Biogenesis and dynamics of the coronavirus replicative structures

Abstract

Coronaviruses are positive-strand RNA viruses that are important infectious agents of both animals and humans. A common feature among positive-strand RNA viruses is their assembly of replication-transcription complexes in association with cytoplasmic membranes. Upon infection, coronaviruses extensively rearrange cellular membranes into organelle-like replicative structures that consist of double-membrane vesicles and convoluted membranes to which the nonstructural proteins involved in RNA synthesis localize. Double-stranded RNA, presumably functioning as replicative intermediate during viral RNA synthesis, has been detected at the double-membrane vesicle interior. Recent studies have provided new insights into the assembly and functioning of the coronavirus replicative structures. This review will summarize the current knowledge on the biogenesis of the replicative structures, the membrane anchoring of the replication-transcription complexes, and the location of viral RNA synthesis, with particular focus on the dynamics of the coronavirus replicative structures and individual replication-associated proteins.

Figure 1

Schematic representation of the coronavirus mouse hepatitis virus (MHV)-A59 genome, replicase polyprotein organization and membrane topology. (A) Schematic representation of the +RNA genome of MHV-A59. The coronavirus genome contains a 5' cap structure and a 3' poly(A) tail, together with untranslated regions (UTRs). The first two-thirds of the genome consist of two large open-reading frames (ORFs), ORF1a and ORF1b, which are translated into two large replicase polyproteins (pp1a and pp1ab). Pp1ab is synthesized via a −1 ribosomal frameshift mechanism at the end of ORF1a (RFS). The final one-third of the genome contains the canonical CoV structural proteins-encoding genes (S, E, M and N), interspaced by several accessory genes (2a, HE, 4, 5a); (B) A schematic representation of pp1ab is shown. The coronavirus polyproteins are processed by viral proteinases residing in nsp3 (PLpro1 and PLpro2; grey arrowheads indicate cleavage sites) and nsp5 (Mpro; black arrowheads indicate cleavage sites), thereby generating 16 mature nsps. Hydrophobic domains (TM1, TM2 and TM3) in nsp3, nsp4 and nsp6 are indicated, together with predicted and identified RNA(-modifying) enzymes: the RNA-dependent RNA polymerase (RdRP; nsp12), the helicase (Hel; nsp13), the exonuclease (ExoN; nsp14), the uridylate-specific endoribonuclease (N; nsp15), and the methyl transferase (MT; nsp16); (C). Schematic representation of the topology of the coronavirus polyprotein. Only the part of the polyprotein is shown that contains the hydrophobic domains (indicated by boxes) residing in nsp3, nsp4 and nsp6. Nsp3 and nsp6 contain hydrophobic domains that do not span the lipid bilayer. Mpro indicates the viral protease residing in nsp5, in between nsp4 and nsp6.

Figure 2

Coronavirus-induced organelle-like replicative structures. (A) Upon coronavirus infection, replicative structures consisting of double-membrane vesicles (DMVs) and convoluted membranes (CMs) are generated; (B) A higher magnification clearly demonstrates that the DMVs contain a double-lipid bilayer. The DMVs are indicated by arrowheads and the CMs by asterisks. The size of the scale bars is indicated.