1Regulatory sequences, constitutive and inducible genes

A regulatory sequence is a stretch of DNA that does not code for a protein but controls when a nearby gene is transcribed, by being a place where regulatory proteins bind. The promoter is where RNA polymerase binds; an operator (in bacteria) is a switch inside or beside the promoter where a repressor can bind.

• A constitutive gene is expressed all the time, at a fairly steady level, because the cell always needs its product. Genes for ribosomal proteins or for the enzymes of glycolysis are examples; they are sometimes called housekeeping genes.
• An inducible gene is normally off and is switched on (induced) when a particular signal is present, such as a nutrient or a hormone.

The point of regulation is economy: making enzymes the cell does not need wastes amino acids and ATP.

2Operons: bacteria switch whole groups of genes

In a bacterium, genes whose products work together are often next to each other, transcribed from one promoter into one mRNA. This unit is an operon:

• a promoter, where RNA polymerase binds;
• an operator, a short sequence overlapping or just after the promoter, where a repressor protein can bind;
• the structural genes, which code for the enzymes.

The repressor is made by a separate regulatory gene, often elsewhere on the chromosome. When the repressor sits on the operator, it physically blocks RNA polymerase, so none of the genes is transcribed. One switch controls all of them: that is coordinate regulation. The position matters. The operator works only because it lies between the promoter and the genes, in RNA polymerase's path.

An inducible operon: lac. E. coli can use the sugar lactose, but needs enzymes to take it in and break it down. Figure 1 shows the two states.

• No lactose: the repressor is active and binds the operator. No transcription. No wasted enzymes.
• Lactose present: a form of lactose binds the repressor and changes its shape, so it can no longer bind the operator. RNA polymerase transcribes the genes. Lactose is the inducer.

Positive control of lac. Removing the repressor is not the whole story. E. coli prefers glucose. When glucose is scarce, a signalling molecule (cAMP) builds up and binds an activator protein, and the activator binds DNA next to the promoter and helps RNA polymerase bind. So the lac genes are transcribed strongly only when lactose is present and glucose is absent. This is positive regulation: a protein that turns transcription up. The repressor is negative regulation: a protein that turns it off.

A repressible operon: trp. The trp operon codes for the enzymes that make the amino acid tryptophan. Here the logic runs the other way. The repressor, as made, is inactive and cannot bind the operator, so the genes are normally on. When tryptophan is plentiful, tryptophan binds the repressor and activates it; the active repressor binds the operator and switches the genes off. The cell stops making what it already has.

A repressor on the operator blocks RNA polymerase: in an inducible operon the signal removes the repressor, in a repressible operon the signal puts it there.

3Coordinate regulation in eukaryotes

Eukaryotes rarely have operons; each gene has its own promoter. Yet genes still work in groups. Genes in different places can carry the same regulatory sequence, so when one transcription factor becomes active it binds all of them and switches them on together. A stress, such as heat, can activate one transcription factor that turns on dozens of protective genes at once. Topic 6.6 looks at how transcription factors bind.

4Epigenetic changes

Epigenetic changes alter gene expression without changing the base sequence, by adding or removing chemical tags. They are reversible. Figure 2 shows the two kinds.

• DNA methylation. Methyl groups are added to cytosine bases, often in a promoter. The bases still pair normally, but transcription factors and RNA polymerase bind poorly, so a heavily methylated gene is usually silenced.
• Histone modification. Chemical groups added to histone tails change how tightly the DNA is wound. Adding acetyl groups generally loosens the packing and makes genes easier to transcribe; other tags can pack DNA more tightly and switch genes off.

Because the tags can be added and removed, epigenetic changes can respond to the environment, such as diet, stress or temperature. Some are copied when a cell divides, so a whole tissue keeps its pattern. They are not mutations: the sequence is the same, and sequencing the gene would show no difference.

5From expression to phenotype

The phenotype of a cell, and of an organism, is decided by which genes are expressed and at what levels.

• Differentiation. Every cell in your body has the same genome, but a red blood cell precursor makes haemoglobin and a pancreatic cell makes digestive enzymes. Each specialised cell expresses its own set of tissue-specific proteins. That difference in expression is the observable difference between cells.
• Sequential expression in development. In an embryo, a signal switches on a transcription factor; that factor switches on further genes, some of which are themselves transcription factors, which switch on the next set. Genes are turned on in a fixed order, so tissues form in the right sequence.
• Function and amount. A phenotype depends both on what a gene product does and on how much of it there is. A cell making a small amount of a pigment enzyme is paler than one making a lot, even if the enzyme is identical.

6Reading operon data

Worked example (invented data). E. coli strains are grown with and without lactose (no glucose), and the activity of the lactose-digesting enzyme is measured in arbitrary units.

• Wild type is inducible: 2 units without lactose rises to 480 with it, a 240-fold increase.
• Mutant 1 makes the enzyme all the time. Claim: the repressor cannot bind the operator. Evidence: activity is high (470) even without lactose. Reasoning: with nothing blocking RNA polymerase, transcription never stops. Either the repressor gene or the operator is altered; these data alone cannot tell which.
• Mutant 2 never makes the enzyme. A repressor that can no longer bind lactose would stay on the operator permanently; so would a broken promoter, or a broken enzyme gene. Again, the data narrow it down but do not settle it, and saying so earns credit.