The HCV Replicase Interactome

Corresponding Author: Antonio Mas Centro Regional de Investigaciones Biomédicas, Universidad de Castilla-La Mancha, Albacete, Spain Email: Antonio.Mas@uclm.es Abstract: Viruses are obligate parasites and can only reproduce within host cells because they lack metabolic pathways to complete their replication cycles. Host factors required in viral replication are mainly those involved in lipid metabolism, cell cycle control and apoptosis, cell-to-cell interactions, immune system regulation, etc. Several inhibitors targeting viral polymerases have been designed. However, the rapid appearance of resistant mutants, as a direct consequence of the viral population structure, diminishes the efficacy of this kind of molecules. To elude the rapid loss of treatment efficiency due to the appearance of resistance mutations, cellular factors have been proposed as a promising therapeutic target to inhibit RNA(+) virus replication. In this review, we focus on those interactions between host factors and HCV replicase, to modulate either cellular metabolism or HCV polymerase activity.


Introduction
Viruses are obligate parasites and can only reproduce within host cells because they lack metabolic pathways to complete their replication cycles. Host factors required in viral replication are mainly those involved in lipid metabolism, cell cycle control and apoptosis, cellto-cell interactions, immune system regulation, etc. Viruses may infect a cell only if the cellular factors that virus needs to replicate are present in the cell (Flint et al., 2015;König and Stertz, 2015).
Positive strand RNA viruses (RNA(+) virus) are classified in the group IV of the Baltimore's classification of viruses. They are the greatest group of pathogenic viruses affecting human and animal health (Flint et al., 2015). RNA(+) include viruses from wellknown families as Coronaviridae (Alpha Coronavirus 1, SARS-related coronavirus, MERS-related coronavirus), Picornaviridae (Hepatitis A virus, Human Rhinovirus, Enterovirus including poliovirus), Flaviviridae (Dengue virus, Yellow Fever virus, Hepatitis C virus), among others (Flint et al., 2015). RNA(+) viruses replicate their RNA genomes through a negative strand intermediate and this reaction is catalyzed by a viral RNA dependent RNA Polymerase (RdRP) (Ferrer-Orta et al., 2015). Consequently, RdRP plays a key role in virus replication cycle (Verdaguer et al., 2014). RNA(+) genome replication is an error prone process and thereby genomic copies will carry mutations that could be selected in the viral offspring following Darwinian forces. Furthermore, RNA(+) virus replicate at large population size, reaching 10 10 -10 12 viruses in an infected individual. Putting these two factors together, error prone replication and population size, RNA(+) viral populations consist of mutant spectra (or mutant clouds) rather than genomes with the same nucleotide sequence. Mutant spectra, usually referred as viral quasispecies and not individual viral particles are the target of evolutionary events (Más et al., 2010).
Several inhibitors targeting viral polymerases have been designed. However, the rapid appearance of resistant mutants, as a direct consequence of the viral population structure, diminishes the efficacy of these kind of molecules (Más et al., 2010). To elude the rapid loss of treatment efficiency due to the appearance of resistance mutations, cellular factors have been proposed as a promising therapeutic target to inhibit RNA(+) virus replication (Lou et al., 2014). Factors of cellular origin cannot mutate and be selected to escape antiviral pressure at the same rate as virus factors. Therefore, host-targeted antivirals show high genetic barrier to escape (Plummer et al., 2015).
Hepatitis C Virus (HCV) is RNA(+) virus with a high-titer and error-prone replication rate leading to the generation of viral populations in which mixtures of almost infinite different variants called quasispecies may coexist (Más et al., 2010). HCV infection is widespread worldwide, showing geographical differences in terms of genetic identity with seven well defined genotypes (Baumert et al., 2016;Clemente-Casares et al., 2011). Independently of the infecting genotype, HCV infection is the main cause for cirrhosis and hepatocellular carcinoma (Westbrook and Dusheiko, 2014). HCV entry into the cell is mediated by the interaction of the glycoproteins from the viral envelope with receptors on the surface of the hepatocyte such as CD81, CLDN1 and OCLN among others (Ding et al., 2014). HCV entry is a complex process governed by viral and cellular factors and several of them contribute to liver tropism and limit host range of this virus. Once the virus has entered the cell, RNA(+) HCV genome is released into the cytoplasm where it is translated at the rough Endoplasmic Reticulum (ER) as a polyprotein (Paul et al., 2014). HCV polyprotein is about 3000 amino acids in length and is co-and post-translationally processed by proteases from cellular and viral origin to give ten mature viral proteins. HCV proteins are structural (core C and envelope proteins E1 and E2) and nonstructural (p7, NS2, NS3, NS4A, NS4B, NS5A and NS5B) proteins. Proteins C, E1 and E2 are main constituents of the virus particle. The p7 viroporin and NS2 participate in virus assembly. Finally, NS3, NS4A, NS4B, NS5A and NS5B form the replicase complex that is sufficient for viral RNA replication (Paul et al., 2014). RNA(+) replication product may be either used for translation, for synthesis of new negative strands, or can be packaged into virus particles that exit the cell via the secretory pathway. Translation and replication take place in opposite directions on the RNA(+) and cannot occur simultaneously. A rigorous control by cis-acting elements in the HCV genome and antigenome as well as cellular proteins and miRNAs mediates the transition from translation to replication (Sagan et al., 2015).
HCV replication takes place in microvesicles derived from ER where replication complex is located. Viral replicase is composed of at least viral proteins NS3, NS4A, NS4B, NS5A and NS5B. NS3 is composed of two domains located at N-terminal and C-terminal ends, showing serine-protease and helicase activities, respectively (Moradpour and Penin, 2013). The serineprotease domain is responsible for polyprotein cleavage in complex with the NS4A protease cofactor, whereas the helicase domain is important for RNA replication because of its RNA unwinding activity. NS4B is a poorly characterized protein with a complex transmembrane topology involved in inducing membrane alterations (Egger et al., 2002). NS5A is a RNA-binding phosphoprotein that exists as both a basal and a hyperphosphorylated form. The phosphorylation status of NS5A appears to be determined by several cellular kinases, including Glycogen Synthase Kinase 3 beta (GSK3β), Protein Kinase A (PKA), Casein Kinases (CK) I and II, polo-like kinase 1 and Mitogen-Activated Protein Kinases (MAPKs) (Colpitts et al., 2015). NS5A function seems to be to interact with other viral replicase components as well as cellular factors (Ross-Thriepland and Harris, 2015). Proteomics and molecular systematics approaches have been reported that more than one hundred proteins interact with NS5A (Tripathi et al., 2013;Li et al., 2014a). Affinity capture was also used for identifying host factors interacting with HCV RNA positive strand (Upadhyay et al., 2013). Some of them have been described above and comprises La protein (Kumar et al., 2013), Heterogeneous Nuclear Ribonucleoprotein L (hnRNP L) (Li et al., 2014b), Nuclear Factor 90 (NF90) (Li et al., 2014b), Vesicleassociated membrane protein-associated protein A and B (VAPA and VAPB) (Evans et al., 2004;Gao et al., 2004;Hamamoto et al., 2005), Polo-like Kinase 1 (Chen et al., 2010), TBC1 domain family member 20 (TBC1D20) (Nevo-Yassaf et al., 2012), Amphiphysin II (Zech et al., 2003), Reticulon 1 and 3 (RTN1 and RTN3) (Tripathi et al., 2013), Protein Phosphatase 2A (Georgopoulou et al., 2006), cyclophilin A (Liu et al., 2009), F-Box and Leucine-rich repeat protein 2 (FBXL2) (Wang et al., 2005), stress granule components (Pène et al., 2015) and the lipid kinase phosphatidylinositol-4 kinase III (Harak et al., 2014) among many others. Some of these are cellular kinases with well known roles in HCV infection in vivo (Reed et al., 1997).
NS5B is the viral RNA-dependent RNA Polymerase (RdRP) responsible for the synthesis of the (+) strand progeny through a (-) strand intermediate (Sesmero and Thorpe, 2015). NS5B X-ray crystal structures have revealed a polymerase-typical right-hand shape with fingers, palm and thumb subdomains (Verdaguer et al., 2014). The catalytic site is totally encircled, as other viral RdRP, with extensive interactions by loops connecting fingers and thumb subdomains (Verdaguer et al., 2014). The C-terminal end has a very hydrophobic peptide that allows NS5B to be anchored to ER membrane. This peptide can be removed to increase recombinant NS5B purification yields without affecting NS5B RdRP activity (López-Jiménez et al., 2014). In vitro RNA synthesis by NS5B can be induced in the presence of a template-primer or initiated by a de novo mechanism (López-Jiménez et al., 2014), the latter being the most likely to occur in vivo. A beta-hairpin from the thumb subdomain protrudes into the catalytic center preventing primer-dependent RNA synthesis (Lesburg et al., 1999). Residues in the tip of this structure act as a platform to initiate RNA synthesis by a de novo mechanism. Once the first phosphodiester bond is formed the beta-hairpin is removed and NS5B can complete genome replication (Appleby et al., 2015).
The cytoplasmic double-stranded RNA binding protein Staufen 1 (Stau1) coimmunoprecipitates HCV NS5B and the host factor Protein Kinase R (PKR), which is critical for interferon-induced cellular antiviral and antiproliferative responses (Dixit et al., 2016). Protein Kinase R (PKR) inhibits translation via eIF2α phosphorylation (Donnelly et al., 2013) and regulation of PKR activity is central for the control of cellular translation by several viruses (Flint et al., 2015). HCV may appropriate Stau1 to its advantage to prevent PKRmediated inhibition of eIF2α, which is required for the synthesis of HCV proteins and also for translocation of viral RNA genome to the polysomes for efficient translation and replication (Dixit et al., 2016). Our laboratory has recently described the interaction of NS5B with the Ser/Thr kinase Akt (Llanos Valero et al., 2016). This interaction has been confirmed by in vitro kinase assays, coimmunoprecipitation of NS5B and Akt, either expressed ectopically or from HCVcc infected cells. The interaction of HCV NS5B with this cellular kinase of the PI3K/Akt/mTOR pathway leads to a subcellular relocalization of Akt from a cytoplasmic to a perinuclear region in a clear colocalization with HCV polymerase. Relocalization was observed in cells transfected with plasmids encoding NS5B and Akt as well as in cells carrying a subgenomic replicon or HCVcc infected cells. NS5A is susceptible to be phosphorylated by Akt and relocalization of Akt with NS5B could drive NS5A phosphorylation at this subcellular region.
Relationship between HCV infection and sex hormones has been previously documented (Giannitrapani et al., 2006;Baden et al., 2014;White et al., 2014). Some estrogen-related drugs inhibits the production of HCV virus particles in an Estrogen Receptor alpha (ER1)-dependent manner (Hayashida et al., 2010). It has been also shown that ER1 may recruit NS5B to the HCV replication complex (Watashi et al., 2007) and our laboratory has described the interaction between HCV NS5B and ER1 in vitro, showing that this protein-protein interaction depends on NS5B oligomerization (Hillung et al., 2012). Cellular DEADbox helicase 5 (DDX5 or p68) also interacts with HCV NS5B (Goh et al., 2004). DDX5 is a RNA-dependent ATPase and it is implicated in cellular processes involving alteration of RNA secondary structure, such as translation initiation. DDX5 has been involved in HCV translation (Ríos-Marco et al., 2016) as well as in replication of other RNA(+) viruses as Japanese Encephalitis virus (Li et al., 2013) and retrovirus (Sithole et al., 2015) and negative strand RNA viruses as influenza virus (Jorba et al., 2008). DDX5 also interacts with Estrogen Receptor 1 (ER1) (Fujita et al., 2003) and with Akt (Zhu et al., 2011). Therefore, it seems to be a complex network comprising interactions among HCV replicase, Akt, DDX5 and ER1 in association with ER membrane that are important for HCV replication. However, experiments to demonstrate a clear localization of these host factors into the HCV replication complex have to be done. Once the mechanism governing these interactions will be decoyed we explore use of host factor inhibitors to treat viral infections. Currently, some inhibitors directed against ER1 (Riggs and Hartmann, 2003;Cuzick et al., 2013) and Akt (Brown, 2016;Nitulescu et al., 2016) are in clinical use or in development for treating other diseases.
Therefore, HCV polymerase interacts with several host factors that are important not only for viral replication process but also to control cell cycle, cell metabolism, etc (Lee et al., 2006). By these interactions, NS5B not only replicates HCV genome but also controls several cellular functions important for virus-cell relationship. Under these premises NS5B is a multifunctional protein so NS5B direct inhibition could lead to HCV replication inhibition by affecting several steps in the replicative cycle of the virus. However, the great genetic diversity of RNA(+) viruses make the appearance of resistant mutants a definite possibility. Targeting one or more of the interactions described above could also blockade HCV replication making it more difficult for the selection of resistant viruses.
Finally, several cellular pathways are shared by different RNA(+) viruses and targeting host factors could be useful for inhibiting viral infections from different viruses.

Conclusion
Viruses needs to replicate inside the cells usurping cellular functions. HCV NS5B, the main component of the viral replicase, not only replicates HCV RNA but also interacts with host factors to subjugate cellular metabolism. A deeper knowledge about NS5B-host interactions will be useful in the design of new, strongest and panviral antiviral strategies with limited side effects.

Acknowledgement
Special thanks to former members of the Molecular Virology lab at Centro Regional de Investigaciones Biomédicas (CRIB) for their past contributions. We apologize colleagues whose work has not been cited in this review.

Funding Information
This work was funded by Universidad de Castilla-La Mancha.

Author's Contributions
All authors equally contributed in this work.

Ethics
This article is original and contains unpublished material. The corresponding author confirms that all of the other authors have read and approved the manuscript and no ethical issues involved.