Co-opted cytosolic proteins form condensate substructures within membranous replication organelles of a positive-strand RNA virus.

Positive-strand RNA viruses co-opt organellar membranes for biogenesis of viral replication organelles (VROs). Tombusviruses also co-opt pro-viral cytosolic proteins to VROs. It is currently not known what type of molecular organization keeps co-opted proteins sequestered within membranous VROs. In this study, we employed tomato bushy stunt virus (TBSV) and carnation Italian ringspot virus (CIRV) - Nicotiana benthamiana pathosystems to identify biomolecular condensate formation in VROs. We show that TBSV p33 and the CIRV p36 replication proteins sequester glycolytic and fermentation enzymes in unique condensate substructures associated with membranous VROs. We find that p33 and p36 form droplets in vitro driven by intrinsically disordered region. The replication protein organizes partitioning of co-opted host proteins into droplets. VRO-associated condensates are critical for local adenosine triphosphate production to support energy for virus replication. We find that co-opted endoplasmic reticulum membranes and actin filaments form meshworks within and around VRO condensates, contributing to unique composition and structure. We propose that p33/p36 organize liquid-liquid phase separation of co-opted concentrated host proteins in condensate substructures within membranous VROs. Overall, we demonstrate that subverted membranes and condensate substructures co-exist and are critical for VRO functions. The replication proteins induce and connect the two substructures within VROs.

Influence of HIV-1 Genomic RNA on the Formation of Gag Biomolecular Condensates.

Biomolecular condensates (BMCs) play an important role in the replication of a growing number of viruses, but many important mechanistic details remain to be elucidated. Previously, we demonstrated that the pan-retroviral nucleocapsid (NC) and HIV-1 pr55Gag (Gag) proteins phase separate into condensates, and that HIV-1 protease (PR)-mediated maturation of Gag and Gag-Pol precursor proteins yields self-assembling BMCs that have HIV-1 core architecture. Using biochemical and imaging techniques, we aimed to further characterize the phase separation of HIV-1 Gag by determining which of its intrinsically disordered regions (IDRs) influence the formation of BMCs, and how the HIV-1 viral genomic RNA (gRNA) could influence BMC abundance and size. We found that mutations in the Gag matrix (MA) domain or the NC zinc finger motifs altered condensate number and size in a salt-dependent manner. Gag BMCs were also bimodally influenced by the gRNA, with a condensate-promoting regime at lower protein concentrations and a gel dissolution at higher protein concentrations. Interestingly, incubation of Gag with CD4+ T cell nuclear lysates led to the formation of larger BMCs compared to much smaller ones observed in the presence of cytoplasmic lysates. These findings suggest that the composition and properties of Gag-containing BMCs may be altered by differential association of host factors in nuclear and cytosolic compartments during virus assembly. This study significantly advances our understanding of HIV-1 Gag BMC formation and provides a foundation for future therapeutic targeting of virion assembly.

The Rous sarcoma virus Gag Polyprotein Forms Biomolecular Condensates Driven by Intrinsically-disordered Regions.

Biomolecular condensates (BMCs) play important roles incellular structures includingtranscription factories, splicing speckles, and nucleoli. BMCs bring together proteins and other macromolecules, selectively concentrating them so that specific reactions can occur without interference from the surrounding environment. BMCs are often made up of proteins that contain intrinsically disordered regions (IDRs), form phase-separated spherical puncta, form liquid-like droplets that undergo fusion and fission, contain molecules that are mobile, and are disrupted with phase-dissolving drugs such as 1,6-hexanediol. In addition to cellular proteins, many viruses, including influenza A, SARS-CoV-2, and human immunodeficiency virus type 1 (HIV-1) encode proteins that undergo phase separation and rely on BMC formation for replication. In prior studies of the retrovirus Rous sarcoma virus (RSV), we observed that the Gag protein forms discrete spherical puncta in the nucleus, cytoplasm, and at the plasma membrane that co-localize with viral RNA and host factors, raising the possibility that RSV Gag forms BMCs that participate in the intracellular phase of the virion assembly pathway. In our current studies, we found that Gag contains IDRs in the N-terminal (MAp2p10) and C-terminal (NC) regions of the protein and fulfills many criteria of BMCs. Although the role of BMC formation in RSV assembly requires further study, our results suggest the biophysical properties of condensates are required for the formation of Gag complexes in the nucleus and the cohesion of these complexes as they traffic through the nuclear pore, into the cytoplasm, and to the plasma membrane, where the final assembly and release of virus particles occurs.

HIV-Induced CPSF6 Condensates.

Viruses are obligate parasites that rely on their host's cellular machinery for replication. To facilitate their replication cycle, many viruses have been shown to remodel the cellular architecture by inducing the formation of membraneless organelles (MLOs). Eukaryotic cells have evolved MLOs that are highly dynamic, self-organizing microenvironments that segregate biological processes and increase the efficiency of reactions by concentrating enzymes and substrates. In the context of viral infections, MLOs can be utilized by viruses to complete their replication cycle. This review focuses on the pathway used by the HIV-1 virus to remodel the nuclear landscape of its host, creating viral/host niches that enable efficient viral replication. Specifically, we discuss how the interaction between the HIV-1 capsid and the cellular factor CPSF6 triggers the formation of nuclear MLOs that support nuclear reverse transcription and viral integration in favored regions of the host chromatin. This review compiles current knowledge on the origin of nuclear HIV-MLOs and their role in early post-nuclear entry steps of the HIV-1 replication cycle.

Condensation Goes Viral: A Polymer Physics Perspective.

The past decade has seen a revolution in our understanding of how the cellular environment is organized, where an incredible body of work has provided new insights into the role played by membraneless organelles. These rapid advancements have been made possible by an increasing awareness of the peculiar physical properties that give rise to such bodies and the complex biology that enables their function. Viral infections are not extraneous to this. Indeed, in host cells, viruses can harness existing membraneless compartments or, even, induce the formation of new ones. By hijacking the cellular machinery, these intracellular bodies can assist in the replication, assembly, and packaging of the viral genome as well as in the escape of the cellular immune response. Here, we provide a perspective on the fundamental polymer physics concepts that may help connect and interpret the different observed phenomena, ranging from the condensation of viral genomes to the phase separation of multicomponent solutions. We complement the discussion of the physical basis with a description of biophysical methods that can provide quantitative insights for testing and developing theoretical and computational models.

Membraneless Organelles and Condensates Orchestrate Innate Immunity Against Viruses.

The cellular defense against viruses involves the assembly of oligomers, granules and membraneless organelles (MLOs) that govern the activation of several arms of the innate immune response. Upon interaction with specific pathogen-derived ligands, a number of pattern recognition receptors (PRRs) undergo phase-separation thus triggering downstream signaling pathways. Among other relevant condensates, inflammasomes, apoptosis-associated speck-like protein containing a caspase-recruitment domain (ASC) specks, cyclic GMP-AMP synthase (cGAS) foci, protein kinase R (PKR) clusters, ribonuclease L-induced bodies (RLBs), stress granules (SGs), processing bodies (PBs) and promyelocytic leukemia protein nuclear bodies (PML NBs) play different roles in the immune response. In turn, viruses have evolved diverse strategies to evade the host defense. Viral DNA or RNA, as well as viral proteases or proteins carrying intrinsically disordered regions may interfere with condensate formation and function in multiple ways. In this review we discuss current and hypothetical mechanisms of viral escape that involve the disassembly, repurposing, or inactivation of membraneless condensates that govern innate immunity. We summarize emerging interconnections between these diverse condensates that ultimately determine the cellular outcome.

Liquid-liquid Phase Separation in Viral Function.

An emerging set of results suggests that liquid-liquid phase separation (LLPS) is the basis for the formation of membrane-less compartments in cells. Evidence is now mounting that various types of virus-induced membrane-less compartments and organelles are also assembled via LLPS. Specifically, viruses appear to use intracellular phase transitions to form subcellular microenvironments known as viral factories, inclusion bodies, or viroplasms. These compartments - collectively referred to as viral biomolecular condensates - can be used to concentrate replicase proteins, viral genomes, and host proteins that are required for virus replication. They can also be used to subvert or avoid the intracellular immune response. This review examines how certain DNA or RNA viruses drive the formation of viral condensates, the possible biological functions of those condensates, and the biophysical and biochemical basis for their assembly.

CASC3 Biomolecular Condensates Restrict Turnip Crinkle Virus by Limiting Host Factor Availability.

The exon-junction complex (EJC) plays a role in post-transcriptional gene regulation and exerts antiviral activity towards several positive-strand RNA viruses. However, the spectrum of RNA viruses that are targeted by the EJC or the underlying mechanisms are not well understood. EJC components from Arabidopsis thaliana were screened for antiviral activity towards Turnip crinkle virus (TCV, Tombusviridae). Overexpression of the accessory EJC component CASC3 inhibited TCV accumulation > 10-fold in Nicotiana benthamiana while knock-down of endogenous CASC3 resulted in a > 4-fold increase in TCV accumulation. CASC3 forms cytoplasmic condensates and deletion of the conserved SELOR domain reduced condensate size 7-fold and significantly decreased antiviral activity towards TCV. Mass spectrometry of CASC3 complexes did not identify endogenous stress granule or P-body markers and CASC3 failed to co-localize with an aggresome-specific dye suggesting that CASC3 condensates are distinct from well-established membraneless compartments. Mass spectrometry and bimolecular fluorescence complementation assays revealed that CASC3 sequesters Heat shock protein 70 (Hsp70-1) and Glyceraldehyde-3-phosphate dehydrogenase (GAPDH), two host factors with roles in tombusvirus replication. Overexpression of Hsp70-1 or GAPDH reduced the antiviral activity of CASC3 2.1-fold and 2.8-fold, respectively, and suggests that CASC3 inhibits TCV by limiting host factor availability. Unrelated Tobacco mosaic virus (TMV) also depends on Hsp70-1 and CASC3 overexpression restricted TMV accumulation 4-fold and demonstrates that CASC3 antiviral activity is not TCV-specific. Like CASC3, Auxin response factor 19 (ARF19) forms poorly dynamic condensates but ARF19 overexpression failed to inhibit TCV accumulation and suggests that CASC3 has antiviral activities that are not ubiquitous among cytoplasmic condensates.

Biomolecular Condensation of the Human Papillomavirus E2 Master Regulator with p53: Implications in Viral Replication.

p53 exerts its tumour suppressor activity by modulating hundreds of genes and it can also repress viral replication. Such is the case of human papillomavirus (HPV) through targeting the E2 master regulator, but the biochemical mechanism is not known. We show that the C-terminal DNA binding domain of HPV16 E2 protein (E2C) triggers heterotypic condensation with p53 at a precise 2/1 E2C/p53 stoichiometry at the onset for demixing, yielding large regular spherical droplets that increase in size with E2C concentration. Interestingly, transfection experiments show that E2 co-localizes with p53 in the nucleus with a grainy pattern, and recruits p53 to chromatin-associated foci, a function independent of the DNA binding capacity of p53 as judged by a DNA binding impaired mutant. Depending on the length, DNA can either completely dissolve or reshape heterotypic droplets into irregular condensates containing p53, E2C, and DNA, and reminiscent of that observed linked to chromatin. We propose that p53 is a scaffold for condensation in line with its structural and functional features, in particular as a promiscuous hub that binds multiple cellular proteins. E2 appears as both client and modulator, likely based on its homodimeric DNA binding nature. Our results, in line with the known role of condensation in eukaryotic gene enhancement and silencing, point at biomolecular condensation of E2 with p53 as a means to modulate HPV gene function, strictly dependent on host cell replication and transcription machinery.

A condensate-hardening drug blocks RSV replication in vivo.

Biomolecular condensates have emerged as an important subcellular organizing principle1. Replication of many viruses, including human respiratory syncytial virus (RSV), occurs in virus-induced compartments called inclusion bodies (IBs) or viroplasm2,3. IBs of negative-strand RNA viruses were recently shown to be biomolecular condensates that form through phase separation4,5. Here we report that the steroidal alkaloid cyclopamine and its chemical analogue A3E inhibit RSV replication by disorganizing and hardening IB condensates. The actions of cyclopamine and A3E were blocked by a point mutation in the RSV transcription factor M2-1. IB disorganization occurred within minutes, which suggests that these molecules directly act on the liquid properties of the IBs. A3E and cyclopamine inhibit RSV in the lungs of infected mice and are condensate-targeting drug-like small molecules that have in vivo activity. Our data show that condensate-hardening drugs may enable the pharmacological modulation of not only many previously undruggable targets in viral replication but also transcription factors at cancer-driving super-enhancers6.