Mutation plays a pivotal role in evolution, as it generates diversity to fuel the selection processes. In more practical sense, animal response to artificial selection and the genetic basis of human disease are two among fields where it can be elucidated better by understanding mutation rate affecting genetic variation (Knightley & Halligan, 2009).
It is observed that there is a distinct mutation rate for each taxonomic group, but lot of variations are observed among groups (Baer et. al, 2007). This fact indicates that the mutation rate is under the subject of evolutionary processes to optimize fitness and to promote adaptation (Sniegowski et. al, 2000). Mutation rate trend (µ, measured as mutation/site/generation) is found to be differ in eukaryotic and microbes (virus and bacteria). The rate is observed to be scale positively with the genome size of eukaryote. Meanwhile, the rate is inversely correlated with microbe genome size, with the product of µ and G (genome size) is constant and it is approximately to be 0.003, known as “Drake’s Rule” (Drake, 1991).
One of the evolutionary theory explaining the pattern of mutation rate variability is ‘mutator hitchhiking theory’. An allele which increase the mutation rate (mutator), such as a naturally occurring allele associated with defective DNA repair (Ness et. al, 2015), will be found more frequently with beneficial allele. This creates non-random association between them, called as linkage disequilibrium. As a beneficial mutation increase in frequency, the mutator will also hitchhike to the higher frequency, even though that the mutator does not exert fitness effect.
The extent of linkage disequilibrium differs according to the type of breeding system. In sexually reproducing organism, homolog chromosome segment is exchanged each other during gamete formation, so that the linkage between mutator and beneficial allele will not last in the successive generations. The association depends on the recombination rate and on the average, it will uncouple in only two generations (Drake et. al, 1998).
In asexual organism, due to lack of genetic recombination, the modifier allele will always be associated with the beneficial mutation. Hence, if the beneficial allele undergoes fixation, mutator allele will also sweep through population, but with the consequence that many deleterious mutations emerge. Increased genetic load from deleterious mutation creates indirect selection for mutator allele, so that the frequency of mutator declines, creating a fluctuation in mutation rate (Baer et. al, 2007). Gerish et. al (2006) gives different view of the consequence for the ‘mutator hitchhiking’, in which that the process leads to extinction. Gerish ‘mutational catastrophe’ model suggested that the increased load from deleterious allele is delayed relative to the effect of beneficial mutation. As a result, mutation rate keeps increasing until selection is not effective anymore to bring it down, with ultimate result in extinction. Thus, the modifier hitchhiking theory is not able to fully resolve the explanation for mutation rate variation in different mode of reproduction.
Sung et. al (2012) stated the unifying theory for the pattern of mutation rate across tree of life, where it is inversely proportional to the average of effective population size and can be understood from ‘drift barrier’ hypothesis . The type of evolutionary forces shape the mutation rate depends on the effective population size (Ne). For example, the effect of genetic drift outweighs natural selection in small population. Selection is not working effectively in small population, so that the mutations with deleterious or beneficial effect in large population will behave stochastically as if it is neutral.
Most of the mutations are having deleterious consequence, so there will be purifying selection favouring low mutation rate. Under finite population size, the effect of drift is greater than the power of selection, so that the selection cannot drop the mutation rate into its lowest possible value. In order to selection to act effectively, the effective size of population needs to be above threshold where selection effect surpasses drift effect, termed as drift barrier. The boundary for mutation to be considered neutral or under effective selection is (Kondrashov et. al, 2006).
This fact has been proposed as the reasoning why mutation rate in prokaryotic is lower than eukaryotic. Microbe’s effective population size is large (often close to its maxima), so the selection can work effectively. Mutation rate per genome among microbe is observed to be constant and low, which indicates that optimum (lowest) value has been reached. Meanwhile, the mutation rate in eukaryotic is higher and this is due to the relaxed selection from finite population size (Baer et. al, 2007).
In population with large Ne, to what extent that mutation rate can be pushed into its lowest limit depends on fitness cost of reduction in mutation. Mutation rate reduction puts a cost for the cell, which is associated with physiochemical of the process. Cell must divert their energy to preserve the genome, such as by increasing the accuracy of genomic replication (proofreading) or by making an intensive mutation repair. Hence, the optimum rate of mutation will be set by balance by the benefit of having less deleterious allele and by the cost of maintaining genomic conservation. This hypothesis has been tested experimentally by exposing Drosophila to X ray radiation over 600 successive generations. The mutation rate was decreasing during exposure and then, returned to the normal after the radiation was removed (Nothel, 1986)
Sniegowski and Raynes (2013) contended the idea of cost of fidelity, with the reasoning that the effect of increasing replication accuracy will constrain cell replication’s speed and thus, put more burden on the fast-replicating organism, but in nature the mutation rate is lower in microbe than in multicellular organism. Drake et. al (1998) suggested that this paradox might be due to the effect of genome size or the number of cell division and this is supported by Knightley et. al (2015) which stated that there is only small variation across taxa when the mutation rate is expressed per cell division, indicating the similar cost of fidelity.
Several experimental approach have been explored to estimate the mutation rate. Nachman and Cowel (2000) used the comparison of neutral locus (silent sites) between species whose divergence time have been known and they found that upon comparison with chimpanzee the human’s spontaneous mutation rate to be in order of 10-8. The similar result was also obtained by Kong et. al (2009). Analysing the de novo mutation from parents-offspring up to 3 generations, they estimated that the rate was about 1.2 x 10-8, after accounting for the father’s age at conception. The more straightforward method is probably using reporter construct with obvious phenotypic change upon mutation. However, the drawback is that the mutations do not always lead to the phenotypic alteration (or the change is too subtle to detect), resulted in underestimation. Another downside is that there will be a loss in accuracy when the rate is extrapolated to the whole genome (Baer et. al, 2007).
The most promising approach to characterize mutation rate is by coupling mutation accumulation (MA) experiment with high-throughput sequencing. MA experiment is done by maintaining several lines from one ancestral line. Each line is then periodically subjected to the extreme population bottleneck which leads to the near absence of selection. Due to the overwhelming effect of drift, different mutations accumulate and fix stochastically, even the ones that possess harmful effect (unless the mutation is lethal or causing sterility). Whole genomic sequencing from multiple generations of MA lines and preserved ancestral line is then done to allow for the direct estimate of mutation rate with narrow confidence limit (Kingihley & Halligan, 2009). MA experiment from Kinghtley (2012) showed C. reinhardtii mutation rate to be in order of 10-10 (one of the lowest in eukaryote) and it aligns with previously calculated mutation rate when regressed with effective population size.
In conclusion, mutation rate variability is shaped by interplay of among several evolutionary forces. The hypothesis of modifier allele hitchhiking is not able to fully accommodate the variability among recombining and non-recombining organism. Meanwhile, more universal explanation involving the effect of effective population size. Effective population size determines whether selection can overcome the drift effect and if it exceeds the drift barrier, the ability of selection to minimize mutation rate is constrained by the ‘cost of fidelity’. Finally, mutation accumulation experiment coupled with high throughput sequencing provides platform for exploring this field further in details.