ABSTRACT
Activation of innate immunity is essential for host cells to restrict the spread of invading viruses and other pathogens. However, attenuation or termination of signaling is also necessary for preventing immune-mediated tissue damage and spontaneous autoimmunity. Here, we identify nucleotide binding oligomerization domain (NOD)-like receptor X1 (NLRX1) as a negative regulator of the mitochondrial antiviral signaling protein (MAVS)-mediated signaling pathway during hepatitis C virus (HCV) infection. The depletion of NLRX1 enhances the HCV-triggered activation of interferon (IFN) signaling and causes the suppression of HCV propagation in hepatocytes. NLRX1, a HCV-inducible protein, interacts with MAVS and mediates the K48-linked polyubiquitination and subsequent degradation of MAVS via the proteasomal pathway. Moreover, poly(rC) binding protein 2 (PCBP2) interacts with NLRX1 to participate in the NLRX1-induced degradation of MAVS and the inhibition of antiviral responses during HCV infection. Mutagenic analyses further revealed that the NOD of NLRX1 is essential for NLRX1 to interact with PCBP2 and subsequently induce MAVS degradation. Our study unlocks a key mechanism of the fine-tuning of innate immunity by which NLRX1 restrains the retinoic acid-inducible gene I-like receptor (RLR)-MAVS signaling cascade by recruiting PCBP2 to MAVS for inducing MAVS degradation through the proteasomal pathway. NLRX1, a negative regulator of innate immunity, is a pivotal host factor for HCV to establish persistent infection.
IMPORTANCE Innate immunity needs to be tightly regulated to maximize the antiviral response and minimize immune-mediated pathology, but the underlying mechanisms are poorly understood. In this study, we report that NLRX1 is a proviral host factor for HCV infection and functions as a negative regulator of the HCV-triggered innate immune response. NLRX1 recruits PCBP2 to MAVS and induces the K48-linked polyubiquitination and degradation of MAVS, leading to the negative regulation of the IFN signaling pathway and promoting HCV infection. Overall, this study provides intriguing insights into how innate immunity is regulated during viral infection.
INTRODUCTION
Innate immunity is the first line of defense of the host against invading viruses and other microbes (1, 2). During virus infection, host cells use pattern recognition receptors (PRRs) to sense viruses as foreign invaders through pathogen-associated molecular pattern (PAMP) recognition to activate the innate immune response (3). Major key PRRs include cytosolic retinoic acid-inducible gene I (RIG-I)-like receptors (RLRs), membrane-bound Toll-like receptors (TLRs), and other nontraditional PRRs. RLRs include three members, RIG-I, melanoma differentiation-associated protein 5 (MDA5), and Laboratory of Genetics and Physiology 2 (LGP2). All three RLRs contain a DExD/H box RNA helicase domain, while LGP2 lacks a caspase recruitment domain (CARD) compared with RIG-I and MDA5. This deficiency makes LGP2 fail to autonomously transduce signaling but potentially modulates RIG-I- and MDA5-mediated innate immune responses (4–6). Although RIG-I and MDA5 sense different types of viruses (7), they share a common adaptor protein, mitochondrial antiviral signaling protein (MAVS) (also known as Cardif, IPS-1, or VISA), residing on the mitochondrial outer membrane, to transduce downstream signaling (8–11). Upon binding to viral RNAs, RIG-I and MDA5 undergo conformational changes to activate MAVS, which subsequently activates interferon (IFN) regulatory factor 3 (IRF3) and NF-κB to produce type I and III IFNs as well as other antiviral cytokines, contributing to the restriction of viral infection (12, 13).
Hepatitis C virus (HCV) infects approximately 130 million to 150 million people worldwide, causing severe liver diseases, including chronic hepatitis, cirrhosis, and hepatocellular carcinoma (14, 15). Currently, no prophylactic vaccine is available. HCV is an enveloped, positive-sense, single-stranded RNA virus belonging to the family Flaviviridae. The genome of HCV encodes a single polyprotein of 3,011 amino acids, which is processed into structural (core, E1, and E2) and nonstructural (p7, NS2, NS3, NS4A, NS4B, NS5A, and NS5B) proteins by host and viral proteases (16). In hepatocytes infected by HCV, HCV can be recognized by the cytosolic sensor RIG-I, a key member of the RLRs, triggering a hepatic innate immune response to restrict HCV infection (17, 18). Interestingly, recent studies suggested that MDA5 and LGP2 may also play roles in the host innate immune response against HCV (6, 19). Nonetheless, MAVS, as the common adaptor protein of RLRs, exerts a central function in anti-HCV signal transduction. Actually, the process in which the HCV NS3/4A protease cleaves MAVS to block the MAVS-mediated signaling pathway crucial for HCV to evade innate immunity to establish persistent infection (20, 21). However, whether host factors are also involved in the negative regulation of the activation of the MAVS-mediated signaling pathway during HCV infection is largely unclear.
Nucleotide binding oligomerization domain (NOD)-like receptor X1 (NLRX1) (also known as NOD5, NOD9, or NOD26) is a member of the NLR family and localizes to the mitochondrial outer membrane. One of the most fundamental functions of the NLR family is to modulate the production of proinflammatory cytokines and chemokines in host cells upon pathogen infections or environmental insults, which is associated with the critical roles of NLRs in autoimmune, autoinflammatory, and infectious diseases (22–24). NLRP3, one of the best-characterized NLRs that form inflammasomes, was shown to drive the production of the proinflammatory cytokines interleukin-1β (IL-1β) and IL-18 in macrophages during HCV infection (25). Unlike other NLRs, NLRX1 does not exhibit inflammasome functions and has unique functions in immune modulation, autophagy, and tumorigenesis. NLRX1 attenuates IRF3 and NF-κB signaling pathways in response to viral infection and TLR activation (26–29). Furthermore, NLRX1 promotes virus-induced autophagy by interactions with another mitochondrial protein, TUFM (Tu translation elongation factor, mitochondrial) (30). Interestingly, two recent studies demonstrated that NLRX1 acts as a tumor suppressor to reduce tumorigenesis through regulating tumor necrosis factor (TNF) and IL-6 signaling pathways (31, 32). However, the potential functions of NLRX1 in HCV propagation and the HCV-induced innate immune response remain unknown.
In the present study, we report that NLRX1 is a proviral host factor for HCV infection and functions as a negative regulator of the HCV-triggered innate immune response. NLRX1 is induced by HCV infection and interacts with MAVS, leading to the K48-linked polyubiquitination and subsequent degradation of MAVS via the proteasomal pathway. Moreover, poly(rC) binding protein 2 (PCBP2), a newly identified NLRX1-interacting protein, plays a pivotal role in the NLRX1-induced degradation of MAVS during HCV infection. The NOD of NLRX1 is essential for interactions with PCBP2 and is required for mediating MAVS degradation. The depletion of PCBP2 also attenuates the host innate immune response against HCV. Our study unlocks a key mechanism of the fine-tuning of innate immunity by which NLRX1 negatively regulates the RLR-MAVS signaling cascade through the PCBP2-mediated proteasomal pathway.
RESULTS
NLRX1 facilitates HCV infection.Although evidence from a recent report indicated that NLRX1 may act as a proviral host factor in infection by human immunodeficiency virus type 1 and some DNA viruses (29), the role of NLRX1 in HCV infection remains obscure. To investigate whether NLRX1 modulates HCV propagation, we constructed a stable NLRX1-silenced cell line using HLCZ01 cells, which support the entire life cycle of HCV (see details in our previous study [33]). NLRX1 was knocked down effectively in HLCZ01 cells (Fig. 1A). Strikingly, the silencing of NLRX1 dramatically decreased the level of intracellular HCV RNA at multiple time points after HCV infection, compared with the control cell line containing the scrambled short hairpin RNA (shRNA) (Fig. 1B). The level of the HCV core protein was also reduced in NLRX1-silenced cells (Fig. 1C). In addition, the knockdown of NLRX1 significantly decreased the percentage of virus-infected cells (Fig. 1D). Importantly, NLRX1 can be specifically induced by HCV infection in a time-dependent manner, whereas Newcastle disease virus (NDV) infection and treatments with RLR stimuli had no effect on the expression of NLRX1 (Fig. 1E). Collectively, these data indicated that NLRX1 is a pivotal host factor for HCV to establish successful infection.
NLRX1 facilitates HCV infection. (A) HLCZ01 cells stably expressing either scrambled shRNA (sh-Ctrl) or NLRX1-targeting shRNA (sh-NLRX1) were analyzed by immunoblotting with an anti-NLRX1 antibody. β-Actin was used as the loading control. (B) HLCZ01 cells stably expressing either sh-Ctrl or sh-NLRX1 were infected with HCV at an MOI of 0.1 at the indicated time points. The intracellular HCV RNA level was determined by real-time PCR and normalized to the value for GAPDH (glyceraldehyde-3-phosphate dehydrogenase). (C and D) The cells were treated as described above for panel B except with HCV infection for 72 h, followed by immunoblotting with an anti-HCV core antibody (C) as well as immunostaining using an anti-HCV NS5A antibody (green) (D). DAPI (4′,6-diamidino-2-phenylindole) was used to counterstain nuclei (blue). (E) HLCZ01 cells were transfected with either HCV 3′-UTR RNA (500 ng) or poly(I·C) (500 ng) for 16 h, treated with IFN-α (200 U/ml) for 12 h, or infected with either NDV or HCV at the indicated time points, followed by immunoblotting with anti-NLRX1 antibody. The NLRX1 blot was analyzed by using ImageJ software. Data in panels B and E are represented as means ± SD from triplicate experiments. **, P < 0.01; ***, P < 0.001.
NLRX1 negatively regulates the HCV-triggered innate immune response.Some studies suggested that NLRX1 is capable of negatively regulating the innate immune response through blocking IFN and cytokine production pathways (26–29). However, another study reported that NLRX1 enhances antiviral activity (34). We speculated that NLRX1 may inhibit the HCV-induced innate immune response, which in turn promotes HCV infection. To assess this hypothesis, we measured the expression levels of a series of genes participating in host antiviral defense. Notably, the knockdown of NLRX1 remarkably enhanced the expression of type I IFN (IFN-β) as well as type III IFN (IL-28A) compared with controls (Fig. 2A and B). Moreover, representative IFN-stimulated genes (ISGs), including ISG56, ISG12a, and ISG60, were also upregulated in NLRX1-silenced cells (Fig. 2C to E). Importantly, ISG12a and ISG56 exhibit anti-HCV activity, as reported in our and other previous studies (35–37). The enhanced expression of ISG12a and ISG56 in NLRX1-silenced cells was consistent with reduced infection by HCV during NLRX1 depletion (Fig. 1). Moreover, silencing of NLRX1 enhanced the activation of TANK-binding kinase 1 (TBK1) and signal transducer and activator of transcription 1 (STAT1) by the HCV 3′ untranslated region (UTR) (Fig. 2F). To rule out the possibility of an off-target effect of NLRX1 shRNA, we generated a NLRX1 mutant with resistance to NLRX1 shRNA function. The NLRX1 shRNA-resistant mutant could be expressed in NLRX1-silenced cells (Fig. 2I). The ectopic expression of the NLRX1 shRNA-resistant mutant inhibited the induction of IFN-β and restored the intracellular levels of HCV RNA compared with those without exogenous NLRX1 expression in NLRX1-silenced cells (Fig. 2J and K). Furthermore, the direct overexpression of NLRX1 in HLCZ01 cells decreased the expression level of IFN-β and increased the abundance of HCV RNA in a dose-dependent manner (Fig. 2G and H). Taken together, these results suggested that NLRX1 negatively regulates the HCV-triggered innate immune response.
NLRX1 negatively regulates HCV-triggered innate immune responses. (A to E) HLCZ01 cells stably expressing either sh-Ctrl or sh-NLRX1 were infected with HCV at an MOI of 0.1 at the indicated time points. The kinetics of induction of IFN-β (A), IL-28A (B), ISG56 (C), ISG12a (D), and ISG60 (E) were analyzed by real-time PCR and normalized to the value for GAPDH. (F) Immunoblot analyses of phosphorylated and total (inactive) TBK1 and STAT1 in lysates of sh-Ctrl and sh-NLRX1 HLCZ01 cells transfected with HCV 3′-UTR RNA (500 ng) for 5 h. β-Actin was used as the loading control. (G and H) HLCZ01 cells were infected with HCV at an MOI of 0.1 for 4 days and then transfected with increasing amounts of a plasmid expressing Flag-NLRX1 for 48 h. The induction of IFN-β (G) or intracellular HCV RNA (H) was analyzed by real-time PCR and normalized the value for GAPDH. (I) Immunoblot analyses of the NLRX1 protein in lysates of cells with or without exogenous mutant NLRX1 (NLRX1-mut) (shNLRX1-resistant) expression. (J and K) HLCZ01 cells stably expressing either sh-Ctrl or sh-NLRX1 were infected with HCV at an MOI of 0.1 at the indicated time points. The plasmid encoding NLRX1-mut was transfected for 48 h before harvesting of cells. The induction of IFN-β (J) or intracellular HCV RNA (K) was analyzed by real-time PCR and normalized to the value for GAPDH. Data are represented as means ± SD from triplicate experiments. *, P < 0.05; **, P < 0.01; ***, P < 0.001.
NLRX1 targets MAVS and induces its degradation during HCV infection.During HCV replication, processed PAMPs are sensed as nonself by PRRs in the host cell, leading to the activation of the innate immune response (1, 3). Previously, we and other groups reported that the RIG-I signaling pathway plays an essential role in the induction of IFNs and ISGs by HCV infection (17, 33, 36). Indeed, the depletion of RIG-I in HLCZ01 cells abrogated HCV-induced IFN-β expression (Fig. 3A). The efficiency of RIG-I knockdown was determined by real-time PCR (Fig. 3B, bottom) and Western blot (Fig. 3B, top) analyses. Based on previous findings and the current results concerning a negative regulatory role of NLRX1, we then tested whether NLRX1 regulates RIG-I-mediated signal transduction in hepatocytes. Upon stimulation with poly(I·C), which mimics viral double-stranded RNA, the levels of IFN mRNAs were dramatically elevated in HLCZ01 cells, whereas the knockdown of NLRX1 remarkably augmented the production of type I IFN (IFN-β) and type III IFN (IL-29 and IL-28A) triggered by poly(I·C) (Fig. 3C to E). In addition, silencing of NLRX1 enhanced the activation of TBK1 and STAT1 by the 3′ UTR of HCV RNA, which is known as a ligand for RIG-I recognition (Fig. 2F), suggesting that NLRX1 may inhibit the activation of the RIG-I signaling pathway. To investigate which component(s) of the RIG-I signaling pathway could be targeted by NLRX1, we cotransfected plasmids encoding NLRX1 and various candidates in a reporter assay system. Interestingly, NLRX1 strongly inhibited the activation of the IFN-β promoter by RIG-I and MAVS in a dose-dependent manner, while the TBK1-mediated activation of the IFN-β promoter was unaffected (Fig. 3F). These data suggested that NLRX1 negatively regulates the RIG-I signaling pathway downstream of MAVS but upstream of TBK1.
NLRX1-mediated suppression of antiviral immunity is RIG-I–MAVS dependent. (A) HLCZ01 cells stably expressing either control shRNA or RIG-I-targeting shRNA were infected with HCV at a gradient of MOIs at the indicated time points. The induction of IFN-β was analyzed by real-time PCR and normalized to the value for GAPDH. (B) Determination of RIG-I knockdown in HLCZ01 cells. The levels of RIG-I mRNA (bottom) and protein (top) were analyzed by real-time PCR and immunoblotting, respectively. (C to E) HLCZ01 cells stably expressing either sh-Ctrl or sh-NLRX1 were transfected with poly(I·C) for 18 h. The induction of IFN-β (C), IL-29 (D), and IL-28A (E) was analyzed by real-time PCR and normalized to the value for GAPDH. (F) NLRX1 inhibits activation of the IFN-β promoter by MAVS and RIG-I. HLCZ01 cells were cotransfected with IFN-β–luciferase (luc), pRL-CMV, and the indicated plasmids for 24 h, followed by a dual-luciferase reporter assay. Data are represented as means ± SD from triplicate experiments. *, P < 0.05; **, P < 0.01; ***, P < 0.001.
Based on the observations of the regulatory functions of NLRX1 described above, we speculated that NLRX1 may function by interacting with MAVS. To evaluate this hypothesis, we coexpressed V5-NLRX1 and Flag-MAVS in HLCZ01 cells, followed by coimmunoprecipitation (co-IP) analysis of the interactions between MAVS and NLRX1. Consistent with our assumption, NLRX1 coprecipitated with MAVS in HLCZ01 cells (Fig. 4A). Basigin (BSG), a plasma membrane protein, did not interact with NLRX1 (Fig. 4A). Furthermore, the association of endogenous NLRX1 and MAVS was also confirmed in HLCZ01 cells, and HCV infection enhanced the NLRX1-MAVS interaction (Fig. 4B), suggesting a potential mechanism for NLRX1 to modulate the MAVS-mediated signaling pathway during HCV infection. Interestingly, the ectopic expression of NLRX1 in the presence of HCV infection effectively decreased the level of MAVS (Fig. 4C). Although MAVS was cleaved dramatically in HCV-infected Huh7.5 cells, only a small portion of the cleaved MAVS protein was observed in HCV-infected HLCZ01 cells (Fig. 4C and D). This suggests that the reduction in the abundance of MAVS by NLRX1 in HLCZ01 cells during HCV infection may not be caused by the virus-induced cleavage of full-length MAVS (Fig. 4C). Moreover, the ectopic expression of NLRX1 also decreased the level of a MAVS-C508R mutant (designated MAVSR) that exhibits resistance to the HCV NS3/4A protease in both Huh7.5.1 and HLCZ01 cells (Fig. 4E and F). To assess whether NLRX1-triggered MAVS downregulation is dependent on the proteasomal pathway, we treated virus-infected cells with the proteasome inhibitor MG132 following the overexpression of NLRX1. The reduction in the level of endogenous MAVS caused by NLRX1 was abrogated by treatment with MG132 (Fig. 4G), indicating that NLRX1 may induce MAVS degradation through the proteasomal pathway. Ubiquitination of a targeted protein is a common event observed during proteasome-mediated protein degradation. Thus, we next investigated the role of the ubiquitination pathway in NLRX1-induced MAVS degradation. The overexpression of NLRX1 significantly enhanced the polyubiquitination of MAVS, compared with the basal level of MAVS polyubiquitination in the control group (Fig. 4H). Strikingly, we furthermore found that HCV infection can trigger the abundant polyubiquitination of MAVS in NLRX1-expressing cells but not in NLRX1-silenced cells (Fig. 4I, left). K48-dependent polyubiquitination is a classical regulatory process for protein degradation. A similar result was observed in a ubiquitination assay with the expression of hemagglutinin (HA)-tagged K48-ubiquitin, which showed that the induction of MAVS polyubiquitination by HCV infection was attenuated in NLRX1-silenced cells compared with that in control cells (Fig. 4I, right). In summary, these results indicated that NLRX1 interacts with MAVS and triggers MAVS degradation via the ubiquitin-mediated proteasomal pathway during HCV infection, which may, at least partially, account for the suppression of HCV-induced innate immune responses caused by NLRX1.
NLRX1 targets MAVS and induces its degradation during HCV infection. (A) Association of NLRX1 with MAVS. HLCZ01 cells coexpressing V5-NLRX1 and Flag-MAVS or Flag-BSG were assessed by co-IP and immunoblotting with the indicated antibodies. (B) HCV enhances the interaction of endogenous NLRX1 with MAVS. HLCZ01 cells were infected with HCV at an MOI of 0.1 at the indicated time points. Co-IP and immunoblotting were performed with the indicated antibodies. (C) HLCZ01 cells were mock infected or infected with HCV (MOI of 1) for 48 h and then transfected with either the vector or the plasmid encoding Flag-NLRX1 (2 μg) for 24 h, followed by immunoblotting with the indicated antibodies. SE, short exposure; LE, long exposure; FL, full-length MAVS; C, cleaved MAVS. (D) Immunoblot analyses of endogenous MAVS in lysates of HLCZ01 and Huh7.5 cells infected with HCV at a gradient of MOIs at the indicated time points. β-Actin was used as the loading control. (E) Huh7.5.1-WT or Huh7.5.1-MAVSR cells were treated as described above for panel C, followed by immunoblotting with the indicated antibodies. (F) HLCZ01 cells were cotransfected with the plasmids encoding V5-NLRX1 and Flag-MAVS or Flag-MAVS(C508R) for 48 h, followed by immunoblotting with the indicated antibodies. (G) HLCZ01 cells were inoculated with HCV (MOI of 1) for 24 h and then transfected with either the vector or the plasmid encoding Flag-NLRX1 (2 μg) for 48 h. The cells were treated with dimethyl sulfoxide or MG132 (25 μM) for the final 6 h of infection, followed by immunoblotting with the indicated antibodies. (H) NLRX1 promotes the polyubiquitination of MAVS. HEK293T cells were cotransfected with the indicated plasmids for 42 h and treated with MG132 (25 μM) for an additional 6 h. Ubiquitination assays and immunoblotting were performed with the indicated antibodies. (I) NLRX1 mediates K48-linked polyubiquitination of MAVS during HCV infection. HLCZ01 cells stably expressing either sh-Ctrl or sh-NLRX1 were infected with HCV (MOI of 0.1) for 48 h, followed by transfection of the indicated plasmids for another 48 h. The cells were treated with MG132 (25 μM) for the final 6 h prior to analysis. Ubiquitination assays and immunoblotting were performed with the indicated antibodies. The blots in panels B, C, and G were analyzed by using ImageJ software, and the data are represented as means ± SD from triplicate experiments. *, P < 0.01; **, P < 0.01; ***, P < 0.001.
The NOD of NLRX1 is crucial for mediating MAVS degradation.To better understand the molecular mechanisms of NLRX1-mediated MAVS degradation, we separately tested the individual functional domains of MAVS and NLRX1. First, we constructed a series of MAVS truncations (the functional domains of MAVS are shown in Fig. 5A) and coexpressed them with full-length NLRX1 in HLCZ01 cells. The result showed that NLRX1 induced the degradation of MAVS domain II (DII), the region lacking domain III (DI and DII) or lacking domain I (DII and DIII), and full-length MAVS (Fig. 5B). Nevertheless, the abundances of MAVS DI and DIII and the region with a deletion of DII were unaffected by NLRX1 overexpression (Fig. 5B). These data suggested that MAVS DII is an essential functional domain involved in NLRX1-mediated MAVS degradation. Next, we sought to find which domain(s) of NLRX1 plays a pivotal role in this process. The major functional domains of NLRX1 are illustrated in Fig. 5C. The ectopic expression of the NOD-containing regions of NLRX1, but not the NLRX1 leucine-rich repeat (LRR) domain, triggered the degradation of MAVS upon HCV infection (Fig. 5D). Moreover, the overexpression of the NLRX1 NOD, as well as other constructs harboring the NOD, remarkably inhibited the HCV-induced expression of IFN-β in HLCZ01 cells (Fig. 5E). These results demonstrated that the NLRX1 NOD is crucial for NLRX1 to trigger MAVS degradation and the subsequent suppression of the activation of the IFN production pathway during HCV infection.
The NOD of NLRX1 is crucial for mediating MAVS degradation. (A) Schematic illustration of MAVS truncations. (B) NLRX1 induces MAVS DII degradation. HLCZ01 cells were cotransfected with the V5-NLRX1 plasmid (1 μg) and the plasmids encoding the indicated domains of MAVS (1 μg) for 48 h. Cell lysates were analyzed by immunoblotting with the indicated antibodies. β-Actin was used as the loading control. FL, full length. (C) Schematic illustration of NLRX1 truncations. (D) The NLRX1 NOD promotes MAVS degradation. HLCZ01 cells were transfected with plasmids encoding the indicated domains of NLRX1 (2 μg) for 24 h and then inoculated with HCV (MOI of 0.1) for another 24 h, followed by immunoblotting with the indicated antibodies. β-Actin was used as the loading control. (E) The NLRX1 NOD inhibits HCV-induced IFN-β production. HLCZ01 cells were infected with HCV (MOI of 0.1) for 72 h, followed by transfection of the plasmids encoding the indicated domains of NLRX1 (2 μg) for 48 h. The induction of IFN-β was analyzed by real-time PCR and normalized to the value for GAPDH. Data are represented as means ± SD from triplicate experiments. *, P < 0.05.
NLRX1 interacts with PCBP2 to regulate the MAVS-mediated signaling pathway.To further explore how NLRX1 triggers MAVS degradation, a bioinformatics analysis was conducted to predict the candidate proteins for NLRX1 interactions. Among the predicted targets, we paid attention to one that has a known function in regulating the MAVS-mediated signaling pathway. PCBP2, an interesting candidate for association with NLRX1, was reported previously to negatively regulate the IFN production pathway via the induction of MAVS degradation in a ubiquitination-dependent manner (38). Based on the similar functions of NLRX1 and PCBP2 in the regulation of the innate immune response, we speculated that PCBP2 may be involved in NLRX1-mediated MAVS degradation and immune inhibition. To assess this hypothesis, we first sought to verify the interaction between NLRX1 and PCBP2 in the experiments. Consistent with our expectation, the NLRX1 protein was detected in a PCBP2 immunoprecipitation (IP)-derived protein lysate obtained from cells overexpressing NLRX1 and PCBP2, while it was undetectable in the control group (Fig. 6A). Notably, an association of endogenous NLRX1 and PCBP2 was found in HLCZ01 cells (Fig. 6B). These data indicated a physiological interaction between NLRX1 and PCBP2, which prompted us to investigate the role of PCBP2 in NLRX1-triggered MAVS degradation. In HCV-infected cells, the overexpression of NLRX1 remarkably reduced the abundance of the MAVS protein, whereas the knockdown of PCBP2 abrogated this effect compared with control treatment (Fig. 6C). The efficiency of PCBP2 silencing was confirmed by Western blotting (Fig. 6C). Moreover, the depletion of PCBP2 reversed the suppression of the MAVS-induced expression of IFN-β caused by NLRX1 (Fig. 6D). To assess whether PCBP2 is involved in the NLRX1-MAVS interaction, we overexpressed PCBP2 in HLCZ01 cells and tested the association of the endogenous NLRX1 and MAVS proteins. The result showed that the overexpression of PCBP2 indeed augmented the NLRX1-MAVS interaction in HCV-infected cells (Fig. 6E). Importantly, the K48-dependent polyubiquitination of MAVS detected by a K48-ubiquitin-specific antibody was observed in cells ectopically expressing either NLRX1 or PCBP2 (Fig. 6F). This effect could be augmented by the simultaneous overexpression of NLRX1 and PCBP2 (Fig. 6F). In addition, the knockdown of PCBP2 abrogated the NLRX1-mediated inhibition of the HCV-induced expression of IFN-β (Fig. 6G). Overall, these results suggested that PCBP2 promotes the NLRX1-MAVS interaction as well as NLRX1-mediated MAVS degradation, which may account for the NLRX1-induced block of the innate immune response.
NLRX1 interacts with PCBP2 to regulate the MAVS-mediated signaling pathway. (A) Association of NLRX1 with PCBP2. HLCZ01 cells coexpressing Flag-NLRX1 and V5-PCBP2 were assessed by co-IP and immunoblotting with the indicated antibodies. (B) Interaction of endogenous NLRX1 with PCBP2. HLCZ01 cells were infected with HCV at an MOI of 0.1 at the indicated time points. Co-IP and immunoblotting were performed with the indicated antibodies. (C) HCV-infected HLCZ01 cells were transfected with either the vector or the plasmid encoding Flag-NLRX1 (2 μg), with or without PCBP2 knockdown, followed by immunoblotting with the indicated antibodies. β-Actin was used as the loading control. (D) HLCZ01 cells were transfected with the indicated plasmids. The induction of IFN-β by MAVS was analyzed by real-time PCR and normalized to the value for GAPDH. (E) PCBP2 facilitates the NLRX1-MAVS interaction. HLCZ01 cells were transfected with either the vector or the V5-PCBP2 plasmid for 12 h and then infected with HCV (MOI of 0.1) for an additional 12 h. Co-IP and immunoblotting were performed with the indicated antibodies. (F) PCBP2 promotes NLRX1-mediated K48-linked polyubiquitination of MAVS. HEK293T cells were cotransfected with the indicated plasmids for 42 h and treated with MG132 (25 μM) for an additional 6 h. Ubiquitination assays and immunoblotting were performed with the indicated antibodies. (G) HLCZ01 cells were infected with HCV for 48 h and then transfected with the indicated plasmids for 24 h. The induction of IFN-β was analyzed by real-time PCR and normalized to the value for GAPDH. Blots in panels B, C, and E were analyzed by using ImageJ software. NS, nonsignificant. Data are represented as means ± SD from triplicate experiments. *, P < 0.05; **, P < 0.01; ***, P < 0.001.
The NLRX1 NOD is essential for interaction with PCBP2.Characterization of the interaction between NLRX1 and PCBP2 is important for further investigation of how NLRX1 functions in host cells. Potential interaction interfaces between NLRX1 and PCBP2 were estimated by using in silico approaches (see Materials and Methods), suggesting that the helices in the NOD of NLRX1 are most likely involved in the interaction with PCBP2 (Fig. 7A). Using co-IP experiments, we verified the prediction that the region containing the NLRX1 X domain and NOD (X-NOD) or containing the NLRX1 NOD and LRR domain (NOD-LRR) interacts with full-length PCBP2 (Fig. 7B). Notably, the NOD of NLRX1 was sufficient to interact with PCBP2, whereas the NLRX1 LRR domain had no binding affinity for PCBP2 (Fig. 7B). These data indicated that the NLRX1 NOD is crucial for the interaction with PCBP2, which is consistent with the results described above showing that the NLRX1 NOD can mediate MAVS degradation. PCBP2 is comprised of three heterogenous nuclear ribonucleoprotein K homology (KH) domains (KH1, KH2, and KH3) and a linker region between KH2 and KH3. To study the functional domain(s) of PCBP2 responsible for the NLRX1-PCBP2 interaction, we generated a series of truncations of PCBP2 (the functional domains of PCBP2 are shown in Fig. 7C). The results of co-IP analyses showed that NLRX1 interacted with the region consisting of KH1 and KH2 of PCBP2 (KH1-KH2) as well as the regions with a deletion of KH1 (Del-KH1) or the linker (Del-Linker) (Fig. 7D). Collectively, these results demonstrated that the NOD of NLRX1 and the KH2 region of PCBP2 are crucial for the NLRX1-PCBP2 interaction.
The NLRX1 NOD is essential for interaction with PCBP2. (A) Bioinformatics analysis was conducted to predict the pivotal domains responsible for the NLRX1-PCBP2 interaction. TM, transmembrane, CTD, C-terminal domain. (B) The NLRX1 NOD interacts with PCBP2. HLCZ01 cells were cotransfected with plasmids encoding V5-PCBP2 and the indicated domains of NLRX1 for 48 h. Co-IP and immunoblotting were performed with the indicated antibodies. * denotes nonspecific bands. (C) Schematic illustration of PCBP2 truncations. (D) The KH2 domain of PCBP2 is crucial for the interaction of PCBP2 with NLRX1. HEK293T cells were cotransfected with the plasmids encoding V5-NLRX1 and the indicated domains of PCBP2 for 48 h. Co-IP and immunoblotting were performed with the indicated antibodies.
PCBP2 mediates the remarkable degradation of MAVS during HCV infection.Although the functions of PCBP2 in modulating innate immune responses induced by some model viruses, including Sendai virus (SeV), vesicular stomatitis virus (VSV), and NDV, were investigated previously (38), the role of PCBP2 in the HCV-triggered innate immune response remains unclear. In fact, the triple interaction of NLRX1, PCBP2, and MAVS can be predicted with in silico approaches (Fig. 8A). Since the association of PCBP2 and NLRX1 was confirmed in the present study, we sought to assess the interaction between PCBP2 and MAVS during HCV infection. Although it was easy to detect the interaction between exogenously expressed PCBP2 and MAVS (Fig. 8B), the endogenous association of PCBP2 and MAVS was undetectable in HLCZ01 cells (Fig. 8C). However, HCV infection can induce the interaction between endogenous PCBP2 and MAVS (Fig. 8C). We also found an interaction between PCBP2 and RIG-I (Fig. 8B), indicating a potential function of PCBP2 in regulating the RIG-I–MAVS signaling pathway. Using a reporter assay system, we found that the increased expression of V5-PCBP2 interfered with the RIG-I- and MAVS-induced activation of the IFN-β promoter in a dose-dependent manner but did not affect the activation of the IFN-β promoter triggered by TBK1 (Fig. 8D). Interestingly, the ectopic expression of PCBP2 remarkably decreased the levels of both exogenous (Fig. 8E) and endogenous (Fig. 8F and G) MAVS in a dose-dependent manner, which could be reversed by MG132 treatment (Fig. 8G). Additionally, the abundance of the RIG-I protein was unaffected by PCBP2 overexpression (Fig. 8F). We also found that HCV infection augmented the degradation of MAVS induced by PCBP2 (Fig. 8H). The finding that the PCBP2 linker region is required for the binding of PCBP2 to MAVS (38) prompted us to study whether this domain is indispensable for PCBP2-mediated MAVS degradation. Consistent with our assumption, the deletion of the linker region of PCBP2 abolished the ability of PCBP2 to induce MAVS degradation (Fig. 8I). Taken together, these results suggested that PCBP2 promotes the degradation of MAVS, depending on its linker domain, during HCV infection.
PCBP2 mediates remarkable degradation of MAVS during HCV infection. (A) Bioinformatics analyses were conducted to predict the pivotal domains responsible for the triple interaction of NLRX1, PCBP2, and MAVS. (B) Association of PCBP2 with RIG-I and MAVS. HLCZ01 cells were transfected with plasmids encoding V5-PCBP2 and Flag–RIG-I or Flag-MAVS, followed by co-IP and immunoblotting with the indicated antibodies. (C) HCV enhances the interaction of endogenous PCBP2 with MAVS. HLCZ01 cells were infected with HCV at an MOI of 0.1 at the indicated time points. Co-IP and immunoblotting were performed with the indicated antibodies. (D) PCBP2 inhibits activation of the IFN-β promoter by MAVS and RIG-I. HLCZ01 cells expressing a gradient of V5-PCBP2 (0, 0.25, 0.5, and 1 μg) were cotransfected with IFN-β–luc, pRL-CMV, and the indicated plasmids for 24 h, followed by a dual-luciferase reporter assay. Data are represented as means ± SD from triplicate experiments. *, P < 0.05. (E) PCBP2 promotes exogenous MAVS degradation. HLCZ01 cells were cotransfected with the indicated plasmids for 48 h, followed by immunoblotting with the indicated antibodies. (F) HLCZ01 cells were transfected with increasing amounts of a plasmid encoding V5-PCBP2 for 48 h, followed by immunoblotting with the indicated antibodies. (G) Cells were treated as described above for panel F, followed by treatment with dimethyl sulfoxide or MG132 (25 μM) for 6 h. Cell lysates were analyzed by immunoblotting with the indicated antibodies. (H) HLCZ01 cells were infected with HCV (MOI of 1) for 24 h and then transfected with either the vector or the Flag-PCBP2 plasmid (0.5 μg) for 48 h. Cell lysates were analyzed by immunoblotting with the indicated antibodies. (I) The PCBP2 linker region is critical for mediating MAVS degradation. HLCZ01 cells were transfected with the indicated plasmids, followed by immunoblotting with the indicated antibodies. Blots in panels C and G were analyzed by using ImageJ software, and the data are represented as means ± SD from triplicate experiments. *, P < 0.05; **, P < 0.01; ***, P < 0.001.
PCBP2 negatively regulates the HCV-triggered innate immune response through NLRX1.Since the MAVS-mediated signaling pathway can be disturbed by PCBP2, we speculated that PCBP2 may modulate the innate immune response triggered by HCV infection. To assess this hypothesis, we compared the activation statuses of the antiviral immune response between PCBP2-silenced and control cells upon HCV infection. The data showed that the level of IFN-β expression induced by HCV was significantly higher in PCBP2-silenced cells than that in control cells in a dose-dependent manner at multiple time points after viral infection (Fig. 9A). Similarly, several representative ISGs, including ISG56 (Fig. 9B), ISG12a (Fig. 9C), ISG60 (Fig. 9D), and G1P3 (Fig. 9E), were induced more dramatically in HCV-infected cells with PCBP2 knockdown than in control cells. To test whether the enhanced activity of innate immunity with PCBP2 knockdown suppresses HCV infection, we measured the levels of HCV genomic RNA in both control and PCBP2-silenced cells. As expected, the knockdown of PCBP2 remarkably inhibited HCV replication in HLCZ01 cells (Fig. 9F). In contrast to the knockdown approach, the overexpression of PCBP2 in HLCZ01 cells also reduced the expression level of IFN-β and enhanced the replication activity of HCV RNA in a dose-dependent manner (Fig. 9G and H). Moreover, the knockdown of NLRX1 abrogated the PCBP2-mediated inhibition of the HCV-induced expression of IFN-β (Fig. 9I) as well as the enhancement of HCV infection by PCBP2 (Fig. 9J). To evaluate the role of PCBP2 in the modulation of the innate immune response induced by a virus other than HCV, we utilized NDV to conduct the experiments. The level of intracellular NDV genomic RNA in PCBP2-silenced cells with NDV infection was significant lower than that in control cells (Fig. 10A). Furthermore, the inductions of IFN-β (Fig. 10B) and several representative ISGs, including ISG12a (Fig. 10C), ISG56 (Fig. 10D), ISG60 (Fig. 10E), and G1P3 (Fig. 10F), were remarkably enhanced in PCBP2-silenced cells compared with control cells during NDV infection. These results indicated that PCBP2 acts as a negative regulator of the HCV-induced innate immune response through NLRX1. PCBP2 may also function in the modulation of the IFN signaling pathway activated by other RNA viruses.
PCBP2 negatively regulates HCV-triggered innate immune responses. (A to E) HLCZ01 cells stably expressing either sh-Ctrl or sh-PCBP2 were infected with HCV at a gradient of MOIs at the indicated time points. The kinetics of induction of IFN-β (A), ISG56 (B), ISG12a (C), ISG60 (D), and G1P3 (E) were analyzed by real-time PCR and normalized to the value for GAPDH. (F) Control and PCBP2-silenced HLCZ01 cells were infected with HCV at a gradient of MOIs at the indicated time points. The intracellular HCV RNA level was determined by real-time PCR. (G and H) HLCZ01 cells were infected with HCV for 4 days and then transfected with increasing amounts of a plasmid expressing Flag-PCBP2 for 48 h. The levels of IFN-β expression (G) and HCV RNA (H) were analyzed by real-time PCR and normalized to the value for GAPDH. (I and J) HLCZ01 cells were infected with HCV for 72 h and then transfected with the indicated plasmids for 48 h. The levels of IFN-β expression (I) and HCV RNA (J) were analyzed by real-time PCR and normalized to the value for GAPDH. Data are represented as means ± SD from triplicate experiments. *, P < 0.05; **, P < 0.01; ***, P < 0.001.
PCBP2 negatively regulates the IFN signaling pathway activated by NDV. (A) HLCZ01 cells stably expressing either sh-Ctrl or sh-PCBP2 were infected with NDV at a gradient of MOIs at the indicated time points. The intracellular NDV RNA level was determined by real-time PCR. (B to F) Cells were treated as described above for panel A. The kinetics of induction of IFN-β (B), ISG12a (C), ISG56 (D), ISG60 (E), and G1P3 (F) were analyzed by real-time PCR and normalized to the value for GAPDH. Data are represented as means ± SD from triplicate experiments. *, P < 0.05; **, P < 0.01; ***, P < 0.001.
DISCUSSION
The activation of innate immunity is essential for host cells to restrict the spread of invading viruses and other pathogens. However, the attenuation or termination of signaling is also necessary for preventing the detrimental effects of an excessive antiviral response and inflammation. RNA virus infection is recognized by cytosolic RLRs to activate the mitochondrial adaptor protein MAVS, leading to the clearance of viruses. In this process, MAVS functions as a common innate immune signaling adaptor to transduce signals from two key cytosolic PRRs, RIG-I and MDA5, and trigger the activation of the kinases TBK1 and IκB kinase (IKK), which subsequently activates IRF3 and NF-κB to produce type I and III IFNs, as well as other antiviral cytokines, for the restriction of viral infection (12, 13). Therefore, the modulation of the MAVS-mediated signaling pathway not only is important for the balance between antiviral activity and the avoidance of damage from overzealous inflammation but also may affect the outcome of RNA virus infection. In our study, we identified NLRX1 as a negative regulator of the MAVS-mediated signaling pathway during HCV infection (Fig. 11). Several lines of evidence are provided in the present study. First, HCV infection rapidly induced NLRX1 expression in hepatocytes, which is frequently observed in many other negative regulators. Second, the association of endogenous NLRX1 and MAVS proteins was found under basal conditions, which was enhanced upon HCV infection. Third, the ectopic expression or knockdown of NLRX1 attenuated or augmented the HCV-triggered innate immune response, respectively, which is consistent with the suppressed replication of HCV in NLRX1-silenced cells. Finally, NLRX1 mediated the K48-linked polyubiquitination and degradation of MAVS in a stimulus-dependent manner.
Work model for NLRX1-mediated MAVS degradation during viral infection. NLRX1 recruits PCBP2 to MAVS and induces the K48-linked polyubiquitination and degradation of MAVS, leading to the negative regulation of the IFN signaling pathway activated by HCV infection, and eventually promoting persistent viral infection.
Previously, there was no report showing that NLRX1 has a function in the regulation of protein degradation. NLRX1 is also not a component of the ubiquitin-proteasome system. Hence, we think that other host factors for the modulation of the ubiquitin-mediated proteasomal pathway may be involved in NLRX1-induced MAVS degradation. Through in silico bioinformatics predictions and experimental confirmation, we verified that PCBP2 is a NLRX1-associated partner and is indispensable for NLRX1 to mediate MAVS degradation (Fig. 11). The depletion of PCBP2 abrogated the NLRX1-triggered degradation of MAVS and the inhibition of the innate immune response in HCV-infected cells. Furthermore, the elimination of PCBP2 expression also enhanced the antiviral response in hepatocytes. PCBP2 belongs to a class of proteins that bind both RNA and DNA with a poly(rC) region. A previous study showed that PCBP2 is capable of mediating the degradation of MAVS via the E3 ubiquitin ligase AIP4 (38). Our findings suggested that PCBP2 plays a critical role in the NLRX1-mediated suppression of the anti-HCV innate immune response, which extends the functions of PCBP2 in chronic viral infections and provides a novel mechanism for the negative regulation of innate immunity. Since both NLRX1 and PCBP2 have no ubiquitin E3 ligase activities, we speculate that one or more specific ubiquitin E3 ligases may participate in NLRX1-triggered MAVS degradation. AIP4, a MAVS-associated ubiquitin E3 ligase, is a candidate for this process, calling for further investigations. In addition, several other ubiquitin E3 ligases, including RING finger protein 125 (RNF125), RNF153, and Smad ubiquitin regulatory factor 2 (SMURF2), were shown to inhibit the RLR signaling pathway by targeting MAVS for degradation following virus infection (39–41), but whether they are implicated in NLRX1-induced MAVS degradation during HCV infection is unknown.
NLRX1 localizes to the mitochondrial outer membrane (26), which provides the space to interact with a known mitochondrial outer membrane protein, MAVS. Upon HCV infection, NLRX1 may act as a scaffold to recruit PCBP2 to MAVS, which is advantageous for subsequently recruiting one or more ubiquitin E3 ligases to the NLRX1-PCBP2-MAVS complex to eventually mediate MAVS degradation. Interestingly, the triple interaction of NLRX1, PCBP2, and MAVS can also be predicted by in silico approaches. Since PCBP2 is a nucleus-residing protein (38), it is worthwhile to explore how PCBP2 translocates from the nucleus to the cytoplasm and interacts with NLRX1 and MAVS in the mitochondrion during viral infection. A recent study reported that insulin receptor tyrosine kinase substrate (IRTKS) mediates the sumoylation of PCBP2 in the nucleus during VSV infection, which causes the cytoplasmic translocation of PCBP2 for the interaction with MAVS (42). It is possible that IRTKS also functions in HCV-infected cells to change the localization of PCBP2, and the enriched cytoplasmic PCBP2 is recruited by NLRX1 to associate with MAVS for inducing MAVS degradation. Further investigation to evaluate this hypothesis is warranted.
In the NLR family, NLRX1 is unique because it does not form an inflammasome but negatively regulates inflammation. NLRX1 associates with TRAF6 and IKK in a stimulus-dependent manner, which causes the inhibition of canonical NF-κB activation (27, 28). NLRP12 and NLRC3, two other members of the NLRs, also exhibit functions in the negative regulation of the NF-κB signaling pathway (24, 43–45). In the present study, we have not explored the potential functions of NLRX1 in regulating the production of proinflammatory cytokines via NF-κB during HCV infection. In fact, it was shown previously that HCV fails to activate NF-κB signaling in plasmacytoid dendritic cells (46), suggesting that some negative regulators may be implicated in the host proinflammatory response. NLRX1 was also shown to interfere with RIG-I–MAVS signal transduction induced by SeV, VSV, and influenza A virus (26, 27). However, another study reported that NLRX1-deficient mice exhibit unaltered antiviral and inflammatory responses triggered by intranasal challenge with influenza A virus or the injection of poly(I·C) compared with wild-type mice (47). The underlying mechanisms are poorly understood. A possible mechanism that has been proposed is that NLRX1 disturbs the interaction between RIG-I and MAVS (26). In our study, we found that the overexpression of NLRX1 attenuated the endogenous RIG-I–MAVS interaction only slightly (data not shown) but mediated a remarkable reduction in the abundance of MAVS during HCV infection. Furthermore, NLRX1 also downregulated the expression of IFN-β induced by ectopically expressed MAVS. Hence, it is likely that the degradation of MAVS triggered by NLRX1 is mainly responsible for the NLRX1-mediated suppression of the innate immune response in HCV-infected cells. Apart from NLRX1, several members of the NLR family also act as negative regulators of the host IFN production pathway. For instance, NLRP4 negatively regulates the innate immune response by inducing TBK1 degradation via a ubiquitin ligase, DTX4 (48). NLRC3 interacts with both STING and TBK1 and disturbs STING translocation and STING-TBK1 association to block type I IFN production (49). Interestingly, NLRX1 also inhibits the STING-mediated signaling pathway by impeding the STING-TBK1 interaction (29). The mechanism discovered in the present study, where NLRX1 attenuates virus-induced IFN signaling by targeting MAVS for degradation, presents a new perspective for understanding the action of the host innate immune system.
Both type I and type III IFNs efficiently inhibit HCV propagation in vitro, and IFN-α-based therapy is capable of curing 40 to 80% of HCV patients (13, 50), suggesting that the activation of the IFN signaling pathway is essential for controlling HCV infection. Although the negative regulation of innate immune signaling is necessary for preventing immune-mediated tissue damage and spontaneous autoimmunity (51), some negative regulators may exert excessive actions to downregulate the antiviral immune response to infections by HCV and other chronic pathogens. The depletion of NLRX1, as well as PCBP2, significantly inhibited HCV propagation in hepatocytes along with the restoration of IFN signaling, indicating that NLRX1 and PCBP2 possibly play roles in HCV pathogenesis.
In summary, the present study identified NLRX1 as a negative regulator of the MAVS-mediated signaling pathway during HCV infection. Mechanistically, NLRX1 recruits PCBP2 to MAVS and induces the K48-linked polyubiquitination and degradation of MAVS, leading to the blocking of the IFN signaling pathway to promote HCV infection. PCBP2 is indispensable for the negative-regulation function of NLRX1. The NOD of NLRX1 is required for NLRX1 to interact with PCBP2 and subsequently induce MAVS degradation. Targeting NLRX1 and PCBP2 may offer a promising strategy to restore antiviral innate immune responses in the host for restricting viral infection.
MATERIALS AND METHODS
Cell culture and reagents.HLCZ01 cells were established in our laboratory (33). Huh7.5 cells were kindly provided by Charles M. Rice (Rockefeller University, New York, NY). Huh7.5.1-MAVS-WT and Huh7.5.1-MAVSR cells were gifts from Jin Zhong (Institut Pasteur of Shanghai, China). HEK293T cells were purchased from Boster. HLCZ01 cells were cultured in collagen-coated tissue culture plates and cultured with Dulbecco's modified Eagle medium (DMEM)–F-12 medium supplemented with 10% (vol/vol) fetal bovine serum (FBS) (Gibco), 40 ng/ml of dexamethasone (Sigma), insulin-transferrin-selenium (ITS) (Lonza), penicillin, and streptomycin. Huh7.5 and HEK293T cells were propagated in DMEM supplemented with 10% FBS, l-glutamine, nonessential amino acids, penicillin, and streptomycin.
Plasmids and antibodies.The shRNAs targeting NLRX1 and PCBP2 were constructed in the pSilencer-neo plasmid (Ambion). The target sequence of the NLRX1 shRNA was 5′-GCACATCTTCCGTCGGGAT-3′. The target sequence of the PCBP2 shRNA was 5′-AGGAGAATCAGTTAAGAAG-3′. NLRX1, MAVS, and PCBP2 cDNAs were synthesized from total cellular RNA isolated from HLCZ01 cells by standard reverse transcription-PCR (RT-PCR) and were subsequently cloned into the pcDNA3.1a vector and/or the p3×FLAG-CMV vector. Multiple domains of NLRX1, MAVS, and PCBP2 were amplified from their full-length templates and cloned into the p3×FLAG-CMV or pcDNA3.1a vector. The primers for the above-described genes or the amplification of domains are shown in Tables 1 to 4. The NLRX1 shRNA-resistant mutant was constructed from the pcDNA3.1a-NLRX1 plasmid by using a QuikChange site-directed mutagenesis kit (Agilent Genomics). The mutation antisense primer and sense primer are 5′-CAGGAAAAACCTCAGGGCATCTCTCCTGAAGATGTGCAGCAGCTGAAA-3′ and 5′-TTTCAGCTGCTGCACATCTTCAGGAGAGATGCCCTGAGGTTTTTCCTG-3′, respectively. The pHA-Ub(K48) and pLV-MAVS-C508R-FLAG plasmids were kindly provided by Zhengfan Jiang (Peking University, Beijing, China) and Jin Zhong (Institut Pasteur of Shanghai, China), respectively. Monoclonal antibodies against β-actin, the Flag tag, and the HA tag were obtained from Sigma. V5 tag monoclonal antibodies were purchased from Invitrogen. Monoclonal antibodies against MAVS, NLRX1, PCBP2, and RIG-I were purchased from Santa Cruz Biotechnology. The STAT1, TBK1, phosphorylated TBK1, and phosphorylated STAT1 antibodies were obtained from Cell Signaling Technology. Mouse monoclonal anti-NS5A and anti-core antibodies were kindly provided by Chen Liu (Rutgers University, Newark, NJ). Goat anti-mouse and goat anti-rabbit IgG-horseradish peroxidase (HRP) secondary antibodies were purchased from Santa Cruz Biotechnology.
Primers for amplification of genes
Primers for amplification of MAVS domains
Primers for amplification of NLRX1 domains
Primers for amplification of PCBP2 domains
Production of HCV stocks.The pJFH1 and pJFH1/GND plasmids were gifts from Takaji Wakita (National Institute of Infectious Diseases, Tokyo, Japan). The linearized DNAs from the pJFH1 and pJFH1/GND plasmids were purified and used as the templates for in vitro transcription using a MEGAscript kit (Ambion, Austin, TX). In vitro-transcribed genomic JFH1 or JFH1/GND RNA was delivered into Huh7.5 cells by electroporation. The transfected cells were transferred to complete DMEM and cultured for the indicated periods. Cells were passaged every 3 to 5 days, and the corresponding supernatants were collected and filtered with a 0.45-μm filter device. The viral titers are presented as focus-forming units per milliliter, determined by the average number of NS5A-positive foci detected in Huh7.5 cells.
Real-time PCR assay.Total cellular RNA was extracted by using TRIzol (Invitrogen, Carlsbad, CA) according to the manufacturer's protocol. The Superscript III first-strand synthesis kit for reverse transcription of RNA was purchased from Invitrogen. After RQ1 DNase (Promega) treatment, the extracted RNA was used as the template for RT-PCR. Real-time PCR was performed as described previously (33). For the absolute quantification of HCV RNA, a standard curve was established by using in vitro-transcribed JFH1 RNA. The primers used for the detection of HCV genomic RNA were described previously (33).
Western blot and immunofluorescence analyses.The Western blotting procedure was reported previously (52). Briefly, cells were washed with phosphate-buffered saline (PBS) and lysed in radioimmunoprecipitation assay (RIPA) buffer (150 mM sodium chloride, 1% Nonidet P-40, 0.5% sodium deoxycholate, 0.1% SDS, and 50 mM Tris-HCl [pH 8.0] supplemented with 2 μg/ml of aprotinin, 2 μg/ml of leupeptin, 20 μg/ml of phenylmethylsulfonyl fluoride, and 2 mM dithiothreitol [DTT]). Forty micrograms of protein was resolved by using an SDS-PAGE gel, transferred onto a polyvinylidene difluoride (PVDF) membrane, and probed with the appropriate primary and secondary antibodies. The bound antibodies were detected by using the SuperSignal West Pico chemiluminescent substrate (Pierce, Rockford, IL). Quantification of blots was performed by using ImageJ software. The protocol for immunofluorescence was described previously (33).
Luciferase reporter assay.Luciferase assays were performed by using a luciferase assay kit (Promega) according to the manufacturer's instructions. The luciferase activity in cells was normalized to protein concentrations as determined by Bradford assays (53).
Coimmunoprecipitation and immunoblotting.Co-IP and immunoblotting procedures were reported previously (35). Briefly, cells were washed three times with ice-cold PBS and lysed in lysis buffer supplemented with a protease inhibitor cocktail. The cell lysates were incubated at 4°C for 30 min and centrifuged at 12,000 × g at 4°C for 15 min. The lysates were diluted to a concentration of 2 μg/μl with PBS before IP. Two hundred micrograms of lysates was immunoprecipitated with the indicated antibodies. The immunocomplexes were captured by adding a protein G-agarose bead slurry to the lysates. The protein bound to the beads was boiled in 2× Laemmli sample buffer and then subjected to 10% SDS-PAGE. The protocol for immunoblotting is described above.
Bioinformatics analysis of protein-protein interactions.To find candidate proteins interacting with NLRX1, the STRING biological database was used to predict the interactions between NLRX1, as well as MAVS, and other proteins (http://string-db.org/ ). The minimum required interaction score was set at the highest confidence level (0.900). The top 20 output targets contain both NLRX1 and PCBP2 in the prediction, dependent on MAVS. We applied in silico approaches to estimate functional domains involved in the NLRX1-PCBP2 interaction and the NLRX1-MAVS interaction. As of April 2017, only structural fragments of NLRX1 (PDB accession number 3UN9 ), PCBP2 (PDB accession numbers 2JZX and 2P2R ), and MAVS (PDB accession number 2VGQ ) were crystallized. First, we used full-length amino acid sequences of NLRX1, PCBP2, and MAVS as inputs for structure prediction through the I-TASSER online server (http://zhanglab.ccmb.med.umich.edu/I-TASSER/ ). The ResProx server was then applied to select the potential models for our analysis. Second, since NLRX1 and MAVS contain transmembrane domains (amino acid positions 365 to 385 in NLRX1 and 515 to 540 in MAVS), we used Chimera V1.1 to reassemble the transmembrane domains and to minimize the structural energy for the improvement of localized atomic structures, and the MolProbity server was thereafter applied to fix the outliers as part of the refinement cycle. Third, Cluspro V2.0 (http://cluspro.bu.edu/ ) was applied to estimate the structural interfaces that mediate the NLRX1-PCBP2 interaction and the NLRX1-MAVS interaction. The protein-protein interaction databases BioGrid (http://thebiogrid.org/ ) and IntAct (http://www.ebi.ac.uk/intact ) were also queried for proteins that potentially interact with either NLRX1, PCBP2, or MAVS. This process resulted in possible models that matched well with data in the literature and structural constraints, while the final models were selected based on the highest structural scores. The structural model of lipid bilayers of the extracellular membrane was added by using Chimera V1.1. Our structural visualization was undertaken by using PyMOL V1.7 (http://www.pymol.org/ ).
Statistical analysis.All results are presented as means ± standard deviations (SD). Comparisons between two groups were performed by using Student's t test.
ACKNOWLEDGMENTS
We thank Charles M. Rice for the Huh7.5 cell line, Takaji Wakita for the pJFH1 plasmid, Zhengfan Jiang for the pHA-Ub(K48) plasmid, and Jin Zhong and Chen Liu for sharing research materials.
This work was supported by the National Natural Science Foundation of China (grants 81730064, 81571985, 81271885, and 31571368), the National Science and Technology Major Project of the Ministry of Science and Technology of China (grants 2017ZX10202201-005 and 2009ZX10004-312), and the Project of Innovation-Driven Plan of Central South University (grant 2016CX031).
FOOTNOTES
- Received 5 August 2017.
- Accepted 20 September 2017.
- Accepted manuscript posted online 27 September 2017.
- Copyright © 2017 American Society for Microbiology.
