Why copy at all? The case for a working note
The previous guide left us with a tension: the genome holds everything, yet a book on a shelf cooks no meal. The cell's answer is the first step of the central dogma, DNA to RNA: rather than dragging its master archive to the workshop, the cell makes a short, expendable copy of just the one gene it needs right now. That copying is transcription, and the act is exactly what the word says — you transcribe a passage by writing it out again in a fresh medium, not by tearing the page out of the original book.
Two things make this clever rather than wasteful. First, the original DNA never leaves the safety of the nucleus or the cell's interior — the archive stays pristine while disposable copies do the rough work out where the machinery lives. Second, the cell can make *many* copies of a busy gene and just one of a rarely-needed one, so the copy step itself becomes a volume knob on how much of each product gets built. A single DNA blueprint can spin off thousands of identical notes, and each note can be thrown away when the job is done.
The machine and where it starts: RNA polymerase meets the promoter
The machine that does the writing is RNA polymerase — a protein complex whose whole job is to crawl along DNA and string together an RNA copy. (It is a cousin of the DNA polymerase you met in the replication rung; both read a DNA template and build a new strand, but this one writes in RNA.) The deep problem it faces is one of *finding the start*. Your DNA is billions of letters long, with a gene buried somewhere in it. How does the machine know where one gene begins, out of all that text — and which of the thousands of genes to copy?
The answer is a landmark written into the DNA just upstream of each gene: the promoter. A promoter is not part of the message; it is a short signature sequence that says "a gene starts here, read this way." Think of it as the title and start-arrow at the top of a recipe. RNA polymerase (often guided by helper proteins) recognizes a promoter, docks onto it, and that binding both *positions* the machine at the right spot and *points* it in the correct direction. No promoter, no transcription — which is exactly why promoters become a master switch when the cell decides which genes to turn on.
Which strand is copied — and the bubble that travels
Here is a point that trips up nearly everyone, so go slowly. DNA has two strands, running antiparallel and held together by base pairing. RNA polymerase reads *only one of them* as its pattern — the template strand (also called the antisense strand). It reads that template and lays down RNA letters that pair with it. The *other* strand, the one it ignores, is the coding strand (or sense strand) — and here is the lovely part: because of base pairing, the RNA copy ends up reading almost exactly like that coding strand, just with U swapped in for T. So the strand the machine never touches is the one whose sequence the message resembles.
coding strand (sense) 5'- A T G C G A T T C -3' <- RNA reads like this (U for T)
| | | | | | | | |
template strand (antisense) 3'- T A C G C T A A G -5' <- the machine actually reads THIS
|
v RNA polymerase pairs A->U, C->G, G->C, T->A
RNA transcript 5'- A U G C G A U U C ... (matches coding strand, T -> U)To read a template, the machine must pry the two strands apart — DNA's letters are tucked *inside* the double helix, unreadable while paired. So RNA polymerase locally unzips about a dozen base pairs, opening a small transcription bubble: a short stretch where the two DNA strands are separated and the template lies exposed. The beautiful part is that this bubble travels with the machine. As the polymerase moves forward it unwinds fresh DNA at its leading edge and lets the strands snap back together behind it, so only a tiny window is ever open at once. The helix is opened just enough, just where it is needed, and immediately re-zipped — the archive is never left dangerously unspooled.
Initiation, elongation, termination
Like most molecular processes, transcription unfolds in three acts: it must start, run, and stop. These are not bureaucratic labels — each act solves a real problem. Starting is about *finding the right place and committing*; running is about *copying fast and accurately*; stopping is about *knowing where the gene ends and letting go*. Walk through them once and the whole process clicks together.
- Initiation. RNA polymerase, with its helper proteins, recognizes and clamps onto the promoter, then opens the first stretch of the transcription bubble. It does not need a primer — unlike DNA polymerase, it can start an RNA chain from scratch. Once it lays down the first few RNA letters, it commits and breaks free of the promoter, leaving the start signal behind.
- Elongation. The machine now slides steadily along the template, the bubble traveling with it. At each step it reads the next template base and adds the matching RNA letter (A pairs with U, G pairs with C, and so on), growing the chain in one direction at roughly a few dozen letters per second. The freshly made RNA peels off behind it as the DNA re-zips.
- Termination. When the machine reaches a stop signal written in the DNA, it releases the finished RNA and falls off the template, and the bubble closes for good. The DNA is left exactly as it was — untouched and ready to be copied again — and a brand-new RNA strand floats free.
The strand that floats free at the end is the primary transcript — the raw, first-draft RNA, copied straight off the gene. The collective name for these three acts, initiation, elongation, and termination, is worth holding onto, because the *very same trio* will reappear when you reach translation in the next rung. Be honest about one simplification, though: in eukaryotes this is a guided, regulated affair involving many helper proteins, and the resulting first draft almost always needs editing before it is ready — which is exactly what the next guide is about.
Why RNA, not more DNA? The working copy is built different
It would seem simpler to copy DNA into more DNA. But the cell deliberately copies into RNA, and the differences between the two are exactly what make RNA a good *disposable working copy* rather than a second archive. Three differences matter. RNA uses the base uracil (U) where DNA uses thymine (T) — chemically nearly the same, but a tell that marks RNA as RNA. RNA's sugar is ribose, carrying one extra oxygen that DNA's deoxyribose lacks; that extra oxygen makes RNA more chemically reactive, and therefore less stable. And RNA is usually single-stranded, where DNA is the famous double helix.
Read those traits as features, not flaws. DNA is built to *last* — two strands so its information is backed up and protected, a stable sugar so it survives for a lifetime. RNA is built to *be used and discarded* — single-stranded so it is nimble and can be read and folded freely, less stable so it does not linger and clutter the cell once its message has been delivered. A short-lived, easily-degraded copy is exactly what you want for an on-demand working note: make it, use it, let it break down. The fragility is the point.
Putting it together: a first draft, not the final word
Stand back and replay the whole scene. RNA polymerase finds a promoter and locks on (initiation); it pries open a traveling transcription bubble and reads the template strand, laying down RNA letters that come out matching the coding strand with U for T (elongation); it hits a stop signal, releases its product, and lets the DNA re-zip behind it (termination). What floats away is a single-stranded RNA primary transcript — a fresh, expendable copy of one gene, made without ever damaging the original.
But notice the word *primary*, or *first draft*. In a eukaryote, what the polymerase produces is not yet a clean, ready-to-use message. It still carries stretches that must be cut out, and it needs protective caps and tails added before it can safely leave the nucleus and be read. Calling it the genome's first draft is honest in both directions: it is a real copy of the gene, and it is unfinished. The editing that turns this raw transcript into a mature, usable messenger RNA is the subject of the next guide — that is where the surprising scissors-and-tape work of RNA processing begins.