The genetic basis of parallel and divergent phenotypic responses in evolving populations of Escherichia coli

Pleiotropy plays a central role in theories of adaptation, but little is known about the distribution of pleiotropic effects associated with different adaptive mutations. Previously, we described the phenotypic effects of a collection of independently arising beneficial mutations in Escherichia coli. We quantified their fitness effects in the glucose environment in which they evolved and their pleiotropic effects in five novel resource environments. Here we use a candidate gene approach to associate the phenotypic effects of the mutations with the underlying genetic changes. Among our collection of 27 adaptive mutants, we identified a total of 21 mutations (18 of which were unique) encompassing five different loci or gene regions. There was limited resolution to distinguish among loci based on their fitness effects in the glucose environment, demonstrating widespread parallelism in the direct response to selection. However, substantial heterogeneity in mutant effects was revealed when we examined their pleiotropic effects on fitness in the five novel environments. Substitutions in the same locus clustered together phenotypically, indicating concordance between molecular and phenotypic measures of divergence.     

Pleiotropic effects of beneficial mutations in Escherichia coli

Micromutational models of adaptation have placed considerable weight on antagonistic pleiotropy as a mechanism that prevents mutations of large effect from achieving fixation. However, there are few empirical studies of the distribution of pleiotropic effects, and no studies that have examined this distribution for a large number of adaptive mutations. Here we examine the form and extent of pleiotropy associated with beneficial mutations in Escherichia coli. To do so, we used a collection of independently evolved genotypes, each of which contains a beneficial mutation that confers increased fitness in a glucose-limited environment. To determine the pleiotropic effects of these mutations, we examined the fitnesses of the mutants in five novel resource environments. Our results show that the majority of mutations had significant fitness effects in alternative resources, such that pleiotropy was common. The predominant form of this pleiotropy was positive--that is, most mutations that conferred increased fitness in glucose also conferred increased fitness in novel resources. We did detect some deleterious pleiotropic effects, but they were primarily limited to one of the five resources, and within this resource, to only a subset of mutants. Although pleiotropic effects were generally positive, fitness levels were lower and more variable on resources that differed most in their mechanisms of uptake and catabolism from that of glucose. Positive pleiotropic effects were strongly correlated in magnitude with their direct effects, but no such correlation was found among mutants with deleterious pleiotropic effects. Whereas previous studies of populations evolved on glucose for longer periods of time showed consistent declines on some of the resources used here, our results suggest that deleterious pleiotropic effects were limited to only a subset of the beneficial mutations available.     

Low concordance of short‐term and long‐term selection responses in experimental Drosophila populations

Experimental evolution is becoming a popular approach to study the genomic selection response of evolving populations. Computer simulation studies suggest that the accuracy of the signature increases with the duration of the experiment. Since some assumptions of the computer simulations may be violated, it is important to scrutinize the influence of the experimental duration with real data. Here, we use a highly replicated Evolve and Resequence study in Drosophila simulans to compare the selection targets inferred at different time points. At each time point, approximately the same number of SNPs deviates from neutral expectations, but only 10% of the selected haplotype blocks identified from the full data set can be detected after 20 generations. Those haplotype blocks that emerge already after 20 generations differ from the others by being strongly selected at the beginning of the experiment and display a more parallel selection response. Consistent with previous computer simulations, our results demonstrate that only Evolve and Resequence experiments with a sufficient number of generations can characterize complex adaptive architectures.     

Correlated responses to clonal selection in populations of Daphnia pulicaria: mechanisms of genetic correlation and the creative power of sex

Genetic correlations among traits alter evolutionary trajectories due to indirect selection. Pleiotropy, chance linkage, and selection can all lead to genetic correlations, but have different consequences for phenotypic evolution. We sought to assess the mechanisms contributing to correlations with size at maturity in the cyclic parthenogen Daphnia pulicaria. We selected on size in each of four populations that differ in the frequency of sex, and evaluated correlated responses in a life table. Size at advanced adulthood, reproductive output, and adult growth rate clearly showed greater responses in high‐sex populations, with a similar pattern in neonate size and r. This pattern is expected only when trait correlations are favored by selection and the frequency of sex favors the creation and demographic expansion of highly fit clones. Juvenile growth and age at maturity did not diverge consistently. The inter‐clutch interval appeared to respond more strongly in low‐sex populations, but this was not statistically significant. Our data support the hypothesis that correlated selection is the strongest driver of genetic correlations, and suggest that in organisms with both sexual and asexual reproduction, adaptation can be enhanced by recombination.     

Resource competition between genetically varied and genetically uniform populations of Daphnia pulex (Leydig): does sexual reproduction confer a short‐term ecological advantage?

Small competitive advantages may suffice to compensate for a large disadvantage in intrinsic growth capacity. This well‐known principle from ecology has recently been applied to the enduring question of how sexual reproduction can persist in the face of invasion by female‐only parthenogens. Small competitive advantages resulting directly from sexual reproduction are predicted to cancel a two‐fold disadvantage in intrinsic growth capacity caused by males (which do not themselves produce offspring) comprising half the sexual population. In this paper we test the principal assumption of this theory, that the genetic variation produced by sexual reproduction confers a competitive advantage over self‐identical asexual invaders. We set up competition between a diverse clonal assembly of Daphnia pulex and genetically uniform populations from single clones. At young ages, the population comprising genetically varied Daphnia had significantly higher birth rates in competition with populations of genetically uniform Daphnia than in competition with itself, indicating competitive release and a Lotka–Volterra competition coefficient α₁₂ < 1. No such difference was apparent under conditions of greater food stress, possibly due to individuals channelling more energy into survival, or for old‐aged populations, possibly as a result of reduced selective pressures for high reproduction in old females. Mean birth rates differed between the clones at all ages in the presence of competition, providing evidence of variation in life history traits between clones. A Lotka–Volterra model predicted empirical estimates of α₁₂ = 0.896 (genetically uniform on varied) and α₂₁ = 1.010 (varied on uniform), which permits immediate coexistence of a sexual population of D. pulex even with an asexual lineage having twice the intrinsic growth capacity. © 2005 The Linnean Society of London, Biological Journal of the Linnean Society, 2005, 85 , 111–123.     

Adaptive mgl-regulatory mutations and genetic diversity evolving in glucose-limited Escherichia coli populations

The mutational adaptation of E. coli to low glucose concentrations was studied in chemostats over 280 generations of growth. All members of six independent populations acquired increased fitness through the acquisition of mutations at the mgl locus, increasing the binding protein-dependent transport of glucose. These mutations provided a strong fitness advantage (up to 10-fold increase in glucose affinity) and were present in most isolates after 140 generations. mgl constitutivity in some isolates was caused by base substitution, short duplication, small deletion and IS1 insertion in the 1041 bp gene encoding the repressor of the mgl system, mglD (galS). But an unexpectedly large proportion of mutations were located in the short mgl operator sequence (mglO), and the majority of mutations were in mglO after 280 generations of selection. The adaptive mglO substitutions in several independent populations were at exactly the positions conserved in the two 8 bp half-sites of the mgl operator, with the nature of the base changes also completely symmetrical. Either mutations were directed to the operator or the particular operator mutations had a selective advantage under glucose limitation. Indeed, isolates carrying mglO mutations showed greater rates of transport for glucose and galactose at low concentrations than those carrying mglD null mutations. mglO mutations avoid cross-talk by members of the GalR-Lacl repressor family, reducing transporter expression and providing a competitive advantage in a glucose-limited environment. Another interesting aspect of these results was that each adapted population acquired multiple mgl alleles, with several populations containing at least six different mgl-regulatory mutations co-existing after 200 generations. The diversity of mutations in the mglO/mglD region, generally in combination with mutations at other loci regulating glucose uptake (malT, mlc, ptsG), provided evidence for multiple clones in each population. Increased fitness was accompanied by the generation of genetic diversity and not the evolution of a single winner clone, as predicted by the periodic selection model of bacterial populations.     

Hidden genetic variance contributes to increase the short‐term adaptive potential of selfing populations

Standing genetic variation is considered a major contributor to the adaptive potential of species. The low heritable genetic variation observed in self‐fertilizing populations has led to the hypothesis that species with this mating system would be less likely to adapt. However, a non‐negligible amount of cryptic genetic variation for polygenic traits, accumulated through negative linkage disequilibrium, could prove to be an important source of standing variation in self‐fertilizing species. To test this hypothesis, we simulated populations under stabilizing selection subjected to an environmental change. We demonstrate that, when the mutation rate is high (but realistic), selfing populations are better able to store genetic variance than outcrossing populations through genetic associations, notably due to the reduced effective recombination rate associated with predominant selfing. Following an environmental shift, this diversity can be partially remobilized, which increases the additive variance and adaptive potential of predominantly (but not completely) selfing populations. In such conditions, despite initially lower observed genetic variance, selfing populations adapt as readily as outcrossing ones within a few generations. For low mutation rates, purifying selection impedes the storage of diversity through genetic associations, in which case, as previously predicted, the lower genetic variance of selfing populations results in lower adaptability compared to their outcrossing counterparts. The population size and the mutation rate are the main parameters to consider, as they are the best predictors of the amount of stored diversity in selfing populations. Our results and their impact on our knowledge of adaptation under high selfing rates are discussed.     

Niche overlap in allotopic and syntopic populations of sexually interacting ground‐hopper species

Abstract There is accumulating evidence that sexual interactions among species (reproductive interference) could have dramatic effects for species’ coexistence. It has been shown that the fitness of individuals can be substantially reduced as a consequence of reproductive interference. This might subsequently lead to displacement of a species (sexual exclusion). On the other hand, some evolutionary and ecological mechanisms might enable species to coexist, such as the divergence of mate recognition systems (reproductive character displacement), habitat partitioning, clumped dispersion patterns or different colonization capabilities. We have previously shown that the two ground‐hopper species Tetrix subulata and Tetrix ceperoi interact sexually in the laboratory as well as in the field. At sites where both species co‐occur niche overlap was high, suggesting that coexistence is maintained by different niche breadths rather than by habitat partitioning. To test the hypothesis that habitat partitioning does not contribute to species’ coexistence, we examined whether allotopic and syntopic populations of these two species differ in niche overlap (competitive release). Our results show that niche overlap is higher in syntopic than in allotopic populations, suggesting that the site‐specific habitat structure (heterogeneity) has a stronger influence on microhabitat utilization than the presence of heterospecifics. Hence, our data do not support the hypothesis that habitat partitioning plays a substantial role for the coexistence of these sexually interacting species.     

Leakiness of genetic markers and susceptibility to post-plating mutagenesis inEscherichia coli

WhenEscherichia coli strain AB1157 is subjected to starvation for threonine or leucine on solid media, threonine-independent or leucine-independent colonies continue to emerge for several days after plating. This process is strongly streptomycin dependent. Under identical conditions arginine-independent colonies do not arise when arginine starvation is imposed. Since thethr1 andleuB6 alleles of AB1157 could be classified as ‘leaky’ while theargE3 allele cannot be so classified, there seems to be a correlation between leakiness of mutant genetic markers and post-plating mutagenesis which counters the effect of the mutations. Some of the threonine-independent variants acquired the ability to increase the leakiness of otherwise nonleaky markers such asargE3 and permit development of arginine independence in arecA-dependent,lexA-independent manner. I show that these variants harbour a mutation, tentatively namedadi (adaptation inducer), at around 72 min on the genetic map, and that theadi mutation increases the intrinsic leakiness of alacZ (ochre) mutation, perhaps by enhanced translational error. These observations are discussed in relation to the phenomenon of ‘adaptive’ mutagenesis, its possible mechanism, and its specificity.     

Antagonism between sexual and natural selection in experimental populations of Saccharomyces cerevisiae

Trade-offs between life-history components are a central concept of evolution and ecology. Sexual and natural selection seem particularly apt to impose antagonistic selective pressures. When sex is not integrated into reproduction, as in Saccharomyces cerevisiae, natural selection can impair or even eliminate it. In this study, a genetic trade-off between the sexual and asexual phases of the yeast life cycle was suggested by sharp declines in the mating and sporulation abilities of unrelated genotypes that were propagated asexually in minimal growth medium and in mice. When sexual selection was applied to populations that had previously evolved asexually, sexual fitness increased but asexual fitness declined. No such negative correlation was observed when sexual selection was applied to an ancestral strain: sexual and asexual fitness both increased. Thus, evolutionary history affected the evolution of genetic correlations, as fitness increases in a population already well adapted to the environment were more likely to come at the expense of sexual functions.     

Interim analysis incorporating short‐ and long‐term binary endpoints

Designs incorporating more than one endpoint have become popular in drug development. One of such designs allows for incorporation of short‐term information in an interim analysis if the long‐term primary endpoint has not been yet observed for some of the patients. At first we consider a two‐stage design with binary endpoints allowing for futility stopping only based on conditional power under both fixed and observed effects. Design characteristics of three estimators: using primary long‐term endpoint only, short‐term endpoint only, and combining data from both are compared. For each approach, equivalent cut‐off point values for fixed and observed effect conditional power calculations can be derived resulting in the same overall power. While in trials stopping for futility the type I error rate cannot get inflated (it usually decreases), there is loss of power. In this study, we consider different scenarios, including different thresholds for conditional power, different amount of information available at the interim, different correlations and probabilities of success. We further extend the methods to adaptive designs with unblinded sample size reassessments based on conditional power with inverse normal method as the combination function. Two different futility stopping rules are considered: one based on the conditional power, and one from P‐values based on Z‐statistics of the estimators. Average sample size, probability to stop for futility and overall power of the trial are compared and the influence of the choice of weights is investigated.     

Genotype‐by‐environment interactions due to antibiotic resistance and adaptation in Escherichia coli

Mutations that are beneficial in one environment can have different fitness effects in other environments. In the context of antibiotic resistance, the resulting genotype‐by‐environment interactions potentially make selection on resistance unpredictable in heterogeneous environments. Furthermore, resistant bacteria frequently fix additional mutations during evolution in the absence of antibiotics. How do these two types of mutations interact to determine the bacterial phenotype across different environments? To address this, I used Escherichia coli as a model system, measuring the effects of nine different rifampicin resistance mutations on bacterial growth in 31 antibiotic‐free environments. I did this both before and after approximately 200 generations of experimental evolution in antibiotic‐free conditions (LB medium), and did the same for the antibiotic‐sensitive wild type after adaptation to the same environment. The following results were observed: (i) bacteria with and without costly resistance mutations adapted to experimental conditions and reached similar levels of competitive fitness; (ii) rifampicin resistance mutations and adaptation to LB both indirectly altered growth in other environments; and (iii) resistant‐evolved genotypes were more phenotypically different from the ancestor and from each other than resistant‐nonevolved and sensitive‐evolved genotypes. This suggests genotype‐by‐environment interactions generated by antibiotic resistance mutations, observed previously in short‐term experiments, are more pronounced after adaptation to other types of environmental variation, making it difficult to predict long‐term selection on resistance mutations from fitness effects in a single environment.     

Phenological responses of Icelandic subarctic grasslands to short‐term and long‐term natural soil warming

The phenology of vegetation, particularly the length of the growing season (LOS; i.e., the period from greenup to senescence), is highly sensitive to climate change, which could imply potent feedbacks to the climate system, for example, by altering the ecosystem carbon (C) balance. In recent decades, the largest extensions of LOS have been reported at high northern latitudes, but further warming‐induced LOS extensions may be constrained by too short photoperiod or unfulfilled chilling requirements. Here, we studied subarctic grasslands, which cover a vast area and contain large C stocks, but for which LOS changes under further warming are highly uncertain. We measured LOS extensions of Icelandic subarctic grasslands along natural geothermal soil warming gradients of different age (short term, where the measurements started after 5 years of warming and long term, i.e., warmed since ≥50 years) using ground‐level measurements of normalized difference vegetation index. We found that LOS linearly extended with on average 2.1 days per °C soil warming up to the highest soil warming levels (ca. +10°C) and that LOS had the potential to extend at least 1 month. This indicates that the warming impact on LOS in these subarctic grasslands will likely not saturate in the near future. A similar response to short‐ and long‐term warming indicated a strong physiological control of the phenological response of the subarctic grasslands to warming and suggested that genetic adaptations and community changes were likely of minor importance. We conclude that the warming‐driven extension of the LOSs of these subarctic grasslands did not saturate up to +10°C warming, and hence that growing seasons of high‐latitude grasslands are likely to continue lengthening with future warming (unless genetic adaptations or species shifts do occur). This persistence of the warming‐induced extension of LOS has important implications for the C‐sink potential of subarctic grasslands under climate change.     

Soil microbiome responses to the short‐term effects of Amazonian deforestation

Slash‐and‐burn clearing of forest typically results in increase in soil nutrient availability. However, the impact of these nutrients on the soil microbiome is not known. Using next generation sequencing of 16S rRNA gene and shotgun metagenomic DNA, we compared the structure and the potential functions of bacterial community in forest soils to deforested soils in the Amazon region and related the differences to soil chemical factors. Deforestation decreased soil organic matter content and factors linked to soil acidity and raised soil pH, base saturation and exchangeable bases. Concomitant to expected changes in soil chemical factors, we observed an increase in the alpha diversity of the bacterial microbiota and relative abundances of putative copiotrophic bacteria such as Actinomycetales and a decrease in the relative abundances of bacterial taxa such as Chlamydiae, Planctomycetes and Verrucomicrobia in the deforested soils. We did not observe an increase in genes related to microbial nutrient metabolism in deforested soils. However, we did observe changes in community functions such as increases in DNA repair, protein processing, modification, degradation and folding functions, and these functions might reflect adaptation to changes in soil characteristics due to forest clear‐cutting and burning. In addition, there were changes in the composition of the bacterial groups associated with metabolism‐related functions. Co‐occurrence microbial network analysis identified distinct phylogenetic patterns for forest and deforested soils and suggested relationships between Planctomycetes and aluminium content, and Actinobacteria and nitrogen sources in Amazon soils. The results support taxonomic and functional adaptations in the soil bacterial community following deforestation. We hypothesize that these microbial adaptations may serve as a buffer to drastic changes in soil fertility after slash‐and‐burning deforestation in the Amazon region.     

Species and community responses to short‐term climate manipulation: Microarthropods in the sub‐Antarctic

Abstract: Both species and community‐level investigations are important for understanding the biotic impacts of climate change, because current evidence suggests that individual species responses are idiosyncratic. However, few studies of climate change impacts have been conducted on entire terrestrial arthropod communities living in the same habitat in the southern Hemisphere, and the effects of precipitation changes on them are particularly poorly understood. Here we investigate the species‐ and community‐level responses of microarthropods inhabiting a keystone plant species, on sub‐Antarctic Marion Island, to experimental reduction in precipitation, warming and shading. These climate manipulations were chosen based on observed climate trends and predicted indirect climate change impacts on this system. The dry‐warm and shade inducing treatments that were imposed effected significant species‐ and community‐level responses after a single year. Although the strongest community‐level trends included a dramatic decline in springtail abundance and total biomass under the dry‐warm and shade treatments, species responses were generally individualistic, that is, springtails responded differently to mites, and particular mite and springtail species responded differently to each other. Our results therefore provide additional support for the dynamic rather than static model for community responses to climate change, in the first such experiment in the sub‐Antarctic. In conclusion, these results show that an ongoing decline in precipitation and increase in temperature is likely to have dramatic direct and indirect effects on this microarthropod community. Moreover, they indicate that while at a broad scale it may be possible to make generalizations regarding species responses to climate change, these generalizations are unlikely to translate into predictable effects at the community level.     

Constraints on adaptation of Escherichia coli to mixed‐resource environments increase over time

Can a population evolved in two resources reach the same fitness in both as specialist populations evolved in each of the individual resources? This question is central to theories of ecological specialization, the maintenance of genetic variation, and sympatric speciation, yet relatively few experiments have examined costs of generalism over long‐term adaptation. We tested whether selection in environments containing two resources limits a population's ability to adapt to the individual resources by comparing the fitness of replicate Escherichia coli populations evolved for 6000 generations in the presence of glucose or lactose alone (specialists), or in varying presentations of glucose and lactose together (generalists). We found that all populations had significant fitness increases in both resources, though the magnitude and rate of these increases differed. For the first 4000 generations, most generalist populations increased in fitness as quickly in the individual resources as the corresponding specialist populations. From 5000 generations, however, a widespread cost of adaptation affected all generalists, indicating a growing constraint on their abilities to adapt to two resources simultaneously. Our results indicate that costs of generalism are prevalent, but may influence evolutionary trajectories only after a period of cost‐free adaptation.     

Changes in antioxidants are critical in determining cell responses to short‐ and long‐term heat stress

Heat stress can have deleterious effects on plant growth by impairing several physiological processes. Plants have several defense mechanisms that enable them to cope with high temperatures. The synthesis and accumulation of heat shock proteins (HSPs), as well as the maintenance of an opportune redox balance play key roles in conferring thermotolerance to plants. In this study changes in redox parameters, the activity and/or expression of reactive oxygen species (ROS) scavenging enzymes and the expression of two HSPs were studied in tobacco Bright Yellow‐2 (TBY‐2) cells subjected to moderate short‐term heat stress (SHS) and long‐term heat stress (LHS). The results indicate that TBY‐2 cells subjected to SHS suddenly and transiently enhance antioxidant systems, thus maintaining redox homeostasis and avoiding oxidative damage. The simultaneous increase in HSPs overcomes the SHS and maintains the metabolic functionality of cells. In contrast the exposure of cells to LHS significantly reduces cell growth and increases cell death. In the first phase of LHS, cells enhance antioxidant systems to prevent the formation of an oxidizing environment. Under prolonged heat stress, the antioxidant systems, and particularly the enzymatic ones, are inactivated. As a consequence, an increase in H₂O₂, lipid peroxidation and protein oxidation occurs. This establishment of oxidative stress could be responsible for the increased cell death. The rescue of cell growth and cell viability, observed when TBY‐2 cells were pretreated with galactone‐γ‐lactone, the last precursor of ascorbate, and glutathione before exposure to LHS, highlights the crucial role of antioxidants in the acquisition of basal thermotolerance.     

The shaping of genetic variation in edge‐of‐range populations under past and future climate change

With rates of climate change exceeding the rate at which many species are able to shift their range or adapt, it is important to understand how future changes are likely to affect biodiversity at all levels of organisation. Understanding past responses and extent of niche conservatism in climatic tolerance can help predict future consequences. We use an integrated approach to determine the genetic consequences of past and future climate changes on a bat species, Plecotus austriacus. Glacial refugia predicted by palaeo‐modelling match those identified from analyses of extant genetic diversity and model‐based inference of demographic history. Former refugial populations currently contain disproportionately high genetic diversity, but niche conservatism, shifts in suitable areas and barriers to migration mean that these hotspots of genetic diversity are under threat from future climate change. Evidence of population decline despite recent northward migration highlights the need to conserve leading‐edge populations for spearheading future range shifts.