Essay: Mutation rate

Mutation acts to promote the evolution of populations. In order to better understand and describe the tree of life, scientists have been investigating mutation rate and its variation. Through the course of the past decades, mutation rates have been estimated for numerous taxa and revealed both intra and interspecific variation as well as heterogeneity across the genome. This variation led to the inevitable conclusion that mutation rate is, in fact, evolvable.
In 1991, Drake performed an experiment to estimate mutation rates in DNA-based microbes, which included bacteriophages, a bacterium, a yeast and a filamentous fungus, and set the foundations for the study of this matter. He observed a narrow range of variation of genomic mutation rate in the studied organisms and concluded that there must have been an equilibrium value across taxa. The experiment also allowed him to draw the conclusion that there was an inverse scaling between average mutation rate per base pair and genome size. Although Drake’s work was an invaluable starting point, some of his assumptions have been proved inaccurate.
The subject of mutation rate is related to that of fitness effects. A new mutation affecting the fitness of an individual is target of selection. One of the initial assumptions regarding selection stated that it would act to lower the mutation rate since advantageous mutations are rare relative to deleterious mutations. This would allow selection to minimise the number of fitness lowering mutations and generate genetic diversity and adaptation.
However, estimates of mutation rate are not zero. This observation led to the thought of a physicochemical barrier as a limiting factor. Acting against the reduction of the mutation rate to zero would be the principle of ‘cost of fidelity’. The maintenance of the accuracy of replication and sequence repair processes would be the reason for the trade-off between selection and the cell costs.
In the past few years, an alternate explanation to the lower limit of mutation rate has arisen: the ‘drift-barrier hypothesis’. This hypothesis suggests that, as selection pushes mutation rates down, said limit is set by random genetic drift rather than physiological limitations. Sung et al. estimated mutation rates in two distinct organisms, a prokaryote with a small genome and a unicellular eukaryote with a large genome. These authors were able to conclude that the reduction in mutation rate caused by selection reaches such a level in which a drift barrier is imposed. The estimates of per-base mutation rate were observed to scale negatively with the effective population size (Ne) leading to the expectation of lower mutation rates in species with larger Ne. When the mutation rate has been significantly reduced by selection, the power of drift, which is also inversely proportional to Ne, is logically the cause of the observed lower limit.
In a cell, the processes that handle its genetic material are numerous and each is an additional step towards DNA sequence alterations. Selection can act on these processes directly and modulate mutation rates by changing DNA replication fidelity, by shaping the cell’s response to mutagen exposure, by altering DNA repair efficiency and by buffering fitness effects of new mutations.
Natural selection can also shift mutation rates by changing the frequencies of ‘mutator’ and ‘antimutator’ alleles. The former increase mutation rates and the latter reduce such rates. In asexual populations, as an advantageous mutation in linkage disequilibrium with a mutator allele reaches fixation, selection will act to allow an increase in frequency of antimutator alleles, so that the fitness effects of deleterious mutations is minimised. In contrast, selection favouring mutator alleles is efficiently weakened by recombination in sexual populations. Even if said allele hitchhikes with a beneficial mutation, recombination will rapidly eliminate such associations.
Several theories have been suggested in order to explain mutation rate variation across taxa: (1) The ‘generation-time hypothesis’, that states that the shorter the generation time, the faster will be the accumulation of mutations and, consequently, the evolution of the lineage; (2) the ‘metabolic-rate hypothesis’, which suggests an increasing mutation rate due to higher metabolic rate, which, in turn, is responsible for producing more mutagen agents, such as oxygen free radicals; and (3) the ‘DNA repair hypothesis’, that associates better DNA repair systems with less mutations.
In order to better assess an estimate of the mutation rate, researchers often use a mutation accumulation (MA) approach. In such an experiment, several lines are allowed to accumulate spontaneous mutations independently for many generations. These lines are initially derived from an ancestral line. During the experiment, the lines are kept at low population sizes so that the effect of selection is minimised. Combined with a whole genome sequencing analysis, the comparison of the DNA sequences of the ancestral and the derived lines allows for an estimation of the mutation rate.
In recent years, MA experiments have been extensively used to study mutation rate evolution. Perfeito et al. measured the genomic mutation rate that generates advantageous mutations in E. coli and observed a value approximately 1000 higher than previous estimates. The authors suggest that the difference is due to clonal interference, and that competition will lead to an underestimation of the mutation rate. Farlow et al. assessed similar mutation rates for S. pombe and S. cerevisiae in a mutation accumulation experiment. However, these species differ significantly in the frequency of each mutation class, indicating that the process of species divergence has not been acting to greatly shift mutation rate
Mutation rate has also been shown to depend on the genomic context of each site. Ness et al. described fine-scale heterogeneity in the mutation rate with clusters of multiple mutations occurring at closely distributed sites in a mutation accumulation experiment using C. reinhardtii. Zhu et al. have confirmed a higher mutation rate at G/C bases, especially in two sequence contexts, in which the rate was twice as high, consistent with cytosine methylation.