Transcription

Transcription

How RNA polymerase copies a gene into RNA, how bacteria start and stop it with one enzyme, and how our cells share the work among three.

Transcription in the nucleus
RNA polymerase II sits on the DNA double helix, holding a short stretch open while it moves along it. A strand of pre-mRNA leaves the polymerase and trails back towards the promoter, where TFIID stays bound to the TATA box. The pre-mRNA’s 5′ end carries a cap. Below, the nuclear envelope is drawn as two lipid bilayers.

Point at a name or a part to light it.

Pol II has left the promoter and moves to the right, reading the template strand. Simplified, not to scale.

At a glance

  1. Transcription copies a gene's DNA into RNA. RNA polymerase reads the template strand 3′ to 5′ and builds the RNA 5′ to 3′, so the RNA matches the coding strand, with U in place of T. It needs no primer, and it proofreads far less than DNA polymerase.
  2. Bacteria have one RNA polymerase. Its σ factor finds the promoter by two short sequences, about 10 and 35 base pairs before the start; a terminator in the RNA stops it.
  3. Our nuclei have three. RNA polymerase II makes pre-mRNAs, Pol I the large rRNAs and Pol III the tRNAs and 5S rRNA.
  4. Pol II needs general transcription factors to start: TFIID binds the TATA box, where there is one, and the others gather round it. Activators bound to enhancers, often far away, decide how often a gene is transcribed.
  5. Bacterial mRNA is translated while it is being made. Our pre-mRNA is capped, spliced and given a poly(A) tail before it leaves the nucleus.

Copying a gene into RNA

A gene's two DNA strands have different roles. RNA polymerase reads one, the template strand (or antisense strand), and the RNA it makes is complementary to it, so the RNA has the same sequence as the other, the coding strand (or sense strand), with uracil in place of thymine. Which strand is the template differs from gene to gene.

RNA polymerase opens a short stretch of the double helix, the transcription bubble, and adds ribonucleotides one at a time, pairing each with the template. Each is joined to the free 3′-OH at the end of the chain, so the RNA grows 5′ to 3′ while the template is read 3′ to 5′. Behind the polymerase the DNA closes again and the new RNA peels away. Unlike DNA polymerase, RNA polymerase can start a chain on its own, without a primer. It also makes more mistakes. It can back up and cut out a wrong nucleotide, but it checks its work far less thoroughly than DNA polymerase, and by the usual estimate it gets about one nucleotide in 10,000 wrong, against about one in 10 million for DNA polymerase; measured in living cells, its errors are rarer still. That matters less, because an RNA is a short-lived copy and is not passed on to daughter cells.

A transcription unit has three parts: the promoter, where the polymerase binds and starts; the transcribed region; and the terminator, where it stops. Positions are counted from the first transcribed nucleotide, +1; upstream positions, in the promoter, carry a minus sign.

in bacteria, for every kind of RNA
1RNA polymerase
in the nucleus of our cells
3RNA polymerases
copied wrongly by RNA polymerase, about
1 in 10,000nucleotides

Transcription in bacteria

E. coli has one RNA polymerase for all its RNAs. Its core enzyme, of five subunits (two α, β, β′ and ω), can make RNA but cannot find a promoter. A σ factor joins it to make the holoenzyme, and σ recognises the promoter by two short sequences: the −35 sequence, TTGACA, and the −10 sequence or Pribnow box, TATAAT. Real promoters differ from these consensus sequences, and the closer a promoter comes to them, the more often its gene is transcribed. Bacteria make other σ factors to switch on whole sets of genes, for heat shock or starvation, for example.

  1. 1

    Initiation

    The holoenzyme binds the promoter and opens about a turn of DNA round the start site. It makes a few short RNAs that it lets go of, then breaks free of the promoter; σ usually leaves.
  2. 2

    Elongation

    The core enzyme moves along the gene at about 40 nucleotides a second, unwinding the DNA ahead of it and rewinding it behind.
  3. 3

    Termination

    At a Rho-independent terminator the new RNA folds into a hairpin rich in G and C, followed by a run of U. The hairpin stalls the polymerase and the weak A–U pairs let the RNA go. At a Rho-dependent terminator, the Rho protein, a helicase that runs along the RNA, catches the stalled polymerase and pulls the RNA off.

Bacterial genes for one pathway often lie together in an operon, transcribed from one promoter into one mRNA that codes for several proteins (a polycistronic mRNA). With no nucleus in the way, ribosomes start translating the mRNA while it is still being made.

Transcription in our cells

The nucleus has three RNA polymerases, each for its own genes, and each needs its own set of transcription factors. Mitochondria have a fourth, a single-chain enzyme like those of some bacteriophages, for their own small genome.

MakesWhereBlocked by α-amanitin
RNA polymerase Ione long precursor of the 28S, 18S and 5.8S rRNAsthe nucleolusno
RNA polymerase IIpre-mRNAs, most snRNAs, microRNAsthe nucleoplasmyes, at low levels
RNA polymerase IIItRNAs, 5S rRNA, U6 snRNA and other small RNAsthe nucleoplasmonly at high levels

Starting Pol II

RNA polymerase II cannot find a promoter or open the DNA by itself. The general transcription factors do this, gathering on the core promoter into a preinitiation complex:

  1. 1

    TFIID binds the promoter

    Its TATA-binding protein binds the TATA box, about 25 to 30 base pairs upstream of the start site, where a promoter has one; many have none and are found by other sequences.
  2. 2

    The complex gathers

    TFIIB, then TFIIF with the polymerase, then TFIIE and TFIIH join, in the order seen in the test tube.
  3. 3

    TFIIH opens and releases

    TFIIH unwinds the DNA at the start site and phosphorylates the polymerase's long tail, its C-terminal domain, which lets it leave the promoter and start elongating.

This makes a basal level of transcription. How often a gene is actually transcribed is set by activators and repressors, proteins that bind short sequences near the promoter or in enhancers, which can lie thousands of base pairs away, upstream, downstream or within an intron. The DNA loops so that proteins bound to an enhancer reach the promoter, and a large complex, Mediator, passes their signal to the polymerase. The DNA must also be freed from its nucleosomes, which is how chromatin controls which genes can be read (Genome, chapter II).

Elongation and termination

As Pol II moves along the gene, the pre-mRNA is processed on the way: it gets its 5′ cap once it is about 25 nucleotides long, introns are spliced out while it is still being made, and its 3′ end is cut just past the poly(A) signal and given a poly(A) tail (RNA processing). Pol II runs on beyond the cut until an exonuclease that chews the leftover RNA catches up with it and it falls off the DNA. Pol I and Pol III stop at their own termination signals.

Check yourselfAnswer, then say how sure you were: a lucky guess still counts as a weak point.

1The coding strand of a gene reads 5′-ATGGCA-3′. What is the sequence of the RNA made from it?

2Which part of the bacterial RNA polymerase recognises the promoter?

3What ends transcription at a Rho-independent terminator in bacteria?

4Which human RNA polymerase makes transfer RNAs?

5Why can rifampicin treat tuberculosis without stopping the patient's own transcription?