Genes VII

Mutations in the regulatory circuit may either abolish expression of the operon or cause unregulated expression. Mutants that cannot be expressed at all are called uninducible. The continued expression of a gene that does not respond to regulation is called constitutive gene expression, and mutants with this property are called constitutive mutants.

Components of the regulatory circuit of the operon can be identified by mutations that affect the expression of all the structural genes and map outside them. They fall into two classes. The promoter and the operator are identified as targets for the regulatory proteins (RNA polymerase and repressor, respectively) by cis-acting mutations. And the locus lacI is identified as the gene that codes for the repressor protein by mutations that eliminate the trans-acting product.

The operator was originally identified by constitutive mutations, denoted O^c, whose distinctive properties provided the first evidence for an element that functions without being represented in a diffusible product.

The structural genes contiguous with an O^c mutation are expressed constitutively because the mutation changes the operator so that the repressor no longer binds to it. So the repressor cannot prevent RNA polymerase from initiating transcription. The operon is transcribed constitutively, as illustrated in Figure 10.8.

The operator can control only the lac genes that are adjacent to it. If a second lac operon is introduced into the bacterium on an independent molecule of DNA, it has its own operator. Neither operator is influenced by the other. So if one operon has a wild-type operator, it will be repressed under the usual conditions, while a second operon with an O^c mutation will be expressed in its characteristic fashion.

These properties define the operator as a typical cis-acting site, whose function depends upon recognition of its DNA sequence by some trans-acting factor. The operator controls the adjacent genes irrespective of the presence in the cell of other alleles of the site. A mutation in such a site, for example, the O^c mutation, is formally described as cis-dominant.

A mutation in a cis-acting site cannot be assigned to a complementation group. (The ability to complement is characteristic only of genes expressed as diffusible products.) When two cis-acting sites lie close together Vfor example, a promoter and an operator Vwe cannot classify the mutations by a complementation test. We are restricted to distinguishing them by their effects on the phenotype.

Cis-dominance is a characteristic of any site that is physically contiguous with the sequences it controls. If a control site functions as part of a polycistronic mRNA, mutations in it will display exactly the same pattern of cis-dominance as they would if functioning in DNA. The critical feature is that the control site cannot be physically separated from the genes that it regulates. From the genetic point of view, it does not matter whether the site and genes are together on DNA or on RNA.

Constitutive transcription is also caused by mutations of the lacI V type, which are caused by loss of function (including deletions of the gene). When the repressor is inactive or absent, transcription can initiate at the promoter. Figure 10.9 shows that the lacI V mutants express the structural genes all the time, irrespective of whether the inducer is present or absent, because the repressor is inactive.

The two types of constitutive mutations can be distinguished genetically. O^c mutants are cis-dominant, whereas lacI^- mutants are recessive. This means that the introduction of a normal, lacI⁺ gene restores control, irrespective of the presence of the defective lacI^- gene.

Mutants of the operon that are uninducible fall into the same two types of genetic classes as the constitutive mutants:

• Promoter mutations are cis-acting. If they prevent RNA polymerase from binding at P_lac, they render the operon nonfunctional because it cannot be transcribed.
• Mutations that abolish the ability of repressor to bind the inducer are described as lacI^s . They are trans-acting. The repressor is "locked in" to the active form that recognizes the operator and prevents transcription. The addition of inducer has no effect because its binding site is absent, and therefore it is impossible to convert the repressor to the inactive form. The mutant repressor binds to all lac operators in the cell to prevent their transcription, and cannot be pried off, irrespective of the properties of any wild-type repressor protein that is present, so it is genetically dominant.

The two types of mutations in lacI can be used to identify the individual active sites in the repressor protein. The DNA-binding site recognizes the sequence of the operator. It is identified by constitutive point mutations that prevent repressor from binding to DNA to block RNA polymerase. The inducer-binding site is identified by point mutations that cause uninducibility, because inducer cannot bind to trigger the allosteric change in the DNA-binding site.

An important feature of the repressor is that it is multimeric. Repressor subunits associate at random in the cell to form the active protein tetramer. When two different alleles of the lacI gene are present, the subunits made by each can associate to form a heterotetramer, whose properties differ from those of either homotetramer. This type of interaction between subunits is a characteristic feature of multimeric proteins and is described as interallelic complementation (see Complementation).

Negative complementation occurs between some repressor mutants, as seen in the combination of lacI ^Vd with lacI⁺ genes. The lacI ^Vd mutation alone results in the production of a repressor that cannot bind the operator, and is therefore constitutive like the lacI ^V alleles. Because the lacI^- type of mutation inactivates the repressor, it is recessive to the wild type. However, the -d notation indicates that this variant of the negative type is dominant when paired with a wild-type allele. Such mutations are said to be trans-dominant; they are also called dominant negatives (see Complementation).

The reason for the dominance is that the lacI^-d allele produces a "bad" subunit, which is not only itself unable to bind to operator DNA, but is also able as part of a tetramer to prevent any "good" subunits from binding. This demonstrates that the repressor tetramer as a whole, rather than the individual monomer, is needed to achieve repression. In fact, we may reverse the argument to say that, whenever a protein has a dominant negative form, this must mean it functions as part of a multimer. The production of dominant negative proteins has become an important technique in eukaryotic genetics.

The lacI^-d mutations lie in the DNA-binding site of the repressor subunit. This explains their ability to prevent mixed tetramers from binding to the operator; a reduction in the number of binding sites reduces the specific affinity for the operator. The map of the lacI gene shown in Figure 10.10 shows that the lacI^-d mutations are clustered at the extreme left end of the gene. This identifies the immediate N-terminal region of the protein as the DNA-binding site. Mutations of the recessive lacI^- type also occur elsewhere in the molecule, but could exert their effects on DNA binding indirectly.

The role of the N-terminal region in specifically binding DNA is shown also by its location as the site of occurrence of "tight binding" mutations. These increase the affinity of the repressor for the operator, sometimes so much that it cannot be released by inducer. They are rare.