Clonal Interference Research Articles

Midostaurin shapes macroclonal and microclonal evolution of FLT3-mutated acute myeloid leukemia

AbstractDespite the use of midostaurin (MIDO) with intensive chemotherapy (ICT) as the frontline treatment for Fms-like tyrosine kinase 3 (FLT3)-mutated acute myeloid leukemia (AML), complete remission rates are close to 60% to 70%, and relapses occur in >40% of cases. Here, we studied the molecular mechanisms underlying refractory/relapsed (R/R) disease in patients with FLT3-mutated AML. We conducted a retrospective and multicenter study involving 150 patients with R/R AML harboring FLT3–internal tandem duplication (ITD) (n = 130) and/or FLT3–tyrosine kinase domain (n = 26) at diagnosis assessed by standard methods. Patients were treated in front line with ICT + MIDO (n = 54) or ICT alone (n = 96) according to the diagnosis date and label of MIDO. The evolution of FLT3 clones and comutations was analyzed in paired diagnosis–R/R samples by targeted high-throughput sequencing. Using a dedicated algorithm for FLT3-ITD detection, 189 FLT3-ITD microclones (allelic ratio [AR] of <0.05) and 225 macroclones (AR ≥ 0.05) were detected at both time points. At R/R disease, the rate of FLT3-ITD persistence was lower in patients treated with ICT + MIDO than in patients not receiving MIDO (68% vs 87.5%; P = .011). In patients receiving ICT + MIDO, detection of multiple FLT3-ITD clones (referred to as “clonal interference”) was associated with a higher FLT3-ITD persistence rate at R/R disease (multiple clones: 88% vs single clones: 57%; P = .049). Considering both treatment groups, if only 24% of FLT3-ITD microclones detected at diagnosis were retained at relapse, 43% became macroclones. Together, these results identify parameters influencing the fitness of FLT3-ITD clones and highlight the importance of using sensitive techniques for FLT3-ITD screening in clinical practice.

Genetic diversity during selective sweeps in non-recombining populations.

Selective sweeps, resulting from the spread of beneficial, neutral, or deleterious mutations through a population, shape patterns of genetic variation at linked neutral sites. While many theoretical, computational, and statistical advances have been made in understanding the genomic signatures of selective sweeps in recombining populations, substantially less is understood in populations with little/no recombination. We present a mathematical framework based on diffusion theory for obtaining the site frequency spectrum (SFS) at linked neutral sites immediately post and during the fixation of moderately or strongly beneficial mutations. We find that when a single hard sweep occurs, the SFS decays as 1/ x for low derived allele frequencies ( x ), similar to the neutral SFS at equilibrium, whereas at higher derived allele frequencies, it follows a 1/ x 2 power law. These power laws are universal in the sense that they are independent of the dominance and inbreeding coefficient, and also characterize the SFS during the sweep. Additionally, we find that the derived allele frequency where the SFS shifts from the 1/ x to 1/ x 2 law, is inversely proportional to the selection strength: thus under strong selection, the SFS follows the 1/ x 2 dependence for most allele frequencies, resembling a rapidly expanding neutral population. When clonal interference is pervasive, the SFS immediately post-fixation becomes U-shaped and is better explained by the equilibrium SFS of selected sites. Our results will be important in developing statistical methods to infer the timing and strength of recent selective sweeps in asexual populations, genomic regions that lack recombination, and clonally propagating tumor populations.

Effects of periodic bottlenecks on the dynamics of adaptive evolution in microbial populations.

Population bottlenecks can impact the rate of adaptation in evolving populations. On the one hand, each bottleneck reduces the genetic variation that fuels adaptation. On the other hand, each founder that survives a bottleneck can undergo more generations and leave more descendants in a resource-limited environment, which allows surviving beneficial mutations to spread more quickly. A theoretical model predicted that the rate of fitness gains should be maximized using ~8-fold dilutions. Here we investigate the impact of repeated bottlenecks on the dynamics of adaptation using numerical simulations and experimental populations of Escherichia coli. Our simulations confirm the model's prediction when populations evolve in a regime where beneficial mutations are rare and waiting times between successful mutations are long. However, more extreme dilutions maximize fitness gains in simulations when beneficial mutations are common and clonal interference prevents most of them from fixing. To examine these predictions, we propagated 48 E. coli populations with 2-, 8-, 100-, and 1000-fold dilutions for 150 days. Adaptation began earlier and fitness gains were greater with 100- and 1000-fold dilutions than with 8-fold dilutions, consistent with the simulations when beneficial mutations are common. However, the selection pressures in the 2-fold treatment were qualitatively different from the other treatments, violating a critical assumption of the model and simulations. Thus, varying the dilution factor during periodic bottlenecks can have multiple effects on the dynamics of adaptation caused by differential losses of diversity, different numbers of generations, and altered selection.

Step-wise evolution of azole resistance through copy number variation followed by KSR1 loss of heterozygosity in Candida albicans.

Antimicrobial drug resistance poses a global health threat, requiring a deeper understanding of the evolutionary processes that lead to its emergence in pathogens. Complex evolutionary dynamics involve multiple mutations that can result in cooperative or competitive (clonal interference) effects. Candida albicans, a major fungal pathogen, displays high rates of copy number variation (CNV) and loss of heterozygosity (LOH). CNV and LOH events involve large numbers of genes and could synergize during evolutionary adaptation. Understanding the contributions of CNV and LOH to antifungal drug adaptation is challenging, especially in the context of whole-population genome sequencing. Here, we document the sequential evolution of fluconazole tolerance and then resistance in a C. albicans isolate involving an initial CNV on chromosome 4, followed by an LOH on chromosome R that involves KSR1. Similar LOH events involving KSR1, which encodes a reductase in the sphingolipid biosynthesis pathway, were also detected in independently evolved fluconazole resistant isolates. We dissect the specific KSR1 codons that affect fluconazole resistance and tolerance. The combination of the chromosome 4 CNV and KSR1 LOH results in a >500-fold decrease in azole susceptibility relative to the progenitor, illustrating a compelling example of rapid, yet step-wise, interplay between CNV and LOH in drug resistance evolution.

Self-replicating artificial neural networks give rise to universal evolutionary dynamics.

In evolutionary models, mutations are exogenously introduced by the modeler, rather than endogenously introduced by the replicator itself. We present a new deep-learning based computational model, the self-replicating artificial neural network (SeRANN). We train it to (i) copy its own genotype, like a biological organism, which introduces endogenous spontaneous mutations; and (ii) simultaneously perform a classification task that determines its fertility. Evolving 1,000 SeRANNs for 6,000 generations, we observed various evolutionary phenomena such as adaptation, clonal interference, epistasis, and evolution of both the mutation rate and the distribution of fitness effects of new mutations. Our results demonstrate that universal evolutionary phenomena can naturally emerge in a self-replicator model when both selection and mutation are implicit and endogenous. We therefore suggest that SeRANN can be applied to explore and test various evolutionary dynamics and hypotheses.

Constitutive expression of the Type VI Secretion System carries no measurable fitness cost in Vibrio cholerae.

The Type VI Secretion System (T6SS) is a widespread and highly effective mechanism of microbial warfare; it confers the ability to efficiently kill susceptible cells within close proximity. Due to its large physical size, complexity, and ballistic basis for intoxication, it has widely been assumed to incur significant growth costs in the absence of improved competitive outcomes. In this study, we precisely examine the fitness costs of constitutive T6SS firing in the bacterium Vibrio cholerae. We find that, contrary to expectations, constitutive expression of the T6SS has a negligible impact on growth, reducing growth fitness by 0.025 ± 0.5% (95% CI) relative to a T6SS- control. Mathematical modeling of microbial populations demonstrates that, due to clonal interference, constitutive expression of the T6SS will often be neutral, with little impact on evolutionary outcomes. Our findings underscore the importance of precisely measuring the fitness costs of microbial social behaviors and help explain the prevalence of the T6SS across Gram-negative bacteria.

How microscopic epistasis and clonal interference shape the fitness trajectory in a spin glass model of microbial long-term evolution.

The adaptive dynamics of evolving microbial populations takes place on a complex fitness landscape generated by epistatic interactions. The population generically consists of multiple competing strains, a phenomenon known as clonal interference. Microscopic epistasis and clonal interference are central aspects of evolution in microbes, but their combined effects on the functional form of the population's mean fitness are poorly understood. Here, we develop a computational method that resolves the full microscopic complexity of a simulated evolving population subject to a standard serial dilution protocol. Through extensive numerical experimentation, we find that stronger microscopic epistasis gives rise to fitness trajectories with slower growth independent of the number of competing strains, which we quantify with power-law fits and understand mechanistically via a random walk model that neglects dynamical correlations between genes. We show that increasing the level of clonal interference leads to fitness trajectories with faster growth (in functional form) without microscopic epistasis, but leaves the rate of growth invariant when epistasis is sufficiently strong, indicating that the role of clonal interference depends intimately on the underlying fitness landscape. The simulation package for this work may be found at https://github.com/nmboffi/spin_glass_evodyn.

How microscopic epistasis and clonal interference shape the fitness trajectory in a spin glass model of microbial long-term evolution

The adaptive dynamics of evolving microbial populations takes place on a complex fitness landscape generated by epistatic interactions. The population generically consists of multiple competing strains, a phenomenon known as clonal interference. Microscopic epistasis and clonal interference are central aspects of evolution in microbes, but their combined effects on the functional form of the population’s mean fitness are poorly understood. Here, we develop a computational method that resolves the full microscopic complexity of a simulated evolving population subject to a standard serial dilution protocol. Through extensive numerical experimentation, we find that stronger microscopic epistasis gives rise to fitness trajectories with slower growth independent of the number of competing strains, which we quantify with power-law fits and understand mechanistically via a random walk model that neglects dynamical correlations between genes. We show that increasing the level of clonal interference leads to fitness trajectories with faster growth (in functional form) without microscopic epistasis, but leaves the rate of growth invariant when epistasis is sufficiently strong, indicating that the role of clonal interference depends intimately on the underlying fitness landscape. The simulation package for this work may be found at https://github.com/nmboffi/spin_glass_evodyn.

A dynamical limit to evolutionary adaptation

Natural selection makes evolutionary adaptation possible even if the overwhelming majority of new mutations are deleterious. However, in rapidly evolving populations where numerous linked mutations occur and segregate simultaneously, clonal interference and genetic hitchhiking can limit the efficiency of selection, allowing deleterious mutations to accumulate over time. This can in principle overwhelm the fitness increases provided by beneficial mutations, leading to an overall fitness decline. Here, we analyze the conditions under which evolution will tend to drive populations to higher versus lower fitness. Our analysis focuses on quantifying the boundary between these two regimes, as a function of parameters such as population size, mutation rates, and selection pressures. This boundary represents a state in which adaptation is precisely balanced by Muller's ratchet, and we show that it can be characterized by rapid molecular evolution without any net fitness change. Finally, we consider the implications of global fitness-mediated epistasis and find that under some circumstances, this can drive populations toward the boundary state, which can thus represent a long-term evolutionary attractor.

Tracing the adaptive evolution of SARS-CoV-2 during vaccine roll-out in Norway.

Vaccination against SARS-CoV-2 has greatly mitigated the impact of the COVID-19 pandemic. However, concerns have been raised about the degree to which vaccination might drive the emergence and selection of immune escape mutations that will hamper the efficacy of the vaccines. In this study, we investigate whether vaccination impacted the micro-scale adaptive evolution of SARS-CoV-2 in the Oslo region of Norway, during the first nine months of 2021, a period in which the population went from near-zero to almost 90 per cent vaccine coverage in the population over 50 years old. Weekly aggregated data stratified by age on vaccine uptake and number of SARS-CoV-2 cases in the area were obtained from the National Immunization Registry and the Norwegian Surveillance System for Communicable Diseases, respectively. A total of 6,438 virus sequences (7.5 per cent of the registered cases) along with metadata were available. We used a causal-driven approach to investigate the relationship between vaccination progress and changes in the frequency of 362 mutations present in at least ten samples, conditioned on the emergence of new lineages, time, and population vaccination coverage. After validating our approach, we identified 21 positive and 12 negative connections between vaccination progress and mutation prevalence, and most of them were outside the Spike protein. We observed a tendency for the mutations that we identified as positively connected with vaccination to decrease as the vaccinated population increased. After modelling the fitness of different competing mutations in a population, we found that our observations could be explained by a clonal interference phenomenon in which high fitness mutations would be outcompeted by the emergence or introduction of other high-fitness mutations.

Clonal Dynamics of FLT3-ITD Positive Acute Myeloid Leukemia Patients with Relapsed/Refractory Disease Following Intensive Chemotherapy +/- Midostaurin

Introduction: Despite the wider use of midostaurin (MIDO) in combination with intensive chemotherapy (ICT) as the 1st-line treatment for FLT3-mutated acute myeloid leukemia (AML), complete remission (CR) rates are close to 60%, and relapses occur in over 40% of cases, demonstrating the ability of leukemic cells to resist and evade therapy ( Stone et al., NEJM 2017). Conventional fragment-length analyses of paired diagnosis/relapse samples have shown that FLT3-internal tandem duplications (ITDs) are retained in 80% and 50% of cases following ICT alone and MIDO+ICT respectively ( Schmalbrock et al., Blood 2021). Only limited data are available on the dynamics of FLT3-ITDs or other co-mutations in refractory patients (pts). Here, we conducted a retrospective study involving 115 pts with relapsed/refractory AML harboring FLT3-ITD at diagnosis. Materials and methods: Clonal evolution was examined in paired diagnosis/progression blood or bone marrow samples from 115 pts with FLT3-ITD+ AML treated with MIDO+ICT (n=33) or ICT alone (n=82). Among them, 21 pts had primary refractory disease (MIDO+ICT, n=8; ICT, n=13) and 94 pts relapsed after achieving CR (MIDO+ICT, n=25; ICT, n=69). FLT3-ITDs and co-mutations were screened on genomic DNA by high-throughput sequencing at both timepoints using a custom-designed panel. For accurate annotation and quantification of FLT3-ITDs from sequencing data, we used the recentlypublished FiLT3r algorithm ( Boudry et al., BMC Bioinformatics 2022). For each ITD detected, FiLT3r allelic ratio (AR) was assessed by the ratio between the mutated allele and the wild-type allele. Results: A total of 226 FLT3-ITDs were detected in 115 pts at AML diagnosis, among which 120 (53%) ITDs were found with an AR below 0.05 ( Figure 1). Among pts who achieved CR and experienced relapse (n=94), 48 had multiple FLT3-ITDs at diagnosis and 46 had a single FLT3-ITD at diagnosis. Overall, we observed a simplification of the FLT3-ITD repertoire upon relapse with the persistence of at least one initial clone in 8/12 [67%] and 24/36 [67%] pts with multiple ITDs receiving MIDO+ICT and ICT alone respectively. In relapsed pts who initially had a single FLT3-ITD clone at diagnosis, the addition of MIDO to ICT was associated with a higher rate of FLT3-ITD negativity compared to pts receiving ICT alone (6/13 [46%] vs 5/33 [15%]; P = 0.05) ( Figure 2). Interestingly, among 21 pts with primary refractory AML, we observed that FLT3-ITD mutation status became negative in 5/8 pts (62%) and 2/13 pts (15%) after induction with MIDO+ICT and ICT alone respectively. We then compared the initial characteristics between retained and lost FLT3-ITDs at AML relapse. Lost FLT3-ITDs had significantly lower AR than retained clones in both treatment groups. In order to limit the impact of sample dilution on the allele burden, we defined adjusted variant allele frequencies (VAFs) as the VAFs of FLT3-ITDs normalized to the VAFs of NPM1 mutations, whenever applicable. In so doing, we observed that adjusted VAFs of retained FLT3-ITDs increased at relapse, regardless of the treatment group (adjusted VAF, diagnosis vs relapse: 0.28 vs 0.86 and 0.88 vs 1.6 in the MIDO+ICT group and ICT alone group; P = 2.3e-03 and P = 2.4e-04). Importantly, an adjusted VAF higher than 1 was strongly suggestive of a homozygous state of FLT3-ITD. Such situation was found to be more prevalent at relapse in both treatment groups. Besides the selection of a dominant FLT3-ITD clone, other relapse-related changes including the acquisition of additional mutations will be presented. Conclusion: Our study of the clonal dynamics of AML with FLT3-ITD mutations provides insights into the mechanisms underlying therapy escape. Our data suggest that clonal interference characterized by multiple FLT3-ITD clones is associated with a greater ability to select a FLT3-ITD-positive clone at relapse in pts receiving MIDO+ICT. Although the addition of MIDO to ICT increases the probability of eradicating a single FLT3-ITD clone, FLT3-ITD+ relapses remain common following this combination, often with the selection of homozygous FLT3-ITD clones and/or the emergence of new mutations. Finally, our data in refractory situation emphasize the need to reassess mutational status at each stage of progression before implementing targeted therapy.

Frequent, infinitesimal bottlenecks maximize the rate of microbial adaptation.

Serial passaging is a fundamental technique in experimental evolution. The choice of bottleneck severity and frequency poses a dilemma: longer growth periods allow beneficial mutants to arise and grow over more generations, but simultaneously necessitate more severe bottlenecks with a higher risk of those same mutations being lost. Short growth periods require less severe bottlenecks, but come at the cost of less time between transfers for beneficial mutations to establish. The standard laboratory protocol of 24-h growth cycles with severe bottlenecking has logistical advantages for the experimenter but limited theoretical justification. Here we demonstrate that contrary to standard practice, the rate of adaptive evolution is maximized when bottlenecks are frequent and small, indeed infinitesimally so in the limit of continuous culture. This result derives from revising key assumptions underpinning previous theoretical work, notably changing the metric of optimization from adaptation per serial transfer to per experiment runtime. We also show that adding resource constraints and clonal interference to the model leaves the qualitative results unchanged. Implementing these findings will require liquid-handling robots to perform frequent bottlenecks, or chemostats for continuous culture. Further innovation in and adoption of these technologies has the potential to accelerate the rate of discovery in experimental evolution.

Long-range linkage effects in adapting sexual populations

In sexual populations, closely-situated genes have linked evolutionary fates, while genes spaced far in genome are commonly thought to evolve independently due to recombination. In the case where evolution depends essentially on supply of new mutations, this assumption has been confirmed by mathematical modeling. Here I examine it in the case of pre-existing genetic variation, where mutation is not important. A haploid population with N genomes, L loci, a fixed selection coefficient, and a small initial frequency of beneficial alleles {f}_{0} is simulated by a Monte-Carlo algorithm. When the number of loci, L, is larger than a critical value of {text{4log}}^{2} left( {Nf_{0} } right), simulation demonstrates a host of linkage effects that decrease neither with the distance between loci nor the number of recombination crossovers. Due to clonal interference, the beneficial alleles become extinct at a fraction of loci 1-2mathrm{log}left(N{f}_{0}right)/{L}^{0.5}. Due to a genetic background effect, the substitution rate varies broadly between loci, with the fastest value exceeding the one-locus limit by the factor of {[{L}^{0.5}/mathrm{log}left(Nsright)]}^{0.75}. Thus, the far-situated parts of a long genome in a sexual population do not evolve as independent blocks. A potential link between these findings and the emergence of new Variants of Concern of SARS-CoV-2 is discussed.

Mathematical Modeling of Clonal Interference by Density-Dependent Selection in Heterogeneous Cancer Cell Lines.

Many cancer cell lines are aneuploid and heterogeneous, with multiple karyotypes co-existing within the same cell line. Karyotype heterogeneity has been shown to manifest phenotypically, thus affecting how cells respond to drugs or to minor differences in culture media. Knowing how to interpret karyotype heterogeneity phenotypically would give insights into cellular phenotypes before they unfold temporally. Here, we re-analyzed single cell RNA (scRNA) and scDNA sequencing data from eight stomach cancer cell lines by placing gene expression programs into a phenotypic context. Using live cell imaging, we quantified differences in the growth rate and contact inhibition between the eight cell lines and used these differences to prioritize the transcriptomic biomarkers of the growth rate and carrying capacity. Using these biomarkers, we found significant differences in the predicted growth rate or carrying capacity between multiple karyotypes detected within the same cell line. We used these predictions to simulate how the clonal composition of a cell line would change depending on density conditions during in-vitro experiments. Once validated, these models can aid in the design of experiments that steer evolution with density-dependent selection.

Deterministic evolution and stringent selection during preneoplasia

The earliest events during human tumour initiation, although poorly characterized, may hold clues to malignancy detection and prevention1. Here we model occult preneoplasia by biallelic inactivation of TP53, a common early event in gastric cancer, in human gastric organoids. Causal relationships between this initiating genetic lesion and resulting phenotypes were established using experimental evolution in multiple clonally derived cultures over 2 years. TP53 loss elicited progressive aneuploidy, including copy number alterations and structural variants prevalent in gastric cancers, with evident preferred orders. Longitudinal single-cell sequencing of TP53-deficient gastric organoids similarly indicates progression towards malignant transcriptional programmes. Moreover, high-throughput lineage tracing with expressed cellular barcodes demonstrates reproducible dynamics whereby initially rare subclones with shared transcriptional programmes repeatedly attain clonal dominance. This powerful platform for experimental evolution exposes stringent selection, clonal interference and a marked degree of phenotypic convergence in premalignant epithelial organoids. These data imply predictability in the earliest stages of tumorigenesis and show evolutionary constraints and barriers to malignant transformation, with implications for earlier detection and interception of aggressive, genome-instable tumours.

Identifying Targets of Selection in Laboratory Evolution Experiments.

Adaptive evolution navigates a balance between chance and determinism. The stochastic processes of mutation and drift generate phenotypic variation; however, once mutations reach an appreciable frequency in the population, their fate is governed by the deterministic action of selection, enriching for favorable genotypes and purging the less-favorable ones. The net result is that replicate populations will traverse similar-but not identical-pathways to higher fitness. This parallelism in evolutionary outcomes can be leveraged to identify the genes and pathways under selection. However, distinguishing between beneficial and neutral mutations is challenging because many beneficial mutations will be lost due to drift and clonal interference, and many neutral (and even deleterious) mutations will fix by hitchhiking. Here, we review the best practices that our laboratory uses to identify genetic targets of selection from next-generation sequencing data of evolved yeast populations. The general principles for identifying the mutations driving adaptation will apply more broadly.

MSphere of Influence: Deterministic and Stochastic Processes Drive Microbial Evolution.

Aleeza C. Gerstein works in the field of microbial evolutionary genetics in human fungal pathogens. In this mSphere of Influence article, she reflects on how the papers "Experimental tests of the roles of adaptation, chance, and history in evolution" by Travisano et al. (1995) and "Pervasive genetic hitchhiking and clonal interference in forty evolving yeast populations" by Lang et al. (2013) provided her with a framework for thinking about the competing stochastic and deterministic evolutionary forces that act on microbial populations at the genotypic and phenotypic levels to cause both parallelism and divergence under different scenarios.

Estimating linkage disequilibrium and selection from allele frequency trajectories.

Genetic sequences collected over time provide an exciting opportunity to study natural selection. In such studies, it is important to account for linkage disequilibrium to accurately measure selection and to distinguish between selection and other effects that can cause changes in allele frequencies, such as genetic hitchhiking or clonal interference. However, most high-throughput sequencing methods cannot directly measure linkage due to short-read lengths. Here we develop a simple method to estimate linkage disequilibrium from time-series allele frequencies. This reconstructed linkage information can then be combined with other inference methods to infer the fitness effects of individual mutations. Simulations show that our approach reliably outperforms inference that ignores linkage disequilibrium and, with sufficient sampling, performs similarly to inference using the true linkage information. We also introduce two regularization methods derived from random matrix theory that help to preserve its performance under limited sampling effects. Overall, our method enables the use of linkage-aware inference methods even for data sets where only allele frequency time series are available.

Antibiotic Cycling Affects Resistance Evolution Independently of Collateral Sensitivity.

Antibiotic cycling has been proposed as a promising approach to slow down resistance evolution against currently employed antibiotics. It remains unclear, however, to which extent the decreased resistance evolution is the result of collateral sensitivity, an evolutionary trade-off where resistance to one antibiotic enhances the sensitivity to the second, or due to additional effects of the evolved genetic background, in which mutations accumulated during treatment with a first antibiotic alter the emergence and spread of resistance against a second antibiotic via other mechanisms. Also, the influence of antibiotic exposure patterns on the outcome of drug cycling is unknown. Here, we systematically assessed the effects of the evolved genetic background by focusing on the first switch between two antibiotics against Salmonella Typhimurium, with cefotaxime fixed as the first and a broad variety of other drugs as the second antibiotic. By normalizing the antibiotic concentrations to eliminate the effects of collateral sensitivity, we demonstrated a clear contribution of the evolved genetic background beyond collateral sensitivity, which either enhanced or reduced the adaptive potential depending on the specific drug combination. We further demonstrated that the gradient strength with which cefotaxime was applied affected both cefotaxime resistance evolution and adaptation to second antibiotics, an effect that was associated with higher levels of clonal interference and reduced cost of resistance in populations evolved under weaker cefotaxime gradients. Overall, our work highlights that drug cycling can affect resistance evolution independently of collateral sensitivity, in a manner that is contingent on the antibiotic exposure pattern.

Mutational Switch-Backs Can Accelerate Evolution of Francisella to a Combination of Ciprofloxacin and Doxycycline.

Combination antimicrobial therapy has been considered a promising strategy to combat the evolution of antimicrobial resistance. Francisella tularensis is the causative agent of tularemia and in addition to being found in the nature, is recognized as a threat agent that requires vigilance. We investigated the evolutionary outcome of adapting the Live Vaccine Strain (LVS) of F. tularensis subsp. holarctica to two non-interacting drugs, ciprofloxacin and doxycycline, individually, sequentially, and in combination. Despite their individual efficacies and independence of mechanisms, evolution to the combination arose on a shorter time scale than evolution to the two drugs sequentially. We conducted a longitudinal mutational analysis of the populations evolving to the drug combination, genetically reconstructed the identified evolutionary pathway, and carried out biochemical validation. We discovered that, after the appearance of an initial weak generalist mutation (FupA/B), each successive mutation alternated between adaptation to one drug or the other. In combination, these mutations allowed the population to more efficiently ascend the fitness peak through a series of evolutionary switch-backs. Clonal interference, weak pleiotropy, and positive epistasis also contributed to combinatorial evolution. This finding suggests that the use of this non-interacting drug pair against F. tularensis may render both drugs ineffective because of mutational switch-backs that accelerate evolution of dual resistance.