School of Biology
DNA sequence response to experimental evolution at different sexual selection conditions
Author(s): Veltsos, P, Cezard, T, Gharbi, K, Snook, RR, Ritchie, MG
We have been maintaining Drosophila pseudoobscura populations in two regimes of high and low sexual selection for 150 generations. Several phenotypes have responded to selection during this limited time, for example courtship behaviour, morphology, fecundity and chemical communication.
We conducted RAD sequencing of 4 biological replicates of each condition and obtained >7000 regions containing SNP variation. We characterise the genes and genomic regions associated with consistent differences in SNPs between the selection regimes. We also comment on the repeatability of the response, with comparisons within each biological replicate.
Department of Biosciences
Environmental variation and the evolution of virulence
Author(s): Anttila, J, Ruokolainen, L, Kaitala, V, Laakso, J
Environmental opportunist pathogens are common in nature. Two well known examples are Vibrio cholera in humans and Flavobacteria in fish. Yet studies on epidemiology and evolution of pathogenicity are centred on obligate pathogens. Environmental opportunist pathogens do not require host-to-host contacts for transmission, and spend most of their time in the outside-host environment where their abundance depends on biotic and abiotic conditions. Such environmental bacteria can spontaneously gain and loose virulence factors that are likely to be associated with increased growth potential and energetic costs. Non-obligate pathogens have not tightly coevolved to evade host immune system, and pathogen transmission is likely to be strongly dose-dependent. This leads to a paradox: in the absence of hosts pathogenicity is selected against due to reduced competitive ability and this in turn prevents transmission to hosts. How can environmental opportunistic pathogens exist? We propose that environmental variation is a plausible mechanism explaining transition to more virulent forms.
We present a model which combines a SIR system to environmental virulent and non-virulent bacterial strains. We assume the pathogenic strain is a fast growing whereas the competing strain is a better competitor. This system is associated with bi-stability between two regimes: virulent strain with infections and non-virulent strain with no infections. Starting from a lower density the virulent strain is unable to grow or cause infections in a stable environment. If both bacterial strains are subjected to environmental variation, the virulent strain can overcome between-strain competition and increase in density close to infective dose and cause sporadic outbreaks or persistent infections, and thus gain significant fitness increases. Infective cycles may in turn promote further evolution from environmental opportunism to obligate pathogenicity, especially in the case of persistent infections.
Evolution of evolvability under fluctuating selection
Author(s): Le Rouzic, A, Alvarez-Castro, JM, Hansen, TF
Empirical evidence suggests that fluctuating selection is a major evolutionary mechanism. The most straightforward consequence of rapid changes of the fitness function is the induced response of the mean phenotype in the population. Yet, repeated back-and-forth evolutionary trajectories are also suspected to affect the genetic architecture underlying the phenotypic characters subject to continuous adaptation. In order to better understand the long-term consequences of fluctuating selection, we modeled the response of complex, multilocus genetic architectures to various natural selection regimes -- stabilizing, directional, and fluctuating. This model accounts for gene-gene interactions (through multilinear epistasis), and thus allows to investigate the dynamics of evolutionary potential at two distinct levels: (i) the standing genetic variation, i.e. the capacity for the population to respond immediately to directional selection, and (ii) the level of canalization (measured as the average effect of new mutations),which reflects the capacity for the population to replenish genetic variation. Both analytical results and individual-based simulations show that fast fluctuations (white noise change in the phenotypic optimum every generation) are essentially similar to stabilizing selection, promoting a degree of genetic canalization and low evolvability. In contrast, when large fluctuations of the phenotypic optimum (beyond the phenotypic range of the population) occur every 10 to 100 generations, equilibrium mutational effects and genetic variance are higher and the population is more evolvable. However, there was no evidence that decanalization and increased evolvability were adaptive, and fluctuating selection remains intrinsically more constraining than genetic drift.
Faculty of Mathematics
Evolutionary rescue in structured populations
Author(s): Uecker, H, Otto, SP, Hermisson, J
As a consequence of environmental deterioration, a population might become maladapted and risk extinction unless it succeeds in adapting to the new conditions. How likely is it that a population escapes extinction through adaptive evolution? Modeling a population in a degrading structured habitat, we analyze the impact of several ecological factors on its survival probability and determine the relative contribution of standing genetic variation and new mutations to evolutionary rescue. We find that in the interplay of various, partially antagonistic effects, the probability of evolutionary rescue can show non-trivial and unexpected dependence on ecological characteristics. The rate of gene flow affects the fate of the population in several ways, resulting in a complex and non-monotonic relationship between migration rate and rescue probability. Counterintuitively, a harsher change or an instantaneous degradation of the total habitat can sometimes lead to a higher survival probability than a less severe or a slowly progressing change.
Host‐parasite coevolution at ecological time scales: bacteria‐phage dynamics within a long‐lived host
Author(s): Koskella, BL
Plants and animals host a diverse microbiota and it is increasingly clear that these dynamic microbial communities significantly affect host phenotype, including resistance to disease. Microbiota act as complex coevolving networks, with high levels of horizontal gene transfer, immigration of new species, selection by the host immune system, and multifaceted antagonistic interactions. A key challenge is to understand how microbial communities are influenced by interactions with their hosts and viral parasites (phages), and how these communities affect the fitness of their hosts. Given their sessile life histories, high risk of infection by microbes, and reliance on commensal microbial species, plant hosts represent an ideal system for testing both fundamental and applied questions in this emerging field. I first determined the relevant ecological structure of phage adaptation in natural microbial communities living within tree hosts and then measured the coevolutionary interaction between phages and bacterial communities within these tree hosts over time. I took an experimental time-shift approach and demonstrate that bacteria are most resistant to phages from the past and least resistant to those from the future. These results provide the first evidence that natural bacterial populations respond rapidly to local phage-mediated selection and that phages play a key role in shaping the microbiota of their eukaryotic hosts.
Evolutionary Dynamics and Biophysics
Observing evolution in action: HIV and its coevolution with the host immune system
Author(s): Neher, RA
The immune system of a HIV infected individual is constantly fighting the virus. In absence of treatment, this battle is ultimately lost in the great majority of cases since the virus changes rapidly to evade the immune pressure. I will show how relevant evolutionary parameters such as typical selection coefficients and recombination rates can estimated from time series data of HIV. Standard population genetic methods are often inappropriate to analyze HIV data, since the viral population experiences strong positive selection at many sites. This implies that the relevant stochastic force in HIV evolution is not genetic drift but selection on linked sites or genetic draft and methods based on the neutral coalescent do not work. I will then discuss recent progress in developing a theory of rapidly adapting populations such as HIV. In models of rapid adaptation, genealogies are different from the standard Kingman coalescent but display frequent multi-merger event. Their statistic is described by the Bolthausen-Sznitman coalescent. These results allow us to calculate levels of neutral diversity, site frequency spectra, and the scale of linkage disequilibrium from parameters of the model.
Centro de Biologia Ambiental/ Departamento de Biologia Animal
Quick adaptation to a new environment erases signature of history in natural populations
Author(s): Fragata, I, Simões, P, Lopes-Cunha, M, Lima, M, Kellen, B, Bárbaro, M, Santos, J, Santos, M, Matos, M
In a world where human activity is changing the climate and habitats at a fast pace it is fundamental to understand how much and how quickly can species adapt. In the last few years we have witnessed the evolutionary response of various species to the effects of global warming. One such case is Drosophila subobscura, which presents clinal variation for body size and inversions, across three continents. Recent evidence shows that temporal changes are occurring in the clinal variation as a response to global warming, with northern populations becoming more similar to southern ones. Both local adaptation and gene flow may be involved, the latter possibly overcoming historical constraints. In this study we propose to measure the contributions of history and selection when populations initially differentiated in Nature are under a similar selective pressure, in order to test if the uniform selection erases prior genetic differences, even in the absence of gene flow. We used as scenario the adaptation to a new, common environment of three populations of Drosophila subobscura initially differentiated along the European latitudinal cline. Quick evolutionary response was observed in all foundations leading to full convergence. All foundations converged to the same adaptive peak, although at different rates and through different paths, suggesting an overall smooth fitness landscape. We concluded that although history had a strong effect during the initial generations, selection quickly overcame it, especially in fitness related traits. The fast loss of differentiation shows that, even in the absence of gene flow adaptation to a common environment can erase the variation observed in nature, a finding that raises concerns in Conservation terms.
School of Biological Sciences
Rapid evolution in experimentally-harvested fish populations
Author(s): Carvalho, GR, Van Wijk, S, Taylor, MI, Creer, S, Dreyer, C, Rodrigues, FM, Ramnarine, IW, Van Oosterhout, C
Rapid contemporary evolution has become more common place in response to human-induced pressures such as introductions and overexploitation. One such increasingly significant pressure arises from size-selective harvesting in commercial fisheries which can induce rapid changes in biological traits. While experimental and wild harvested populations often show clear shifts in body size and maturation associated with fishing, the relative contributions of genetic and environmental factors to these shifts remain uncertain and have been much debated. Observations of so-called fisheries-induced evolution (FIE), to date are based solely on phenotypic measures, such as size data, and no genetic data are available. Here, we quantify genetic versus environmental change in response to size-selective harvesting in guppies (Poecilia reticulata) across three generations of selection for small and large body-sized individuals. We document for the first time significant changes at individual genetic loci, some of which have previously been associated with body size. In contrast, variation at neutral microsatellite markers was unaffected by selection, providing direct genetic evidence for rapid evolution induced by size-selective harvesting. These findings demonstrate FIE in an experimental system, with major implications for the sustainability of harvested populations, as well as impacts on size-structured communities and ecosystem processes. Such findings prompt scientists and managers to reconsider the capacity of harvested stocks to adapt to, and recover from, harvesting and predation.
Rapid evolution of insecticide resistance genes in mosquito populations: a quantitative approach
Author(s): Milesi, P, Labbe, P, Weill, M, Lenormand, T
How adaptation appears and is later refined by natural selection has been the object of intense theoretical work. However, the testing of these theories is limited by our ability to estimate the strength of natural selection in nature. Selection by insecticide treatments of the Ester resistance gene in populations of the mosquito Culex pipiens in Montpellier area (southern France) is a contemporary example of rapid evolution of an adaptation to environmental changes in natural populations. We observed in twenty years, the emergence and replacement of three resistance alleles at the Ester locus due to the selection pressure of insecticide uses. With continuous sampling for >40 years on a 50 km transect corresponding to a gradient of insecticides treatment, we were able to follow the dynamics of these different alleles in natural populations. Using a population genetics model taking into account environmental variables, we were able to estimate the intensity of the selection pressure, as well as many parameters (genetic and environmental) affecting the dynamics of these alleles. In addition, we were able to analyze the evolution of allele frequencies under various environmental conditions. For example the prohibition by the European instances of OP’s insecticides in 2007 was a full-scale test of the quality of our predictions. I will present this work, which links ecology and evolution, through the integration of environmental data into a genetic model, to analyze the dynamics of adaptive alleles in natural populations. We quantified the selection and environmental parameters conditioning resistance genes dynamic and showed that the relationship between environmental factors and allele dynamics in natural populations is not straightforward.
Rapid evolution of metal tolerance revealed by resurrecting dormant propagules
Author(s): Spaak, P, Turko, P
We investigated the evolutionary response of an ecologically important freshwater crustacean, Daphnia, to a rapidly changing toxin environment. From the 1920s until the 1960s, the use of leaded gasoline caused the aquatic concentration of this toxin to increase at least 5-fold, presumably exerting rapid selective pressure on aquatic organisms to develop resistance. Similarly, we predicted that the banning of leaded gasoline, and the corresponding return to low lead levels, would result in the loss of resistance if this resistance carried a cost. These questions were addressed directly by using the resurrection ecology approach, whereby dormant propagules from a focal time period are hatched or germinated and compared to those from another time period. We hatched several Daphnia genotypes from each of two Swiss lakes, during times of higher (1960s / 1980s) and lower (2000s) lead stress, and compared their life histories under different laboratory levels of this stressor. The results were clear: modern Daphnia genotypes (hatched from a period of low lead pollution) had significantly reduced fitness, measured as the population growth rate (r), when exposed to lead, while those genotypes hatched from times of high lead pollution did not display this reduction. We conclude that Daphnia in these lakes were able to rapidly adapt to increasing lead concentrations, and just as rapidly lost this adaptation when the stressor was removed. These results are analyzed in context of the recent theoretical advances by Ellner et al. (2011), who developed a framework for the explicit comparison of ecological and evolutionary rates, and are used to demonstrate a previously unrecognized feature of their model. We also discuss avenues of further research into the genetic underpinnings of this adaptation, taking into account recent research on ABC transporters and technological advances in sequencing technology.