The Valley-of-Death: reciprocal sign epistasis constrains adaptive trajectories in a constant, nutrient limiting environment

Highly parallelized laboratory evolution of wine yeasts for enhanced metabolic phenotypes

Adaptive Laboratory Evolution (ALE) of microorganisms can improve the efficiency of sustainable industrial processes important to the global economy. However, stochasticity and genetic background effects often lead to suboptimal outcomes during laboratory evolution. Here we report an ALE platform to circumvent these shortcomings through parallelized clonal evolution at an unprecedented scale. Using this platform, we evolved 104 yeast populations in parallel from many strains for eight desired wine fermentation-related traits. Expansions of both ALE replicates and lineage numbers broadened the evolutionary search spectrum leading to improved wine yeasts unencumbered by unwanted side effects. At the genomic level, evolutionary gains in metabolic characteristics often coincided with distinct chromosome amplifications and the emergence of side-effect syndromes that were characteristic of each selection niche. Several high-performing ALE strains exhibited desired wine fermentation kinetics when tested in larger liquid cultures, supporting their suitability for application. More broadly, our high-throughput ALE platform opens opportunities for rapid optimization of microbes which otherwise could take many years to accomplish.

Have you tried turning it off and on again? Oscillating selection to enhance fitness-landscape traversal in adaptive laboratory evolution experiments

Adaptive Laboratory Evolution (ALE) is a powerful tool for engineering and understanding microbial physiology. ALE relies on the selection and enrichment of mutations that enable survival or faster growth under a selective condition imposed by the experimental setup. Phenotypic fitness landscapes are often underpinned by complex genotypes involving multiple genes, with combinatorial positive and negative effects on fitness. Such genotype relationships result in mutational fitness landscapes with multiple local fitness maxima and valleys. Traversing local maxima to find a global maximum often requires an individual or sub-population of cells to traverse fitness valleys. Traversing involves gaining mutations that are not adaptive for a given local maximum but are necessary to 'peak shift' to another local maximum, or eventually a global maximum. Despite these relatively well understood evolutionary principles, and the combinatorial genotypes that underlie most metabolic phenotypes, the majority of applied ALE experiments are conducted using constant selection pressures. The use of constant pressure can result in populations becoming trapped within local maxima, and often precludes the attainment of optimum phenotypes associated with global maxima. Here, we argue that oscillating selection pressures is an easily accessible mechanism for traversing fitness landscapes in ALE experiments, and provide theoretical and practical frameworks for implementation.

Changes in the distribution of fitness effects and adaptive mutational spectra following a single first step towards adaptation

Historical contingency and diminishing returns epistasis have been typically studied for relatively divergent genotypes and/or over long evolutionary timescales. Here, we use Saccharomyces cerevisiae to study the extent of diminishing returns and the changes in the adaptive mutational spectra following a single first adaptive mutational step. We further evolve three clones that arose under identical conditions from a common ancestor. We follow their evolutionary dynamics by lineage tracking and determine adaptive outcomes using fitness assays and whole genome sequencing. We find that diminishing returns manifests as smaller fitness gains during the 2^nd step of adaptation compared to the 1^st step, mainly due to a compressed distribution of fitness effects. We also find that the beneficial mutational spectra for the 2^nd adaptive step are contingent on the 1^st step, as we see both shared and diverging adaptive strategies. Finally, we find that adaptive loss-of-function mutations, such as nonsense and frameshift mutations, are less common in the second step of adaptation than in the first step.

Analyses of both natural and experimental evolution suggest that adaptation depends on the evolutionary past and adaptive potential decreases over time. Here, by tracking yeast adaptation with DNA barcoding, the authors show that such evolutionary phenomena can be observed even after a single adaptive step.

Ensemble epistasis: thermodynamic origins of nonadditivity between mutations

Epistasis-when mutations combine nonadditively-is a profoundly important aspect of biology. It is often difficult to understand its mechanistic origins. Here, we show that epistasis can arise from the thermodynamic ensemble, or the set of interchanging conformations a protein adopts. Ensemble epistasis occurs because mutations can have different effects on different conformations of the same protein, leading to nonadditive effects on its average, observable properties. Using a simple analytical model, we found that ensemble epistasis arises when two conditions are met: (1) a protein populates at least three conformations and (2) mutations have differential effects on at least two conformations. To explore the relative magnitude of ensemble epistasis, we performed a virtual deep-mutational scan of the allosteric Ca2+ signaling protein S100A4. We found that 47% of mutation pairs exhibited ensemble epistasis with a magnitude on the order of thermal fluctuations. We observed many forms of epistasis: magnitude, sign, and reciprocal sign epistasis. The same mutation pair could even exhibit different forms of epistasis under different environmental conditions. The ubiquity of thermodynamic ensembles in biology and the pervasiveness of ensemble epistasis in our dataset suggests that it may be a common mechanism of epistasis in proteins and other macromolecules.

Genomic Background Governs Opposing Responses to Nalidixic Acid upon Megaplasmid Acquisition in Pseudomonas

Horizontal gene transfer is a significant driver of evolutionary dynamics across microbial populations. Although the benefits of the acquisition of new genetic material are often quite clear, experiments across systems have demonstrated that gene transfer events can cause significant phenotypic changes and entail fitness costs in a way that is dependent on the genomic and environmental context. Here, we test for the generality of one previously identified cost, sensitization of cells to the antibiotic nalidixic acid after acquisition of an ∼1-Mb megaplasmid, across Pseudomonas strains and species. Overall, we find that the presence of this megaplasmid sensitizes many different Pseudomonas strains to nalidixic acid but that this same horizontal gene transfer event increases resistance of Pseudomonas putida KT2440 to nalidixic acid across assays as well as to ciprofloxacin under competitive conditions. These phenotypic results are not easily explained away as secondary consequences of overall fitness effects and appear to occur independently of another cost associated with this megaplasmid, sensitization to higher temperatures. Lastly, we draw parallels between these reported results and the phenomenon of sign epistasis for de novo mutations and explore how context dependence of effects of plasmid acquisition could impact overall evolutionary dynamics and the evolution of antimicrobial resistance.IMPORTANCE Numerous studies have demonstrated that gene transfer events (e.g., plasmid acquisition) can entail a variety of costs that arise as by-products of the incorporation of foreign DNA into established physiological and genetic systems. These costs can be ameliorated through evolutionary time by the occurrence of compensatory mutations, which stabilize the presence of a horizontally transferred region within the genome but which also may skew future adaptive possibilities for these lineages. Here, we demonstrate another possible outcome, that phenotypic changes arising as a consequence of the same horizontal gene transfer (HGT) event are costly to some strains but may actually be beneficial in other genomic backgrounds under the right conditions. These results provide a new viewpoint for considering conditions that promote plasmid maintenance and highlight the influence of genomic and environmental contexts when considering amelioration of fitness costs after HGT events.

Every which way? On predicting tumor evolution using cancer progression models

Successful prediction of the likely paths of tumor progression is valuable for diagnostic, prognostic, and treatment purposes. Cancer progression models (CPMs) use cross-sectional samples to identify restrictions in the order of accumulation of driver mutations and thus CPMs encode the paths of tumor progression. Here we analyze the performance of four CPMs to examine whether they can be used to predict the true distribution of paths of tumor progression and to estimate evolutionary unpredictability. Employing simulations we show that if fitness landscapes are single peaked (have a single fitness maximum) there is good agreement between true and predicted distributions of paths of tumor progression when sample sizes are large, but performance is poor with the currently common much smaller sample sizes. Under multi-peaked fitness landscapes (i.e., those with multiple fitness maxima), performance is poor and improves only slightly with sample size. In all cases, detection regime (when tumors are sampled) is a key determinant of performance. Estimates of evolutionary unpredictability from the best performing CPM, among the four examined, tend to overestimate the true unpredictability and the bias is affected by detection regime; CPMs could be useful for estimating upper bounds to the true evolutionary unpredictability. Analysis of twenty-two cancer data sets shows low evolutionary unpredictability for several of the data sets. But most of the predictions of paths of tumor progression are very unreliable, and unreliability increases with the number of features analyzed. Our results indicate that CPMs could be valuable tools for predicting cancer progression but that, currently, obtaining useful predictions of paths of tumor progression from CPMs is dubious, and emphasize the need for methodological work that can account for the probably multi-peaked fitness landscapes in cancer.

Cancer progression models and fitness landscapes: a many-to-many relationship

Motivation

The identification of constraints, due to gene interactions, in the order of accumulation of mutations during cancer progression can allow us to single out therapeutic targets. Cancer progression models (CPMs) use genotype frequency data from cross-sectional samples to identify these constraints, and return Directed Acyclic Graphs (DAGs) of restrictions where arrows indicate dependencies or constraints. On the other hand, fitness landscapes, which map genotypes to fitness, contain all possible paths of tumor progression. Thus, we expect a correspondence between DAGs from CPMs and the fitness landscapes where evolution happened. But many fitness landscapes-e.g. those with reciprocal sign epistasis-cannot be represented by CPMs.

Results

Using simulated data under 500 fitness landscapes, I show that CPMs' performance (prediction of genotypes that can exist) degrades with reciprocal sign epistasis. There is large variability in the DAGs inferred from each landscape, which is also affected by mutation rate, detection regime and fitness landscape features, in ways that depend on CPM method. Using three cancer datasets, I show that these problems strongly affect the analysis of empirical data: fitness landscapes that are widely different from each other produce data similar to the empirically observed ones and lead to DAGs that infer very different restrictions. Because reciprocal sign epistasis can be common in cancer, these results question the use and interpretation of CPMs.

Availability and implementation

Code available from Supplementary Material.

Contact

ramon.diaz@iib.uam.es.

Supplementary information

Supplementary data are available at Bioinformatics online.

Adaptive Roles of SSY1 and SIR3 During Cycles of Growth and Starvation in Saccharomyces cerevisiae Populations Enriched for Quiescent or Nonquiescent Cells

Over its evolutionary history, Saccharomyces cerevisiae has evolved to be well-adapted to fluctuating nutrient availability. In the presence of sufficient nutrients, yeast cells continue to proliferate, but upon starvation haploid yeast cells enter stationary phase and differentiate into nonquiescent (NQ) and quiescent (Q) cells. Q cells survive stress better than NQ cells and show greater viability when nutrient-rich conditions are restored. To investigate the genes that may be involved in the differentiation of Q and NQ cells, we serially propagated yeast populations that were enriched for either only Q or only NQ cell types over many repeated growth–starvation cycles. After 30 cycles (equivalent to 300 generations), each enriched population produced a higher proportion of the enriched cell type compared to the starting population, suggestive of adaptive change. We also observed differences in each population’s fitness suggesting possible tradeoffs: clones from NQ lines were better adapted to logarithmic growth, while clones from Q lines were better adapted to starvation. Whole-genome sequencing of clones from Q- and NQ-enriched lines revealed mutations in genes involved in the stress response and survival in limiting nutrients (ECM21, RSP5, MSN1, SIR4, and IRA2) in both Q and NQ lines, but also differences between the two lines: NQ line clones had recurrent independent mutations affecting the Ssy1p-Ptr3p-Ssy5p (SPS) amino acid sensing pathway, while Q line clones had recurrent, independent mutations in SIR3 and FAS1. Our results suggest that both sets of enriched-cell type lines responded to common, as well as distinct, selective pressures.

Quantitative Trait Loci (QTL)-Guided Metabolic Engineering of a Complex Trait

Engineering complex phenotypes for industrial and synthetic biology applications is difficult and often confounds rational design. Bioethanol production from lignocellulosic feedstocks is a complex trait that requires multiple host systems to utilize, detoxify, and metabolize a mixture of sugars and inhibitors present in plant hydrolysates. Here, we demonstrate an integrated approach to discovering and optimizing host factors that impact fitness of Saccharomyces cerevisiae during fermentation of a Miscanthus x giganteus plant hydrolysate. We first used high-resolution Quantitative Trait Loci (QTL) mapping and systematic bulk Reciprocal Hemizygosity Analysis (bRHA) to discover 17 loci that differentiate hydrolysate tolerance between an industrially related (JAY291) and a laboratory (S288C) strain. We then used this data to identify a subset of favorable allelic loci that were most amenable for strain engineering. Guided by this "genetic blueprint", and using a dual-guide Cas9-based method to efficiently perform multikilobase locus replacements, we engineered an S288C-derived strain with superior hydrolysate tolerance than JAY291. Our methods should be generalizable to engineering any complex trait in S. cerevisiae, as well as other organisms.

Variation in virus effects on host plant phenotypes and insect vector behavior: what can it teach us about virus evolution?

Virus infection can elicit changes in host plant cues that mediate vector orientation, feeding, and dispersal. Given the importance of plant cues for vector-mediated virus transmission, it is unlikely that selection is blind to these effects. Indeed, there are many examples of viruses altering plant cues in ways that should enhance transmission. However, there are also examples of viruses inducing transmission-limiting plant phenotypes. These apparently mal-adaptive effects occur when viruses experience host plant environments that also limit infectivity or within-host multiplication. The apparent link between virus effects and pathology argues for consideration of prior evolutionary relationships between viruses and host plants in order to understand how viruses might evolve to manipulate vector behavior via effects on host plant cues.

Heterozygote Advantage Is a Common Outcome of Adaptation in Saccharomyces cerevisiae

Adaptation in diploids is predicted to proceed via mutations that are at least partially dominant in fitness. Recently, we argued that many adaptive mutations might also be commonly overdominant in fitness. Natural (directional) selection acting on overdominant mutations should drive them into the population but then, instead of bringing them to fixation, should maintain them as balanced polymorphisms via heterozygote advantage. If true, this would make adaptive evolution in sexual diploids differ drastically from that of haploids. The validity of this prediction has not yet been tested experimentally. Here, we performed four replicate evolutionary experiments with diploid yeast populations (Saccharomyces cerevisiae) growing in glucose-limited continuous cultures. We sequenced 24 evolved clones and identified initial adaptive mutations in all four chemostats. The first adaptive mutations in all four chemostats were three copy number variations, all of which proved to be overdominant in fitness. The fact that fitness overdominant mutations were always the first step in independent adaptive walks supports the prediction that heterozygote advantage can arise as a common outcome of directional selection in diploids and demonstrates that overdominance of de novo adaptive mutations in diploids is not rare.

Strong Selection Significantly Increases Epistatic Interactions in the Long-Term Evolution of a Protein

Epistatic interactions between residues determine a protein’s adaptability and shape its evolutionary trajectory. When a protein experiences a changed environment, it is under strong selection to find a peak in the new fitness landscape. It has been shown that strong selection increases epistatic interactions as well as the ruggedness of the fitness landscape, but little is known about how the epistatic interactions change under selection in the long-term evolution of a protein. Here we analyze the evolution of epistasis in the protease of the human immunodeficiency virus type 1 (HIV-1) using protease sequences collected for almost a decade from both treated and untreated patients, to understand how epistasis changes and how those changes impact the long-term evolvability of a protein. We use an information-theoretic proxy for epistasis that quantifies the co-variation between sites, and show that positive information is a necessary (but not sufficient) condition that detects epistasis in most cases. We analyze the “fossils” of the evolutionary trajectories of the protein contained in the sequence data, and show that epistasis continues to enrich under strong selection, but not for proteins whose environment is unchanged. The increase in epistasis compensates for the information loss due to sequence variability brought about by treatment, and facilitates adaptation in the increasingly rugged fitness landscape of treatment. While epistasis is thought to enhance evolvability via valley-crossing early-on in adaptation, it can hinder adaptation later when the landscape has turned rugged. However, we find no evidence that the HIV-1 protease has reached its potential for evolution after 9 years of adapting to a drug environment that itself is constantly changing. We suggest that the mechanism of encoding new information into pairwise interactions is central to protein evolution not just in HIV-1 protease, but for any protein adapting to a changing environment.

Evolution is often viewed as a process that occurs “mutation by mutation”, suggesting that the effect of each mutation is independent of that of others. However, in reality the effect of a mutation often depends on the context of other mutations, a dependence known as “epistasis”. Even though epistasis can constrain protein evolution, it is actually very common. Such interactions are particularly pervasive in proteins that evolve resistance to a drug via mutations that create defects, and that must be repaired with compensatory mutations. We study how epistasis between protein residues evolves over time in a new and changing environment, and compare these findings to protein evolution in a constant environment. We analyze the sequences of the human immunodeficiency virus type 1 (HIV-1) protease enzyme collected over a period of 9 years from patients treated with anti-viral drugs (as well as from patients that went untreated), and find that epistasis between residues continues to increase as more potent anti-viral drugs enter the market, while epistasis is unchanging in the proteins exposed to a constant environment. Yet, the proteins adapting to the changing landscape do not appear to be constrained by the epistatic interactions and continue to manage to evade new drugs.

Evolutionary rescue by compensatory mutations is constrained by genomic and environmental backgrounds

Since deleterious mutations may be rescued by secondary mutations during evolution, compensatory evolution could identify genetic solutions leading to therapeutic targets. Here, we tested this hypothesis and examined whether these solutions would be universal or would need to be adapted to one's genetic and environmental makeups. We performed experimental evolutionary rescue in a yeast disease model for the Wiskott–Aldrich syndrome in two genetic backgrounds and carbon sources. We found that multiple aspects of the evolutionary rescue outcome depend on the genotype, the environment, or a combination thereof. Specifically, the compensatory mutation rate and type, the molecular rescue mechanism, the genetic target, and the associated fitness cost varied across contexts. The course of compensatory evolution is therefore highly contingent on the initial conditions in which the deleterious mutation occurs. In addition, these results reveal biologically favored therapeutic targets for the Wiskott–Aldrich syndrome, including the target of an unrelated clinically approved drug. Our results experimentally illustrate the importance of epistasis and environmental evolutionary constraints that shape the adaptive landscape and evolutionary rate of molecular networks.

Elucidating the molecular architecture of adaptation via evolve and resequence experiments

Evolve and resequence (E&R) experiments use experimental evolution to adapt populations to a novel environment, then next-generation sequencing to analyse genetic changes. They enable molecular evolution to be monitored in real time on a genome-wide scale. Here, we review the field of E&R experiments across diverse systems, ranging from simple non-living RNA to bacteria, yeast and the complex multicellular organism Drosophila melanogaster. We explore how different evolutionary outcomes in these systems are largely consistent with common population genetics principles. Differences in outcomes across systems are largely explained by different starting population sizes, levels of pre-existing genetic variation, recombination rates and adaptive landscapes. We highlight emerging themes and inconsistencies that future experiments must address.