Biology subjects

Mustonen, V.

Publications and source records attributed to Mustonen, V..

At least 19 recordsLinked to original sources

bioRxiv2026

How far can microbial monocultures predict growth in multi-strain communities?

Previous research in microbes successfully predicted biculture growth based on monoculture growth curves. Still, the usual model-based approach does not seem to extend to communities involving more than two bacterial strains. Here, we use a model-blind machine-learning approach to predict community-wide yield, area under the growth curve, maximum relative growth rate and its timing in communities involving up to five strains of Escherichia coli. First, we identify the highest per-capita growth rate and its timing in monocultures as major predictors of multi-strain community growth. Next, we show that a random forest trained on communities involving a low number of strains is able to predict the outcomes of communities involving a higher number of strains. Finally, we reveal diminishing returns in using more and more complex communities in order to predict the behaviour of a higher-level community, because monoculture- and biculture-based growth predictions are often already accurate. This finding relativises the need for experiments involving high numbers of strains when studying the growth dynamics of multi-strain communities.

ecology↗

bioRxiv2026

Eco-evolutionary dynamics and environmental detoxification jointly shape bacterial community response to antibiotic perturbation

Microbial communities frequently encounter recurrent antibiotic disturbance, yet how ecological, evolutionary, and environmental processes jointly shape the response remains unresolved. Here we experimentally disentangle these mechanisms using a 23-species bacterial community exposed to ampicillin pulses with pre-pulse and resistance priming. Pre-pulse priming preconditioned community composition toward resistant taxa, reducing compositional change during the main pulse. Resistance priming was the dominant determinant of community response, buffering compositional change, relaxing selection, and altering gene expression. This buffering arose from both evolution of higher resistance and accelerated ampicillin detoxification by a dominant degrader, which transiently reduced antibiotic effects and favoured non-degrading taxa. Yet buffering did not translate into better recovery: because resistance was coupled to competitive dominance, diversity remained comparable to or below that of ancestral communities, and dominance was reinforced. Together, we show that antibiotic history reshapes microbial disturbance response through coupled eco-evolutionary and environmental feedbacks, generating a trade-off between resistance and recovery.

evolutionary biology↗

bioRxiv2026

Ecological tristability driven by total carbon availability over resource complexity in a synthetic microbial community

Even though complex microbial communities are ubiquitous and provide essential services for natural and human-associated ecosystems, our knowledge about their assembly and dynamics is incomplete. There is an ongoing debate whether the behavior of complex communities can be predicted from the outcome of pairwise competition of species, and whether communities reach alternative stable states depending on the level and complexity of resource provided for growth. To estimate the effect of two resource gradients, total carbon availability and resource complexity, on the compositional dynamics of a complex microbial community, we conducted a 16-day serial passage experiment, transferring a 16-species synthetic community in 96 different resource environments. We observed that although both resource dimensions influenced community composition, total carbon exerted a considerably larger effect. Additionally, we saw the emergence of a tristable pattern along the total carbon gradient, a feature not observed for the resource complexity gradient. Using monoculture assays, we identified lag phase duration as the dominant predictor of competitive success at carbon extremes, with maximum growth rate increasing in importance as lag times converged. Total carbon availability thus structured community state transitions and regulated which growth trait governed competitive sorting. These results suggest the importance of total carbon level over resource complexity and identifying dominant species for the quest to successfully manage, maintain and manipulate complex microbial communities.

ecology↗

bioRxiv2026

In silico prediction of metabolic trait robustness in microbial cells

In industrial applications, microbial strains undergo notable biomass expansion subjecting them to Darwinian selection. The consequent adaptive evolution may threaten the often tediously developed desired traits of the strains, such as flavor or platform chemical production. Yet, it remains unresolved how to predict the evolutionary trait robustness. Here, we propose TRAEV (Trait Robustness Against EVolution), a computational framework for in silico prediction of evolutionary trajectories and robustness of desired metabolic traits in application environments. TRAEV uses constraint-based metabolic model simulations to predict environment-dependent trait-fitness dependencies and Monte Carlo-based perturbation analysis to account for the stochasticity of adaptive evolution. First, TRAEV predicts the immediate phenotypic adaptation to the new environment, and then, the evolutionary trajectories of fitness and desired trait by sampling enzyme usage changes. From the predicted trajectories, trait robustness is quantified as two scores: Robustness Score (RS) and Trade-off Score (TS). RS and TS are the mean of a normalized desired trait and the mean of the product of normalized changes in the desired trait and in fitness, respectively, over intermediate metabolic states along the evolutionary trajectory. We validated TRAEV by demonstrating that it predicted the relative robustness of heterologous pigmentation of genetically engineered Saccharomyces cerevisiae strains in synthetic defined chemical environments aligned with experimental observations. We then further applied TRAEV to predictively assess the robustness of desired aroma generation trait in a wine must environment by multiple S. cerevisiae strains if they were developed via a laboratory selection process. Thus, we showed how TRAEV predictions could guide such strain development. TRAEV can be integrated into computational strain design workflows across microbial strains, metabolic traits, and application environments. Ultimately, model-predicted evolutionary robustness of desired traits can guide both strain and process development and help avoiding production losses and enhancing the economic attractiveness of industrial applications using microbial cells. Author summaryMicrobial cells developed to express desired traits by genetic engineering or selection processes are used in a wide range of applications, from biotechnological chemical production to food and beverage fermentations. However, when the cells replicate in the application environment the traits are subject to Darwinian selection and consequent adaptive evolution. As a result, desired traits can be rapidly lost. To mitigate such undesired evolutionary changes, we developed TRAEV (Trait Robustness Against EVolution), a computational framework for in silico assessment of evolutionary trajectories and robustness of desired traits in application environments by integrating constraint-based modeling and Monte Carlo-based perturbation analysis methods. We demonstrate the usability of TRAEV by validating its predictions with experimental data on the robustness of heterologous pigmentation of engineered Saccharomyces cerevisiae strains, and applying it to predict the robustness of aroma generation in wine must by multiple S. cerevisiae strains if they were developed by laboratory selection in specific conditions. As being applicable across microbial strains, metabolic traits, and application environments, we believe that TRAEV can help to avoid production losses, and thus, contribute to the development of economically attractive industrial applications using microbial cells.

evolutionary biology↗

bioRxiv2025

Normative assembly rule reveals fairness in microbial communities

Understanding and predicting how communities assemble is a paramount challenge in ecology. Here we address these questions normatively by comparing the ecological distribution of growth surplus to a game-theoretically fair distribution based on each species Shapley value. By analyzing in total 56 distinct community outcomes, we assess how fairly biomass is distributed in microbial communities displaying both competitive and cooperative interactions in different environmental conditions. We find examples of fair communities that closely follow their Shapley value across all environments as well as counterexamples where the true abundances deviate from the species objective contribution to community biomass. Our results give unique empirical insights into the distributive function of ecological dynamics and lay down the theoretical foundations of what might become a normative community assembly theory.

ecology↗

bioRxiv2025

Evolution induced state shifts in a long-term microbial community experiment

Biological communities are complex, dynamic systems that underpin ecosystem functionality1, yet their long-term dynamics and predictability remain poorly understood2. Understanding how Darwinian evolution shapes these systems through eco-evolutionary feedbacks is a central challenge in ecology and evolution. Experimental studies using simplified microbial assemblages have yielded important insights into the ecological principles governing community states3-5. However, an important knowledge gap is how selection within member species drives changes of community state in multispecies systems. Here, we present a four-year evolution experiment involving a 23-species synthetic bacterial community propagated in two environments: a control medium and the same medium supplemented with the antibiotic streptomycin. Through combined analyses of community composition and genome evolution, we quantified the temporal changes in species abundances and the evolutionary trajectories of individual community members. The extended duration of the experiment enabled the detection of adaptive mutations and community state shifts that occur only over long evolutionary timescales. We show that community dynamics are environment dependent and reproducible across replicates, and that evolution of streptomycin resistance in a previously streptomycin-sensitive species on its own can induce abrupt community state shifts. Our results provide a direct demonstration of eco-evolutionary feedbacks within a multi-species community, revealing how a single adaptive mutation can reorganize complex ecological networks.

ecology↗

bioRxiv2025

Non-ergodicity in ecology and evolution

Stochasticity plays an important role in all biological systems. The standard way to deal with stochasticity involves averaging over an ensemble of independent realizations. However, such mean statistics need not accurately reflect the typical outcomes in any finite sample unless the system satisfies the property of ergodicity, which guarantees that each trajectory will over time experience the same statistics as the entire ensemble. Here, we argue that, in contrast, non-ergodicity might instead be the rule rather than exception in real biological systems and investigate its implications for eco-evolutionary dynamics through three case-studies. First, we show how demographic stochasticity leads to ergodicity breaking where the asymptotic growth rate carries a signature of the initial condition. This motivates us to define a mutant establishment threshold, which quantifies a critical population size above which the typical mutant population starts to grow. Second, we consider environmental stochasticity and demonstrate that autocorrelated environmental variation does not average out in time, which has the important consequence that the fitness of a genotype cannot be simply averaged over the environments. Finally, we show how in a metapopulation structure the evolutionary dynamics within a typical subpopulation can deviate from the ensemble dynamics in the entire metapopulation, which is sufficient to explain the evolution and persistence of cooperation despite a fitness cost.

evolutionary biology↗

bioRxiv2025

Data-driven symbolic regression reveals microbial growth laws

O_LIConnecting mathematical models with empirically measured microbial growth has remained challenging, as numerous competing models based on different theoretical approaches can fit observations. Therefore, we develop a method to automatically propose growth models from microbial data alone. We validate this approach using an available dataset of E. coli grown on known resources, and study sixteen species across various concentrations of a rich medium. C_LIO_LIThe inherently interpretable approach of symbolic regression infers explicit dynamical models directly from growth data. Using symbolic regression natively, does not favour biologically interpretable models, but we find cumulative population gain to be a more informative machine learning feature than population size. C_LIO_LIRandom Forest machine learning allows us to relate this finding to the approximation of a constant-rate per capita resource consumption. This suggests that the area under the growth curve (AUC) measured in routine experiments provides information on the effective resource dynamics governing microbial growth. Finally, we use theoretical insights to inform the symbolic regression algorithm and favour biologically interpretable models. C_LIO_LIOverall, we found that balancing between data fit, parsimony and biological relevance favoured both the simplest, linear approximation, and models based on a Monod dynamics, with either one or two underlying resources. Therefore, our approach to read growth laws off of microbial batch cultures provides insights on data-driven modelling. C_LI

ecology↗

bioRxiv2025

Universal bacterial clade dynamics dominate under predation despite altered phenotypes and mutation targets

Recent studies have revealed bacterial genome-wide evolution to be complex and dynamic even in a constant environment, characterized by the emergence of new clades competing or temporarily coexisting as each clade undergoes evolutionary change. Previous studies on predator-prey dynamics tracking simple ecological and phenotypic metrics have shown predation to fundamentally alter prey evolution, facilitating defense evolution followed by coevolution and frequency dependent selection between defended and undefended prey genotypes. Here we sought to consolidate these fields by examining genome-wide evolution in five bacterial prey species separately subjected to long-term evolution under ciliate predation. We hypothesized that the presence of predation could change the pattern of clonal dynamics, for example, by more frequently producing selective sweeps if predation-defense-related mutations are under strong selection. For all species, we found mutational signals of prey adaptation, with phenotypic data and genomic mutation targets demonstrating changes in composition between the experimental treatments. Intriguingly, despite higher variant counts, overall temporal clade dynamics across the coevolved prey species were strikingly similar to those of bacteria evolving alone, with constant emergence, competition and quasi-stable coexistence of clades. This study shows that long-term molecular evolution in bacterial prey under predation is more interesting and less predictable than we might expect based on existing coevolutionary theories.

evolutionary biology↗

bioRxiv2025

A novel DNA repair protein, N-Myc downstream regulated gene 1 (NDRG1), links stromal tumour microenvironment to chemoresistance

In pancreatic ductal adenocarcinoma cancer (PDAC) drug resistance is a severe clinical problem and patients relapse within a few months after receiving the standard-of-care chemotherapy. One contributing factor to treatment resistance is the desmoplastic nature of PDAC; the tumours are surrounded by thick layers of stroma composing up to 90% of the tumour mass. This stroma, which is mostly comprised of extracellular matrix (ECM) proteins, is secreted by cancer-associated fibroblasts (CAFs) residing in the tumour microenvironment. However, the mechanistic basis by which the tumour stroma directly contributes to chemoresistance remains unclear. Here, we show that CAF-secreted ECM proteins induce chemoresistance by blunting chemotherapy-induced DNA damage. Mechanistically, we identify N-myc downstream regulated gene 1 (NDRG1) as a key protein required for stroma-induced chemoresistance that responds to signals from the ECM and adhesion receptors. We further show that NDRG1 is a novel DNA repair protein that physically interacts with replication forks, maintains DNA replication and functions to resolve stalled forks caused by chemotherapy. More specifically, NDRG1 reduces R-loops, RNA-DNA hybrids that are known to cause genomic instability. R-loops occur during replication-transcription conflicts in S-phase and after chemotherapy treatments, thus posing a major threat to normal replication fork homeostasis. We identify NDRG1 as highly expressed in PDAC tumours, and its high expression correlates with chemoresistance and poor disease-specific survival. Importantly, knock-out of NDRG1 or inhibition of its phosphorylation restores chemotherapy-induced DNA damage and resensitizes tumour cells to treatment. In conclusion, our data reveal an unexpected role for CAF-secreted ECM proteins in enhancing DNA repair via NDRG1, a novel DNA repair protein, directly linking tumour stroma to replication fork homeostasis and R-loop biology, with important therapeutic implications for restoring DNA damage response pathways in pancreatic cancer. Summary paragraphDrug resistance is a severe clinical problem in stroma-rich tumours, such as pancreatic ductal adenocarcinoma (PDAC), and patients often relapse within a few months on chemotherapy1-9. The stroma, comprised of extracellular matrix (ECM) proteins, is secreted by cancer-associated fibroblasts (CAFs) residing in the tumour microenvironment10-13. Prior work show that ECM proteins provide survival benefits to cancer cells14,15. However, the precise role of CAF-secreted ECM in resistance to DNA damaging chemotherapies remains poorly understood. Here, we link ECM proteins to chemoresistance by enhanced DNA damage repair (DDR). Mechanistically, we identify N-myc downstream-regulated gene 1 (NDRG1) as a key effector downstream of ECM and the integrin-Src-SGK1-signalling axis that mediates enhanced DDR. We show that NDRG1 loss, mutation of conserved His194, or inhibition of NDRG1 phosphorylation by SGK1 lead to replication fork stalling, increased R-loops, and higher transcription-replication conflicts, resulting in genomic instability and sensitivity to chemotherapies. Our analysis of PDAC patient cohorts16 found that high NDRG1 expression correlates with chemoresistance and poor patient survival. In conclusion, we uncover an unexpected role for CAF-secreted ECM proteins in promoting therapeutic resistance by enhancing DDR and establish NDRG1 as a novel DNA repair protein directly linking tumour stroma to DDR.

cancer biology↗

bioRxiv2024

Long deletion signatures in repetitive genomic regions track somatic evolution and enable sensitive detection of microsatellite instability

Deficiency in the mismatch repair system (MMRd) causes microsatellite instability (MSI) in cancers and determines eligibility for immunotherapy. Here, we show that MMRd tumours harbour long-deletion signatures ([≥]2-5+ base pairs deleted in repetitive regions), which provide new insights into MSI evolution and enable sensitive MSI detection particularly in challenging clinical samples. Long deletions, accumulated through stepwise DNA slippage errors, are significantly more prevalent in metastatic MMRd tumours compared to primary tumours. Importantly, we show that long-deletion signatures harbour features that are distinct from background noise, making them robustly detectable even in shallow whole genome sequencing (sWGS, [~]0.1X coverage) of formalin-fixed samples. We constructed a machine learning classifier that uses these distinct features to detect Microsatellite Instability in LOw-quality (MILO) samples. MILO achieved 100% accuracy in detecting MSI in sWGS data with only 2%-15% tumour purity and demonstrated promise in identifying MMRd clones in precancerous intestinal lesions. We propose that MILO could be clinically used for the sensitive monitoring of MMRd cancer evolution from early to late stages, using minimal sequencing data from both archival and fresh-frozen samples with low tumour content. SignificanceMutational signatures characterised by long deletions in repetitive genomic regions provide a sensitive route to detect and track MMRd clone evolution, even with low purity shallow whole genome sequencing data.

bioinformatics↗

bioRxiv2023

Quantifying massively parallel microbial growth with spatially mediated interactions

Quantitative understanding of microbial growth is an essential prerequisite for successful control of pathogens as well as various biotechnology applications. Even though the growth of cell populations has been extensively studied, microbial growth remains poorly characterized at the spatial level. Indeed, even isogenic populations growing at different locations on solid growth medium typically show significant location-dependent variability in growth. Here we show that this variability can be attributed to an interplay between populations interacting with their local environment and the diffusion of nutrients and energy sources coupling the environments, i.e. interpopulation interactions are mediated via the shared environment. We use a dual approach, first applying machine learning regression models to discover that location dominates growth variability at specific times, and, in parallel, developing explicit population growth models to describe this spatial effect. In particular, treating nutrient and energy source concentration as a latent variable allows us to develop a mechanistic resource consumer model that captures growth variability across the shared environment. As a consequence, we are able to determine intrinsic growth parameters for each local population, removing confounders common to location-dependent variability in growth. Importantly, our explicit low parametric model for the environment paves the way for massively parallel experimentation with configurable spatial niches for testing specific eco-evolutionary hypotheses. Author summaryImage-based platforms allow obtaining population size estimates for massively parallel growth experiments on substrate plates at relatively low cost. However, such population size data has been shown to display a high degree of spatial variability, which occurs even with isogenic populations. Here we first quantified the importance of spatial location on growth variation using a machine learning approach, and then developed spatially aware population growth models to explain the spatial structure of the growth data. Ultimately, we showed that a spatial consumer-resource model with local microhabitats connected via diffusion can fully explain the observed spatial variation in growth while allowing the inference of intrinsic growth parameters of specific populations. This result provides a method for systematic extraction of spatial growth models and paves the way for massively parallel eco-evolutionary experimentation.

ecology↗

bioRxiv2022

Turnover shapes evolution of birth and death rates

Population turnover, a key trait shaped by the organisms life history strategy, plays an important role in eco-evolutionary dynamics by fixing the timescale for individual birth and death events as well as in determining the level of demographic stochasticity related to growth. Yet, the standard theory of population genetics, and the models heavily used in the related data analysis, have largely ignored the role of turnover. Here we propose a reformulation of population genetics starting from the first principles of birth and death and show that the role of turnover is evolutionarily important. We derive a general stochastic differential equation for the frequency dynamics of competing birth-death processes and determine the appropriate turnover corrections for the essential results regarding fixation, establishment, and substitution of mutants. Our results reveal how both the absolute and relative turnover rates influence evolution. We further describe a deterministic turnover selection, the turnover flux, which operates in small populations. Finally, we analyse the evolution of mean turnover and show how it explains the key eco-evolutionary mechanisms underlying demographic transitions. In conclusion, our results explicitly show how competing life-history strategies, demographic stochasticity, ecological feedback, and evolution are inseparably intertwined, thus calling for a unified theory development starting from the underlying mechanisms of birth and death.

evolutionary biology↗

bioRxiv2021

Machine-learning prediction of resistance to sub-inhibitory antimicrobial concentrations from Escherichia coli genomes

Escherichia coli is an important cause of bacterial infections worldwide, with multidrug resistant strains incurring substantial costs on human lives. Besides therapeutic concentrations of antimicrobials in healthcare settings, the presence of sub-inhibitory antimicrobial residues in the environment and in the clinics selects for antimicrobial resistance (AMR), but the underlying genetic repertoire is less well understood. We used machine-learning to predict the population doubling time and growth yield of 1,432 genetically diverse E. coli expanding under exposure to three sub-inhibitory concentrations of six classes of antimicrobials from single nucleotide genetic variants, accessory gene variation and the presence of known AMR genes. We could predict cell yields in the held-out test data with an average correlation (Spearmans {rho}) of 0.63 (0.32 - 0.90 across concentrations) and cell doubling time with an average correlation of 0.47 (0.32 - 0.74 across concentrations), with moderate increases in sample size unlikely to improve predictions further. This points to the remaining missing heritability of growth under antimicrobials exposure being explained by effects that are too rare or weak to be captured unless sample size is dramatically increased, or by effects other than those conferred by the presence of individual SNPs and genes. Predictions based on whole genome information were generally superior to those based only on known AMR genes, and also accurate for AMR resistance at therapeutic concentrations. We also pinpointed genes and SNPs determining the predicted growth and thereby recapitulated the known AMR determinants. Finally, we estimated the effect sizes of resistance genes across the entire collection of strains, disclosing growth effects for known resistance genes for each strain. Our results underscore the potential of predictive modelling of growth patterns from genomic data under sub-inhibitory concentrations of antimicrobials, although the remaining missing heritability poses an issue for achieving the accuracy and precision required for clinical use. ImportancePredicting bacterial growth from genome sequences is important not only for a rapid characterization of strains in clinical diagnostic applications but for the identification of novel targets for drug discovery. Previous studies examined the relationship between bacterial growth and genotype in mutant libraries for laboratory strains, yet no study has so far examined the prediction power of genome sequences for bacterial growth in natural strains. In this study, we used a high throughput phenotypic assay to measure bacterial growth of a systematic collection of natural Escherichia coli strains and then employed machine learning models to predict bacterial growth from genomic data under non-therapeutic sub-inhibitory concentrations of antimicrobials that are common in nonclinical settings. Our results revealed a moderate to strong correlation between predicted and actual values for different antimicrobials concentrations. Furthermore, the quantified effect of resistance genes on bacterial growth indicate these genes are still effective at sublethal antimicrobial concentrations.

microbiology↗

bioRxiv2021

The mutational signatures of formalin fixation on the human genome

BackgroundFormalin fixation and paraffin embedding (FFPE) of patient material remains standard practice in clinical pathology labs around the world. Clinical archives of patient material near-exclusively consist of FFPE blocks. The ability to perform high quality genome sequencing on FFPE-derived DNA would accelerate a broad spectrum of medical research. However, formalin is a recognised mutagen and sequencing of DNA derived from FFPE material is known to be riddled with artefactual mutations. ResultsHere we derive genome-wide mutational signatures caused by formalin fixation, and provide a computational method to correct mutational profiles for these formalin-induced artefacts. We show that the FFPE-signature is dominated by C>T transitions caused by cytosine deamination, and has very high similarity to COSMIC signature SBS30 (base excision repair deficiency due to inactivation mutations in NTHL1). Further, we demonstrate that chemical repair of formalin-induced DNA lesions, a process that is routinely performed as part of sequencing library preparation, leads to a signature highly similar to COSMIC signature SBS1 (spontaneous deamination of methylated cytosine). Next, we design FFPEsig, a computational method to remove the formalin-induced artefacts from mutational counts. We prove the efficacy of this method by generating synthetic FFPE samples using 2,780 cancer genomes from the Pan-Cancer Analysis of Whole Genome (PCAWG) project, and via analysis of FFPE-derived genome sequencing data from colorectal cancers. ConclusionsFormalin fixation leaves a predictable mutational footprint across the genome. The application of our FFPEsig software corrects the mutational profile for the influence of formalin, enabling robust mutational signature analysis in FFPE-derived patient material.

bioinformatics↗

bioRxiv2020

Strong Selective Agents Determine Resistance Evolution in a Multidrug Therapeutic Regime

The impact of fitness landscape features on evolutionary outcomes has attracted considerable interest in recent decades. However, evolution often occurs under time-dependent selection in so-called fitness seascapes where the landscape is under flux. Fitness seascapes are an inherent feature of natural environments, where the landscape changes owing both to the intrinsic fitness consequences of previous adaptations and extrinsic changes in selected traits caused by new environments. The complexity of such seascapes may curb the predictability of evolution. However, empirical efforts to test this question utilising a comprehensive set of regimes are lacking. Here we employed an in vitro microbial model system to investigate differences in evolutionary outcomes between time-invariant and -dependent environments, including all possible temporal permutations, with three subinhibitory antimicrobials and a viral parasite (phage) as selective agents. Expectedly, time-invariant environments caused stronger directional selection for resistances compared to time-dependent environments. Intriguingly, however, multidrug resistance outcomes in both cases were largely driven by two strong selective agents (rifampicin and phage) out of four agents in total. These agents either caused cross-resistance or obscured the phenotypic effect of other resistance mutations, modulating the evolutionary outcome overall in time-invariant environments and as a function of exposure epoch in time-dependent environments. This suggests that identifying strong selective agents and their pleiotropic effects is critical for predicting evolution in fitness seascapes, with ramifications for evolutionarily informed strategies to mitigate drug resistance evolution.

evolutionary biology↗

bioRxiv2020

Drug-induced resistance evolution necessitates less aggressive treatment

Evolution of drug resistance to anticancer, antimicrobial and antiviral therapies is widespread among cancer and pathogen cell populations. Classical theory posits strictly that genetic and phenotypic variation is generated in evolving populations independently of the selection pressure. However, recent experimental findings among antimicrobial agents, traditional cytotoxic chemotherapies and targeted cancer therapies suggest that treatment not only imposes selection but also affects the rate of adaptation via altered mutational processes. Here we analyze a model with drug-induced increase in mutation rate and explore its consequences for treatment optimization. We argue that the true biological cost of treatment is not limited to the harmful side-effects, but instead realizes even more profoundly by fundamentally changing the underlying eco-evolutionary dynamics within the microenvironment. Factoring in such costs (or collateral damage) of control is at the core of successful therapy design and can unify different evolution-based approaches to therapy optimization. Using the concept of evolutionary rescue, we formulate the treatment as an optimal control problem and solve the optimal elimination strategy, which minimizes the probability of evolutionary rescue. Our solution exploits a trade-off, where increasing the drug concentration has two opposing effects. On the one hand, it reduces de novo mutations by decreasing the size of the target cell population faster; on the other hand, a higher dosage generates a surplus of treatment-induced mutations. We show that aggressive elimination strategies, which aim at eradication as fast as possible and which represent the current standard of care, can be detrimental even with modest drug-induced increases (fold change[≤] 10) to the baseline mutation rate. Our findings highlight the importance of dose dependencies in resistance evolution and motivate further investigation of the mutagenicity and other hidden collateral costs of therapies that promote resistance. Author summaryThe evolution of drug resistance is a particularly problematic and frequent outcome of cancer and antimicrobial therapies. Recent research suggests that these treatments may enhance the evolvability of the target population not only via inducing intense selection pressures but also via altering the underlying mutational processes. Here we investigate the consequences of such drug-induced evolution by considering a mathematical model with explicitly dose-dependent mutation rate. We identify, characterize and exploit a trade-off between decreasing the target population size as fast as possible and generating a surplus of treatment-induced de novo mutations. By formulating the treatment as an optimal control problem over the evolution of the target population, we find the optimal treatment strategy, which minimizes the probability of evolutionary rescue. We show that this probability changes non-monotonically with the cumulative drug concentration and is minimized at an intermediate dosage. Our results are immediately amenable to experimental investigation and motivate further study of the various mutagenic and other hidden collateral costs of treatment. Taken together, our results add to the ongoing criticism of the standard practice of administering aggressive, high-dose therapies and stimulate further clinical trials on alternative treatment strategies.

evolutionary biology↗

bioRxiv2020

Genomic epidemiology and evolution of Escherichia coli in wild animals

Escherichia coli is a common bacterial species in the gastrointestinal tracts of warm-blooded animals and humans. Pathogenic and antimicrobial resistance in E. coli may emerge via host switching from animal reservoirs. Despite its potential clinical importance, knowledge of the population structure of commensal E. coli within wild hosts and the epidemiological links between E. coli in non-human hosts and E. coli in humans is still scarce. In this study, we analysed the whole genome sequencing data of a collection of 119 commensal E. coli recovered from the guts of 68 mammal and bird species in Mexico and Venezuela in the 1990s. We observed low concordance between the population structures of E. coli colonizing wild animals and the phylogeny, taxonomy and ecological and physiological attributes of the host species, with distantly related E. coli often colonizing the same or similar host species and distantly related host species often hosting closely related E. coli. We found no evidence for recent transmission of E. coli genomes from wild animals to either domesticated animals or humans. However, multiple livestock- and human-related virulence factor genes were present in E. coli of wild animals, including virulence factors characteristic for Shiga toxin-producing E. coli (STEC) and atypical enteropathogenic E. coli (aEPEC), where several isolates from wild hosts harboured the locus of enterocyte effacement (LEE) pathogenicity island. Moreover, E. coli in wild animal hosts often harboured known antibiotic resistance determinants, including against ciprofloxacin, aminoglycosides, tetracyclines and beta-lactams, with some determinants present in multiple, distantly related E. coli lineages colonizing very different host animals. We conclude that although the genome pools of E. coli colonizing wild animal and human gut are well separated, they share virulence and antibiotic resistance genes and E. coli underscoring that wild animals could serve as reservoirs for E. coli pathogenicity in human and livestock infections. ImportanceEscherichia coli is a clinically importance bacterial species implicated in human and livestock associated infections worldwide. The bacterium is known to reside in the guts of humans, livestock and wild animals. Although wild animals are recognized to serve as potential reservoirs for pathogenic E. coli strains, the knowledge of the population structure of E. coli in wild hosts is still scarce. In this study we used the fine resolution of whole genome sequencing to provide novel insights into the evolution of E. coli genomes within a broad range of wild animal species (including mammals and birds), the co-evolution of E. coli strains with their hosts and the genetics of pathogenicity of E. coli strains in wild hosts. Our results provide evidence for the clinical importance of wild animals as reservoirs for pathogenic strains and necessitate the inclusion of non-human hosts in the surveillance programs for E. coli infections.

microbiology↗