Major vault protein: A virus-induced host factor against viral replication through the induction of type-I interferon

Authors


  • Potential conflict of interest: Nothing to report.

  • Supported by research grants from the Major State Basic Research Development Program of China (“973” project No. 2009CB522506, and No. 2012CB518900), National Mega Project on Major Infectious Diseases Prevention (2012ZX10004503-004, 2012ZX10002006-003, and 2012ZX10004-207), National Mega Project on Major Drug Development (2011ZX09401-302). The funding agencies had no role in the study design, data collection, analysis, decision to publish, or preparation of the article.

Abstract

Major vault protein (MVP) is the major constituent of vaults and is involved in multidrug resistance, nucleocytoplasmic transport, and cell signaling. However, little is known about the role of MVP during viral infections. In this study, high levels of MVP were found in peripheral blood mononuclear cells, sera, and liver tissue from patients infected with hepatitis C virus (HCV) relative to healthy individuals. HCV infections resulted in elevated levels of MVP messenger RNA (mRNA) and protein expression in the hepatocyte cell lines Huh7.5.1 and Huh7. Further studies demonstrated that the nuclear factor kappa B (NF-κB) and Sp1 pathways are involved in the induction of MVP expression by HCV. Interestingly, MVP expression suppressed HCV replication and protein synthesis by way of induction of type-I interferon mRNA expression and protein secretion. Upon investigating the mechanisms behind this event, we found that MVP enhanced the expression of interferon regulatory factor 7 (IRF7), but not IRF3. Translocation of activated IRF7 and NF-κB from the cytosol to the nucleus was involved in this process. Furthermore, vesicular stomatitis virus, influenza A virus, and enterovirus 71 also induced MVP production, and MVP in turn hampered viral replication and production. Conclusion: MVP is a novel virus-induced host factor and its expression up-regulates type-I interferon production, leading to cellular antiviral responses. (HEPATOLOGY 2012;56:57–66)

Type-I interferon (IFN-α/β) is a critical first line of defense that limits virus expression and replication. Recent advances have provided a clearer picture of the virus-triggered type-I IFN signaling pathways. Host cells use pattern recognition receptors, including Toll-like receptors (TLRs) and retinoic acid-inducible gene 1 protein (RIG-I)-like helicases such as RIG-I and melanoma differentiation-associated protein 5 (MDA5), to detect viral infection.1, 2 For example, RIG-I and MDA5 are cytoplasmic viral RNA (vRNA) sensors that bind mitochondrial Cardif (also known as MAVS, VISA, and IPS-1) through caspase recruitment domains initiating signaling cascades that lead to interferon regulatory factor (IRF)3/7 and nuclear factor kappa B (NF-κB) activation.2, 3 Unlike RIG-I and MDA5, TLRs are transmembrane receptors that dimerize and are conformationally altered by specific ligands. TLR7 and TLR9 are responsible for detection of viral single-strand RNA (ssRNA) and DNA, respectively,4 by binding to their ligands and interacting with the adaptor protein MyD88. MyD88 recruits IRAK family members, leading to the activation of TRAF6, a ubiquitin E3 ligase. TRAF6 catalyzes or forms protein complexes that mediate IRF7 and NF-κB activation.5, 6 In addition, most virus-triggered signaling pathways are dependent on IRF3/7 and NF-κB activation to induce type-I IFNs. The subsequent secreted type-I IFNs bind to cognate receptors on the surface of surrounding cells to activate the Jak/STAT pathway, resulting in the transcription of IFN-stimulated genes (ISGs) that are involved in the inhibition of virus replication.7

The 110-kDa major vault protein (MVP), identified as lung resistance-related protein, is the predominant component of a large, cytoplasmic ribonucleoprotein particle, the vault complex.8, 9 Vaults are barrel-shaped and highly conserved in a wide range of phylogenetic groups.10 Mammalian vaults are comprised of three proteins, MVP, vault poly(ADP-ribose) polymerase, and telomerase-associated protein 1, as well as one or more small untranslated RNAs, vRNA.11 MVP constitutes at least 70% of the total vault particle mass and is sufficient to direct the formation of a basic vault structure.9

MVP overexpression has been identified in multidrug-resistant cancer cells, suggesting that MVP may be involved in multidrug resistance.12 Another function of MVP is nucleocytoplasmic transport. Vaults are located in the cytoplasm in association with the cytoskeleton. However, nearly 5% of vaults are occasionally localized to nuclear pore complexes.13 Recently, several studies confirmed the importance of MVP in the regulation of multiple cellular processes, such as immune responses, signal transmission, and cellular stress responses. MVP associates with endogenous phosphatase and tensin homolog (PTEN) with the help of calcium (Ca)2+, suggesting MVP participates in the phosphoinositide 3-kinase pathway and cell growth.14 MVP can also associate with COP1 and suppress c-Jun-mediated AP-1 transactivation.15 MVP interacts with Src in response to endothelial growth factor (EGF) and functions as a signaling scaffold protein for the extracellular signal-related kinase (ERK) / mitogen-activated protein kinase (MAPK) pathway.16 MVP forms complexes with hypoxia-inducible factor (HIF)-1α promoting HIF-1α degradation.17 Interestingly, IFN-γ can enhance MVP expression, implying that MVP is related to immune responses.18

Although several functions of MVP have been described, a clear role for MVP in innate immune responses to viral infections has not been established. In this study we show that viruses such as hepatitis C virus (HCV), vesicular stomatitis virus (VSV), influenza A virus (IAV), and enterovirus 71 (EV71) enhanced MVP expression. In turn, MVP limited viral expression and replication by inducing type-I IFNs. Our results describe a novel mechanism for induction of type-I IFN during viral infections.

Abbreviations

EV71, enterovirus 71; HCV, hepatitis C virus; IAV, influenza A virus; IFN, interferon; ISG, IFN-stimulated gene; MDCK, Madin-Darby canine kidney; MOI, multiplicity of infection; MVP, major vault protein; PBMC, peripheral blood mononuclear cell; RLR, RIG-I like receptors; TLR, Toll-like receptors; VSV, vesicular stomatitis virus.

Patients and Methods

A detailed description of the experimental procedures can be found in the online Supporting Information.

Results

MVP Expression Is Elevated in HCV Patients.

To investigate MVP expression during HCV infection, peripheral blood mononuclear cells (PBMCs) were isolated from HCV patients (n = 11) and healthy individuals (n = 11). As determined by real-time reverse-transcription polymerase chain reaction (RT-PCR), MVP messenger RNA (mRNA) levels were approximately 7-fold higher in HCV patients than in healthy individuals (Fig. 1A; Supporting Table 1). Significant differences in serum MVP protein levels were also observed between HCV patients (n = 108) and healthy individuals (n = 108) as determined by enzyme-linked immunosorbent assay (ELISA) (mean ± standard error of the mean [SEM] 119.39 ± 25.32 versus 67.99 ± 17.92 pg/mL) (Fig. 1B; Supporting Table 2).

Figure 1.

MVP expression in healthy individuals and patients infected with HCV. (A) Total RNA was extracted from freshly isolated PBMCs from healthy individuals (n = 11) or HCV patients (n = 11). MVP mRNA was detected by real-time RT-PCR. Data represent means ± SEM from samples tested in triplicate. Boxplots illustrate medians with 25% and 75% and error bars for 5% and 95% percentiles. (**P < 0.01). (B) Serum MVP levels in healthy individuals (n = 108) and patients (n = 108) were detected by ELISA. Samples were tested in duplicate and concentrations determined from standard curves. Data represent means ± SEM. Boxplots illustrate medians with 25% and 75% and error bars for 5% and 95% percentiles.(**P < 0.01). (C) Immunohistochemical staining of MVP in normal (n = 20) and HCV patient (n = 20) livers. (D) Freshly isolated PBMCs from healthy donors were treated with PBS, IFN-α (300 IU/mL), or IFN-γ (300 IU/mL) for 24 hours. MVP was measured by real-time RT-PCR (left panel) and western blot (right panel). Experiments were repeated three times with similar results. Data represent means ± SD, n = 3.

Furthermore, MVP expression in liver biopsy specimens from HCV patients (n = 20) and healthy individuals (n = 20) was examined using immunohistochemistry. HCV patients showed higher levels of MVP staining than did healthy individuals (Fig. 1C; Supporting Table 3). These data indicate that MVP expression is elevated in patients infected with HCV.

Because IFN-α is used as a clinical treatment for HCV infections, we tested whether IFN-α had an effect on MVP expression. Freshly isolated PBMCs from healthy donors were stimulated with or without IFN-α (300 IU/mL) for 24 hours. MVP mRNA levels were measured using real-time RT-PCR and western blot analyses. There was no association between IFN-α treatments and MVP expression (Fig. 1D). Because it has been reported that IFN-γ can enhance MVP expression,18 IFN-γ treatment was included as a positive control for comparison.

HCV Induces MVP Expression in Huh7 and Huh7.5.1 Cells.

Because high levels of MVP expression were observed in HCV patients, we next examined changes in endogenous MVP levels in response to HCV infection in vitro in two HCV-permissive cell lines (Huh7 and Huh7.5.1) using the JFH-1 infectious HCV clone. To explore the kinetics of MVP induction, MVP mRNA and protein expression levels were measured at various timepoints after infection with JFH-1 at a multiplicity of infection of 1 (multiplicity of infection [MOI] = 1) as described.19, 20 HCV infection of Huh7 (Fig. 2A) and Huh7.5.1 (Fig. 2B) cells resulted in increased MVP mRNA and protein expression. MVP mRNA elevation was detected as early as 6 days after infection for Huh7 cells and 4 days for Huh7.5.1 cells (Fig. 2A,B, upper panel). MVP levels continued to increase throughout the course of the experiment. MVP protein levels showed a similar trend as determined by western blot (Fig. 2A,B, lower panel).

Figure 2.

Effect of HCV on the expression of MVP. Huh7 (A) and Huh7.5.1 (B) cells were infected with HCV (MOI = 1) for the indicated times. MVP and HCV RNA levels were quantified by real-time RT-PCR (upper panel) and protein levels of MVP and HCV Core were detected by western blot (lower panel). Huh7 (C) and Huh7.5.1 (D) cells were infected with HCV for 9 days and 4 days, respectively. MVP and HCV RNA levels were quantified by real-time RT-PCR (upper panel) and protein levels were detected by western blot (lower panel). Experiments were performed three times with similar results (lower panels of A-D). Data represent means ± SD, n = 3 (upper panels of A-D).

We next infected Huh7 and Huh7.5.1 cells with various doses of HCV and harvested the cells 9 and 4 days after infection, respectively. MVP mRNA levels were positively correlated with the doses of HCV in both Huh7 (Fig. 2C, upper panel) and Huh7.5.1 (Fig. 2D, upper panel) cells. Viral infections resulted in maximal induction of MVP at an MOI of 10 in Huh7 cells, but at an MOI of 1 in Huh7.5.1 cells. High infection rates of HCV (MOI = 10) in Huh7.5.1 cells attenuated MVP induction. Western blot analyses mirrored the MVP transcript data (Fig. 2C,D, lower panels). Additionally, the induction of MVP by HCV was significantly higher in Huh7.5.1 cells than in Huh7 cells, likely because Huh7.5.1 cells are highly permissive for HCV replication.

Transcriptional Regulation of MVP by HCV Protein NS5A.

HCV encodes 10 viral proteins. To investigate which viral protein plays a role in MVP regulation, Huh7.5.1 cells were cotransfected with each of 10 HCV gene-expressing plasmids and pMVP-Luc that contains the MVP promoter region spanning +300 to −1,440 bp. The luciferase activity assays indicated that NS5A and Core protein stimulated MVP promoter activity, whereas the other HCV proteins had no significant effects (Fig. 3A, upper panel). Real-time RT-PCR and western blot analyses also indicated that NS5A increased MVP mRNA and protein expression (Fig. 3A, lower panel).

Figure 3.

Functional analysis of cis-regulatory elements involved in the activation of the MVP promoter regulated by NS5A. (A) Huh7 cells were cotransfected with MVP-luciferase reporter plasmid (pMVP-Luc) and the indicated viral proteins for 48 hours prior to luciferase assays (upper panel). Huh7 cells were transfected with the indicated plasmids for 48 hours prior to real-time RT-PCR and western blot analyses (lower two panels). (B) Schematic diagram of individual MVP-promoter cis-regulatory elements and MVP truncated or site-specific mutants (left) and the results from luciferase activity assays (right). Huh7 cells were transfected with the indicated plasmids for 48 hours prior to luciferase assays. (C) Huh7 cells were transfected with the indicated plasmids and treated with or without NF-κB inhibitor Bay11-7082 for 48 hours prior to luciferase assays (top panel), RT-PCR (middle panel), and western blot (bottom panel) analyses. (D) Experiments were performed similar to those in (C), except SP1-specific siRNA (siRNA-SP1) was used. (E,F) Chromatin immunoprecipitation analyses were performed using two separate MVP promoter amplicons, SP1 binding site (E) and NF-κB binding site (F). Cells were transfected with pCMV-NS5A or control vector. Total input and IgG served as positive and negative controls, respectively. All experiments were repeated at least three times with similar results. Bar graphs represent means ± SD, n = 3 (**P < 0.01; *P < 0.05).

To further investigate the transcriptional regulation of MVP, we examined the MVP promoter region for possible consensus cis-elements using software from http://www.gene-regulation.com. Three promoter truncations and two binding-site mutants (NF-κB and Sp1 binding sites) were generated from the full-length MVP promoter plasmid (pMVP-Luc; Fig. 3B). In reporter assays, NS5A expression stimulated expression of the wildtype promoter and the two truncated mutants (Fig. 3B). However, mutation of either NF-κB or Sp1 abolished NS5A-stimulated promoter activity (Fig. 3B). To determine whether MVP expression is NF-κB and Sp1-dependent, we chose an NF-κB inhibitor (Bay11) and designed two specific small interfering RNAs (siRNAs) for Sp1 (siRNA-Sp1#1 and siRNA-Sp1#2) and tested their efficiency (Supporting Fig. 2). SiRNA-Sp1#1 was selected for experiments described below. The MVP promoter activity was suppressed by NF-κB inhibitor (Bay11; Fig. 3C) and by Sp1-specific siRNA (Fig. 3D). RT-PCR and western blot analyses validated the promoter assay results (Fig. 3C,D). Chromatin immunoprecipitation was used to confirm NF-κB and Sp1 binding to the MVP promoter. Antibodies against NF-κB and Sp1 immunoprecipitated chromatin fragments containing nucleotides −347 to −288 and nucleotides −175 to −58 of the MVP promoter, respectively. In contrast, a nonspecific IgG did not immunoprecipitate the MVP promoter (Fig. 3E,F). Together, these results demonstrate that NF-κB and Sp1 signaling pathways are involved in the up-regulation of MVP expression in response to NS5A.

Anti-HCV Effect on MVP Expression in Huh7 and Huh7.5.1 Cells.

Because HCV induced MVP production, we investigated whether MVP plays a role in HCV replication. In real-time RT-PCR and western blot assays, overexpression of MVP significantly inhibited HCV RNA replication and viral protein (Core and NS3) expression in Huh7 (Fig. 4A) and Huh7.5.1 cells (Fig. 4B). In addition, overexpression of MVP and concurrent treatment with IFN-α (300 IU/mL) synergistically inhibited HCV replication (Fig. 4A,B). Conversely, higher levels of HCV RNA replication and viral protein expression were present in Huh7 and Huh7.5.1 cells when MVP was knocked down by MVP-specific short hairpin RNA (shRNA) (Fig. 4C,D). Moreover, the antiviral activity of IFN-α was inhibited by MVP-specific shRNA (Fig. 4C,D). To ensure that the suppression of HCV replication observed were not due to the off-target effects of shRNA constructs, further experiments were perform with a special sh-MVP-resistant version of pCMV-MVP, ΔpCMV-MVP. Coexpression of pCMV-MVP or ΔpCMV-MVP with shRNA-MVP followed by a western blot analysis revealed that although pCMV-MVP expression was inhibited by sh-MVP, ΔpCMV-MVP was resistant to sh-MVP mediated down-regulation (Supporting Fig. 2). These data suggest that MVP positively regulates cellular antiviral immunity against HCV.

Figure 4.

MVP inhibits HCV replication. (A) Huh7 (A) and Huh7.5.1 (B) cells were infected with HCV (MOI = 1) for 7 days and 4 days, respectively. Cells were transfected with pCMV-MVP for 24 hours and treated with or without IFN-α (300 IU/mL) for 24 hours prior to real-time RT-PCR (upper panel) and western blot (lower panel) analyses. (C) Experiments were performed as in A except cells were transfected with MVP-specific shRNA (shRNA-MVP). (D) Experiments were performed as in B except cells were transfected with shRNA-MVP. In the real-time RT-PCR experiments of (A,B), the vector control was designated as 1, and for those in (C,D), IFN treatment was designated as 1. Experiments were performed three times with similar results (lower panels of A-D). Graphs represent means ± SD, n = 3 (upper panels of A-D) (**P < 0.01; *P < 0.05).

MVP Induces Type-I IFN Production Through Activation of the IRF7 and NF-κB Signaling Pathway.

To determine the mechanism of the MVP-mediated anti-HCV action, we performed microarray analyses to identify MVP-regulated genes in Huh7 and Huh7.5.1 cells. A series of IFN-inducible genes and NF-κB-regulated genes were up-regulated after transfection with pCMV-MVP (Supporting Table 4). Thus, we determined whether MVP had the ability to induce the production of type-I IFNs in Huh7 cells. Real-time RT-PCR analyses revealed that intracellular IFN-α and IFN-β mRNA levels were enhanced by MVP expression (Fig. 5A,B, left panel). Elevated levels of secreted IFN-α and IFN-β in culture supernatants were also observed by ELISA (Fig. 5A,B, right panel). These results indicate that MVP regulated type-I IFN production.

Figure 5.

MVP is involved in type-I IFN production and IRF7/NF-κB activation. (A,B) Huh7 cells and PBMCs were transfected with pCMV-MVP or control vector for 48 hours prior to real-time RT-PCR (left panel) and ELISA (right panel) analyses for IFN-α (A) and IFN-β (B). (C) Huh7 cells were transfected as in (A) and subjected to real-time RT-PCR (left panel) and western blot (right panel) analyses. (D) Huh7 cells were transfected with NF-κB dual-luciferase reporter plasmid and an empty vector or pCMV-MVP for 48 hours prior to luciferase assays (upper panel). TNFα treatment was included as a positive control for comparison. Western blot analysis of TNF-α-induced IκB degradation and phosphorylation (lower panel). (E) Huh7 cells were transfected as in (A) and subjected to western blot analyses. (F) Huh7 cells were transfected with pCMV-MVP or control vector plasmid. Cytosolic and nuclear extracts were prepared at the indicated timepoints and subjected to western blot analyses. Lamin A and β-actin were used as markers for nuclear and cytosolic fractions, respectively. All experiments were repeated at least three times with similar results. Bar graphs represent means ± SD, n = 3 (**P < 0.01; *P < 0.05).

Induction of type-I IFNs requires coordinated and cooperative actions of the transcription factors IRF3/7 and NF-κB. Real-time RT-PCR and western blot experiments showed that MVP expression greatly enhanced IRF7 expression, but had little effect on IRF3, TLR9, TLR7, and MyD88 expression (Fig. 5C). The effect of MVP expression on IRF7 expression was not cell-type-specific because similar results were observed in a wide range of cell types (PBMCs, human rhabdomyosarcoma [RD], Madin-Darby canine kidney [MDCK], Huh7.5.1, and L-02 cells; Supporting Fig. 3).

Because type-I IFN production also needs NF-κB activation, we determined whether MVP expression affected NF-κB expression. In reporter assays, MVP significantly stimulated NF-κB activation (Fig. 5D). We also examined the effect of MVP on the IkappaB kinase (IKK)/NF-κB pathway. Western blot experiments indicated that MVP expression induced IKKα and IKKβ expression and IκBα phosphorylation and degradation (Fig. 5E). Furthermore, we examined the effect of MVP on the translocation of IRF7 and NF-κB from the cytosol to the nucleus, which is a hallmark of type-I IFN production. Western blot analyses revealed that IRF7 and NF-κB protein levels were lowered in the cytosol and elevated in the nucleus 24 hours after MVP expression, but IRF3 expression was unchanged (Fig. 5F). Similar results were obtained by immunofluorescence assays (Supporting Fig. 4).

We further examined whether the expression of ISGs was affected by MVP expression. Real-time RT-PCR and western blot analyses indicated that MVP expression enhanced the expression of conventional ISGs, such as Stat 1/2, PKR, and OAS, and that treatment of MVP-transfected Huh7 cells with IFN-α resulted in the strongest induction of ISG expression (Supporting Fig. 5). These observations strongly suggest a potential role of MVP in the expression regulation of type-I IFNs and ISGs, thereby implicating MVP in the modulation of type-I IFN-based antiviral defenses.

MVP Is Required for Virus-Triggered IRF7 and NF-κB Activation.

We determined whether endogenous MVP expression was required for virus-triggered IRF7 and NF-κB activation and type-I IFN induction. Knockdown of MVP expression inhibited VSV-induced type-I IFN mRNA expression and protein secretion (Fig. 6A,B). As expected, reporter assays indicated that MVP knockdown inhibited VSV-triggered activation of the NF-κB promoter (Fig. 6C). Results from real-time RT-PCR experiments confirmed that MVP knockdown also inhibited virus-induced expression of endogenous IRF7 (Fig. 6D). Consistently, MVP knockdown suppressed VSV-induced translocation of IRF7 and NF-κB to the nucleus, whereas IRF3 was not affected (Fig. 6E). Taken together, these data suggest that MVP positively regulates virus-triggered activation of IRF7, NF-κB, and type-I IFNs.

Figure 6.

MVP is required for virus-induced signaling. (A,B) Freshly isolated PBMCs were transfected with the indicated plasmids for 12 hours, infected with VSV or mock infected for 36 hours, and subjected to real-time RT-PCR (left panels) and ELISA (right panels) analyses for IFN-α (A) and IFN-β (B). (C,D) Huh7 cells were transfected with the indicated plasmids for 12 hours, infected with VSV or mock infected for 36 hours, and subjected to luciferase assays (C) or real-time RT-PCR for IRF7 mRNA levels (D). (E) Huh7 cells were transfected with the indicated plasmids for 12 hours, infected with VSV for 36 hours, and fractionated into cytoplasmic (left panel) and nuclear (right panel) extracts for western blot analysis. Experiments were performed three times with similar results (E). Graphs represent means ± SD, n = 3 (A-D) (**P < 0.01; *P < 0.05).

VSV, IAV, and EV71 Induce MVP Expression and MVP Hampers Viral Replication and Production.

To examine whether other viruses also induced MVP expression, Huh7, MDCK, and RD cells were infected with VSV, IAV, and EV71, respectively. Real-time RT-PCR and western blot analyses indicated that all three viruses were able to induce MVP mRNA and protein expression (Fig. 7A-C).

Figure 7.

The relationship between MVP expression and VSV, IAV, and EV71. (A-C) VSV, IAV, and EV71 induced MVP expression. (A) Huh7 cells were infected with VSV (MOI = 1) for 6 hours. (B) MDCK cells were infected with influenza A virus (H3N2; MOI = 0.1) for 24 hours. (C) RD cells were infected with EV71 (MOI = 1) for 6 hours. MVP expression was quantified by real-time RT-PCR (upper panels) and western blots (lower panel). (D) Huh7 cells were transfected with the indicated plasmids, infected with VSV (MOI = 1) 24 hours later, and treated with or without 300 IU/mL IFN-α. Supernatants were harvested at 24 hours postinfection and analyzed for VSV production using standard plaque assays. (E) MDCK cells were transfected with the indicated plasmids, infected with IAV (MOI = 0.1) 48 hours later, and treated with or without 300 IU/mL IFN-α. Supernatants were harvested at 24 hours postinfection and analyzed for IAV production using hemagglutination assays. (F) RD cells were transfected with the indicated plasmids and further treated with or without 300 IU/mL IFN-α 24 hours later. Thirty-six hours after transfection, cells were infected with EV71 (MOI = 1). Cells were harvested at 12 hours postinfection. EV71 RNA levels were determined by real-time RT-PCR with EV71 VP1-specific primers. Experiments were performed three times with similar results (lower panels of A-C). All graphs represent means ± SD, n = 3 (**P < 0.01; *P < 0.05).

VSV is a pathogen that is extremely sensitive to the action of type-I IFNs.21 Thus, we investigated the effect of MVP on VSV production. MVP expression in huh7 cells resulted in decreased virus titers (Fig. 7D). Because MVP exerts its antiviral function by inducing type-I IFN, we hypothesized that MVP expression would not induce antiviral activity in an IFN-deficient cell line. To test our hypothesis, Vero cells were utilized, as they are deficient for type-I IFNs due to genomic deletions.22 As expected, MVP expression had no influence on viral replication in the type-I IFN-deficient Vero cells (Supporting Fig. 6A).

We assessed the effects of MVP on IAV production in MDCK cells. Virus titers were significantly reduced in cells overexpressing MVP as compared with control cells. Virus titers were lowest in infected cells overexpressing MVP and cotreated with IFN-α (Fig. 7E). We also examined the effect of MVP on viral transcription and replication by measuring the production of three different forms of IAV RNA (mRNA, cRNA, and vRNA) using an approach described previously.23 Real-time RT-PCR analyses showed that the levels of NP-specific mRNA, complementary RNA [cRNA], and vRNA were suppressed by MVP expression (Supporting Fig. 6B).

We examined the effects of MVP on EV71 replication in RD cells. MVP effectively suppressed viral VP1 mRNA (Fig. 7F) and protein (Supporting Fig. 6C) levels. The combination of MVP overexpression and IFN-α treatment resulted in a significant synergistic reduction of viral VP1 in RD cells. In conclusion, MVP exhibits antiviral activity toward a broad range of viral infections.

Discussion

In this study we identified a previously undescribed mechanism in which inducible MVP plays an important role in endogenous type-I IFN production in response to viral infections. MVP exhibited strong antiviral activity toward a broad range of viral infections in a type-I IFN-dependent manner.

Currently, two reports have shown that viral infections induce vault expression.24, 25 One study compared the global gene expression differences between HepG2 cells and HepG2.2.15 cells using human genome-wide oligonucleotide microarrays. HepG2.2.15 cells are stable HBV-producing cells and HepG2 is the uninfected parental cell line. MVP was one of eight genes up-regulated in HepG2.2.15 cells.24 The other study used a new method, subtractive hybridization of noncoding RNA transcripts, or SHORT, to identify differentially expressed noncoding RNAs between uninfected cells and Epstein-Barr virus-infected cells. Vault RNAs were strongly induced following Epstein-Barr virus infection.25 In this study we identified cellular signaling pathways that contribute to MVP expression during HCV infection.

We examined MVP mRNA levels in PBMCs of healthy participants and HCV patients and found that MVP mRNA levels were significantly higher in HCV patients (Fig. 1A,B). MVP protein expression was elevated in the sera and liver tissues of HCV patients (Fig. 1D). Huh7 and Huh7.5.1 cell lines are an effective cell culture model of HCV infection.19, 20 Thus, we next infected Huh7 cells and Huh7.5.1 cells with an HCV genotype 2a replicon (JFH-1) and measured the expression of MVP. JFH-1 expression induced MVP mRNA and protein expression in a time- and dose-dependent manner (Fig. 2).

It remains controversial whether PBMCs are suitable for HCV infection analyses. Some studies indicate that PBMCs can function as an important extrahepatic reservoir or a possible site for HCV replication.26, 27 However, another study found that HCV RNA levels in PBMCs are dependent on serum concentrations of HCV, suggesting that HCV cannot replicate in PBMCs.28 In this study we treated PBMCs from healthy donors with HCV and found that HCV was able to stimulate MVP expression (Supporting Fig. 1). Moreover, HCV RNA levels in PBMCs increased in a time- and dose-dependent manner. However, it is unclear whether the increase in HCV RNA levels in PBMCs was due to HCV replication or passive adsorption.

To further investigate the mechanism of MVP induction by HCV, we tested the effect of 10 individual HCV proteins on MVP promoter activity. Only the viral proteins NS5A and Core protein induced MVP promoter activity. NS5A expression also enhanced MVP mRNA and protein expression (Fig. 3A). However, the increase in MVP promoter activity by NS5A was less than the increase in MVP protein levels (Fig. 3A). Interestingly, a prior study also found that IFN-γ modestly enhanced MVP promoter activity, but greatly elevated MVP protein expression (up to 11-fold).18 The authors verified that IFN-γ influences the half-life of the newly translated MVP protein. Further studies will determine whether NS5A similarly regulates MVP protein stability.

A critical step in innate immune responses to viral infections is the production of type-I IFN, which is primarily controlled at the transcriptional level, wherein IRF3/IRF7 and NF-κB play pivotal roles.29 Inactive IRF3/IRF7 predominantly resides in the cytoplasm, but translocates to the nucleus upon phosphorylation induced by viruses recognized by pattern recognition receptors.30 On the other hand, NF-κB proteins are present in the cytoplasm in association with the inhibitory IκB proteins.31 Upon viral infection, phosphorylated IKK complexes can phosphorylate IκBα, which is subsequently ubiquitinated and degraded by way of the proteasome pathway.32 NF-κB is then released and translocated to the nucleus. Both activated IRF3/IRF7 and NF-κB bind specific promoter elements of the type-I IFN genes, resulting in type-I IFN induction. It was previously reported that HCV impairs IRF-7 activation and IFN-α induction in IHH, the immortalized human hepatocytes.33 Our data show that HCV-induced MVP up-regulated IRF-7 expression in Huh7 cells. The discrepancy may come from different cell types used. We further confirmed MVP induced IRF-7 activation in a wide range of cell types, including PBMCs, RD, MDCK, Huh7.5.1, and L-02 cells.

The localization of MVP is very intriguing. Generally, MVP/vaults reside in the cytoplasm, but they can also be found in the nucleus depending on cell status and type.34 In human MCF-7 breast cancer cells, approximately 5% of the MVP is localized to the nucleus.35 Moreover, in 253J cells the majority of MVP is nuclear.16 MVP expression is sufficient to form the basic barrel-shaped vault structure with a hollow interior.36 A mass was occasionally found in the inner hollow cavity. The structure and localization of MVP has led to the hypothesis that MVP may function as a nuclear-cytoplasmic transport vehicle. In this study, we demonstrated that MVP overexpression resulted in IRF7 and NF-κB translocation to the nucleus (Fig. 5), whereas MVP knockdown inhibited VSV-induced activation of IRF7 and NF-κB (Fig. 6). Consistently, MVP overexpression inhibited replication and production of HCV, VSV, IAV, and EV71 (Fig. 7). It is likely that MVP is downstream of TLRs and RIG-I like receptors (RLRs), as is RIG-I, in the signaling cascade, functioning in IRF7 and NF-κB transport to the nucleus, and resulting in type-I IFN production during viral infection.

We propose a working model of the role of MVP in the virus-triggered induction of type-I IFNs and intracellular antiviral immunity (Fig. 8). In this model, viral infection strongly induces MVP expression through the NF-κB and Sp1 pathways. The enhanced MVP expression results in type-I IFN expression by promoting IRF7 and NF-κB translocation into the nucleus, leading to inhibition of viral replication and production. Although more studies are needed to understand the delicate regulatory mechanisms of MVP in viral replication and antiviral responses, our findings reveal a previously undescribed role for MVP in the regulation of cellular antiviral responses.

Figure 8.

A hypothetical model for MVP induction following viral infection and the relationship between MVP levels and type-I IFN antiviral immunity. Solid arrows represent signaling pathways identified in this study. Broken arrows indicate potential signaling pathways. Viruses (HCV, VSV, IAV, and EV71) induce MVP expression through the NF-κB and SP1 signal pathways. In turn, MVP suppresses viral replication and production by activating IRF7 and NF-κB, which leads to type-I IFN production.

Acknowledgements

We thank the Core Facility at College of Life Sciences, Wuhan University, for the use of their confocal microscope. We thank Ms. Hu Ying for technique help.

Ancillary