Hepatitis C virus non-structural proteins modulate cellular kinases for increased cytoplasmic abundance of host factor HuR and facilitate viral replication

Host protein HuR translocation from nucleus to cytoplasm following infection is crucial for the life cycle of several RNA viruses including hepatitis C virus (HCV), a major causative agent of hepatocellular carcinoma. HuR assists the assembly of replication-complex on the viral-3′UTR, and its depletion hampers viral replication. Although cytoplasmic HuR is crucial for HCV replication, little is known about how the virus orchestrates the mobilization of HuR into the cytoplasm from the nucleus. We show that two viral proteins, NS3 and NS5A, act co-ordinately to alter the equilibrium of the nucleo-cytoplasmic movement of HuR. NS3 activates protein kinase C (PKC)-δ, which in-turn phosphorylates HuR on S318 residue, triggering its export to the cytoplasm. NS5A inactivates AMP-activated kinase (AMPK) resulting in diminished nuclear import of HuR through blockade of AMPK-mediated phosphorylation and acetylation of importin-α1. Cytoplasmic retention or entry of HuR can be reversed by an AMPK activator or a PKC-δ inhibitor. Our findings suggest that efforts should be made to develop inhibitors of PKC-δ and activators of AMPK, either separately or in combination, to inhibit HCV infection.

Introduction and ABL1 kinases is important for centrosomal accumulation of HuR and, therefore, genomic stability [31]. In addition to phosphorylation, HuR is methylated on R217 by CARM1, which correlates with its cytoplasmic localization in non-small cell lung carcinoma [32]. Ubiquitination on K313 and K326 leads to the dissociation of HuR from p21, MKP1, and SIRT1 mRNA [33]; whereas neddylation on K283, K313, and K326 promotes nuclear localization and protein stability [34].
Subcellular localization of HuR is also regulated by AMP-activated kinase (AMPK) [35]. Inhibition of AMPK has been associated with cytoplasmic localization of HuR, leading to increased stability of the HuR-bound transcripts p21, cyclin B1, cyclin A, and IL-20. AMPK phosphorylates importin-α1 on S105 and acetylates it on K22, thereby preventing its interaction with HuR and import to the nucleus [36].
Cells adopt multiple strategies to regulate the subcellular localization of HuR and, consequently, the stability of its cognate transcripts under different conditions. Here, we investigated the viral mechanism driving the nuclear export of HuR, the cytoplasmic partitioning of HuR, and the effect of its altered localization. We report that PKC-δ and AMPK work in a coordinated manner to achieve cytoplasmic localization of HuR via direct modification and indirect regulation of its nuclear carrier. Thus, HCV modulates two separate kinases to promote replication and pathogenesis. Relocalization of HuR and its interaction with cytoplasmic RNAs could reveal the molecular basis of HCV-induced hepatocellular carcinoma and provide an attractive target for preventing its progression.

Phosphorylation of HuR regulates its cellular localization upon HCV infection
It has been shown that HuR relocalizes from nucleus to cytoplasm after 48h of HCV infection and is important for HCV replication [9,13]. These studies have been done in Huh7.5 and Huh7.5.1 cells. Therefore, we first studied the time kinetics of this relocalization using nuclearcytoplasmic fraction of mock and HCV-JFH1 virus infected Huh7.5 cells at 12h and 24h post infection (S1A and S1B Fig). We calculated the ratio of nuclear to cytoplasmic HuR intensities and found that the increase in cytoplasmic HuR begins from 12h post infection and increases till 24h. These time points correlate with the replication phase of the virus wherein translation to replication switch occurs 12-18h post infection and highlight the importance of HuR relocalization for HCV replication. Viruses are known to modify host proteins [37]. To study the post-translational modifications of host proteins after HCV infection, total protein extracts prepared from mock (control) and HCV-RNA transfected Huh7.5 cells were analyzed by twodimensional gel electrophoresis and HuR-specific spots were detected using anti-HuR antibody. In control cells, two spots were observed for HuR, one at the acidic end towards pH 3 and the other at the basic end towards pH 10. The spots had the same molecular weight, suggesting the presence of two sub-populations of HuR with separate charged modifications that did not alter overall protein weight. An additional spot, slightly more acidic than the one at pH 10 (marked with an asterisk in Fig 1A), was observed upon HCV RNA transfection (Fig 1A  and 1B), indicating a negatively charged modification on HuR.
Phosphorylation is a post-translational modification that imparts a negative charge on proteins. Therefore, we examined the involvement of phosphorylation on HuR following HCV infection. Some phosphorylation sites known to impact HuR subcellular localization include T118, Y200, S221, S242, and S318. The phospho-dead mutants of these sites were generated using site-directed mutagenesis, wherein the serine or threonine residues were replaced by alanine (T118A, S221A, S242A, S318A), and tyrosine was replaced by phenylalanine (Y200F). Mock and transfected cells were harvested after 48h and cell lysates were precipitated using TCA precipitation. The precipitated proteins were run in 2 dimensions as described in the methods, followed by Western blotting of the gel. Western blotting was done with anti-HuR and anti-Actin antibodies. Appropriate HRP-conjugate secondary antibodies were used. (B) HCV infection was detected by western blotting using Anti-HCV core antibody for same lysates run on 12% SDS-PAGE. (C) Schematic for experimental protocol followed. (D) Huh 7.5 cells were transfected with Myc-tagged overexpression construct of WT HuR, S318A HuR mutant, S242A HuR mutant, Y200F HuR mutant, T118A HuR mutant and S221A HuR mutant and mock transfected. Cells were processed for immunofluorescence staining using Alexa Fluor conjugated secondary antibody against Myc (Green) and viral protein Core (Red). The nucleus was stained with DAPI (Blue). Scale bar represents 10 μm. (E) Nuclear and cytoplasmic ratio for overexpressed HuR (Myc) was quantified for images in (D) using Zen 2.3 lite software. (F) Images of HCV transfected cells following the schematic in (C). (G) Nuclear and cytoplasmic ratio for overexpressed HuR (Myc) was quantified for images in (F) using Zen 2.3 lite software. n = 10. Student t-test was performed for statistical analysis. * = p<0.05, ** = p<0.01, *** = p<0.001. https://doi.org/10.1371/journal.ppat.1011552.g001

HCV-proteins regulate cellular kinases
The mutants were verified by sequencing and employed in the relocalization assay. Endogenous HuR translocates from the nucleus to the cytoplasm 48h after HCV-RNA transfection [9]. If phosphorylation of a particular residue is crucial for guiding this translocation, the corresponding phospho-dead mutation would prevent the cytoplasmic relocalization of the mutant protein upon viral infection. Wild-type (WT) HuR overexpression was used as a positive control for HuR localization. Using immunostaining, the subcellular localization of WT and mutant HuR proteins was detected in uninfected control and HCV-infected cells. Myctagged WT and mutant HuR was overexpressed in Huh7.5 cells, followed by HCV-RNA transfection ( Fig 1C). An anti-Myc-tag antibody was used to visualize the overexpressed protein, the HCV Core protein was used to identify HCV-infected cells, and DAPI was used to mark the nuclei. In untransfected control cells, WT and all mutant HuR proteins localized to the nucleus (Fig 1D and 1E). HCV-RNA transfection led to the relocalization of WT HuR to the cytoplasm. The same was observed for HuR mutants S221A and S242A. In contrast, mutants S318A, Y200F, and T118A remained localized to the nucleus even after 48h of HCV-RNA transfection (Fig 1F and 1G). This finding suggests that phosphorylation of S318, Y200, and T118 might be involved in relocalization of HuR to the cytoplasm upon HCV infection. The relocalization of WT and S318A HuR was also analysed by nuclear-cytoplasmic fractionation, wherein WT/S318A HuR expressing Huh7.5 cells were either Mock transfected or transfected with HCV-RNA and localization of HuR examined by western blotting using anti-HuR antibody (S1C and S1D Fig). We observed that the ratio of Nuclear to cytoplasmic intensity of overexpressed WT HuR decreased, but that of S318A HuR slightly increased upon HCV-RNA transfection, suggesting and corroborating the imaging results of nuclear retention of S318A HuR upon HCV RNA transfection. The effect on relocalization of WT and mutant HuR was marginal because the percentage of cells co-transfected with the HuR overexpression construct and HCV-RNA is very low and therefore, immunofluorescence staining serves as a better technique for monitoring the relocalization. Also, in the HCV infected cells, there could be other signalling in addition to 318 site phosphorylation which might influence HuR relocalization.

HuR is differentially phosphorylated upon HCV infection
To validate HCV induced HuR phosphorylation, total phosphorylated proteins were immunoprecipitated using pan-anti-phospho-Ser/Thr and pan-anti-phospho-Tyr antibodies. The presence of HuR in the pull-down fraction of both antibodies confirmed HuR phosphorylation in cells (Fig 2A and 2B). HCV core exhibited slight phosphorylation in both Ser/Thr and Tyr pull down. HCV Core protein has Ser, Thr and Tyr motifs which are known and predicted to be phosphorylated. One of the Tyrosine motifs is conserved in all the HCV genotypes and is also reported to be essential for HCV assembly (Tyr 136) [38]. There is also tyrosine phosphorylation site prediction in HCV Core (Tyr 86) [39]. To identify the phosphorylation sites on HuR, the latter was immunoprecipitated from control or HCV-RNA-transfected cells, and the pulldown fraction was subjected to mass-spectrometric analysis (Fig 2C and 2D). The results revealed six previously characterized sites, including T80, S88, T143, S146, S202, and S318, as well as two new sites, S135 and T293 (Fig 2E). The abundance of phosphorylated peptides at a particular site was calculated as the fraction of phosphorylated peptides to total peptides observed for that site. Quantification was performed for S318, which was previously found to be regulated by HCV infection, as well as for S202, which is known to regulate HuR localization, and S135, which has not yet been characterized. The percentage of phosphorylated S318 increased in all three sets of HCV-RNA transfected cells compared to control cells ( Fig 2F). S202 phosphorylation varied among separate sets of experiments ( Fig 2G); whereas S135 phosphorylation was reduced upon HCV-RNA transfection in all three sets (Fig 2H). Differential phosphorylation at distinct HuR sites might explain the absence of an overall variation (Fig 2A  and 2B). Because S318 phosphorylation was higher following HCV RNA-transfection and was found to be important for HuR relocalization (Fig 1F and 1G), this site was selected for further studies.

Phosphorylation of S318 influences HCV replication
The location of the S318 site on HuR was visualized using Pymol software. The X-ray structure of HuR RRM3 (PDB ID: 6GD2) complexed with RNA indicated that the S318 site was on the surface of the protein and interacted directly with the bound RNA ( Fig 3A). Accordingly, phosphorylation of S318 might affect the RNA binding of HuR. To study the impact of phosphorylation on the affinity for HCV-RNA, we used surface plasmon resonance. Biotinylated HCV 3'-UTR RNA was immobilized on a streptavidin-coated SPR chip (SA Chip) and recombinant purified WT, S318A, and S318D HuR proteins were passed over as analytes to calculate the binding affinity (Fig 3B-3D). The Kd values for WT and S318A HuR were 25.7 nM and 46.2 nM, respectively; whereas the phospho-mimic mutant S318D showed approximately fivefold better affinity (Kd 6.36 nM) (Fig 3E). This result indicates that, along with relocalization to the cytoplasm, S318 phosphorylation of HuR also augments the binding affinity for HCV 3'-UTR. To assess the impact of S318 phosphorylation on HCV replication, WT, S318A, and S318D constructs were overexpressed in Huh7.5 cells, followed by HCV-RNA transfection ( Fig 3F). The effect on replication was checked by analyzing the levels of HCV negative-strand RNA compared to the vector-overexpression control. Viral replication was increased upon overexpression of WT HuR but not S318A HuR, which did not relocalize to the cytoplasm and was rescued by overexpression of S318D HuR (Fig 3G and 3H). We analysed the S318D distribution in untreated cells and observed increased but not complete cytoplasmic localization as compared to WT HuR (S2 Fig). We further aimed to overexpress the WT and mutant HuR in the background of endogenous HuR silencing. For this, we generated a mutation in the overexpression constructs at the site of siHuR seed sequence binding (S3A Fig). The siRNA resistant overexpression constructs were transfected in the background of siHuR transfection. This was followed by HCV-JFH1 RNA transfection. 48h post transfection, the levels of HCV negative strand RNA were determined to assess HCV replication. The treatment of cells with siHuR reduced the HCV replication (S3B Fig). In this background, the overexpression of WT HuR increased the viral replication, S318A overexpression was unable to induce this increase while S318D overexpression exhibited a rescue in the RNA levels (S3C and S3D Fig). These findings confirm the importance of S318 phosphorylation in HuR localization and hence, HCV replication.

PKC-δ activity is required for HCV-mediated cytoplasmic export of HuR
Once the involvement of S318 phosphorylation in guiding HuR localization was ascertained, we determined the role of the kinase responsible for this modification. An earlier study suggested that PKC-δ participated in the phosphorylation of S318 on HuR and its nucleocytoplasmic shuttling [17,26]. Therefore, the involvement of PKC-δ in regulating HuR localization upon HCV infection was investigated. The levels of total and phosphorylated PKC-δ at different time points after HCV RNA transfection were analyzed. An increase in p-PKC-δ/total PKC-δ after 48h of HCV-RNA transfection was observed, suggesting the activation of PKC-δ upon HCV RNA transfection ( Fig 4A). This effect was not observed upon the transfection of pSGR-JFH1/Luc-GND RNA which is a replication defective sub-genomic HCV-JFH1 RNA (S4A Fig), suggesting the effect of replicative viral RNA in inducing PKC-δ phosphorylation. We also observed an increase in cleavage of PKC-δ upon HCV RNA transfection (S4B Fig). This cleavage product retains the catalytic subunit of PKC-δ, while freeing it from the regulatory subunit and is known to be transported to the nucleus [40]. This cleaved subunit might initiate the cascade to HuR phosphorylation for its cytoplasmic export. To visualize the interaction between HuR and PKC-δ in cells, immunoprecipitation of HuR was performed in mock (control) and HCV-virus infected cells, and the presence of PKC-δ in the pull-down

PLOS PATHOGENS
HCV-proteins regulate cellular kinases fraction confirmed their physical interaction ( Fig 4B). The increase in HuR associated with PKC-δ indicated a stronger interaction in cells upon HCV infection. The association was further strengthened by reverse pull-down, wherein immunoprecipitation of PKC-δ was performed and increased HuR association was observed upon HCV infection ( Fig 4C). To correlate this finding with HuR localization, an inhibitor of PKC-δ (rottlerin) was used. The

PLOS PATHOGENS
HCV-proteins regulate cellular kinases effect of PKC-δ inhibition on HCV-induced HuR relocalization was assessed. Huh7.5 cells were transfected with HCV-JFH1 RNA and 3 μM rottlerin was added to the medium after 6h of transfection (Fig 4D). The cells were incubated with the inhibitor for 48h, after which HuR localization was visualized by confocal microscopy. In the absence of the inhibitor, HuR relocated from the nucleus to the cytoplasm upon HCV RNA transfection (Fig 4E and 4F). The presence of the inhibitor did not alter the localization of HuR in control cells, but it prevented its relocalization to the cytoplasm upon HCV RNA transfection. These results confirmed the involvement of PKC-δ in altering the localization of HuR upon HCV infection.

PKC-δ regulates the HCV life cycle through HuR
Given that HuR is involved in viral RNA replication, we investigated the effect of rottlerin on HCV-RNA levels. Huh7.5 cells were transfected with HCV-JFH1 RNA and 3 μM rottlerin was added to the medium for 24h (Fig 5A). After 24h, the abundance of HCV-negative strand RNA in the cells was quantified by real-time PCR, which showed that addition of rottlerin decreased HCV-RNA levels by > 90% (Fig 5B). Similarly, confocal microscopy indicated that the percentage of HCV positive cells dropped from~20% in untreated cells to <2% in rottlerin-treated cells (Fig 5C and 5D). In all assays, the PKA inhibitor KT5720 was used as a negative control, which did not show any effect on HuR localization upon HCV infection. However, we observed an increase in replication upon PKA inhibitor treatment which could be because of the regulation of localization of another negative regulator of HCV replication, PTB [41]. The effect on viral replication was also assayed using siRNA mediated knockdown of PKC-δ. Huh7.5 cells were transfected with HCV-JFH1 RNA after 24h of siRNA transfection. 48h post HCV RNA transfection, HCV-negative strand RNA was quantified. We observed a dose-dependent decrease in HCV-negative strand RNA with increasing concentration of siRNA targeting PKC-δ (Fig 5E and 5F).
The expression of PKC-δ in healthy and hepatocellular carcinoma patients (Virus induced HCC patient positive for either HCV, HBV or both) in The Cancer Genome Atlas (TCGA) database was analyzed. Increased PKC-δ expression in virus-induced hepatocellular carcinoma patients suggested its involvement in disease progression (Fig 5G), and survival probability was inversely proportional to PKC-δ expression levels ( Fig 5H). This result suggested that patients with elevated PKC-δ would have more and early relocalization of HuR upon HCV infection, which would lead to enhanced viral replication and, hence, increased likelihood of severe disease progression.

HCV non-structural proteins increase the cytoplasmic abundance of HuR
Activation of PKC-δ and relocalization of HuR to cytoplasm seemed to be a viral strategy for its efficient replication. Therefore, candidate viral protein(s) involved in PKC-δ activation were investigated. Over-expression of either the structural protein, Core or all the non-structural proteins together through pSGR-Luc construct was performed in Huh7.5 cells and localization of HuR was assessed after 48h of overexpression. Viral non-structural proteins were found to be sufficient to cause the relocalization of HuR (Fig 6A and 6B). Among non-structural proteins, NS3 is a major pathogenic protein that interacts with multiple host proteins and kinases [42][43][44][45]. To assess the physical interaction between NS3 and PKC-δ, interaction studies for NS3 and PKC-δ were performed in Huh7.5 cells. Myc-tagged NS3 overexpression construct was transfected in Huh7.5 cells and 48h post transfection, PKC-δ was pulled down from the cell lysate to assess co-immunoprecipitation of myc-tagged NS3. PKC-δ pull-down could immunoprecipitate NS3 from the cell lysate, establishing their physical association (Fig 6C). This interaction was further confirmed by the colocalization of NS3 and PKC-δ in HCV-JFH1

PLOS PATHOGENS
HCV-proteins regulate cellular kinases

PLOS PATHOGENS
HCV-proteins regulate cellular kinases RNA transfected cells with a colocalization coefficient of 0.898 (Fig 6D). The colocalization assay was performed using HCV infection as well. The staining for HCV-NS3 and HCV-Core was performed to examine their interaction with PKC-δ (Fig 6E). This yielded a colocalization coefficient of 0.184 for Core and PKC-δ colocalization, and a colocalization coefficient of 0.85 for NS3 and PKC-δ colocalization, suggesting the interaction between PKC-δ and NS3 and not Core in HCV infected cells. These assays provide the evidence of direct interaction of HCV-NS3 protein with PKC-δ, which could activate it for HuR phosphorylation at S318 and hence its cytoplasmic localization upon HCV infection.

AMPK is involved in cytoplasmic retention of HuR upon HCV infection
Viral infection leads to HuR relocalization to the cytoplasm; however, at the same time, it is essential for the virus that HuR is prevented from reentry in the nucleus. The nuclear localization signal for HuR is not very well characterized. AMPK has been shown to phosphorylate and acetylate importin-α1, which carries newly synthesized HuR to the nucleus. Here, we assessed whether this mechanism was disrupted during HCV infection, thus allowing HuR to be retained in the cytoplasm. Huh7.5 cells were treated with an AMPK inhibitor (Compound C) or an AMPK activator (A769662) following HCV-RNA transfection. HuR localization was assessed by confocal microscopy 48h after drug treatment. Inhibition of AMPK activity did not alter HuR localization, whereas activation of AMPK led to increase in nuclear and decrease in cytoplasmic HuR and prevented complete cytoplasmic relocalization of HuR, even upon HCV RNA transfection (Fig 7A and 7B). NS5A overexpression alone could not cause HuR relocalization to cytoplasm (S5 Fig), suggesting that the trigger of NS3 mediated S318 phosphorylation could be required for initial HuR transport to the cytoplasm.
To correlate AMPK activity with hepatocellular carcinoma progression, TCGA database AMPK levels were analyzed. A decrease in total AMPKα1 levels was observed upon virusinduced tumor progression (Fig 7C). No significant correlation was observed between survival probability and AMPK-α1 levels ( Fig 7D). The viral non-structural protein NS5A inhibits AMPK activity [46,47], and we showed here that AMPK activation could prevent complete cytoplasmic relocalization of HuR upon HCV infection.
Our results uncovered the participation of multiple viral proteins in achieving increased cytoplasmic abundance of HuR. We describe the underlying mechanism wherein viral RNA translates upon entry into the cell, producing viral proteins. Viral NS3 protein activates PKCδ, which phosphorylates HuR on S318, leading to its relocalization to the cytoplasm, where it is available for the replication of cytoplasmic viral RNA. At the same time, viral NS5A protein inhibits AMPK, which retains relocalized and newly synthesized HuR in the cytoplasm (Fig 8). The proposed mechanism depicts a coordinated effort to regulate a host RNA binding protein, so it can be exploited to guide the viral life cycle.

Discussion
HCV is a hepatotropic virus that can cause chronic infection, leading to hepatocellular carcinoma. Host factors play an essential role in regulating the virus life cycle and cellular pathogenesis. HCV is a positive-strand RNA virus, whose RNA is replicated and translated in the PKC-δ antibody. HCV-Core protein served as the marker for HCV-RNA transfection. β-actin was used as loading control. (G) Expression level of PKC-δ (PRKCD gene) in healthy tissue and Virus induced primary HCC tissue samples from TCGA database. (H) Kaplan Meier plot for survival probability for Virus induced primary HCC patients with varying level of expression of PKC-δ. Data taken from TCGA database. Student t-test was performed for statistical analysis. * = p<0.05, ** = p<0.01, *** = p<0.001. https://doi.org/10.1371/journal.ppat.1011552.g005

PLOS PATHOGENS
HCV-proteins regulate cellular kinases Immunofluorescence staining was carried out at 48h post transfection using Alexa Fluor conjugated secondary antibodies against NS3 (Green) and PKC-δ (Red). The nucleus was counterstained with DAPI. Scale bar represents 10 μm. An enlarged image of infected cell and the line profile for PKC-δ and NS3 intensity over the indicated arrow in

PLOS PATHOGENS
HCV-proteins regulate cellular kinases cytoplasm. RNA binding proteins (RBPs) such as HuR bind to viral RNA and regulate these processes.
HCV is a completely cytoplasmic virus with no nuclear history of viral RNA. Therefore, it mediates the relocalization of RBPs to the cytoplasm to enable RNA binding. We show that phosphorylation of the S318 residue on HuR is essential for this relocalization. This site lies in RRM3 of HuR, which is critical for its binding to HCV-RNA. Indeed, we demonstrate that phosphorylation of S318 promotes the binding of HuR to HCV-RNA. Altered S318 phosphorylation levels have been detected in various pathologies such as colon cancer [27], suggesting that the same could happen in hepatocellular carcinoma. We report that S318 phosphorylation the enlarged image is depicted in the inset. Colocalization coefficient was calculated using Zeiss software. (E) Huh7.5 cells were infected with HCV-JFH1 virus. Immunofluorescence staining was carried out at 48h post transfection using Alexa Fluor conjugated secondary antibodies against NS3, Core (Green) and PKC-δ (Red). The nucleus was counterstained with DAPI. Scale bar represents 10 μm. Colocalization coefficient was calculated using Zeiss software. https://doi.org/10.1371/journal.ppat.1011552.g006

PLOS PATHOGENS
HCV-proteins regulate cellular kinases is mediated by PKC-δ. Inhibition of PKC-δ by lenvatinib has recently been associated with decreased tumorigenicity in Huh7 and Hep3B hepatocellular carcinoma cells [48]. One of the underlying mechanisms behind the pro-proliferative effect of PKC-δ could be its regulation of HuR localization in cells, which directly influences the cellular transcripts involved in tumorigenicity. These reports and the evidence presented herein point to the key role of S318 phosphorylation in the transport of HuR to the cytoplasm and in HCV infection.
HuR is not known to be mutated in many cancers. However, its localization and binding to different RNA transcripts affect cancer progression [49]. Reduced AMPK activity, which leads to cytoplasmic localization of HuR, is known to stabilize oncogenic molecules that promote progression of the proliferative phenotype. Here, we show that this mechanism is activated also by HCV, pointing to the significant role of HuR cytoplasmic localization in the progression of HCV-mediated hepatocellular carcinoma. HuR targets include invasive and metastatic RNAs (e.g., MMP-9, MTA1, and uPA), angiogenic RNAs (e.g., VEGF-1 and HIF-1), and cell proliferative RNAs (e.g., EGF, cyclin A, cyclin B1, cyclin E, and cyclin D1), through which it can influence hepatocellular carcinoma progression [50,51]. By stabilizing these target RNAs and increasing their protein products, HuR promotes the transformation from non-cancerous to cancerous cells. Hence, HuR has been proposed as a potential target for cancer treatment [52].
AMPK regulates the availability of HuR to the above RNAs in the cytoplasm. AMPK levels and activity are reduced in patients with hepatocellular carcinoma, and low p-AMPK levels

PLOS PATHOGENS
HCV-proteins regulate cellular kinases correlate with higher tumor occurrence [53,54]. Regulation of pathogenic RNAs by cytoplasmic HuR may be one of the mechanisms underlying this correlation. AMPK is a doubleedged sword, which influences cell proliferation in either way. A recent report has shown that the effect of AMPK on cell proliferation is a function of nutrient abundance. In cancer cells subjected to chronic nutrient deprivation, AMPK inactivation limits cell proliferation [55]. This limitation could be explained by direct activation of mTORC1 and mTORC2. Another possibility is the cytoplasmic retention of HuR, which is known to stabilize and enhance the translation of rictor, thereby activating mTORC2 [56]. This could be one of the mechanisms of regulation of viral replication in cells and hepatocellular carcinoma progression.
PKC-δ and AMPK are important cellular kinases that direct cell fate under various conditions. HCV exploits these kinases to regulate replication and pathogenesis through HuR. Preventing HCV induced cytoplasmic relocalization of HuR by either targeted PKC-δ inhibition or AMPK activation or by blocking the interaction of non-structural proteins with these kinases could offer novel therapeutic intervention against hepatocellular carcinoma progression.

Plasmids and constructs
pJFH1: Plasmid encoding all HCV proteins. The plasmid was linearized using XbaI to synthesise JFH1 RNA (Obtained from Apath,LLC).
pSGR-JFH1/Luc and pSGR-JFH1/Luc-GND: Bicistronic plasmid where the luciferase expression is driven by HCV-IRES while HCV non-structural proteins are expressed by EMCV IRES. GDD to GND mutation in the viral polymerase inhibits its activity and renders the RNA replication deficient. The plasmid was linearised by XbaI to synthesise SGR-JFH1/Luc RNA (A kind gift from Dr. Ralf Bartenschlager, Heidelberg University).
HuR mutants: HuR mutants were generated in the pcDNA3.1-myc-HuR background by site directed mutagenesis and the mutations were confirmed by sequencing.

Cell lines and transfections
Huh7.5 cells were maintained in Dulbecco's Modified Eagle Medium DMEM (Sigma) with 10% Fetal Bovine Serum. Lipofectamine 2000 (Invitrogen) was used for JFH1 RNA transfections and TurboFect (Thermo Scientific) was used for DNA transfections according to manufacturer's protocol. For generation of the infectious virus, HCV-JFH1 RNA was in vitro transcribed and transfected into Huh7.5 cells. Supernatant from the transfected cells was concentrated and used for viral infection. Uninfected Huh7.5 cells were used as a mock control.

HCV virus preparation
Huh7.5 cells electroporated with HCV-JFH1 RNA. 72-96h post electroporation, cell supernatant was collected and viral RNA copy number in supernatant determined using standard graph of known number of RNA molecules. Thereafter MOI of infection was determined after calculating specific infectivity of the virus. All virus experiments were done at an MOI of 0.01.

Nuclear-cytoplasmic fractionation
Nuclear and cytoplasmic extracts were prepared using the SIGMA CelLytic NuCLEAR Extraction Kit as per manufacturer's protocol. Briefly, cells were lysed in hypotonic lysis buffer and treated with 1% IGEPAL as per the manufacturer's recommendations (NXTRACT, SIGMA) to obtain the cytoplasmic extract. The nuclear pellet was then washed thoroughly with lysis buffer and lysed in the extraction buffer. The protein concentration of each extract was determined by Bradford assay. Equal amounts of protein were resolved on a SDS-10% PAGE followed by western blot using the desired antibodies.

2D-gel electrophoresis
The protein lysates for Mock and HCV transfected samples were precipitated using 10% TCA and pellets were washed with acetone. The protein pellets were resuspended in sample rehydration buffer (

Immunofluorescence staining
For immunofluorescence staining,~0.2*10 6 Huh7.5 cells were seeded on coverslips in a 12-well plate for 14h followed by transfection of HCV-JFH1 RNA (1μg) or HuR overexpression constructs as per the experiment protocol described. After desired time of infection/ RNA transfection, cells were washed twice with 1X PBS and fixed using 4% formaldehyde in PBS for 20 min at room temperature. After permeabilization by 0.1% Triton X-100 for 2 min at room temperature, cells were incubated with 3% BSA at room temperature for 1h for blocking, followed by incubation with the indicated antibody for 2h at 4˚C and then detected by Alexa-633-conjugated anti-mouse or Alexa-488 conjugated anti-rabbit secondary antibody for 30 min (Invitrogen). DAPI stain was used for visualisation of nuclei. Images were taken using Zeiss microscope and image analysis was done using the Zeiss LSM or ZEN software tools.

PLOS PATHOGENS
HCV-proteins regulate cellular kinases anti-mouse or anti-rabbit IgG; Sigma). Mouse-monoclonal anti-β-actin-peroxidase antibody (A3854, Sigma) was used as a control for equal loading of total cell extracts. Antibody complexes were detected using the ImmobilonTM Western systems (Millipore).
NS3 antibody used for immunostaining and co-immunoprecipitation was a kind gift from Prof. Guangxiang (George) Luo, University of Alabama.

Immunoprecipitation
Huh7.5 cells were scraped in polysome lysis buffer (100 mM KCl, 5 mM MgCl 2 , 10 mM HEPES pH 7.0, 0.5% NP-40, 1 mM DTT, 100 U/ml RNasin) and kept on cyclomixer for 1h, followed by spinning at 16000×g for 15 min at 4˚C. The supernatant obtained was precleared with Protein G Sepharose beads for 1h at 4˚C. The samples were spun at 1000×g for 2 min to pellet the beads, and the supernatant was removed. Protein G Sepharose beads were incubated with 1 μg of respective antibody overnight at 4˚C in a total volume of 200 μl of polysome lysis buffer and added to the precleared lysates followed by incubation for 3h with continuous mixing on a rotator device at 4˚C. The beads were washed three times with polysome lysis buffer. SDS sample buffer was then added to the beads and boiled to release the immunoprecipitated protein, and the supernatant was electrophoresed on a SDS-12% PAGE.

Surface plasmon resonance
Surface plasmon resonance spectroscopy was performed using a BIAcore3000 optical biosensor (GE Healthcare Lifescience) to study the binding kinetics of HuR with HCV 3'UTR RNA. Biotin-labelled HCV 3'UTR RNA was immobilized on streptavidin-coated sensor chips (GE Healthcare Lifescience) to a final concentration of 300 Resonance Units (RU)/flow cell. RNAprotein interactions were carried out in a continuous flow of Tris buffer (25 mM Tris (pH 7.5), 100 mM KCl, 7 mM β-mercaptoethanol, and 10% glycerol) at 25˚C at a flow rate of 10 μl/min. Increasing concentrations of HuR protein loaded on the biosensor chip for 100 s (characterized as the association phase), followed by a dissociation phase of 300 s with buffer alone. For normalising background non-specific interaction, a blank surface without any RNA was used for simultaneous injections of the sample during the experiment. BIAevaluation software (version 3.0) was used to determine the on rate, kon (M -1 s -1 ), and off rate, koff (s -1 ), using a 1:1 Langmuir binding model. The binding affinity, Kd was determined using the following equation: Kd = koff/kon.

Immunoprecipitation of HuR followed by Mass spectrometry
Huh7.5 cells were transfected with HCV-JFH1 RNA and harvested after 48h of transfection. Cell lysates were prepared in IP lysis buffer (Pierce, Thermo Scientific) adding 1× halt-protease and phosphatase inhibitor cocktail (Thermo Fisher Scientific, USA) with moderate vortexing followed by mild sonication. The lysate was centrifuged at 14,000 × g for 45 min at 4˚C to obtain the supernatant. Total protein was quantified by the BCA method (Thermo Fisher Scientific, USA). An equivalent amount (4 mg) of proteins from each condition was incubated with 3μg of anti-HuR antibody in end to end rotation for overnight at 4˚C. The immunocomplex was captured by using protein G Sepharose 4 fast flow beads (17-0618-01/ GE Healthcare). The immunoprecipitated complex was washed with IP lysis buffer twice and mili-Q once. Bound protein complexes were eluted in 50 μl SDS-PAGE 2X Laemmli sample buffer. The samples were separated using SDS-12% PAGE and stained with fresh coomassie stain (0.1% w/v) and then destained. Further in-gel mass spectrometry analysis was performed.
Gel slices corresponding to each lane were dissected in separate micro centrifuge tube and destained with wash buffer (50 mM ammonium bicarbonate (ABC) and 50% ACN). Before

PLOS PATHOGENS
HCV-proteins regulate cellular kinases digestion gel slices were reduced with 10 mM DTT (56˚C, 30 min), alkylated using 50 mM IAA (30 min, RT, dark), and dehydrated using 100% ACN. Finally, gel pieces were digested with MS grade trypsin (Pierce™ Trypsin Protease MS grade, cat no. 1862743, USA) and incubated at 37˚C for 18 h. Digested peptides were extracted by progressively adding 60% ACN with 0.1% FA for two times, followed by vortexing and ultra-sonication. The digested peptides were pooled, vacuum dried, and reconstituted in 40 μl of solvent A (2% (v/v) ACN, 0.1% (v/v) FA in water) and subjected to LC−MS/MS experiments using Sciex 5600 + Triple-TOF mass spectrometer coupled with ChromXP reversed-phase 3 μm C18-CL trap column (350 μm × 0.5 mm, 120 Å, Eksigent, AB Sciex) and nanoViper C18 separation column (75 μm × 250 mm, 3 μm, 100 Å; Acclaim Pep Map, Thermo Scientific, USA) in Eksigent nanoLC (Ultra 2D plus) system. The binary mobile solvent system was comprised of solvent A (2% (v/v) ACN, 0.1% (v/v) FA in water) and solvent B (98% (v/v) ACN, 0.1% (v/v) FA). The peptides were separated with 300 nl/min flow rate in a 60 min gradient with a total run time of 75 minutes. The data collection was carried out using the standard data-dependent IDA method. Each cycle had an acquisition duration of 250 and 100 ms for MS1 (m/z 350-1250 Da) and MS/MS (100-1500 m/z) scans, for a total cycle time of 2.8 s. Each fraction was tested twice.
The proteomics data were deposited to the ProteomeXchange Consortium through the PRIDE [57] partner repository, with the dataset identifier PXD035903.

Phospho-peptide and phospho-site identification
All raw files (.wiff) were processed to ProteinPilot software (version 4.5, SCIEX) using the Paragon algorithm (version 4. 5. 0. 0,1654). The phospho-peptides along with phospho-sites were identified against the entire HuR (elavl1) protein sequence. The following identification parameters were used: (a) trypsin was employed for proteolytic cleavage (b) Iodoacetic acid was used for Cys alkylation and (c) phosphorylation emphasis was used as a special factor. Peptides and proteins were verified at <1% false discovery rate (FDR) and with unused ProtScore of > 0.05. vested, and nuclear cytoplasmic fractionation performed. Western blotting was done for nuclear and cytoplasmic fraction to determine the ratio of HuR in nucleus and cytoplasm. GAPDH was used as a marker for cytoplasmic fraction and Histone H3 as a marker for nuclear fraction. Densitometry was performed and the numbers in bottom indicate Nuclear to cytoplasmic ratio of HuR after normalising with H3 and GAPDH. (C, D) Huh7.5 cells were transfected with vector control (pcDNA3.1), WT HuR overexpression construct or S318A HuR overexpression construct as indicated. 16h post transfection, cells were infected with HCV-JFH1 virus and 48h post infection, cells were harvested, and nuclear cytoplasmic fractionation performed. Western blotting was done for nuclear and cytoplasmic fraction to determine the ratio of HuR in nucleus and cytoplasm. The lower band in HuR blot represents endogenous HuR and the upper band represents overexpressed HuR. GAPDH was used as a marker for cytoplasmic fraction and Histone H3 as a marker for nuclear fraction. Densitometry was performed for overexpressed HuR (o/e HuR) and the numbers in bottom indicate Nuclear to cytoplasmic ratio of o/e HuR after normalising with H3 and GAPDH.