The von Hippel–Lindau (VHL) gene is a well-defined tumor suppressor linked to human heredity cancer syndromes. As a component of the VHL-elongin B/C E3 ligase complex, pVHL performs its tumor function by targeting proteins for proteasomal degradation. It is largely unknown whether pVHL functions in antiviral immunity. In this article, we identify that pVHL negatively regulates innate antiviral immunity, which acts mainly by inducing degradation of mitochondrial antiviral-signaling protein (MAVS, also known as Cardif, IPS-1, or VISA). Overexpression of pVHL abrogated the cellular response to viral infection, whereas knockdown of pVHL exerted the opposite effect. pVHL targeted the K420 residue of MAVS to catalyze the formation of K48-linked polyubiquitin chains, leading to proteasomal degradation of MAVS. After viral infection, Mavs levels remained low in wild type zebrafish embryos but became much higher in vhl-deficient (vhl−/−) zebrafish embryos. Higher MAVS levels correlated with a greatly exaggerated antiviral response. In this work, we demonstrate that pVHL exhibits a previously unknown role in innate antiviral immunity.

The von Hippel–Lindau (VHL) gene is a classic tumor suppressor. Inactivation of VHL is linked to human hereditary VHL diseases that are autosomal-dominant, neoplastic diseases (1). VHL diseases are associated with various tumor types, including clear cell renal cell carcinomas, CNS and retinal hemangioblastomas, pheochromocytomas, and pancreatic neuroendocrine tumors, in addition to pancreatic and renal cysts (2, 3). The VHL tumor suppressor protein (pVHL) forms a ternary complex with elongin C and elongin B termed VHL-elongin B/C (VCB) complex, which resembles the yeast Skp1-Cullen-1/Cdc53-F-box protein complex (4). The VCB complex and the similar Skp1-Cullen-1/Cdc53-F-box protein complex both possess ubiquitin ligase activities and are capable of targeting proteins for proteasomal degradation.

The best-characterized function of pVHL is acting as a substrate recognition subunit of the VCB E3 ligase complex that targets the proline hydroxylated hypoxia-induced factor (HIF)-1/2α for proteasomal degradation under hypoxia conditions, which is thought to be the major pVHL mechanism for tumor suppression (5). Targets other than HIF-α have been tentatively identified, including estrogen receptor α (6), Kruppel-like factor receptor 4 (7), ERK5 (8), and androgen receptor (9), to rationalize the multiple symptoms exhibited in VHL disease. However, apart from HIF-α, the mechanisms and downstream effects of pVHL-induced ubiquitination of non–HIF-α targets are still obscure. Identification of additional targets of pVHL action and elucidation of their biological effects will shed new light on the physiological functions of pVHL.

Mitochondrial antiviral-signaling protein (MAVS, also known as Cardif, IPS-1, or VISA) was the first identified mitochondrial protein linked to innate antiviral immunity (1013). As an adaptor protein, MAVS anchors itself to the mitochondrial outer membrane, where it interacts with Mda5 or RIG-1 sensor proteins that had been triggered by viral infection (14). Mda5 and RIG-1 are two cytoplasmic RNA sensors that recognize viral RNA released during viral replication (15). The interaction of the N-terminal CARD domain of MAVS with RIG-1 CARDs or Mda5 CARDs activates MAVS. Activated MAVS then signals downstream to the kinases IKK and TBK1, which induce the activities of transcription factors, including NF-κB and IFN regulatory factor 3 (IRF3), eventually leading to the production of type I IFNs and proinflammatory cytokines (14, 16).

Studies have shown that ubiquitination-mediated protein degradation plays a crucial role in the modulation of MAVS-mediated signaling (17). This regulation is initiated after three enzymes catalyze the covalent attachment of ubiquitin moieties to MAVS: a ubiquitin-activating enzyme (E1), a ubiquitin-conjugating enzyme (E2), and a ubiquitin ligase (E3) that recognizes MAVS specifically (17). E3 ligase RNF125 can ubiquitinate and trigger degradation of Mda5, RIG-1, and MAVS to suppress the RIG-1–like-helicase–mediated response (18). The E3 ligase TRIM25 induces ubiquitination and proteasomal degradation of MAVS to suppress type I IFN production (19). The E3 ligase Smurf2 negatively modulates the antiviral response by targeting MAVS for proteasomal degradation (20). In addition, some factors have been shown to mediate degradation of MAVS through their effects on other E3 ligases. PCBP2 recruits the E3 ligase AIP4 to polyubiquitinate and degrade MAVS (21). Ndfip1 negatively modulates the antiviral response by enhancing Smurf1 E3 ligase–mediated MAVS degradation (22).

As a classic tumor suppressor with E3 ligase activity, pVHL has not previously been linked to the antiviral response. In this study, we identify pVHL as a negative regulator in MAVS-mediated antiviral signaling. Overexpression of pVHL abrogated the cellular response to viral infection, whereas knockdown of pVHL exerted the opposite effect. pVHL targeted the K420 residue of MAVS to catalyze the formation of K48-linked polyubiquitin chains, leading to proteasomal degradation of MAVS. Mavs levels remained low in wild type (WT) zebrafish embryos after viral infection but became much higher in vhl-deficient (vhl−/−) zebrafish embryos; the latter result was accompanied by a greatly exaggerated antiviral response. Collectively, the data presented in this work point to a previously unknown function of pVHL in innate antiviral immunity.

HEK293T, Cos7, and HCT116 cells were originally obtained from American Type Culture Collection and were cultured in DMEM (HyClone) with 10% FBS and grown at 37°C in a humidified incubator containing 5% CO2.

RIG-1, Mda5, MAVS, and TBK1 expression vectors were purchased from Addgene and subcloned into different expression vectors by PCR amplification. IRF3 and mouse MAVS were amplified by PCR from human or mouse cDNA pool. IRF-E reporter was kindly provided by Hong-Bing Shu (23). VHL expression vectors, pSUPER-VHL–short hairpin RNA (shRNA)-1 and pSUPER-VHL-shRNA2, have been described previously (9). Zebrafish mavs and vhl were amplified by PCR from zebrafish cDNA pool.

The Abs used are as follows: Anti-Myc (9E10; Santa Cruz), anti-GAPDH (0411; Santa Cruz), Anti-HA (Covance), Anti-Flag (F1804; Sigma), Anti-α-tubulin (EPR1333; Epitomics), anti-MAVS (166583; Santa Cruz), VHL (ABclonal). Zebrafish Mavs Ab (anti-DrMavs) was developed by injecting rabbits with GST-tagged partial zebrafish Mavs.

Mito-Tracker Green kit was purchased from Beyotime Institute of Biotechnology (Haimen, Jiangsu). MG132 was purchased from Calbiochem. Ni-NTA beads were purchased from Novagen.

Sendai virus (SeV), vesicular stomatitis virus (VSV)-GFP, and HSV-1 viruses were propagated and amplified by infection of Vero cells. Postinfection for 24 h, the supernatant was collected and was clarified by centrifugation. Viral titers were determined by plaque assay on Vero cells. SeV was provided by Xinwen Chen, VSV-GFP was provided by Minzhou Chen, and HSV-1 was provided by Chunfu Zheng.

For HEK293T cell infection, the cells were incubated with DMEM without FBS containing SeV, HSV-1, or VSV-GFP at multiplicity of infection (MOI) from 1 to 100. Postinfection for 1 h, the cells were washed twice and then were cultured cells in DMEM containing 1–10% FBS.

Cells were grown in 24-well plates and transfected with various amounts of plasmids by VigoFect (Vigorous Biotech, Beijing, China), as well as with pTK-Renilla used as an internal control. After the cells were transfected for 18–24 h, the luciferase activity was determined by the dual-luciferase reporter assay system (Promega). Data were normalized to Renilla luciferase. Data are reported as mean ± SEM from three independent experiments performed in triplicate. The statistical analysis was performed using GraphPad Prism 5 (unpaired t test; GraphPad Software, San Diego, CA).

Coimmunoprecipitation and Western blot analysis were performed as described previously (24). Anti-Flag and anti-HA Ab–conjugated agarose beads were purchased from Sigma. The Fuji Film LAS4000 mini-luminescent image analyzer was used to photograph the blots. Multi Gauge V3.0 was used for quantifying the protein levels based on the band density obtained in Western blot analysis.

The total RNA was extracted from HEK293T cells by TRIzol reagent (Invitrogen) and the total RNA extracted from zebrafish embryos by SV Total RNA Isolation System (Promega). The cDNA synthesis was carried out using a first-strand cDNA synthesis kit (Fermentas).

The primers used for RT-PCR are as follows: human IFN-β: 5′-CATTACCTGAAGGCCAAGGA-3′ and 5′-CAATTGTCCAGTCCCAGAGG-3′; human IFN-stimulated gene 15 (ISG15): 5′-ATGGGCTGGGACCTGACGG-3′ and 5′- TTAGCTCCGCCCGCCAGGCT-3′; human GAPDH (internal control): 5′-AGCCACATCGCTCAGACAC-3′ and 5′-GCCCAATACGACCAAATCC-3′; zebrafish ifn1: 5′-AAGTTTTTAGTCCTGACATTGGATCA-3′ and 5′-TCCCAGTTCACCGAGTTCATG-3′; zebrafish ifn3: 5′-TTCTGCTTTGTGCAGGTTTG-3′ and 5′-GGTATAGAAACGCGGTCGTC-3′; zebrafish ef1α (internal control): 5′-CTTCTCAGGCTGACTGTGC-3′ and 5′-CCGCTAGCATTACCCTCC-3′ (25).

The Cos-7 cells were transfected with RFP-tagged VHL expression vector for 24 h, then were infected with SeV for 8 h. Subsequently, the cells were incubated with the MitoTracker Green (Beyotime Institute of Biotechnology) for 45 min and were photographed under a Zeiss confocal microscope (LSM-710).

One pair of zebrafish vhl knockout TALEN was designed based on the exon 1 sequence of zebrafish vhl gene. The left target sequence is 5′-CAACGTTCTGTTCTGT-3′; the right target sequence is 5′-CCAGACGGGCTTGAC-3′. Vhl-TALENs were constructed by FastTALE TALEN Assembly kit (Sidansai Biotechnology, Shanghai, China) and the mRNAs of vhl-TALENs were synthesized by Amplicap Sp6 High Yield Message Maker kit (Cell Script).

The synthesized mRNA was injected into zebrafish embryos at one-cell stage. For determining the efficiency of TALEN-mediated vhl knockout, the total DNA was extracted from injected embryos at 24-h postfertilization stage; PCR amplification was conducted and the products were checked by nondenature PAGE gel. If the positive results were observed, this batch of embryos was maintained until they matured (F0). F0 fishes were cross-bred with WT zebrafish to obtain F1. If F1 zebrafish were confirmed that one copy of vhl was mutated by genotyping, they were maintained for generating F2. DNA was extracted from the F2 tails, and PCR products were examined by nondenature PAGE gel assay at first; then the positive PCR products were further confirmed by sequence analysis. The vhl heterozygous zebrafish (vhl+/−) were bred for generating vhl homozygous null embryos (vhl−/−). The vhl homozygous null embryos could be selected by their morphology after 5 d postfertilization (dpf), which were also further genotyped by sequence analysis.

Spring viremia of carp virus (SVCV) were propagated in carp epithelium papulosum cyprinid cells (originally obtained from American Type Culture Collection) at 28°C. After one round of frozen (−70°C) and thawed, the supernatant was collected and was clarified by centrifugation. Viral titers were determined by 50% tissue culture infection dose assay (TCID50) assay on epithelium papulosum cyprinii cells. Vhl heterozygous zebrafish (vhl+/−) were mated and vhl homozygous null embryos (vhl−/−) were selected based on morphology at 5-dpf stage. Two mutant lines were used for viral infection assays, and their corresponding sibling WT embryos (vhl+/+) were used for the control. For viral infection of zebrafish embryos, ∼2 × 108 TCID50/ml SVCV were added to fish water in six-well plates at triplicate.

In an attempt to define the role of pVHL in viral infection, we initially performed luciferase reporter assays using IRF-E (23), a reporter driven by five copies of IRF-E of the ISG54 gene promoter, which is activated by IRF3 (26, 27). As expected, SeV infection (MOI = 1) enhanced IRF-E reporter activity dramatically in HEK293T cells (Fig. 1A). However, ectopic expression of pVHL inhibited the enhancement of activity of the IRF-E reporter by SeV infection (Fig. 1A). This phenomenon suggests that pVHL might negatively regulate cellular innate antiviral immune response.

FIGURE 1.

pVHL negatively regulates MAVS-mediated signaling. (A) pVHL suppresses IRF-E reporter activity activated by SeV infection in a dose-dependent manner in HEK293T cells. HEK293T cells were transfected with the IRF-E luciferase reporter (0.1 μg/well) together with an increasing amount of Myc-VHL vector (0.1, 0.2, and 0.4 μg/well). After 24 h, the cells were infected by SeV (MOI = 1) for 8 h and then luciferase assays were performed. (B) pVHL suppresses IRF-E reporter activity activated by overexpression of RIG-1 in a dose-dependent manner in HEK293T cells. HEK293T cells were transfected with IRF-E luciferase reporter (0.1 μg/well) and Flag-RIG-1 vector (0.2 μg/well) together with an increasing amount of Myc-VHL vector (0.1, 0.2, and 0.4 μg/well). After 18–24 h, luciferase assays were performed. (C) pVHL suppresses IRF-E reporter activity resulting from overexpression of Mda5 in a dose-dependent manner in HEK293T cells. HEK293T cells were transfected with IRF-E luciferase reporter (0.1 μg/well) and Flag-MDA5 vector (0.2 μg/well) together with an increasing amount of Myc-VHL vector (0.1, 0.2, and 0.4 μg/well). After 18–24 h, luciferase assays were performed. (D) pVHL suppresses IRF-E reporter activity resulting from overexpression of MAVS in a dose-dependent manner in HEK293T cells. HEK293T cells were transfected with IRF-E luciferase reporter (0.1 μg/well) and Flag-MAVS vector (0.2 μg/well) together with an increasing amount of Myc-VHL vector (0.1, 0.2, and 0.4 μg/well). After 18–24 h, luciferase assays were performed. (E) pVHL has no obvious effect on IRF-E reporter activity resulting from overexpression of TBK1 in HEK293T cells. HEK293T cells were transfected with IRF-E luciferase reporter (0.1 μg/well) and HA-TBK1 vector (0.2 μg/well) together with an increasing amount of Myc-VHL vector (0.1, 0.2, and 0.4 μg/well). After 18–24 h, luciferase assays were performed. (F) pVHL has no obvious effect on IRF-E reporter activity activated by overexpression of IRF3 in HEK293T cells. HEK293T cells were transfected with IRF-E luciferase reporter (0.1 μg/well) and Myc-IRF3 vector (0.2 μg/well) together with an increasing amount of Myc-VHL vector (0.1, 0.2, and 0.4 μg/well). After 18–24 h, luciferase assays were performed. Data are presented as mean ± SEM of three independent experiments performed in triplicate.

FIGURE 1.

pVHL negatively regulates MAVS-mediated signaling. (A) pVHL suppresses IRF-E reporter activity activated by SeV infection in a dose-dependent manner in HEK293T cells. HEK293T cells were transfected with the IRF-E luciferase reporter (0.1 μg/well) together with an increasing amount of Myc-VHL vector (0.1, 0.2, and 0.4 μg/well). After 24 h, the cells were infected by SeV (MOI = 1) for 8 h and then luciferase assays were performed. (B) pVHL suppresses IRF-E reporter activity activated by overexpression of RIG-1 in a dose-dependent manner in HEK293T cells. HEK293T cells were transfected with IRF-E luciferase reporter (0.1 μg/well) and Flag-RIG-1 vector (0.2 μg/well) together with an increasing amount of Myc-VHL vector (0.1, 0.2, and 0.4 μg/well). After 18–24 h, luciferase assays were performed. (C) pVHL suppresses IRF-E reporter activity resulting from overexpression of Mda5 in a dose-dependent manner in HEK293T cells. HEK293T cells were transfected with IRF-E luciferase reporter (0.1 μg/well) and Flag-MDA5 vector (0.2 μg/well) together with an increasing amount of Myc-VHL vector (0.1, 0.2, and 0.4 μg/well). After 18–24 h, luciferase assays were performed. (D) pVHL suppresses IRF-E reporter activity resulting from overexpression of MAVS in a dose-dependent manner in HEK293T cells. HEK293T cells were transfected with IRF-E luciferase reporter (0.1 μg/well) and Flag-MAVS vector (0.2 μg/well) together with an increasing amount of Myc-VHL vector (0.1, 0.2, and 0.4 μg/well). After 18–24 h, luciferase assays were performed. (E) pVHL has no obvious effect on IRF-E reporter activity resulting from overexpression of TBK1 in HEK293T cells. HEK293T cells were transfected with IRF-E luciferase reporter (0.1 μg/well) and HA-TBK1 vector (0.2 μg/well) together with an increasing amount of Myc-VHL vector (0.1, 0.2, and 0.4 μg/well). After 18–24 h, luciferase assays were performed. (F) pVHL has no obvious effect on IRF-E reporter activity activated by overexpression of IRF3 in HEK293T cells. HEK293T cells were transfected with IRF-E luciferase reporter (0.1 μg/well) and Myc-IRF3 vector (0.2 μg/well) together with an increasing amount of Myc-VHL vector (0.1, 0.2, and 0.4 μg/well). After 18–24 h, luciferase assays were performed. Data are presented as mean ± SEM of three independent experiments performed in triplicate.

Close modal

Cells use pattern-recognition receptors to sense viral nucleic acids. RIG-1–like helicases, including RIG-1 and Mda5, function as cytoplasmic RNA sensors and share a common adaptor protein MAVS. Upon viral infection, RIG-1–like helicases bind to viral RNA and activate MAVS, which, in turn, activates the IRF3 and NF-κB to induce type I IFN expression. TBK1 acts as an upstream factor of IRF3 and is required for induction of IFN-α triggered by overexpression of MAVS (12). To define the specific step whereby pVHL affects the cellular antiviral innate immune signaling, we performed subsequent coexpression experiments. When RIG-1, Mda5, MAVS, TBK1, or IRF3 were overexpressed in HEK293T cells without coexpression of pVHL, the activity of an IRF-E reporter was upregulated significantly (Fig. 1B–F). However, coexpression of pVHL together with RIG-1, MDA5, or MAVS appreciably suppressed the enhancement of IRF-E activity observed when no pVHL was present (Fig. 1B–D). In contrast, coexpression of pVHL together with TBK1 or IRF3 had no obvious inhibitory effect on the activity of IRF-E upregulated by TBK1 or IRF3 (Fig. 1E, 1F). These results imply that pVHL might negatively regulate cellular innate antiviral immune response through its effect on MAVS, RIG-1, or MDA5, but pVHL absolutely does not suppress this response through any effect on TBK1 and IRF3.

Given that pVHL has been well defined as a component of the VHL-elongin B/C E3 ligase complex, we examined whether pVHL could affect the protein stabilities of RIG-1, MDA5, and MAVS. Coexpression of pVHL together with RIG-1 or MDA5 resulted in a mild reduction of RIG-1 or Mda5 protein levels, but coexpression of pVHL together with MAVS caused a dramatic reduction in MAVS protein levels, consistent with pVHL’s suppressive effect on the cellular antiviral innate immune response (Fig. 2A). Because MAVS acts as the common adaptor used by RIG-1 and MDA5 to mediate the cellular antiviral innate immune response, we focused our subsequent assays on the effect of pVHL on MAVS. As shown in Fig. 2B, coexpression of either human MAVS or mouse MAVS together with an increasing amount of human pVHL in HEK293T cells resulted in a dramatic reduction of MAVS protein levels. Furthermore, overexpression of pVHL in HEK293T cells also caused reduction of endogenous MAVS induced by SeV infection (Fig. 2C). Notably, knockdown of endogenous pVHL by VHL–shRNA-1 and VHL–shRNA-2 in HEK293T cells caused endogenous MAVS to increase, as compared with the control (GFP-shRNA; Fig. 2D). Taken together, these results suggest that pVHL promotes MAVS degradation.

FIGURE 2.

pVHL induces the degradation of MAVS. (A) Overexpression of HA-VHL (2 μg/well) caused ectopically expressed MAVS (2 μg/well) to be dramatically reduced in HEK293T cells. (B) Overexpression of pVHL (2 μg/well) caused ectopically expressed human MAVS (2 and 3 μg/well), as well as mouse MAVS (2 and 3 μg/well), to be reduced in a dose-dependent manner in HEK293T cells. (C) The endogenous MAVS induced by SeV infection (MOI = 1) in HEK293T cells was reduced by ectopic expression of VHL in a dose-dependent manner (0.5, 1, 2, and 4 μg/well). Lane 1, without SeV addition and pVHL overexpression; lanes 2–6, with SeV addition and an increasing amount of pVHL overexpression. (D) Knockdown of endogenous VHL in HCT116 cells by VHL–shRNA-1 and VHL–shRNA-2 (4 μg/well) caused an increase in endogenous MAVS. Data are presented based on three independent experiments. IB, immunoblotting.

FIGURE 2.

pVHL induces the degradation of MAVS. (A) Overexpression of HA-VHL (2 μg/well) caused ectopically expressed MAVS (2 μg/well) to be dramatically reduced in HEK293T cells. (B) Overexpression of pVHL (2 μg/well) caused ectopically expressed human MAVS (2 and 3 μg/well), as well as mouse MAVS (2 and 3 μg/well), to be reduced in a dose-dependent manner in HEK293T cells. (C) The endogenous MAVS induced by SeV infection (MOI = 1) in HEK293T cells was reduced by ectopic expression of VHL in a dose-dependent manner (0.5, 1, 2, and 4 μg/well). Lane 1, without SeV addition and pVHL overexpression; lanes 2–6, with SeV addition and an increasing amount of pVHL overexpression. (D) Knockdown of endogenous VHL in HCT116 cells by VHL–shRNA-1 and VHL–shRNA-2 (4 μg/well) caused an increase in endogenous MAVS. Data are presented based on three independent experiments. IB, immunoblotting.

Close modal

To gain a more complete picture of pVHL in cellular antiviral innate immunity, we further examined the effect of pVHL on RIG-1 and MDA5. Coexpression of RIG-1 together with an increasing amount of pVHL in HEK293T cells resulted in a reduction of RIG-1 protein levels (Supplemental Fig. 1A). Notably, the reduction of RIG-1 by pVHL was not as dramatic as that of MAVS by pVHL (Fig. 2B versus Supplemental Fig. 1A). However, coexpression of MDA5 together with an increasing amount of pVHL in HEK293T cells did not cause an obvious reduction of MDA5 protein levels (Supplemental Fig. 1D). These results suggest that pVHL might also promote RIG-1 degradation, but not of MDA5.

To determine whether pVHL acts as an E3 ligase for mediating MAVS degradation, we subsequently examined the interaction between pVHL and MAVS. We transfected Myc-tagged VHL (Myc-VHL) with Flag-tagged MAVS (Flag-MAVS) or Flag empty vector into HEK293T cells and then performed coimmunoprecipitation assays. Flag-MAVS could efficiently pull down Myc-VHL (Fig. 3A), indicating that pVHL interacts with MAVS. To determine whether these proteins interact endogenously, we performed coimmunoprecipitation assays from HEK293T cell extracts in the absence or presence of SeV infection using a polyclonal Ab against pVHL. The results show that endogenous MAVS could interact with pVHL when the cells were infected by SeV (MOI = 1). For unknown reasons, we could not express MAVS in bacteria in our laboratory (data not shown); therefore, we could not determine whether pVHL interacts with MAVS directly.

FIGURE 3.

pVHL interacts with MAVS. (A) Ectopically expressed Myc-VHL interacts with ectopically expressed Flag-MAVS. HEK293T cells were transfected with the indicated expression vectors (3 μg/well), and coimmunoprecipitation was performed with anti-Flag (M2) Ab. (B) Endogenous VHL interacts with endogenous MAVS after viral infection (SeV, MOI = 1). Anti-VHL Ab was used for coimmunoprecipitation, and rabbit Ig (IgG) was used as control. (C) Schematic of MAVS domains interacted with pVHL. The extent of the interaction is indicated by the number of plus (+) signs; minus sign (−) indicates no interaction. (D and E) Domain mapping for MAVS domains interacting with pVHL. HEK293T cells were transfected with the indicated expression vectors (3 μg/well), and anti-Flag Ab was used for coimmunoprecipitation. (F) Schematic of pVHL domains interacted with MAVS. The extent of the interaction is indicated by the number of plus (+) signs. (G) Domain mapping for pVHL domains interacting with MAVS. HEK293T cells were transfected with the indicated expression vectors (3 μg/well), and anti-Flag Ab was used for coimmunoprecipitation. IB, immunoblotting; IP, immunoprecipitation; WCL, whole cell lysates.

FIGURE 3.

pVHL interacts with MAVS. (A) Ectopically expressed Myc-VHL interacts with ectopically expressed Flag-MAVS. HEK293T cells were transfected with the indicated expression vectors (3 μg/well), and coimmunoprecipitation was performed with anti-Flag (M2) Ab. (B) Endogenous VHL interacts with endogenous MAVS after viral infection (SeV, MOI = 1). Anti-VHL Ab was used for coimmunoprecipitation, and rabbit Ig (IgG) was used as control. (C) Schematic of MAVS domains interacted with pVHL. The extent of the interaction is indicated by the number of plus (+) signs; minus sign (−) indicates no interaction. (D and E) Domain mapping for MAVS domains interacting with pVHL. HEK293T cells were transfected with the indicated expression vectors (3 μg/well), and anti-Flag Ab was used for coimmunoprecipitation. (F) Schematic of pVHL domains interacted with MAVS. The extent of the interaction is indicated by the number of plus (+) signs. (G) Domain mapping for pVHL domains interacting with MAVS. HEK293T cells were transfected with the indicated expression vectors (3 μg/well), and anti-Flag Ab was used for coimmunoprecipitation. IB, immunoblotting; IP, immunoprecipitation; WCL, whole cell lysates.

Close modal

To determine which domains of MAVS bind to pVHL, we performed domain mapping using coimmunoprecipitation assays. The C terminus of MAVS (aa 361–540) was observed to interact with pVHL and required the transmembrane domain of MAVS for the interaction (Fig. 3A–E). Moreover, we mapped the domains of pVHL that bind to MAVS. The full-length pVHL could interact with MAVS, but the C terminus appeared to be necessary for strong binding to MAVS (Fig. 3F, 3G). Together, these results suggest that pVHL interacts with MAVS.

In addition, we further confirmed that pVHL could also interact with RIG-1 and MDA5 (Supplemental Fig. 1B, 1E).

To determine whether pVHL mediates proteasomal degradation of MAVS, we initially used the proteasome inhibitor, MG132, to examine whether it could block pVHL-mediated MAVS degradation. As shown in Fig. 4A, addition of MG132 could effectively block degradation of MAVS induced by pVHL. Subsequently, we performed ubiquitination assays to determine whether pVHL could catalyze polyubiquitin chain formation on MAVS. As shown in Fig. 4B, only pVHL could indeed induce MAVS polyubiquitination, but Smurf1 and Nedd4 did not do so, suggesting that pVHL serves as a specific E3 ligase of MAVS.

FIGURE 4.

pVHL promotes MAVS proteasomal degradation by targeting lysine 420 of MAVS. (A) The proteasome inhibitor, MG132, blocks VHL-mediated MAVS degradation. HEK293T cells were transfected with the indicated expression vectors (2 μg/well); after 18–24 h, the cells were harvested for Western blot analysis; 6 h before cell harvest, MG132 (20 μM) was added to the culture medium; DMSO was used as a carrier control. (B) pVHL catalyzes MAVS to form polyubiquitin chains. HEK293T cells were transfected with the indicated expression vectors (His-ubiquitin, 3 μg/well; Flag-MAVS, 6 μg/well; Myc-VHL/Myc-Smurf1/Myc-Nedd4, 3 μg/well); the cell lysates underwent affinity purification using Ni2+-NTA resin with anti-Flag Ab used for detection. (C) pVHL (0.2 μg/well) suppresses IRF-E reporter (0.1 μg/well) activity activated by WT MAVS (0.2 μg/well), K325/K331R mutant (0.2 μg/well), and K371R mutant (0.2 μg/well), but not by the K420R mutant (0.2 μg/well) in HEK293T cells. Data are presented as mean ± SEM of three independent experiments performed in triplicate. (D) pVHL (3 μg/well) catalyzes formation of polyubiquitin chains on WT MAVS (6 μg/well) and the K371R mutant (6 μg/well), but not on the K420R mutant (6 μg/well). (E) pVHL (2 μg/well) induces degradation of WT MAVS (2 μg/well), but not of K420R mutant (2 μg/well). Data are presented based on three independent experiments. WCL, whole cell lysates.

FIGURE 4.

pVHL promotes MAVS proteasomal degradation by targeting lysine 420 of MAVS. (A) The proteasome inhibitor, MG132, blocks VHL-mediated MAVS degradation. HEK293T cells were transfected with the indicated expression vectors (2 μg/well); after 18–24 h, the cells were harvested for Western blot analysis; 6 h before cell harvest, MG132 (20 μM) was added to the culture medium; DMSO was used as a carrier control. (B) pVHL catalyzes MAVS to form polyubiquitin chains. HEK293T cells were transfected with the indicated expression vectors (His-ubiquitin, 3 μg/well; Flag-MAVS, 6 μg/well; Myc-VHL/Myc-Smurf1/Myc-Nedd4, 3 μg/well); the cell lysates underwent affinity purification using Ni2+-NTA resin with anti-Flag Ab used for detection. (C) pVHL (0.2 μg/well) suppresses IRF-E reporter (0.1 μg/well) activity activated by WT MAVS (0.2 μg/well), K325/K331R mutant (0.2 μg/well), and K371R mutant (0.2 μg/well), but not by the K420R mutant (0.2 μg/well) in HEK293T cells. Data are presented as mean ± SEM of three independent experiments performed in triplicate. (D) pVHL (3 μg/well) catalyzes formation of polyubiquitin chains on WT MAVS (6 μg/well) and the K371R mutant (6 μg/well), but not on the K420R mutant (6 μg/well). (E) pVHL (2 μg/well) induces degradation of WT MAVS (2 μg/well), but not of K420R mutant (2 μg/well). Data are presented based on three independent experiments. WCL, whole cell lysates.

Close modal

To determine which lysine sites are catalyzed by pVHL to form polyubiquitin chains, we first did a sequence alignment for mouse, rat, human, and zebrafish MAVS proteins (Supplemental Fig. 2). Among mouse, rat, and human MAVS proteins, lysine sites including K7, K10, K325, K331, K348, K371, and K420 are evolutionarily conserved (Supplemental Fig. 2). We made three mutants (K325/331R, K371R, and K420R) with lysine mutated to arginine and examined the effect of pVHL on the function of these mutants using IRF-E luciferase reporter assays. As shown in Fig. 4C, overexpression of pVHL still suppressed the induction role of the K325/331R and K371R mutants on IRF-E luciferase reporter activity. However, overexpression of pVHL had no obvious effect on the induction role of the mutant K420R on IRF-E luciferase reporter activity (Fig. 4C), suggesting that the K420 residue of MAVS might be the target site of pVHL. To confirm that K420 residue of MAVS is the polyubiquitination target site catalyzed by pVHL, we performed ubiquitination assays. Overexpression of pVHL could promote polyubiquitination of the MAVS mutant K371R, as well as the WT MAVS, but not of the K420R mutant (Fig. 4D). Notably, overexpression of pVHL did not induce degradation of the K420R mutant (Fig. 4E). Taken together, these data suggest that pVHL promotes MAVS proteasomal degradation by targeting lysine 420 residue of MAVS for addition of a polyubiquitin chain. Interestingly, consistent with the notion that pVHL induces RIG-1 degradation, pVHL could also catalyze RIG-1 to form polyubiquitin chains (Supplemental Fig. 1C).

To assess the functional importance of pVHL-mediated MAVS degradation, we examined the effect of pVHL on the cellular antiviral response. When HEK293T cells were infected with fluorescently labeled VSV (VSV-eGFP) at MOI = 10, no obvious difference was observed between the cells transfected with Myc empty vector (control) and the cells transfected with Myc-VHL expression vector (MOI = 10; Fig. 5A, left two panels). However, when HEK293T cells were infected with VSV-eGFP at MOI = 1, overexpression of pVHL caused more cells to be infected by VSV-eGFP as assessed by number of GFP+ cells (MOI = 1; Fig. 5A, right two panels). In contrast, when pVHL was knocked down in HEK293T cells by transfection with either pSUPER-VHL1 or pSUPER-VHL2, the number of cells infected by VSV-eGFP at an MOI = 10 was greatly reduced (MOI = 10; Fig. 5B).

FIGURE 5.

pVHL negatively regulates the cellular antiviral response. (A) Overexpression of pVHL in HEK293T cells enhances VSV infection. HEK293T cells were transfected with a Myc-VHL expression vector (0.5 μg/well) or Myc empty vector control (0.5 μg/well); after 24 h, VSV-eGFP viruses were added to culture medium (MOI = 10 or MOI = 1); 8 h later, the cells were checked and photographed under a Nikon-TE2000U inverted fluorescent microscope. (B) Knockdown of pVHL by pSUPER-VHL1 and pSUPER-VHL2 in HEK293T cells diminishes VSV viral infection. HEK293T cells were transfected with pSUPER-VHL1 (1 μg/well), pSUPER-VHL2 (1μg/well), or pSUPER empty vector control (1 μg/well); after 24 h, VSV-eGFP virus was added to the culture medium (MOI = 10); 8 h later, the cells were checked and photographed under a Nikon-TE2000U inverted fluorescent microscope. (A and B) Data are presented based on three independent experiments. (C) Overexpression of pVHL suppresses IFN-β expression induced by SeV infection (MOI = 1), but not by HSV-1 infection (MOI = 1). HEK293T cells were transfected with Myc-VHL expression vector (4 μg/well) or empty vector control (4 μg/well); after 24 h, SeV or HSV-1 viruses were added to culture medium; 8 h later, the cells were harvested for total RNA extraction; IFN-β expression was detected by semiquantitative RT-PCR assays. (D) Knockdown of pVHL by pSUPER-VHL1 and pSUPER-VHL2 enhances IFN-β expression induced by VSV infection in HEK293T cells. HEK293T cells were transfected with pSuper-VHL1(6 μg/well), pSuper-VHL2 (6 μg/well), or a scrambled control (6 μg/well); after 24 h, VSV viruses (MOI = 1) were added to the culture medium; 8 h later, the cells were harvested for total RNA extraction; IFN-β expression was detected by semiquantitative RT-PCR assays. (E) Overexpression of pVHL suppresses ISG15 expression induced by SeV infection. HEK293T cells were transfected with Myc-VHL expression vector (4 μg/well) or empty vector control (4 μg/well); after 24 h, SeV viruses (MOI = 1) were added to culture medium; 8 h later, the cells were harvested for total RNA extraction; ISG15 expression was detected by semiquantitative RT-PCR assays. (C–E) Data are presented as mean ± SEM of three independent experiments performed in triplicate. (F) Without SeV infection, pVHL does not localize in mitochondria; the mitochondria are marked by the white arrows. Cos-7 cells were transfected with RFP-tagged VHL (4 μg/well). (G) pVHL translocates into mitochondria after SeV infection; the mitochondria are marked by the white arrows. Cos-7 cells were transfected with RFP-tagged VHL expression vector (4 μg/well); after 24 h, SeV viruses were added to the culture medium (MOI = 10); 8 h later, the cells were stained with MitoTracker Green (1:5000 dilution) for 45 min and photographed under a Zeiss confocal microscope (LSM-710). (F and G) Data are presented based on three independent experiments. Original magnification ×4 in (A) and (B) and ×63 in (F) and (G).

FIGURE 5.

pVHL negatively regulates the cellular antiviral response. (A) Overexpression of pVHL in HEK293T cells enhances VSV infection. HEK293T cells were transfected with a Myc-VHL expression vector (0.5 μg/well) or Myc empty vector control (0.5 μg/well); after 24 h, VSV-eGFP viruses were added to culture medium (MOI = 10 or MOI = 1); 8 h later, the cells were checked and photographed under a Nikon-TE2000U inverted fluorescent microscope. (B) Knockdown of pVHL by pSUPER-VHL1 and pSUPER-VHL2 in HEK293T cells diminishes VSV viral infection. HEK293T cells were transfected with pSUPER-VHL1 (1 μg/well), pSUPER-VHL2 (1μg/well), or pSUPER empty vector control (1 μg/well); after 24 h, VSV-eGFP virus was added to the culture medium (MOI = 10); 8 h later, the cells were checked and photographed under a Nikon-TE2000U inverted fluorescent microscope. (A and B) Data are presented based on three independent experiments. (C) Overexpression of pVHL suppresses IFN-β expression induced by SeV infection (MOI = 1), but not by HSV-1 infection (MOI = 1). HEK293T cells were transfected with Myc-VHL expression vector (4 μg/well) or empty vector control (4 μg/well); after 24 h, SeV or HSV-1 viruses were added to culture medium; 8 h later, the cells were harvested for total RNA extraction; IFN-β expression was detected by semiquantitative RT-PCR assays. (D) Knockdown of pVHL by pSUPER-VHL1 and pSUPER-VHL2 enhances IFN-β expression induced by VSV infection in HEK293T cells. HEK293T cells were transfected with pSuper-VHL1(6 μg/well), pSuper-VHL2 (6 μg/well), or a scrambled control (6 μg/well); after 24 h, VSV viruses (MOI = 1) were added to the culture medium; 8 h later, the cells were harvested for total RNA extraction; IFN-β expression was detected by semiquantitative RT-PCR assays. (E) Overexpression of pVHL suppresses ISG15 expression induced by SeV infection. HEK293T cells were transfected with Myc-VHL expression vector (4 μg/well) or empty vector control (4 μg/well); after 24 h, SeV viruses (MOI = 1) were added to culture medium; 8 h later, the cells were harvested for total RNA extraction; ISG15 expression was detected by semiquantitative RT-PCR assays. (C–E) Data are presented as mean ± SEM of three independent experiments performed in triplicate. (F) Without SeV infection, pVHL does not localize in mitochondria; the mitochondria are marked by the white arrows. Cos-7 cells were transfected with RFP-tagged VHL (4 μg/well). (G) pVHL translocates into mitochondria after SeV infection; the mitochondria are marked by the white arrows. Cos-7 cells were transfected with RFP-tagged VHL expression vector (4 μg/well); after 24 h, SeV viruses were added to the culture medium (MOI = 10); 8 h later, the cells were stained with MitoTracker Green (1:5000 dilution) for 45 min and photographed under a Zeiss confocal microscope (LSM-710). (F and G) Data are presented based on three independent experiments. Original magnification ×4 in (A) and (B) and ×63 in (F) and (G).

Close modal

To further confirm the role of pVHL in the cellular antiviral response, we examined the expression of IFN-β, a marker gene induced by viral infection (28). As shown in Fig. 5C, overexpression of pVHL in HEK293T cells suppressed IFN-β induction by SeV. In contrast, overexpression of pVHL in HEK293T cells had no obvious effect on IFN-β induction by HSV-1, a DNA virus. As reported, MAVS acts as adaptor protein to transmit signals from both RIG-1 and MDA5, which function as cytoplasmic RNA sensors that recognize viral RNA released during virus replication (29). Thus, the observation that overexpression of pVHL only suppressed IFN-β induction by RNA virus but not by a DNA virus are consistent with the role of pVHL in promoting MAVS degradation. Conversely, knockdown of pVHL in HEK293T cells by either pSUPER-VHL1 or pSUPER-VHL2 enhanced IFN-β induction by VSV infection (Fig. 5D).

Furthermore, we examined the expression of ISG15, a ubiquitin-like modifier gene induced by type I IFN that has been shown to mediate protection in a number of different viral infection models (30). As shown in Fig. 5E, overexpression of pVHL in HEK293T cells suppressed ISG15 expression induced by SeV.

Interestingly, without virus infection, RFP-tagged pVHL mainly distributed in cytoplasm but was outside of mitochondria (Fig. 5F, white arrows). However, SeV infection caused translocation of pVHL to mitochondria completely (Fig. 5G, white arrows), further highlighting the role of pVHL in inducing MAVS degradation and negatively regulating antiviral immunity (31). Taken together, these data suggest that pVHL negatively regulates cellular antiviral response.

To determine the physiological role of pVHL in vivo in response to viral infection, we took advantage of the zebrafish model by using infection with the SVCV (32, 33). Using the TALEN technique, we generated two zebrafish lines with mutated vhl (mutant 1 and mutant 2; Supplemental Fig. 3) (34). In the mutant 1, two nucleotides in exon I of vhl were deleted (Supplemental Fig. 3C). In the mutant 2, two nucleotides were inserted in exon 1 (Supplemental Fig. 3C).

Similar to reported zebrafish vhl mutants generated by N-ethyl-N-nitrosourea treatment (35), these two mutants displayed a severe hyperventilation and cardiophysiological response and developed polycythemia (35) (Supplemental Fig. 3D). The homozygous null embryos (vhl−/−) survived up to 8–11 d and ultimately developed severe edema and died (35) (Supplemental Fig. 3D).

At 5 dpf, based on the general morphology (hyperventilation, polycythemia, etc.), it was easy to separate homozygous null embryos (vhl−/−) from vhl+/+ and vhl+/− offspring produced by mating adult heterozygous zebrafish (vhl+/−; Supplemental Fig. 2D). Therefore, we used 5 dpf embryos for viral infection assays. SVCV infection could cause obvious pathological effects in both vhl-deficient zebrafish embryos (vhl−/−; mutant 1 and mutant 2) and WT embryos (vhl+/+; Supplemental Fig. 2E). All embryos used for viral infection assays, including the homozygous null embryos (vhl−/−) and their WT siblings (vhl+/+), were regenotyped by sequencing.

Similar to that observed for mammalian VHL, zebrafish Vhl could still induce zebrafish Mavs degradation when overexpressed in HEK293T cells (Fig. 6A). Consistently, the protein level of Mavs in the homozygous embryos (vhl−/−; mutant 1) was higher than that of WT embryos (vhl+/+; Fig. 6B, left two lanes). Interestingly, SVCV infection enhanced the endogenous Mavs protein level, resulting in much higher levels in the homozygous embryos (vhl−/−) compared with that observed in WT embryos (vhl+/+; Fig. 6B, right two lanes) (32, 33). Of note, the homozygous embryos (vhl−/−; mutant1) had the highest level of ifn1, a well-defined gene for zebrafish response to viral infection (33, 36), but the heterozygous embryos (vhl+/−) had a middle level of ifn1, and WT embryos (vhl+/+) had a lowest level of ifn1 (Fig. 6C). This expression pattern is correlated with the level of vhl in zebrafish embryos, implying the physiological role of vhl in antiviral immune response.

FIGURE 6.

Zebrafish Vhl (DrVhl) induces zebrafish Mavs degradation (DrMavs) and negatively regulates zebrafish antiviral response. (A) Zebrafish Vhl promotes zebrafish Mavs degradation. HEK293T cells were transfected with HA-DrMAVS (2 μg/well) together with Myc empty vector (2 μg/well) or Myc-DrVhl expression vector (2 μg/well); after 18–24 h, the cells were harvested and examined by Western blot analysis. (B) Zebrafish Vhl-deficient embryos (vhl−/−; Mutant 1) exhibit higher endogenous Mavs protein compared with the WT zebrafish (vhl+/+) with or without SVCV infection. After 5 dpf, SVCV viruses (∼2 × 108 TCID50/ml) in water were added to zebrafish vhl WT and null embryos and incubated for 24 h, and endogenous Mavs was detected by Western blot analysis using a polyclonal anti-DrMavs Ab. (C) Expression of ifn1 (Drinf1) in Vhl-deficient embryos (vhl−/−; mutant 1) is much higher than that in both Vhl heterozygous embryos (vhl+/−; p < 0.0001) and WT embryos (vhl+/+; p < 0.0001); expression of ifn1 (Drinf1) in Vhl heterozygous embryos (vhl+/−; mutant 1) is also higher than that in WT embryos (vhl+/+; p = 0.0002). (D) SVCV infection (∼2 × 108 TCID50/ml) in zebrafish promotes upregulation of ifn1 (Drinf1) much more dramatically in vhl-null embryos (vhl−/−) (Mutant 1) compared with that in vhl WT embryos (vhl+/+) (p < 0.0001). (E) SVCV infection (∼2 × 108 TCID50/ml) in zebrafish promotes upregulation of ifn3 (Drinf3) much more dramatically in vhl-null embryos (vhl−/−; mutant 1) compared with that in vhl WT embryos (vhl+/+; p = 0.0097). After 5 dpf, SVCV viruses (∼2 × 108 TCID50/ml) in water were added to zebrafish vhl WT and null embryos (mutant 1) and incubated for 24 h after which inf1 (Drifn1) and ifn3 (Drifn3) expression levels were detected by semiquantitative RT-PCR analysis. Data are presented as mean ± SEM of three independent experiments performed in triplicate.

FIGURE 6.

Zebrafish Vhl (DrVhl) induces zebrafish Mavs degradation (DrMavs) and negatively regulates zebrafish antiviral response. (A) Zebrafish Vhl promotes zebrafish Mavs degradation. HEK293T cells were transfected with HA-DrMAVS (2 μg/well) together with Myc empty vector (2 μg/well) or Myc-DrVhl expression vector (2 μg/well); after 18–24 h, the cells were harvested and examined by Western blot analysis. (B) Zebrafish Vhl-deficient embryos (vhl−/−; Mutant 1) exhibit higher endogenous Mavs protein compared with the WT zebrafish (vhl+/+) with or without SVCV infection. After 5 dpf, SVCV viruses (∼2 × 108 TCID50/ml) in water were added to zebrafish vhl WT and null embryos and incubated for 24 h, and endogenous Mavs was detected by Western blot analysis using a polyclonal anti-DrMavs Ab. (C) Expression of ifn1 (Drinf1) in Vhl-deficient embryos (vhl−/−; mutant 1) is much higher than that in both Vhl heterozygous embryos (vhl+/−; p < 0.0001) and WT embryos (vhl+/+; p < 0.0001); expression of ifn1 (Drinf1) in Vhl heterozygous embryos (vhl+/−; mutant 1) is also higher than that in WT embryos (vhl+/+; p = 0.0002). (D) SVCV infection (∼2 × 108 TCID50/ml) in zebrafish promotes upregulation of ifn1 (Drinf1) much more dramatically in vhl-null embryos (vhl−/−) (Mutant 1) compared with that in vhl WT embryos (vhl+/+) (p < 0.0001). (E) SVCV infection (∼2 × 108 TCID50/ml) in zebrafish promotes upregulation of ifn3 (Drinf3) much more dramatically in vhl-null embryos (vhl−/−; mutant 1) compared with that in vhl WT embryos (vhl+/+; p = 0.0097). After 5 dpf, SVCV viruses (∼2 × 108 TCID50/ml) in water were added to zebrafish vhl WT and null embryos (mutant 1) and incubated for 24 h after which inf1 (Drifn1) and ifn3 (Drifn3) expression levels were detected by semiquantitative RT-PCR analysis. Data are presented as mean ± SEM of three independent experiments performed in triplicate.

Close modal

To further determine the role of vhl in response to viral infection, we examined expression of ifn1 and ifn3, two well-defined genes for zebrafish in response to viral infection (33, 36). Moreover, SVCV infection caused ifn1 and ifn3 to increase much more dramatically in vhl-deficient zebrafish embryos (vhl−/−; mutant 1) than in WT embryos (vhl+/+; Fig. 6D, 6E). Taken together, these data suggest that zebrafish vhl negatively regulates the antiviral response probably through mediation of Mavs degradation in vivo.

Since VHL was identified in 1993 as the genetic basis for VHL disease, it has proved to be of widespread interest (1, 37). pVHL has been well characterized and performs its tumor suppressive function by targeting hypoxia-inducible factors for ubiquitination and proteasomal degradation, which accounts for the cellular oxygen sensing (10, 38). Large phenotypic variations exist in VHL patients. Coupled with the fact that some VHL mutants can still cause VHL diseases while retaining the ability to induce degradation of hypoxia-inducible factors, mechanisms other than those mediating HIF-degradation have been proposed to explain pVHL function (1). However, these mechanisms still need to be further defined (10).

Although the impact of HIFs on immune responses has been observed (39), the mechanisms by which VHL affects the cellular immunity response, either directly or through HIFs, is still largely unknown. Interestingly, it was reported that pVHL could inhibit NF-κB phosphorylation through its effect on the NF-κB agonist Card 9, which links pVHL to control of NF-κB activity and tumorigenesis (40). However, the role of pVHL in the cellular antiviral response has not been defined. In this study, we have found that pVHL negatively regulates IRF3 activity in response to viral infection. Using additional assays, we have identified MAVS as a major target of pVHL for ubiquitination and proteasomal degradation. Intriguingly, after viral infection, pVHL colocalized to mitochondria onto which MAVS were anchored. Furthermore, in vhl-deficient zebrafish embryos, the protein level of zebrafish Mavs rose much higher after viral infection, compared with its level in the WT zebrafish embryos. In addition, vhl-deficient zebrafish embryos exhibited a greatly exaggerated antiviral response. Collectively, these data suggest that pVHL plays an important role in innate antiviral immunity. However, how this negative regulatory role correlates with its tumor suppressive function is still unknown. Investigation of the correlations between pVHL’s tumor suppressive function and its negative regulation of antiviral immunity should uncover new avenues for understanding the relationship between tumor development and innate immunity.

Notably, during the identification of pVHL targets in the cellular antiviral innate immunity signaling pathway (Mda5/RIG-1—MAVS—TBK1—IRF3), we found that in addition to MAVS, RIG-1 and MDA5 also appear to be the targets of pVHL. Overexpression of pVHL together with RIG-1 not only suppressed the activity of the IRF-E reporter enhanced by RIG-1 (Fig. 1B), but also induced protein degradation of RIG-1 (Fig. 2A, Supplemental Fig. 1A), even though their degradation rate by pVHL was not as dramatic as that of MAVS. In addition, pVHL could also interact with RIG-1 and catalyze RIG-1 to form polyubiquitin chains, further suggesting that RIG-1 might be another target of pVHL in the cellular antiviral innate immunity. For MDA5, it appeared that pVHL could not mediate its protein degradation, but pVHL could still interact with MDA5 and suppressed MDA5-activated IRF-E reporter activity. Thus, MDA5 might also be a target of pVHL in the cellular antiviral innate immunity. pVHL may affect MDA5 function through a different mechanism other than inducing MDA5 degradation. Further investigation of the effect of pVHL on the RIG-1 and MDA5 will help to better understand the function of pVHL in cellular innate antiviral immunity.

The behavior of pVHL in cellular antiviral signaling is quite similar to that of E3 ligase RNF125, which can ubiquitinate and trigger degradation of MDA5, RIG-1, and MAVS simultaneously to suppress the cellular antiviral response (18). However, MAVS might play a greater role as a pVHL target than RIG-1 and MDA5, as indicated by the observations: the inhibitory ability of pVHL on MAVS-enhanced IRF3 activity was higher and the degradation rate of MAVS by pVHL was higher. However, because of the multiple steps and multiple regulation points of cellular RIG-1–like-helicase–mediated antiviral signaling, we still cannot rule out other targets of pVHL involved in this signaling pathway (17, 31, 41, 42).

As a well-known disorder, VHL disease has attracted much attention from clinicians and biologists (4346), culminating in the collection of much clinical and epidemiological data. However, most data pertain to angiogenic tumors (43, 45, 46). So far, no epidemiological data describe the relationship between VHL diseases and viral infection. Elucidation of the correlation between VHL disease and viral infection should provide new clues for understanding the physiological function of pVHL.

We are grateful to Drs. Hong-bin Shu, Xinwen Chen, Minzhou Chen, and Chunfu Zheng for the generous gifts of reagents.

This work was supported by Chinese Academy of Sciences Major Scientific and Technological Project XDA08010208 and National Natural Science Foundation of China Grant 31461163003.

The online version of this article contains supplemental material.

Abbreviations used in this article:

dpf

day postfertilization

E1

ubiquitin-activating enzyme

E2

ubiquitin-conjugating enzyme

E3

ubiquitin ligase

Flag-MAVS

Flag-tagged MAVS

HIF

hypoxia-induced factor

IRF

IFN regulatory factor

ISG

IFN-stimulated gene

MAVS

mitochondrial antiviral-signaling protein

MOI

multiplicity of infection

pVHL

VHL tumor suppressor protein

SeV

Sendai virus

shRNA

short hairpin RNA

SVCV

spring viremia of carp virus

VCB

VHL-elongin B/C

VHL

von Hippel–Lindau

VSV

vesicular stomatitis virus

WT

wild type.

1
Gossage
L.
,
Eisen
T.
,
Maher
E. R.
.
2015
.
VHL, the story of a tumour suppressor gene.
Nat. Rev. Cancer
15
:
55
64
.
2
Kaelin
W. G.
 Jr.
2007
.
The von Hippel-Lindau tumor suppressor protein: an update.
Methods Enzymol.
435
:
371
383
.
3
Frew
I. J.
,
Moch
H.
.
2015
.
A clearer view of the molecular complexity of clear cell renal cell carcinoma.
Annu. Rev. Pathol.
10
:
263
289
.
4
Robinson
C. M.
,
Ohh
M.
.
2014
.
The multifaceted von Hippel-Lindau tumour suppressor protein.
FEBS Lett.
588
:
2704
2711
.
5
Shen
C.
,
Kaelin
W. G.
 Jr.
2013
.
The VHL/HIF axis in clear cell renal carcinoma.
Semin. Cancer Biol.
23
:
18
25
.
6
Jung
Y. S.
,
Lee
S. J.
,
Yoon
M. H.
,
Ha
N. C.
,
Park
B. J.
.
2012
.
Estrogen receptor α is a novel target of the Von Hippel-Lindau protein and is responsible for the proliferation of VHL-deficient cells under hypoxic conditions.
Cell Cycle
11
:
4462
4473
.
7
Gamper
A. M.
,
Qiao
X.
,
Kim
J.
,
Zhang
L.
,
DeSimone
M. C.
,
Rathmell
W. K.
,
Wan
Y.
.
2012
.
Regulation of KLF4 turnover reveals an unexpected tissue-specific role of pVHL in tumorigenesis.
Mol. Cell
45
:
233
243
.
8
Arias-González
L.
,
Moreno-Gimeno
I.
,
del Campo
A. R.
,
Serrano-Oviedo
L.
,
Valero
M. L.
,
Esparís-Ogando
A.
,
de la Cruz-Morcillo
M. A.
,
Melgar-Rojas
P.
,
García-Cano
J.
,
Cimas
F. J.
, et al
.
2013
.
ERK5/BMK1 is a novel target of the tumor suppressor VHL: implication in clear cell renal carcinoma.
Neoplasia
15
:
649
659
.
9
Wang
J.
,
Zhang
W.
,
Ji
W.
,
Liu
X.
,
Ouyang
G.
,
Xiao
W.
.
2014
.
The von Hippel-Lindau protein suppresses androgen receptor activity.
Mol. Endocrinol.
28
:
239
248
.
10
Kawai
T.
,
Takahashi
K.
,
Sato
S.
,
Coban
C.
,
Kumar
H.
,
Kato
H.
,
Ishii
K. J.
,
Takeuchi
O.
,
Akira
S.
.
2005
.
IPS-1, an adaptor triggering RIG-I- and Mda5-mediated type I interferon induction.
Nat. Immunol.
6
:
981
988
.
11
Meylan
E.
,
Curran
J.
,
Hofmann
K.
,
Moradpour
D.
,
Binder
M.
,
Bartenschlager
R.
,
Tschopp
J.
.
2005
.
Cardif is an adaptor protein in the RIG-I antiviral pathway and is targeted by hepatitis C virus.
Nature
437
:
1167
1172
.
12
Seth
R. B.
,
Sun
L.
,
Ea
C. K.
,
Chen
Z. J.
.
2005
.
Identification and characterization of MAVS, a mitochondrial antiviral signaling protein that activates NF-kappaB and IRF 3.
Cell
122
:
669
682
.
13
Xu
L. G.
,
Wang
Y. Y.
,
Han
K. J.
,
Li
L. Y.
,
Zhai
Z.
,
Shu
H. B.
.
2005
.
VISA is an adapter protein required for virus-triggered IFN-beta signaling.
Mol. Cell
19
:
727
740
.
14
Cai
X.
,
Chen
Z. J.
.
2014
.
Prion-like polymerization as a signaling mechanism.
Trends Immunol.
35
:
622
630
.
15
Cai
X.
,
Chen
J.
,
Xu
H.
,
Liu
S.
,
Jiang
Q. X.
,
Halfmann
R.
,
Chen
Z. J.
.
2014
.
Prion-like polymerization underlies signal transduction in antiviral immune defense and inflammasome activation.
Cell
156
:
1207
1222
.
16
Sun
Q.
,
Sun
L.
,
Liu
H. H.
,
Chen
X.
,
Seth
R. B.
,
Forman
J.
,
Chen
Z. J.
.
2006
.
The specific and essential role of MAVS in antiviral innate immune responses.
Immunity
24
:
633
642
.
17
Bhoj
V. G.
,
Chen
Z. J.
.
2009
.
Ubiquitylation in innate and adaptive immunity.
Nature
458
:
430
437
.
18
Arimoto
K.
,
Takahashi
H.
,
Hishiki
T.
,
Konishi
H.
,
Fujita
T.
,
Shimotohno
K.
.
2007
.
Negative regulation of the RIG-I signaling by the ubiquitin ligase RNF125.
Proc. Natl. Acad. Sci. USA
104
:
7500
7505
.
19
Castanier
C.
,
Zemirli
N.
,
Portier
A.
,
Garcin
D.
,
Bidère
N.
,
Vazquez
A.
,
Arnoult
D.
.
2012
.
MAVS ubiquitination by the E3 ligase TRIM25 and degradation by the proteasome is involved in type I interferon production after activation of the antiviral RIG-I-like receptors.
BMC Biol.
10
:
44
.
20
Pan
Y.
,
Li
R.
,
Meng
J. L.
,
Mao
H. T.
,
Zhang
Y.
,
Zhang
J.
.
2014
.
Smurf2 negatively modulates RIG-I-dependent antiviral response by targeting VISA/MAVS for ubiquitination and degradation.
J. Immunol.
192
:
4758
4764
.
21
You
F.
,
Sun
H.
,
Zhou
X.
,
Sun
W.
,
Liang
S.
,
Zhai
Z.
,
Jiang
Z.
.
2009
.
PCBP2 mediates degradation of the adaptor MAVS via the HECT ubiquitin ligase AIP4.
Nat. Immunol.
10
:
1300
1308
.
22
Wang
Y.
,
Tong
X.
,
Ye
X.
.
2012
.
Ndfip1 negatively regulates RIG-I-dependent immune signaling by enhancing E3 ligase Smurf1-mediated MAVS degradation.
J. Immunol.
189
:
5304
5313
.
23
Zhong
B.
,
Yang
Y.
,
Li
S.
,
Wang
Y. Y.
,
Li
Y.
,
Diao
F.
,
Lei
C.
,
He
X.
,
Zhang
L.
,
Tien
P.
,
Shu
H. B.
.
2008
.
The adaptor protein MITA links virus-sensing receptors to IRF3 transcription factor activation.
Immunity
29
:
538
550
.
24
Chen
Z.
,
Liu
X.
,
Mei
Z.
,
Wang
Z.
,
Xiao
W.
.
2014
.
EAF2 suppresses hypoxia-induced factor 1α transcriptional activity by disrupting its interaction with coactivator CBP/p300.
Mol. Cell. Biol.
34
:
1085
1099
.
25
McCurley
A. T.
,
Callard
G. V.
.
2008
.
Characterization of housekeeping genes in zebrafish: male-female differences and effects of tissue type, developmental stage and chemical treatment.
BMC Mol. Biol.
9
:
102
.
26
Nakaya
T.
,
Sato
M.
,
Hata
N.
,
Asagiri
M.
,
Suemori
H.
,
Noguchi
S.
,
Tanaka
N.
,
Taniguchi
T.
.
2001
.
Gene induction pathways mediated by distinct IRFs during viral infection.
Biochem. Biophys. Res. Commun.
283
:
1150
1156
.
27
Navarro
L.
,
Mowen
K.
,
Rodems
S.
,
Weaver
B.
,
Reich
N.
,
Spector
D.
,
David
M.
.
1998
.
Cytomegalovirus activates interferon immediate-early response gene expression and an interferon regulatory factor 3-containing interferon-stimulated response element-binding complex.
Mol. Cell. Biol.
18
:
3796
3802
.
28
Wang
B. X.
,
Fish
E. N.
.
2012
.
The yin and yang of viruses and interferons.
Trends Immunol.
33
:
190
197
.
29
Yoneyama
M.
,
Onomoto
K.
,
Jogi
M.
,
Akaboshi
T.
,
Fujita
T.
.
2015
.
Viral RNA detection by RIG-I-like receptors.
Curr. Opin. Immunol.
32
:
48
53
.
30
Skaug
B.
,
Chen
Z. J.
.
2010
.
Emerging role of ISG15 in antiviral immunity.
Cell
143
:
187
190
.
31
Arnoult
D.
,
Soares
F.
,
Tattoli
I.
,
Girardin
S. E.
.
2011
.
Mitochondria in innate immunity.
EMBO Rep.
12
:
901
910
.
32
Sanders
G. E.
,
Batts
W. N.
,
Winton
J. R.
.
2003
.
Susceptibility of zebrafish (Danio rerio) to a model pathogen, spring viremia of carp virus.
Comp. Med.
53
:
514
521
.
33
López-Muñoz
A.
,
Roca
F. J.
,
Sepulcre
M. P.
,
Meseguer
J.
,
Mulero
V.
.
2010
.
Zebrafish larvae are unable to mount a protective antiviral response against waterborne infection by spring viremia of carp virus.
Dev. Comp. Immunol.
34
:
546
552
.
34
Bedell
V. M.
,
Ekker
S. C.
.
2015
.
Using engineered endonucleases to create knockout and knockin zebrafish models.
Methods Mol. Biol.
1239
:
291
305
.
35
van Rooijen
E.
,
Voest
E. E.
,
Logister
I.
,
Korving
J.
,
Schwerte
T.
,
Schulte-Merker
S.
,
Giles
R. H.
,
van Eeden
F. J.
.
2009
.
Zebrafish mutants in the von Hippel-Lindau tumor suppressor display a hypoxic response and recapitulate key aspects of Chuvash polycythemia.
Blood
113
:
6449
6460
.
36
Li
S.
,
Lu
L. F.
,
Feng
H.
,
Wu
N.
,
Chen
D. D.
,
Zhang
Y. B.
,
Gui
J. F.
,
Nie
P.
,
Zhang
Y. A.
.
2014
.
IFN regulatory factor 10 is a negative regulator of the IFN responses in fish.
J. Immunol.
193
:
1100
1109
.
37
Latif
F.
,
Tory
K.
,
Gnarra
J.
,
Yao
M.
,
Duh
F. M.
,
Orcutt
M. L.
,
Stackhouse
T.
,
Kuzmin
I.
,
Modi
W.
,
Geil
L.
, et al
.
1993
.
Identification of the von Hippel-Lindau disease tumor suppressor gene.
Science
260
:
1317
1320
.
38
Xiao
W.
2015
.
The hypoxia signaling pathway and hypoxic adaptation in fishes.
Sci. China. Life Sci. 58: 148–155
.
39
Palazon
A.
,
Goldrath
A. W.
,
Nizet
V.
,
Johnson
R. S.
.
2014
.
HIF transcription factors, inflammation, and immunity.
Immunity
41
:
518
528
.
40
Yang
H.
,
Minamishima
Y. A.
,
Yan
Q.
,
Schlisio
S.
,
Ebert
B. L.
,
Zhang
X.
,
Zhang
L.
,
Kim
W. Y.
,
Olumi
A. F.
,
Kaelin
W. G.
 Jr.
2007
.
pVHL acts as an adaptor to promote the inhibitory phosphorylation of the NF-kappaB agonist Card9 by CK2.
Mol. Cell
28
:
15
27
.
41
Takeuchi
O.
,
Akira
S.
.
2007
.
Recognition of viruses by innate immunity.
Immunol. Rev.
220
:
214
224
.
42
Hou
F.
,
Sun
L.
,
Zheng
H.
,
Skaug
B.
,
Jiang
Q. X.
,
Chen
Z. J.
.
2011
.
MAVS forms functional prion-like aggregates to activate and propagate antiviral innate immune response.
Cell
146
:
448
461
.
43
Haddad
N. M.
,
Cavallerano
J. D.
,
Silva
P. S.
.
2013
.
Von Hippel-Lindau disease: a genetic and clinical review.
Semin. Ophthalmol.
28
:
377
386
.
44
Keefe
S. M.
,
Nathanson
K. L.
,
Rathmell
W. K.
.
2013
.
The molecular biology of renal cell carcinoma.
Semin. Oncol.
40
:
421
428
.
45
Bausch
B.
,
Jilg
C.
,
Gläsker
S.
,
Vortmeyer
A.
,
Lützen
N.
,
Anton
A.
,
Eng
C.
,
Neumann
H. P.
.
2013
.
Renal cancer in von Hippel-Lindau disease and related syndromes.
Nat. Rev. Nephrol.
9
:
529
538
.
46
Schmid, S., S. Gillessen, I. Binet, M. Brandle, D. Engeler, J. Greiner, C. Hader, K. Heinimann, P. Kloos, W. Krek, et al. 2014. Management of von Hippel-Lindau disease: an interdisciplinary review. Oncol. Res. Treat. 37: 761–771
.

The authors have no financial conflicts of interest.

Supplementary data