key: cord-0325762-y63ckysf
authors: Brown, Shelby L.; May, Jared P.
title: Phase separation of both a plant virus movement protein and cellular factors support virus-host interactions
date: 2021-05-11
journal: bioRxiv
DOI: 10.1101/2021.05.11.443547
sha: ea6db964daa45555c49c8d97df113cd5bbb0e549
doc_id: 325762
cord_uid: y63ckysf

Phase separation concentrates biomolecules, which should benefit RNA viruses that must sequester viral and host factors during an infection. Here, the p26 movement protein from Pea enation mosaic virus 2 (PEMV2) was found to phase separate and partition in nucleoli and G3BP stress granules (SGs) in vivo. Electrostatic interactions drive p26 phase separation as mutation of basic (R/K-G) or acidic (D/E-G) residues either blocked or reduced phase separation, respectively. During infection, p26 must partition inside the nucleolus and interact with fibrillarin (Fib2) as a pre-requisite for systemic trafficking of viral RNAs. Partitioning of p26 in pre-formed Fib2 droplets was dependent on p26 phase separation suggesting that phase separation of viral movement proteins supports nucleolar partitioning and virus movement. Furthermore, viral ribonucleoprotein complexes containing p26, Fib2, and PEMV2 RNA were formed via phase separation in vitro and could provide the basis for self-assembly in planta. Interestingly, both R/K-G and D/E-G p26 mutants failed to support systemic trafficking of a Tobacco mosaic virus (TMV) vector in Nicotiana benthamiana suggesting that p26 phase separation, proper nucleolar partitioning, and systemic movement are intertwined. p26 also partitioned in SGs and G3BP over-expression restricted PEMV2 accumulation >20-fold. Expression of phase separation-deficient G3BP only restricted PEMV2 5-fold, demonstrating that G3BP phase separation is critical for maximum antiviral activity. AUTHOR SUMMARY Phase separation of several cellular proteins is associated with forming pathological aggregates and exacerbating neurodegenerative disease progression. In contrast, roles for viral protein phase separation in RNA virus lifecycles are less understood. Here, we demonstrate that the p26 movement protein from Pea enation mosaic virus 2 phase separates and partitions with phase-separated cellular proteins fibrillarin and G3BP. The related orthologue from Groundnut rosette virus has been extensively studied and is known to interact with fibrillarin in the nucleolus as a pre-requisite for virus movement. We determined that basic residues and electrostatic interactions were critical for p26 phase separation. Furthermore, mutation of charged residues prevented the rescue of a movement-deficient Tobacco mosaic virus vector in Nicotiana benthamiana. Stress granules form through phase separation and we found that p26 could partition inside stress granules following heat shock. Phase separation of the stress granule nucleator G3BP was required for maximum antiviral activity and constitutes a host response that is dependent on cellular protein phase separation. Collectively, we demonstrate that phase separation of a plant virus protein facilitates virus-host interactions that are required for virus movement and phase separation of cellular proteins can simultaneously restrict virus replication.

were purified (Fig. 2C ) and tested. First, all basic or acidic residues were mutated to glycine 147 (R/K-G or D/E-G, respectively). Since high-salt blocks IDR-GFP phase separation, simultaneous 148 mutation of either basic or acidic residues was predicted to inhibit phase separation. Indeed, 

Again, VLIMFYW-S phase separated like wild-type and was sensitive to high-salt (Fig. 2G ).

However, VLIMFYW-S condensates failed to recover in FRAP assays (Fig. 2J ). These results

suggest that hydrophobic residues contribute to the limited fluidity of p26 phase separations or 

Phase separation-deficient p26 mutants fail to systemically traffic a virus vector.

To determine whether phase separation-deficient p26 mutants could support virus trafficking, a (Fig. 5B ). As expected, ΔNTF2-G3BP failed to phase 273 separate and form SGs following heat shock (Fig. 5B) . When co-expressed with p26:GFP, 274 recruitment of p26 to G3BP SGs was observed following heat shock (Fig. 5B) . To determine if 284 accumulation was reduced >20-fold during G3BP over-expression demonstrating that G3BP 285 exerts strong antiviral activity towards PEMV2 (Fig. 5E ). Virus accumulation was largely 286 restored (only 5-fold inhibition) during overexpression of ΔNTF2-G3BP demonstrating that 287 phase separation of G3BP is required for maximal antiviral activity (Fig. 5E) 

The Tobacco mosaic virus (TMV) expression vector pJL-TRBO has been previously 

Fluorescence recovery after photobleaching (FRAP). A ~2 µm diameter region was 397 photobleached with 100% laser power with subsequent recovery measured at 5 s intervals.

Background regions and unbleached reference condensates were recorded as controls. FRAP 

How viruses use the endoplasmic reticulum for entry, replication, and 485 assembly

Viruses as modulators of mitochondrial functions

Editorial: Viral Interactions with the Nucleus

Modification of intracellular membrane structures for virus 491 replication

What lava lamps and vinaigrette can teach us about cell biology

Liquid phase separation

Matter over mind: Liquid phase separation and neurodegeneration. 496 The Journal of biological chemistry

Phase Separation, and Membrane-Less Organelles: Are 498 Post-Transcriptional Modifications Modulating Organelle Dynamics?

Pi-Pi contacts are an 503 overlooked protein feature relevant to phase separation

Protein 505 Phase Separation: A New Phase in Cell Biology

Phase separation of RNA-binding proteins in physiology and disease: An 507 introduction to the JBC Reviews thematic series

Stress-Triggered Phase Separation Is an Adaptive, Evolutionarily Tuned Response

Both G3BP1 and 513 G3BP2 contribute to stress granule formation. Genes to cells : devoted to molecular & cellular 514 mechanisms

Distinct stages in stress granule 518 assembly and disassembly

Considerations and Challenges in Studying Liquid-602

Liquid Phase Separation and Biomolecular Condensates

Grammar Governing the Driving Forces for Phase Separation of Prion-like RNA Binding 605

Molecular 609 interactions underlying liquid-liquid phase separation of the FUS low-complexity domain. Nature 610 structural & molecular biology

Identification of a nuclear localization signal and 612 nuclear export signal of the umbraviral long-distance RNA movement protein

Charge versus sequence for nuclear/nucleolar 615 localization of plant ribosomal proteins

A charge-617 dependent mechanism is responsible for the dynamic accumulation of proteins inside nucleoli

The 620 nucleolus functions as a phase-separated protein quality control compartment

Two RNA-binding sites in plant 623 fibrillarin provide interactions with various RNA substrates

Nascent Pre-rRNA Sorting via 626

Phase Separation Drives the Assembly of Dense Fibrillar Components in the Human Nucleolus

RNA transcription modulates 629 phase transition-driven nuclear body assembly. Proceedings of the National Academy of 630 Sciences

TRBO: A High-Efficiency Tobacco Mosaic Virus RNA-Based Overexpression 632 Vector

A plant virus-encoded protein facilitates long-634 distance movement of heterologous viral RNA. Proceedings of the National Academy of 635 Sciences

The stress granule component 637 G3BP is a novel interaction partner for the nuclear shuttle proteins of the nanovirus pea necrotic 638 yellow dwarf virus and geminivirus abutilon mosaic virus

RasGAP-associated endoribonuclease G3BP assembles stress granules

RNA-Induced Conformational Switching and Clustering of G3BP Drive Stress Granule 644 Assembly by Condensation

Negri bodies and other virus 646 membrane-less replication compartments

Functional 649 characterization of Negri bodies (NBs) in rabies virus-infected cells: Evidence that NBs are sites 650 of viral transcription and replication

Measles virus 652 nucleo-and phosphoproteins form liquid-like phase-separated compartments that promote 653 nucleocapsid assembly

Building Viral Replication Organelles: Close 657 Encounters of the Membrane Types

Phosphoregulation of Phase Separation by the SARS-CoV-2 N Protein Suggests a Biophysical 660 Basis for its Dual Functions

Phase Separation of C9orf72 Dipeptide Repeats Perturbs Stress Granule Dynamics

Polymer physics of intracellular phase 665 transitions

Analyzing the G3BP-like gene family of Arabidopsis 667 thaliana in early turnip mosaic virus infection

Structure-Mediated RNA Decay by UPF1 669 and G3BP1

3' cap-independent translation enhancers of plant viruses

RNA virus evasion of nonsense-mediated decay

May JP, Simon AE. Targeting of viral RNAs by Upf1-mediated RNA decay pathways

Intrinsically 679 disordered proteins drive membrane curvature

Entropic 681 Control of Receptor Recycling Using Engineered Ligands

Actively Mediate Liquid-liquid Phase Separation. Bio-protocol

Aureusvirus P14 is 686 an efficient RNA silencing suppressor that binds double-stranded RNAs without size specificity

EzColocalization: An ImageJ plugin for visualizing and 689 measuring colocalization in cells and organisms

Note that the majority of plant 699 mesophyll cells is taken up by a single large vacuole. Differential interference contrast (DIC) 700 microscopy was used for p26:GFP samples to visualize cell borders. Bar scale: 20 µm. (C) 701 FRAP analysis of p26:GFP was performed by photobleaching cytoplasmic condensates and 702 monitoring fluorescence recovery. A representative p26:GFP condensate is shown before 703 photobleaching

Average FRAP intensity is shown from seven FRAP experiments and shaded area represents 705 95% confidence interval

Fig. 2. p26 is intrinsically disordered and phase separates through electrostatic 708 interactions. (A) PEMV2 p26 contains a large intrinsically disordered region (IDR) spanning 709 amino acids 1-132. The dispensable cell-to-cell movement protein, p27, is highly ordered

The p26 IDR was fused to the N-terminus of GFP for bacterial expression and contained an N-711 terminal histidine tag. The p26 IDR sequence is shown with highlighted residues corresponding 712 to basic (blue) or acidic (red) residues. (C) Recombinant proteins used in this study were 713 analyzed by SDS-PAGE to assess size and purity

Bar scale: 20 µm. (F) Phase diagram for IDR-GFP gives an apparent 722 C sat = 4 µM and sensitivity to high NaCl concentrations. Results are representative of two 723 independent experiments. (G) IDR mutants (8 µM) were examined using in vitro phase 724 separation assays. R/K-G formed irregular aggregates at high concentration (24 µM) and D/E-G 725 showed reduced phase separation compared to IDR-GFP. R-K and VLIMFYW-S mutants 726 appeared like wild-type IDR. Bar scale: 20 µm (H) D/E-G had significantly reduced turbidity 727 (OD 600 ) under crowding conditions when compared to IDR-GFP at 8 µM and 24 µM 728 concentrations. Data represents three independent replicates for each condition. Bars denote 729 standard deviations. *** P<0.001 unpaired t test (I) Mean condensate sizes for all mutants 730 (excluding R/K-G) were plotted by cumulative distribution frequency. Particle sizes were 731 measured from three representative 20x fields using ImageJ. P values represent results from 732 two-tailed Mann-Whitney tests compared to IDR-GFP. ns: not significant. (J) FRAP was 733 performed for in vitro condensates. 24 µM protein was used for

Data represents 7-10 separate FRAP measurements for each mutant. Shaded areas represent 737 95% confidence intervals

Prior to imaging, leaves were infiltrated with 5 µg/mL DAPI to stain 742 nuclei

Bar scale: Top 20 µm; Bottom 10 µm. (B) Nuclear localization of p26:GFP or D/E-G was 744 quantified using Mander's overlap coefficient (MOC) using ImageJ and EzColocalization

Bar 746 scale: 50 µm. Error bars denote standard deviations. ****P<0.0001 unpaired t test. (C) Fib2 747 contains an N-terminal glycine-and arginine-rich (GAR) domain that is intrinsically disordered. 748 (D) Either the Fib2 GAR domain (Fib2 GAR ) or full-length Fib2 (Fib2 FL ) were fused to mCherry and 749 purified from E. coli and analyzed by SDS-PAGE

and Fib2 FL were examined by confocal microscopy after inducing phase 751 separation with 10% PEG-8000 alone or in the presence of 1 M NaCl. 8 µM protein was used 752 for all assays. Bar scale: 20 µm. (F) FRAP analyses of Fib2 GAR and Fib2 FL condensates

Fib2 GAR and Fib2 FL droplets were pre-formed prior to addition of PEMV2-Cy5 at a 500:1 756 protein:RNA molar ratio. PEMV2 RNA was only sorted to Fib2 FL condensates. Bar scale: 20 µm. 757 (H) IDR-GFP droplets were pre-formed prior to addition of PEMV2-Cy5 or TCV-Cy5 at a 500:1 758 protein:RNA molar ratio. Bar scale: 20 µm. (I) The fraction of IDR-GFP signal that was positive 759 for Cy5-labelled RNA was determined by MOC analysis using EzColocalization [80]. ns: not 760 significant by unpaired t test. Bars denote standard deviations. Three 20x fields were quantified 761 for each condition. (J) Fib2 GAR droplets were pre-formed using 24 µM protein before the addition 762 of 4 µM IDR-GFP or R/K-G. Sorting of IDR-GFP to

RNA were mixed at a 500:500:1 molar ratio after pre-forming 765

condensates under crowding conditions. Droplets containing all 766 components were observed. Bar scale: 10 µm. Images in all panels are representative of at 767 least three independent experiments

GFP fusions were inserted into pJL-TRBO to test whether 772 systemic trafficking could be restored. (B) Following agroinfiltration of N. benthamiana leaves, 773 TMV infections were established in local leaves. Free GFP, or GFP-fusion proteins were 774 visualized and detected in local leaves at 4 dpi by UV exposure (Left) or western blotting 775 (Right)

Fig. 5. p26 is sorted into G3BP phase separations that restrict PEMV2 accumulation

Scale bar: 20 µm. Inset shows western 788 blot using anti-RFP antibodies to detect full-length G3BP and ΔNTF2-G3BP. Rubisco was used 789 as a loading control (C) G3BP:RFP was agroinfiltrated into systemically-infected TMV:p26:GFP 790 plants to determine if p26 partitions in G3BP SGs during a virus infection. p26:GFP co-localized 791 with G3BP SGs as labelled by white arrows. Scale bar: 20 µm. (D) Native G3BP expression was 792 measured in Mock-or PEMV2-infected N. benthamiana by RT-qPCR. The co-agroinfiltrated p14 793 RNA silencing suppressor was used as a reference gene. Data is from three biological 794 replicates. *P<0.05; student's t-test. Bars denote standard error. (E) PEMV2 was agroinfiltrated 795 alone, or alongside either G3BP or ΔNTF2-G3BP. At 3 dpi, total RNAs were extracted and used 796 for RT-qPCR targeting PEMV2 or p14 (reference gene)

Welch ANOVA with multiple comparisons was used to determine if observed differences were 799 significant