Suppression

XI. SUPPRESSION

Suppressors are the class of second-site mutations that, in the presence of the original mutation, modify the original phenotype of the strain to be enough like that of wild-type to be scored as such. The virtue of suppressor analysis, especially non-informational ones (see below), is that they suggest alternate ways for a cell to solve a problem and therefore provide insight into the cell's choices - it's biochemical and physiological makeup. All suppressors run two general risks: (1) they often entail the loss of a normal gene product function and (2) they often involve the acquisition of new, possibly deleterious, functions. Sometimes revertants containing suppressor mutations are referred to as "pseudo- revertants", but this terminology does not reflect the traditional use of the term reversion (which is a restoration of an apparently wild-type phenotype). The following is an incomplete list of some of the sorts of events that might occur as suppressor mutations:

Informational: An informational suppressor is one that does not change the mutated gene's sequence, but rather affects the way the information (of the original mutant gene) is expressed as a product. Examples of this class of suppressors are altered tRNAs that now misread the mutant codon. The suppressor formed could suppress missense, nonsense, or frameshift mutations, depending on the nature of the tRNA alteration. Similarly, there are enzymes that modify and process tRNAs to active forms. An alteration in these modification enzymes can occasionally give rise to an altered tRNA that misreads and thus suppresses certain mutations. Alterations that affect the fidelity of ribosomes, the ram mutations, can suppress some forms of base substitution mutations. Very recently, a specific change in one of the genes encoding 16S rRNA has been shown to act as a suppressor of UGA signals. Whether or not other codon-specific suppressors map to ribosomal genes is presently unclear.
The above are all genotypic suppressors, because they are due to an alteration in the genotype. There is also a phenomenon termed phenotypic suppression, where an added chemical causes errors in translation (e.g. streptomycin) or transcription (e.g. 5-fl-U), so that "functional" products are occasionally synthesized with only the mutant gene present. The term "phenotypic" refers to the fact that you are not affecting the genotype of the organism (they are still mutant), but you are affecting their ability to make some functional product. Obviously these chemicals mess up the translation of everything in the cell to some extent, but the ability to make a bit of the limiting protein allows a bit of growth.
All the following classes of suppressors are non-informational:
Revertants in the same gene as the original mutation: A second alteration in the originally mutated codon, giving another codon distinct from both the mutant and the wild-type, is a form of suppressor mutation (at least under the definition I am using). Base changes elsewhere in the affected gene might cause an amino acid change that allows the gene product to function in a more wild-type fashion. If the phenotype of the original mutation was due only to its polarity onto downstream genes, then any mutation that eliminates that polarity would have the property of suppressing the original mutation. Such events might involve the restoration of proper reading frame, reinitiation of translation, reinitiation of transcription, or deletion of the polar block. This latter example is something of a special case since it causes the elimination of the original mutation. Finally, if the phenotype of the original mutation was mutant because it caused a toxic (and therefore trans-dominant) polypeptide to be synthesized, any mutation that eliminated the polypeptide would be a suppressor.

The following are possible classes of suppressors whose analysis would provide insight into the biochemistry or physiology of the organism.

Increase in the amount of a partially active protein: If the original mutation causes the synthesis of a protein with lower specific activity, this phenotype might be suppressed by (a) amplifying the gene that encodes the partially active protein (by some sort of duplication mechanism); (b) altering the expression and regulation of that gene product by either eliminating a repressor (or some other effect on the regulatory system); (c) generating new or more efficient promoters. The result of any of these would be that the product was more abundantly synthesized.

Elimination of "super-inhibition": There are cases where a mutation generates a protein that is particularly sensitive to an inhibitory agent. Elimination of this phenotype might come by either further altering the protein to eliminate that inhibition or altering the uptake or processing of the inhibitory agent so that it no longer reaches a level that causes a mutant phenotype.

Increasing the substrate level for a mutationally altered protein: If the original mutation causes the gene product to have a poor Km for its substrate, then any mutation that increases the level of that substrate, or decreases the competition for that substrate by altering for example the Km of the competing enzyme, would have the effect of suppressing the mutant phenotype.

Generation of a new protein: While at first glance this seems an impossible situation, there are a number of ways that a "new" gene can be activated to make the desired protein. If a "silent" copy of the mutated gene exists, this simply need be turned on. If an isozyme of the altered enzyme exists and its normal regulation prevents it from substituting for the altered enzyme, then that regulation itself could be altered to cause the substitute protein to be synthesized at the desired time. It is occasionally possible to alter an existing protein in such a precise way that it acquires a slightly different activity and can run a reaction it did not previously run.

New supply of the product: In some organisms, it has been demonstrated that there are "channeled pathways" whereby particular metabolites never exist free in the cell, but are maintained internally in an enzyme complex. It is conceivable that a particular mutant phenotype can be eliminated by destroying another channelled system, thus releasing its metabolites to the system which has been damaged. Damaging another metabolic pathway to cause the accumulation of a product not normally accumulated at high level could have the phenotype of suppressing the original mutation.

As a related aside, it has been argued that very few metabolites are actually floating arond free in the cell, but spend most of their time being passed from protein to protein. The argument is largely theoretical and is based on the observation that the vast majority of small molecules in the cell are the product of a single enzyme and the substrate of a single enzyme, so that it would make no sense to let the molecule "float around". Only those molecules that are substrates for more than one enzyme might profitably be accumulated, so that the various enzymes had access to them. To the extent this model is correct, there are a number of significant implications: (a) The idea of "concentration in the cell" has little meaning for a substrate that is not free. Channeled substrates could reach very high local concentrations (suitable for the Km of the next enzyme), while being very non-abundant in the cell. (b) This would allow a cell to maintain a much lower level of all sorts of stuff than presently envisaged, with savings in energy and "response time". (c) Those molecules that do float around might be expected to be utilized as signals of metabolism.

The take-home lessons with suppressors are: (1) When you start with a mutant that fails to grow, any alteration that allows it to grow, even a little, is acceptable and detectable. Revertants are therefore cells that are significantly better than the mutant but are not necessarily of a fully wild-type phenotype. They are strains that have the requisite amount of growth to be detectable under the arbitrarily chosen conditions. (2) Just as it is apparent that there are a very wide variety of mechanisms for "restoring" a wild-type phenotype, so too are there a wide variety of ways of causing a mutant phenotype. Again, do not assume that simply because a mutation causes a loss of enzyme activity that the mutation is in the gene encoding that enzyme. As argued above, the biochemical and physiological analysis of a suppressor-containing strain provides insight into how the cell really "thinks".

A final mention should be made of the subtle distinction between complementation and suppression. In the section on mapping by primes (Other Types of Mapping), the statement was made that one could "complement" a particular mutant phenotype by moving in a range of primes that cover the entire bacterial chromosome. The reason the term complementation was in quotes, was because such an analysis is not necessarily complementation. What you are in fact asking is, "is there any region of the chromosome, which when made diploid, restores a pseudo-wild-type phenotype". Typically, what you are doing is complementation. There are cases, however, where such an analysis might indicate two different regions of the chromosome that had the property of restoring a pseudo-wild-type phenotype when introduced into the mutant recipient. An analysis is properly considered to be complementation if it is known that the introduced region is the homologous region to that which is mutated in the recipient. This subtle effect can become a significant problem if one is doing "shot-gun cloning" from one organism into a heterologous organism, demanding Complementation of a mutant phenotype. In this case, arguably there can be no true complementation, and any form of Suppression would give the desired phenotype. Any clones presumed to carry the appropriate gene will need to be characterized in another way before proceeding with confidence.

[See sample problem 26,27, and 28]

[Previous] | [Next]


		This page was last built with Frontier and Web Warrior on a Macintosh on Thu, Sep 21, 2000 at 1:02:10 PM.