The Environment Affects Epistatic Interactions to Alter the Topology of an Empirical Fitness Landscape

Abstract

The fitness effect of mutations can be influenced by their interactions with the environment, other mutations, or both. Previously, we constructed 32 ( = 25) genotypes that comprise all possible combinations of the first five beneficial mutations to fix in a laboratory-evolved population of Escherichia coli. We found that (i) all five mutations were beneficial for the background on which they occurred; (ii) interactions between mutations drove a diminishing returns type epistasis, whereby epistasis became increasingly antagonistic as the expected fitness of a genotype increased; and (iii) the adaptive landscape revealed by the mutation combinations was smooth, having a single global fitness peak. Here we examine how the environment influences epistasis by determining the interactions between the same mutations in two alternative environments, selected from among 1,920 screened environments, that produced the largest increase or decrease in fitness of the most derived genotype. Some general features of the interactions were consistent: mutations tended to remain beneficial and the overall pattern of epistasis was of diminishing returns. Other features depended on the environment; in particular, several mutations were deleterious when added to specific genotypes, indicating the presence of antagonistic interactions that were absent in the original selection environment. Antagonism was not caused by consistent pleiotropic effects of individual mutations but rather by changing interactions between mutations. Our results demonstrate that understanding adaptation in changing environments will require consideration of the combined effect of epistasis and pleiotropy across environments.

Relative importance of evolutionary dynamics depends on the composition of microbial predator–prey community

“…Alternatively, slow-growing PT and TT colony types could have been organised in the bottom of mixed biofilms resulting in enhanced protection against protist predation (Kim et al, 2014) or could have hitchhiked along with SM and WS colony types in the mixed biofilms (Popat et al, 2012;. Although further experiments are needed to test these hypotheses, our results suggest that concurrent selection by two protists potentially changes the topology of bacterial fitness landscape in ways that allow bacterial adaptation against multiple enemies (Flynn et al, 2013).…”

Environment Determines Epistatic Patterns for a ssDNA Virus

“…All three possible comparisons showed significant differences (P , 0.001 on the basis of two-sided, Welch two-sample t-tests with a Bonferroni correction for three tests). Diminishingreturns epistasis appears to be common in bacterial (MacLean et al 2010;Chou et al 2011;Khan et al 2011;Sousa et al 2012;Flynn et al 2013) and viral (Bull et al 2000;Pearson et al 2012) systems. This pattern has typically been observed by changing the background fitness by means of modifying the genetic background, while here we reduced fitness by altering the environment.…”

“…The environment determines fitness effects of mutations (Remold and Lenski 2004;Lalíc et al 2011;Vale et al 2012;Flynn et al 2013) and therefore also affects epistatic patterns (You and Yin 2002;Lehner 2011;Flynn et al 2013;Lalíc and Elena 2013;Wang et al 2013). Rokyta et al (2011) used a model based on Fisher's geometric model that assumed additive phenotypic effects for mutations with an intermediate phenotypic optimum to explain the complex epistatic patterns among pairs of individually beneficial mutations for the single-stranded (ss)DNA bacteriophage ID11.…”

Survival and innovation: The role of mutational robustness in evolution

“…Indeed, copious examples in the literature show that the effect of mutations on fitness is conditional to the genotype-byenvironment interaction, which re-shapes the fitness landscape of a population [15e18]. Fluctuating conditions, including changes in population sizes or in the source of carbon, could change the interaction between mutations (e.g., sign epistasis) and the distribution of mutations fitness effects, allowing population take previously prohibited evolutionary paths [19,20], or precluding them from adopting previously adaptive solutions [21,22]. For example, Flynn and colleagues showed that the positive adaptive interaction of five mutations that were beneficial in Escherichia coli in a specific environment changed to being deleterious when tested in a subset of 1920 environments other than the original one [21].…”