Heal The Girls: June 2019

Flavivirus host factors: importance of the ER in viral replication

With the emergence of ZIKV in the Americas there has been a renewed interest in flaviviruses, in particular those that are transmitted by insects which historically only generated limited interest in the research community due to their inability to infect vertebrates or only causing relative mild illness.

In recent years however, the increase in infections caused by a number of flavivirus’ including West Nile Virus (WNV), Yellow Fever Virus (YFV), Dengue Virus (DENV) and Zika Virus (ZIKV) transmitted by insects such as Aedes sp. and Culex sp. in the Pacific islands, the Americas and more recently in Africa renewed an interest in these viruses and the advent of new technologies allows to study the virus-host interactions and assists in the identification of potential therapeutic targets.

Flavivirus infection and the EMC: pro-apoptotic during WNV infection whilst supporting replication of DENV, ZIKV and YFV?

The Endoplasmic Reticulum (ER) membrane complex (EMC) was originally discovered as part of a complex allowing the tethering Mitochondria to the ER and thus facilitating the exchange of lipids between the ER and the outer mitochondrial membrane (OMM) but later also being required for the assembly of multipass ER membrane proteins as well as the ER associated degradation (ERAD) pathway. Whilst it has been demonstrated that all EMC proteins interact with the mitochondrial translocase of the OMM (TOM) protein 5 (TOM-5), the role of the EMC in the assembly of proteins is less well characterised.

Genomic screens using RNAi and a CRISPR/Cas9a assay targeting 19052 genes, the replication of both ZIKV and DENV has been shown to depend on the presence of at least four components of the EMC, namely EMC-1, -3, -4 and -5, suggesting that the ability of maintaining tethering mitochondria to the ER and/or to process multipass ER membrane proteins is a significant factor for ZIKV and DENV replication in HeLa cells and 293T cells.

Loss of EMC decreases the level of both intracellular E protein and viral RNA as early as 40 min following the infection of HeLa cells with either DENV-2/NGC and ZIKV MR766, similar to Axl depleted cells, suggesting that EMC is a significant factor for viral binding and/or viral entry, with viral entry being suggested be the limiting factor as opposed to binding of viral particles. One mechanism might be that the loss EMC might prevent decapsidation of the viral genome by targeting endosomes to the lysosome and thus induce the degradation of viral RNA (see below).

The dependence of ZIKV and DENV on the integrity of the EMC therefore might extend beyond facilitating viral entry to the formation of the viral replication centre as well as the release of viral particles via COPII independent pathways similar to Mouse Hepatitis Virus. Further studies using the DENV and ZIKV replicon systems are however needed to characterize the role of EMC in viral replication and release as well in initiating the ERAD response. In the case of WNV, the expression of seven genes –including EMC-2 and -3- has been shown to be crucial to WNV induced cell death in HeLa and 293FT cells via the ERAD pathway whereas in DENV-NGC1 and various ZIKV strains (MR766, PR 2019 and Cambodia) infected HeLa cells the knockout of EMC-1, -2, -4, or-5 reduces viral replication, suggesting that the EMC supports viral replication as determined by intracellular staining for the viral E protein at 48 hrs p.i. . Closer examination suggests that in addition to a potential role of EMC in the formation of viral RC and/or transport of viral particles to the cell surface, EMC has also a role in viral entry, specifically after viral binding but prior to viral endocytosis. Although the role of the EMC is only poorly characterized, it might be possible that the knockout of EMC might promote the degradation of viral particles and/or the RC by targeting viral RC (and late endosomes containing viral particles following viral entry) to lysosomes and thus promote the degradation rather than release of mature viral particles. This hypothesis is supported by findings that the position and timing of endosome fission is dependent on the ER contact site. In Cos-7 cells expressing mCherry-Rab7 (a marker for late endosomes) and GFP- Sec61β (a marker for the ER), a small cargo containing Rab7⁺ compartment buds from a larger vacuolar Rab7⁺ compartment with an ER tubule localised perpendicular to the fission site that “cups” the bud just prior fission. Closer examination of these sites revealed that prior fission components of the retromer complex including FAM21 (which is involved in endosomal sorting) co-localise to the site of fission, indicating that the localization of proteins to the fission site determines the sorting of cargo. Future work however is needed to determine the role of the EMC in the sorting of endosomes and the role of EMC during the formation of viral RC; thus any role of EMC in the role of the development of ZIKV and DENV RC is hypothetical.

An additional role for EMC in the replication of both ZIKV and DENV might the recruitment of mitochondria and thus the facilitation of lipophagy. The transfer of phospholipids from the ER to Mitochondria is believed to be non-vesicular and to occur at sites of close contact between the ER and Mitochondria. In S. cerevisiae, deletion of EMC components leads to a decreased transfer of phosphatidylserine (PS) from the ER to Mitochondria which as a consequence contain decreased levels of both PS and Phosphatidyletholamine (PE), thus leading to decreased cell growth. Since both PS and PE are also involved in the formation of lipid droplets (LD), EMC deficiency might also impact viral replication by decreased formation of LD and thus decreased lipophagy. Again, more research is needed to verify the involvement of EMC in LD synthesis during ZIKV and DENV infection.

In the case of WNV, the deletion of EMC-2 partially prevents WNV induced apoptosis, indicating that EMC-2 is a factor for viral induced apoptosis. Besides the potential involvement of ERAD in WNV induced apoptosis not much is known and indeed speculative. One possible mechanism is that viral proteins associate with EMC-2 and thus increase ER stress, inducing ERAD and apoptosis due to lipid depletion. On the other hand, deletion of EMC-2 mitigates WNV induced apoptosis. It might be necessary therefore to monitor the ER stress response in WNV infected EMC-2^-/-cells by artificially inducing the accumulation of unfolded proteins in the ER.

The importance for the ER for ZIKV replication is further strengthened by observations that in cells infected with various members of the Flaviviridae –including but not limited to WNV, DENV, YFV, and Japanese Encephalitis Virus (JEV)- viral proteins are localised to the ER and indeed viral proteins are processed at the ER prior viral assembly. As has been discussed in prior posts, the localisation of viral proteins from JEV at the ER induces the ER stress response concomitant with the formation of autophagosomes and ZIKV infection of primary human fibroblasts has been associated with an increase in autophagosomes.

The role of the ER in the replication of Flavivirus’ has been further supported by recent findings that the expression of sgRNAs related to the ERAD response (including EMC-4 and -6), ER translocation machinery or the Oligosaccharyl transferase complex (OST) decreases the replication of WNV (Kunjin), ZIKV H/PF/2019, YFV (12D vaccine strain), JEV and DENV-2 in 293T cells as well as in a Drosophila cell line, DL-1. Two of the genes tested encode for two of the five components of the cellular Signal Peptidase Complex (SPC), namely SPCS-1 and SPCS-3. Indeed, WNV, JEV, DENV and ZIKV do not replicate in SPCS-1^-/-293T and SPCS-1^-/-Huh 7.5 cells and both WNV and DENV-2 are not replicating in U2OS cells transfected with either siRNA targeting SPCS-1 or SPCS-3, suggesting that viral polyprotein consisting of the structural and non-structural proteins requires to be processed at the ER by SPCS-1 and/or SPCS-3.

Figure: Prototype Flavivirus genome

This notion is supported by results showing that in SPCS-1^-/-293T cells infected with WNV the levels of both the viral prM and E protein are reduced at 12 hrs p.i. and non-cleaved prM, E proteins are detectable at 24 hrs p.i., whereas cleavage of the viral C protein in SPCS-1^-/-293T cells is not affected with similar finding in cells transfected with a prM-E-C plasmid, indicating therefore that the cellular Signal Peptidase complex is required for the cleavage of prM and E (but not C). Further experiments revealed that the leader sequence preceding the viral E protein however is not cleaved by SPCS (in contrast to the leader sequence preceding prM), suggesting that the cleavage of prM is necessary for the processing of E in a sequential manner. In a similar way the processing of the viral NS1 protein depends on the previous processing of prM but itself is not dependent processed by SPCS-1.

In line with these results, the loss of EMC might induce the accumulation of misfolded viral proteins and/or decreased incorporation of viral proteins into the ER and thus decrease viral replication.

Figure: Localisation of the signal sequence of prM in the context of structural and non-structural protein
localisation in the ER

In conclusion, the ER -in particular the EMC and Signal Peptidase Complex- plays a pivotal role in the replication of ZIKV and DENV. Whilst the connection between the EMC and viral replication is still obscure, the role of ER localised cellular signal peptidases is better characterized (at least for WNV) although questions remain, in particular if the cleavage of prM leader sequence induces a structural change that increases the stability of E which would be consistent with a chaperone-like role for prM in the folding of E in cells infected with Tick Borne Encephalitis Virus and evidenced by lower expression levels of both prM and E in SPCS-1^-/-293T cells expressing prM and E derived from WNV

compared to wt cells.

In addition to EMC and SPCS-1/-3, the genome wide CRISPR/Cas9a assay based screen identified other ER resident proteins that are required for viral replication, including components of OST and the ER translocation machinery such as OST-C and Sec61β whose contribution to viral replication is still undetermined.

It might be also of interest to explore the question if prM co-localises and/or interact directly with SPSC-1 or other components of Signal Peptidase complex? As mentioned above, the absence of the EMC might induce the relocalisation of endosomes to lysosomes and thus affect sorting. One of the questions to be answered therefore is if the absence of the EMC -or components of the EMC- induces the localisation of viral RC to lysosomes and thus prevents the release of viral particles. Further studies are needed to address these and other questions.

The importance of Coronavirus RTC for antiviral signaling

Internalised viruses are recognised by pathogen recognition receptors (PRRs) which sense Pathogen-associated Molecular Patterns (PAMP). Generally, PRRs are located within the endosomes -as it the case for Toll-like receptor (TLR)-7 or MyD88 - or alternatively in the cytoplasm, as it the case for retinoic acid inducible gene I (RIG-I) like receptors, and melanoma differentiation associated gene 5 (MDA5). In the case of RNA viruses, viral RNA is recognised by proteins belonging to the cytoplasmic RNA-induced silencing complex (RISC), which consists of RIG-1, MDA-5, protein kinase RNA activator (PACT), transactivation response RNA binding protein (TRBP), and Dicer. RIG-1, PACT, and MDA-5 recognize viral RNA and induce the expression of various cytokines, in particular Interferon(s) (IFN) and ultimately a broad array of IFN stimulated genes (ISG) as part of the antiviral response. In contrast, TRBP and Dicer process pre-microRNAs into mature microRNAs (miRNAs) by recruiting Argonaute 2 (Ago2) that target specific mRNA species -such as viral RNA- for regulation and targeting steroid-responsive promoters.

PACT, allowing the degradation of viral RNA as well as stimulating an immune response, can induce both TBRP and Dicer and thus targets viral microRNAs, viral small RNAs, and viral dsRNA.

ssRNA or dsRNA can induce an antiviral Interferon or NF-κB response via TLR or MAVS

In order to prevent host cell derived mRNAs from being recognised by cellular PRRs, cellular mRNA molecules not only possess a 5’ cap structure but are also methylated at their 5’ end; any RNA contained no 5’ cap, a non-methylated, or incompletely methylated 5’cap can therefore be recognised by cellular PAMP receptors and induce an antiviral response, in addition to viral dsRNA intermediates. In this system, RIG-1 preferentially recognizes short blunt end or 5’ triphosphate RNAs whereas MDA-5 recognizes dsRNA intermediates as well as non-methylated RNAs lacking 2’O methylated ribose. Antiviral signaling by these pathways can be inhibited by viruses using either the cellular machinery for capping viral RNAs, and thus prevent them from being recognised by the RISC, or alternatively encoding enzymes which cap the viral RNA in a manner similar to cellular enzymes.

In addition to stimulating the expression of antiviral genes, the recognition of single-stranded viral RNA by TLR-7 also induces the formation of autophagy vesicles via the interaction between the TLR adaptor MyD88 and the Beclin-1, the latter being required for the formation of the Phagosome/Autophagosomes.

Autophagy and the antiviral response

The formation of autophagy vesicles has been postulated to facilitate the presentation of viral antigens by the cellular MHC Class I (e.g. HSV-1 gB), MHC Class II complex (e.g. EBNA-1) and target viral RNA to endosomes (and thus being recognised by TLR-7) as well as facilitating the degradation of viral components. Indeed, infecting cells with a HSV-1 mutant deficient for the viral autophagy inhibitory ICP 34.5 protein not only fails to inhibit autophagy but also has decreased infectious titers, suggesting that autophagy degrades HSV-1 components in addition to facilitate the presentation of viral antigens.

Coronavirus RTC and the antiviral response

In the case of Coronavirus infected cells, the major pathogen associated pattern recognised by the PARRs is the dsRNA intermediate and the viral ssRNA both which are located with the replication-transcription complexes (RTCs) derived from the ER and induced by the expression of the viral nsp-3,-, and -6 proteins as discussed earlier. In these RTCs, the viral RNAs are not only synthesized but also modified in order to prevent the induction of the antiviral response. In order to prevent the viral RNA from being recognised by RIG-1 or MDA-5, the Coronavirus RNA contain a 5’cap structure which is added to newly synthesized viral RNA within the RTC by the viral RNA-triphosphatase (nsp13), 2’O-methyltransferase (nsp16), as well as a N7-Methyltransferase (nsp14). Failure to methylate viral RNA in cells infected with a nsp16 deficient virus induces a IFN type 1, MDA-5 dependent, antiviral response concomitant with elevated levels of ISGs. In addition to bind MDA-5, uncapped viral RNA binds IFN induced protein with tetratricopeptide repeats (IFIT)-1, thus preventing the translation of viral RNA. So far the precise contribution of the viral N7-Methyltransferase remains unknown although the equivalent in West Nile Virus has been shown to be required for 2’O-methyltransferase mediated methylation of WNV viral RNA. The N-terminus of nsp14 encodes for the viral 3’ -to-5’ exoribo-nuclease (ExoN) which hydrolyses both ssRNA and dsRNA and excises nucleotide mismatches in dsRNA intermediates, thus providing proofreading of newly synthesized RNA. Accordingly, inhibiting or deleting ExoN increases the potency of antivirals such as 5-Fluorouracil. Since both ExoN and 2’O-methyltransferase activity positively regulated by nsp10, inhibiting nsp10 might be an interesting target for antiviral therapy. In addition to ExoN, both the Arteriviridae and the Coronaviridae express another enzyme that can hydrolyze RNA, EndoU. In contrast to ExoN the function of EndoU (nsp15) is less well defined and it has been proposed that it might be involved in cleaving free -“mislocalised”-RNA in order to prevent the recognition by PRRs, although EndoU can be found both in RTCs as well as in the cytoplasm. One possibility, which the author of these lines suggests, is that EndoU might be associated with Endosomes that contain viral RNA (see below). These Endosomes might contain viral RNA as a result of the induction of the formation of autophagy vesicles by nsp-3,-4, and -6.

Although the Coronaviral nsps-3,-4, and -6 induce the formation of autophagy like (LC3-II negative) vesicles, so far the formation of autolysosomes and subsequent degradation of viral components in infected cells has not been demonstrated. Since the formation of LC3-II positive vesicles has been demonstrated in cells transfected with nsp6 derived from SARS-CoV, MHV, and avian IBV it might be possible that viral RNA derived from autophagy vesicles can be found in endosomes and induce TLR-7 mediated antiviral signaling and/or viral components transferred to multivesicular class II loading compartments. If this is the case, TLR-7/-8 mediated anitviral signaling might be inhibited by the orf4a protein (in the case of COVID-19 and SARS-CoV) whereas the MHC- Class II mediated activation of cytotoxic T lymphocytes might not be inhibited and indeed contribute in particular to the disease outcome of SARS-CoV or COVID-19 infected patients.

Coronavirus nsp-3,-4, and -6 might contribute to the induction of autophagy vesicle induced antiviral response counteracted by orf4a

The infection of microglia cells with a neurotropic strain of MHV, MHV-JHM, indeed lead to a sustained up-regulation of both MHC Class I and Class II molecules not only during viral induced inflammation but also following viral clearance, similar to patients which have recovered from SARS. The formation of LC3-II positive vesicles therefore might induce an antiviral response that might be partially blocked by viral proteins. Some evidence suggests that the formation of LC3-II is more pronounced in cells transfected with SARS-CoV derived nsp6 compared to IBV derived nsp6 but if this has any implication for inducing an antiviral response has to be investigated.

Last but not least, a short note on STAT1 and 2 mediated signaling, STAT 1 and 2 signaling can be inhibited by SARS-CoV orf6 protein which in contrast to the above mentioned proteins is not residing within the RTC but at the ER and Golgi membrane where it sequesters STAT1 and 2 and thus prevents nuclear entry of these. In this case, orf6 therefore prevents the activation of IFN-β stimulated genes.

The emerging COVID-19 has been shown to infect a wide variety of cells of the immune response, including dendritic cells, macrophages and T-Lymphocytes. It remains to be seen if the antiviral genes encoded by COVID-19 are differ in their ability to block antiviral signaling compared to those found in SARS-CoV or HCoV-NL63. It also remains to be seen if the antiviral signaling in dromedary camels or in bats infected with COVID-19 is less vulnerable to these proteins or not - not an easy task in dromedaries- and if the differences in the antiviral response account for the relatively benign outcome in these animals. In humans, it would be interesting to see if TLR-8 (restricted to myeloid dendrite cells, monocytes, and monocyte derived dendritic cells) is differently affected by COVID-19 derived proteins than TLR-7 (plasmacytoid dendrite cells). One might argue that these are marginal questions, but I would argue that it might shed some light on the pathogenesis of MERS.

The importance of antagonizing the antiviral response does not end at preventing the viral RNA from being recognised. A recent study indicated that the HCoV-OC43 Nucleocapsid protein binds microRNA9 and prevents a NF-κΒ dependent antiviral response and thus TLR-8 antiviral signaling; . In addition, the COVID-19 derived orf4b and SARS-CoV N protein also block NF-κΒ. Moreover, the SARS-CoV nsp1 papain-like protease inhibits IRF3 induced expression of IFN-β. It remains of course to be seen if the mechanisms described for SARS-CoV or HCoV-OC43 also apply to COVID-19 and more importantly if these mechanisms differ among various host species.

Coronavirus derived proteins block the antiviral signaling at various stages

As the reader of these lines can see, there is much to be learned from a virus family which until 2003 has been considered to be only a marginal virus by many.

Further reading

Kindler E, & Thiel V (2019). To sense or not to sense viral RNA-essentials of coronavirus innate immune evasion. Current opinion in microbiology, 20C, 69-75 PMID: 24908561

Zinzula L, & Tramontano E (2019). Strategies of highly pathogenic RNA viruses to block dsRNA detection by RIG-I-like receptors: hide, mask, hit. Antiviral research, 100 (3), 615-35 PMID: 24129118

Redfern AD, Colley SM, Beveridge DJ, Ikeda N, Epis MR, Li X, Foulds CE, Stuart LM, Barker A, Russell VJ, Ramsay K, Kobelke SJ, Li X, Hatchell EC, Payne C, Giles KM, Messineo A, Gatignol A, Lanz RB, O'Malley BW, & Leedman PJ (2019). RNA-induced silencing complex (RISC) Proteins PACT, TRBP, and Dicer are SRA binding nuclear receptor coregulators. Proceedings of the National Academy of Sciences of the United States of America, 110 (16), 6536-41 PMID: 23550157

English, L., Chemali, M., Duron, J., Rondeau, C., Laplante, A., Gingras, D., Alexander, D., Leib, D., Norbury, C., Lippé, R., & Desjardins, M. (2009). Autophagy enhances the presentation of endogenous viral antigens on MHC class I molecules during HSV-1 infection Nature Immunology, 10 (5), 480-487 DOI: 10.1038/ni.1720

English L, Chemali M, & Desjardins M (2009). Nuclear membrane-derived autophagy, a novel process that participates in the presentation of endogenous viral antigens during HSV-1 infection. Autophagy, 5 (7), 1026-9 PMID: 19556870

Paludan C, Schmid D, Landthaler M, Vockerodt M, Kube D, Tuschl T, & Münz C (2005). Endogenous MHC class II processing of a viral nuclear antigen after autophagy. Science (New York, N.Y.), 307 (5709), 593-6 PMID: 15591165

Cavignac Y, & Esclatine A (2010). Herpesviruses and autophagy: catch me if you can! Viruses, 2 (1), 314-33 PMID: 21994613

Taylor GS, & Rickinson AB (2007). Antigens and autophagy: the path less travelled? Autophagy, 3 (1), 60-2 PMID: 17102586

Züst R, Cervantes-Barragan L, Habjan M, Maier R, Neuman BW, Ziebuhr J, Szretter KJ, Baker SC, Barchet W, Diamond MS, Siddell SG, Ludewig B, & Thiel V (2011). Ribose 2'-O-methylation provides a molecular signature for the distinction of self and non-self mRNA dependent on the RNA sensor Mda5. Nature immunology, 12 (2), 137-43 PMID: 21217758

Chen Y, Su C, Ke M, Jin X, Xu L, Zhang Z, Wu A, Sun Y, Yang Z, Tien P, Ahola T, Liang Y, Liu X, & Guo D (2011). Biochemical and structural insights into the mechanisms of SARS coronavirus RNA ribose 2'-O-methylation by nsp16/nsp10 protein complex. PLoS pathogens, 7 (10) PMID: 22022266

Cao J, & Zhang X (2012). Comparative in vivo analysis of the nsp15 endoribonuclease of murine, porcine and severe acute respiratory syndrome coronaviruses. Virus research, 167 (2), 247-58 PMID: 22617024

Li T, Xie J, He Y, Fan H, Baril L, Qiu Z, Han Y, Xu W, Zhang W, You H, Zuo Y, Fang Q, Yu J, Chen Z, & Zhang L (2006). Long-term persistence of robust antibody and cytotoxic T cell responses in recovered patients infected with SARS coronavirus. PloS one, 1 PMID: 17183651

Hamo L, Stohlman SA, Otto-Duessel M, & Bergmann CC (2007). Distinct regulation of MHC molecule expression on astrocytes and microglia during viral encephalomyelitis. Glia, 55 (11), 1169-77 PMID: 17600339

Frieman M, Yount B, Heise M, Kopecky-Bromberg SA, Palese P, & Baric RS (2007). Severe acute respiratory syndrome coronavirus ORF6 antagonizes STAT1 function by sequestering nuclear import factors on the rough endoplasmic reticulum/Golgi membrane. Journal of virology, 81 (18), 9812-24 PMID: 17596301

Lai FW, Stephenson KB, Mahony J, & Lichty BD (2019). Human coronavirus OC43 nucleocapsid protein binds microRNA 9 and potentiates NF-κB activation. Journal of virology, 88 (1), 54-65 PMID: 24109243

Heal The Girls

Pages

InfoLinks

Flavivirus host factors: importance of the ER in viral replication

The importance of Coronavirus RTC for antiviral signaling