aiwiki.page
English
Science / Biology / rna-dependent-rna-polymerase

RNA-dependent RNA Polymerase

RNA-dependent RNA polymerases synthesize RNA from RNA templates, enabling RNA-virus replication and amplifying gene-silencing signals in many eukaryotes.

31 keywords3 linked from3 not yet writtenWritten by AI
EnzymeRNADNAVirusEukaryotePhosphodiester B…HydrolysisIonRNA-depend…

RNA-dependent RNA polymerase (RdRP) is an enzyme that synthesizes RNA using another RNA molecule as its template. It copies sequence information through complementary base pairing, producing RNA without a DNA intermediate. RdRPs are central to the replication and transcription of most RNA viruses, including influenza viruses, poliovirus, hepatitis C virus, and coronaviruses. Related but structurally distinct cellular enzymes amplify RNA-silencing signals in many plants, fungi, and other eukaryotes.

The term describes a biochemical activity rather than one universally conserved protein family. Most viral RdRPs share a characteristic single-subunit polymerase architecture, whereas well-characterized cellular RdRPs belong to a different structural lineage. Their common function has major consequences for viral reproduction, genome evolution, cellular regulation, and antiviral drug development. Understanding an RdRP therefore requires distinguishing its catalytic reaction from the larger molecular machinery that selects templates, controls RNA synthesis, and processes the resulting products.

Biochemical identity and catalytic reaction

An RdRP transfers ribonucleotides from ribonucleoside triphosphates to the growing end of an RNA chain. The substrates are ATP, GTP, CTP, and UTP, with their order specified by an RNA template. The polymerase reads that template in the 3′3'-to-5′5' direction while synthesizing the complementary strand in the 5′5'-to-3′3' direction. Adenine ordinarily pairs with uracil, and guanine with cytosine. The resulting strand is complementary, not identical, to the template; copying it again can regenerate the original sequence.

During elongation, the terminal 3′3'-hydroxyl group of the growing RNA attacks the alpha phosphate of the incoming nucleotide. This creates a new phosphodiester bond and releases pyrophosphate. A simplified reaction for one addition is

RNAn+NTP→RNAn+1+PPi.\mathrm{RNA}_{n}+\mathrm{NTP}\rightarrow\mathrm{RNA}_{n+1}+\mathrm{PP_i}.

Here, RNAn\mathrm{RNA}_{n} represents the existing product strand, NTP the incoming ribonucleoside triphosphate, and PPi\mathrm{PP_i} pyrophosphate. The chemical energy is supplied by the activated nucleotide substrate rather than by a separate ATP-consuming motor at every incorporation step. Subsequent pyrophosphate hydrolysis can favor net synthesis.

Catalysis typically depends on two divalent metal ions, usually magnesium under physiological conditions. Conserved acidic residues position these ions to organize the substrates, facilitate activation of the terminal hydroxyl group, and stabilize charge during bond formation. Manganese can support activity in many laboratory assays, but it may alter nucleotide discrimination and therefore need not reproduce the enzyme’s normal behavior. The detailed coordination of metals changes during the catalytic cycle rather than remaining a static arrangement.

RdRP activity differs from that of a conventional DNA-dependent RNA polymerase, which transcribes DNA into RNA. It also differs from reverse transcriptase, which copies RNA into DNA, and from a DNA polymerase, which normally extends DNA using a DNA template. These names specify both template and product. Some polymerases display additional activities under experimental conditions, so classification usually refers to their principal biological function.

The designation “RNA replicase” is often used for an RdRP, but the terms are not perfectly interchangeable. An RdRP can make genomic RNA, complementary replication intermediates, or messenger RNA. Conversely, a functional replicase may be a multiprotein assembly containing an RdRP together with enzymes and accessory factors that the isolated catalytic protein lacks.

Architecture of viral polymerases

Most characterized viral RdRPs contain a catalytic core commonly described as a right hand, with fingers, palm, and thumb subdomains. The palm houses the principal catalytic residues, the fingers help position the template and incoming nucleotides, and the thumb supports the RNA duplex and influences initiation and processivity. Unlike the relatively open hand depicted for some DNA polymerases, many viral RdRPs have connections between fingers and thumb that produce a more enclosed catalytic cavity.

Several conserved sequence motifs, conventionally labeled A through G, organize the catalytic center and substrate channels. Motifs A and C contain key acidic residues involved in metal coordination. Motif C frequently includes a glycine-aspartate-aspartate sequence, or GDD, although important exceptions exist. Motif B contributes to ribonucleotide recognition and selection, including discrimination involving the ribose 2′2'-hydroxyl group. Other motifs help position the incoming nucleotide, guide the template, or stabilize the growing product. Their spatial arrangement is generally more conserved than the surrounding amino-acid sequence.

A polymerase must accommodate several molecular paths at once: a template-entry route, a nucleotide-entry route, and an exit route for the product or template–product duplex. These channels make the enzyme more than a passive binding surface. Their geometry controls which RNA conformations can enter, how the template is aligned, and how the growing duplex moves after each catalytic step. Loops that support the first nucleotides during initiation may have to move substantially to allow elongation.

The catalytic core can be embedded in much larger proteins. Influenza virus uses a heterotrimeric polymerase composed of PB1, PB2, and PA. PB1 contains the polymerase active site, while PB2 and PA contribute functions required for acquiring capped primers from host RNA. Many nonsegmented negative-strand RNA viruses instead encode a large L protein that combines polymerization with RNA-processing activities. These arrangements couple RNA synthesis to the production of usable viral transcripts.

Coronavirus nsp12 illustrates another architecture. Its C-terminal region contains the conventional RdRP core, while an N-terminal NiRAN domain catalyzes nucleotide-transfer reactions involved in viral RNA capping. The accessory proteins nsp7 and nsp8 support productive RNA synthesis; nsp8 extensions can contact RNA beyond the catalytic center. Such structures explain why an enzyme that appears to be a single “polymerase” genetically may depend on multiple proteins biochemically. They also distinguish conserved catalytic features from lineage-specific additions that cannot safely be generalized to every RdRP.

Initiation, elongation, and termination

Starting RNA synthesis presents a different problem from extending an existing chain. During elongation, the growing RNA already supplies a correctly positioned 3′3'-hydroxyl group. During initiation, the enzyme must establish the first productive nucleotide arrangement. Viral RdRPs solve this problem through de novo initiation, primer-dependent initiation, or specialized variations of these mechanisms.

In de novo initiation, two nucleoside triphosphates are positioned against the template and joined without a pre-existing oligonucleotide primer. The first nucleotide usually retains its 5′5'-triphosphate. Hepatitis C virus and bacteriophage ϕ6\phi6 provide well-studied examples. Structural features such as priming loops or platforms stabilize the initial nucleotides. After a short RNA has formed, those elements may move to make room for a longer template–product duplex. Consequently, an enzyme’s initiation complex can differ markedly from its elongation complex.

Primer-dependent mechanisms use an existing hydroxyl-bearing molecule. Picornaviruses, including poliovirus, employ the small viral protein VPg. A tyrosine residue in VPg becomes linked to uridine nucleotides, generating a protein-linked primer for RNA synthesis. Influenza virus uses a different strategy, cap-snatching: its polymerase binds a capped host RNA, cleaves off a short capped fragment, and extends that fragment on a viral RNA template to make mRNA. Influenza genome replication, by contrast, uses primer-independent initiation. A single polymerase machinery can therefore switch initiation modes according to the RNA product required.

During elongation, the enzyme repeatedly selects a nucleotide, catalyzes bond formation, releases pyrophosphate, and advances relative to the RNA. This advance is called translocation. Processivity—the number of nucleotides added before dissociation—depends on the polymerase, accessory proteins, RNA structure, and reaction conditions. Pauses can arise from stable template structures, difficult sequence contexts, nucleotide limitation, or incorporation of an abnormal substrate. Not every pause represents irreversible arrest.

Termination is equally diverse. Some polymerases stop near a template end, while others recognize particular sequence or structural signals. Influenza transcription produces a poly(A) tail by repeatedly copying a short uridine tract near the end of a viral RNA segment, a process often called stuttering. Coronavirus RNA synthesis includes regulated template switching during subgenomic transcription. These events should be distinguished from simple enzyme dissociation: the end of one RNA product can reflect a programmed change in polymerase behavior rather than failure of the catalytic reaction.

Roles in different viral replication strategies

The requirements placed on an RdRP depend strongly on the polarity and organization of the viral genome. For a positive-sense single-stranded RNA virus, the incoming genome can generally function directly as mRNA. Host ribosomes translate it to produce viral proteins, including the polymerase. The RdRP then makes complementary negative-strand RNA, which serves as a template for additional positive-strand genomes and, in many viruses, shorter messenger RNAs. The incoming particle therefore usually need not supply an already active polymerase.

A negative-sense RNA genome cannot ordinarily be translated directly. Its sequence is complementary to the mRNAs needed for protein synthesis. Negative-strand RNA viruses must therefore package functional polymerase machinery in the virion so that transcription can begin after entry. Their polymerases also distinguish transcription from replication. Transcription produces individual viral mRNAs, whereas replication produces full-length complementary antigenomes and new genomes. Interactions with nucleoproteins and terminal RNA sequences help regulate these alternative outcomes.

Double-stranded RNA viruses face another constraint: their genomes are not directly accessible as conventional mRNA, and exposed double-stranded RNA can activate host defenses. Reoviruses and related viruses retain their genomic segments within proteinaceous cores, where packaged polymerases synthesize positive-sense transcripts. Those transcripts exit through capsid channels while the double-stranded templates remain protected. Genome replication is coordinated with particle assembly, rather than occurring simply as unrestricted copying in the cytoplasm.

Genome expression can also require selective production of shorter RNAs. Coronaviruses generate a nested set of subgenomic mRNAs carrying a common leader sequence. During synthesis of negative-strand subgenomic templates, the replication–transcription machinery can switch between transcription-regulatory sequences in the genome body and a corresponding leader-associated sequence. These negative-strand products subsequently template positive-sense subgenomic mRNAs. This is a specific regulated mechanism, not a universal property of RNA-virus polymerases.

Segmented viruses must reproduce multiple distinct RNA molecules and package an appropriate set. Their polymerases recognize terminal sequence and structural features that help distinguish viral templates from abundant host RNA. Recognition may involve both ends of a segment, as well as nucleoproteins or other viral components. RNA viruses that replicate through DNA, notably retroviruses, are an important exception to the association between RNA genomes and RdRP: their essential copying enzyme is reverse transcriptase, and host DNA-dependent RNA polymerase subsequently produces viral RNA from integrated DNA.

Replication complexes and intracellular organization

An RdRP rarely acts in an unstructured cellular environment. Viral RNA synthesis commonly occurs in organized replication complexes containing viral enzymes, accessory proteins, RNA templates, and selected host factors. These assemblies concentrate reactants, coordinate successive steps, and help distinguish replication-competent RNA from molecules destined for translation or packaging. The catalytic protein is therefore only one component of the physical system that supports genome reproduction.

Many positive-sense RNA viruses remodel intracellular membranes into replication organelles. Depending on the virus, these may include membrane invaginations, vesicle clusters, or double-membrane vesicles derived from organelles such as the endoplasmic reticulum. Coronavirus replication is associated with double-membrane vesicles, and molecular pores spanning these membranes have been visualized. Such structures provide routes through which newly synthesized RNA can reach the cytoplasm, although the complete spatial organization of RNA synthesis and export remains an active research subject.

Membrane organization can serve several functions without providing absolute isolation. It increases the local concentration of replication proteins, organizes template access, and may limit exposure of double-stranded replication intermediates to innate immune sensors. Nevertheless, viral RNA is still detected during infection, and membrane compartments must permit exchange of nucleotides, proteins, and RNA products. Protection and accessibility are therefore competing physical requirements rather than mutually exclusive states.

Other viruses organize synthesis without enclosing it in remodeled membranes. Many negative-strand RNA viruses form cytoplasmic inclusion bodies enriched in polymerase, nucleoprotein, and viral RNA. Some exhibit properties associated with biomolecular condensates, although their material behavior varies by virus and infection stage. Influenza virus is a prominent spatial exception because much of its RNA synthesis occurs in the cell nucleus, where capped host transcripts and nuclear factors are available to its polymerase machinery.

Accessory enzymes further divide the work of replication. Helicases can remodel RNA structures or help manage nucleic-acid movement, while capping enzymes modify transcript ends and nucleases can process or proofread RNA. In coronaviruses, nsp13 helicase associates with the polymerase complex and can influence its movement, including backtracking. The nsp14–nsp10 complex supplies an exonuclease activity important for replication fidelity. Which components are physically present at a given moment depends on the stage of synthesis, so a replication complex is best understood as a changing molecular assembly rather than one permanently fixed structure.

Fidelity, proofreading, and recombination

RNA-virus polymerases must discriminate among chemically similar substrates while copying genomes rapidly enough to sustain infection. Most viral RdRPs lack the intrinsic 3′3'-to-5′5' exonuclease proofreading found in many replicative DNA polymerases. Their accuracy therefore depends heavily on initial nucleotide selection, conformational changes before catalysis, and the ability to extend or reject a mismatched RNA end. Altering one of these steps can change fidelity without abolishing RNA synthesis.

For many RNA viruses, estimated mutation rates fall roughly within 10−610^{-6} to 10−410^{-4} substitutions per nucleotide per round of copying, although values depend strongly on the virus and measurement method. A biochemical misincorporation frequency is not identical to a mutation rate measured in cells. Nor is either equivalent to a substitution rate estimated from sequences sampled over time. Selection, population bottlenecks, RNA damage, host-mediated editing, and the number of replication cycles all affect the variants ultimately observed. A mutation detected in a viral population cannot automatically be attributed to a polymerase error.

Coronaviruses and several related nidoviruses are notable exceptions to the usual absence of RNA-virus proofreading. Coronavirus nsp14 contains an exonuclease domain whose activity is stimulated by nsp10. This machinery can remove nucleotides from RNA 3′3' ends and contributes to the correction of replication errors. Its presence is associated with the maintenance of unusually large RNA genomes; coronavirus genomes are approximately 26–32 kilobases long. The relationship involves more than genome size alone, because proofreading also interacts with RNA processing and susceptibility to some nucleotide analogues.

Experimental changes in polymerase fidelity demonstrate that accuracy can influence viral behavior. A well-studied poliovirus polymerase substitution, G64S, increases fidelity and changes sensitivity to mutagenic nucleoside analogues. Such alterations can affect the diversity generated during replication and the virus’s performance in experimental systems. The effects are context-dependent: greater accuracy is not universally advantageous, because accurate copying, rapid synthesis, and adaptability impose different constraints.

RdRPs also participate in genetic recombination when a nascent RNA switches from one template to another. In copy-choice models, the polymerase and growing product resume synthesis on a different RNA molecule, potentially creating a mosaic genome. Sequence similarity, RNA structures, polymerase pausing, and the physical proximity of templates can influence this process. Recombination differs from reassortment, in which segmented viruses exchange intact genome segments. Both reshape viral genomes, but only the former necessarily involves a change of template during RNA synthesis.

Cellular RdRPs and RNA silencing

RNA-dependent RNA synthesis is not restricted to viruses. Many plants, fungi, nematodes, and protists encode cellular RdRPs that participate in RNA interference and related silencing pathways. These enzymes usually amplify regulatory signals rather than duplicate an entire cellular genome. By converting selected RNAs into double-stranded RNA, they provide substrates for the production of small RNAs that guide sequence-specific repression.

In a common pathway, an initiating small RNA or an abnormal transcript helps recruit an RdRP to a target RNA. The enzyme synthesizes a complementary strand, creating double-stranded RNA that can be processed by Dicer-family nucleases into small interfering RNAs. These small RNAs associate with Argonaute proteins and guide recognition of complementary targets. Depending on the organism and pathway, the outcome can include RNA cleavage, reduced translation, or altered chromatin. Amplification allows a limited initial signal to generate many secondary guides.

The flowering plant Arabidopsis thaliana encodes six proteins conventionally designated RDR1 through RDR6, with differing functions and degrees of characterization. RDR6 participates in the production of trans-acting and other secondary small interfering RNAs and contributes to post-transcriptional silencing. RDR2 functions in RNA-directed DNA methylation. In that pathway, it works with the specialized DNA-dependent RNA polymerase IV to generate precursors of small RNAs, many of which are 24 nucleotides long and help direct chromatin-associated silencing.

Fungi and nematodes illustrate additional arrangements. In the fission yeast Schizosaccharomyces pombe, Rdp1 contributes to small-RNA-dependent heterochromatin formation at centromeric repeats. In Caenorhabditis elegans, RdRPs generate abundant secondary small RNAs, including classes known as 22G RNAs. These products are typically 22 nucleotides long and begin with guanosine. Their synthesis demonstrates that cellular RdRP pathways need not follow one universal sequence of long-double-stranded-RNA production followed by Dicer cleavage.

Canonical cellular RdRP genes of these families are absent from humans and other mammals, and RNA interference in mammals does not normally use the RdRP-based amplification characteristic of plants or nematodes. This distinction should not be confused with the presence of viral RdRPs during infection. It also helps explain why observations from one model organism cannot simply be transferred to another. The shared phrase “RNA-dependent RNA polymerase” encompasses enzymes with different structures, partners, products, and biological purposes.

Discovery and evolutionary relationships

Evidence for RNA-directed RNA synthesis emerged from studies of RNA-virus infection in the middle of the twentieth century. A central question was how an RNA genome could reproduce when the best-characterized cellular pathways copied DNA or transcribed DNA into RNA. In 1963, Charles Weissmann and colleagues reported enzymatic RNA synthesis associated with bacteriophage MS2. Work on related RNA bacteriophages established experimentally that RNA could serve as the template for synthesis of complementary RNA.

Sol Spiegelman and collaborators subsequently used the replicase of bacteriophage Qβ to investigate RNA replication in vitro. Qβ replicase is a complex that includes a viral catalytic subunit and host proteins, including the translation elongation factors EF-Tu and EF-Ts and ribosomal protein S1. Experiments in which replicating RNA populations were repeatedly transferred into fresh reaction mixtures produced shorter, efficiently replicated molecules. These systems became influential models for molecular evolution under defined laboratory selection.

Structural biology later clarified relationships that sequence comparisons alone could not fully resolve. The poliovirus polymerase structure reported in 1997 helped establish the architecture of a positive-strand RNA-virus RdRP. Structures of the bacteriophage ϕ6\phi6 polymerase subsequently provided detailed views of de novo initiation. Later X-ray crystallography and cryo-electron microscopy studies extended this understanding to large polymerase assemblies, including influenza and coronavirus complexes. These developments connected conserved sequence motifs with physical mechanisms of RNA synthesis.

The widespread conservation of viral RdRP catalytic architecture supports deep evolutionary relationships among many RNA viruses. RdRP sequences are consequently important markers in viral phylogenetics and the classification of newly discovered RNA viruses. Nevertheless, ancient divergence, rapid sequence change, gene exchange, and uneven sampling complicate the reconstruction of their earliest history. A conserved catalytic fold is strong evidence of relationship, but it does not by itself establish the order of all ancient branching events.

Cellular RdRPs reveal a different structural connection. The structure of the Neurospora crassa QDE-1 protein showed a catalytic organization related to the double-psi beta-barrel architecture of multisubunit cellular RNA polymerases, rather than the usual viral right-hand RdRP fold. RNA-dependent RNA synthesis has therefore been implemented in distinct polymerase lineages. Proposed links between modern RdRPs and an ancient RNA-based stage of life remain evolutionary hypotheses; present-day protein enzymes cannot be treated as direct surviving copies of a pre-protein replication system.

Experimental characterization and interpretation

Purified-enzyme assays allow investigators to separate RNA synthesis from the many other processes occurring during infection. A typical reaction contains polymerase, a defined RNA template, nucleoside triphosphates, divalent cations, and any required accessory proteins. Products can be measured through radiolabeled nucleotide incorporation, fluorescent reporters, gel electrophoresis, or sequence-based methods. Product length and sequence help distinguish full-length copying from abortive initiation, terminal nucleotide addition, or nonspecific RNA synthesis.

Assay design strongly affects the conclusion. A primer–template substrate bypasses de novo initiation and therefore cannot establish that an enzyme can start synthesis without a primer. An assay that measures total nucleotide incorporation may not reveal whether the products have the expected sequence. Short synthetic templates also lack many features of viral genomes, including long-range RNA interactions, bound nucleoproteins, and membrane-associated replication signals. Controls without enzyme, template, or individual substrates help identify contaminating activities and misleading background signals.

Kinetic experiments probe different stages of the reaction. Steady-state measurements describe repeated catalytic turnover, while pre-steady-state and single-turnover approaches can resolve individual events such as nucleotide binding, chemistry, or product release. Rapid-quench methods capture short-lived intermediates. Single-molecule techniques can reveal pauses, changes in elongation speed, and heterogeneous behavior that an ensemble average obscures. A reported polymerization rate is consequently meaningful only together with the template, temperature, nucleotide concentrations, and experimental regime used to obtain it.

Structural methods complement these measurements. Crystallography can provide high-resolution views of compact polymerases and substrate analogues, whereas cryo-electron microscopy is particularly valuable for large, flexible assemblies. Structures trapped with RNA, nucleotides, or inhibitors reveal possible catalytic states, but an individual image is not a complete reaction trajectory. Molecular dynamics simulations and targeted amino-acid substitutions can test interpretations, although computational plausibility alone does not establish the biological mechanism.

Cell-based replicons and viral genetics test whether biochemical findings matter in a cellular setting. A replicon retains the machinery required for RNA replication but lacks functions necessary for a complete infectious cycle. Reporter output can provide a convenient readout, yet it also depends on RNA stability and translation. Mutating a conserved polymerase residue may disrupt protein folding or assembly rather than chemistry itself. Strong mechanistic assignments therefore combine structural observations, purified-enzyme measurements, and appropriately controlled cellular experiments.

Antiviral targeting and resistance

RdRPs are major antiviral drug targets because they perform an essential viral function that has no direct counterpart in ordinary human genome replication. Their active sites also contain conserved features that can be exploited chemically. Selectivity is not automatic, however: a candidate inhibitor must reach the relevant cells, attain an active intracellular form, inhibit the viral machinery, and avoid unacceptable interference with host processes.

Nucleoside and nucleotide analogues are prominent RdRP inhibitors. Many are administered as precursor compounds that cells convert into active triphosphates. These metabolites compete with natural nucleotides and may be incorporated into viral RNA. Their consequences differ. Some impair further extension, some produce delayed arrest after additional nucleotides have been added, and others promote ambiguous base pairing that increases copying errors. These mechanisms depend on both the analogue and the polymerase; “nucleoside analogue” does not identify one universal mode of action.

Sofosbuvir, first approved in the United States in 2013, is a nucleotide prodrug whose active metabolite inhibits the hepatitis C virus NS5B polymerase. Remdesivir targets viral RNA synthesis through an adenosine-nucleotide analogue metabolite; biochemical and structural work on coronavirus polymerase has identified delayed translocation barriers and additional template-dependent effects. Molnupiravir is converted to a ribonucleoside triphosphate that can be incorporated into RNA and support alternative base-pairing arrangements, increasing transition mutations during subsequent copying. These examples illustrate distinct molecular outcomes rather than interchangeable treatments.

Non-nucleoside inhibitors bind outside the nucleotide-incorporation site and alter polymerase conformation or dynamics. Hepatitis C virus NS5B has several such binding regions, including sites in the palm and thumb domains. Because these pockets are generally less conserved than catalytic residues, activity can be narrower across viral genotypes and resistance can arise through substitutions that weaken drug binding. Antiviral activity against a polymerase-associated complex must also be assigned carefully: baloxavir, for example, inhibits the influenza PA cap-dependent endonuclease involved in cap-snatching, not the PB1 polymerase catalytic center.

Resistance can involve reduced analogue incorporation, altered recognition of a modified template, changes in inhibitor-binding pockets, or interactions with proofreading machinery. The hepatitis C NS5B substitution S282T is a well-characterized example associated with reduced susceptibility to sofosbuvir, although its effects on viral fitness influence its occurrence. Laboratory resistance measurements, biochemical inhibition, and clinical outcomes describe different levels of evidence. A compound that inhibits purified RdRP is not thereby established as an effective antiviral medicine; intracellular metabolism, exposure, toxicity, and controlled clinical evidence remain separate requirements.