The lac operon is an operon required for the transport and metabolism of lactose in Escherichia coli and some other enteric bacteria. It consists of three adjacent structural genes, a promoter, a terminator, and an operator. The lac operon is regulated by several factors including the availability of glucose and of lactose. Gene regulation of the lac operon was the first genetic regulatory mechanism to be elucidated and is often used as the canonical example of prokaryotic gene regulation.
Structure of the operon
- lacZ encodes β-galactosidase (LacZ), an intracellular enzyme that cleaves the disaccharide lactose into glucose and galactose.
- lacY encodes β-galactoside permease (LacY), a membrane bound transport protein that pumps lactose into the cell.
- lacA encodes β-galactoside transacetylase (LacA), an enzyme that transfers an acetyl group from acetyl-CoA to β-galactosides.
Only lacZ and lacY appear to be necessary for lactose catabolism.
Specific control of the lac genes depends on the availability of the substrate lactose to the bacterium. The proteins are not produced by the bacterium when lactose is unavailable as a carbon source. The lac genes are organized into an operon; that is, they are oriented in the same direction immediately adjacent on the chromosome and are co-transcribed into a single polycistronic mRNA molecule. Transcription of all genes starts with the binding of the enzyme RNA polymerase (RNAP), a DNA-binding protein, to a specific DNA binding site immediately upstream of the genes, the promoter. From this position RNAP proceeds to transcribe all three genes (lacZYA) into mRNA. The DNA sequence of the E. coli lac operon, the lacZYA mRNA, and the lacI genes are available from GenBank (view).
The regulatory response to lactose requires an intracellular regulatory protein called the lactose repressor. The lacI gene encoding repressor lies nearby the lac operon and is always expressed (constitutive). If lactose is missing from the growth medium, the repressor binds very tightly to a short DNA sequence just downstream of the promoter near the beginning of lacZ called the lac operator. Repressor bound to the operator interferes with binding of RNAP to the promoter, and therefore mRNA encoding LacZ and LacY is only made at very low levels. When cells are grown in the presence of lactose, a lactose metabolite called allolactose binds to the repressor, causing a change in its shape. Thus altered, the repressor is unable to bind to the operator, allowing RNAP to transcribe the lac genes and thereby leading to high levels of the encoded proteins.
The diagram below summarizes these statements.
Three-letter mnemonics are used to describe phenotypes in bacteria including E. coli.
- Lac (the ability to use lactose),
- His (the ability to synthesize the amino acid histidine)
- Mot (swimming motility)
- Str (response to the antibiotic streptomycin)
In the case of Lac, wild type cells are Lac+ and are able to use lactose as a carbon and energy source, while Lac- mutant derivatives cannot use lactose. The same three letters are typically used (lower-case, italicized) to label the genes involved in a particular phenotype, where each different gene is additionally distinguished by an extra letter. The lac genes encoding enzymes are lacZ, lacY, and lacA. The fourth lac gene is lacI, encoding the lactose repressor---I stands for inducibility.
One may distinguish between structural genes encoding enzymes, and regulatory genes encoding proteins that affect gene expression. Current usage expands the phenotypic nomenclature to apply to proteins: thus, LacZ is the protein product of the lacZ gene, β-galactosidase. Various short sequences that are not genes also affect gene expression, including the lac promoter, lac p, and the lac operator, lac o. Although it is not strictly standard usage, mutations affecting lac o are referred to as lac oc, for historical reasons.
A number of lactose derivatives or analogs have been described that are useful for work with the lac operon. These compounds are mainly substituted galactosides, where the glucose moiety of lactose is replaced by another chemical group.
- Isopropyl-β-D-thio-galactoside (IPTG) is frequently used as an inducer of the lac operon for physiological work. IPTG binds to repressor and inactivates it, but is not a substrate for β-galactosidase. One advantage of IPTG for in vivo studies is that since it cannot be metabolized by E. coli its concentration remains constant and the rate of expression of lac p/o-controlled genes, is not a variable in the experiment. In addition, IPTG is transported efficiently independent of whether the lacY gene is functional.
- Phenyl-β-D-galactose (phenyl-Gal) is a substrate for β-galactosidase, but does not inactivate repressor and so is not an inducer. Since wild type cells produce very little β-galactosidase, they cannot grow on phenyl-Gal as a carbon and energy source. Mutants lacking repressor are able to grow on phenyl-Gal. Thus, minimal medium containing only phenyl-Gal as a source of carbon and energy is selective for repressor mutants and operator mutants. If 108 cells of a wild type strain are plated on agar plates containing phenyl-Gal, the rare colonies which grow are mainly spontaneous mutants affecting the repressor. The relative distribution of repressor and operator mutants is affected by the target size. Since the lacI gene encoding repressor is about 50 times larger than the operator, repressor mutants predominate in the selection.
- Other compounds serve as colorful indicators of β-galactosidase activity.
- Allolactose is an isomer of lactose and is the inducer of the lac operon. Lactose is galactose-(β1->4)-glucose, whereas allolactose is galactose-(β1->6)-glucose. Lactose is converted to allolactose by β-galactosidase in an alternative reaction to the hydrolytic one. A physiological experiment which demonstrates the role of LacZ in production of the "true" inducer in E. coli cells is the observation that a null mutant of lacZ can still produce LacY permease when grown with IPTG but not when grown with lactose. The explanation is that processing of lactose to allolactose (catalyzed by β-galactosidase) is needed to produce the inducer inside the cell.
Classification of regulatory mutants
A conceptual breakthrough of Jacob and Monod was to recognize the distinction between regulatory substances and sites where they act to change gene expression. A former soldier, Jacob used the analogy of a bomber that would release its lethal cargo upon receipt of a special radio transmission or signal. A working system requires both a ground transmitter and a receiver in the airplane. Now, suppose that the usual transmitter is broken. This system can be made to work by introduction of a second, functional transmitter. In contrast, he said, consider a bomber with a defective receiver. The behavior of this bomber cannot be changed by introduction of a second, functional aeroplane.
To analyze regulatory mutants of the lac operon, Jacob developed a system by which a second copy of the lac genes (lacI with its promoter, and lacZYA with promoter and operator) could be introduced into a single cell. A culture of such bacteria, which are diploid for the lac genes but otherwise normal, is then tested for the regulatory phenotype. In particular, it is determined whether LacZ and LacY are made even in the absence of IPTG. This experiment, in which genes or gene clusters are tested pairwise, is called a complementation test.
This test is illustrated in the figure (lacA is omitted for simplicity). First, certain haploid states are shown (i.e. the cell carries only a single copy of the lac genes). Panel (a) shows repression, (b) shows induction by IPTG, and (c) and (d) show the effect of a mutation to the lacI gene or to the operator, respectively. In panel (e) the complementation test for repressor is shown. If one copy of the lac genes carries a mutation in lacI, but the second copy is wild type for lacI, the resulting phenotype is normal---no LacZ is expressed without IPTG. Mutations affecting repressor are said to be recessive to wild type (and that wild type is dominant), and this is explained by the fact that repressor is a small protein which can diffuse in the cell. The copy of the lac operon adjacent to the defective lacI gene is effectively shut off by protein produced from the second copy of lacI.
If the same experiment is carried out using an operator mutation, a different result is obtained (panel (f)). The phenotype of a cell carrying one mutant and one wild type operator site is that LacZ and LacY are produced even in the absence of the inducer IPTG. The operator mutation is dominant. When the operator site where repressor must bind is damaged by mutation, the presence of a second functional site in the same cell makes no difference to expression of genes controlled by the mutant site.
A more sophisticated version of this experiment uses marked operons to distinguish between the two copies of the lac genes and show that the unregulated gene(s) are the ones next to the mutant operator (panel (g). For example, suppose that one copy is marked by a mutation inactivating lacZ so that it can only produce the LacY protein, while the second copy carries a mutation affecting lacY and can only produce LacZ. In this version, only the copy of the lac operon that is adjacent to the mutant operator is expressed without IPTG. We say that the operator mutation is cis-dominant, it is dominant to wild type but affects only the copy of the operon which is immediately adjacent to it.
This explanation is misleading in an important sense, because it proceeds from a description of the experiment and then explains the results in terms of a model. But in fact, it is often true that the model comes first, and an experiment is fashioned specifically to test the model. Jacob and Monod first imagined that there must be a site in DNA with the properties of the operator, and then designed their complementation tests to show this.
The dominance of operator mutants also suggests a procedure to select them specifically. If regulatory mutants are selected from a culture of wild type using phenyl-Gal, as described above, operator mutations are rare compared to repressor mutants because the target-size is so small. But if instead we start with a strain which carries two copies of the lac genes (that is diploid for lac), the repressor mutations (which still occur) are not recovered because complementation by the wild type genes confers a wild type phenotype. While mutation of a single repressor gene produces no change in phenotype, mutation of a single operator can be determined by a reduction in color intensity.
Regulation by cyclic AMP
The experimental microorganism used by François Jacob and Jacques Monod was the common laboratory bacterium, E. coli, but many of the basic regulatory concepts that were discovered by Jacob and Monod are fundamental to cellular regulation in organisms. The key idea is that proteins are not synthesized when they are not needed--- E. coli conserves cellular resources and energy by not making the three Lac proteins when there is no need to metabolize lactose, such as when other sugars like glucose are available. The following section discusses how E. coli controls certain genes in response to metabolic needs.
During World War II, Monod was testing the effects of combinations of sugars as nutrient sources for E. coli. He found that bacteria grown with two different sugars often displayed two phases of growth. For example, if glucose and lactose were both provided, glucose would be metabolized first (growth phase I, see Figure 2) and then lactose (growth phase II). This phenomenon is called diauxie.
Metabolism of lactose does not occur during the first part of the diauxic growth curve because β-galactosidase is not made when both glucose and lactose are present in the medium.
Explanation of this depended on the characterization of additional mutations affecting the lac genes other than those explained by the classical model. Two other genes, cya and crp, were known that map far away from lac, and when mutant result in a decreased level of expression in the presence of IPTG and even in a strain which is mutant for repressor or operator. The discovery of cyclic AMP in 1957 (in eukaryotic cells) and a decade later in E. coli led to the demonstration that mutants defective in one of these genes could be restored to full activity by the addition of cyclic AMP to the medium.
The cya gene encodes adenylate cyclase, which produces cyclic AMP. In a cya mutant, the absence of cyclic AMP makes the expression of the lacZYA genes more than ten times lower than normal. Addition of cyclic AMP corrects the low Lac expression characteristic of cya mutants. The second gene, crp, encodes a protein called catabolite activator protein (CAP) or cAMP receptor protein (CRP). [It is remarkable that more than almost forty years later, different geneticists use different terms for the same gene depending on how they feel about the two competing groups involved in the original discovery.]
This dual regulation causes the lactose metabolism enzymes to be made in small quantities in the presence of both glucose and lactose (sometimes called leaky expression) due to lactose inhibiting LacI from binding to the operator, but at high cAMP concentrations and in the presence of lactose there are high levels of expression (Phase II in Figure 2). Leaky expression is necessary in order to allow for metabolism of some lactose after the glucose source is expended, but before lac expression is fully activated.
- When lactose is absent then there is very little Lac enzyme production (the operator has LacI bound to it).
- When lactose is present but a preferred carbon source (like glucose) is also present then a small amount of enzyme is produced (LacI is not bound to the operator).
- When lactose is the favoured carbon source (for example in the absence of glucose) cAMP-CAP bind to the promoter and Lac enzyme production is maximised.
The delay between growth phases reflects the time needed to produce sufficient quantities of lactose-metabolizing enzymes. First, the CAP regulatory protein has to assemble on the lac promotor, resulting in an increase in the production of lac mRNA. More available copies of the lac mRNA results in the production (see translation) of significantly more copies of LacZ (β-galactosidase, for lactose metabolism) and LacY (lactose permease to transport lactose into the cell). After a delay needed to increase the level of the lactose metabolizing enzymes, the bacteria enter into a new rapid phase of cell growth.
Two puzzles of catabolite repression relate to how cAMP level is actually coupled to the presence of glucose, and secondly, why the cells should even bother. After lactose is cleaved it actually forms glucose and galactose (easily converted to glucose). In metabolic terms, lactose is just as good a carbon and energy source as glucose. The cAMP level is related not to intracellular glucose concentration but to the rate of glucose transport, which influences the activity of adenylate cyclase. (In addition, glucose transport also leads to direct inhibition of the lactose permease.) As to why E. coli works this way, one can only speculate. All enteric bacteria ferment glucose, which suggests they encounter it frequently. It is possible that a small difference in efficiency of transport or metabolism of glucose v. lactose makes it advantageous for cells to regulate the lac operon in this way.
Multimeric nature of repressor and the complex operator
The lac repressor is a tetramer of identical subunits. Each subunit contains a helix-turn-helix (HTH) motif capable of binding to DNA. The operator site where repressor binds is a DNA sequence with inverted repeat symmetry. The two DNA half-sites of the operator together bind to two of the subunits of the tetrameric repressor. Although the other two subunits of repressor are not doing anything in this model, this property was not understood for many years.
Eventually it was discovered that two additional (minor) operators are involved in lac regulation. One (O3) lies in the end of the lacI gene and the other (O2) is about 400 bp downstream in the early part of lacZ. These two sites were not found in the early work because they have redundant functions and individual mutations do not affect repression very much. Single mutations to either O2 or O3 have only 2 to 3-fold effects. However, their importance is demonstrated by the fact that a double mutant defective in both O2 and O3 is dramatically de-repressed (by about 70-fold).
In the current model, repressor is bound simultaneously to both the main operator O1 and to either O2 or O3. The intervening DNA loops out from the complex. The redundant nature of the two minor operators suggests that it is not a specific looped complex that is important. One idea is that the system works through tethering. If bound repressor releases from O1 momentarily, binding to a minor operator keeps it in the vicinity, so that it may rebind quickly. This would increase the affinity of repressor for O1.
Mechanism of induction
The repressor is an allosteric protein, i.e. it can assume either one of two slightly different shapes, which are in equilibrium with each other. In one form the repressor is capable of binding to the operator DNA, and in the other form it cannot bind to the operator. According to the classical model of induction, binding of the inducer, either allolactose or IPTG, to the repressor affects the distribution of repressor between the two shapes. Thus, repressor with inducer bound is stabilized in the non-DNA-binding conformation. However, this simple model cannot be the whole story, because repressor is bound quite stably to DNA, yet it is released rapidly by addition of inducer. Therefore it seems clear that repressor can also bind inducer while still bound to DNA. It is still not entirely known what the exact mechanism of binding is.
Use in molecular biology
The lac gene and its derivatives are amenable to use as a reporter gene in a number of bacterial-based selection techniques such as two hybrid analysis, in which the successful binding of a transcriptional activator to a specific promoter sequence must be determined. In LB plates containing X-gal, the colour change from white colonies to a shade of blue, corresponds to about 20-100 β-galactosidase units, while tetrazolium lactose and MacConkey lactose media have a range of 100-1000 units, being most sensitive in the high and low parts of this range respectively.  Since MacConkey lactose and tetrazolium lactose media both rely on the products of lactose breakdown, they require the presence of both lacZ and lacY genes. The many lac fusion techniques which include only the lacZ gene are thus suited to the X-gal plates  or ONPG liquid broths 
- ONPG (β-Galactosidase)Test (May 2005). Retrieved on 2007-02-20.
- Joung J, Ramm E, Pabo C (2000). "A bacterial two-hybrid selection system for studying protein-DNA and protein-protein interactions". Proc Natl Acad Sci U S A 97 (13): 7382-7. PMID 10852947.
- Jacob F; Monod J (June 1961). "Genetic regulatory mechanisms in the synthesis of proteins". J Mol Biol. 3: 318-56. PMID 13718526.
Transcription (Prokaryotic, Eukaryotic)