Reovirus uses temporospatial compartmentalization to orchestrate core versus outercapsid assembly

Abstract

Reoviridae virus family members, such as mammalian orthoreovirus (reovirus), encounter a unique challenge during replication. To hide the dsRNA from host recognition, the genome remains encapsidated in transcriptionally active proteinaceous core capsids that transcribe and release +RNA. De novo +RNAs and core proteins must repeatedly assemble into new progeny cores in order to logarithmically amplify replication. Reoviruses also produce outercapsid (OC) proteins μ1, σ3 and σ1 that assemble onto cores to create highly stable infectious full virions. Current models of reovirus replication position amplification of transcriptionally-active cores and assembly of infectious virions in shared factories, but we hypothesized that since assembly of OC proteins would halt core amplification, OC assembly is somehow regulated. Using kinetic analysis of virus +RNA, core and OC proteins, core assembly and whole virus assembly, assembly of OC proteins was found to be temporally delayed. All viral RNAs and proteins were made simultaneously, eliminating the possibility that delayed OC RNAs or proteins account for delayed OC assembly. High resolution fluorescence and electron microscopy revealed that core amplification occurred early during infection at peripheral core-only factories, while all OC proteins associated with lipid droplets (LDs) that coalesced near the nucleus in a μ1–dependent manner. Core-only factories transitioned towards the nucleus despite cycloheximide-mediated halting of new protein expression, while new core-only factories developed in the periphery. As infection progressed, OC assembly occurred at LD-and nuclear-proximal factories. Silencing of OC μ1 expression with siRNAs led to large factories that remained further from the nucleus, implicating μ1 in the transition to perinuclear factories. Moreover, late during infection, +RNA pools largely contributed to the production of de-novo viral proteins and fully-assembled infectious viruses. Altogether the results suggest an advanced model of reovirus replication with spatiotemporal segregation of core amplification, OC complexes and fully assembled virions.

Dinucleotide biases in RNA viruses that infect vertebrates or invertebrates

“…6C ). Limited data on reoviruses indicated that all proteins are expressed at similar levels ( 42 ).…”

“…However, M is expressed at higher levels than S and ORF1a/ORF1b but shows a weaker bias against CpG ( 48 , 49 ). Finally, the little differences among ORFs observed in picornaviruses might be expected on the basis that the single ORF is translated as a polyprotein and subsequently cleaved, whereas for reoviruses limited experimental evidence suggests that all proteins are expressed at similar levels ( 42 ). Thus, it is difficult to draw definite conclusions about why individual ORFs differ in CpG content.…”

Seeing Biomolecular Condensates Through the Lens of Viruses

“…Another new area of research is how viral condensates interact with organelles and other cellular components to mediate viral replication and assembly functions. Some common themes emerge in the ways in which viral condensates use lipid membranes, in the form of either lipid droplets (113)(114)(115)(116) or ER membranes (117), to maximize virus production.…”

Proteasome activity is required for reovirus entry into cells

“…Anti-reovirus polyclonal antisera were generated by immunizing rabbits with UV-inactivated mixture of purified virions of T1L and T3D. Anti-core antisera were obtained from the Maya Shmulevitz laboratory (University of Alberta) (20). mAb 8H6 and 4A3, previously described, were obtained from the Terry Dermody laboratory (University of Pittsburgh) (21).…”