The more traditional approach in Evolutionary Biology is to observe organisms in their natural environments, and to deduce from behavioral, phenotypic, and genetic data, the selection pressures they are exposed to and the adaptation they undergo. A challenge of this approach is that it is difficult to establish causalities because experimental manipulations are often impossible in natural systems. Conducting experimental evolution in laboratory setups is an elegant way around this problem. It allows the experimenter to define and manipulate selection pressures, and to observe the emergence of new mutations and to follow their fates in real time. Combined with computational analysis, modelling evolution in silico, experimental evolution has become an extremely powerful approach to establish causal links between selection pressures, mutational patterns and adaptions, and thus to understand Evolution in Action. This approach is furthest developed in small single-celled organisms and their communities, and has been successful in explaining patterns of resistance evolution against antibiotics, evolutionary innovations, and community dynamics in expanding populations, and was also applied to many other areas of Evolutionary Biology. Important to note is that computational simulation/modelling approaches are also gaining increasing attraction in macroecology as a powerful method to predict evolution in these more complex systems. For Research Core Area 3, the following two projects addressing different aspects of the topic were selected:
Table of contents
Tracking Genomic and Phenotypic Changes Induced by Experimental Manipulation of Natural Pollinator Communities
Plant–pollinator interactions are essential for plant reproductive success and as food resource for pollinators. However, the dramatic decline in pollinators currently threatens the existence of many flowering plants. How – and whether – plants react to the loss of pollinators has therefore attracted lots of attention in research on wild and crop plant communities. A rapid evolutionary response of plants could be facilitated by standing genetic variation. Our project uses an experimental evolution approach to test this prediction and track the dynamics of phenotypic and genetic changes induced by the loss of pollinators. To this end, we will artificially manipulate the abundance and diversity of natural pollinators in a common garden setting using fast cycling Brassica rapa as a model plant.
This experiment will be conducted across six generations, and changes in phenotypic traits and allele frequencies will be measured as a time series. After being raised in a phytotron, plants will be first exposed to natural pollinator communities. Brassica rapa reproduction will be controlled in three different treatments. The abundance and diversity of natural pollinators will be manipulated by varying the opening time of experimental cages allowing the exclusion of pollinators as follows: ten minutes, permanently open, and a control condition (random hand pollination). Only plants firstly visited by pollinators will be recruited to generate the next generation. These steps will be repeated six times during the natural season of pollinator activity (March to October) in years 1 and 2. A total of 1’728 individuals will be tracked (3 treatments * 3 replicates * 32 individuals * 6 generations). For each cycle, phenotypic measurements of plant and flower morphology, as well as scent collection will be taken. Leaf tissue will be collected for DNA extraction and SNP-chip genotyping (Illumina Brassica 60X). Our experiment will provide insights into phenotypic and genetic evolution induced by the loss of pollinators and allow for the fit of population-genetic models to time-series data.
PhD-Student: Elisabeth Authier
Is Dolutegravir-Based Combination Antiretroviral Therapy Evolution Proof in Sub-Saharan Africa
Antiretroviral therapy (ART) of HIV in sub-Saharan Africa (SSA) has saved millions of lives but is jeopardized by exponentially increasing antiretroviral resistance, which has exceeded the WHO thresholds in many countries. Resistance evolution has been facilitated by weaknesses of both the employed drugs and the implementation of ART in SSA. The most promising strategy to address ART-resistance is the planned large-scale introduction of the integrase-inhibitor dolutegravir (DTG) in SSA. Evolution of DTG resistance has proven to be very difficult due to its high efficacy, genetic barrier, and the fitness costs of resistance mutations – though monotherapy trials suggest that resistance is not completely impossible. In resource-rich settings, these favorable properties and the personalized HIV therapy have almost completely prevented DTG resistance. We will address the crucial, currently completely open questions: Are the favorable properties of DTG sufficient to compensate for its imperfect administration in SSA? Which public health measures can prevent DTG-resistance? To this end, we will:
Establish a generic mathematical model for the interaction between the epidemiological dynamics of HIV in SSA and the evolution of resistance on complex fitness/resistance landscapes, i.e. resistance mechanisms involving multiple drugs, resistance, and compensatory mutations.
Provide an evidence synthesis, based on published in vitro and in vivo data, and nucleotide sequences on the fitness and resistance landscape of HIV to DTG-based combination therapy.
By integrating (2) and the rich information from IeDEA (International epidemiology Database to Evaluate AIDS) collaboration into (1), predict how likely the evolution and transmission of DTG resistance is under currently practiced treatment practices, determine the potential public health impact of DTG resistance, and identify the key knowledge gaps preventing more precise estimates. Finally, we will assess to what extent public health interventions can prevent the spread of DTG resistance in SSA.
Thereby, we will not only address one of the most important open questions in HIV medicine upon which the sustainability of ART for over 20 million individuals depends, but also provide a real-world and real-time case study of complex fitness landscapes, and the interaction between evolutionary and epidemiological processes.
PhD-Student: Tom Loosli