RNA polymerase follows a three-stage sequence: it binds to a promoter and unwinds the DNA (initiation), builds an RNA strand one nucleotide at a time (elongation), then detaches when it hits a termination signal. Within each of those stages, there are precise substeps that follow the same order every single time a gene is transcribed.
Initiation: Binding, Unwinding, and First Nucleotides
Transcription begins when RNA polymerase locates a promoter, a specific DNA sequence near the start of a gene. In bacteria, the enzyme does this on its own with help from a sigma factor. In eukaryotes, the process requires an assembly line of helper proteins called general transcription factors. First, a protein called TBP (TATA-binding protein) recognizes the promoter and lands on the DNA. Then additional factors arrive in sequence, ultimately recruiting RNA polymerase II and positioning it correctly on the double-stranded DNA. Only after a final set of factors joins does the complex pry open the DNA and become ready to synthesize RNA.
Regardless of organism, the steps within initiation follow the same logic:
- Promoter recognition and binding. RNA polymerase (with its associated factors) makes initial contact with the promoter DNA, forming what’s called a closed complex because the two DNA strands are still paired together.
- DNA unwinding. The enzyme forces the two strands apart over a short stretch, roughly from position −10 to +2 relative to the transcription start site. This creates a “transcription bubble” of about 12 to 14 base pairs of single-stranded DNA, forming an open complex. Structural studies show that parts of the enzyme physically widen by about 15 ångströms to make room for this unwound DNA to slide into the active site.
- First RNA bonds. With the template strand exposed, incoming ribonucleotides pair with the DNA bases, and the enzyme links the first few nucleotides together. These early attempts are often “abortive,” meaning the enzyme makes short RNA fragments (around 2 to 9 nucleotides) and releases them before starting over. Only when it successfully synthesizes a stretch long enough to stabilize the complex does it commit to full transcription and release its hold on the promoter.
Elongation: The Nucleotide Addition Cycle
Once RNA polymerase clears the promoter, it enters elongation, which is where the bulk of the RNA molecule gets built. The enzyme moves along the DNA template strand in the 3′ to 5′ direction, synthesizing the new RNA in the 5′ to 3′ direction. Each nucleotide added follows a repeating four-step cycle:
- Translocation. The enzyme shifts forward by one base pair along the DNA.
- Nucleotide binding. The next matching ribonucleotide (ATP, UTP, GTP, or CTP) enters the active site. A flexible part of the enzyme called the trigger loop swings into a closed position, locking the nucleotide in place.
- Bond formation. A phosphodiester bond forms between the new nucleotide and the growing RNA chain. The closed trigger loop helps catalyze this reaction by precisely positioning the reactants and excluding water.
- Byproduct release. The trigger loop swings back open (a movement of about 15 to 25 ångströms), releasing pyrophosphate, the chemical byproduct of the bond-forming reaction. This reopening also clears the active site for the next nucleotide.
This cycle repeats for every single nucleotide in the transcript. In human cells, RNA polymerase II typically moves at speeds between 1,250 and 3,500 nucleotides per minute, though it often starts slower (around 500 nucleotides per minute for the first 10,000 to 15,000 bases of a gene) and accelerates further downstream. Bacterial RNA polymerase works at comparable speeds, typically around 40 to 80 nucleotides per second.
Error Correction During Elongation
RNA polymerase does not have the same high-fidelity proofreading system that DNA polymerase uses, but it is not helpless against mistakes. When the enzyme incorporates the wrong nucleotide, the mismatched base pair destabilizes the active site and causes the enzyme to stall. Rather than continuing forward, RNA polymerase slides backward along the DNA and RNA, a process called backtracking. This reverse movement pushes the mismatched nucleotide out of the active site and threads the 3′ end of the RNA through a secondary channel in the enzyme.
On its own, the enzyme can cleave the extruded RNA to remove one or two incorrect nucleotides and resume synthesis. For longer backtracks, bacteria use specialized rescue factors (called GreA and GreB) that stimulate the enzyme’s cleavage activity. GreA handles short backtracks of one or two nucleotides, the kind that typically result from a single misincorporation. GreB rescues polymerases that have slid further back. Eukaryotes have an analogous factor called TFIIS. After clipping away the error, the enzyme is repositioned with a correct 3′ end and can resume the addition cycle.
Termination: Signals That Stop Transcription
Elongation continues until RNA polymerase encounters a signal that triggers it to stop synthesizing RNA and release from the DNA. How this happens differs between bacteria and eukaryotes.
Bacterial Termination
Bacteria use two mechanisms. In intrinsic (rho-independent) termination, the RNA itself folds into a hairpin structure, a short stem-loop formed by complementary sequences near the end of the transcript, followed by a string of uracil residues. The hairpin destabilizes the enzyme’s grip on the RNA, and the weak bonds between the uracil-rich RNA and the adenine-rich DNA template let the transcript slip free. In rho-dependent termination, a ring-shaped protein called rho latches onto the RNA and chases the polymerase. When the polymerase pauses, rho catches up and physically pulls the RNA out of the enzyme, forcing dissociation.
Eukaryotic Termination
Eukaryotic termination is more varied because eukaryotes have three different RNA polymerases, each with its own mechanism. For RNA polymerase II, which transcribes protein-coding genes, termination is linked to the processing of the RNA’s tail end. When the enzyme transcribes past a polyadenylation signal (a sequence that marks where the messenger RNA should be cut and given its poly-A tail), the RNA is cleaved. This cleavage exposes a free end on the remaining RNA still attached to the polymerase. A specialized enzyme then chews through that leftover RNA from the exposed end, racing toward the polymerase. When it catches up, the polymerase is knocked off the DNA.
RNA polymerase III, which makes small RNAs like transfer RNA, uses a simpler system: it terminates at a stretch of thymine residues in the DNA coding strand, similar in principle to bacterial intrinsic termination. RNA polymerase I, which produces the precursor to ribosomal RNA, relies on specific protein factors that bind to terminator sequences downstream of the gene and block the polymerase’s path.
The Complete Sequence at a Glance
Putting it all together, the order of RNA polymerase’s actions runs like this: recognize and bind the promoter, unwind the DNA to form an open complex, synthesize the first short RNA fragments, clear the promoter to commit to elongation, repeatedly translocate and add nucleotides (correcting errors by backtracking when needed), and finally encounter a termination signal that causes the enzyme to release both the completed RNA and the DNA template. Every transcript in every cell follows this same fundamental sequence, whether the gene encodes a tiny transfer RNA or a protein-coding messenger RNA spanning millions of base pairs.