Gene Expression and Regulation

Making a protein from a gene takes two steps. In transcription the gene's DNA is copied into messenger RNA, mRNA. In translation the mRNA is read to build a chain of amino acids. Together the two steps are called gene expression, and the flow of information from DNA to RNA to protein is sometimes called the central dogma (see MedlinePlus).

DNA → RNA → protein

Francis Crick set out the idea in 1958. His version also said that information never flows from proteins back to nucleic acids. Scientists have since found exceptions to the simple picture; prions, for example, are infectious proteins that copy themselves without any DNA or RNA in between (see NHGRI).

Another exception runs the first arrow backwards. In 1970 Howard Temin and David Baltimore each found an enzyme in tumour viruses that makes DNA from an RNA template (see Temin and Mizutani; Baltimore). This is reverse transcription, and Temin, Baltimore and Renato Dulbecco shared the 1975 Nobel Prize for their work on tumour viruses and the cell's genetic material (see Nobel Prize).

A gene starts at a promoter, a stretch of DNA just upstream of the gene where RNA polymerase and helper proteins called transcription factors bind to start transcription (see NHGRI). RNA polymerase then moves along the gene and builds an RNA copy.

Two strands, one message. RNA polymerase reads the template strand and builds RNA that pairs with it. So the mRNA has the same sequence as the other DNA strand, the coding strand, except that RNA uses U where DNA has T. That is why gene sequences are usually written as the coding strand; the code tables list the DNA triplets that pair with each RNA codon (see Britannica).

Processing. In eukaryotes the first copy is pre-mRNA. It is cut and rejoined to make mature mRNA, and the cell adds a cap to the 5′ end and a tail of adenine bases, the poly(A) tail, to the 3′ end. The cap and tail help keep the mRNA stable (see Britannica). The cutting removes introns, and the exons left behind are joined into the finished message (see NHGRI; NHGRI). Then the mRNA leaves the nucleus for the cytoplasm (see MedlinePlus).

A ribosome, made of RNA and protein, is where proteins are built (see NHGRI). It reads the mRNA three bases at a time. Each group of three is a codon, and a codon usually stands for one amino acid (see MedlinePlus).

Transfer RNA is the adaptor between the two languages (see NHGRI). Each tRNA carries an amino acid at one end and an anticodon at the other, three bases that pair with a matching codon (see NHGRI). The ribosome joins the amino acids one by one until it meets a stop codon (see MedlinePlus).

Reading frames. Because codons have no commas between them, where you start decides how the bases are grouped. AUGGCC is read AUG GCC from the first base, but UGG CC… from the second. The start codon, AUG, sets the frame (see Britannica).

Four bases, three at a time, give 4 × 4 × 4 = 64 codons. Sixty-one stand for amino acids and three, UAA, UAG and UGA, mean stop (see NHGRI; Britannica). AUG codes for methionine and also marks the start (see Britannica).

With 61 codons and only 20 amino acids, most amino acids have several codons. Methionine and tryptophan have one each; the other 18 have two to six. This redundancy is why the code is called degenerate. The code is almost universal, with rare variations, for example in mitochondria (see Britannica).

DNA and Genetics showed that a mutation is a change in the DNA sequence. What it does to the protein depends on what kind of change it is.

• Silent. Because the code is redundant, a base change can turn one codon into another for the same amino acid (see Britannica). GAG and GAA both mean glutamic acid, so the protein is unchanged.
• Missense. One amino acid is swapped for another (see MedlinePlus). Some missense changes are harmless, depending on where they fall (see NHGRI).
• Nonsense. The change makes an early stop codon, so the protein comes out short and may not work (see MedlinePlus).
• Frameshift. Adding or losing bases in a number that is not a multiple of three shifts the reading frame (see NHGRI). Every amino acid after it changes, and the protein usually does not work (see MedlinePlus).

Sickle cell. In sickle cell hemoglobin, HbS, valine replaces glutamic acid at position 6 of the beta-globin protein, written Glu6Val. That single missense change makes red blood cells bend into a sickle shape (see MedlinePlus). Population Genetics shows why the allele stays common where malaria spreads.

Bacteria often group genes for one job together under one promoter, so they switch on and off as a unit. This is an operon. Between the promoter and the genes sits the operator. When a repressor protein sits on the operator, RNA polymerase cannot transcribe the genes. A molecule called an inducer can bind the repressor and pull it off (see Britannica).

François Jacob and Jacques Monod worked this out for the lac operon of E. coli in 1961 (see Jacob and Monod; StatPearls). Its genes make the enzymes for using lactose, a milk sugar, and they are held off when there is no lactose (see Britannica). The operon is inducible: lactose switches it on, so the enzymes are made only when there is lactose to use (see StatPearls).

And glucose? Given both sugars, E. coli uses glucose first (see Inada and others). The standard model adds a second switch, a protein called CAP, also known as CRP, the cAMP receptor protein. The small signal molecule cAMP builds up when glucose is absent and binds CAP, and the cAMP-CAP pair helps RNA polymerase bind the promoter (see Ro and others). So with lactose present, low glucose also raises cAMP, which binds CAP and helps RNA polymerase bind, so transcription is high. With glucose around, cAMP stays low and transcription stays low. Glucose also acts another way: it keeps much of the lactose out of the cell, which is called inducer exclusion. Inada, Kimata and Aiba found that in E. coli given both sugars, this effect explained the glucose effect better than a fall in cAMP (see Inada and others). Both mechanisms are part of the picture.

Each of your cells expresses only a fraction of its genes. Control is most common at transcription (see MedlinePlus).

• Transcription factors. Signals from the environment or other cells switch on proteins called transcription factors, which bind a gene's regulatory regions and raise or lower its transcription (see MedlinePlus).
• Enhancers. Promoters work with other DNA regions, enhancers, and the DNA bends to bring an enhancer close to the promoter to start transcription (see NHGRI).
• Epigenetics. Chemical tags change gene activity without changing the DNA sequence. A methyl group attached to DNA usually turns a gene off. Tags on histones, the proteins DNA wraps around, change how tightly it is packed, and some tags are kept when a cell divides (see MedlinePlus).
• Alternative splicing. Exons from one gene can be joined in different combinations, giving different but related mRNAs (see NHGRI), and so several proteins from one gene.

Differentiation. Genes are turned on and off in different patterns during development, which makes a brain cell look and act different from a liver cell or a muscle cell (see MedlinePlus). Epigenetic tags help keep those choices; proteins that build bone, for example, are not made in muscle cells (see MedlinePlus).

The path from gene to protein was worked out in about twenty years of experiments, and the surprises kept coming after that.