Microbial Experimental Evolution - a proving ground for evolutionary theory and a tool for discovery

Abstract

Microbial experimental evolution uses controlled laboratory populations to study the mechanisms of evolution. The molecular analysis of evolved populations enables empirical tests that can confirm the predictions of evolutionary theory, but can also lead to surprising discoveries. As with other fields in the life sciences, microbial experimental evolution has become a tool, deployed as part of the suite of techniques available to the molecular biologist. Here, I provide a review of the general findings of microbial experimental evolution, especially those relevant to molecular microbiologists that are new to the field. I also relate these results to design considerations for an evolution experiment and suggest future directions for those working at the intersection of experimental evolution and molecular biology.

Keywords: adaptation; directed evolution; experimental evolution; microbiome evolution; selection.

Figure 1. Mechanisms of propagation for experimental evolution

(A) Batch culture requires the regular dilution of culture into fresh media. These experiments are relatively easy to establish, since a range of vessels commonly used in a microbiology laboratory can be used for batch culture. These experiments can be scaled to a large number of replicates, for example when using 96‐well plates. (B) Chemostat culture systems include mechanisms for the constant supply of fresh medium. This provides for the continuous cultures of populations and constant growth without large fluctuations in populations size or growth phase. (C) Microfluidics provides the most precise control over the supply of media and supplements to cell cultures. Microfluidics may need to be custom designed, and the number of replicates will be limited. (D) Emulsion cultures take advantage of small cell‐containing vesicles that form when mixing an oil, surfactant and cells. The number of cells in each vesicle is determined by the ratio cell, surfactant and oil. The cells can be mixed back into a single population by vortexing and centrifuging the solution. One advantage of evolving cells in a large number of small populations is that this can select for yield per‐vesicle rather than rapid growth 144. (E) Mutation accumulation introduces a regular, single‐cell bottleneck into each replicate population. This achieved by streaking out cells on a petri dish and then choosing a single colony (founded by a single cell) to streak out the next plate. (F) Microbial cultures can be introduced into a model organism, often a plant or a mouse, and left to propagate for a number of generations before it is recovered from the organism. The recovered cells can be analysed or subjected to further propagation in the organism. This mode of experimental evolution allows for the testing of unanticipated organism‐specific features of the environment that are difficult to replicate in the laboratory.

Figure 2. Three consistent results from evolution experiments

(A) Genetic parallelism. A signature of natural selection is the repeated evolution of mutations in the same genes in independent populations. The expected number of multi‐hit genes mutated across six replicate populations in a hypothetical 1000‐generation experiment without natural selection (grey shaded) and an example of the number of multi‐hit genes in a population with selection (orange line) 6. (B) Diminishing returns epistasis. The fitness effect of a beneficial mutation is negatively correlated with the fitness of the genetic background in which it occurs (figure adapted from 25). (C) Stable polymorphism can evolve, whereby multiple ecotypes, each adapted to a different niche in the microcosm, coexist in the population. Figure adapted from 27. One possible outcome of experimental evolution is that populations will adapt by successive sweeps of beneficial mutation, occasionally hampered by clonal interference (D).

Figure 3. Tracking the dynamics of mutations that underlie adaptation using DNA sequencing

Each line shows the trajectory of a mutation that arises during evolution. (A) Whole metagenome sequencing can track all mutations in the genome, but is limited to tracking mutations that attain a high frequency, typically > 1–10%. (B) Amplicon sequencing can track the change in frequency of as many as 500,000 distinct genetic lineages, but does not convey the identity of the beneficial mutations that cause adaptation.