The capsid protein of the Flaviviridae family members is involved in nucleocapsid formation and virion assembly. However, the influence of C protein-interacting partners on the outcome of pestivirus infections is poorly defined. In this study, the hemoglobin subunit beta (HB) was identified as a C protein-binding protein using glutathione S-transferase (GST) pulldown and subsequent mass spectrometry analysis of PK-15 cells which are permissive cells of classical swine fever virus (CSFV). Coimmunoprecipitation and confocal microscopy confirmed that HB interacts and colocalizes with the C protein in cytoplasm. Silencing of HB with small interfering RNAs promoted CSFV growth and replication, whereas overexpression of HB suppressed CSFV replication and growth. Interestingly, HB interacts with retinoic acid inducible gene I (RIG-I) and increases its expression, resulting in increased production of type I interferon (IFN). Overall, our results suggest that cellular HB antagonizes CSFV growth and genomic replication probably by triggering IFN signaling and might represent a novel antiviral restriction factor. This study reports for the first time about the novel role of HB in innate immunity.
Keywords: classical swine fever virus; C protein; hemoglobin subunit beta; interferon-beta; antiviral protein; virus-host interaction
Introduction
Classical swine fever virus (CSFV) is the causative agent of classical swine fever (CSF), a highly contagious disease in pigs responsible for significant economic losses worldwide. CSFV belongs to the genus Pestivirus of the family Flaviviridae. The RNA genome of CSFV is a single-stranded, positive-sense RNA molecule approximately 12.3 kb in length that encodes a polyprotein that is processed to yield 12 mature proteins (NH2-Npro-C-Erns-E1-E2-p7-NS2-NS3-NS4A-NS4B-NS5A-NS5B-COOH) (1). The structural proteins of CSFV consist of the glycoproteins Erns, E1 and E2, and the capsid (C) protein, a nucleocapsid protein of unknown symmetry (2).
The C protein of pestiviruses is a small, highly basic polypeptide that has been shown to bind to RNA with low affinity and specificity (3, 4). The C protein of CSFV influences the regulation of cellular transcription (5) and interacts with host SUMOylation proteins (6). The ability of the C protein to bind the cellular IQGAP1 protein during infection plays a critical role in determining the virulence of the virus (7). Recently, some studies showed that the C protein is dispensable for infectious virion formation and virus propagation in vitro (8, 9).
Studies of the C protein of hepatitis C virus (HCV), another member of the Flaviviridae family, provide further insight into the possible functions of the C protein of CSFV. The HCV C protein has been shown to be crucial for virus particle production because it is the structural component of the viral nucleocapsid and required for formation of the active HCV replication/assembly complex in host cells (10, 11). A number of studies have revealed that a variety of host proteins, such as vimentin, interact with the C protein (12, 13) and influence the pathogenesis of HCV by modulating processes such as cell signaling, transformation and proliferation, cellular and viral gene expression, apoptosis, and lipid metabolism (14-19).
Hemoglobin is the metalloprotein responsible for oxygen transport and is present in the red blood cells of all vertebrates (20) except members of the fish family Channichthyidae (21). The hemoglobin molecule is comprised of four globular protein subunits (α2β2), each of which is composed of a protein chain tightly associated with a non-protein heme group. Each protein chain is arranged as a set of alpha-helix structural segments connected in structure known as a globin fold arrangement, so-called because this folding motif occurs in other heme/globin proteins, such as myoglobin (22). Hemoglobin has an oxygen binding capacity of 1.34 ml O2 per gram of hemoglobin (23), which increases the total blood oxygen capacity 70-fold over that of dissolved oxygen in blood. The mammalian hemoglobin molecule can bind (carry) up to four oxygen molecules at one time (24). Hemoglobin is also involved in the transport of other gases; it carries some of the body's respiratory carbon dioxide (about 10% of the total) as carbaminohemoglobin, as well as the important regulatory molecule, nitric oxide, bound to a globin protein thiol group, releasing it at the same time as oxygen (25). Hemoglobin is also found outside red blood cells and their progenitor lines. Other cells that contain hemoglobin include A9 dopaminergic neurons in the substantia nigra, macrophages, alveolar cells, and mesangial cells in the kidney. In these tissues, hemoglobin functions as an antioxidant and a regulator of iron metabolism (26).
To date, some studies have focused on the roles of the structural glycoproteins of CSFV in virus replication and virulence (2, 27, 28), but information on C protein-interacting proteins and their impact on the outcome of CSFV infection is limited. Here, we show that the cellular hemoglobin subunit beta (HB) interacts specifically with the CSFV C protein and antagonizes CSFV growth and genomic replication.
Materials and methods
Cells, viruses, and virus titer assays. HEK293T cells, SK6 cells and PK-15 cells were grown in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum (FBS) known to be free of bovine pestivirus and anti-pestivirus antibodies. The CSFV strain Shimen was propagated in CSFV-permissive PK-15 cells and SK6 cells. Virus titers in the culture supernatants of CSFV-infected PK-15 and SK6 cells were determined by the Reed-Muench method (29).
Plasmids. The p3-Flag-HB plasmid encoding the HB protein (GenBank No. NM_001144841.1) with a Flag tag at its N-terminus was constructed by cloning HB cDNA into a p3-Flag-CMV10 vector (E7658, Sigma, USA) using EcoRI and KpnI restriction enzymes. The CSFV C protein cDNA (GenBank No. AY775178.2) was cloned into the pCMV-Myc vector (Clontech, USA) using EcoRI and XhoI restriction enzymes to generate the pMyc-C plasmid. Retinoic acid inducible gene I (RIG-I) (GenBank No. NMïŸ213804.2) or melanoma differentiation-associated gene 5 (MDA5) (GenBank No. NMïŸ001100194.1) protein cDNA was cloned into the pCMV-Myc vector (Clontech, USA) using NotI and XhoI or EcoRI and XhoI restriction enzymes to generate the pMyc-RIG-I or pMyc-MDA5 plasmid. For bacterial expression of the GST-tagged C protein, the C protein gene was subcloned into the pGEX-6p-1 vector (28-9546-48, GE Healthcare, USA), creating pGEX-C. A series of HB deletion mutants was generated from p3-Flag-HB by conventional PCR techniques using the mutagenesis primers listed in Table 1. All plasmids were verified by sequencing.
Plasmid DNA transfection. Cells in 6- or 12-well plates (Nunc, USA) cultured in a humidified 37°C CO2 incubator were transfected with plasmids (4 μg each) using X-tremeGENE HP DNA transfection reagent (No. 06366236001, Roche, Germany), according to the manufacturer's instructions. At 4 h post-transfection (hpt), the transfection mix was replaced with complete growth medium and incubated for an additional 48 h before being used for assays.
Virus infection and treatment. At 48 h after DNA or siRNA transfection, cells were infected with CSFV Shimen strain at a multiplicity of infection (MOI) of 0.1 or Sendai virus (SeV) at an MOI of 1. After 2 h, the viral inoculum was removed and the infected cells were washed twice with PBS (pH 7.4) and re-fed with DMEM containing 2% FBS. At various time points post-infection (hpi), cell-free culture supernatants and cell lysates were harvested and stored at -80°C until use. To examine the effects of IFN-β, SK6 cells were treated with or without IFN-β (1000 IU/ml) (ab87929, Abcam, England) at 12 h prior to infection with CSFV, and CSFV titer and genomic replication were analyzed.
Glutathione S-transferase (GST) pulldown assays. For GST pulldown assays, GST or GST-C proteins expressed in E. coli BL21(DE3) cells were conjugated to glutathione beads (10049253, GE Biosciences, USA) and blocked for 1 h in 5% BSA. The beads were then washed twice with TIF buffer [20 mM Tris-HCl (pH 8.0), 150 mM NaCl, 1 mM MgCl2, 0.1% NP-40, 10% glycerol, 0.1 mM DTT, and 1 mg/ml protease inhibitor] and incubated with recombinant Flag-tagged HB harvested from transfected HEK293T cells for 2 h at 4°C. The beads were washed at least six times with TIF buffer, followed by elution and detection of the proteins by SDS-PAGE and immunoblotting.
Coimmunoprecipitation (co-IP). HEK293T cells were transfected with the indicated constructs, as described above. The transfected cells were harvested 48 h after transfection, washed three times with cold phosphate-buffered saline (PBS) (pH 7.4), and lysed with NP40 buffer [50 mM Tris (pH 8.0), 150 mM NaCl, 0.5% NP40, 0.5 mM EDTA] containing 1 mM PMSF and 1 mg/ml protease inhibitor cocktail (Roche, Germany) at 4°C for 1 h. Clarified extracts were pre-cleared with Protein A/G beads (SC-2003, Santa Cruz, USA), and then incubated with Protein A/G beads plus anti-Flag monoclonal antibody (mAb) (F3040, Sigma, USA) for 4 h. The beads were then washed with NP40 buffer, boiled in sample buffer and subjected to SDS-PAGE followed by immunoblotting analysis with anti-Flag (F7425, Sigma, USA) and anti-Myc polyclonal antibodies (pAb) (E022050-1, Earthox LLC, USA).
Real-time RT-PCR. Following the indicated treatment, total cellular RNA was extracted from CSFV-infected cells using TRIzol (15596026, Invitrogen, USA) and treated with DNase I to remove potential genomic DNA contamination. The isolated RNA was then reverse transcribed to cDNA using Moloney murine leukemia virus reverse transcriptase (TaKaRa, Japan) according to the manufacturer's instructions. Quantification of genomic copies of CSFV was performed by a previously described quantitative real-time RT-PCR assay (30).
Quantitative real-time PCR was performed by using SYBR Permix Ex Taq II (DRR081A, TaKaRa, Japan) with Light Cycler 480 II real-time PCR system (Roche, Germany). The primers used to detect interferon-beta (IFN-β) were listed in Table 1. The relative abundance of each target was obtained by normalization with the endogenous GAPDH.
RNA interference. Small interfering RNAs (siRNAs) targeting HB or RIG-I were used at a final concentration of 400 or 200 nM, unless otherwise stated. Cells were transfected with X-tremeGENE siRNA Transfection Reagent (4476093001, Roche, Germany), as described previously (31). The siRNA target sequences of HB were GGACGAAGTTGGTGGTGAG (siHB-1) and GGTGCATCTGTCTGCTGAG (siHB-2). The siRNA target sequences of RIG-I were GGTACAAAGTTGCAGGCA (siRIG-I-1), GCAAACAGCATCCTTATAA (siRIG-I-2) and CCATAACTCTTGGAGGCTT (siRIG-I-3). Western blotting was used to analyze the endogenous swine RIG-I expression using goat anti-RIG-I PAb (SC-48932, Santa Cruz).
Confocal imaging. HEK293T cells were cotransfected with pMyc-C (2 μg) and p3-Flag-HB (2 μg). After 48 h incubation, transfected or infected CSFV cells were fixed with 4% paraformaldehyde in PBS for 30 min and permeabilized with 0.1% Triton X-100 for 15 min. The cells were then incubated with anti-Myc mAb (E022050-1, Earthox, USA) or anti-C mAb (produced in-house) (32) for 2 h, followed by incubation for 2 h with anti-Flag pAb (F7425, Sigma, USA) or anti-HB pAb (SC-22718, Santa Cruz, USA). The cells were then incubated with anti-mouse IgG (whole molecule)-FITC antibody produced in goat (F2012, Sigma, USA) and anti-rabbit IgG (whole molecule)-TRITC antibody produced in goat (T6778, Sigma, USA) or an anti-goat IgG (whole molecule)-TRITC antibody (T7028, Sigma). Cells were stained with 4, 6-diamidino-2-phenylindole (DAPI) for 15 min and examined using a Leica SP2 confocal system (Leica Microsystems, Germany). Signal colocalization was analyzed with the program Colocalizer Pro (Colocalization Research Software, Boise, ID).
Luciferase (Luc) reporter assay. HEK293T cells (105) in 24-well plates were transfected with 500 ng of pIFN-β-Luc, 10 ng of pRL-SV40-Renilla (Promega, USA) as an internal control, and 1 μg of each of the indicated expression vectors. After 24 h, cells were transfected with or without 1.0 μg of poly(I:C). PK-15 cells (105) in 24-well plates were transfected with 1 μg of pIFN-β-Luc, 20 ng of pRL-SV40-Renilla as an internal control and 400 nM siHB. After 24 h, cells were infected with CSFV or SeV for 48 h. Reporter gene activity was analyzed using the Dual-Luciferase Reporter 1000 Assay System (Promega, USA) and measured with a TD-20/20 Luminometer (Turner Designs, USA) according to the manufacturer's instructions. Three independent experiments were carried out in duplicate. Error bars represent standard deviations from the mean (mean ± SD).
Protein extraction from swine tissues and isolation of porcine peripheral blood mononuclear cells (PBMCs). Tissues were collected from 3-month-old healthy crossbreed pigs and stored at ï€80°C. The experimental and animal handling protocols were approved by the Ethics Committee on Experimental Animal Usage and Animal Welfare, Harbin Veterinary Research Institute, CAAS. Frozen tissues were broken into small pieces and washed to remove blood using PBS. Then the tissue pieces were ground into homogenate with a mortar and pestle. The tissues were washed three times with ice-cold PBS and lysed with red blood cell lysis buffer (C3702, Beyotime, China) to remove red blood cells for 30 min at 4°C. The lysates were then clarified by centrifugation at 10,000 rpm for 10 min at 4°C. The tissues were lysed by lysis buffer containing 7 M urea, 2 M thiourea, 4% (w/v) CHAPS, 65 mM DTT, 0.2% Bio-Lyte 3/10 and 1 mM PMSF at a volume ratio of 1:10 for 1 h on ice, and the lysates were then clarified by centrifugation at 10,000 rpm for 30 min at 4°C. The supernatants were collected and the protein concentration was determined by the Bradford assay (PA102, Tiangen, China).
Heparinized blood was diluted (1:3) in sterile PBS (pH 7.4), and subsequently layered over pig lymphocyte separation medium (LTS1110, TBDscience, China) according to the manufacturer's protocol. After centrifugation at 2000 - g for 20 min at room temperature, the mononuclear cell band at the interface was removed and washed twice with basal medium by centrifugation at 2000 - g for 10 min at room temperature. PBMCs were lysed with red blood cell lysis buffer (C3702, Beyotime, China) to remove red blood cells for 30 min at 4°C. The lysates were then clarified by centrifugation at 10,000 rpm for 10 min at 4°C. And then PBMCs were lysed with NP40 buffer for 30 min on ice and the lysates were then clarified by centrifugation at 10,000 rpm for 30 min at 4°C. The supernatants were collected and the protein concentration was determined as described above.
Statistical analysis. Statistical analyses were performed using SPSS 13.0 software. Student's t-test or one-way ANOVA were used to compare viral titers and genomic replication. A P value of < 0.05 was considered significant.
Results
CSFV C-interacting proteins identified by GST pulldown and mass spectrometry. To identify cellular proteins that potentially associate with the CSFV C protein in PK-15 cells, GST pulldown was coupled with mass spectrometry (MS). Plasmids encoding GST-tagged C protein or the GST tag alone were expressed in Escherichia coli BL21(DE3) cells and GST or GST-C proteins were affinity purified using GST beads. The beads were then added to PK-15 cell lysates from and incubated for 4 h. Bound proteins were eluted from the beads and resolved by SDS-PAGE followed by Coomassie staining.
In addition to several common bands detected in both GST and GST-C lanes, at least two specific bands were detected in the GST-C protein (Fig. 1A, identified by an asterisk). These protein bands were excised and subjected to MS analysis. Several cellular proteins were identified as C protein binding proteins in this analysis, and their identities, including the number of separate peptides detected by MALDI-TOF/MS analysis, are shown in Table 2. The present study focuses on the cellular protein HB, which was detected in Band 1.
To determine the tissue and cell distribution of HB, the expression of swine HB protein was assessed by using immunoblotting. The results showed that HB was expressed in all swine tissues and cells examined, including kidney, liver, lung, spleen, tonsil and PBMCs (Fig. 1B).
HB interacts and colocalizes with C protein. To confirm the interaction between HB and C proteins, co-IP experiments were performed in HEK293T cells transiently coexpressing 3-Flag-tagged HB and Myc-tagged C proteins. Cells coexpressing 3-Flag and Myc-C proteins were used as a control for specificity. Co-IP with anti-Flag mAb showed that Myc-C protein formed a complex with 3-Flag-HB but not with 3-Flag (Fig. 2A). Since the above experiments did not exclude the possibility that C protein-HB association might be mediated indirectly through association with other cellular protein(s), a GST pulldown assay was performed using glutathione beads conjugated to GST-C or GST protein. The presence of GST-C protein, but not GST, resulted in the binding of HB (Fig. 2B), indicating that the interaction of cellular HB with the CSFV C protein is likely due to a direct physical association. To determine whether C protein interacts with cellular HB in the context of CSFV infection, virus-infected PK-15 cell lysates were immunoprecipitated with an anti-C mAb and probed for the presence of HB using anti-HB pAb. HB was readily detected in CSFV-infected PK-15 cells (Fig. 2C), indicating that HB indeed interacts with endogenous C protein in CSFV-infected PK-15 cells. To examine the colocalization of C protein with HB, HEK293T cells were cotransfected with plasmids expressing 3-Flag-HB and Myc-C proteins and the subcellular localization of C protein and HB was examined by confocal microscopy (Fig. 2D). Both Myc-C protein and 3-Flag-HB were distributed throughout the cytoplasm, and C protein colocalized extensively with HB. To confirm that endogenous HB colocalizes with C protein, PK-15 cells were infected with CSFV for 48 h and analyzed by confocal microscopy. Confocal images of the cells immunostained with anti-C and anti-HB antibodies showed colocalization of HB with CSFV C protein (Fig. 2E). On the basis of digital analysis of multiple cell images, 90% and 98% of HB-labeled pixels colocalized with exogenous and endogenous C protein, respectively. The colocalization coefficients were 0.90 and 0.98, respectively. Collectively, these findings confirm that HB is an interacting partner of CSFV C protein.
The C-terminal portion of HB is necessary for interaction with C protein. To determine which domain is necessary for the interaction with C protein, a number of swine HB deletion mutants were generated (Fig. 3A) and assessed for their capacity to interact with C protein. Notably, the C-terminal HB deletion mutant (Δ100-146) lost the ability to associate with C protein (Fig. 3B).
Depletion of HB by siRNAs enhances CSFV growth and C protein expression. To investigate the relevance of the C-HB interaction to the CSFV life cycle, specific siRNAs were used to target HB in PK-15 cells, resulting in efficient knockdown of the protein (Fig. 4A). Knockdown of HB resulted in upregulation of CSFV C protein compared to cells treated with a scrambled siRNA (siScr) or mock-treated cells (Mock) (Fig. 4B). In addition, HB-silenced cells exhibited an increased viral titer in the supernatant (Fig. 4C). These results indicate that knockdown of cellular HB enhances virus growth and protein expression.
Overexpression of HB suppresses CSFV growth and C protein expression. The observation that depletion of HB resulted in an enhanced CSFV protein expression and growth prompted examination of the effects of HB overexpression on CSFV. PK-15 cells were transiently transfected with 3-Flag-HB and subsequently infected with CSFV. CSFV C protein was reduced when HB was upregulated (Fig. 5A). A reduction in viral titers was observed in the supernatants of these cells (Fig. 5B). Together with the results of the knockdown experiments, these findings highlight the antagonistic effect of cellular HB expression on CSFV growth and C protein expression.
HB affects viral genomic replication and CSFV inhibits HB expression. To examine whether HB plays a role in CSFV RNA synthesis, PK-15 cells that were depleted of HB by siRNA were infected with CSFV. Examination of the genomic copies of CSFV RNA in HB-knockdown cells revealed that the synthesis of viral RNA was significantly increased compared to control cells (No treat, Mock, and siScr) (Fig. 6A). Conversely, overexpression of HB resulted in the downregulation of CSFV genomic replication (Fig. 6B). Taken together, these results suggest that HB has a significant inhibitory effect on CSFV genomic replication. Furthermore, the expression of HB in PK-15 cells infected with CSFV was assessed and the results indicated that CSFV inhibits the expression of the endogenous HB (Fig. 6C).
HB activates IFN-β production. IFN actually results in an increase in dimethylsulfoxide-induced hemoglobin synthesis (33). To examine whether HB affects IFN production, the effect of overexpression of HB on poly(I:C)-triggered reporter gene activation was examined. HB increased the poly(I:C)-induced IFN-β transcription level in HEK293T cells (Fig. 7A), indicative of a positive role for HB in the cellular antiviral response. To validate the inhibitory effect of endogenous HB, siRNAs were used to reduce the HB expression in PK-15 cells. HB suppression (Fig. 7B) correlated with decreased IFN-β in SeV- or CSFV-infected PK-15 cells, indicating that HB positively regulates IFN-β transcription. We found that IFN-β mRNA was induced by HB or HB and CSFV infection in PK-15 cells (Fig. 7D). In addition, the expression of HB in PK-15 cells pretreated with IFN-β was examined and the results showed that IFN-β increased the expression of the endogenous HB (Fig. 7C). Collectively, HB activates IFN-β production.
RIG-I interacts with and is upregulated by HB. The dsRNA produced by RNA viruses during viral replication is recognized by sensor molecules, such as RIG-I, MDA5 and LGP2. These sensors initiate signaling cascades, including the activation of transcription factors [IRF-3 (IFN regulatory factor 3), IRF-7] and IFN genes (34). Thus, the potential for HB to interact with the positive regulators of IFN activity, RIG-I and MDA5, was investigated. To this end, HEK293T cells were cotransfected with expression plasmids encoding HB and either RIG-I or MDA5. HB was found to selective interact with RIG-I, but not with MDA5 (Fig. 8A). This finding prompted us to investigate whether HB affects the expression of MDA5 or RIG-I. Western blotting and quantitative analysis revealed that overexpression of HB resulted in an increased expression of RIG-I, but not MDA5 (Fig. 8B and 8C). To explore the effect of the C protein on the expression of RIG-I and HB, HEK293T cells were cotransfected with pMyc-C, pMyc-RIG-I and p3ï‚´Flag-HB plasmids. The expression levels of both HB and RIG-I were reduced with the increased expression of the C protein (Fig. 8D). These data suggest HB inhibits CSFV growth via trigger of RIG-I activation and the C protein inhibits RIG-I expression.
Depletion of RIG-I by siRNAs or lack of IFN-β enhances CSFV growth, genomic replication and C protein expression. To further address the role of endogenous RIG-I in CSFV infection, the endogenous RIG-I expression in PK-15 cells was knocked down by siRNA targeting swine RIG-I gene (Fig. 9A). The results showed that knockdown of RIG-I increased CSFV growth and replication in PK-15 cells (Fig. 9B and 9C). To further ascertain whether IFN has effects on CSFV growth and replication, SK6 cells, which are deficient in type I IFN production (35), were infected with CSFV after treatment with IFN-β. The results indicated that SK6 cells were more efficiently infected with CSFV than those pretreated with IFN-β (Fig. 9D, 9E and 9F). Further, to determine whether HB is still effective at suppressing virus infection in the absence of RIG-I, we detected the expression of HB, RIG-I and C proteins and CSFV growth and replication in PK-15 cells transfected with siRNA and p3-Flag-HB after infection with CSFV (Fig. 9G, 9H and 9I). The results showed that HB was unable to suppress CSFV infection in the absence of RIG-I in PK-15 cells. Overall, depletion of RIG-I or lack of IFN-β enhances CSFV growth, replication and C protein expression.
Discussion
Because of the limited genome coding capacity, viruses depend on host factors to carry out important functions. During the course of evolution, individual viruses have acquired a unique array of multifunctional proteins. In an attempt to identify host cellular proteins that interact with the C protein of CSFV and understand the functional importance of these interactions, GST pulldown and MS were employed to isolate host factors that interact with the CSFV C protein. Among the cellular proteins screened, the HB protein was identified and selected for further analysis. Using confocal analysis, C protein and HB were found to interact (Fig. 2C) and colocalize in the cytoplasm (Fig. 2D). It is likely that the CSFV C protein translocates to the nucleus during later phases of infection, because the CSFV C protein has been shown to influence cellular transcription (5).
Hemoglobin is the red blood cell protein that carries oxygen. It was found in red blood cells of all vertebrates as well as the tissues of some invertebrates, such as insects, worms, clams, crabs, octopus, snails and starfish (36). In this study, we showed that HB exists in all tissues and cells of swine tested. Previous studies paid little attention to hemoglobin in mammalian tissues and cells other than red blood cells, possibly due to the relatively lower abundance.
It has been shown that interaction of the HCV C protein with cellular molecules impairs the host's immune response through mechanisms that result in suppression of IL-12 synthesis in human macrophages (37), T cell dysfunction (38), and inhibition of T-lymphocyte activation and proliferation (25, 39, 40). In Friend leukemia cells treated with interferon, the induction and production of leukemia viruses are inhibited while induction of HB is slightly increased (33). Based on these findings, we speculated that the interaction between C and HB influences interferon signaling. It has been shown that viral invasion triggers an array of host antiviral innate immune responses, resulting in the production of various cytokines and chemokines. To detect viral infections, cells use pattern-recognition receptors to sense viral nucleic acids (41). RIG-I-like helicases (RLHs), including RIG-I and MDA5, function as cytoplasmic RNA sensors that recognize viral RNAs released during virus replication (42). We found that HB mediated IFN signaling by interacting with RIG-I and that HB-mediated suppression of CSFV growth and replication involved upregulation of RIG-I and IFN-β. And CSFV growth and replication were increased when RIG-I was knocked down in PK-15 cells. To examine the change of HB in PK-15 cells infected with CSFV or treated with IFN-β, we detected the expression of HB at different time points after infection with CSFV or treatment with IFN-β. We showed that CSFV/C inhibited the expression of HB and that IFN-β upregulated the expression of HB and inhibited the CSFV growth and replication and that the C protein antagonized HB-mediated upregulation of RIG-I. We proposed a possible relationship among CSFV/C, HB, RIG-I and IFN-β, as illustrated in Fig. 10. In short, CSFV or C protein alone inhibits the expression of HB, HB activates RIG-I, RIG-I activates IFN-β, and IFN-β inhibits CSFV infection and increases HB expression. What and how to active HB and how HB activates RIG-I signaling during viral infection need to be studied further.
The key findings of this study are the identification of cellular HB as a novel interacting partner of the CSFV C protein, and the demonstration that HB inhibits the replication and growth of CSFV through activation of the IFN signaling pathway. Additional studies are needed to define the role of HB as a novel antiviral restriction factor.
Acknowledgements
We are grateful to Drs. Changjiang Weng and Henggui Liu for their helpful suggestions. This study was supported by the National Natural Science Foundation of China (31201921).
