Abstract
Viral pathogens are an ongoing threat to public health worldwide. Analysing their dependence on host biosynthetic pathways could lead to effective antiviral therapies1. Here we integrate proteomic analyses of polysomes with functional genomics and pharmacological interventions to define how enteroviruses and flaviviruses remodel host polysomes to synthesize viral proteins and disable host protein production. We find that infection with polio, dengue or Zika virus markedly modifies polysome composition, without major changes to core ribosome stoichiometry. These viruses use different strategies to evict a common set of translation initiation and RNA surveillance factors from polysomes while recruiting host machineries that are specifically required for viral biogenesis. Targeting these specialized viral polysomes could provide a new approach for antiviral interventions. For example, we find that both Zika and dengue use the collagen proline hydroxylation machinery to mediate cotranslational modification of conserved proline residues in the viral polyprotein. Genetic or pharmacological inhibition of proline hydroxylation impairs nascent viral polyprotein folding and induces its aggregation and degradation. Notably, such interventions prevent viral polysome remodelling and lower virus production. Our findings delineate the modular nature of polysome specialization at the virus–host interface and establish a powerful strategy to identify targets for selective antiviral interventions.
This is a preview of subscription content, access via your institution
Access options
Access Nature and 54 other Nature Portfolio journals
Get Nature+, our best-value online-access subscription
$29.99 / 30 days
cancel any time
Subscribe to this journal
Receive 51 print issues and online access
$199.00 per year
only $3.90 per issue
Buy this article
- Purchase on Springer Link
- Instant access to full article PDF
Prices may be subject to local taxes which are calculated during checkout
Similar content being viewed by others
Data availability
The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium through the PRIDE45 partner repository with the dataset identifier PXD024546.
Change history
21 October 2021
A Correction to this paper has been published: https://doi.org/10.1038/s41586-021-04102-0
References
Kumar, N. et al. Host-directed antiviral therapy. Clin. Microbiol. Rev. 33, e00168 (2020).
Stern-Ginossar, N., Thompson, S. R., Mathews, M. B. & Mohr, I. Translational control in virus-infected cells. Cold Spring Harb. Perspect. Biol. 11, a033001 (2019).
Nicholson, B. L. & White, K. A. Functional long-range RNA–RNA interactions in positive-strand RNA viruses. Nat. Rev. Microbiol. 12, 493–504 (2014).
Aviner, R. & Frydman, J. Proteostasis in viral infection: unfolding the complex virus–chaperone interplay. Cold Spring Harb. Perspect. Biol. 12, a034090 (2020).
Jaafar, Z. A. & Kieft, J. S. Viral RNA structure-based strategies to manipulate translation. Nat. Rev. Microbiol. 17, 110–123 (2019).
Mizuno, C. M. et al. Numerous cultivated and uncultivated viruses encode ribosomal proteins. Nat. Commun. 10, 752 (2019).
Macejak, D. G. & Sarnow, P. Association of heat shock protein 70 with enterovirus capsid precursor P1 in infected human cells. J. Virol. 66, 1520–1527 (1992).
Yang, W. et al. G3BP1 inhibits RNA virus replication by positively regulating RIG-I-mediated cellular antiviral response. Cell Death Dis. 10, 946 (2019).
Limjindaporn, T. et al. Interaction of dengue virus envelope protein with endoplasmic reticulum-resident chaperones facilitates dengue virus production. Biochem. Biophys. Res. Commun. 379, 196–200 (2009).
Sweeney, T. R., Abaeva, I. S., Pestova, T. V. & Hellen, C. U. T. The mechanism of translation initiation on type 1 picornavirus IRESs. EMBO J. 33, 76–92 (2014).
Lee, K.-M., Chen, C.-J. & Shih, S.-R. Regulation mechanisms of viral IRES-driven translation. Trends Microbiol. 25, 546–561 (2017).
Zhang, Y., Gao, W., Li, J., Wu, W. & Jiu, Y. The role of host cytoskeleton in flavivirus infection. Virol. Sin. 34, 30–41 (2019).
Dougherty, J., Tsai, W.-C. & Lloyd, R. Multiple poliovirus proteins repress cytoplasmic RNA granules. Viruses 7, 6127–6140 (2015).
Michalski, D. et al. Zika virus noncoding sfRNAs sequester multiple host-derived RNA-binding proteins and modulate mRNA decay and splicing during infection. J. Biol. Chem. 294, 16282–16296 (2019).
Moon, S. L. et al. Flavivirus sfRNA suppresses antiviral RNA interference in cultured cells and mosquitoes and directly interacts with the RNAi machinery. Virology 485, 322–329 (2015).
Ooi, Y. S. et al. An RNA-centric dissection of host complexes controlling flavivirus infection. Nat. Microbiol. 4, 2369–2382 (2019).
Lenarcic, E. M., Landry, D. M., Greco, T. M., Cristea, I. M. & Thompson, S. R. Thiouracil cross-linking mass spectrometry: a cell-based method to identify host factors involved in viral amplification. J. Virol. 87, 8697–8712 (2013).
Pallarés, H. M. et al. Zika virus subgenomic flavivirus RNA generation requires cooperativity between duplicated RNA structures that are essential for productive infection in human cells. J. Virol. 94, e00343 (2020).
Zeng, J. et al. The Zika virus capsid disrupts corticogenesis by suppressing Dicer activity and miRNA biogenesis. Cell Stem Cell 27, 618–632 (2020).
Rodríguez Pulido, M., Serrano, P., Sáiz, M. & Martínez-Salas, E. Foot-and-mouth disease virus infection induces proteolytic cleavage of PTB, eIF3a,b, and PABP RNA-binding proteins. Virology 364, 466–474 (2007).
Hunt, S. L., Hsuan, J. J., Totty, N. & Jackson, R. J. unr, a cellular cytoplasmic RNA-binding protein with five cold-shock domains, is required for internal initiation of translation of human rhinovirus RNA. Genes Dev. 13, 437–448 (1999).
Andreev, D. E. et al. Glycyl-tRNA synthetase specifically binds to the poliovirus IRES to activate translation initiation. Nucleic Acids Res. 40, 5602–5614 (2012).
Iwasaki, S., Floor, S. N. & Ingolia, N. T. Rocaglates convert DEAD-box protein eIF4A into a sequence-selective translational repressor. Nature 534, 558–561 (2016).
Li-Weber, M. Molecular mechanisms and anti-cancer aspects of the medicinal phytochemicals rocaglamides (=flavaglines). Int. J. Cancer 137, 1791–1799 (2015).
Yang, S. N. Y. et al. RK-33 is a broad-spectrum antiviral agent that targets DEAD-box RNA helicase DDX3X. Cells 9, 170 (2020).
Kramer, G., Shiber, A. & Bukau, B. Mechanisms of cotranslational maturation of newly synthesized proteins. Annu. Rev. Biochem. 88, 337–364 (2019).
Gamerdinger, M., Hanebuth, M. A., Frickey, T. & Deuerling, E. The principle of antagonism ensures protein targeting specificity at the endoplasmic reticulum. Science 348, 201–207 (2015).
Taguwa, S. et al. Zika virus dependence on host Hsp70 provides a protective strategy against infection and disease. Cell Rep. 26, 906–920 (2019).
Taguwa, S. et al. Defining Hsp70 subnetworks in dengue virus replication reveals key vulnerability in flavivirus infection. Cell 163, 1108–1123 (2015).
Kirk, T. Z., Evans, J. S. & Veis, A. Biosynthesis of type I procollagen. Characterization of the distribution of chain sizes and extent of hydroxylation of polysome-associated pro-alpha-chains. J. Biol. Chem. 262, 5540–5545 (1987).
DiChiara, A. S. et al. Mapping and exploring the collagen-I proteostasis network. ACS Chem. Biol. 11, 1408–1421 (2016).
Ishikawa, Y. & Bächinger, H. P. A molecular ensemble in the rER for procollagen maturation. Biochim. Biophys. Acta 1833, 2479–2491 (2013).
Ueno, T. et al. Enhancement of procollagen biosynthesis by p180 through augmented ribosome association on the endoplasmic reticulum in response to stimulated secretion. J. Biol. Chem. 285, 29941–29950 (2010).
Li, Y., Li, Q., Wong, Y. L., Liew, L. S. Y. & Kang, C. Membrane topology of NS2B of dengue virus revealed by NMR spectroscopy. Biochim. Biophys. Acta 1848, 2244–2252 (2015).
Bretscher, L. E., Jenkins, C. L., Taylor, K. M., DeRider, M. L. & Raines, R. T. Conformational stability of collagen relies on a stereoelectronic effect. J. Am. Chem. Soc. 123, 777–778 (2001).
Naduthambi, D. & Zondlo, N. J. Stereoelectronic tuning of the structure and stability of the trp cage miniprotein. J. Am. Chem. Soc. 128, 12430–12431 (2006).
Guo, J. et al. pVHL suppresses kinase activity of Akt in a proline-hydroxylation-dependent manner. Science 353, 929–932 (2016).
Burrill, C. P., Strings, V. R. & Andino, R. Poliovirus: generation, quantification, propagation, purification, and storage. Curr. Protoc. Microbiol. 128, 15H.1.1–15H.1.27 (2013).
Kinney, R. M. et al. Construction of infectious cDNA clones for dengue 2 virus: strain 16681 and its attenuated vaccine derivative, strain PDK-53. Virology 230, 300–308 (1997).
Cox, J. & Mann, M. MaxQuant enables high peptide identification rates, individualized p.p.b.-range mass accuracies and proteome-wide protein quantification. Nat. Biotechnol. 26, 1367–1372 (2008).
Cox, J. et al. Andromeda: a peptide search engine integrated into the MaxQuant environment. J. Proteome Res. 10, 1794–1805 (2011).
Shah, P. S. et al. Comparative flavivirus–host protein interaction mapping reveals mechanisms of dengue and Zika virus pathogenesis. Cell 175, 1931–1945 (2018).
Gamarnik, A. V. & Andino, R. Switch from translation to RNA replication in a positive-stranded RNA virus. Genes Dev. 12, 2293–2304 (1998).
Di Veroli, G. Y. et al. An automated fitting procedure and software for dose–response curves with multiphasic features. Sci. Rep. 5, 14701 (2015).
Perez-Riverol, Y. et al. The PRIDE database and related tools and resources in 2019: improving support for quantification data. Nucleic Acids Res. 47, D442–D450 (2019).
Cox, J. & Mann, M. 1D and 2D annotation enrichment: a statistical method integrating quantitative proteomics with complementary high-throughput data. BMC Bioinf. 13, S12 (2012).
Imami, K. et al. Phosphorylation of the ribosomal protein RPL12/uL11 affects translation during mitosis. Mol. Cell 72, 84–98 (2018).
Simsek, D. et al. The mammalian ribo-interactome reveals ribosome functional diversity and heterogeneity. Cell 169, 1051–1065 (2017).
Fontaine, K. A. et al. The cellular NMD pathway restricts Zika virus infection and is targeted by the viral capsid protein. MBio 9, e02126-18 (2018).
Göertz, G. P. et al. Subgenomic flavivirus RNA binds the mosquito DEAD/H-box helicase ME31B and determines Zika virus transmission by Aedes aegypti. Proc. Natl Acad. Sci. USA 116, 19136–19144 (2019).
Kanodia, P. et al. A rapid and simple quantitative method for specific detection of smaller coterminal RNA by PCR (DeSCo-PCR): application to the detection of viral subgenomic RNAs. RNA 26, 888–901 (2020).
Lloyd, R. E. Nuclear proteins hijacked by mammalian cytoplasmic plus strand RNA viruses. Virology 479–480, 457–474 (2015).
Jagdeo, J. M. et al. Heterogeneous nuclear ribonucleoprotein M facilitates enterovirus infection. J. Virol. 89, 7064–7078 (2015).
Gustin, K. E. & Sarnow, P. Effects of poliovirus infection on nucleo-cytoplasmic trafficking and nuclear pore complex composition. EMBO J. 20, 240–249 (2001).
Silvera, D., Gamarnik, A. V. & Andino, R. The N-terminal K homology domain of the poly(rC)-binding protein is a major determinant for binding to the poliovirus 5′-untranslated region and acts as an inhibitor of viral translation. J. Biol. Chem. 274, 38163–38170 (1999).
Dave, P. et al. Strand-specific affinity of host factor hnRNP C1/C2 guides positive to negative-strand ratio in Coxsackievirus B3 infection. RNA Biol. 16, 1286–1299 (2019).
Acknowledgements
This research was supported by fellowships from Rothschild-Yad Hanadiv Foundation, the European Molecular Biology Organization (EMBO ALTF 289-2015) and the Human Frontier Science Program (HFSP LT000050-2016) to R. Aviner; National Institutes of Health (NIH) grants AI127447 and GM056433 to J.F.; NIH grants R01 AI36178, AI40085, P01 AI091575); the University of California (California Center for Antiviral Drug Discovery; CCADD); and DoD-DARPA Prophecy to R. Andino. J.F. is a CZ Biohub Investigator.
Author information
Authors and Affiliations
Contributions
R. Aviner, R. Andino and J.F. designed the study. R. Aviner carried out experiments, analysed data and performed statistical analyses. K.H.L. performed LC–MS/MS data acquisition. R. Aviner, R. Andino and J.F. wrote the manuscript, and all authors approved the manuscript.
Corresponding authors
Ethics declarations
Competing interests
All authors declare no competing interests.
Additional information
Peer review information Nature thanks Andreas Pichlmair, Alexander Khromykh and the anonymous, reviewer(s) for their contribution to the peer review of this work. Peer reviewer reports are available.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Extended data figures and tables
Extended Data Fig. 1 Suppression of global translation is associated with preferential synthesis of viral proteins.
a, b, Schematics of vRNA and polyprotein organization for poliovirus (a), ZIKV, and DENV (b). IRES, internal ribosome entry site; ORF, open reading frame. c, Poliovirus, ZIKV and DENV infection is associated with shut-off of global translation and continued synthesis of viral proteins. Top, immunoblots of cells infected with poliovirus, ZIKV or DENV (MOI = 5) and treated with puromycin for 15 min at the indicated time points, to label nascent chains. Results are representative of 4 independent repeats. Bottom, time course of infectious particle production detected by plaque assays (grey) and densitometry of viral proteins (black) or puromycin-labelled nascent chains (pink). Data are means ± s.d. d, Related to Fig. 1b. Polysome profile analysis (top) and immunoblots (bottom) of infected and control cells subjected to sucrose gradient fractionation followed by either qPCR analysis of negative strand viral RNA or puromycin treatment of gradient fractions. Results are representative of 3 independent repeats. e, Cells infected as in d were lysed in buffer containing EDTA and no Mg2+ and subjected to polysome profile analysis and qPCR of positive strand viral RNA. Results are representative of 2 independent repeats.
Extended Data Fig. 2 Overview of polysome mass spectrometry in infected and control cells.
a, Pairwise comparisons of two biological replicates of polysome MS from mock-infected cells. Each dot represents the level of a single protein as detected by MS. iBAQ, intensity based absolute quantification. r, Spearman correlation coefficient. b, iBAQ-normalized MS intensities of proteins associated with the indicated cellular complexes and organelles, plotted as percent of total, for either lysate (input) or polysome MS analyses. Data are means ± s.d. n = 2 and 3 for Lysate(input) and Polysomes, respectively. c, Venn bar diagram of proteins detected by polysome MS across all conditions and time points. d, Distribution of ribosomal protein intensities (grey) and median intensities of viral polyproteins (pink), as detected by polysome MS. e, To identify co-sedimenting non-polysomal proteins, we performed polysome MS on uninfected cells pretreated with 1 µM puromycin for 3 min. Puromycin treated and untreated controls were subjected to polysome profile analysis on sucrose gradients, in triplicates. Dashes lines indicate fractions pooled for MS analysis. Because samples are normalized for total protein content, non-polysomal proteins, whose migration in the gradient is unaffected by puromycin, will appear to be enriched, whereas polysomal proteins, which will shift to lighter fractions in a puromycin dependent manner, will appear depleted or unchanged. To determine fold-changes (FC), we subtracted the means of each log2-transformed iBAQ value in puromycin versus control. FC distribution is shown in green for ribosomal proteins and grey for non-ribosomal proteins (histogram). FC values were then used for 1D annotation enrichment analysis46, where the FC distribution in each category is compared to the global distribution using a two-sided Wilcoxon-Mann-Whitney test. FDR values for each category are plotted as a function of mean FC (scatterplot). f, Most proteins detected by polysome MS are either previously identified ribosome-interacting partners47,48 or known to be involved in protein biogenesis.
Extended Data Fig. 3 Polysome composition is gradually and extensively remodelled during infection.
a, Infection is not associated with changes to the core ribosome composition. Volcano plots of median-adjusted iBAQ intensities of ribosomal proteins measured by polysome MS in infected and uninfected cells. Two-sided Student’s t-test, FDR < 0.01, S0 = 1. b, Differential detection of peptides mapping to ribosomal proteins may reflect changes in post-translational modifications. Same analysis as in a, only at the peptide level. Red, ribosomal peptides with statistically significant increases in polysome MS of DENV at 60 hpi versus mock. Bottom bar graph, number of changing peptides mapping to each indicated protein, plotted as percent of all detected peptides that map to that protein. All changing peptides were found to harbor at least one—and most more than two—potential PTM sites based on Phosphosite data. c, Pairwise comparisons of non-ribosomal proteins detected by polysome MS (left) and total proteome analysis of matching lysate samples (right) in infected cells (y axis) versus uninfected control (x axis). r, Spearman correlation coefficient. d, Euclidean distance hierarchical clustering of Spearman correlation coefficients across all polysome MS samples. Highest correlations are in white, lowest in red. e, Protein rank plots for total proteome of infected cells. Viral polyproteins are indicated in colour. f, Related to Fig. 1d, line plots show medians±interquartile range of fold change from mock in log2-transformed iBAQ intensities, based on clustering shown in Fig. 1d.
Extended Data Fig. 4 Differentially associated polysome interactors in mock versus the terminal time point for each virus.
a–c, Volcano plots comparing mock polysome MS with that of poliovirus (a), ZIKV (b) and DENV (c) at 4, 36 and 60 h, respectively. Two-sided Student’s t-test, FDR < 0.01, S0 = 1.
Extended Data Fig. 5 Antiviral host factors are evicted from polysomes by distinct yet convergent mechanisms.
a, Major RNA QC factors are strongly evicted from polysomes by all three viruses. STRING protein-protein interaction network for host factors involved in RNA surveillance and quality control depleted from (blue) or recruited to (pink) polysomes during infection. Differentially associated proteins (ANOVA FDR < 0.01 across all polysome MS samples) were grouped based on annotations in Fig. 1g. Each node is divided in three to reflect individual changes for each virus and coloured based on changes in polysome association from Mock. Node size represents relative abundance, which is the maximum iBAQ intensity measured for each indicated protein across all polysome MS samples. b, Line plots of the change over time in polysome association of G3BP complexes (top) and UPF1/MOV10 (bottom) during infection. Each line represents a single replicate of polysome MS time course. Dotted grey line, translation shut-off. c, G3BP1 is cleaved during poliovirus but not ZIKV or DENV infection. Immunoblot analysis of whole-cell lysates from cells infected with poliovirus, ZIKV or DENV (MOI = 5). Time courses are 1–6 h for poliovirus and 12–72 h for ZIKV and DENV. Arrowhead marks the position of G3BP1 cleavage fragment. Results are representative of 2 independent repeats. d, Volcano plot comparing polysome interactors in cells infected with poliovirus for 4h versus mock. Host proteins known to be cleaved by poliovirus protease are indicated. e, Polysome MS captures most direct interactors of vRNA. Venn diagram of proteins detected by polysome MS and those detected by pull-down of poliovirus, ZIKV and DENV vRNA from the indicated studies17,18. f, High affinity flavivirus RNA interactors are depleted from polysomes during infection. Violin plots of proteins detected by both vRNA pulldowns and polysome MS. Proteins were divided into two groups based on their polysome association in this study: depleted or recruited by more than 2 folds after infection with ZIKV or DENV at 36 or 60 h, compared to mock. y-axis, protein abundance in vRNA pulldown-MS. P, Mann Whitney U test p-value. For each violin plot, white circles indicate the median of each distribution, black rectangles indicate interquartile range and vertical lines indicate the 95% confidence intervals. g, Polysome interactors depleted during ZIKV infection consist of high affinity ZIKV 3’UTR or capsid interactors. Pairwise comparison of the fold change in polysomes interactors (x axis) versus their protein abundance in ZIKV vRNA pulldown-MS (y axis). Known ZIKV sfRNA and capsid interactors from14,15,49,50 are coloured blue and purple, respectively. h, Schematic of the production of subgenomic flaviviral RNA (sfRNA) by limited 5′-to-3′ XRN1 exonucleolytic cleavage of full-length genomes. i, j, Most ZIKV sfRNA does not associate with polysomes. Cells were infected with ZIKV (MOI = 5) for 24 h and lysates were fractionated on sucrose gradients followed by: qPCR analysis using primers that amplify either full-length genome (pink) or 3’UTR sequence present in both full-length genome and sfRNA (blue). Values for each fraction are plotted as percent of total across the entire gradient. Data are means ± s.d (n = 3) (i); or PCR analysis for specific detection of either gRNA or sfRNA using blocking oligonucleotides (j) 51. k, Most ZIKV capsid does not associate with polysomes. Levels of ZIKV capsid detected by MS analysis of either polysomes or lysate input samples from cells infected with ZIKV for 36 h. Values are sum of peptides mapping to the capsid region of the viral polyprotein divided by the sum of peptides mapping to ribosomal proteins, with average ratios in the lysate set as 1. Data are means ± s.d (n = 3). P, two-tailed Student’s t-test p-value. l, A similar set of RNA QC factors is evicted from polysomes by either viral protease cleavage (poliovirus) or sfRNA/capsid sequestration (ZIKV) to prevent translational silencing of gRNA.
Extended Data Fig. 6 Rearrangement of translation initiation complexes during viral infection.
a, Line plots of the change over time in polysome association of initiation factors 3 (EIF3), 4G1 (EIF4G1), 4A1 (EIF4A1) and poly-A-binding protein (PABPC1). Included in the quantification of EIF3 are the 8 subunits of the core octamer complex. Data are means ± s.d. (n = 3) Asterisk denotes statistically significant differences in EIF4A1 polysome association between 24 and 36 hpi for ZIKV and 48 to 60 hpi for DENV (p=0.0037 and 0.0071, respectively, two-tailed Student’s t-test). Dotted grey line, translation shut-off. b, EIF4G1 c-terminal fragment, generated by poliovirus protease cleavage to inhibit cap-dependent translation while promoting polio-virus-specific translation initiation21, is retained on polysomes during poliovirus infection. Mapping of polysome-associated EIF4G1 peptides, as detected by polysome MS of cells infected with poliovirus, ZIKV or DENV for 4, 36 and 60 h, respectively, plotted as cumulative fraction from N- to C-termini. Protein–protein interaction domains are shown below. c, EIF3, a multisubunit complex, forms two distinct subcomplexes in virus-infected cells. Line plots of the change over time in polysome association of EIF3 subunits, divided into the octamer (top) and yeast-like core (bottom) subcomplexes, coloured blue and cyan on the structure of a preinitiating 48S ribosome (right; PDB 6FEC). Each line represents a single replicate of polysome MS time course. Dotted grey line, translation shut-off. d, Toxicity of RocA at 36 h. Data are means ± s.d. (n = 3). e, Heat map of the median fold change from mock in polysome association of RNA helicases during infection with each of the three viruses.
Extended Data Fig. 7 RNA-binding proteins control the fate of polysome-associated poliovirus vRNA.
Many RNA-binding proteins have been implicated in regulating poliovirus translation and replication by binding and restructuring its 5′ UTR52. a, Multiple RNA-binding proteins are uniquely recruited to polysomes by poliovirus. STRING protein–protein interaction network for host factors involved in alternative splicing and mRNA processing depleted from (blue) or recruited to (pink) polysomes during infection. Differentially associated proteins (ANOVA FDR<0.01 across all polysome MS samples) were grouped based on annotations in Fig. 1g. Each node is divided into three to reflect individual changes for each virus and coloured based on change in polysome association from Mock. Node size represents relative abundance, which is the maximum iBAQ intensity measured for each indicated protein across all polysome MS samples. b, To identify novel RNA interactors regulating poliovirus biogenesis, we used the temporal pattern of change in CSDE1/Unr, a bonafide poliovirus ITAF21, across all polysome MS samples. We then generated a list of the top 15 polysome-associated RNA-binding proteins based on similarity in temporal patterns using Euclidian distances. Shown are medians±interquartile range of the top 15 interactors identified. Specific protein names are noted on the right. Dotted grey line, translation shut-off. Seven of the top 15 hits are known interactors, including bona fide ITAFs e.g. PCBP2, STRAP/UnrIP21 and glycyl-tRNA synthetase (GARS)22. Recruitment of additional factors that mirror the same temporal pattern, for example, HNRNPR, SRSF10 and TRA2B, is likely to reflect previously unknown involvement in poliovirus biogenesis. c, Line plots of the change over time in polysome association of known poliovirus 5′ UTR binders, plotted as fold change from mock. Each line is interquartile range. Dotted grey line, translation shut-off. Green, factors that stimulate poliovirus translation; pink, factors that inhibit poliovirus translation and stimulate replication; grey, alternative splicing factor cleaved by poliovirus protease to facilitate infection53. Values for GARS were recalculated manually based on unique peptides only, as many peptides also map to IARS. Cytosolic ITAFs are recruited earlier than nuclear factors (3 versus 4 hpi), consistent with a later disruption of nuclear pores by poliovirus protease54. This coincides with cleavage of PCBP2, which converts it from activator to inhibitor of translation55, as well as increased association of nuclear HNRNPC, an essential component of viral replication that competes with and displaces ITAFs from the 5′ UTR of picornaviruses54,56. d, Poliovirus translation peaks by 3 hpi while replication lags by about an hour. (+)vRNA and 3CD polymerase were quantified by qPCR and immunoblot analysis, respectively, in cells infected with poliovirus (MOI = 5) for the indicated times. Shown are representative blots from 2 independent repeats, and qPCR values of (+)vRNA normalized to GAPDH (pink) and 3CD densitometry normalized to actin (black) (n = 1). e, Schematic of the change in poliovirus 5′ UTR-binding proteins upon transition from translation to replication. Early during translation, viral genome binds cytosolic ITAFs that stimulate its translation. Later during infection, poliovirus 3C protease cleaves nuclear pore proteins, and alternative splicing factors are released into the cytoplasm to inhibit new rounds of translation initiation on polysome-associated vRNA. This releases vRNA from polysomes for use as a template in replication. Thus, the observed rearrangement of 5′ UTR-binding proteins between 3 and 4 hpi acts as a timer that shifts viral genomes from translation to replication.
Extended Data Fig. 8 Cytosolic and ER proteostasis of viral nascent chains.
a, Line plots of the change over time in polysome association of cytosolic HSP70 chaperones and co-chaperones. Each line represents a single replicate of polysome MS time course. Dotted grey line, translation shut-off. b, Efficiency of KD in Huh7 cells based on qPCR using primers specific to the indicated transcripts, normalized to GAPDH (n = 2). c, Toxicity of JG40 and JG345 at 36 h. Data are means ± s.d. (n = 3). d, JG345 inhibits poliovirus, ZIKV and DENV, whereas JG40 only inhibits ZIKV and DENV. Virus production by plaque assays of cells infected with poliovirus, ZIKV or DENV (MOI = 0.1) for 5, 24 and 36 h, respectively, in the presence or absence of HSP70 inhibitor, plotted as % of DMSO control. Data are means ± s.d. (n = 3). e, Virus production by plaque assays of cells infected with poliovirus (MOI = 0.1) in the presence or absence of HSP70 inhibitors. Data are means ± s.d. (n = 3). f, Eviction of nascent chain-associated complex (NAC) from polysomes during flavivirus infection is associated with increased binding of ER-resident factors. Line plots of the change over time in polysome association of NAC subunits (NACA and BTF3, blue) and ER-resident factors with GOCC ER lumen annotation (pink). Data are means ± s.d. for NAC or median ± interquartile range for GOCC ER Lumen. Dotted grey line, translation shut-off. g, Cumulative fraction of polysome interactors with GOCC ER Part annotation. P, two-sided Mann Whitney U test p-value. h, Efficiency of KD in Huh7 cells based on qPCR using primers specific to the indicated transcripts, normalized to GAPDH (n = 2). i, Toxicity of indicated compounds at 36 h. Data are means ± s.d. (n = 3).
Extended Data Fig. 9 Collagen prolyl hydroxylases facilitate infection of ZIKV and DENV.
a, Collagen-specific chaperone, but not collagens or an RNA-binding protein required for their translation, is recruited to polysomes by ZIKV and DENV. Line plots of the change over time in polysome association of collagen-specific chaperone (SERPINH1/HSP47), all collagens identified by polysome MS, and collagen translation regulator RRBP1. Each line represents a single replicate of polysome MS time course. Dotted grey line, translation shut-off. b, Related to Fig. 3c. Polysome profiles (top) and immunoblot analysis of gradient fractions treated with puromycin to label nascent chains (bottom). Results are representative of 2 independent repeats. c, P4HA1 co-sedimentation with ZIKV polysomes is disrupted by ribosome runoff. Wild-type cells were infected with ZIKV (MOI = 5) for 24 h then treated with 20 µM harringtonine for 10 min followed by addition of 1 µM puromycin for another 10 min. Lysates were subjected to sucrose gradient fractionation and puromycin labelling. Shown are rRNA absorbance profiles (top) and immunoblots (bottom). Results are representative of 2 independent repeats. d, Efficiency of KD in Huh7 cells based on qPCR using primers specific to the indicated transcripts, normalized to GAPDH. (n = 2). e, f, Time course of virus production by plaque assays in wild-type and cP4H-KD cells for ZIKV (e) and DENV (f) at MOI = 0.1 (n = 3). g, Wild-type or cP4H-KD cells were infected with ZIKV at the indicated MOIs. 200 µM ascorbic acid (VitC) was added at the start of infection and again 12 h later, and virus production at 24 h was monitored using plaque assays. Data are means ± s.d. (n = 3).
Extended Data Fig. 10 Proline hydroxylation affects viral protein stability.
a, Multiple DENV proline residues are hydroxylated cotranslationally by cP4H. Bubble chart shows individual hydroxyproline sites, as detected by polysome MS in wild-type cells infected with DENV for 60 h. Bubble size is proportional to posterior error probability (PEP) of hydroxyproline sites detected by MS. b, Loss of proline hydroxylation prevents full induction of translation shut-off by ZIKV infection. Wild-type and cP4H-KD cells were infected with ZIKV (MOI = 10) for 24 h and metabolically labelled for 1h followed by autoradiography (left) and quantification (right). Data are means ± s.d. (n = 4). P, two-tailed Student’s t-test p-value. c, Proline hydroxylation stabilizes multiple viral proteins during infection. Wild-type or cP4H-KD cells were infected with ZIKV (MOI = 10) and MG132, CHX or MG132+CHX were added at 24 h for additional 12 h prior to immunoblot analysis. Results are representative of 3 independent repeats. d, Lack of proline hydroxylation does not affect NS3 protease activity or NS1 protein stability/oligomerization. Strep-tagged ZIKV proteins were transfected into wild-type or cP4H-KD cells and analyzed by nonreducing immunoblot at 24 h. Representative of 3 independent repeats. e, Proline hydroxylation reduces aggregation of ZIKV and DENV NS2B. Strep-tagged NS2B3 was transfected into wild-type and cP4H-KD cells and MG132 was added at 24 h for additional 12 h, prior to immunoblot analysis. Results are representative of 2 independent repeats. f, Wild-type or cP4H-KD cells were infected with DENV (MOI = 10) and MG132 was added at 36 for additional 12 h prior to immunoblot analysis. Results are representative of 3 independent repeats.
Supplementary information
Supplementary Figure
This file contains the uncropped blots.
Supplementary Table 1
Mass spectrometry data, polysome mass spectrometry and global proteome profiling.
Supplementary Table 2
Polysome mass spectrometry after puromycin pretreatment (related to Extended Data Fig. 2e).
Supplementary Table 3
Ribosomal proteins in polysome mass spectrometry (related to Extended Data Fig. 3a, b).
Supplementary Table 4
Differentially associated polysome interactors (related to Fig. 1d).
Supplementary Table 5
Hydroxyproline sites in mass spectrometry data.
Supplementary Table 6
Oligonucleotides and shRNA.
Rights and permissions
About this article
Cite this article
Aviner, R., Li, K.H., Frydman, J. et al. Cotranslational prolyl hydroxylation is essential for flavivirus biogenesis. Nature 596, 558–564 (2021). https://doi.org/10.1038/s41586-021-03851-2
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1038/s41586-021-03851-2
This article is cited by
Comments
By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.
