Evolution of the mutational process under relaxed selection inCaenorhabditis elegans

Abstract

The mutational process varies at many levels, from within genomes to among taxa. Many mechanisms have been linked to variation in mutation, but understanding of the evolution of the mutational process is rudimentary. Physiological condition is often implicated as a source of variation in microbial mutation rate and may contribute to mutation rate variation in multicellular organisms.Deleterious mutations are a ubiquitous source of variation in condition. We test the hypothesis that the mutational process depends on the underlying mutation load in two groups of Caenorhabditis elegans mutation accumulation (MA) lines that differ in their starting mutation loads. “First-Order MA” (O1MA) lines maintained under minimal selection for ∼250 generations were divided into high-fitness and low-fitness groups and sets of “second-order MA” (O2MA) lines derived from each O1MA line were maintained for ∼150 additional generations. Genomes of 48 O2MA lines and their progenitors were sequenced. There is significant variation among O2MA lines in base-substitution rate (µbs), but no effect of initial fitness, whereas the indel rate is greater in high-fitness O2MA lines. Overall, µbs is positively correlated with recombination and proximity to short tandem repeats and negatively correlated with 10 bp and 1 Kb GC content. However, probability of mutation is well-predicted by the three-nucleotide motif. ∼90% of the variance in standing nucleotide variation is explained by mutability. Total mutation rate increased in the O2MA lines, as predicted by the “drift barrier” model of mutation rate evolution. These data, combined with experimental estimates of fitness, suggest that epistasis is synergistic.

Low Base-Substitution Mutation Rate but High Rate of Slippage Mutations in the Sequence Repeat-Rich Genome of Dictyostelium discoideum

“…But a bias towards insertions relative to deletions has been observed for some eukaryotes (DENVER et al 2004;LYNCH et al 2008;SUNG et al 2012a;SUNG et al 2012b;FARLOW et al 2015; BEHRINGER AND HALL 2016a). Our results indicate that deletion events outnumber insertions in D. discoideum, consistent with a recent study in C. elegans (SAXENA et al 2019).…”

A broad mutational target explains a fast rate of phenotypic evolution

“…We therefore directly inquired whether this deletion (and the other mutations) occurred recurrently at a detectable frequency by analyzing sequence data of other MA lines (Saxena et al 2019: 75 other MA lines, 3469 nuclear mutations). We did not find any other mutation at the corresponding positions and the closest mutations were at least 4 kb away (Supplementary File 7).…”

Selection and drift determine phenotypic stasis despite genetic divergence

“…These simulations consider a fixed number (100) of freely recombining loci while controlling the probability that each loci is mutated at each generation (x-axis; equivalent to changing the mutational target size). In C.elegans , given a per-site SNP mutation rate on the order of 10 − 9 (Saxena et al (2018)), a mutation rate of 10 − 4 -10 − 3 would correspond to a total target size of between 1Mb and 10Mb (for reference, the average gene length is around 5Kb). Even when considering such large loci, the number of mutations reaching 5% frequency in the population is small.…”