polyubiquitination of RIG-I (Figure 4A). In vitro ubiquitination assays confirmed that TRIM4 could ubiquitinate RIG-I with UbcH5b as an E2 ubiquitin ligase (Figure 4B). Taken together, these data suggest that TRIM4 mediates polyubiquitination of RIG-I.
TRIM4 mediates K63-linked polyubiquitination of RIG-I. (A) Wild-type TRIM4 but not TRIM4(C27S) enhances polyubiquitination of RIG-I in HEK293 cells. Cells (2 × 106) were transfected with the indicated plasmids. Twenty-four hours after transfection, cell lysates were immunoprecipitated with anti-HA. The immunoprecipitates were analyzed by immunoblots with anti-ubiquitin, anti-HA, or anti-Flag as indicated. (B) UbcH5b is an E2 enzyme for TRIM4-mediated RIG-I polyubiquitination in vitro. RIG-I and TRIM4 were translated in vitro, and Biotin-ubiquitin, E1, and the indicated E2s were added for ubiquitination assays. Ubiquitin-conjugated proteins were detected by immunoblot with HRP-streptavidin. The input levels of the translated proteins were determined by immunoblot analysis. (C) TRIM4 increased K63-linked but decreased K48-linked polyubiquitination of RIG-I. HEK293 cells were transfected with the indicated plasmids. Coimmunoprecipitation and immunoblots were performed with the indicated antibodies. (D) Effects of TRIM4-RNAi on SeV-induced polyubiquitination of endogenous RIG-I. HEK293 cells stably transduced with a control or TRIM4-RNAi were either uninfected or infected with SeV for the indicated periods. Cell lysates were immunoprecipitated with anti-RIG-I and the immunoprecipitates were analyzed by immunoblots with the indicated antibodies. (E) Effects of TRIM4(C27S) on SeV-induced IFN-β promoter activation. HEK293 cells (1 × 105) were transfected with the IFN-β promoter reporter plasmids (0.1 μg) and the indicated amounts of TRIM4(C27S) expression plasmid. Eighteen hours after transfection, cells were left uninfected or infected with SeV for 12 h before luciferase assays were performed.
It has been well established that K48-linked polyubiquitination often mediates degradation of the substrates, whereas K63-linked polyubiquitination regulates the localization and signaling activity of the substrates. Using ubiquitin mutants, we found that overexpression of TRIM4 dramatically increased K63-linked polyubiquitination of RIG-I and decreased K48-linked polyubiquitination of RIG-I (Figure 4C). We further found that knockdown of TRIM4 inhibited basal as well as SeV-induced K63-linked polyubiquitination of RIG-I, but increased K48-linked polyubiquitination (Figure 4D). These results suggest that TRIM4 mediates K63-linked polyubiquitination of RIG-I. Consistent with its role in TRIM4-mediated K63-linked polyubiquitination of RIG-I, the enzymatic inactive mutant TRIM4(C27S), acting as a dominate negative mutant, inhibited SeV-induced activation of the IFN-β promoter in a dose-dependent manner in HEK293 cells (Figure 4E).
TRIM4 targets RIG-I for ubiquitination mostly at K154, K164, and K172 residues
To map the residues of RIG-I that are ubiquitinated by TRIM4, we first determined which domains of RIG-I are ubiquitinated by TRIM4. As shown in Figure 5A, overexpression of TRIM4 markedly increased ubiquitination of full-length RIG-I and its N-terminal CARD domain, but not its C-terminal helicase domain. These results suggest that TRIM4 targets residues in the CARD domain of RIG-I for ubiquitination.
TRIM4 targets Lys 164 and 172 of RIG-I for polyubiquitination. (A) TRIM4 increases ubiquitination of the CARD domain of RIG-I. HEK293 cells were transfected with the indicated plasmids. Cell lysates were immunoprecipitated with anti-HA (αHA). The immunoprecipitates were analyzed by immunoblot with anti-Myc (αMyc). Expression of the transfected proteins in cell lysates was analyzed by immunoblots with the indicated antibodies. (B) TRIM4 enhances polyubiquitination of wild-type RIG-I but not its mutations in HEK293 cells. HEK293 cells (2 × 106) were transfected with the indicated plasmids. Twenty-four hours after transfection, cell lysates were immunoprecipitated with anti-RIG-I. The immunoprecipitates were analyzed by immunoblots with anti-ubiquitin, anti-HA, or anti-Flag as indicated. (C) Effects of K154R, K164R, and K172R mutants of RIG-I on SeV-induced IFN-β promoter activation. HEK293 cells (1 × 105) were transfected with the IFN-β promoter reporter plasmids (0.1 μg) and the indicated expression plasmids. Reporter assays were performed at 24 h after transfection.
To identify which residues of RIG-I are targeted by TRIM4, we mutated individually three lysine residues in the CARD domain of RIG-I to arginine. Recently, it has been demonstrated that K154, K164, and K172 are the essential sites for RIG-I-induced activation of the IFN-β promoter (Gack, 2007; Gao et al., 2009). Interestingly, mutation of K164 or K172 of RIG-I markedly reduced its ubiquitination mediated by TRIM4 in HEK293 cells, whereas mutation of K154 had a minor effect (Figure 5B). These data suggest that TRIM4 primarily targets Lys 154, 164, and 172 for ubiquitination of RIG-I.
We then tested the functions of mutants K154R, K164R, and K172R individually and in various combinations. As shown in Figure 5C, mutation of K154, K164, or K172 to arginine markedly reduced the ability of RIG-I to activate the IFN-β promoter, and the double mutants K154/164R, K154/172R, K164/172R, and the triple mutant K154R/K164R/K172R caused further loss of the ability to induce activation of the IFN-β promoter. In these experiments, wild-type TRIM4 but not TRIM4(C27S) potentiated RIG-I-induced activation of the IFN-β promoter. In addition, the potentiation of RIG-I-mediated activation of the IFN-β promoter by TRIM4 was markedly decreased after mutation of the above mentioned lysine residues (Figure 5C). These results suggest that TRIM4-mediated ubiquitination of RIG-I at K154, K164, and K172 is important for RIG-I-mediated induction of IFN-β.
TRIM4 plays a redundant role with TRIM25 and Riplet in RIG-I-mediated signaling
Previously, it has been demonstrated that two E3 ubiquitin ligases, TRIM25 and Riplet, also mediate K63-linked polyubiquitination of RIG-I at K154, K164, and K172 and modulate RIG-I-mediated signaling (Gack, 2007; Gao et al., 2009). In HEK293, human monocytic THP1, and U937 cells, TRIM4, TRIM25, and Riplet were co-expressed, and their expression levels were not markedly changed before and after SeV infection (Figure 6A). In addition, knockdown of TRIM4 did not cause noticeable change in levels of TRIM25 or Riplet before and after SeV infection (Figure 6B).
TRIM4 plays a redundant role with TRIM25 and Riplet. (A) Expression levels of endogenous TRIM4, TRIM25, and Riplet in HEK293, THP-1, and U937 cells. HEK293, THP-1, and U937 cells were either uninfected or infected with SeV for the indicated periods. Cell lysates were analyzed by immunoblots with the indicated antibodies. (B) Expression levels of endogenous TRIM25 and Riplet in TRIM4 knockdown cells. HEK293 cells stably transduced with a control or TRIM4-RNAi were either uninfected or infected with SeV for 12 h. Cell lysates were analyzed by immunoblots with the indicated antibodies. (C) Effects of knockdown of TRIM4, TRIM25, and Riplet on SeV-induced activation of the IFN-β promoter. HEK293 cells (1 × 105) were transfected with the IFN-β promoter reporter plasmids (0.1 μg) and TRIM4-RNAi, TRIM25-RNAi, Riplet-RNAi, double mix (1:1), or triple mix (1:1:1) plasmids (the total amounts of RNAi plasmids for each transfection were the same). Eighteen hours after transfection, cells were left uninfected or infected with SeV for 12 h before reporter assays were performed. Data shown are presented as mean ± SD, n = 3. **P < 0.01. (D) TRIM4 competes with TRIM25 for interaction with RIG-I-CARD. HEK293 cells (2 × 106) were transfected with the indicated plasmids (4 μg each). Coimmunoprecipitation and immunoblot analysis were performed with the indicated antibodies (upper panels). Expression of the transfected proteins was analyzed by immunoblots with anti-HA and anti-Flag (lower panels).
To determine whether these E3 ubiquitin ligases play redundant roles in RIG-I-mediated signaling, we examined the effects of knockdown of TRIM4, TRIM25, and Riplet individually or combinationally. We found that knockdown of any of them could inhibit SeV-induced activation of the IFN-β promoter. Knockdown of two or three of them at accumulative equal doses had a similar inhibition on SeV-induced activation of the IFN-β promoter (Figure 6C). In addition, in coimmunoprecipitation experiments, TRIM4 and TRIM25 competitively interacted with RIG-I-CARD (Figure 6D). These results indicated that TRIM4 and TRIM25 play redundant roles in RIG-I signaling.
Discussion
Recognition of viral PAMPs by PRRs leads to induction of type I IFNs and downstream proteins, which are critically involved in antiviral innate immune response as well as initiation of adaptive immunity. PRR-mediated induction of type I IFNs is heavily regulated by post-translational modifications, particularly phosphorylation and ubiquitination. In this study, we performed expression screens of ubiquitin-related enzymes and identified TRIM4 as a positive regulator in virus-triggered type I IFN induction pathways.
Overexpression of TRIM4 activated IRF3, NF-κB, and the IFN-β promoter and markedly potentiated SeV-triggered signaling, including transcriptional induction of the IFNB1, CCL5, and ISG56 genes. Knockdown of TRIM4 inhibited virus-induced activation of IRF3 and NF-κB, as well as IFN-β induction. However, knockdown of TRIM4 had no marked effects on TNF-α- or IL-1β-induced activation of NF-κB or IFN-γ-induced activation of the IRF1 promoter. These results reveal a critical role for TRIM4 in innate immune response against viruses.
RIG-I is a cytoplasmic sensor for viral RNA. Several lines of evidence suggest that TRIM4 regulates innate antiviral response by targeting RIG-I for K63-linked polyubiquitination. Firstly, reporter assays indicated that knockdown of TRIM4 inhibited RIG-I- but not MDA5-, VISA-, TBK1-, or IRF3-mediated signaling. Secondly, coimmunoprecipitation experiments indicated that TRIM4 was associated with RIG-I. Thirdly, overexpression of TRIM4 but not its enzymatic inactive mutant causes K63-linked polyubiquitination, whereas knockdown of TRIM4 inhibited basal and SeV-induced K63-linked polyubiquitination of RIG-I. These results suggest that TRIM4 regulates virus-triggered IFN induction by mediating K63-linked polyubiquitination of RIG-I.
In our experiments, we found that overexpression of TRIM4 increased K63-linked but decreased K48-linked polyubiquitination, whereas knockdown of TRIM4 inhibited basal and SeV-induced K63-linked polyubiquitination, but increased K48-linked polyubiquitination. The simplest explanation for these observations is that K63- and K48-linked polyubiquitination can compete for the same lysine residue(s) of RIG-I and these two types of modifications of RIG-I are reciprocally regulated.
As a crucial cytosolic RNA receptor, RIG-I activity is regulated by diverse post-translational mechanisms. It has been shown that CYLD interacts with the CARDs of RIG-I and removes K63-linked polyubiquitin chains from RIG-I, which inhibits downstream signaling (Friedman et al., 2008; Zhang et al., 2008). RNF125 mediates the attachment of K48-linked polyubiquitin chains to RIG-I, targeting it for proteasomal degradation and thereby inhibiting the antiviral immune response (Arimoto et al., 2007). Previously, it has been demonstrated that two E3 ubiquitin ligases, TRIM25 and Riplet, mediate K63-linked polyubiquitination of RIG-I at K154, K164, and K172 and modulate RIG-I-mediated signaling (Gack, 2007; Gao et al., 2009). Our results indicated that TRIM4 also targets RIG-I at K154, K164, and K172 for K63-linked polyubiquitination. Interestingly, TRIM4, TRIM25, and Riplet were co-expressed in several different types of cells. Knockdown of any one individually inhibited SeV-induced activation of the IFN-β promoter to similar degrees as combined knockdown of two or three of them at accumulative equal doses. Coimmunoprecipitation experiments suggest that TRIM4 and TRIM25 competitively interacted with RIG-I-CARD. These studies suggest that TRIM4 and TRIM25 play redundant roles in RIG-I-mediated signaling.
In conclusion, this study reveals that TRIM4 is an important regulator involved in virus-induced IFN induction pathways. TRIM4 modulates virus-triggered IFN induction by mediating K63-linked ubiquitination of RIG-I. Our findings provide additional insights to the mechanisms of regulation of cellular antiviral response.
Materials and methods
Reagents and antibodies
Recombinant TNFα, IL-1β, and IFN-γ (R&D Systems); mouse monoclonal antibodies against FLAG (Sigma), HA (Origene), β-actin (Sigma), rabbit anti-TRIM4 (Abcam), and mouse anti-TRIM25 (BD) were purchased from the indicated companies. Mouse anti-TRIM4 was raised against recombinant human full-length TRIM4 protein. SeV and anti-RIG-I were previously described (Zhong et al., 2008). Mouse anti-Riplet was kindly provided by Dr Dan-Ying Chen from Peking University.
Constructs
NF-κB, ISRE, IRF1 promoter, and IFN-β promoter luciferase reporter plasmids and mammalian expression plasmids for RIG-I, MDA5, VISA, TBK1, and IRF3 were previously described (Zhong et al., 2008; Wang et al., 2010; Li et al., 2012). Mammalian expression plasmids for HA- or FLAG-tagged TRIM4 and its mutants were constructed by standard molecular biology techniques. Mammalian expression plasmids for Riplet and TRIM25-RNAi and Riplet-RNAi plasmids were provided by Dr Dan-Ying Chen from Peking University.
Expression cloning
The cDNA expression clones encoding 352 ubiquitin-related enzymes were obtained from the Ubiquitin-GFC Transfection Array (Origene, Cat# UBGB19601). The clones were transfected together with an IFN-β luciferase reporter plasmid into HEK293 cells. Reporter assays were performed at 16 h after transfection to identify clones that could markedly activate the IFN-β promoter reporter.
Quantitative real-time PCR
Total RNA was isolated from cells using Trizol reagent (Takara), reverse-transcribed, and subjected to qPCR analysis to measure mRNA expression levels of tested genes. Gene-specific primer sequences were as follow: IFNB1: 5′-TTGTTGAGAACCTCCTGGCT-3′ (forward), 5′-TGACTATGGTCCAGGCACAG-3′ (reverse); RANTES: 5′-GGCAGCCCTCGCTGTCATCC-3′ (forward), 5′-TTGGGCGCCGGAAAGCTGTAGAT-3′ (reverse); ISG56: 5′-GCCGCATCGCCGTCTCCTAC-3′ (forward), 5′-GCAGCAGGGTGTGGTGTCCG-3′ (reverse); GAPDH: 5′-GAGTCAACGGATTTGGTCGT-3′ (forward), 5′-GACAAGCTTCCCGTTCTCAG-3′ (reverse).
RNAi experiments
Double-stranded oligonucleotides corresponding to the target sequences were cloned into the pSuper.Retro RNAi plasmid (Oligoengine, Inc.). The sequences targeting human TRIM4 cDNA were as follow: #1: 5′-GAAGTTGAGAGTAGAGATA-3′; #2: 5′-GAGATTGAACAAAGAAGAA-3′; #3: 5′-GAAGACAGTGTGCCAGATA-3′. A pSuper.retroRNAi plasmid targeting GFP was used as control for all RNAi-related experiments.
Transfection and reporter assays
HEK293 cells (1 × 105) were seeded on 24-well dishes and transfected on the following day by a standard calcium phosphate precipitation method. An empty control plasmid was added to ensure that each transfection received the same amount of total DNA. To normalize for transfection efficiency, 0.05 μg of pRL-TK (Renilla luciferase) reporter plasmid was added to each transfection. Approximately 18 h after transfection, luciferase assays were performed using a dual-specific luciferase assay kit (Promega). Firefly luciferase activities were normalized on the basis of Renilla luciferase activities.
Fluorescent confocal microscopy
HEK293 cells were transfected with the indicated plasmids by lipofectamine 2000 (Invitrogen). Twenty-four hours after transfection, the cells were infected with SeV for 4 h followed by fixation with 4% paraformaldehyde for 15 min at 4°C. The cells were then observed with an OLYMPUS confocal microscope under a 100× oil objective.
VSV plaque assay
HEK293 cells (1 × 105) were transfected with the indicated plasmids for 36 h prior to VSV infection (MOI at 0.1). At 1 h after infection, cells were washed with PBS for three times and then medium was added. The supernatants were harvested at 24 h after washing. The supernatants were diluted by 1:106 to infect confluent BHK21 cells cultured on 24-well plates. At 1 h post-infection, the supernatant was removed, and 3% methylcellulose was overlayed. Three days after infection, the overlay was removed, and cells were fixed with 4% formaldehyde for 20 min, and stained with 0.2% crystal violetin. Plaques were counted, averaged, and multiplied by the dilution factor to determine viral titer as PFU/ml.
Coimmunoprecipitation and immunoblot analysis
For transient transfection and coimmunoprecipitation experiments, HEK293 cells (1 × 106) were transfected for 18−24 h. The transfected cells were lysed in 1 ml of lysis buffer (20 mM Tris, pH 7.5, 150 mM NaCl, 1% Triton, 1 mM EDTA, 10 μg/ml aprotinin, 10 μg/ml leupeptin, and 1 mM phenylmethylsulfonyl fluoride). For each immunoprecipitation, a 0.4 ml aliquot of the lysate was incubated with 0.5 μg of the indicated antibody or control IgG and 25 μl of 1:1 slurry of Protein G Sepharose (GE Healthcare) for 2 h. Sepharose beads were washed three times with 1 ml of lysis buffer containing 0.5 M NaCl. The precipitates were analyzed by standard immunoblot procedures. For endogenous coimmunoprecipitation experiments, HEK293 cells (5 × 107) were infected with SeV for the indicated times or left uninfected. The coimmunoprecipitation and immunoblot experiments were performed as described above.
In vitro ubiquitination assays
The tested proteins were expressed with a TNT Quick Coupled Transcription/Translation Systems kit (Promega) following instructions of the manufacturer. Ubiquitination was analyzed with an ubiquitination kit (Enzo Life Sciences) following protocols recommended by the manufacturer.
Statistical analysis
Differences between experimental and control groups were determined by student's t-test. P-values <0.05 were considered statistically significant.
Funding
This work was supported by grants from the Ministry of Science and Technology of China (2012CB910201 and 2010CB911802) and the National Natural Science Foundation of China (31221061, 31130020, and 91029302).
Conflict of interest: none declared.
