The Impact of Macroscopic Epistasis on Long-Term Evolutionary Dynamics

Genetic interactions can strongly influence the fitness effects of individual mutations, yet the impact of these epistatic interactions on evolutionary dynamics remains poorly understood. Here we investigate the evolutionary role of epistasis over 50,000 generations in a well-studied laboratory evolution experiment in Escherichia coli. The extensive duration of this experiment provides a unique window into the effects of epistasis during long-term adaptation to a constant environment. Guided by analytical results in the weak-mutation limit, we develop a computational framework to assess the compatibility of a given epistatic model with the observed patterns of fitness gain and mutation accumulation through time. We find that a decelerating fitness trajectory alone provides little power to distinguish between competing models, including those that lack any direct epistatic interactions between mutations. However, when combined with the mutation trajectory, these observables place strong constraints on the set of possible models of epistasis, ruling out many existing explanations of the data. Instead, we find that the data are consistent with a “two-epoch” model of adaptation, in which an initial burst of diminishing-returns epistasis is followed by a steady accumulation of mutations under a constant distribution of fitness effects. Our results highlight the need for additional DNA sequencing of these populations, as well as for more sophisticated models of epistasis that are compatible with all of the experimental data.     

EPISTATIC INTERACTIONS DETERMINE THE MUTATIONAL PATHWAYS AND COEXISTENCE OF LINEAGES IN CLONAL ESCHERICHIA COLI POPULATIONS

Understanding how diversity emerges in a single niche is not fully understood. Rugged fitness landscapes and epistasis between beneficial mutations could explain coexistence among emerging lineages. To provide an experimental test of this notion, we investigated epistasis among four pleiotropic mutations in rpoS, mglD, malT, and hfq present in two coexisting lineages that repeatedly fixed in experimental populations of Escherichia coli. The mutations were transferred into the ancestral background individually or in combination of double or triple alleles. The combined competitive fitness of two or three beneficial mutations from the same lineage was consistently lower than the sum of the competitive fitness of single mutants—a clear indication of negative epistasis within lineages. We also found sign epistasis (i.e., the combined fitness of two beneficial mutations lower than the ancestor), not only from two different lineages (i.e., hfq and rpoS) but also from the same lineage (i.e., mglD and malT). The sign epistasis between loci of different lineages indeed indicated a rugged fitness landscape, providing an epistatic explanation for the coexistence of distinct rpoS and hfq lineages in evolving populations. The negative and sign epistasis between beneficial mutations within the same lineage can further explain the order of mutation acquisition.     

Selection history and epistatic interactions impact dynamics of adaptation to novel environmental stresses

In rapidly changing environments, selection history may impact the dynamics of adaptation. Mutations selected in one environment may result in pleiotropic fitness trade-offs in subsequent novel environments, slowing the rates of adaptation. Epistatic interactions between mutations selected in sequential stressful environments may slow or accelerate subsequent rates of adaptation, depending on the nature of that interaction. We explored the dynamics of adaptation during sequential exposure to herbicides with different modes of action in Chlamydomonas reinhardtii. Evolution of resistance to two of the herbicides was largely independent of selection history. For carbetamide, previous adaptation to other herbicide modes of action positively impacted the likelihood of adaptation to this herbicide. Furthermore, while adaptation to all individual herbicides was associated with pleiotropic fitness costs in stress-free environments, we observed that accumulation of resistance mechanisms was accompanied by a reduction in overall fitness costs. We suggest that antagonistic epistasis may be a driving mechanism that enables populations to more readily adapt in novel environments. These findings highlight the potential for sequences of xenobiotics to facilitate the rapid evolution of multiple-drug and -pesticide resistance, as well as the potential for epistatic interactions between adaptive mutations to facilitate evolutionary rescue in rapidly changing environments.     

Evidence that Adaptation in Drosophila Is Not Limited by Mutation at Single Sites

Adaptation in eukaryotes is generally assumed to be mutation-limited because of small effective population sizes. This view is difficult to reconcile, however, with the observation that adaptation to anthropogenic changes, such as the introduction of pesticides, can occur very rapidly. Here we investigate adaptation at a key insecticide resistance locus (Ace) in Drosophila melanogaster and show that multiple simple and complex resistance alleles evolved quickly and repeatedly within individual populations. Our results imply that the current effective population size of modern D. melanogaster populations is likely to be substantially larger (≥100-fold) than commonly believed. This discrepancy arises because estimates of the effective population size are generally derived from levels of standing variation and thus reveal long-term population dynamics dominated by sharp—even if infrequent—bottlenecks. The short-term effective population sizes relevant for strong adaptation, on the other hand, might be much closer to census population sizes. Adaptation in Drosophila may therefore not be limited by waiting for mutations at single sites, and complex adaptive alleles can be generated quickly without fixation of intermediate states. Adaptive events should also commonly involve the simultaneous rise in frequency of independently generated adaptive mutations. These so-called soft sweeps have very distinct effects on the linked neutral polymorphisms compared to the standard hard sweeps in mutation-limited scenarios. Methods for the mapping of adaptive mutations or association mapping of evolutionarily relevant mutations may thus need to be reconsidered.     

Genomic signatures of strain selection and enhancement in Bacillus atrophaeus var. globigii, a historical biowarfare simulant

Background

Despite the decades-long use of Bacillus atrophaeus var. globigii (BG) as a simulant for biological warfare (BW) agents, knowledge of its genome composition is limited. Furthermore, the ability to differentiate signatures of deliberate adaptation and selection from natural variation is lacking for most bacterial agents. We characterized a lineage of BGwith a long history of use as a simulant for BW operations, focusing on classical bacteriological markers, metabolic profiling and whole-genome shotgun sequencing (WGS).

Results

Archival strains and two “present day” type strains were compared to simulant strains on different laboratory media. Several of the samples produced multiple colony morphotypes that differed from that of an archival isolate. To trace the microevolutionary history of these isolates, we obtained WGS data for several archival and present-day strains and morphotypes. Bacillus-wide phylogenetic analysis identified B. subtilis as the nearest neighbor to B. atrophaeus. The genome of B. atrophaeus is, on average, 86% identical to B. subtilis on the nucleotide level. WGS of variants revealed that several strains were mixed but highly related populations and uncovered a progressive accumulation of mutations among the “military” isolates. Metabolic profiling and microscopic examination of bacterial cultures revealed enhanced growth of “military” isolates on lactate-containing media, and showed that the “military” strains exhibited a hypersporulating phenotype.

Conclusions

Our analysis revealed the genomic and phenotypic signatures of strain adaptation and deliberate selection for traits that were desirable in a simulant organism. Together, these results demonstrate the power of whole-genome and modern systems-level approaches to characterize microbial lineages to develop and validate forensic markers for strain discrimination and reveal signatures of deliberate adaptation.     

Evolution after Introduction of a Novel Metabolic Pathway Consistently Leads to Restoration of Wild-Type Physiology

Organisms cope with physiological stressors through acclimatizing mechanisms in the short-term and adaptive mechanisms over evolutionary timescales. During adaptation to an environmental or genetic perturbation, beneficial mutations can generate numerous physiological changes: some will be novel with respect to prior physiological states, while others might either restore acclimatizing responses to a wild-type state, reinforce them further, or leave them unchanged. We examined the interplay of acclimatizing and adaptive responses at the level of global gene expression in Methylobacterium extorquens AM1 engineered with a novel central metabolism. Replacing central metabolism with a distinct, foreign pathway resulted in much slower growth than wild-type. After 600 generations of adaptation, however, eight replicate populations founded from this engineered ancestor had improved up to 2.5-fold. A comparison of global gene expression in wild-type, engineered, and all eight evolved strains revealed that the vast majority of changes during physiological adaptation effectively restored acclimatizing processes to wild-type expression states. On average, 93% of expression perturbations from the engineered strain were restored, with 70% of these occurring in perfect parallel across all eight replicate populations. Novel changes were common but typically restricted to one or a few lineages, and reinforcing changes were quite rare. Despite this, cases in which expression was novel or reinforced in parallel were enriched for loci harboring beneficial mutations. One case of parallel, reinforced changes was the pntAB transhydrogenase that uses NADH to reduce NADP⁺ to NADPH. We show that PntAB activity was highly correlated with the restoration of NAD(H) and NADP(H) pools perturbed in the engineered strain to wild-type levels, and with improved growth. These results suggest that much of the evolved response to genetic perturbation was a consequence rather than a cause of adaptation and that physiology avoided “reinventing the wheel” by restoring acclimatizing processes to the pre-stressed state.     

Characterization of the RpoS Status of Clinical Isolates of Salmonella enterica

The stationary-phase-inducible sigma factor, σ^S (RpoS), is the master regulator of the general stress response in Salmonella and is required for virulence in mice. rpoS mutants can frequently be isolated from highly passaged laboratory strains of Salmonella. We examined the rpoS status of 116 human clinical isolates of Salmonella, including 41 Salmonella enterica serotype Typhi strains isolated from blood, 38 S. enterica serotype Typhimurium strains isolated from blood, and 37 Salmonella serotype Typhimurium strains isolated from feces. We examined the abilities of these strains to produce the σ^S protein, to express RpoS-dependent catalase activity, and to resist to oxidative stress in the stationary phase of growth. We also carried out complementation experiments with a cloned wild-type rpoS gene. Our results showed that 15 of the 41 Salmonella serotype Typhi isolates were defective in RpoS. We sequenced the rpoS allele of 12 strains. This led to identification of small insertions, deletions, and point mutations resulting in premature stop codons or affecting regions 1 and 2 of σ^S, showing that the rpoS mutations are not clonal. Thus, mutant rpoS alleles can be found in freshly isolated clinical strains of Salmonella serotype Typhi, and they may affect virulence properties. Interestingly however, no rpoS mutants were found among the 75 Salmonella serotype Typhimurium isolates. Strains that differed in catalase activity and resistance to hydrogen peroxide were found, but the differences were not linked to the rpoS status. This suggests that Salmonella serotype Typhimurium rpoS mutants are counterselected because rpoS plays a role in the pathogenesis of Salmonella serotype Typhimurium in humans or in the transmission cycle of the disease.     

A Genetic Incompatibility Accelerates Adaptation in Yeast

During mismatch repair (MMR) MSH proteins bind to mismatches that form as the result of DNA replication errors and recruit MLH factors such as Mlh1-Pms1 to initiate excision and repair steps. Previously, we identified a negative epistatic interaction involving naturally occurring polymorphisms in the MLH1 and PMS1 genes of baker’s yeast. Here we hypothesize that a mutagenic state resulting from this negative epistatic interaction increases the likelihood of obtaining beneficial mutations that can promote adaptation to stress conditions. We tested this by stressing yeast strains bearing mutagenic (incompatible) and non-mutagenic (compatible) mismatch repair genotypes. Our data show that incompatible populations adapted more rapidly and without an apparent fitness cost to high salt stress. The fitness advantage of incompatible populations was rapid but disappeared over time. The fitness gains in both compatible and incompatible strains were due primarily to mutations in PMR1 that appeared earlier in incompatible evolving populations. These data demonstrate a rapid and reversible role (by mating) for genetic incompatibilities in accelerating adaptation in eukaryotes. They also provide an approach to link experimental studies to observational population genomics.     

Functional Heterogeneity of RpoS in Stress Tolerance of Enterohemorrhagic Escherichia coli Strains

The stationary-phase sigma factor (RpoS) regulates many cellular responses to environmental stress conditions such as heat, acid, and alkali shocks. On the other hand, mutations at the rpoS locus have frequently been detected among pathogenic as well as commensal strains of Escherichia coli. The objective of this study was to perform a functional analysis of the RpoS-mediated stress responses of enterohemorrhagic E. coli strains from food-borne outbreaks. E. coli strains belonging to serotypes O157:H7, O111:H11, and O26:H11 exhibited polymorphisms for two phenotypes widely used to monitor rpoS mutations, heat tolerance and glycogen synthesis, as well as for two others, alkali tolerance and adherence to Caco-2 cells. However, these strains synthesized the oxidative acid resistance system through an rpoS-dependent pathway. During the transition from mildly acidic growth conditions (pH 5.5) to alkaline stress (pH 10.2), cell survival was dependent on rpoS functionality. Some strains were able to overcome negative regulation by RpoS and induced higher β-galactosidase activity without compromising their acid resistance. There were no major differences in the DNA sequences in the rpoS coding regions among the tested strains. The heterogeneity of rpoS-dependent phenotypes observed for stress-related phenotypes was also evident in the Caco-2 cell adherence assay. Wild-type O157:H7 strains with native rpoS were less adherent than rpoS-complemented counterpart strains, suggesting that rpoS functionality is needed. These results show that some pathogenic E. coli strains can maintain their acid tolerance capability while compromising other RpoS-dependent stress responses. Such adaptation processes may have significant impact on a pathogen's survival in food processing environments, as well in the host's stomach and intestine.     

Parallel adaptation to climate above the 35th parallel

Independent or parallel evolution of similar traits is key to understanding the genetics and limitations of adaptation. Adaptation from the same genetic changes in different populations defines parallel evolution. Such genetic changes can derive from standing ancestral variation or de novo mutations and excludes instances of adaptive introgression. In this issue of Molecular Ecology, Walden et al.(2020) investigate the scale of parallel climate adaptation from standing genetic variation between two North American Arabidopsis lyrata lineages, each formed by a distinct evolutionary history during the last glacial cycle. By identifying adaptive variants correlated with three ecologically significant climatic gradients, they show that instead of the same genetic variants or even genes, parallel evolution is only observed at the level of biological processes. The evolution of independent adaptive variants to climate in two genetically close lineages is explained by their different post‐glacial demographic histories. Separate glacial refugia and strong population bottlenecks were probably sufficient to change the landscape of shared allele frequencies, hindering the possibility of parallel evolution.     

Adaptation to High Ethanol Reveals Complex Evolutionary Pathways

Tolerance to high levels of ethanol is an ecologically and industrially relevant phenotype of microbes, but the molecular mechanisms underlying this complex trait remain largely unknown. Here, we use long-term experimental evolution of isogenic yeast populations of different initial ploidy to study adaptation to increasing levels of ethanol. Whole-genome sequencing of more than 30 evolved populations and over 100 adapted clones isolated throughout this two-year evolution experiment revealed how a complex interplay of de novo single nucleotide mutations, copy number variation, ploidy changes, mutator phenotypes, and clonal interference led to a significant increase in ethanol tolerance. Although the specific mutations differ between different evolved lineages, application of a novel computational pipeline, PheNetic, revealed that many mutations target functional modules involved in stress response, cell cycle regulation, DNA repair and respiration. Measuring the fitness effects of selected mutations introduced in non-evolved ethanol-sensitive cells revealed several adaptive mutations that had previously not been implicated in ethanol tolerance, including mutations in PRT1, VPS70 and MEX67. Interestingly, variation in VPS70 was recently identified as a QTL for ethanol tolerance in an industrial bio-ethanol strain. Taken together, our results show how, in contrast to adaptation to some other stresses, adaptation to a continuous complex and severe stress involves interplay of different evolutionary mechanisms. In addition, our study reveals functional modules involved in ethanol resistance and identifies several mutations that could help to improve the ethanol tolerance of industrial yeasts.     

Evolution of polygenic traits under global vs local adaptation

Observations about the number, frequency, effect size, and genomic distribution of alleles associated with complex traits must be interpreted in light of evolutionary process. These characteristics, which constitute a trait’s genetic architecture, can dramatically affect evolutionary outcomes in applications from agriculture to medicine, and can provide a window into how evolution works. Here, I review theoretical predictions about the evolution of genetic architecture under spatially homogeneous, global adaptation as compared with spatially heterogeneous, local adaptation. Due to the tension between divergent selection and migration, local adaptation can favor “concentrated” genetic architectures that are enriched for alleles of larger effect, clustered in a smaller number of genomic regions, relative to expectations under global adaptation. However, the evolution of such architectures may be limited by many factors, including the genotypic redundancy of the trait, mutation rate, and temporal variability of environment. I review the circumstances in which predictions differ for global vs local adaptation and discuss where progress can be made in testing hypotheses using data from natural populations and lab experiments. As the field of comparative population genomics expands in scope, differences in architecture among traits and species will provide insights into how evolution works, and such differences must be interpreted in light of which kind of selection has been operating.     

Pleiotropic Effects of a Mitochondrial–Nuclear Incompatibility Depend upon the Accelerating Effect of Temperature in Drosophila

Interactions between mitochondrial and nuclear gene products that underlie eukaryotic energy metabolism can cause the fitness effects of mutations in one genome to be conditional on variation in the other genome. In ectotherms, the effects of these interactions are likely to depend upon the thermal environment, because increasing temperature accelerates molecular rates. We find that temperature strongly modifies the pleiotropic phenotypic effects of an incompatible interaction between a Drosophila melanogaster polymorphism in the nuclear-encoded, mitochondrial tyrosyl-transfer (t)RNA synthetase and a D. simulans polymorphism in the mitochondrially encoded tRNA^Tyr. The incompatible mitochondrial–nuclear genotype extends development time, decreases larval survivorship, and reduces pupation height, indicative of decreased energetic performance. These deleterious effects are ameliorated when larvae develop at 16° and exacerbated at warmer temperatures, leading to complete sterility in both sexes at 28°. The incompatible genotype has a normal metabolic rate at 16° but a significantly elevated rate at 25°, consistent with the hypothesis that inefficient energy metabolism extends development in this genotype at warmer temperatures. Furthermore, the incompatibility decreases metabolic plasticity of larvae developed at 16°, indicating that cooler development temperatures do not completely mitigate the deleterious effects of this genetic interaction. Our results suggest that the epistatic fitness effects of metabolic mutations may generally be conditional on the thermal environment. The expression of epistatic interactions in some environments, but not others, weakens the efficacy of selection in removing deleterious epistatic variants from populations and may promote the accumulation of incompatibilities whose fitness effects will depend upon the environment in which hybrids occur.     

Adaptation of gastrointestinal nematode parasites to host genotype: single locus simulation models

Background

Breeding livestock for improved resistance to disease is an increasingly important selection goal. However, the risk of pathogens adapting to livestock bred for improved disease resistance is difficult to quantify. Here, we explore the possibility of gastrointestinal worms adapting to sheep bred for low faecal worm egg count using computer simulation. Our model assumes sheep and worm genotypes interact at a single locus, such that the effect of an A allele in sheep is dependent on worm genotype, and the B allele in worms is favourable for parasitizing the A allele sheep but may increase mortality on pasture. We describe the requirements for adaptation and test if worm adaptation (1) is slowed by non-genetic features of worm infections and (2) can occur with little observable change in faecal worm egg count.

Results

Adaptation in worms was found to be primarily influenced by overall worm fitness, viz. the balance between the advantage of the B allele during the parasitic stage in sheep and its disadvantage on pasture. Genetic variation at the interacting locus in worms could be from de novo or segregating mutations, but de novo mutations are rare and segregating mutations are likely constrained to have (near) neutral effects on worm fitness. Most other aspects of the worm infection we modelled did not affect the outcomes. However, the host-controlled mechanism to reduce faecal worm egg count by lowering worm fecundity reduced the selection pressure on worms to adapt compared to other mechanisms, such as increasing worm mortality. Temporal changes in worm egg count were unreliable for detecting adaptation, despite the steady environment assumed in the simulations.

Conclusions

Adaptation of worms to sheep selected for low faecal worm egg count requires an allele segregating in worms that is favourable in animals with improved resistance but less favourable in other animals. Obtaining alleles with this specific property seems unlikely. With support from experimental data, we conclude that selection for low faecal worm egg count should be stable over a short time frame (e.g. 20 years). We are further exploring model outcomes with multiple loci and comparing outcomes to other control strategies.     

The Underlying Structure of Adaptation under Strong Selection in 12 Experimental Yeast Populations

The aims of this study were to determine (i) whether adaptation under strong selection occurred through mutations in a narrow target of one or a few nucleotide sites or a broad target of numerous sites and (ii) whether the programs of adaptation previously observed from three experimental populations were unique or shared among populations that underwent parallel evolution. We used archived population samples from a previous study, representing 500 generations of experimental evolution in 12 populations under strong selection, 6 populations in a high-salt environment and 6 populations in a low-glucose environment. Each set of six populations included four with sexual reproduction and two with exclusively asexual reproduction. Populations were sampled as resequenced genomes of 115 individuals and as bulk samples from which frequencies of mutant alleles were estimated. In a high-salt environment, a broad target of 11 mutations within the proton exporter, PMA1, was observed among the six populations, in addition to expansions of the ENA gene cluster. This pattern was shared among populations that underwent parallel evolution. In a low-glucose environment, two programs of adaptation were observed. The originally observed pattern of mutation in MDS3/MKT1 in population M8 was a narrow target of a single nucleotide, unique to this population. Among the other five populations, the three mutations were shared in a broad target, sensing/signaling genes RAS1 and RAS2. RAS1/RAS2 mutations were not observed in the high-salt populations; PMA1 mutations were observed only in a high-salt environment.     

Fur-dependent detoxification of organic acids by rpoS mutants during prolonged incubation under aerobic, phosphate starvation conditions

The activity of amino acid-dependent acid resistance systems allows Escherichia coli to survive during prolonged incubation under phosphate (P_i) starvation conditions. We show in this work that rpoS-null mutants incubated in the absence of any amino acid survived during prolonged incubation under aerobic, P_i starvation conditions. Whereas rpoS⁺ cells incubated with glutamate excreted high levels of acetate, rpoS mutants grew on acetic acid. The characteristic metabolism of rpoS mutants required the activity of Fur (ferric uptake regulator) in order to decrease the synthesis of the small RNA RyhB that might otherwise inhibit the synthesis of iron-rich proteins. We propose that RpoS (σ^S) and the small RNA RyhB contribute to decrease the synthesis of iron-rich proteins required for the activity of the tricarboxylic acid (TCA) cycle, which redirects the metabolic flux toward the production of acetic acid at the onset of stationary phase in rpoS⁺ cells. In contrast, Fur activity, which represses ryhB, and the lack of RpoS activity allow a substantial activity of the TCA cycle to continue in stationary phase in rpoS mutants, which decreases the production of acetic acid and, eventually, allows growth on acetic acid and P_i excreted into the medium. These data may help explain the fact that a high frequency of E. coli rpoS mutants is found in nature.     

Detecting short spatial scale local adaptation and epistatic selection in climate‐related candidate genes in European beech (Fagus sylvatica) populations

Detecting signatures of selection in tree populations threatened by climate change is currently a major research priority. Here, we investigated the signature of local adaptation over a short spatial scale using 96 European beech (Fagus sylvatica L.) individuals originating from two pairs of populations on the northern and southern slopes of Mont Ventoux (south‐eastern France). We performed both single and multilocus analysis of selection based on 53 climate‐related candidate genes containing 546 SNPs. F_ST outlier methods at the SNP level revealed a weak signal of selection, with three marginally significant outliers in the northern populations. At the gene level, considering haplotypes as alleles, two additional marginally significant outliers were detected, one on each slope. To account for the uncertainty of haplotype inference, we averaged the Bayes factors over many possible phase reconstructions. Epistatic selection offers a realistic multilocus model of selection in natural populations. Here, we used a test suggested by Ohta based on the decomposition of the variance of linkage disequilibrium. Overall populations, 0.23% of the SNP pairs (haplotypes) showed evidence of epistatic selection, with nearly 80% of them being within genes. One of the between gene epistatic selection signals arose between an F_ST outlier and a nonsynonymous mutation in a drought response gene. Additionally, we identified haplotypes containing selectively advantageous allele combinations which were unique to high or low elevations and northern or southern populations. Several haplotypes contained nonsynonymous mutations situated in genes with known functional importance for adaptation to climatic factors.     

The rpoS Gene Is Predominantly Inactivated during Laboratory Storage and Undergoes Source-Sink Evolution in Escherichia coli Species

The rpoS gene codes for an alternative RNA polymerase sigma factor, which acts as a general regulator of the stress response. Inactivating alleles of rpoS in collections of natural Escherichia coli isolates have been observed at very variable frequencies, from less than 1% to more than 70% of strains. rpoS is easily inactivated in nutrient-deprived environments such as stab storage, which makes it difficult to determine the true frequency of rpoS inactivation in nature. We studied the evolutionary history of rpoS and compared it to the phylogenetic history of bacteria in two collections of 82 human commensal and extraintestinal E. coli strains. These strains were representative of the phylogenetic diversity of the species and differed only by their storage conditions. In both collections, the phylogenetic histories of rpoS and of the strains were congruent, indicating that horizontal gene transfer had not occurred at the rpoS locus, and rpoS was under strong purifying selection, with a ratio of the nonsynonymous mutation rate (Ka) to the synonymous substitution rate (Ks) substantially smaller than 1. Stab storage was associated with a high frequency of inactivating alleles, whereas almost no amino acid sequence variation was observed in RpoS in the collection studied directly after isolation of the strains from the host. Furthermore, the accumulation of variations in rpoS was typical of source-sink dynamics. In conclusion, rpoS is rarely inactivated in natural E. coli isolates within their mammalian hosts, probably because such strains rapidly become evolutionary dead ends. Our data should encourage bacteriologists to freeze isolates immediately and to avoid the use of stab storage.