Beneficial Mutations
From SkepticWiki
[edit] Introduction
One common piece of creationist propaganda is to claim that beneficial mutations are impossible. Obviously if this were true than adaptive evolution would also be impossible, since then natural selection would have nothing to work with and would function purely as a conservative force. This creationist claim is, of course, not true, as we shall demonstrate below by listing a representative collection of mutations that are favored by natural selection.
We should note that what is beneficial depends on the environmental context. If bacteria are exposed to streptomycin, then a mutation allowing them to thrive in such an environment is undoubtedly beneficial; in the absence of streptomycin, the same mutation would be of no benefit and might well be harmful.
Therefore the theory of evolution predicts that we should expect beneficial mutations to be rare, because natural selection has already had billions of years to operate. We are unlikely to see a beneficial mutation occur in a lineage that has had plenty of time to adapt to its environment, because in such cases evolution will have already happened, the beneficial mutations will already have taken place and been favored by natural selection, and any change is very likely to be neutral or for the worse. So it is no coincidence that the examples we give below are responses to changes in the environment, or are observed in organisms that have been tinkered with by scientists to make them inferior to the wild type.
[edit] Example: A simple experiment
A simple experiment suitable for the classroom demonstrating the existence of benefical mutations is detailed here. In brief, students take a population of Bacillus thuringiensis that is not immune to the antibiotic streptomycin. They establish this by a pair of control experiments: one to check that the population of B. thuringiensis can't grow on agar jelly impregnated with streptomycin (the negative control) and the other to establish that they can grow on normal agar jelly.
They then prepare agar jelly with a streptomycin gradient, so that at one end there isn't any streptomycin, and the other end is as rich in streptomycin as the negative control. Left long enough, a mutation will arise that causes streptomycin immunity, allowing the lucky mutant and its offspring to survive in the streptomycin-rich area of the agar.
Two such mutations are known in detail (see Snyder L, Champness W (1997) Mutations of bacteria, Molecular Genetics of Bacteria, pp. 77–80, American Society of Microbiology). One changes the gene for ribosomal protein S12, so that streptomycin can't inhibit the function of the ribosome (which is how streptomycin kills bacteria); another changes a gene coding for a protein in the cell membrane so that the streptomycin isn't allowed into the cell.
[edit] Example: Vancomycin-dependent enterococcus
Certain strains of Enterococcus take adaptation to antibiotics even further. Instead of merely evolving to be resistant to vancomycin[1], they go one step better and feed off it. Indeed, they become dependent on it as an essential component of their diet.[2]
The mutations involved were, therefore, beneficial only after vancomycin came into use; before then, such traits would be lethal. We can therefore definitively date the origin of these strains to some point after the first use of vancomycin against Enterococcus, that is, some time after 1968.
Moreover, when vancomycin is withdrawn from vancomycin-dependent enterococcus, in an attempt to starve it out, then revertant mutations to a non-dependent form become beneficial, and have been observed in laboratory tests:
- Vancomycin-dependent enterococci illustrate the power of biologic adaptation. Not only do these bacteria acquire the ability to actually utilize vancomycin for cell wall synthesis, but they also have the ability to revert to their resistant phenotypes when the drug is withdrawn.[3]
The mutations involved have been studied at the DNA level in the vancomycin-dependent strains:
- Each strain had either a 5-bp insertion at codon 41 resulting in an early stop codon, an in-frame 6-bp deletion resulting in the loss of two residues (KDVA243-246->KA), or single amino acid substitutions (E13->G, G99->R, V241->D, D295->G, P313->L)[4]
[edit] Example: Other forms of antibiotic resistance
In the previous examples, we gave details of an experiment in which the control experiments proved that the population did not, initially, possess immunity to streptomycin, and the example of vancomycin-dependent bacteria which could not have existed before the use of vancomycin as an antibiotic.
There are of course many more examples of bacteria acquiring immunity to antibiotics outside the laboratory, producing the so-called "superbugs", but presented with this fact, creationists will sometimes offer the alternative hypothesis that the various genes for resistance were all created by God "in the beginning". Obviously, the experiments and observations discussed above show that this hypothesis is not necessary; we shall also prove that it is not possible.
We shall not debate the theological point. An omniscient and spiteful Creator could indeed have foreseen every antibiotic that humans would ever come up with, and put genes for immunity to every antibiotic into the original genome of each species of bacteria.
The problem is that in the 6000 years (taking the minimum creationist estimate) between creation and the invention of antibiotics, natural selection would be neutral with respect to the antibiotic-resistant versions of these genes, which would be no better or worse than non-resistant versions of these genes. Given the high rate of mutation and the rapid rate of reproduction in bacteria, the odds against every resistant gene surviving the process of genetic drift are astronomical even in one species of bacterium, let alone in every species that we attack with antibiotics.
We shall omit the detailed mathematics, but the reader who wishes to verify this can do so by reference to the key result that for neutral mutations, the rate of substitution per site per generation in the gene pool of a population is equal to the rate of mutation per site per generation in the genotype of an individual.
For further discussion, see our article on Front-Loaded Evolution.
[edit] Example: Self-inflicted DNA damage to enhance population diversity
A single bacterium can grow into an entire colony. If no mutations occurred, this bacterial colony would be genetically uniform. Evolutionary theory argues that such a monolithic culture would be highly sensitive to stresses and be less able to adapt to changing environments. It has been shown that a single bacterium can rapidly generate an entire biofilm community with a genetically diverse population using a self inflicted DNA damaging mechanism, oxidative stress. [5]
Oxidative stress[6] is the general condition where highly reactive oxygen species are produced. These species (e.g., hydrogen peroxide, hydroxyl radicals, superoxide anion) directly damage lipid membranes, proteins and DNA. When this damage is severe, cell death occurs. In fact, oxidative stress is the primary mechanism by which white blood cells kill microbes in innate immune system.
As a defense against oxidative stress, bacteria have classes of proteins which can repair DNA damage that results from oxidative stress. One such mechanism is the repair of DNA double strain breaks by proteins such as recA, recB, recC, ruvB, and ruvC proteins. Double strain breaks are the cleavage of a DNA molecule into two separate fragments. These repair mechanisms introduce errors into the DNA strands, thereby resulting in a potential gene mutation. By occurring multiple times during proliferation, it is possible to generate from a single bacterium a population of bacteria with a wide degree of genetic diversity.
It was recently discovered that Pseudomonas aeruginosa will capitalize upon this mechanism and actually generate their own oxidative stress as a means of enhancing the diversity of the bacterial population. This enhanced diversity has been shown in the laboratory to increase the rate of evolution of antibiotic resistance[7]. By possessing a mechanism that causes mutation as a means of increasing genetic diversity, we have an example that poses a difficult question to creationists. If there are no beneficial mutations, why was P. aeruginosa "designed" with a mechanism to damage its DNA?
[edit] Example: Nylonase
Another clear example of a beneficial mutation is the bacteria that have evolved to eat nylon-6. [8][9]
Evidently this is a beneficial variation, since it allows the colonization of a new niche; and obviously the mutation can't have existed since some imaginary act of fiat creation 6000 years ago, since bacteria with this variation can only feed on nylon-6, and nylon was only invented in 1935.
One interesting aspect of this mutation is that it consists of the insertion of a single nucleotide, causing a frame-shift, so that the protein being coded for, after the frame-shift, is nothing like the protein coded for in the wild type: the mutation is small in terms of nucleotides, but makes a radical change to the protein.
[edit] Example: The Ames test
One important application of the existence of beneficial mutations is the Ames test, a common method of biological assay to discover the mutagenic and carcinogenic potential of chemicals (see Ames, Lee, and Durston, 1973. An improved bacterial test system for the detection and classification of mutagens and carcinogens. Proc. Natl. Acad. Sci. USA 70: 782-786.)
The concept is very simple. Take a line of bacteria (Salmonella enterica) with a broken his operon. An operon is a set of genes that get transcribed together; the his operon consists of the genes by which bacteria make their own histidine. The result, then, of having a broken his operon is that the bacteria can't grow on a standard growth medium, but require a growth medium laced with histidine, so they can get it in their diet.
So, to test a substance to see how mutagenic it is, you take a plate of growth medium which only has enough histidine in it to last for a few generations; and you also add a standard quantity of the suspected mutagen.
It follows that the bacteria will starve to death after a few generations, unless some of the bacteria undergo a beneficial mutation repairing the his operon and allowing them to live without histidine. Each bacterium so affected will survive when the unaffected bacteria starve for want of histidine. So if you count the number of colonies of bacteria on the plate after the histidine runs out, you can find out how many beneficial mutations have occured. As there is nothing special (on the DNA level) about mutations that repair the his operon, this gives us a measure of the chemical's mutagenic potential. You can read more about the Ames test here
This sort of biological assay depends crucially on the existence of beneficial mutations. Imagine trying to do the experiment the other way round, starting with bacteria with an intact his operon, and seeing how often the mutagen broke the operon. The problem is, how on Earth would we tell? We should have to comb each Petri dish for dead bacteria, and then perform an autopsy to discover if they died due to a broken his operon or some other reason.
In the Ames test, by contrast, it is easy to spot the beneficial mutations because a single bacterium undergoing repair of the his operon can be the parent to an entire colony of bacteria, which makes them easy to spot.
This is one more reason to be thankful that our scientific techniques were developed by scientists rather than dogmatic creationists: if Ames had held the creationist dogma of "no beneficial mutations", he would have dismissed his test as impossible a priori.
[edit] Example: Reversion
The repair of the his operon in the Ames test is just one case of the phenomenon known as reversion. If one mutation breaks a gene, then another subsequent mutation can repair it, unless the effect of breaking the gene is absolutely lethal. So, for example, the effects of a single nucleotide substitution can be undone by another single nucleotide substitution in the same place; the effects of the insertion of a single nucleotide, causing a frame-shift can be undone by the deletion of a single nucleotide shifting the reading frame back.
Now, if the breakage of a gene is harmful, then a subsequent mutation mending the gene again must necessarily be beneficial.
[edit] Example: Re-evolution of b-galactosidase
In experiments published in 1992 (see BG Hall, 1982, Evolution on a Petri Dish: The evolved b-galactosidase system as a model for studying acquisitive evolution in the laboratory. Evolutionary Biology 15: 85-150) Barry Hall entirely deleted the gene for b-galactosidase from the lac operon of bacteria, so that they couldn't metabolise lactose. Grown in nutrients rich in lactose, so that there is a benefit to being able to metabolize it, mutations in a gene known as ebg make it able to produce a protein that performs the function of b-galactosidase, and the control region of the ebg gene also evolves so that the gene is only transcribed in the presence of lactose.
[edit] Example: Adaptation to temperature regimes in E. coli
The results are reported in Bennett, A.F., Lenski, R.E., & Mittler, J.E. (1992). Evolutionary adaptation to temperature I. Fitness responses of Escherichia coli to changes in its thermal environment. Evolution, 46:16-30.[10]
- Adaptation of the groups to temperature was measured by improvement in fitness relative to the ancestor, as estimated by competition experiments. All four experimental groups showed improved relative fitness in their own thermal environment (direct response of fitness) ... No necessary tradeoff between direct and correlated responses of fitness was apparent: for example, the improved fitness of the 42°C group at 42°C was not accompanied by a loss of fitness at 37°C or 32°C
It is clear that since the experimenters began with a single ancestor for all the E. coli used in the experiment, these variations cannot be simply the result of selection acting on pre-existing variation in the population, but must be the result of mutations favored by selection.
[edit] Example: Glucose transport genes in yeast
So far we have mentioned beneficial substitution, insertion and deletion mutations. Beneficial duplication of genes has also been observed, as for example in Brown,Todd, and Rosenzweig, Multiple Duplications of Yeast Hexose Transport Genes in Response to Selection in a Glucose-Limited Environment.[11] The title of the paper really says it all: brewer's yeast (Saccharomyces cerevisiae) was kept in a glucose-poor environment for 450 generations. By the end of that time:
- Relative to the strain used as the inoculum, the predominant cell type at the end of this experiment sustains growth at significantly lower steady-state glucose concentrations and demonstrates markedly enhanced cell yield per mole glucose, significantly enhanced high-affinity glucose transport, and greater relative fitness in pairwise competition.
Analysis of the underlying changes to the DNA showed that this adaptation to gluscose-poor conditions was produced by multiple duplication of genes coding for glucose transport proteins.
[edit] Example: Evolution of colonialism in algae
Chlorella vulgaris, a green alga, provides an interesting example of adaptation. Normally, it is a single-celled form. However, when it is grown along with the predator Ochromonas vallescia, it evolves to be a colonial organism, at first growing in large colonies, and then evolving into a stable eight-celled form which is too large for O. vallescia to eat, but small enough that every member of the colony has good access to nutrients. The benefit of not being eaten is obvious; the novelty of this behavior in Chlorella is a matter of simple observation, since the experimenters began with a single-celled strain; and the fact that this trait, once evolved, is maintained in C. vulgaris cultures when they are grown in the absence of O. vallescia, shows that this is the result of a change in their genes rather than an instinctive reponse to the presence of predators.
One description of such an experiment can be found in Boraas, Seale, Boxhorn, Phagotrophy by a flagellate selects for colonial prey: A possible origin of multicellularity, Evolutionary Ecology 1998, 12, pp. 153-164[12].
[edit] Example: Mutations in "higher" organisms
It is a lot easier to observe beneficial mutations in lab bacteria or in small eukaryotes such as yeast and algae, thanks to their high rate of reproduction and the ability to keep a population of millions in a Petri dish or chemostat. It is one thing to breed E. coli for 2000 generations; it is quite another thing to breed white mice for 2000 generations.
However, we can often see the effects of mutations in the wild. Consider, for example, resistance to the poison warfarin in rats. The advantages of this variety are clear --- when rats are being poisoned by warfarin. When they are not, then the allele resposible for resistance is disadvantageous, as it increases the rats' dependency on vitamin K; and, indeed, in areas where the use of warfarin is discontinued, the occurrence of this allele in the population drops rapidly to zero.
It follows that this allele can't have been hanging around for 6000 years or more since a hypothetical act of fiat creation, since it would have disappeared entirely from the rat population long before warfarin was invented and it became beneficial.
Similar remarks can be made about, for example, those varieties of plants that thrive on the metal-rich spoil heaps of mining operations. Since these varieties flourish only on spoil heaps, and are elsewhere outcompeted by varieties without this trait, the alleles responsible would have been eliminated by natural selection before there were spoil heaps to grow on.
[edit] Creationist responses
Creationists have a number of stock excuses when presented with evidence like this.
The first is to claim, without evidence, that such mutations existed ever since God miracled them into existence in the beginning. For this reason, we have confined our list to cases where we know by direct observation or the application of a little common sense that this cannot be the case.
The second common creationist excuse is to claim, without evidence, that the new phenotypes were not produced by random mutation, but by the equally random process of recombination. There are a number of problems with this excuse. First, many of the experiments listed involve bacteria that reproduce asexually and do not undergo recombination. Second, many of the experiments are performed on a clonal line: a population descended from a single individual. Third, this excuse definitely won't wash now that scientists can study the mutations involved on the level of DNA, and can verify that the new forms are the result of single nucleotide substitutions, insertions, deletions, and duplications: errors in replication or recombination of DNA which are by definition mutations.
The most common response, however, when a creationist demands to see evidence of beneficial mutations, and is shown such evidence, is to start complaining that it is not evidence for something else, such as "macroevolution" (which is the accumulation of many mutations over thousands or millions of years) or "increase of information" (a term that creationists are unable or unwilling to define: see our main article on Mutations and Information for further details).
In principle, there is a fourth possible creationist response to being shown evidence of beneficial mutations, which would be to say: "Oh, it seems that I was misinformed. Thank you for putting me right". We have never ever seen this happen. The best one can expect under these circumstances is that the creationist stops claiming that there are no beneficial mutations, and moves on to being wrong about some other aspect of biology.
