|Home | About | Journals | Submit | Contact Us | Français|
Hepatitis C virus (HCV) nonstructural protein 5A (NS5A) is a component of the replication complex consisting of several host and viral proteins. We have previously reported that human butyrate-induced transcript 1 (hB-ind1) recruits heat shock protein 90 (Hsp90) and FK506-binding protein 8 (FKBP8) to the replication complex through interaction with NS5A. To gain more insights into the biological functions of hB-ind1 in HCV replication, we assessed the potential cochaperone-like activity of hB-ind1, because it has significant homology with cochaperone p23, which regulates Hsp90 chaperone activity. The chimeric p23 in which the cochaperone domain was replaced with the p23-like domain of hB-ind1 exhibited cochaperone activity comparable to that of the authentic p23, inhibiting the glucocorticoid receptor signaling in an Hsp90-dependent manner. Conversely, the chimeric hB-ind1 in which the p23-like domain was replaced with the cochaperone domain of p23 resulted in the same level of recovery of HCV propagation as seen in the authentic hB-ind1 in cells with knockdown of the endogenous hB-ind1. Immunofluorescence analyses revealed that hB-ind1 was colocalized with NS5A, FKBP8, and double-stranded RNA in the HCV replicon cells. HCV replicon cells exhibited a more potent unfolded-protein response (UPR) than the parental and the cured cells upon treatment with an inhibitor for Hsp90. These results suggest that an Hsp90-dependent chaperone pathway incorporating hB-ind1 is involved in protein folding in the membranous web for the circumvention of the UPR and that it facilitates HCV replication.
Hepatitis C virus (HCV) is the major causative agent of non-A, non-B hepatitis in humans and infects approximately 170 million people worldwide (64). HCV belongs to the genus Hepacivirus of the family Flaviviridae and is classified into six major genotypes (39). The virus forms small, round, enveloped particles and possesses a genome consisting of a single positive-stranded RNA with a nucleotide length of 9.6 kb. The viral genome encodes a single precursor polyprotein consisting of approximately 3,000 amino acids, which in turn is posttranslationally processed into 10 viral proteins by host and viral proteases. The structural proteins are cleaved from the N-terminal one-fourth of the polyprotein by the host signal peptidase and signal peptide peptidase (36, 43, 44), resulting in the maturation of capsid protein, two envelope proteins, and viroporin p7. The nonstructural protein 2 (NS2) protease cleaves its own carboxyl terminus, and then NS3 cleaves the appropriate downstream positions to produce NS3, NS4A, NS4B, NS5A, and NS5B (24, 60), which form the replication complex, together with several host proteins (14, 35).
NS5A is a membrane-anchored zinc-binding phosphoprotein that appears to possess diverse functions, including the suppression of host defense and the regulation of virus replication (1, 15, 58), but its biological function remains unclear. Several groups, including ours, have suggested that the molecular chaperone, heat shock protein 90 (Hsp90), and several cochaperones participate in the replication complex of HCV through interaction with NS5A or other NS proteins (45, 56, 65). Hsp90 is the highly conserved and ubiquitously expressed protein that acts as a key regulator for the turnover and the activities of more than 200 signaling proteins, including steroid receptors and cell-signaling kinases (66). The chaperone activity of Hsp90 contributes to the refolding of an unfolded protein in an ATP-dependent manner, and the execution of Hsp90-dependent protein folding requires the formation of a multichaperone complex containing other chaperones (e.g., Hsp70, Hsp104, and Hsp40) and cochaperones (e.g., p23, Hop, and immunophilins) (4, 18, 48). Geldanamycin or its derivatives, which are represented as specific inhibitors of Hsp90, can destabilize and then degrade client proteins (41, 55).
The host chaperone mechanism is involved in the folding of viral polymerase to support viral replication (6, 27). Moreover, host chaperones have been reported to play roles in the assembly of viral particles and the sorting of virus proteins (9, 32, 38). We have previously reported that Hsp90 chaperone activities and chaperone-associated proteins are required for the efficient propagation of HCV (45, 56) and that human butyrate-induced transcript 1 (hB-ind1) is involved in the propagation of HCV through interactions with NS5A and Hsp90 via the coiled-coil domain and the FXXW motif, respectively (56). hB-ind1 was first reported to be a multiple-membrane-spanning protein consisting of 362 amino acids that possesses a significant homology with a cochaperones, p23, that regulates Hsp90 function by its cochaperone activity (11). However, the roles of hB-ind1 in the life cycle of HCV have not been precisely clarified. In this study, we investigated the role of the Hsp90-related chaperone system, including hB-ind1, in the regulation of the RNA replication and particle production of HCV.
The plasmids encoding hB-ind1, NS5A, Hsp90, and FK506-binding protein 8 (FKBP8) were prepared by methods described previously (45, 56). The DNA fragments encoding hB-ind1 mutants were prepared by PCR with the introduction of a silent mutation that is resistant to the short hairpin RNA in the hB-ind1 knockdown cells, as described previously (56). The human p23 gene and glucose-regulated protein 78 (GRP78) promoter region (−151 to +22) were amplified by PCR from the total cDNA and genomic DNA of Huh7 cells, respectively. The DNA fragments encoding mutants of hB-ind1 and p23 were prepared by the method of splicing by overlap extension (26) and introduced into pEF FLAGGs pGKpuro (28). The GRP78 promoter region was introduced between the KpnI and HindIII sites of pGL3-basic (Promega, Madison, WI) and designated pGRP78-luc. The reporter plasmid carrying a firefly luciferase gene under the control of the GR promoter (pGR-luc) was purchased from Panomics (Fremont, CA). The internal-control plasmid encoding a Renilla luciferase (pRL-TK) was purchased from Promega. The plasmid pFK-I389 neo/NS3-3′/NK5.1 (47) was kindly provided by R. Bartenschlager. The plasmids used in this study were confirmed by sequencing them with an ABI Prism 3130 genetic analyzer (Applied Biosystems, Tokyo, Japan).
All cell lines were cultured at 37°C under a humidified atmosphere and 5% CO2. The human embryonic kidney 293T and hepatocellular carcinoma Huh7 cell lines were maintained in Dulbecco's modified Eagle's medium (DMEM) (Sigma, St. Louis, MO) supplemented with 100 U/ml penicillin, 100 μg/ml streptomycin, and 10% fetal calf serum (FCS). The human hepatocellular carcinoma cell line Huh7.5.1 was kindly provided by F. Chisari (70) and was maintained in DMEM containing nonessential amino acids, 100 U/ml penicillin, 100 μg/ml streptomycin, and 10% FCS. The Huh9-13 cell line, which is a Huh7 cell line harboring a subgenomic HCV RNA replicon (35), was maintained in DMEM containing 10% FCS, nonessential amino acids, and 1 mg/ml G418 (Nakalai Tesque, Kyoto, Japan). The hB-ind1 knockdown cell line Huh-KD and control cell line Huh-ctrl were described previously (56). Huh-KD cells were transfected with each of the expression plasmids encoding wild-type or mutant hB-ind1 and cultured for 1 week in the presence of 10 μg/ml of puromycin. The remaining cells were used for the experiments described below. The viral RNA of JFH1 was introduced into Huh7.5.1 cells according to the method of Wakita et al. (62) for preparation of the infectious HCV particles in cell culture.
The rabbit anti-hB-ind1 antibody was prepared as described previously (56). Mouse monoclonal antibodies to HCV NS5A, influenza virus hemagglutinin (HA) and FLAG tags, and β-actin were purchased from Austral Biologicals (San Ramon, CA), Covance (Richmond, CA), and Sigma, respectively. Mouse anti-protein disulfide isomerase (PDI) immunoglobulin G2a (IgG2a) was from Affinity Bioreagents (Golden, CO). Mouse anti-double-stranded RNA (dsRNA) IgG2a (J1 and K2) antibodies were from Biocenter Ltd. (Szirak, Hungary). Alexa Fluor 488 (AF488)-conjugated anti-mouse IgG1, AF647-conjugated anti-rabbit IgG, and AF594-conjugated anti-mouse IgG2a and IgG2b antibodies were from Invitrogen (San Diego, CA).
Transfection and immunoprecipitation analyses were carried out as described previously (25, 45). Immunoprecipitates boiled in loading buffer were subjected to 12.5% sodium dodecyl sulfate-polyacrylamide gel electrophoresis. The proteins were transferred to polyvinylidene difluoride membranes (Millipore, Bedford, MA) and were reacted with the appropriate antibodies. The immune complexes were visualized with Super Signal West Femto substrate (Pierce, Rockford, IL) and detected by an LAS-3000 image analyzer system (Fujifilm, Tokyo, Japan). The protein bands of GRP78 and β-actin were quantified by Multi Gauge software (Fujifilm), and the values of GRP78 expression were normalized with those of β-actin.
HCV RNA was estimated by the method described previously (56). Total RNA was prepared from cells by using an RNeasy minikit (Qiagen, Tokyo, Japan). First-strand cDNA was synthesized using an RNA LA PCR in vitro cloning kit (Takara Bio Inc., Shiga, Japan) and random primers. Each cDNA was estimated with Platinum SYBR green qPCR SuperMix UDG (Invitrogen) according to the manufacturer's protocol. Fluorescent signals were analyzed with an ABI Prism 7000 (Applied Biosystems). The internal ribosomal entry site regions of HCV and mRNAs of GAPDH (glyceraldehyde-3-phosphate dehydrogenase), GRP78, and growth arrest- and DNA damage-inducible gene 153 (GADD153) were amplified using the primer pairs 5′-GAGTGTCGTGCAGCCTCCA-3′ and 5′-CACTCGCAAGCACCCTATCA-3′, 5′-GAAGGTGAAGGTCGGAGTC-3′ and 5′-GAAGGTGAAGGTCGGAGTC-3′, 5′-CGCCAAGCGGCTCATC-3′ and 5′-AACCACCTTGAACGGCAAGA-3′, and 5′-AGCTGGAACCTGAGGAGAGA-3′ and 5′-TGGATCAGTCTGGAAAAGCA-3′, respectively. The values of the HCV genome or each mRNA were normalized with those of GAPDH mRNA. Each PCR product was detected as a single band of the correct size on agarose gel electrophoresis (data not shown).
In vitro transcription and RNA transfection. The plasmid pFK-I389 neo/NS3-3′/NK5.1 was linearized by treatment with ScaI and then transcribed in vitro using the MEGAscript T7 kit (Applied Biosystems) according to the manufacturer's protocol. The in vitro-transcribed RNA was electroporated into cells at 4 million cells/0.4 ml under conditions of 270 V and 960 μF using a Gene Pulser (Bio-Rad, Hercules, CA). The colony formation assay was carried out by a method described previously (45).
Cells cultured on glass slides were fixed with 4% paraformaldehyde in phosphate-buffered saline (PBS) at room temperature for 30 min. After being washed twice with PBS, the cells were permeabilized for 20 min at room temperature with PBS containing 0.25% saponin and blocked with PBS containing 0.2% gelatin (gelatin-PBS) for 60 min at room temperature. The cells were incubated with gelatin-PBS containing rabbit anti-hB-ind1 antibody, mouse anti-NS5A IgG1, mouse anti-PDI IgG2a, mouse anti-FKBP8 IgG2b, or mouse anti-dsRNA IgG2a (J1 and K2) at 37°C for 60 min; washed three times with PBS containing 1% Tween 20; and incubated with gelatin-PBS containing AF488-conjugated anti-mouse IgG1 or AF647-conjugated anti-rabbit or AF594-conjugated anti-mouse IgG2a or IgG2b antibodies at 37°C for 60 min. Finally, the cells were washed three times with PBS containing 1% Tween 20 and observed with a FluoView FV1000 laser scanning confocal microscope (Olympus, Tokyo, Japan).
Correlative fluorescence microscopy-electron microscopy (FM-EM) allows individual cells to be examined both in an overview with FM and in a detailed subcellular-structure view with EM (51). The endogenous hB-ind1 and NS5A were stained and observed in the HCV replicon cells by the correlative FM-EM method as described previously (45).
Each plasmid was transfected into Huh7, Huh9-13, and interferon (IFN)-cured cells seeded in a 12-well plate, and the cells were treated with 1 μM dexamethasone (Sigma) for 12 h or with 17-dimethylamino-ethylamino-17-demethoxygeldanamycin (DMAG) (Sigma) for 6 h at 36 h posttransfection and lysed in 200 μl of passive lysis buffer (Promega). Luciferase activity was measured in 20-μl aliquots of the cell lysates using a Dual-Luciferase Reporter Assay System (Promega). Firefly luciferase activity was standardized with that of Renilla luciferase cotransfected with the internal-control plasmid pRL-TK. The resulting values were expressed as the increase in relative light units (RLU).
Results were expressed as the mean ± standard deviation. The significance of differences in the means was determined by Student's t test.
Although we had previously reported that hB-ind1 regulates HCV RNA replication through interaction with NS5A and Hsp90, the molecular mechanisms underlying the regulation of HCV replication remained to be clarified. To gain more insights into the potential cochaperone activity of hB-ind1 in the Hsp90 chaperone system, we prepared expression plasmids encoding a wild-type p23 and three p23 mutants—one in which the FXXW motif was replaced with AXXA (p23AxxA), one in which the cochaperone domain of p23 was replaced with the p23-like domain of hB-ind1 (cp23), and one in which both substitutions were made (cp23AxxA) (Fig. (Fig.1A).1A). HA-tagged Hsp90 was coexpressed with FLAG-tagged p23 or the FLAG-tagged p23 mutants in 293T cells (Fig. (Fig.1B).1B). Hsp90 was coimmunoprecipitated with wild-type p23 and a cp23 mutant, but not with the p23AxxA or cp23AxxA mutants, indicating that the FXXW motif of hB-ind1, as is the case with that of p23 (67), is also involved in binding to Hsp90. Hsp90 participates in the folding and stabilization of the ligand-binding domain of the glucocorticoid receptor (GR), together with p23 and other cofactors (49). p23 was shown to act not only in the activation (30), but also in the inhibition, of GR signaling (67). To examine whether hB-ind1 has the ability to work as a cochaperone in an Hsp90-dependent manner, each of the plasmids encoding p23 or the p23 mutants was cotransfected with a reporter plasmid carrying a firefly luciferase gene under the control of the GR promoter (pGR-luc), together with an internal-control plasmid (pRL-TK), and GR-mediated transcriptional activity was determined at 12 h after treatment with dexamethasone, a ligand of GR. Expression of the p23 or cp23 mutant, but not of the AXXA mutants, significantly inhibited GR-mediated transcription (Fig. (Fig.1C).1C). These results indicate that the p23-like domain of hB-ind1 possesses cochaperone activity comparable to that of p23.
Previous reports have suggested that HCV NS5A interacts with several host proteins, including FBL2 (63), vesicle-associated membrane protein-associated protein subtype A (VAP-A) (61), VAP-B (25), FKBP8 (45), and hB-ind1 (56), and that these interactions participate in the replication of HCV. We have shown that hB-ind1 interacts with NS5A and Hsp90 through the coiled-coil domain and the FXXW motif in the p23-like domain, respectively, and that coexpression of FKBP8 enhances the interaction of Hsp90 with hB-ind1 (56). To determine the effect of the mutation in the p23-like domain of hB-ind1 on interaction with Hsp90, NS5A, and FKBP8, we prepared an expression plasmid encoding wild-type hB-ind1 and three hB-ind1 mutants, one in which the p23-like domain was replaced with the cochaperone domain of p23 (chB-ind1), one in which the FXXW motif was replaced with AXXA (hB-ind1AxxA), and one in which both replacements were made (chB-ind1AxxA) (Fig. (Fig.2A).2A). The FLAG-tagged wild-type or mutant hB-ind1 was coexpressed with HA-tagged Hsp90 (Fig. (Fig.2B,2B, left) or HA-tagged NS5A (Fig. (Fig.2B,2B, right) in 293T cells and immunoprecipitated with anti-FLAG antibody. Hsp90 was coprecipitated with wild-type hB-ind1 and the chB-ind1 mutant, but not with the hB-ind1AXXA and chB-ind1AXXA mutants (Fig. (Fig.2B,2B, left), confirming that the FXXW motif is crucial for the interaction with Hsp90. In contrast, NS5A was coprecipitated with each of the hB-ind1 proteins, suggesting that mutation in the p23-like domain of hB-ind1 has no effect on the binding of hB-ind1 to NS5A through the coiled-coil domain (Fig. (Fig.2B,2B, right). To determine the effect of FKBP8 expression on the interaction between hB-ind1 and Hsp90, FLAG-tagged wild-type hB-ind1 or the chB-ind1 mutant was coexpressed with HA-tagged FKBP8 and/or Hsp90 in 293T cells and immunoprecipitated with anti-FLAG antibody. The amounts of Hsp90 coprecipitated with hB-ind1 or chB-ind1 were increased by coexpression of FKBP8 (Fig. (Fig.2C).2C). To further examine the interaction of hB-ind1 with Hsp90 and NS5A at an endogenous expression level in Huh9-13 cells harboring an HCV subgenomic RNA replicon, lysates of the replicon cells were subjected to immunoprecipitation analysis. Endogenous Hsp90 and NS5A were specifically coimmunoprecipitated with endogenous hB-ind1 (Fig. (Fig.2D).2D). These results suggest that the p23-like domain of hB-ind1 is interchangeable with the cochaperone domain of p23 during complex formation with NS5A, Hsp90, and FKBP8.
The p23-like domain of hB-ind1 has been suggested to be required for HCV propagation (56). However, the involvement of the cochaperone activity of hB-ind1 in HCV propagation has not been examined. To assess the effect of cochaperone activity in the p23-like domain of hB-ind1 on the RNA replication and particle production of HCV, each of the expression plasmids encoding the FLAG-tagged wild-type or mutant hB-ind1 carrying the silent mutations resistant to small interfering RNA was transfected into hB-ind1 knockdown (Huh-KD) cells and cultured for a week in the presence of puromycin. The expressions of FLAG-tagged hB-ind1 and the mutants in the Huh-KD cells were comparable to that of the endogenous hB-ind1 in the control (Huh-ctrl) cells transfected with an empty vector (Fig. (Fig.3A)3A) . Subgenomic HCV replicon RNA transcribed from pFK-I389 neo/NS3-3′/NK5.1 was transfected into these cells and cultured for 4 weeks in the presence of G418. Although the number of colonies was reduced in the Huh-KD cells compared with the Huh-ctrl cells after transfection with an empty vector, as described previously (56), the colony numbers were recovered by the expression of the hB-ind1 or chB-ind1 mutant, but not by that of the hB-ind1AXXA or chB-ind1AXXA mutants (Fig. (Fig.3B).3B). Similarly, intracellular HCV RNA and infectious viral titers in the culture supernatants of Huh-KD cells infected with JFH1 virus were partially recovered by the expression of the hB-ind1 or chB-ind1 mutant, but not by that of the hB-ind1AXXA or chB-ind1AXXA mutant (Fig. (Fig.3C).3C). These results suggest that cochaperone activity in the p23-like domain of hB-ind1 is required for HCV propagation and that the cochaperone domain of p23 can substitute for the p23-like domain of hB-ind1.
Our previous report revealed the interplay among hB-ind1, Hsp90, FKBP8, and NS5A and showed that these interactions play an important role in HCV replication (56). However, the subcellular localization of the endogenous hB-ind1 in the replicon cells and JFH1 virus-infected cells has not been precisely assessed. To determine the subcellular localization of hB-ind1 in the context of HCV replication, the expression of hB-ind1 and NS5A in the replicon cells and JFH1 virus-infected cells was examined by immunofluorescence analyses (Fig. (Fig.4A).4A). Endogenous hB-ind1 was colocalized with the endoplasmic reticulum (ER)-marker PDI and NS5A as dot-like structures in the Huh9-13 replicon cells (Fig. (Fig.4A,4A, top) and in cells infected with JFH1 virus (Fig. (Fig.4A,4A, bottom), and these dot-like structures disappeared in concert with the loss of NS5A expression by treatment with IFN-α in the replicon cells and was not observed in the mock-infected Huh7.5.1 cells. Furthermore, FKBP8 (Fig. (Fig.4B,4B, top) and dsRNA (Fig. (Fig.4B,4B, bottom) were colocalized with hB-ind1 and NS5A in the dot-like structures in Huh9-13 replicon cells. These results indicate that HCV replicating RNA is localized with hB-ind1, FKBP8, and NS5A in the dot-like compartments. HCV RNA replication or expression of viral proteins leads to formation of the convoluted membranous structures designated the membranous web (14, 23). The large structures of the replication complexes in the replicon cells indicate membranous webs with restricted motility (68). To further analyze the subcellular compartments, including hB-ind1 and NS5A, the same field of the Huh9-13 replicon cells was observed under FM and EM by using the correlative FM-EM technique (Fig. (Fig.5A,5A, upper two rows). The large structures that included hB-ind1 and NS5A in the replicon cells were observed under FM and EM (white-boxed areas) and further magnified (black-boxed areas). Convoluted membranous structures that consisted of small vesicles and that were similar to the membranous web were observed. Another field of view yielded similar results (Fig. (Fig.5A,5A, lower two rows). The membranous web resembling the convoluted structures was not observed in the Huh9-13 cells depleted of viral RNA by IFN treatment (Fig. (Fig.5B).5B). Together, these results suggest that hB-ind1 interacts with NS5A on the membranous web in cells replicating HCV RNA.
Hsp90 regulates the folding and stability of proteins in all eukaryotes (59), and inhibition of the chaperone pathway suppresses correct protein folding, which leads to induction of proteasome-mediated degradation of the unfolded proteins and the unfolded protein response (UPR). Our previous (46) and present studies (Fig. (Fig.44 and and5)5) showed that several cochaperone components are recruited in the membranous web, suggesting that the Hsp90 chaperone system participates in the replication complex to circumvent the induction of the UPR and to maintain the folding of the host and viral proteins in a replication-competent state. To determine the induction of the UPR by HCV replication, Huh9-13 replicon cells were transfected with a reporter plasmid carrying a firefly luciferase gene under the control of the GRP78 promoter, which is activated by the induction of the UPR, together with an internal-control plasmid. Although the GRP78 promoter activity was slightly enhanced in the Huh9-13 cells compared to that in the parental cells, a fourfold increase of GRP78 promoter activity in the replicon cells was observed after treatment with an Hsp90 inhibitor, DMAG, in contrast to the twofold increase in similarly treated parental Huh7 cells, and the activation of the GRP78 promoter was canceled by treatment with IFN-α despite DMAG treatment (Fig. (Fig.6A),6A), suggesting that the Hsp90 chaperone system participates in the circumvention of the UPR induced by the replication of HCV RNA. In addition, activation of GRP78 at transcriptional and translational levels after treatment with DMAG was higher in the HCV replicon cells than in the parental cells or in cured cells, which were depleted of HCV RNA by treatment with IFN-α (Fig. (Fig.6B).6B). Furthermore, DMAG treatment enhanced the transcription of the UPR marker protein GADD153 at a higher level in the replicon cells than in the parental Huh7 or the cured cells (Fig. (Fig.6C).6C). These results suggest that the Hsp90-dependent chaperone system plays a crucial role in the folding of the host and viral proteins involved in HCV replication and in the regulation of UPR induction.
Studies of the relationship between Hsp90 and steroid receptors, such as GR, have revealed the activities of cochaperones (52, 67). Cochaperones, such as p23, appear to interact with and dissociate from Hsp90 and the client protein complex in a defined order. These cochaperones participate in the chaperone complex in a late step and promote the dissociation of the client proteins from Hsp90 to facilitate formation of the chaperone complex in the next chaperone cycle (16-18). In this study, we have shown that hB-ind1 participates in HCV replication and that the p23-like domain of hB-ind1 possesses cochaperone activity comparable to that of the cochaperone domain of p23, suggesting that hB-ind1 is involved in the recycling of the chaperone complex in the membranous web to maintain the function of the replication complex of HCV.
Previous studies have indicated that HCV proteins rearrange the ER membrane into the small convoluted membranous vesicles that are collectively known as the membranous web, and these vesicles have been suggested to be the intracellular compartments in which HCV replication takes place (14, 23, 68). In the living replicon cells, two forms of replication complexes, small and large vesicles, are detected, both of which include the viral replication complexes (68). Large vesicles, corresponding to membranous webs, exhibit restricted motility, while small vesicles show fast movement (68), and FM and EM have revealed that NS5A is colocalized with hB-ind1, as well as FKBP8 (45), in the membranous webs. hB-ind1 was first identified as a regulator of Rac1 that activates JNK and NF-κB (11). Rac1 is a member of the Rho GTPase family and plays crucial roles in cytoskeletal dynamics, membrane ruffling, and gene transcription through the effectors of the Rho GTPase family members. IQGAP1 and PAK1 are Rac1 effectors that bind to Rac proteins and are also involved in the replication of HCV (5, 7, 19, 31, 50). The tetratricopeptide repeat domain of immunophilin family members, such as FKBP8, has been shown to interact with Hsp90 (12, 45) and the GR-Hsp90 complex that leads to association with dynein for retrograde transport, along with microtubules (12). Hsp90 has been shown to play an important role in the interaction of transcriptase with genomic RNA of hepatitis B virus (27) and the nuclear transportation of the polymerase of influenza virus (40). Flock house virus also recruits Hsp90 in the polymerase synthesis in the early step of infection (9). Hsp90 may be involved in the regulation of the movement and arrangement of the HCV replication complexes through interaction with Rac1, hB-ind1, and FKBP8. Further investigation is needed to clarify the role of the Hsp90 chaperone system in the life cycle of HCV.
The surrounding membranes, including the membranous web, may protect the viral replication complex and RNA genome against digestion by the host proteases and nucleases (69). The replication complex is composed of viral nonstructural proteins and host proteins, including chaperone and cochaperone proteins. HCV NS5A has been shown to interact with various host proteins, including cochaperones, such as FKBP8 and hB-ind1, and to recruit a chaperone, Hsp90, into the replication complex through interaction with these cochaperones. Recruitment of the chaperone complex into the replication complex is crucial for the correct folding of newly synthesized viral proteins to maintain the efficient replication of the viral genome. HCV replication has been shown to be improved by the adaptive mutations suppressing the phosphorylation status of NS5A in the replicon cells (3). Although suppression of the hyperphosphorylation of NS5A by treatment with kinase inhibitors improves the replication of the replicons that have no adaptive mutations (42), several kinase inhibitors have been shown to suppress the replication of the HCV replicon carrying the adaptive mutations (29), and phosphorylation of NS5A by casein kinase II was shown to improve virus production but not HCV RNA replication (57). Hsp90 is capable of directly modulating the activities of several kinases (37, 53, 54), and thus, it might be feasible that cochaperones, including hB-ind1 and FKBP8, participate in the propagation of HCV by regulating the phosphorylation status of NS5A in cooperation with Hsp90.
The host chaperone system regulates the quality of client proteins, and impairment of the chaperone activity induces accumulation of misfolded proteins and affects the natural cellular function and viability (20, 21, 33). In this study, DMAG treatment induced a higher level of UPR in HCV replicon cells than in parental and cured cells, indicating that the Hsp90 chaperone system participates in the maintenance of correct folding of the viral and host proteins in the replication complex in the membranous web and in the circumvention of the UPR induced by HCV replication. Treatment with geldanamycin or its derivatives has been shown to inhibit GRP94, which is the Hsp90 paralog located in the ER (10), and to disrupt the ER chaperone pathway, leading to the induction of ER-associated protein degradation, transcriptional attenuation, and eventually induction of apoptosis (34). ER chaperones, such as GRP94, may also participate in the correct folding of the viral and host proteins in the replication complex for efficient replication of the HCV genome.
Geldanamycin and its derivatives have been reported to remarkably inhibit poliovirus replication in vivo without any emergence of drug-resistant escape mutants (22), suggesting that an inhibitor of the chaperone system may be a promising candidate for the treatment of viral infectious diseases with low risk of the emergence of drug-resistant viruses. In addition, Hsp90 inhibitors exhibit anticancer activities through the suppression of various cell signals essential for cancer growth and the enhancement of radiation sensitivity (2, 8, 13). In conclusion, our data indicate that hB-ind1 is included within the HCV replication complex and regulates HCV RNA replication through its own cochaperone activity. Hsp90 and cochaperones, including hB-ind1 and FKBP8, which are required for efficient HCV replication, should be ideal targets for the treatment of chronic hepatitis C with a low frequency of emergence of drug-resistant breakthrough viruses.
We thank H. Murase for her secretarial work. We also thank R. Bartenschlager, T. Wakita, and F. V. Chisari for providing the plasmids and cell lines.
This work was supported in part by grants-in-aid from the Ministry of Health, Labor, and Welfare; the Ministry of Education, Culture, Sports, Science, and Technology; the Global Center of Excellence Program; the Foundation for Biomedical Research and Innovation; and the Naito Foundation.
Published ahead of print on 5 August 2009.