Flaviviruses such as dengue (DENV), Zika (ZIKV) or West Nile virus (WNV) express their genomes as a single multi-pass transmembrane proteins at the endoplasmic reticulum (ER) membrane. Subsequently, the polyprotein is cleaved by the viral NS2B-NS3 protease and the cellular signal peptidase into the individual structural (C, prM, E) and non-structural (NS1-5) proteins required for replication and assembly into new virions (Neufeldt et al., 2018). Similar to cellular transmembrane proteins, the viral polyprotein relies on the host cell machinery for targeting to the ER, translocation across the membrane and insertion of transmembrane domains (TMDs) into the lipid bilayer during its translation. These processes are facilitated by the signal recognition particle (SRP), the translocon-associated protein (TRAP) complex (also known as signal-sequence receptor complex), the Sec61 translocon and the signal peptidase complex. These cell components were all identified as essential host factors for flavivirus infection in genetic screens underscoring their importance for virus replication (Krishnan et al., 2008; Marceau et al., 2016; Sessions et al., 2009; Zhang et al., 2016). For most transmembrane proteins, recognition of a hydrophobic TMD or signal sequence by the SRP followed by transfer to the Sec61 translocation channel are sufficient for membrane targeting and accurate topogenesis (Shao and Hegde, 2011). Recently, it was shown that the ER membrane protein complex (EMC) plays a role in the insertion and/or stabilization of certain transmembrane proteins (Chitwood et al., 2018; Guna et al., 2018; Shurtleff et al., 2018). Intriguingly, several genome-wide CRISPR knockout (KO) screens for flavivirus dependency factors showed strong enrichment of the EMC (Ma et al., 2015; Marceau et al., 2016; Savidis et al., 2016; Zhang et al., 2016). However, its role in the virus life cycle is not yet understood.

The EMC was originally discovered in a genetic screen for yeast mutants modulating the unfolded protein response (UPR) (Jonikas et al., 2009). Genetic interaction mapping and biochemical purification revealed a six-subunit core complex (EMC1-6) conserved in yeast and mammals (Christianson et al., 2011; Jonikas et al., 2009). Other than its more general association with UPR and ER-associated degradation (ERAD), the EMC was shown to interact co-translationally and enable biogenesis of a subset of multi-pass transmembrane proteins (Shurtleff et al., 2018). Biochemical characterization of select EMC client proteins, such as tail-anchored proteins and G protein-coupled receptors (GPCRs), revealed that the EMC is capable of inserting TMDs into the ER membrane (Chitwood et al., 2018; Guna et al., 2018). Here, we investigated how the mosquito-borne flaviviruses, such as DENV, ZIKV and WNV, depend on the EMC for their polyprotein biogenesis during infection.

To validate the enrichment of EMC components in the flavivirus CRISPR screens, we generated isogenic EMC subunit KO cell lines in Huh7.5.1 and HEK293FT, two cell types that were used for the genetic screens (Figure 1—figure supplement 1A). Upon infection with DENV expressing Renilla luciferase (DENV-Luc), we measured a ~ 10,000 fold reduction in viral replication in Huh7.5.1 EMC1-6 KO cells (Figure 1A) and a 10–1,000-fold reduction in HEK293FT EMC1-5 KO cells (Figure 1B). The differences in the extent of the phenotypes may be attributed to the presence of compensatory pathways or the degree of disruption of the entire protein complex between the two cell types upon knockout of a single subunit. The strong deficit in DENV replication in the absence of the EMC persisted over a 30–72 hr period (Figure 1—figure supplement 1B). Complementation of EMC2 and EMC4 KO cells with the respective cDNA completely restored levels of DENV infection (Figure 1C and Figure 1—figure supplement 1C). Lastly, we tested the effect of EMC4 KO on several flaviviruses and saw a significant decrease of viral RNA for DENV, WNV and ZIKV but not hepatitis C virus (HCV), which is a more distantly related member of the Flaviviridae family (Figure 1D and Figure 1—figure supplement 1D). This is congruent with results of an HCV CRISPR screen that did not identify the EMC as a critical host factor (Marceau et al., 2016). This finding highlights that there is a specific, evolutionarily conserved dependency on the EMC among mosquito-borne flaviviruses and not a universal disruption of trafficking to and/or translation or replication at the ER membrane, which would affect a wider range of viruses.

Figure 1 with 1 supplement see all

Download asset Open asset

Next, we sought to determine at which step of the viral life cycle the EMC is required. As the EMC is important for the expression of a large set of transmembrane proteins, its disruption leads to an altered cell surface proteome, which may affect viral attachment to cells and/or entry. We measured effects on viral binding and uptake by quantifying the amount of viral RNA after virus incubation with WT and EMC4 KO cells at 2 and 6 hr post-infection (hpi), where we did not observe a difference (Figure 2A). Only at 24hpi, a timepoint when viral replication had occurred, there was a difference in RNA load evident between WT and EMC4 KO cells. Furthermore, we used a DENV replicon expressing luciferase, which bypasses entry and does not produce viral particles, thus allowing the analysis of the effect of EMC KO specifically on translation and replication. In this assay, we saw decreased luminescence in EMC4 KO cells starting at 12 hr post-electroporation suggesting that the EMC is important during or prior to viral replication (Figure 2B). We additionally performed the replicon assay in the presence of MK-0608, a nucleoside analogue that inhibits the viral polymerase (Chen et al., 2015), and observed that the luminescence signal in EMC4 KO cells was lower than in replication-inhibited WT cells suggesting that deletion of EMC has an effect on viral translation, which happens prior to replication (Figure 2—figure supplement 1A). To further dissect effects on translation or genome replication of DENV, we conducted an immunoblot timecourse series, where WT and EMC4 KO cells were infected at a high multiplicity of infection (moi = 32) and lysates were harvested at early timepoints post-infection (6-12hpi), when no or only little DENV replication had occurred. Viral protein expression was lower in EMC4 KO compared to WT cells as early as 6hpi further supporting a translation defect (Figure 2C). To rule out any contribution of low levels of replication to the observed phenotype, we also performed the immunoblot experiment in the presence of MK-0608 and still saw different amounts of DENV protein accumulation between WT and EMC4 KO cells indicating that the EMC is indeed important for viral protein expression (Figure 2—figure supplement 1B). Interestingly, while the EMC has been implicated in the stable biogenesis of transmembrane proteins, we found reduced expression of both TMD and non-TMD containing viral proteins. This suggests that the defect likely occurs at the polyprotein stage thus affecting all individual, cleaved proteins.

Figure 2 with 1 supplement see all

Download asset Open asset

The expression of flaviviral proteins is a complex process that requires several steps. DENV translation is initiated in a cap-dependent manner, whereupon the single-stranded positive-sense RNA genome is translated into a single ~ 3400 amino acid long multi-pass transmembrane protein. This polyprotein is co- and post-translationally processed by host and viral proteases producing the individual structural and non-structural proteins. We first tested whether the defect occurs at the stage of polyprotein translation. We used ribosome profiling (Ribo-seq), which measures the density of ribosomes on a given mRNA species with high positional resolution, to compare translation efficiencies (TE) and successful full-length synthesis of both cellular and viral proteins in WT and EMC4 KO cells (Ingolia et al., 2009). Any changes in TE or premature ribosome stalling in EMC KO vs WT cells would be reflected by changes in ribosomal footprints (as measured by ribosome profiling) normalized to changes in mRNA abundance (as measured by RNA-seq). In the uninfected condition, we did not observe dramatic changes in TE for cellular mRNAs in EMC4 KO vs WT cells (Figure 3A and Figure 3—source data 1). This is in line with previously reported results using CRISPRi knockdown against EMC subunits suggesting that deletion of EMC minimally impacts the synthesis rates of its client proteins (Shurtleff et al., 2018). Similarly, we did not find any substantial differences in viral TE in EMC4 KO cells relative to WT cells at 18 (0.43x) and 44hpi (2.17x) (Figure 3B and C and Figure 3—source data 1). Instead, we observed that DENV RNA was among the highest translated (top 0.1%) mRNAs in both WT and EMC4 KO cells (Figure 3—figure supplement 1A). The lower number of ribosomal footprints aligning to the viral genome in EMC4 KO cells was largely due to the reduced number of viral RNA copies present in the cells. Moreover, footprints were found to map throughout the viral genome in EMC4 KO cells, indicating that synthesis of full-length polyprotein still occurred (Figure 3—figure supplement 1B). Therefore, we concluded that translation of DENV polyproteins is not strongly impacted by loss of EMC, which led us to explore potential post-translational defects.

Figure 3 with 2 supplements see all

Download asset Open asset

Figure 3—source data 1

Ribosome Profiling data including read counts and fold changes for ribosome footprints and RNA-seq used to generate Figure 3A–C.

https://doi.org/10.7554/eLife.48469.009

Download elife-48469-fig3-data1-v2.xlsx

Proper insertion of TMDs and correct folding are required for stable expression of transmembrane proteins, while misfolded proteins are generally targeted to the proteasome for degradation via the ERAD pathway (Olzmann et al., 2013). To probe whether an increased fraction of viral proteins is degraded by the proteasome in EMC deficient cells, we infected WT and EMC4 KO cells with DENV in the presence of Bortezomib (BZ), a proteasome inhibitor, thus preventing removal of misfolded proteins from the cells. Indeed, BZ treatment led to an increase of detectable viral proteins in EMC4 KO cells but not WT cells (Figure 3D). Addition of BZ, however, did not restore DENV replication arguing that the proteasome-targeted proteins are non-functional (Figure 3—figure supplement 2). Together, this suggests that in the absence of EMC viral proteins are still translated but are unable to achieve a stable conformation resulting in their degradation and inability to establish efficient replication.

For cellular multi-pass client proteins, such as transporters and GPCRs, the EMC co-translationally engages with specific TMDs in order to support insertion into the membrane and allow stable expression of the entire transmembrane protein (Chitwood et al., 2018; Shurtleff et al., 2018). We used two orthogonal methods, a fluorescence-based protein stability assay and viral evolution, to examine which region of the viral polyprotein requires the EMC for successful expression. First, we utilized a dual-color fluorescent reporter as previously described (Chitwood et al., 2018). The C terminus of different segments of the DENV polyprotein was fused to eGFP followed by a viral T2A sequence and mCherry (G2C) (Figure 4A). Translation of the fusion constructs generates two protein products at a 1:1 ratio due to peptide bond skipping by the ribosome at the T2A site: 1) a polyprotein segment fused to eGFP and 2) mCherry. Any post-translational instability of the eGFP fusion protein due to misfolding results in an eGFP:mCherry ratio of less than 1. As expected, the cytosolic G2C-only construct displayed a 1:1 ratio in both WT and EMC4 KO cells, while the β₁-adrenergic receptor (ADRB1) fused to G2C exhibited drastic destabilization in EMC4 KO compared to WT cells as was previously reported (Figure 4B) (Chitwood et al., 2018). Applying this approach to different DENV polyprotein G2C fusion constructs, we observed that the segments containing only the structural proteins (capsid, prM and E) or NS1-3 were not destabilized in EMC4 KO Huh7.5.1 cells (Figure 4B). By contrast, constructs encoding both NS4A and NS4B (but not either of them alone) showed reduced stability in EMC4 KO cells (Figure 4B). Similar results were obtained in HEK293FT (Figure 4—figure supplement 1). NS4A has 3 TMDs with the third (termed 2K peptide) being highly hydrophobic and serving as a signal sequence for the translocation of NS4B into the ER membrane (Lin et al., 1993). Subsequently, the NS4A-2K junction is cleaved by the viral protease and the 2K-NS4B junction by the signal peptidase. As the 2K-targeted NS4B-5-G2C construct did not display a stability defect in EMC deficient cells, the results suggest that EMC mediates stable expression by first engaging within NS4A prior to the 2K peptide and that a defect in NS4A is extended to the NS4B region. Stability of NS4A-4B-G2C was restored to near WT levels by complementation of EMC4 KO cells with EMC4 cDNA (Figure 4C). Moreover, treatment of EMC4 KO cells with BZ increased the eGFP:mCherry ratio of NS4A-4B-G2C in EMC4 KO cells indicating that this region of the polyprotein is misfolded and targeted for degradation in the absence of EMC (Figure 4D). Therefore, we linked the DENV replication phenotype in EMC deficient cells to a specific molecular defect in the NS4A-4B region of the polyprotein.

Figure 4 with 1 supplement see all

Download asset Open asset

In a second, orthogonal approach, we performed viral adaptation to evolve EMC-independent DENV mutants by passaging the virus consecutively on EMC4 KO Huh7.5.1 cells every 3–4 days. After 20–22 passages we observed cell death in 3 out of 6 independent adaptation experiments. Sequencing of RNA extracted from supernatant and alignment to the WT16681 DENV genome revealed two non-synonymous point mutations in NS4A and NS4B, which were conserved across all three isolates (Figure 5A). The A6665G mutation led to a tyrosine to cysteine and the A7558T mutation to an asparagine to tyrosine substitution. The majority of other SNPs was silent (Figure 5—source data 1). A6665G is predicted to be inside a TMD helix encoding region of NS4A suggesting that this TMD relies on the EMC for proper ER membrane insertion (Miller et al., 2007). Only one of 1,359 DENV serotype two genomes from the NCBI database has a G at this position encoding cysteine, while other sequences all encode tyrosine. This evolutionary conservation suggests that this residue indeed has an important function. Similarly, the frequency of A at position 7558 is very conserved with 99.6%, and there is no T encoding tyrosine in the 1359 sequenced strains. Interestingly, A7558T resides in a non-TMD region at the end of NS4B, thus likely not supporting membrane insertion but folding in a different manner (Miller et al., 2006). For all three EMC adapted isolates, replication levels in EMC4 KO cells were similar to levels in WT cells, while WT16681 DENV displayed a ~ 100 fold reduction in viral RNA in EMC4 KO relative to WT cells (Figure 5—figure supplement 1A). The same effect was observed in a 72 hr viral growth curve (Figure 5B). The adapted virus appeared less fit than the WT DENV at earlier timepoints but reached similar levels of replication at 72 hpi suggesting that the adaptive mutations did not cause dramatic growth defects.

Figure 5 with 1 supplement see all

Download asset Open asset

Figure 5—source data 1

Identified SNPs in EMC KO adapted DENV isolates relative to the DENV 16681 reference genome.

A minimum variant frequency cutoff of 80% was used.

https://doi.org/10.7554/eLife.48469.013

Download elife-48469-fig5-data1-v2.xlsx

To dissect whether both mutations are required for optimal replication in EMC KO background, we introduced them individually or in tandem into the DENV-Luc infectious clone and generated virus. While A6665G and A7558T individually restored DENV replication in EMC4 KO cells to 30% and 86% of WT levels, respectively, the double mutant displayed complete rescue of viral growth in EMC KO cells (Figure 5C). Similarly, we measured ~ 10 and 100-fold higher replication levels for the single and double mutant DENV-Luc, respectively, compared to WT DENV-Luc in EMC KO cells over a 72 hr period, while replication in WT cells was comparable (Figure 5—figure supplement 1B). We also introduced the two adaptive mutations into NS4A-5-G2C, which showed increased stability in EMC4 KO cells compared to the WT construct thus reverting the molecular destabilization phenotype (Figure 5D). Remarkably, the G2C fluorescence assay and viral adaptation independently pinpointed NS4A-4B of the polyprotein as the region, which requires the EMC. The results of the two experiments suggest that a defect in the stable insertion of the NS4A TMD2 could lead to an aberrant topology/instability of NS4B. By contrast, polyprotein segments starting with 2K-NS4B were stably expressed in EMC KO cells as was shown using NS4B-5-G2C construct (Figure 4A). Therefore, the EMC is required for correct insertion, processing and/or folding of the NS4A-4B region.

To probe further how absence of EMC causes failure of DENV replication, we transfected a subgenomic NS2B-4B construct (containing functional NS2B-NS3 for polyprotein cleavage) with a C-terminal fusion to eGFP into WT and EMC4 KO cells and immunoblotted against the different processed products. While there was no notable difference in expression levels between the higher molecular weight intermediates (NS2B-NS3-NS4A-NS4B-eGFP or NS4A-NS4B-eGFP), we observed reduced accumulation of processed NS4B-eGFP in EMC4 KO cells (Figure 6A). BZ treatment indicated rapid degradation of cleaved NS4B-eGFP in EMC deficient cells. Furthermore, if the EMC acts to insert or stabilize NS4A-4B, we hypothesized that direct binding of NS4B to EMC post polyprotein-processing should be detectable in WT cells. We performed a co-immunoprecipitation using FLAG-tagged EMC4 and detected NS4B but not prM or NS5 in the pulldown (Figure 6B). This supports the finding that EMC physically engages with NS4A-4B to facilitate its biogenesis and stable expression. Interestingly, we also detected physical interaction between the EMC and NS4B of the adapted DENV (Figure 6C). This suggests that the two amino acid substitutions in the NS4A and NS4B viral proteins do not ablate the signal triggering EMC engagement. The mutations may rather allow the virus to maintain proper folding and stability in the absence of the EMC.

Figure 6

Download asset Open asset

Finally, we examined the topology of the NS4A-4B transmembrane protein region in WT and EMC deficient cells using a protease protection assay. We generated different constructs, in which eGFP was C-terminally fused after different TMDs that had been determined experimentally (Figure 6D) (Miller et al., 2006). Transfected cells were left untreated or treated with digitonin and proteinase K followed by immunoblot. In the case of luminal localization, eGFP is protected against protease digestion, while cytoplasmic eGFP is degraded. Our results showed that in WT cells eGFP fused after NS4B TMD1 and TMD4 is luminal, which is consistent with the previously determined topology of NS4B (Figure 6D) (Miller et al., 2006). By contrast, we did not observe luminal localization for NS4B TMD1 and TMD4 in EMC4 KO cells, which indicates that EMC is required for the achieving the correct topology and thus stable expression of NS4B. We additionally probed the topology of NS4A using a deglycosylation assay similar to a previous study (Miller et al., 2007). We introduced an opsin tag containing a glycan acceptor site downstream of NS4A TMD2, which had been determined to be in the ER lumen and should therefore be glycosylated. We observed efficient glycosylation in WT cells (as expected) but very poor glycosylation in EMC4 KO cells, indicating an aberrant topology (Figure 6E). Moreover, we sought to test whether the adaptive mutations restore the topology of NS4A-4B or allow the virus to replicate with a different topology in the absence of the EMC. Since a topological defect was observed early in NS4A-4B and membrane translocation occurs co-translationally, we focused on the A6665G mutation, which is located in TMD2 of NS4A, rather than the A7558T mutation near the C-terminus of NS4B. As the A6665G mutation led to a significant increase in replication in EMC deficient cells, we would expect any rescue of correct topology to be visible. However, no increase of glycosylation was observed in EMC4 KO cells using the NS4A-TMD2-opsin construct containing the adaptive mutation (Figure 6E). Therefore, we conclude that the adaptive mutation in NS4A does not immediately restore the correct topology but instead allows the virus to replicate with NS4A-4B assuming an aberrant topology. However, we cannot exclude that the adaptive mutations, which were selected by passaging full virus, only restore the wild-type topology in the context of full-length viral polyprotein, for example through interactions with other viral proteins.

Based on our results and previous work (Chitwood and Hegde, 2019; Shurtleff et al., 2018) we propose the following model for the role of the EMC in flavivirus infection. During translation of the polyprotein the EMC engages with TMDs in NS4A and NS4B to ensure their accurate topology, correct folding and stable expression (Figure 6F). In the absence of the EMC, TMDs in the NS4A-4B region are not accurately or stably integrated in the ER membrane leading to misfolding and degradation of processed NS4B (and potentially also NS4A). Based on the molecular weight of the detected bands in the protease protection assay (Figure 6D), we expect that cleavage between 2K and NS4B by the signal peptidase still occurs in EMC deficient cells, which is consistent with previously observed processing of dengue NS4A-2K-NS4B (Cahour et al., 1992). This suggests that the N-terminus of NS4B is still localized to the ER lumen (at least while being translated) in EMC KO cells. However, unlike in WT cells, where the TMD1 and TMD4 of NS4B are luminal, their termini are localized to the cytoplasmic site of the ER membrane in EMC KO cells highlighting that EMC is required for achieving the correct topology of NS4B. Our data suggests that EMC does not affect ER targeting and translocation of 2K but it still remains an open question whether EMC facilitates proper insertion of TMDs into the membrane or if it is required for maintaining the correct topology in a chaperone-like function post membrane integration.

NS4A and NS4B fulfill crucial functions for viral replication. NS4A induces ER membrane rearrangements important for the formation of viral replication complexes (Miller et al., 2007; Roosendaal et al., 2006) while NS4B interacts with the NS3 helicase and also suppresses the cellular interferon response (Muñoz-Jordán et al., 2005; Zou et al., 2015). Additionally, it was demonstrated that the NS4A-2K-NS4B precursor interacts with NS1, important for viral replication (Płaszczyca et al., 2019).

Recently, another study found that biogenesis of DENV and ZIKV NS4A and NS4B is dependent on the EMC (Lin et al., 2019). While Lin et al. used ectopic expression of individual viral proteins to identify NS4A and NS4B as EMC client proteins, we performed a fluorescent based stability assay expressing different fragments of the viral polyprotein leading to similar conclusions. However, we only observed destabilization when NS4B was preceded by NS4A arguing that the NS4A-4B intermediate triggers an EMC-dependent process. Both studies showed that the EMC physically interacts with NS4B to facilitate stable expression and that, in absence of EMC, NS4B is misfolded and targeted for degradation as demonstrated by proteasomal inhibition. Lin et al. showed that NS4B is still membrane associated in EMC deficient cells, while we provided further molecular insights that NS4A-4B assumes an aberrant topology at the ER membrane. Furthermore, bioinformatic prediction of NS4B transmembrane regions revealed that the two N-terminal domains are only weakly hydrophobic, which could explain the dependence of NS4B on the EMC for stable expression (Lin et al., 2019). This is consistent with the systematic analysis of cellular EMC client proteins, which showed an underrepresentation of hydrophobic and an overrepresentation of charged and aromatic amino acids within their TMDs compared to all TMDs in the proteome (Shurtleff et al., 2018). Consequently, Lin et al. demonstrated that substitution of all polar and charged residues inside TMD1 and TMD2 of NS4B to hydrophobic leucines led to complete rescue of ectopic NS4B expression in EMC depleted cells. However, it remains unclear whether these substitutions would allow for viral replication in the full polyprotein context. By contrast, our EMC KO adapted mutations do not reside in TMD1 and TMD2 of NS4B but in NS4A TMD2 and near the C-terminus of NS4B, and were shown to increase NS4A-4B expression levels as well as restore viral replication. Overall, both studies are complementary and in large agreement, and together provide a detailed picture of how the EMC ensures stable expression of two viral multi-pass transmembrane proteins.

Interestingly, we found that the A6665G mutation inside NS4A TMD2 does not restore the wild-type topology. The A6665G mutation also arose during DENV adaptation to cell lines deficient in the oligosaccharyltransferase (OST) complex, another critical host factor inside the ER membrane (Marceau et al., 2016; Puschnik et al., 2017). The OST complex has been shown to be important for RNA genome replication in a catalytically independent function and was suggested to act as structural scaffold for viral replication complexes. It is thus possible that the tyrosine to cysteine substitution generally modulates ER membrane integration of NS4A-4B or exposes a binding interface important for replication complex formation, thus likely rescuing replication without restoring the normal topology. Since membrane proteins are co-translationally translocated across and inserted into the membrane, it is unlikely that the A7558T mutation downstream of the apparent topology defect restores proper membrane integration. Possibly, the substituting tyrosine may stabilize the cytosolic orientation of the C-terminus through expansion of the ‘aromatic belt’ (Granseth et al., 2005) to allow for efficient viral protease cleavage or facilitate other critical protein-protein interactions.

The isolation of EMC-independent DENV mutants additionally enables drug screening for EMC modulators by testing compounds against WT DENV-Luc and then counter-screening hits with EMC-adapted virus. EMC inhibitors may be a useful chemical tool to study EMC function and also provide a potential antiviral therapeutic against flaviviruses. Host-directed antivirals may have the advantage of a broader activity against multiple viruses and a higher barrier of drug resistance. Only 50% of the adaptation experiments yielded escape mutants containing two mutations after > 20 passages. For comparison, for direct-acting antivirals (e.g. viral polymerase inhibitors), one point mutation is often sufficient to overcome the mechanism of action (Beaucourt and Vignuzzi, 2014).

In addition to characterizing the role of the EMC in the viral life cycle, our data also help to further elucidate its cellular function. The emerging role of the EMC to facilitate membrane protein insertion and biogenesis has been implicated for several cellular transmembrane proteins (Chitwood et al., 2018; Coelho et al., 2019; Guna et al., 2018; Shurtleff et al., 2018; Volkmar et al., 2019). For a subset of tail-anchored proteins, the EMC inserts the sole C-terminal TMD after translation is completed in the cytosol (Guna et al., 2018). Moreover, the EMC co-translationally inserts the first N-terminal TMD of many GPCRs with an N_exo topology (N terminus in the ER lumen) (Chitwood et al., 2018). By contrast, we found EMC-dependency within DENV NS4A-4B, a region with several TMDs, which are not close to the termini of the polyprotein. Moreover, as NS4A is already targeted to and inserted into the ER membrane, it argues against an insertase mechanism (as shown for tail-anchored proteins and GPCRs) to ensure NS4B expression. This provides experimental evidence that the EMC is not only capable of supporting stable expression of TMDs near the N and C termini but also of internal TMDs, arguing that the EMC may directly facilitate the biogenesis of a wide variety of implicated transmembrane proteins beyond tail-anchored proteins and GPCRs as suggested previously (Shurtleff et al., 2018). It is noteworthy that the DENV polyprotein is distinct from human transmembrane proteins in that it contains multiple signal peptidase cleavage sites along the polyprotein, including at the 2K/NS4B site between NS4A and NS4B, which removes the 2K peptide acting as signal sequence for NS4B (Lin et al., 1993; Miller et al., 2006). Interestingly, it was shown for ADRB1 that addition of a signal sequence or an N_cyt TMD upstream of the first N_exo TMD makes the EMC dispensable for biogenesis (Chitwood et al., 2018), whereas NS4B is EMC-dependent despite the upstream 2K signal peptide within the polyprotein. This observation suggests that EMC-mediated integration/stabilization of TMDs may be context-dependent. Generally, for both cellular multi-pass transmembrane proteins and the DENV polyprotein, only one or a subset of TMDs seem to require the EMC for accurate topology, while the other TMDs rely on the Sec61 translocon. The exact interplay between Sec61 and the EMC during transmembrane protein biogenesis warrants further investigation. The simultaneous dependency of flaviviruses on both Sec61 and EMC as highlighted by the CRISPR KO screens could allow an in-depth analysis of the sequence and TMD features that may determine membrane insertion via Sec61 or EMC.

Lastly, the EMC was originally discovered to be associated with the UPR (Jonikas et al., 2009), and upregulation of UPR during flavivirus infection has been reported (Peña and Harris, 2011; Reid et al., 2018; Su et al., 2002). It is possible that competition between viral polyproteins and cellular proteins for the machinery important for proper transmembrane protein expression (such as EMC, OST complex and ERAD components) is linked to this phenomenon. The induction of ER stress can further trigger autophagy, which has also been shown to be beneficial for flavivirus infection (Abernathy et al., 2019; Lee et al., 2018; Senft and Ronai, 2015). Therefore, hijacking the EMC may have direct and indirect implications for flavivirus replication.

Overall, our study gives detailed insights into how flaviviruses exploit the EMC to support proper polyprotein expression, and we provide further evidence for the cellular function of the EMC in ensuring accurate transmembrane protein topology by utilizing DENV as a cell biological tool.

Ribosome profiling/RNAseq data have been deposited as NCBI BioProject under the accession code PRJNA526529.

1. 1. Ngo AM
   2. Shurtleff MJ
   3. Popova KD
   4. Kulsuptrakul J
   5. Weissman JS
   6. Puschnik AS

   (2019) NCBI BioProject

   ID PRJNA526529. Ribosome profiling and RNA-seq of dengue virus infected HEK293 cell lines.

   https://www.ncbi.nlm.nih.gov/bioproject/?term=PRJNA526529

1. 1. Abernathy E
   2. Mateo R
   3. Majzoub K
   4. van Buuren N
   5. Bird SW
   6. Carette JE
   7. Kirkegaard K

   (2019) Differential and convergent utilization of autophagy components by positive-strand RNA viruses

   PLOS Biology 17:e2006926.

   https://doi.org/10.1371/journal.pbio.2006926

   • PubMed
   • Google Scholar
2. 1. Beaucourt S
   2. Vignuzzi M

   (2014) Ribavirin: a drug active against many viruses with multiple effects on virus replication and propagation. Molecular basis of Ribavirin resistance

   Current Opinion in Virology 8:10–15.

   https://doi.org/10.1016/j.coviro.2014.04.011

   • PubMed
   • Google Scholar
3. 1. Brinkman EK
   2. Chen T
   3. Amendola M
   4. van Steensel B

   (2014) Easy quantitative assessment of genome editing by sequence trace decomposition

   Nucleic Acids Research 42:e168.

   https://doi.org/10.1093/nar/gku936

   • PubMed
   • Google Scholar
4. Book

   1. Buentzel J
   2. Thoms S

   (2017)

   The Use of Glycosylation Tags as Reporters for Protein Entry into the Endoplasmic Reticulum in Yeast and Mammalian Cells

   In: Schrader M, editors. Peroxisomes: Methods and Protocols. New York: Springer. pp. 221–232.

   • Google Scholar
5. 1. Cahour A
   2. Falgout B
   3. Lai CJ

   (1992)

   Cleavage of the dengue virus polyprotein at the NS3/NS4A and NS4B/NS5 junctions is mediated by viral protease NS2B-NS3, whereas NS4A/NS4B may be processed by a cellular protease

   Journal of Virology 66:1535–1542.

   • PubMed
   • Google Scholar
6. 1. Chen YL
   2. Yokokawa F
   3. Shi PY

   (2015) The search for nucleoside/nucleotide analog inhibitors of dengue virus

   Antiviral Research 122:12–19.

   https://doi.org/10.1016/j.antiviral.2015.07.010

   • PubMed
   • Google Scholar
7. 1. Chitwood PJ
   2. Juszkiewicz S
   3. Guna A
   4. Shao S
   5. Hegde RS

   (2018) EMC is required to initiate accurate membrane protein topogenesis

   Cell 175:1507–1519.

   https://doi.org/10.1016/j.cell.2018.10.009

   • PubMed
   • Google Scholar
8. 1. Chitwood PJ
   2. Hegde RS

   (2019) The role of EMC during membrane protein biogenesis

   Trends in Cell Biology 29:371–384.

   https://doi.org/10.1016/j.tcb.2019.01.007

   • PubMed
   • Google Scholar
9. 1. Christianson JC
   2. Olzmann JA
   3. Shaler TA
   4. Sowa ME
   5. Bennett EJ
   6. Richter CM
   7. Tyler RE
   8. Greenblatt EJ
   9. Harper JW
   10. Kopito RR

   (2011) Defining human ERAD networks through an integrative mapping strategy

   Nature Cell Biology 14:93–105.

   https://doi.org/10.1038/ncb2383

   • PubMed
   • Google Scholar
10. 1. Coelho JPL
    2. Stahl M
    3. Bloemeke N
    4. Meighen-Berger K
    5. Alvira CP
    6. Zhang ZR
    7. Sieber SA
    8. Feige MJ

    (2019) A network of chaperones prevents and detects failures in membrane protein lipid bilayer integration

    Nature Communications 10:672.

    https://doi.org/10.1038/s41467-019-08632-0

    • PubMed
    • Google Scholar
11. 1. Granseth E
    2. von Heijne G
    3. Elofsson A

    (2005) A study of the membrane-water interface region of membrane proteins

    Journal of Molecular Biology 346:377–385.

    https://doi.org/10.1016/j.jmb.2004.11.036

    • PubMed
    • Google Scholar
12. 1. Guna A
    2. Volkmar N
    3. Christianson JC
    4. Hegde RS

    (2018) The ER membrane protein complex is a transmembrane domain insertase

    Science 359:470–473.

    https://doi.org/10.1126/science.aao3099

    • PubMed
    • Google Scholar
13. 1. Ingolia NT
    2. Ghaemmaghami S
    3. Newman JR
    4. Weissman JS

    (2009) Genome-wide analysis in vivo of translation with nucleotide resolution using ribosome profiling

    Science 324:218–223.

    https://doi.org/10.1126/science.1168978

    • PubMed
    • Google Scholar
14. 1. Jonikas MC
    2. Collins SR
    3. Denic V
    4. Oh E
    5. Quan EM
    6. Schmid V
    7. Weibezahn J
    8. Schwappach B
    9. Walter P
    10. Weissman JS
    11. Schuldiner M

    (2009) Comprehensive characterization of genes required for protein folding in the endoplasmic reticulum

    Science 323:1693–1697.

    https://doi.org/10.1126/science.1167983

    • PubMed
    • Google Scholar
15. 1. Krishnan MN
    2. Ng A
    3. Sukumaran B
    4. Gilfoy FD
    5. Uchil PD
    6. Sultana H
    7. Brass AL
    8. Adametz R
    9. Tsui M
    10. Qian F
    11. Montgomery RR
    12. Lev S
    13. Mason PW
    14. Koski RA
    15. Elledge SJ
    16. Xavier RJ
    17. Agaisse H
    18. Fikrig E

    (2008) RNA interference screen for human genes associated with west nile virus infection

    Nature 455:242–245.

    https://doi.org/10.1038/nature07207

    • PubMed
    • Google Scholar
16. 1. Lee YR
    2. Kuo SH
    3. Lin CY
    4. Fu PJ
    5. Lin YS
    6. Yeh TM
    7. Liu HS

    (2018) Dengue virus-induced ER stress is required for autophagy activation, viral replication, and pathogenesis both in vitro and in vivo

    Scientific Reports 8:489.

    https://doi.org/10.1038/s41598-017-18909-3

    • PubMed
    • Google Scholar
17. 1. Lin C
    2. Amberg SM
    3. Chambers TJ
    4. Rice CM

    (1993)

    Cleavage at a novel site in the NS4A region by the yellow fever virus NS2B-3 proteinase is a prerequisite for processing at the downstream 4A/4B signalase site

    Journal of Virology 67:2327–2335.

    • PubMed
    • Google Scholar
18. 1. Lin DL
    2. Inoue T
    3. Chen YJ
    4. Chang A
    5. Tsai B
    6. Tai AW

    (2019) The ER membrane protein complex promotes biogenesis of dengue and zika virus Non-structural Multi-pass transmembrane proteins to support infection

    Cell Reports 27:1666–1674.

    https://doi.org/10.1016/j.celrep.2019.04.051

    • PubMed
    • Google Scholar
19. 1. Ma H
    2. Dang Y
    3. Wu Y
    4. Jia G
    5. Anaya E
    6. Zhang J
    7. Abraham S
    8. Choi JG
    9. Shi G
    10. Qi L
    11. Manjunath N
    12. Wu H

    (2015) A CRISPR-Based screen identifies genes essential for West-Nile-Virus-Induced cell death

    Cell Reports 12:673–683.

    https://doi.org/10.1016/j.celrep.2015.06.049

    • PubMed
    • Google Scholar
20. 1. Marceau CD
    2. Puschnik AS
    3. Majzoub K
    4. Ooi YS
    5. Brewer SM
    6. Fuchs G
    7. Swaminathan K
    8. Mata MA
    9. Elias JE
    10. Sarnow P
    11. Carette JE

    (2016) Genetic dissection of Flaviviridae host factors through genome-scale CRISPR screens

    Nature 535:159–163.

    https://doi.org/10.1038/nature18631

    • PubMed
    • Google Scholar
21. 1. Miller S
    2. Sparacio S
    3. Bartenschlager R

    (2006) Subcellular localization and membrane topology of the dengue virus type 2 Non-structural protein 4B

    Journal of Biological Chemistry 281:8854–8863.

    https://doi.org/10.1074/jbc.M512697200

    • PubMed
    • Google Scholar
22. 1. Miller S
    2. Kastner S
    3. Krijnse-Locker J
    4. Bühler S
    5. Bartenschlager R

    (2007) The non-structural protein 4A of dengue virus is an integral membrane protein inducing membrane alterations in a 2K-regulated manner

    Journal of Biological Chemistry 282:8873–8882.

    https://doi.org/10.1074/jbc.M609919200

    • PubMed
    • Google Scholar
23. 1. Muñoz-Jordán JL
    2. Laurent-Rolle M
    3. Ashour J
    4. Martínez-Sobrido L
    5. Ashok M
    6. Lipkin WI
    7. García-Sastre A

    (2005) Inhibition of alpha/beta interferon signaling by the NS4B protein of flaviviruses

    Journal of Virology 79:8004–8013.

    https://doi.org/10.1128/JVI.79.13.8004-8013.2005

    • PubMed
    • Google Scholar
24. 1. Neufeldt CJ
    2. Cortese M
    3. Acosta EG
    4. Bartenschlager R

    (2018) Rewiring cellular networks by members of the Flaviviridae family

    Nature Reviews Microbiology 16:125–142.

    https://doi.org/10.1038/nrmicro.2017.170

    • PubMed
    • Google Scholar
25. 1. Olzmann JA
    2. Kopito RR
    3. Christianson JC

    (2013) The mammalian endoplasmic reticulum-associated degradation system

    Cold Spring Harbor Perspectives in Biology 5:a013185.

    https://doi.org/10.1101/cshperspect.a013185

    • PubMed
    • Google Scholar
26. 1. Peña J
    2. Harris E

    (2011) Dengue virus modulates the unfolded protein response in a time-dependent manner

    Journal of Biological Chemistry 286:14226–14236.

    https://doi.org/10.1074/jbc.M111.222703

    • PubMed
    • Google Scholar
27. 1. Płaszczyca A
    2. Scaturro P
    3. Neufeldt CJ
    4. Cortese M
    5. Cerikan B
    6. Ferla S
    7. Brancale A
    8. Pichlmair A
    9. Bartenschlager R

    (2019) A novel interaction between dengue virus nonstructural protein 1 and the NS4A-2K-4B precursor is required for viral RNA replication but not for formation of the membranous replication organelle

    PLOS Pathogens 15:e1007736.

    https://doi.org/10.1371/journal.ppat.1007736

    • PubMed
    • Google Scholar
28. 1. Puschnik AS
    2. Marceau CD
    3. Ooi YS
    4. Majzoub K
    5. Rinis N
    6. Contessa JN
    7. Carette JE

    (2017) A Small-Molecule Oligosaccharyltransferase Inhibitor with Pan-flaviviral Activity

    Cell Reports 21:3032–3039.

    https://doi.org/10.1016/j.celrep.2017.11.054

    • PubMed
    • Google Scholar
29. 1. Reid DW
    2. Campos RK
    3. Child JR
    4. Zheng T
    5. Chan KWK
    6. Bradrick SS
    7. Vasudevan SG
    8. Garcia-Blanco MA
    9. Nicchitta CV

    (2018) Dengue virus selectively annexes endoplasmic Reticulum-Associated translation machinery as a strategy for Co-opting host cell protein synthesis

    Journal of Virology 92:e01766.

    https://doi.org/10.1128/JVI.01766-17

    • PubMed
    • Google Scholar
30. 1. Roosendaal J
    2. Westaway EG
    3. Khromykh A
    4. Mackenzie JM

    (2006) Regulated cleavages at the West Nile virus NS4A-2K-NS4B junctions play a major role in rearranging cytoplasmic membranes and Golgi trafficking of the NS4A protein

    Journal of Virology 80:4623–4632.

    https://doi.org/10.1128/JVI.80.9.4623-4632.2006

    • PubMed
    • Google Scholar
31. 1. Savidis G
    2. McDougall WM
    3. Meraner P
    4. Perreira JM
    5. Portmann JM
    6. Trincucci G
    7. John SP
    8. Aker AM
    9. Renzette N
    10. Robbins DR
    11. Guo Z
    12. Green S
    13. Kowalik TF
    14. Brass AL

    (2016) Identification of zika virus and dengue virus dependency factors using functional genomics

    Cell Reports 16:232–246.

    https://doi.org/10.1016/j.celrep.2016.06.028

    • PubMed
    • Google Scholar
32. 1. Senft D
    2. Ronai ZA

    (2015) UPR, autophagy, and mitochondria crosstalk underlies the ER stress response

    Trends in Biochemical Sciences 40:141–148.

    https://doi.org/10.1016/j.tibs.2015.01.002

    • PubMed
    • Google Scholar
33. 1. Sessions OM
    2. Barrows NJ
    3. Souza-Neto JA
    4. Robinson TJ
    5. Hershey CL
    6. Rodgers MA
    7. Ramirez JL
    8. Dimopoulos G
    9. Yang PL
    10. Pearson JL
    11. Garcia-Blanco MA

    (2009) Discovery of insect and human dengue virus host factors

    Nature 458:1047–1050.

    https://doi.org/10.1038/nature07967

    • PubMed
    • Google Scholar
34. 1. Shao S
    2. Hegde RS

    (2011) Membrane protein insertion at the endoplasmic reticulum

    Annual Review of Cell and Developmental Biology 27:25–56.

    https://doi.org/10.1146/annurev-cellbio-092910-154125

    • PubMed
    • Google Scholar
35. 1. Shurtleff MJ
    2. Itzhak DN
    3. Hussmann JA
    4. Schirle Oakdale NT
    5. Costa EA
    6. Jonikas M
    7. Weibezahn J
    8. Popova KD
    9. Jan CH
    10. Sinitcyn P
    11. Vembar SS
    12. Hernandez H
    13. Cox J
    14. Burlingame AL
    15. Brodsky JL
    16. Frost A
    17. Borner GH
    18. Weissman JS

    (2018) The ER membrane protein complex interacts cotranslationally to enable biogenesis of multipass membrane proteins

    eLife 7:e37018.

    https://doi.org/10.7554/eLife.37018

    • PubMed
    • Google Scholar
36. 1. Su HL
    2. Liao CL
    3. Lin YL

    (2002) Japanese encephalitis virus infection initiates endoplasmic reticulum stress and an unfolded protein response

    Journal of Virology 76:4162–4171.

    https://doi.org/10.1128/jvi.76.9.4162-4171.2002

    • PubMed
    • Google Scholar
37. 1. Volkmar N
    2. Thezenas ML
    3. Louie SM
    4. Juszkiewicz S
    5. Nomura DK
    6. Hegde RS
    7. Kessler BM
    8. Christianson JC

    (2019) The ER membrane protein complex promotes biogenesis of sterol-related enzymes maintaining cholesterol homeostasis

    Journal of Cell Science 132:jcs223453.

    https://doi.org/10.1242/jcs.223453

    • PubMed
    • Google Scholar
38. 1. Zhang R
    2. Miner JJ
    3. Gorman MJ
    4. Rausch K
    5. Ramage H
    6. White JP
    7. Zuiani A
    8. Zhang P
    9. Fernandez E
    10. Zhang Q
    11. Dowd KA
    12. Pierson TC
    13. Cherry S
    14. Diamond MS

    (2016) A CRISPR screen defines a signal peptide processing pathway required by flaviviruses

    Nature 535:164–168.

    https://doi.org/10.1038/nature18625

    • PubMed
    • Google Scholar
39. 1. Zou J
    2. Lee leT
    3. Wang QY
    4. Xie X
    5. Lu S
    6. Yau YH
    7. Yuan Z
    8. Geifman Shochat S
    9. Kang C
    10. Lescar J
    11. Shi PY

    (2015) Mapping the interactions between the NS4B and NS3 proteins of dengue virus

    Journal of Virology 89:3471–3483.

    https://doi.org/10.1128/JVI.03454-14

    • PubMed
    • Google Scholar

Author details

1. Ashley M Ngo

   Chan Zuckerberg Biohub, San Francisco, United States

   Contribution

   Formal analysis, Validation, Investigation, Methodology, Writing—review and editing

   Competing interests

   No competing interests declared

2. Matthew J Shurtleff

   Department of Cellular and Molecular Pharmacology, University of California, San Francisco, San Francisco, United States

   Contribution

   Formal analysis, Funding acquisition, Validation, Investigation, Methodology, Writing—review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0001-9846-3051

3. Katerina D Popova

   Department of Cellular and Molecular Pharmacology, University of California, San Francisco, San Francisco, United States

   Contribution

   Formal analysis, Validation, Investigation, Methodology, Writing—review and editing

   Competing interests

   No competing interests declared

4. Jessie Kulsuptrakul

   Chan Zuckerberg Biohub, San Francisco, United States

   Contribution

   Validation, Investigation

   Competing interests

   No competing interests declared

5. Jonathan S Weissman

   1. Department of Cellular and Molecular Pharmacology, University of California, San Francisco, San Francisco, United States
   2. Howard Hughes Medical Institute, University of California, San Francisco, San Francisco, United States

   Contribution

   Conceptualization, Resources, Supervision, Funding acquisition, Writing—review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0003-2445-670X

6. Andreas S Puschnik

   Chan Zuckerberg Biohub, San Francisco, United States

   Contribution

   Conceptualization, Resources, Formal analysis, Supervision, Funding acquisition, Validation, Investigation, Visualization, Methodology, Writing—original draft, Writing—review and editing

   For correspondence

   andreas.puschnik@czbiohub.org

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-9605-9458

• 3,797

  views

• 622

  downloads

• 33

  citations

Views, downloads and citations are aggregated across all versions of this paper published by eLife.