The mechanisms by which viral mRNA is made and the RNA genome is replicated in cells infected by RNA viruses appear even more diverse than the structure and organization of viral RNA genomes (Fig. 6.1). Nevertheless, each mechanism of viral RNA synthesis meets two essential requirements common to all infectious cycles: (i) during replication, the RNA genome must be copied from one end to the other with no loss of nucleotide sequence; and (ii) viral mRNAs that can be translated efficiently by the cellular protein synthetic machinery must be made.
In this chapter, we consider the mechanisms of viral RNA synthesis, the switch from mRNA production to genome replication, and the origins of genetic diversity. Much of our understanding of viral RNA synthesis comes from experiments with purified components. Because it is possible that events proceed differently in infected cells, the results of such in vitro studies are used to build models for the different steps in RNA synthesis, which must then be tested in vivo. While many models exist for each reaction, those presented in this chapter were selected because they are consistent with experimental results obtained in different laboratories or have been validated with simplified systems in cells in culture. The general principles of RNA synthesis deduced from such studies are illustrated with a few viruses as examples.
The Nature of the RNA Template
Secondary Structures in Viral RNA
RNA molecules are not simple linear chains but can form secondary structures that are important for RNA synthesis, translation, and assembly (Fig. 6.2). Structural features in RNA are identified by scanning the nucleotide sequence with software designed to fold the nucleic acid into energetically stable structures. Comparative sequence analysis can predict RNA secondary structures. For example, comparison of the RNA sequences of several related viruses might establish that the structure, but not the sequence, of a stem-loop is conserved. Direct evidence for specific RNA structures comes from experiments in which RNAs are treated with enzymes or chemicals that attack single- or double-stranded regions specifically. The results of such analyses can confirm that the predicted stem regions are base paired while loops are unpaired. Structures of RNA hairpins and pseudoknots have been determined by X-ray crystallography or nuclear magnetic resonance (Fig. 6.2C).
PRINCIPLES Synthesis of RNA from RNA templates
Viral RNA genomes must be copied to provide both genomes for assembly into progeny virus particles and mRNAs for the synthesis of viral proteins.
Viral RNA genomes may be naked in the virus particle [typically (+) strand RNAs] or organized into nucleocapsids in which proteins are bound to the genomic RNAs.
Viral RNA-dependent RNA polymerases, like the other three types of nucleic acid enzymes, resemble a right hand consisting of palm, fingers, and thumb domains, with the active site located in the palm.
Some viral RNA polymerases can initiate RNA synthesis without a primer, while others are primer dependent.
Primers for viral RNA polymerases may be capped fragments of cellular pre-mRNAs or protein-linked nucleotides.
Specificity of viral RNA polymerases for viral RNAs is conferred by the recognition of RNA sequences or structures.
Host cell proteins are required for the activity of viral RNA polymerases.
The single-stranded RNA genome of hepatitis delta virus is copied by host cell DNA-dependent RNA polymerase, an exception to the rule that RNA viruses are copied by RNA-dependent RNA polymerases.
Viral RNA synthesis takes place in specific structures in the cell, either nucleocapsids, subviral particles, or membrane-bound replication complexes.
RNA synthesis is error prone, and this process, together with reassortment and recombination, yields diversity that is required for viral evolution.
TERMINOLOGY
What should we call RNA polymerases and the processes they catalyze?
Historically, viral RNA-dependent RNA polymerases were given two different names depending on their activities during infection. The term replicase was used to describe the enzyme that copies the viral RNA to produce additional genomes, while the enzyme that synthesizes mRNA was called transcriptase. In some cases, this terminology indicates true differences in the enzymes that carry out synthesis of functionally different RNAs, but for other RNA viruses, genomic replication and mRNA synthesis are the same reaction (see the figure). For double-stranded RNA viruses, mRNA synthesis produces templates that can also be used for genomic replication. As these formerly applied terms can therefore be inaccurate and misleading, they are not used here. The name RNA-dependent RNA polymerase (RdRP) will be used in this textbook to describe the enzymes that carry out genome replication and mRNA synthesis from viral RNA templates.
The production of mRNAs from viral RNA templates is often designated transcription. However, this term refers to a specific process, the copying of genetic information carried in DNA into RNA. Consequently, it is not used herein to describe synthesis of the mRNAs of viruses with RNA genomes; this process will be called mRNA synthesis. Similarly, use of the term promoter is reserved to designate sequences controlling transcription of DNA templates.
Naked or Nucleocapsid RNA
The genomes of (−) strand viruses are organized into nucleocapsids in which protein molecules, including the RdRP and accessory proteins, are bound to the genomic RNAs at regular intervals. These tightly wound ribonucleoproteins are very stable and generally resistant to RNase. The RdRPs of (−) strand viruses copy viral RNAs only when they are present in the nucleocapsid, such as that formed by the N protein of vesicular stomatitis virus bound to genomic RNA. In contrast, the genomes of many (+) strand RNA viruses are not coated with proteins in the virus particle [exceptions are the (+) strand RNA genomes of members of the Coronaviridae, Arteriviridae, and Retroviridae]. This difference is consistent with the fact that mRNAs are produced from the genomes of (−) strand RNA viruses upon cell entry, whereas the genomes of (+) strand RNA viruses are translated directly.
The viral nucleoproteins (NP) are cooperative, single-stranded-RNA-binding proteins, as are the single-stranded nucleic acid-binding proteins required during DNA-directed DNA synthesis. Their function during replication is to keep the RNA single stranded and prevent base pairing between the template and product, so that additional rounds of RNA synthesis can occur. The nucleoproteins of nonsegmented (−) strand RNA viruses have a two-lobe architecture that forms a positively charged groove that binds and shields the genomic RNA (Fig. 6.3). Interactions between nucleoproteins lock monomers tightly, resulting in rigid NP-RNA assemblies. The NP structures from segmented (−) strand RNA viruses are more varied and display less coordinated contacts between nucleoprotein subunits. These differences may explain why the NP-RNAs of these viruses are more susceptible to RNase digestion than those of nonsegmented (−) strand RNA viruses. The varied structures of the NP-RNA complexes also influence access of the viral RNA polymerase to the template. The RdRP of segmented (−) strand RNA viruses can bind the NP-RNA template directly, whereas those of nonsegmented (−) strand RNA viruses cannot: a phosphoprotein