DNA is transcribed into RNA by an enzyme called RNA polymerase, which reads one strand of the DNA double helix and builds a complementary RNA copy, one nucleotide at a time, in three stages: initiation, elongation, and termination. This process is how your cells convert the genetic instructions stored in DNA into working molecules that can build proteins and carry out cellular functions.
What Transcription Actually Produces
Transcription creates a single-stranded RNA molecule using one strand of the DNA double helix as a template. The RNA copy follows the same base-pairing rules as DNA replication, with one key difference: wherever the DNA template has adenine, the RNA strand gets uracil instead of thymine. The four RNA building blocks are ATP, CTP, GTP, and UTP, and they’re linked together in a chain that grows from the 5′ end to the 3′ end.
Not all transcription produces the same type of RNA. In eukaryotic cells (everything from yeast to humans), three different versions of RNA polymerase handle different jobs. RNA polymerase I makes the large ribosomal RNAs that form the structural core of ribosomes. RNA polymerase II makes messenger RNA (mRNA) from protein-coding genes, which is the type most people mean when they talk about transcription. RNA polymerase III makes transfer RNAs (tRNAs) and the smallest ribosomal RNA, along with a few other small RNA molecules involved in splicing and protein transport.
Stage 1: Initiation
Transcription begins when RNA polymerase locates and binds to a specific DNA sequence called a promoter, which marks the starting point for a gene. How this happens differs between bacteria and more complex organisms.
In bacteria, the process is relatively straightforward. RNA polymerase slides along the DNA until it finds a promoter sequence, then binds tightly. A detachable piece of the enzyme called the sigma factor is responsible for recognizing these promoter signals. Once locked on, the polymerase pries open the double helix to expose a short stretch of nucleotides, joins the first two RNA building blocks together, and begins synthesizing the RNA chain.
In eukaryotic cells, initiation is more involved. RNA polymerase II can’t recognize the promoter on its own. It needs a team of helper proteins called general transcription factors to assemble at the promoter first. The process typically starts when a protein called TFIID binds to a short DNA sequence known as the TATA box, located about 35 to 50 nucleotides before the point where transcription will begin. Once the full initiation complex is assembled, another factor called TFIIH (which works as both a helix-opener and a chemical switch) pulls apart the two DNA strands and adds phosphate groups to the tail of RNA polymerase II. That chemical modification essentially releases the polymerase to start moving along the gene. Interestingly, many eukaryotic genes lack a classic TATA box entirely, yet their promoters still recruit the same binding protein and initiate transcription at comparable levels.
Stage 2: Elongation
Once RNA polymerase clears the promoter, it enters elongation, the phase where the bulk of the RNA molecule is actually built. The polymerase moves along the DNA template strand one nucleotide at a time, unwinding the double helix just ahead of itself and re-zipping it behind. At each position, it matches the correct RNA nucleotide to the exposed DNA base and adds it to the growing chain, powered by breaking high-energy chemical bonds on the incoming nucleotide.
Speed varies dramatically between organisms and even between individual genes. In bacteria, RNA polymerase adds roughly 50 nucleotides per second. In human cells, RNA polymerase II averages between 1,250 and 3,500 nucleotides per minute (about 20 to 58 per second), but the rate for individual genes can range from as low as roughly 6 nucleotides per second to nearly 60. This variation matters because transcription speed influences how the RNA folds, how it gets processed, and ultimately how much protein a gene produces.
Transcription is also far less accurate than DNA replication. RNA polymerase makes errors at a rate more than 10,000 times higher than DNA polymerase. Because each mRNA molecule is then used to produce many copies of a protein, a single transcription error can be amplified over a thousandfold through translation. Cells tolerate this because mRNA molecules are temporary. They’re used and then broken down, so errors don’t persist the way a DNA mutation would.
Stage 3: Termination
Termination is the process by which RNA polymerase stops transcribing, releases the finished RNA, and detaches from the DNA. Bacteria use two distinct mechanisms to accomplish this.
In intrinsic (or Rho-independent) termination, the RNA itself does the work. The newly made transcript contains a sequence that folds into a stable hairpin loop, followed by a string of uracil residues. This hairpin structure physically disrupts the connection between the polymerase, the DNA, and the RNA, causing the whole complex to fall apart.
In Rho-dependent termination, a ring-shaped protein called Rho latches onto the growing RNA at specific sequences and uses energy from ATP to chase after the polymerase. When the polymerase pauses, Rho catches up, collides with it, and pulls the RNA free. A helper protein called NusG can stabilize the connection between Rho and the polymerase, essentially acting as an anchor that helps the Rho motor do its job. An alternative model suggests that in some cases, Rho rides along already attached to the polymerase, scanning the RNA as it emerges and triggering termination when it detects the right signal sequence.
Eukaryotic termination is less well defined but involves signals in the RNA that direct cleavage and release of the transcript downstream of the gene.
Processing the RNA After Transcription
In eukaryotic cells, the initial RNA transcript (called pre-mRNA) isn’t ready to use right away. It goes through three major processing steps before it becomes a mature messenger RNA that can leave the nucleus and direct protein production.
First, a protective cap made of a modified guanosine molecule is added to the front (5′ end) of the transcript. This cap helps the cell’s protein-making machinery recognize the mRNA and also shields it from being broken down prematurely. Second, a long tail of adenine nucleotides (the poly-A tail) is added to the back end, which further stabilizes the molecule and helps control how long it lasts in the cell. Third, stretches of non-coding sequence called introns are cut out by a molecular machine called the spliceosome, and the remaining coding segments (exons) are joined together. This splicing step is what allows a single gene to produce multiple different proteins by including or excluding different exon combinations.
Bacteria skip all of this. Their mRNA is ready for translation the moment it’s made, and ribosomes often begin building proteins on the RNA while transcription is still in progress.
How Cells Control Which Genes Get Transcribed
Not every gene is transcribed all the time. Cells use regulatory DNA sequences called enhancers and silencers to dial transcription up or down. Enhancers can be located thousands of nucleotides away from the gene they regulate, either upstream or downstream, and their effect on gene expression depends on that distance in a non-linear way. At short distances, whether the enhancer sits before or after the gene can change the timing and pattern of transcription. Multiple enhancers can also work together, amplifying expression beyond what any single one could achieve alone, and they can sometimes reach promoters outside their typical range through looping of the DNA.
Silencers work by the opposite logic, recruiting proteins that condense the DNA and make it physically inaccessible to RNA polymerase. Together, this system of activating and repressing elements lets a single genome produce hundreds of distinct cell types, each with its own transcription profile, from the same underlying DNA sequence.
Transcribing DNA in the Lab
Researchers routinely transcribe DNA outside of living cells using a technique called in vitro transcription. The basic recipe requires four components: a purified DNA template containing a promoter sequence, the four ribonucleotide building blocks (ATP, CTP, GTP, and UTP), a buffer solution containing magnesium and potassium ions to support enzyme activity, and an RNA polymerase. Lab protocols typically use RNA polymerases from bacteriophages (viruses that infect bacteria) because they’re small, efficient, and recognize very specific promoter sequences, making the system easy to control. This approach is used to produce RNA for everything from research probes to mRNA vaccines.