The nature of Viral Replication Centers: Coronavirus and Hepatitis C Virus RTCs

Positive strand RNA viruses are replicating in association with cytoplasmic membranes of infected cells which contains in general not viral RNA but also the machinery required for genome replication, namely the RNA dependent RNA Polymerase (RdRp) as well as accessory enzymes such as a viral RNA helicase. Whilst in most cases the RNA transcription complex (RTC) is localized and associated with the ER, in some cases the RTC is associated with the outer mitochondrial membrane (OMM) (Nodaviridae), the endosome/lysosome (Togaviridae), or even Peroxisomes and Chloroplasts (Tomboviridae); a special case are the lipid droplets formed upon Hepatitis C Virus infection, which are localized to the actin/microtubuli of the cytoskeleton. The replication enzymes -virus specific- become associated to the respective RTC early in infection process however not by the viral RNA but by the non-structural proteins (nsp’s) whose expression is sufficient to induce the formation of membrane structures which ultimately form the RTC - in other words, the replication proteins are targeted to the respective organelle by the nsp rather than the viral RNA. In order to form these structures, most of the nsp’s are integral proteins with multiple or single transmembrane domains (as in the case of Coronaviral nsp-3/-4/-6 proteins and Hepatitis C Virus NS4A/NS5A) or are attached to the membrane via a binding peptide (Semliki Forest virus nsp1). 

                         

Examples of Positive Strand RNA viruses inducing rearrangements of cellular membranes 

Coronavirus: Formation of DMV s containing the RTC adjacent to the ER

The RTC of the Coronaviridae consists of double-membrane vesicles (DMVs) in which the nsps required for the replication of the viral genome are anchored in addition to the viral genome and -once transcription of the genome starts- intermediate dsRNA can be detected by in situ hybridization or EU labeling of nascent newly synthesized RNA. In addition to viral proteins and viral RNA, the DMVs also contain host derived proteins such as the previously discussed EDAM1 or RNA binding proteins such as hnRNP, Polypyrimidine tract-binding protein (PTB), T-cell intracellular antigen-1 (TIA-1), and TIA-1-related protein (TIAR). RNA binding proteins are believed to be required to switch from replicating the viral genome to viral transcription by binding to the transcription regulating sequences; indeed, silencing of PTB has a negative effect on Coronaviral RNA accumulation but increases viral titer. The precise mechanism of targeting the viral RNA to sites of DMV formation/RTC assembly is not known but it has been postulated that the viral Nucleocapsid protein might be involved since it binds viral RNA and can be detected within DMVs.

The DMVs itself are derived from the cellular ER and regularly form a reticulovesicular network of modified ER membranes into which the viral nsp-3,-4, and -6 are inserted (also referred to as convoluted membranes) and induce a process known as “membrane zipping” or “membrane pairing”  which is succeeded  by the wrapping of membrane cisternae around cytoplasmic constituents in a process resembling the formation of autophagosomes or -to be more precise- the formation of  EDEMosomes in which the DMV forms a bud which eventually is released from the ER. Similar to the latter, the DMVs are believed to contain LC3-I although as I pointed out before it might be possible that the DMVs might be LC3-II positive. 
This notion is supported by evidence suggesting that the Coronaviral nsp6 protein can induce the formation of LC3-II positive  omegasomes that mature into autophagosomes; the induction of autophagy seems to be dispensable for the replication of SARS-CoV, MHV, and the avian Infectious Bronchitis Virus. If RTCs are also LC3-II positive than one of the question which needs to be answered is how the DMV/RTC escapes degradation - or maybe the answer in the end might be that LC3-II positive cannot be detected precisely because they are unstable and degraded. Since EDEMosomes are normally targeted for degradation in a LC3-II independent manner, the question remains why the EDEMosome-like Coronaviral DMVs are not degraded as well.

Further experiments might be worthwhile doing are to investigate if the DMVs undergo membrane fission in a RabGTPase dependent or independent manner. Here the use of the various RabGTPase mutants available might give us some answers - again is anybody up to the challenge?  Interestingly, two Coronavirus nsps, nsp-2 and -3, are located outside of the mature DMV are involved in inhibiting host cell pathways involving Prohibitin-1 and -, two proteins involved in modulating transcription as well as in maintaining mitochondrial function and morphology. It remains to be seen to which extent DMV bound nsp-2/-3 interferes with these processes.

Coronavirus nsp-3, -4, and -6 induce the formation of LC3-I positive EDEMosome like DMVsand hypothetically may form LC3-II positive DMVs

                    Hepatitis C Viruses: Role of the NS4B and NS5A proteins

The infection of cells with Hepatitis C Virus (HCV), a member of the Flaviviridae, triggers the formation of n membranous vesicles of heterogeneous size and morphology, which are associated with viral replication. Although originally it has been postulated that both structural and non-structural proteins are required for the formation of these vesicular structures, more recent results indicate that non-structural proteins NS4A and NS5B are inducing a membranous web and are able to recruit viral RNA independent of the structural Core-E1-E2-p7 complex as well as the NS3-4A complex. The entire HCV polyprotein however is required for membrane budding into the rER.

HCV NS4B is a 261 AA long protein, consisting of N- and C- terminal cytoplasmic domains and a central domain with four transmembrane helices and localizes to the ER in infected as well as transfected cells, although a small proportion of NS4B localizes either at the surface of Lipid droplets (LDs) or ER membranes associated with LDs.

Mutation analysis of full length NS4B showed that hydrophobic amino acids in amphipathic helix bind to and are critical for binding LDs and mutation of these residues caused a severe reduction in viral replication, suggesting that NS4B induced recruitment of LDs is critical for viral replication.  

In contrast to NS4B, the viral NS5A protein has multiple functions during the replication of HCV. As NS4B, NS5A localizes to the ER, where it interacts with viral proteases (NS2 and NS3/4A), the viral RdRp (NS5B) and the NS4B protein as well with the viral RNA (specifically 3’ untranslated region (UTR) ) and the viral Core protein. In addition, NS5A binds to at least one host cell protein that is found on the surface of LDs, Tail-Interacting Protein 47 (TIP47) via the N terminal PAT domain of TIP47, and thus relocalizes LDs to sites of HCV replication. siRNA mediated knockdown of TIP47 or mutations of NS5A prevent the recruitment of LDs and reduces viral replication whereas overexpression of TIP47 has the opposite effect, thus highlighting the importance of LDs for the replication of HCV. Although the function of NS5A and NS4B with regards to the recruitment of LDs seems to be similar on the first glance, NS5A binds to TIP47 localized on the surface of LDs followed by insertion of NS5A into the ER (although contradictory results suggest that only NS5A bound to viral RNA and/or Core recruits TIP47 positive LDs) by a mechanism that the author of these lines refers to as “capture”.

   

HCV NS4B and NS5A proteins "capture" LDs

So what is the role of TIP47? In non-infected cells, LDs are lipid storage organelles containing neutral lipids (triacylglycerides and sterol esters) which are surrounded by a phospholipid monolayer containing a number of associated proteins, including TIP47 that belongs to the LD associated Perilipin, ADRP and TIP47 (PAT) family. PAT domain containing proteins are coating the surface of LDs and involved in the turnover and stability of LDs. TIP47 is also found on late endosomes to which it is recruited by Rab9-GTPase which might facilitate the formation of transport vesicles. If these results stand the test of time, then the recruitment of TIP47 by NS5A to sites of viral replication might also recruit Rab9-GTPase, or the related Rab5-GTPase, which might be required for the formation of viral particles and trafficking to the endosome, the site of viral egress and released. Indeed, Rab9-GTPase expression is upregulated in HCV infected cells although NS5A does induce only a slight relocalisation of Rab9-GTPase suggesting that endosome-to-Golgi trafficking might be increased, whereas Rab5-GTPase co-localises with both NS5A and NS4B at the ER.  
TIP47 was originally described to function in the retrograde vesicular-membrane transport, whereby lysosomal enzymes are delivered to endosomes prior to recycling in the Golgi and thus be involved in the formation of transport vesicles. It seems feasible that this is precisely the role of TIP47 in HCV replication; to allow the transport of new viral particles. Personally I am curious to see if in TIP47 knockout cells transport vesicles are accumulating in the cytoplasm of the cell. It is well known that newly synthesized HCV particles are transported from early to late endosomes and subsequently to the Golgi, so is TIP47 maybe required for this step? Who knows….

One final word: the knowledge of the intricate details of the composition of the RTC not only of positive strand RNA viruses but also for DNA viruses, negative strand RNA viruses, or Retroviral agents might lead not only to better understanding of the interactions of viruses with the host cell but also lead to the discovery and development of new antiviral drugs as described for HCV. Until then, however, let us marvel in the wonderful world of viruses and their interactions with the host!

Further reading

Hagemeijer, M., Verheije, M., Ulasli, M., Shaltiel, I., de Vries, L., Reggiori, F., Rottier, P., & de Haan, C. (2009). Dynamics of Coronavirus Replication-Transcription Complexes Journal of Virology, 84 (4), 2134-2149 DOI: 10.1128/JVI.01716-09

Knoops, K., Kikkert, M., Worm, S., Zevenhoven-Dobbe, J., van der Meer, Y., Koster, A., Mommaas, A., & Snijder, E. (2008). SARS-Coronavirus Replication Is Supported by a Reticulovesicular Network of Modified Endoplasmic Reticulum PLoS Biology, 6 (9) DOI: 10.1371/journal.pbio.0060226


Maier HJ, Hawes PC, Cottam EM, Mantell J, Verkade P, Monaghan P, Wileman T, & Britton P (2013). Infectious bronchitis virus generates spherules from zippered endoplasmic reticulum membranes. mBio, 4 (5) PMID: 24149513


Sola I, Galán C, Mateos-Gómez PA, Palacio L, Zúñiga S, Cruz JL, Almazán F, & Enjuanes L (2011). The polypyrimidine tract-binding protein affects coronavirus RNA accumulation levels and relocalizes viral RNAs to novel cytoplasmic domains different from replication-transcription sites. Journal of virology, 85 (10), 5136-49 PMID: 21411518

Sola I, Mateos-Gomez PA, Almazan F, Zuñiga S, & Enjuanes L (2011). RNA-RNA and RNA-protein interactions in coronavirus replication and transcription. RNA biology, 8 (2), 237-48 PMID: 21378501

Cottam EM, Maier HJ, Manifava M, Vaux LC, Chandra-Schoenfelder P, Gerner W, Britton P, Ktistakis NT, & Wileman T (2011). Coronavirus nsp6 proteins generate autophagosomes from the endoplasmic reticulum via an omegasome intermediate. Autophagy, 7 (11), 1335-47 PMID: 21799305

Maier HJ, & Britton P (2012). Involvement of autophagy in coronavirus replication. Viruses, 4 (12), 3440-51 PMID: 23202545

Verheije MH, Hagemeijer MC, Ulasli M, Reggiori F, Rottier PJ, Masters PS, & de Haan CA (2010). The coronavirus nucleocapsid protein is dynamically associated with the replication-transcription complexes. Journal of virology, 84 (21), 11575-9 PMID: 20739524

Nagy PD, & Pogany J (2011). The dependence of viral RNA replication on co-opted host factors. Nature reviews. Microbiology, 10 (2), 137-49 PMID: 22183253

Egger D, Wölk B, Gosert R, Bianchi L, Blum HE, Moradpour D, & Bienz K (2002). Expression of hepatitis C virus proteins induces distinct membrane alterations including a candidate viral replication complex. Journal of virology, 76 (12), 5974-84 PMID: 12021330

Tanaka T, Kuroda K, Ikeda M, Wakita T, Kato N, & Makishima M (2013). Hepatitis C virus NS4B targets lipid droplets through hydrophobic residues in the amphipathic helices. Journal of lipid research, 54 (4), 881-92 PMID: 23315449

Foster, T., Belyaeva, T., Stonehouse, N., Pearson, A., & Harris, M. (2010). All Three Domains of the Hepatitis C Virus Nonstructural NS5A Protein Contribute to RNA Binding Journal of Virology, 84 (18), 9267-9277 DOI: 10.1128/JVI.00616-10

Vogt DA, Camus G, Herker E, Webster BR, Tsou CL, Greene WC, Yen TS, & Ott M (2013). Lipid droplet-binding protein TIP47 regulates hepatitis C Virus RNA replication through interaction with the viral NS5A protein. PLoS pathogens, 9 (4) PMID: 23593007

Ploen D, Hafirassou ML, Himmelsbach K, Sauter D, Biniossek ML, Weiss TS, Baumert TF, Schuster C, & Hildt E (2013). TIP47 plays a crucial role in the life cycle of hepatitis C virus. Journal of hepatology, 58 (6), 1081-8 PMID: 23354285

 Lai CK, Jeng KS, Machida K, & Lai MM (2010). Hepatitis C virus egress and release depend on endosomal trafficking of core protein. Journal of virology, 84 (21), 11590-8 PMID: 20739534

Kohler JJ, Nettles JH, Amblard F, Hurwitz SJ, Bassit L, Stanton RA, Ehteshami M, & Schinazi RF (2014). Approaches to hepatitis C treatment and cure using NS5A inhibitors. Infection and drug resistance, 7, 41-56 PMID: 24623983