This week’s BEACON Researchers at Work blog post is by University of Washington postdoc Jenna Gallie.
Left: Haley Lindsey; right: Jenna Gallie.
Life has existed on Earth for nearly four billion years. Given that organisms have been evolving continuously, why are they still not perfectly adapted to their environments? One reason is that environments are constantly changing. This is particularly evident at present, as Earth’s climate is changing rapidly. In addition, there are many other situations where the survival of organisms depends on their capacity to adapt to dynamic environments. For example, drug treatment regimes drastically alter the environment experienced by infectious microbes.
A metaphor that can aid understanding of the evolutionary effects of changing environments is that of the ‘fitness landscape.’ The fitness landscape is a map from genotype to fitness for a given environment, where elevation gauges fitness (see Fig. 1 for a simple example). The genotypes involved form a network where neighbours are mutationally accessible. Populations take ‘steps’ through mutation, and are driven up slopes by natural selection (as fitter genotypes prevail). The operation of mutation and selection generates an evolutionary ‘path’ on the fitness landscape. This path may lead to an ‘adaptive peak’, a genotype where all mutational neighbours are less fit (e.g. ‘ab’ in Fig. 1B). Once a peak is reached the population can remain static if the environment is constant.
Fig. 1: The fitness landscape (an example with two mutations). (A) Four genotypes (AB, aB, Ab, and ab) are linked by single mutations. (B) The landscape results from the inclusion of the fitness ‘heights’ of the genotypes. To get from genotype AB to ab, there are two ‘paths’ available (red and blue). The blue path is accessible (each successive genotype has higher fitness), whereas the red path is not. (Figure created by Ben Kerr and Jenna Gallie).
Since the topography of the fitness landscape is dependent on the environment, the evolutionary path can be altered by environmental change. When the environment shifts, the relative fitness of genotypes may change, altering the landscape. If the environmental shift is instantaneous, any corresponding change in the landscape is immediate. Conversely, more gradual environmental changes may result in transitions through several forms of the landscape as each stage of the environmental change is realized. These additional landscape forms may alter the accessibility of mutational paths. In this way, the rate of environmental change can have profound effects on evolutionary outcomes. This is a particularly pressing issue in the face of modern climate change, which is occurring faster than ever before. Many predictions focus on how today’s organisms will survive in conditions projected by global warming models in many years time. However, for organisms with short generation times (e.g. annual plants), many generations will pass – and so populations may evolve – before the projected conditions are realized. To improve predictions regarding the biological effects of climate change, it is essential to understand how the rate of environmental change affects evolution.
For the past 1.5 years, I have been a postdoctoral researcher in Dr. Ben Kerr’s laboratory at the University of Washington. Haley Lindsey, Ben Kerr and I have been using a combination of experimental evolution and molecular genetics to explore the effect of environmental change on adaptation, with particular emphasis on how the rate of environmental change influences evolutionary outcomes. In our research we use a well-established microbial model system: populations of Escherichia coli and the antibiotic rifampicin. Microbial systems are often used in experimental evolution as a combination of short generation times (minutes to hours) and large population sizes (~1010 organisms) enable evolution to be directly observed in real time. Additionally, microbial populations can be frozen in a state of suspended animation almost indefinitely, and then revived for comparison with derived genotypes. We chose rifampicin as an environmental stressor because its effects on E. coli populations are well understood; mutations conferring resistance to rifampicin usually occur in specific regions of the rpoB gene, which encodes the major subunit of RNA polymerase (the enzyme that catalyses transcription).
Fig. 2: Representation of the three differing rates of drug addition during our experiment.
Using the E. coli-rifampicin system, we devised an experiment to explore the effects of the rate of environmental change on adaptation: increasing amounts of rifampicin were added to replicate E. coli populations that were propagated over many generations. The populations were divided into three treatment groups that differed only in the rate at which rifampicin was added (Fig. 2). The first treatment group (‘Rapid’) received the full concentration of rifampicin immediately, while in the second group (‘Gradual’) the rifampicin concentration was slowly increased over the course of the experiment. The final group (‘Moderate’) was subjected to an intermediate rate of rifampicin increase. Importantly, the final concentration of rifampicin was the same in each treatment group. All surviving populations from each treatment group were periodically frozen throughout the experiment, allowing the evolutionary changes occurring in each treatment group to be analysed. By comparing the changes found in survivors from each treatment group, we aimed to determine if and how evolutionary outcomes were influenced by the rate of rifampicin addition.
The results obtained so far are promising. Notably, faster addition of rifampicin led to a lower survival rate, showing that rapid environmental change can lead to higher rates of extinction. A second interesting finding was that survivors from the Moderate and Gradual treatment groups contained a wider variety of rpoB mutations than those from the Rapid treatment group. We are currently exploring whether specific evolutionary paths taken under the Moderate and Gradual treatments were accessible to populations exposed to the Rapid treatment. Preliminary data suggests that a greater variety of mutational paths were available under the Moderate and Gradual treatments than under the Rapid treatment. Together, these findings indicate that rapid environmental shifts can severely constrain evolutionarily outcomes. Our findings highlight the need to consider the rate of environmental change in other situations. In particular, rates of change should be considered when making predictions about the biological effects of climate change.
For more information about Jenna’s work, you can contact he
r at jgallie at uw dot edu.
