often lack template specificity. The replication complex in the infected cell may contribute to template specificity by concentrating reaction components to create an environment that copies viral RNAs selectively. Replication of viral RNAs on membranous structures might contribute to such specificity (Chapter 14).
Unwinding the RNA Template
Base-paired regions in viral RNA must be disrupted to permit copying by RdRP. RNA helicases, which are encoded in the genomes of many RNA viruses, are thought to unwind the genomes of double-stranded RNA viruses, as well as secondary structures in template RNAs. They also prevent extensive base pairing between template RNA and the nascent complementary strand. The RNA helicases of several viruses that are important human pathogens, including the flaviviruses hepatitis C virus and dengue virus, have been studied extensively because they are potential targets for therapeutic intervention. To facilitate the development of new agents that inhibit these helicases, their three-dimensional structures have been determined by X-ray crystallography. These molecules comprise three domains that mediate hydrolysis of NTPs and RNA binding (Fig. 6.15). Between the domains is a cleft that is large enough to accommodate single-stranded but not double-stranded RNA. Unwinding of double-stranded RNA probably occurs as one strand of RNA passes through the cleft and the other is excluded.
The bacteriophage ϕ6 RNA polymerase can separate the strands of double-stranded RNA without the activity of a helicase. Examination of the structure of the enzyme suggests how such melting might be accomplished. This RdRP has a plow-like protuberance around the entrance to the template channel that is thought to separate the two strands, allowing only one to enter the channel.
Role of Cellular Proteins
Host cell components required for viral RNA synthesis were initially called “host factors,” because nothing was known about their chemical composition. Evidence that cellular proteins are essential components of a viral RdRP first came from studies of the bacteriophage Qβ enzyme. This viral RdRP is a multisubunit enzyme, consisting of a 65-kDa virus-encoded protein and four host proteins: ribosomal protein S1, translation elongation proteins (EF-Tu and EF-Ts), and an RNA-binding protein. Proteins S1 and EF-Tu contain RNA-binding sites that enable the RNA polymerase to recognize the viral RNA template. The 65-kDa viral protein has sequence and structural similarity to known RdRPs, but exhibits no RNA polymerase activity in the absence of the host proteins.
Figure 6.15 Structure of a viral RNA helicase. The RNA helicase of the flavivirus yellow fever virus is shown in surface representation, colored red, white, or blue depending on the distance of the amino acid from the center of the molecule. A model for melting of double-stranded RNA is shown (PDB file 1YKS).
Polioviral RNA synthesis also requires host cell proteins. When purified polioviral RNA is incubated with a cytoplasmic extract prepared from uninfected permissive cells, the genomic RNA is translated and the viral RNA polymerase is made. If the RNA synthesis inhibitor guanidine hydrochloride is included in the reaction, the polymerase assembles on the viral genome, but initiation is blocked. The RdRP-template assembly can be isolated free of guanidine, but RNA synthesis does not occur unless a new cytoplasmic extract is added, indicating that soluble cellular proteins are required for initiation. A similar conclusion comes from studies in which polioviral RNA was injected into oocytes derived from the African clawed toad Xenopus laevis: the viral RNA cannot replicate in Xenopus oocytes unless it is coinjected with a cytoplasmic extract from human cells. These observations can be explained by the requirement of the viral RNA polymerase for one or more mammalian proteins that are absent in toad oocytes.
One of these host cell proteins required for poliovirus RNA synthesis is poly(rC)-binding protein, which binds to a cloverleaf structure that forms in the first 108 nucleotides of the viral (+) strand RNA (Fig. 6.10). Formation of a ribonucleoprotein composed of the 5′ cloverleaf, 3CD, and poly(rC)-binding protein is essential for initiation of viral RNA synthesis. Interaction of poly(rC)-binding protein with the cloverleaf facilitates the binding of polioviral protein 3CD to the opposite side of the same cloverleaf.
Another host protein that is essential for polioviral RNA synthesis is poly(A)-binding protein 1. This protein brings together the ends of the viral genome by interacting with poly(rC)-binding protein 2, 3CDpro, and the 3′ poly(A) tail of poliovirus RNA (Fig. 6.10). Formation of this circular ribonucleoprotein is required for (−) strand RNA synthesis.
Interactions among cellular and viral proteins can now be identified readily by mass spectrometry, and their function in viral genome replication can be determined by silencing their production by RNA interference or disrupting the gene using CRISPR (clustered regularly interspaced short palindromic repeat)/Cas9. These approaches have been used to identify diverse cell proteins that participate in viral RNA-directed RNA synthesis in cells infected with a variety of (+), (–), and double-stranded RNA viruses.
Paradigms for Viral RNA Synthesis
Exact replicas of the RNA genome must be made for assembly of infectious viral particles. However, the mRNAs of most RNA viruses are not complete copies of the viral genome. The reproductive cycle of these viruses must therefore include a switch from mRNA synthesis to the production of full-length genomes. The majority of mechanisms for this switch regulate either the initiation or the termination of RNA synthesis.
(+) Strand RNA
The genome and mRNA of some (+) strand RNA viruses are identical. The genome RNAs of the Picornaviridae and Flaviviridae are translated upon entry into the cytoplasm to produce viral proteins, including the RdRP and accessory proteins. The (+) strand RNA genome is copied to a (−) strand, which in turn is used as a template for the synthesis of additional (+) strands (Fig. 6.1). Newly synthesized (+) strand RNA molecules can serve as templates for further genomic replication, as mRNAs for the synthesis of viral proteins, or as genomic RNAs to be packaged into progeny virions. Because picornaviral mRNA is identical in sequence to the viral RNA genome, all RNAs needed for the reproduction of these viruses can be made by a simple set of RNA synthesis reactions (Fig. 6.1). Such simplicity comes at a price, however, because synthesis of individual viral proteins cannot be regulated. However, polioviral gene expression can be controlled by the rate and extent of polyprotein processing. For example, the precursor of the viral RdRP, 3CD, cannot polymerize RNA, but is a protease that cleaves at certain Gln-Gly amino acid pairs in the polyprotein. Therefore, regulating the processing of the precursor 3CD controls the concentration of RNA polymerase.
The mechanisms of mRNA synthesis of other (+) strand RNA viruses allow structural and nonstructural proteins (generally needed in greater and lesser quantities, respectively) to be made separately. The nonstructural proteins are synthesized from full-length (+) strand (genomic) RNA, while structural proteins are translated from subgenomic mRNA(s). This strategy is characteristic of the replication cycles of coronaviruses, caliciviruses, and alphaviruses. Translation of the Sindbis virus (+) strand RNA genome yields the nonstructural proteins that synthesize a full-length (−) strand (Fig. 6.16). The (–) strand RNA molecules contain not only a 3′-terminal sequence for initiation of (+) strand RNA synthesis, but also an internal initiation site, used for
