RNA polymerase is the enzyme that reads a DNA template strand and assembles a complementary strand of RNA, one nucleotide at a time. That single job, copying genetic information from DNA into RNA, is transcription. But describing RNA polymerase as a mere copying machine sells it short. The enzyme unwinds DNA, selects the correct building blocks, proofreads its own work, pauses strategically, and coordinates with dozens of other proteins that process the RNA even as it emerges. Understanding how it accomplishes all of this reveals why transcription is far more than a passive readout of genetic code.
The Chemistry at the Active Site
At its core, RNA polymerase catalyzes a single chemical reaction over and over: it joins an incoming ribonucleotide to the growing RNA chain by forming a bond between the new nucleotide and the one already at the end. The energy for this comes from the incoming nucleotide itself, which arrives carrying extra phosphate groups that get clipped off during the reaction. What makes this possible at a molecular level is a pair of magnesium ions sitting in the enzyme’s active center. These two metal ions work together, positioning the reactants and stabilizing the transition state so the bond forms efficiently. The same pair of ions also participates when the enzyme needs to cut the RNA chain during error correction, essentially switching roles depending on whether RNA is being built or trimmed.
1PubMed Central. Unified two-metal mechanism of RNA synthesis and degradation by RNA polymeraseHow Transcription Begins
Before RNA polymerase can start copying, it has to find the right stretch of DNA to read. Genes don’t come with neon signs, so the enzyme relies on helper proteins that recognize specific sequences called promoters. In bacteria, a detachable protein called a sigma factor binds to RNA polymerase and guides it to the promoter. The sigma factor is responsible for recognizing the promoter sequence, prying the two DNA strands apart at that spot, and getting the first few nucleotides of RNA assembled. Once transcription is underway, the sigma factor lets go, and the core enzyme continues on its own.
2PubMed Central. Bacterial Sigma Factors and Anti-Sigma Factors: Structure, Function and DistributionEukaryotic cells, which include everything from yeast to humans, handle initiation differently. Instead of a single sigma factor, RNA polymerase II (the version that transcribes protein-coding genes) assembles into a large preinitiation complex with more than 30 proteins and general transcription factors. Structural studies have captured this entire complex and show it weighing in at roughly 1.5 million daltons, a molecular machine of staggering size built just to get transcription started at the right place.
3PubMed Central. Structure of an RNA polymerase II preinitiation complexPrying Open the Double Helix
DNA is a double helix, and to read one strand, RNA polymerase needs to separate the two strands locally. This step, called promoter melting, is driven by the enzyme’s own thermal motions rather than by burning a chemical fuel like ATP. The enzyme essentially rocks and flexes on the DNA until the strands come apart at the promoter. Researchers have tracked these motions and found that they explain how RNA polymerase searches along DNA and then melts it open without any external energy source.
4PubMed Central. RNA polymerase motions during promoter meltingOnce the strands are separated, about 12 to 14 base pairs of DNA remain unwound in a “transcription bubble” that travels with the enzyme. The template strand threads through the active site, where incoming nucleotides are matched to it by standard base-pairing rules. The non-template strand is held aside until the enzyme has passed, at which point the two DNA strands re-anneal behind it.
Moving Along the DNA During Elongation
After initiation, RNA polymerase enters the elongation phase. It advances along the template one nucleotide at a time in a cycle of adding a nucleotide, then stepping forward by exactly one position. This stepping motion, called translocation, is not a simple slide. Detailed simulations of RNA polymerase II show that a structural element inside the enzyme called the bridge helix actively pushes the RNA-DNA hybrid forward, and specific amino acid residues on the bridge helix interact directly with the nucleic acids to accelerate the step.
5PubMed Central. Millisecond dynamics of RNA polymerase II translocation at atomic resolutionThe translocation cycle also involves brief backward and forward fluctuations. The enzyme does not march in a perfectly smooth line; it jitters. Research on the human mitochondrial RNA polymerase, a structurally different single-subunit enzyme, suggests that the binding of the next nucleotide is what locks the enzyme in its forward position, a process sometimes described as a Brownian ratchet. Without that next nucleotide clicking into place, the enzyme tends to drift back, and translocation proceeds without productive result.
6PubMed. Collective Variables and Facilitated Conformational Opening during Translocation of Human Mitochondrial RNA Polymerase (POLRMT) from Atomic SimulationsCatching Mistakes Through Backtracking
RNA polymerase does not have the same elaborate proofreading machinery that DNA polymerase uses during DNA replication, but it is not error-blind either. When a wrong nucleotide gets incorporated, the mismatched base pair destabilizes the enzyme’s grip, and RNA polymerase can slide backward along the DNA and RNA. This reverse movement, called backtracking, pushes the most recently added stretch of RNA out through a secondary channel in the enzyme.
7PubMed. Backtracking and proofreading in DNA transcriptionOnce backtracked, the enzyme can recover in two ways. For shallow backtracks of just a few nucleotides, it often diffuses forward again on its own. For deeper backtracks, the protruding RNA tail gets cleaved off, which resets the enzyme’s active site to a fresh 3′ end and allows elongation to resume. In eukaryotes, this cleavage is strongly enhanced by a factor called TFIIS, which enables RNA polymerase II to recover from backtracks of essentially any depth. Without TFIIS, deep backtracks can become dead ends. A related subunit called A12.2 plays an analogous role in RNA polymerase I, though it works somewhat differently: rather than enabling cleavage at any depth, it slows the rate of backward diffusion and supports cleavage of up to about 20 nucleotides.
8PubMed Central. Mechanisms of backtrack recovery by RNA polymerases I and IIBacktracking-based proofreading produces error rates that match what researchers observe in living cells, so while it is not as precise as DNA replication, it keeps transcription accurate enough that mistakes in RNA rarely cause serious problems. Cells can afford a few defective RNA copies because each gene is transcribed many times, and individual bad copies get diluted out or degraded.
7PubMed. Backtracking and proofreading in DNA transcriptionPausing as a Regulatory Tool
RNA polymerase does not always charge ahead at full speed. Certain DNA sequences cause the enzyme to pause, and these pauses are not accidents. Sequences with high G/C content followed by A/T-rich stretches can destabilize the RNA-DNA hybrid inside the enzyme, prompting it to backtrack briefly and stall. The RNA transcript itself can also fold into hairpin structures that physically jam the exit channel.
9PubMed Central. Pause & Go: from the discovery of RNA polymerase pausing to its functional implicationsThese pauses serve real biological purposes. In bacteria, pausing gives regulatory proteins time to catch up with the polymerase and bind the emerging RNA before it folds into the wrong shape. A well-studied example involves the trp operon in Bacillus subtilis, where a pause at a specific position gives a regulatory protein called TRAP enough time to grab the nascent transcript and fold it into a structure that blocks translation of a downstream gene. When that pause is disrupted by mutations, the gene gets overexpressed because TRAP cannot bind fast enough.
10PubMed. RNA polymerase pausing regulates translation initiation by providing additional time for TRAP-RNA interactionHow Transcription Ends
Stopping transcription cleanly is just as important as starting it. In bacteria, termination occurs by two major pathways. Intrinsic termination relies on a hairpin structure that forms in the RNA near a stretch of uracil residues. The hairpin destabilizes the elongation complex, and the weak bonds holding the RNA to the DNA template let go, releasing the transcript. The second pathway uses a protein called Rho that threads onto the RNA and chases down the polymerase, pulling it off the DNA. Additional factors like NusA and NusG fine-tune both pathways: NusA helps weak terminator hairpins form, NusG stabilizes pauses at termination sites to extend the window for hairpin formation, and Rho can even prevent competing RNA structures from interfering with the terminator hairpin.
11PubMed Central. Factor-stimulated intrinsic termination: getting by with a little help from some friendsEukaryotic termination is more complex and varies depending on which RNA polymerase is involved, but the general principle is similar: signals in the DNA or RNA trigger disassembly of the transcription machinery and release of the finished transcript.
Three Eukaryotic Polymerases, Three Different Jobs
Bacteria get by with a single type of RNA polymerase. Eukaryotes divide the work among three. RNA polymerase I is dedicated to making ribosomal RNA, the structural backbone of ribosomes. RNA polymerase II handles messenger RNA (the transcripts that encode proteins) along with several types of small regulatory RNAs. RNA polymerase III transcribes transfer RNAs and 5S ribosomal RNA.
12PubMed. Structural differentiation of the three eukaryotic RNA polymerasesAll three share a common structural core and clearly descend from the same ancestral enzyme, but they have evolved distinct features suited to their particular workloads. RNA polymerase I, for instance, transcribes only a handful of genes but does so at extremely high rates to keep up with the cell’s demand for ribosomes. Its unique properties reflect that specialization.
13PubMed Central. Functional divergence of eukaryotic RNA polymerases: unique properties of RNA polymerase I suit its cellular rolePlants push the count even higher. In addition to Pol I, II, and III, plants have evolved two extra RNA polymerases, Pol IV and Pol V, that produce non-coding RNAs involved in gene silencing. These polymerases generate small transcripts that guide the cell’s machinery to add chemical tags (methyl groups) to DNA, shutting down specific genes. They appear to have diverged from Pol II over evolutionary time and taken on entirely new regulatory functions.
14PubMed Central. RNA Pol IV and V in gene silencing: Rebel polymerases evolving away from Pol II’s rulesGetting Through Chromatin
In eukaryotic cells, DNA is wrapped around protein spools called histones, forming structures called nucleosomes. This packing presents a physical barrier: RNA polymerase II has to push through or displace nucleosomes to read the gene underneath. When a single polymerase encounters a nucleosome, it can transcribe through it while the histones stay largely in place, forming a transient DNA loop around the enzyme. But when multiple polymerases hit the same nucleosome in quick succession, they collectively strip the histones away entirely.
15PubMed Central. Mechanism of transcription through a nucleosome by RNA polymerase IIElongation factors help the polymerase push through. Structural work has shown that the factors Elf1 and Spt4/5 reshape the leading edge of the transcribing complex, reducing friction between the polymerase and the nucleosome. They prevent the enzyme from getting stuck at specific positions on the nucleosome where strong contacts would otherwise stall progress.
16PubMed. Structural insight into nucleosome transcription by RNA polymerase II with elongation factorsCryo-electron microscopy structures of the polymerase caught mid-nucleosome show that half a turn of DNA gets peeled off the histone surface as the enzyme approaches, feeding it into the active-site cleft. The polymerase grips the nucleosome at two contact points flanking its central cleft, effectively wrestling the DNA away from the histones as it reads through.
17Nature Communications. Structure of transcribing RNA polymerase II-nucleosome complexThe CTD Tail and Co-Transcriptional Processing
RNA polymerase II has a feature the other polymerases lack: a long, flexible tail on its largest subunit called the C-terminal domain, or CTD. In vertebrates, this tail consists of 52 repeats of a seven-amino-acid sequence, and it acts as a landing platform for the proteins that process the RNA transcript while it is still being made. The CTD gets tagged with phosphate groups at different positions during different stages of transcription, and each modification recruits a different set of processing factors.
18PubMed Central. The RNA polymerase II CTD coordinates transcription and RNA processingEarly in transcription, the fifth position in the repeat gets phosphorylated, which recruits the capping enzyme that adds a protective cap to the front end of the new RNA. As the polymerase moves into elongation, the second position gets phosphorylated by a different enzyme, and this change recruits the factors responsible for adding a poly-A tail to the RNA’s back end and cutting it loose from the polymerase. Disrupting this second phosphorylation scrambles the recruitment of polyadenylation factors and produces abnormal RNA ends.
19PubMed. Phosphorylation of serine 2 within the RNA polymerase II C-terminal domain couples transcription and 3′ end processingThis coupling of transcription with RNA processing is one of the reasons eukaryotic gene expression is so tightly controlled. The polymerase is not just making RNA; it is orchestrating the maturation of that RNA in real time, and the CTD phosphorylation cycle acts as a kind of molecular clock that times each processing step to the right phase of transcription.
An Ancient and Conserved Machine
Multisubunit RNA polymerases in bacteria, archaea, and eukaryotes all descend from a single common ancestor. The structural similarities are striking: bacterial RNA polymerase and eukaryotic RNA polymerase II share the same overall architecture, the same relative positioning of subunits, and even similar folding patterns within individual subunits.
20PubMed. RNA polymerase: structural similarities between bacterial RNA polymerase and eukaryotic RNA polymerase IIArchaeal RNA polymerases sit somewhere in between, sharing features with both bacterial and eukaryotic versions and providing a window into how the enzyme diversified as the three domains of life diverged.
21PubMed. Cycling through transcription with the RNA polymerase F/E (RPB4/7) complex: structure, function and evolution of archaeal RNA polymeraseNot all RNA polymerases are multisubunit machines, though. Human mitochondria, for example, use a single-subunit RNA polymerase called POLRMT that is distantly related to the RNA polymerases of certain bacterial viruses rather than to the cell’s own nuclear enzyme. This is a relic of mitochondria’s origin as engulfed bacteria, and the enzyme works with a small set of accessory proteins rather than the large preinitiation complexes seen in the nucleus.
22PubMed. Human mitochondrial RNA polymerase: structure-function, mechanism and inhibitionWhen a Stalled Polymerase Signals for DNA Repair
RNA polymerase II serves an unexpected second function beyond transcription: it acts as a damage sensor for DNA. When the enzyme runs into a lesion on the template strand, it stalls. That stalled polymerase triggers a repair pathway called transcription-coupled nucleotide excision repair, which specifically removes the damaged DNA and patches the gap. This ensures that actively transcribed genes, the ones the cell is using right now, get repaired faster than silent regions of the genome. Defects in this pathway are linked to diseases in which patients are extremely sensitive to DNA-damaging agents.
23PubMed Central. Differential processing of RNA polymerase II at DNA damage correlates with transcription-coupled repair syndrome severityDrugs That Target RNA Polymerase
Because RNA polymerase is essential for life, it is a prime target for antibiotics and toxins. Rifampicin, one of the cornerstone drugs for treating tuberculosis, works by physically plugging the RNA exit channel of bacterial RNA polymerase, preventing the growing transcript from extending past a few nucleotides. Because the bacterial and human enzymes are structurally distinct enough at the binding site, rifampicin shuts down bacterial transcription without harming the patient’s cells.
24PubMed. Low-molecular weight inhibitors of bacterial DNA-dependent RNA polymeraseOn the eukaryotic side, alpha-amanitin, the toxin in death cap mushrooms, targets RNA polymerase II specifically. It jams the enzyme’s bridge helix, the same structural element involved in translocation, preventing the enzyme from stepping forward along the DNA. A single meal of death cap mushrooms can deliver enough alpha-amanitin to effectively shut down messenger RNA production in the liver, which is why the poisoning is so often fatal. Understanding these inhibitors at a structural level has been instrumental in figuring out how the enzyme’s moving parts work during normal transcription.