key: cord-0306641-sxovb1t2
authors: Sadasivan, Jibin; Vlok, Marli; Wang, Xinying; Nayak, Arabinda; Andino, Raul; Jan, Eric
title: Targeting Nup358/RanBP2 by a viral protein disrupts stress granule formation
date: 2022-05-19
journal: bioRxiv
DOI: 10.1101/2022.05.19.492599
sha: 106133bc0392c104d84c4018a81c9328d68d49ef
doc_id: 306641
cord_uid: sxovb1t2

Viruses have evolved mechanisms to modulate cellular pathways to facilitate infection. One such pathway is the formation of stress granules (SG), which are ribonucleoprotein complexes that assemble during translation inhibition following cellular stress. Inhibition of SG assembly has been observed under numerous virus infections across species, suggesting a conserved fundamental viral strategy. However, the significance of SG modulation during virus infection is not fully understood. The 1A protein encoded by the model dicistrovirus, Cricket Paralysis Virus (CrPV), is a multifunctional protein that can bind to and degrade Ago-2 in an E3 ubiquitin ligase-dependent manner to block the antiviral RNA interference pathway and inhibit SG formation. Moreover, the R146 residue of 1A is necessary for SG inhibition and CrPV infection in both Drosophila S2 cells and adult flies. Here, we uncoupled CrPV-1A’s functions and provide insight into its underlying mechanism for SG inhibition. CrPV-1A mediated inhibition of SGs requires the E3 ubiquitin-ligase binding domain and the R146 residue, but not the Ago-2 binding domain. Wild-type but not mutant CrPV-1A R146A localizes to the nuclear membrane which correlates with nuclear enrichment of poly(A)+ RNA. Transcriptome changes in CrPV-infected cells are dependent on the R146 residue. Finally, Nup358/RanBP2 is targeted and degraded in CrPV-infected cells in an R146-dependent manner and the depletion of Nup358 blocks SG formation. We propose that CrPV utilizes a multiprong strategy whereby the CrPV-1A protein interferes with a nuclear event that contributes to SG inhibition in order to promote infection. AUTHOR SUMMARY Viruses often inhibit a cellular stress response that leads to the accumulation of RNA and protein condensates called stress granules. How this occurs and why this would benefit virus infection are not fully understood. Here, we reveal a viral protein that can block stress granules and identify a key amino acid residue in the protein that inactivates this function. We demonstrate that this viral protein has multiple functions to modulate nuclear events including mRNA export and transcription to regulate stress granule formation. We identify a key host protein that is important for viral protein mediate stress granule inhibition, thus providing mechanistic insights. This study reveals a novel viral strategy in modulating stress granule formation to promote virus infection.

inactivates this function. We demonstrate that this viral protein has multiple functions to 48 modulate nuclear events including mRNA export and transcription to regulate stress 49 granule formation. We identify a key host protein that is important for viral protein 50 mediate stress granule inhibition, thus providing mechanistic insights. This study reveals 51 a novel viral strategy in modulating stress granule formation to promote virus infection.

7 120 deaminase 1 (ADAR1) and cyclic GMP-AMP synthase (cGAS) have been found in SGs, 121 termed antiviral SGs (avSGs)6776, which may act as an antiviral hub to co-ordinate 122 immune responses to limit viral replication (40-43). Influenza A virus RNA and RIG-1 123 have been found in avSGs during infection, which is thought to trigger the RIG-I- suggesting that the SG modulation is linked to a nuclear event(s) (64). In summary,

CrPV-1A is a multifunctional protein that modulates several host cell processes to 174 promote infection. Whether the specific functions of CrPV-1A are mutually exclusive or 175 interdependent have yet to be examined.

In this study, we use overexpression and mutagenesis approaches to uncouple 177 the relationship between the multiple functions of CrPV-1A. We show that CrPV-1A's 178 ability to inhibit SGs is dependent on the BC Box ubiquitin complex-interacting domain 179 and independent of the Ago-2 binding TALOS element. We also demonstrate that with the Cul-2-Rbx1-EloBC complex ( Fig 1A) . We previously showed that expression of similar to that observed when wild-type CrPV-1A is expressed (Fig 2A-B GFP is expressed (Fig 2A-B) . These results strongly showed that CrPV-1A-mediated 236 SG inhibition is independent of Ago-2 binding. 237 We next investigated whether the BC box domain of CrPV-1A is required for SG To determine whether the effects on SG inhibition are due to differences in CrPV-

1A protein levels, we monitored wild-type and mutant CrPV-1A protein levels in 255 transfected cells by immunoblotting using anti-GFP and anti-CrPV-1A, which we raised 256 against purified recombinant CrPV-1A protein (Fig 3A) . The individual CrPV-1A and 257 GFP proteins were detected at similar levels after transfection, indicating that the wild- enrichment, similar to that observed when wild-type CrPV-1A is expressed (Fig 4A-B) . process and other macromolecular metabolic processes (Fig 6E) . 350 We compared the transcriptome data with previous transcriptome data of S2 infection, thus suggesting that replication was recovered (Fig 8A-B) . 391 To support these findings, we performed qRT-PCR of RNA extracted from wild- (Fig S3) . These results showed that Nup358 is necessary for Pateamine A 408 and arsenite stress-induced Rin foci formation in S2 cells.

Nup358 is degraded in a proteasome-dependent manner in CrPV-infected cells 411 Given that the CrPV-1A BC box domain is required for SG inhibition, we next 412 investigated whether CrPV-1A mediates Nup358 degradation. To follow expression, we 20 413 generated an antibody raised against Drosophila Nup358. Immunoblotting for Nup358 414 showed a distinct protein band that migrated at >245 kDa, which was not detected in 415 Nup358 dsRNA treated cells, thus showing specificity of the antibody (Fig 9A) . In CrPV-416 infected cells, Nup358 protein levels were decreased as compared to mock-infected 417 cells (Fig 9B) . Notably, in mutant CrPV(R146A)-infected S2 cells, Nup358 protein levels proteosome inhibition compared to wild-type infection (Fig 9B) . Importantly, treating specific mutations singly or in combination with R146A, we demonstrate that CrPV-1A's 473 ability to block SG formation is not dependent on its ability to bind to Ago-2 (F114A 474 mutation) (Fig 3) . It is also clear that the TALOS domain does not contribute to CrPV-23 475 1A's effects on enrichment of nuclear poly(A)+ mRNA (Fig 4) . However, our results of stress-induced SG formation (Fig. 7, 8) . Moreover, all of these effects are dependent 

Rin granules were counted using Image J using a quantitatively measured 622 threshold intensity and defined circularity using Image J Intensity measurements were 623 done using Image J (94). Box plots and graphs generated using GraphPad Prism is 624 used to represent the data. 

Principles and Properties of Stress Granules

RNA-binding proteins TIA-1

TIAR link the phosphorylation of eIF-2α to the assembly of mammalian stress 707 granules

Stress granules: Sites of mRNA triage that regulate 709 mRNA stability and translatability

Large G3BP-induced granules 712 trigger eIF2α phosphorylation

Eukaryotic Stress Granules : The Ins and Out of Translation 714 What are Stress Granules ?

Modulated Stress Granules Contain a Diverse Proteome and Substructure

Proximity Mapping Reveals the Subcellular Organization of mRNA-Associated 720

Proximity RNA Labeling by APEX-Seq Reveals 722 the Organization of Translation Initiation Complexes and Repressive RNA 723

Granule Transcriptome Reveals Principles of mRNA Accumulation in Stress 727

Stress granules and cell signaling: More than 729 just a passing phase

Cytoplasmic stress granules: Dynamic modulators of cell 732 signaling and disease

Translational Control in Mammalian Cells

The histidyl-tRNA synthetase-related sequence in the 737 eIF-2 alpha protein kinase GCN2 interacts with tRNA and is required for activation 738 in response to starvation for different amino acids

Phosphorylation of eukaryotic translation 741 initiation factor 2 mediates apoptosis in response to activation of the double-742 stranded RNA-dependent protein kinase

Regulated 745 translation initiation controls stress-induced gene expression in mammalian cells

Translation Initiation Control by Heme-Regulated 748

Eukaryotic Initiation Factor 2α Kinase in Erythroid Cells under Cytoplasmic 749

Metazoan stress granule assembly is 751 mediated by P-eIF2α-dependent and -independent mechanisms

Evidence that ternary 754 complex (eIF2-GTP-tRNAiMet)-Deficient preinitiation complexes are core 755 constituents of mammalian stress granules

Uncoupling 758 stress granule assembly and translation initiation inhibition

Translation inhibition and stress granules in the 761 antiviral immune response

Dance with the Devil: 764 Stress Granules and Signaling in Antiviral Responses

Inhibition of Cytoplasmic 767 mRNA Stress Granule Formation by a Viral Proteinase

Virus Hijacks P-Body and Stress Granule Components around Lipid Droplets

Inhibition of Stress Granule 773 Formation by Middle East Respiratory Syndrome Coronavirus 4a Accessory 774

Protein Facilitates Viral Translation, Leading to Efficient Virus Replication

Influenza A virus inhibits 777 cytoplasmic stress granule formation

Middle East Respiratory Coronavirus Accessory Protein 4a Inhibits PKR-Mediated 38

Formation of Antiviral Cytoplasmic Granules during Orthopoxvirus Infection

KSHV inhibits stress granule 786 formation by viral ORF57 blocking PKR activation

Foot-and-mouth disease virus 789 counteracts on internal ribosome entry site suppression by G3BP1 and inhibits 790 G3BP1-mediated stress granule assembly via post

Foot-and-Mouth Disease Virus Leader Protease Cleaves G3BP1 and 794 G3BP2 and Inhibits Stress Granule Formation

Interaction of TIA-1/TIAR with West Nile and dengue 796 virus products in infected cells interferes with stress granule formation and 797 processing body assembly

2A-like" sequences in virus 881 genomes

The RNA 883 silencing endonuclease Argonaute 2 mediates specific antiviral immunity in 884

RNA interference 886 directs innate immunity against viruses in adult Drosophila

Biochemical and single-889 molecule analyses of the RNA silencing suppressing activity of CrPV-1A

Disruption 892 of Stress Granule Formation by the Multifunctional Cricket Paralysis Virus 1A 893

Eukaryotic initiation factor 2α-independent pathway of stress granule induction by 896 the natural product pateamine A

Inhibition of ribosome recruitment induces stress granule formation independently 44 899 of eukaryotic initiation factor 2α phosphorylation

Stimulation of mammalian translation initiation factor eIF4A activity by a small 903 molecule inhibitor of eukaryotic translation

Rasputin, more promiscuous than ever: 906 A review of G3BP

Modulation of Stress Granules and P Bodies during Dicistrovirus 909 Infection

The 911 heat shock response restricts virus infection in Drosophila. Sci Rep

signaling pathway is required but not sufficient for the antiviral 915 response of drosophila

Mechanisms of mRNA export

RanBP2/Nup358 Provides a Major Binding Site for NXF1-p15 Dimers at the

Nuclear Pore Complex and Functions in Nuclear mRNA Export

Formation of Non-Nucleoplasmic Proteasome 926

Foci during the Late Stage of Hyperosmotic Stress. Cells

Proteasome inhibitors: valuable new tools for cell biologists

The Integral Role of 930 RNA in Stress Granule Formation and Function

RNA nuclear export is 933 blocked by poliovirus 2A protease and is concomitant with nucleoporin cleavage

Picornavirus 2A protease 936 regulates stress granule formation to facilitate viral translation

McCormick 939 C. Influenza A Virus Host Shutoff Disables Antiviral Stress-Induced Translation 940

Uncoupling 942 of nucleo-cytoplasmic RNA export and localization during stress

The cytoplasmic filaments of the 945 nuclear pore complex are dispensable for selective nuclear protein import

KIF5B 948 and Nup358 Cooperatively Mediate the Nuclear Import of HIV-1 during Infection

Selective 951 recruitment of nucleoporins on vaccinia virus factories and the role of Nup358 in 952 viral infection

Spatiotemporal Proteomic Analysis of Stress Granule Disassembly Using APEX 955

Reveals Regulation by SUMOylation and Links to ALS Pathogenesis

Sumoylation is Required for the 47

Cytoplasmic Accumulation of a Subset of mRNAs

Sumoylation of eIF4A2 affects stress granule formation

RanBP2/Nup358 enhances miRNA activity by sumoylating Argonautes

The nucleoporin RanBP2 has 966 SUMO1 E3 ligase activity

2A 968 peptides provide distinct solutions to driving stop-carry on translational recoding

The 971 "cleavage" activities of foot-and-mouth disease virus 2A site-directed mutants and 972 naturally occurring "2A-like" sequences

The 5′

Untranslated Region of a Novel Infectious Molecular Clone of the Dicistrovirus 975

Cricket Paralysis Virus Modulates Infection

Switch from Cap-to Factorless IRES-Dependent 0 and +1 48

Frame Translation during Cellular Stress and Dicistrovirus Infection. Preiss T, 978 editor

NIH Image to ImageJ: 25 years of 980 image analysis

Unipro UGENE: a unified bioinformatics 983 toolkit

Unipro UGENE NGS pipelines and components for variant calling, RNA-seq 986 and ChIP-seq data analyses

Salmon provides fast and 988 bias-aware quantification of transcript expression

iDEP: An integrated web application for differential 991 expression and pathway analysis of RNA-Seq data

InteractiVenn: A 994 web-based tool for the analysis of sets through Venn diagrams

ClueGO: A Cytoscape plug-in to decipher functionally grouped gene ontology and 998 pathway annotation networks

Cytoscape: A software Environment for integrated models of biomolecular 1001 interaction networks

PDB 6C3R) (below) highlighting the domains selected for mutagenesis. (B) Schematic 1017 of CrPV-1A-2A-GFP RNA containing the CrPV 5' and 3'UTRs. (C) Confocal 1018 immunofluorescence images of S2 cells transfected with control 5'cap-GFP-poly (A)+, 1019 wild type or R146A mutant CrPV-1A-2A-GFP RNAs (16 hours) followed by one-hour 1020 treatment in the presence or absence of 500 µM sodium arsenite. The arrows show 1021 transfected cells. Shown are representative transfected cells detecting GFP 1022 fluorescence (green), Rin antibody staining (red), Hoechst dye staining for nucleus 1023 (blue) and merged images

At least 50 cells were counted for each condition from three 1026 independent experiments. Data are mean ± SD. P > 0.05 (ns) p < 0.0001(****) by a 1027 one-way ANOVA (nonparametric) with a Bonferroni

Figure 2 CrPV-1A mediated stress granule inhibition requires the BC Box domain 1030 and is independent of the Ago-2 binding domain. (A) Images of transiently 1031 transfected S2 cells with the indicated in vitro transcribed RNAs (16 hours), followed by 1032 one-hour sodium arsenite treatment (500 M). Shown are representative transfected 1033 cells detecting GFP fluorescence (green), Rin antibody staining (red) and Hoechst dye 1034 staining for nucleus (blue) and merged images. The arrows show transfected cells

Images were taken using the Leica Sp5 confocal microscope with a 63X objective lens 1036 and 2X zoom (B) Box plot of the number of Rin foci per cell. At least 50 cells were 1037 counted for each condition from three independent experiments

021 (*),p < 0.0001(****) by a one-way ANOVA (nonparametric) with 1039 a Bonferroni's post hoc-test

Immunoblots of 1042 lysates from S2 cells transfected with the indicated reporter RNAs (16 hours post 1043 transfection). A light exposure (top) and longer exposure (bottom) of an anti-GFP 1044 immunoblot and CrPV-1A immunoblot is shown. (B) Autoradiography of

C) Percent quantification of [ 35 S] Met/Cys labelled CrPV-1A proteins. Data are 1047 mean ± SD from three independent experiments

Immunoblots of lysates from S2 cells infected with (M) mock, CrPV or CrPV R146A 1050

Figure 4 CrPV-1A localizes to the nucleus and induces poly(A)+ RNA 1053 accumulation in the nucleus. (A) Confocal immunofluorescence images of S2 cells 1054 transfected with in vitro transcribed RNA encoding CrPV-1A

CrPV-1A antibody staining 1056 (red), fluorescence in situ hybridization using Cy5-oligo(dT) probes (cyan) and Hoechst 1057 dye (blue). The arrows show transfected cells

confocal microscope with a 63X objective lens and 2X zoom (B) Box plot of the fraction 1059 of nuclear to total Cy5-oligo(dT) fluorescence intensity in each cell. At least 50 cells 1060 were counted for each condition from two independent experiments

SD. p > 0.05 (ns), p < 0.021 (*), p < 0.002(**), p < 0.0001(****) by a one-way ANOVA 1062 (nonparametric) with a Bonferroni's post hoc-test

Z-stack confocal 1065 images of S2 cells infected with (A) wild-type CrPV or (B) CrPV(R146A) virus. From left 1066 to right are Z-stack images through the cells. Cells were fixed and stained with Lamin 1067 (green), CrPV-1A(red) and Hoechst (blue)

Figure 6 Transcriptional profiling of CrPV and CrPV(R146A) infected S2 cells

Principal component analysis of transcriptional signatures from cells infected with Mock

Bar graph indicating the number of differentially expressed 1073 genes for each comparison identified by DESeq2 (C) Volcano plots showing changes in 1074 gene expression with fold change (FC) in expression intensity of DEGs

Gene ontology analysis filtered by molecular function for CrPV infected cells (F) Venn 1077 diagram showing comparison of dicistrovirus transcriptome datasets

Figure 7 RNA export modulates CrPV infection. (A) Fluorescence in situ 1080 hybridization using Cy5-oligo(dT) (blue) of S2 cells incubated with dsRNA targeting 1081 RNA export factor NXF1 or control GFP and Hoechst dye (blue), followed by mock 1082 infection or infection with wild-type or R146A mutant virus for 8 hours

Viral yield from wild-type and mutant (R146A) CrPV infected S2 cells was accessed by 1084 fluorescence foci unit (FFU). Shown are averages from two independent experiments

Figure 8 Nup358 promotes CrPV infection in an R146-dependent manner

Immunoblots of S2 cells treated with dsRNA targeting GP210, mTor, Rae1

Nup358 or control FLuc, followed by mock infection or infection with wild-type 1089 or mutant CrPV virus for 8 hours (MOI 1). (B) Quantification of VP2 intensity normalized 1090 to tubulin. The intensity values are normalized to the VP2/Tubulin intensity in FLuc 1091 control knockdown cells. (C) CrPV viral RNA levels by qRT

Data are mean ± SD relative to WT p > 0.05 (ns), p < 0.002(**), p < 1093 0.0001(****) by a one-way ANOVA (nonparametric) with a Bonferroni's post hoc test

Fluorescence in situ hybridization using Cy5-oligo(dT) or antibody staining of Rin

of S2 cells treated with control dsRNA or Nup358 dsRNA in the presence of DMSO or 1096

Hoechst staining is shown in blue. The arrows show Nup358 1097 knockdown cells Shown are representative images of at least two independent 1098 experiments

Rachel DaSilva, Reid Warsaba and Christina Young) for discussions 1136 and critical reading of the paper. This study was supported by a CIHR operating grant 1137 (PJT-178342) and an NSERC grant (RGPIN-2017-04515) to EJ, an NIH grant (A132131 1138 -R01AI137471) to RA

Xinying Wang 1144 Formal analysis: Jibin Sadasivan, Marli Vlok

Writing -review & editing: Jibin Sadasivan

 1131 We thank Eric Lecuyer for generously providing the Rin antibody. We acknowledge the 1132