Accelerated evolution in networked metapopulations of Pseudomonas aeruginosa

Natural populations are often spatially structured as metapopulations, meaning they are geographically subdivided and connected by migration. Examples include infectious diseases that transmit through contact networks among hosts (Leventhal et al., 2015), species ranges that can be discontiguous, especially at range edges where peripheral populations are connected via dispersal to a more well-connected central population (Sagarin et al., 2006; Pennington et al., 2021), and patients (along with their colonizing pathogens) that move among wards in a hospital (Myall et al., 2021). How metapopulation topology, the arrangement of connections among subpopulations, influences the dynamics of adaptation is not well understood. The central issues concern the impact of topology on rates of adaptation, the time to generate and fix beneficial mutations (Tkadlec et al., 2019; Yagoobi and Traulsen, 2021; Frean et al., 2013), and the population genetic mechanisms responsible.

The conventional view, based on a model of reproduction involving offspring replacing parents en masse in non-overlapping generations and leading to global competition among individuals, is that the topology of connections among subpopulations has little influence on rates of adaptation (Guillaume, 2011; Maruyama, 1970; Slatkin, 1981; Slatkin, 1985). Migration in these models tends to slow but not prevent adaptation across a metapopulation relative to a well-mixed, fully connected system when the rate of selection, s, exceeds that of migration, m. Rates of adaptation in spatially structured populations can be slowed further by clonal interference, the competition for fixation among independently arising beneficial mutations (Gordo and Campos, 2006), an effect that can be particularly strong in large populations because mutation supply rates, being the product of population size, N, and the genome-wide mutation rate, U, are often high in large populations.

By contrast, topology can strongly modulate the dynamics of adaptation in finite populations where individuals reproduce in overlapping generations and compete locally with each other, either decreasing or increasing rates of adaptation relative to a well-mixed system depending on the arrangement of subpopulations (Yagoobi and Traulsen, 2021; Lieberman et al., 2005; Pavlogiannis et al., 2018; Tkadlec et al., 2021; Hindersin and Traulsen, 2015). Much attention has been paid to how ‘star’ topologies, comprised of a central ‘hub’ connected to peripheral ‘leaf’ subpopulations, can increase the probability of fixation for beneficial mutations above that expected for a well-mixed population because this result is unexpected from standard models (Maruyama, 1970; Slatkin, 1981) and may reasonably capture features of many real-world scenarios such as contact networks of infectious diseases, the edge of species ranges, or patient transfers between regional care centres and large urban hospitals (Donker et al., 2017; Nekkab et al., 2020). Rates of adaptation, measured as the establishment time for beneficial mutations destined to fix, are usually slower in stars than well-mixed populations (Frean et al., 2013) but can be accelerated when mutation supply rates in each subpopulation (which is NU, as defined above, times m, the migration rate into a subpopulation) are <1 (Tkadlec et al., 2019; Yagoobi and Traulsen, 2021), possibly because rare beneficial mutations become concentrated in the hub and so are less likely to be stochastically lost when rare, a phenomenon called drift loss.

Empirical evidence on the impact of topology on rates of adaptation is limited. In microbial evolution experiments where population sizes are typically very large, various forms of spatial structure usually slow down adaptation relative to a well-mixed condition (Miralles et al., 1999; Habets et al., 2006; Habets et al., 2007; Perfeito et al., 2008; Kryazhimskiy et al., 2012; Bailey, 2021). The one exception is a study where faster adaptation occurred in structured compared to unstructured populations, a result attributed to the ability of the structured population to explore higher fitness peaks in a rugged fitness landscape (Nahum et al., 2015). The only study to evaluate the impact of topology per se on adaptation across a range of population sizes tracked an initially rare beneficial mutation introduced into one peripheral ‘leaf’ through both ‘star’ and well-mixed metapopulations (Chakraborty et al., 2023). The mutation spread more rapidly through star metapopulations than well-mixed populations when migration rates were low (m < 0.01%) and this difference could be exaggerated by reducing population sizes (from N ~ 10⁷ to 10⁵ CFU/ml) and biasing migration from leaves towards the hub. These results are consistent with a mechanism for acceleration involving a reduced probability of drift loss in stars: beneficial mutations rising to high frequency in their introduced subpopulation become concentrated in the hub via migration, allowing them to spread rapidly to other leaves (Chakraborty et al., 2023).

Is acceleration possible under more prolonged and open-ended evolution, where mutations arise naturally at any location across the metapopulation and compete with each other for fixation through clonal interference? Evolutionary graph theory, which is grounded in a model of reproduction that generates local competition among individuals, suggests the answer depends on the extent of clonal interference associated with migration (Frean et al., 2013; Sharma et al., 2023). Only when clonal interference is sufficiently low to allow single, initially rare beneficial mutations to gain access to the hub do we expect to see acceleration (Tkadlec et al., 2019; Yagoobi and Traulsen, 2021; Sharma et al., 2023). To test this prediction, we tracked rates of adaptation of the opportunistic pathogen Pseudomonas aeruginosa (PA14) in the presence of subinhibitory concentrations of the fluoroquinolone antibiotic ciprofloxacin for approximately 100 generations in 4-patch star and well-mixed populations across a range of mutation supply rates. We manipulate mutation supply rates in each subpopulation by adjusting both the effective population size and migration rate (Figure 1; see Materials and methods). Each treatment combination is replicated eight times.

Figure 1

Download asset Open asset

Our results are shown in Figures 2 and 3. Fitness, measured as both growth rate (r) and stationary phase density (K) relative to the ancestral strain, was faster in star metapopulations relative to well-mixed populations when effective population sizes were small (main effect of network, p = 0.06 and p < 0.01 for relative r and K, respectively; effect sizes (Cohen’s d for the difference between star and well-mixed treatments) r = 0.929, 95% CI: [−0.035 to 1.89]; K = 1.53, 95% CI: [0.586 to 2.47]), irrespective of migration rate (relative r: p = 0.73 and relative K: p = 0.97) and even after accounting for the removal of outliers (Figure 3—figure supplement 2). Larger metapopulations adapted faster than small ones, as expected (Wilke, 2004; Orr, 2000), but there was no effect of topology on rates of fitness increase at large effective population sizes (main effect of network treatment for relative r: p = 0.95 and relative K: p = 0.60) across the two migration rates (relative r: p = 0.42 and relative K: p = 0.51). These results demonstrate that star topologies can accelerate adaptation by as much as 1.5x over that of well-mixed metapopulations when mutation supply is low but not high.

Figure 2 with 1 supplement see all

Download asset Open asset

Figure 3 with 2 supplements see all

Download asset Open asset

We previously showed that a single beneficial mutation spreads more rapidly in stars than well-mixed populations when selection is strong relative to migration because a rare beneficial mutant is less likely to be lost due to drift in a newly colonized patch, consistent with the first mechanism (Chakraborty et al., 2023). Establishment times, being the time required to substitute a beneficial mutation that is destined to fix (Frean et al., 2013), may also be faster in stars than well-mixed populations because beneficial mutants experience reduced clonal interference owing to the fact that leaf subpopulations in stars exchange migrants only with the hub whereas all subpopulations – leaves and hub – share migrants equally in a well-mixed metapopulation. Reduced clonal interference could allow beneficial mutants arising in a leaf both to fix more rapidly there and to seed other leaves via the hub (Kuo et al., 2025). Adaptation could be further accelerated if reduced clonal interference in stars is accompanied by a lower probability of drift loss for rare beneficial mutants colonizing a new patch.

To evaluate these hypotheses, we first checked that clonal interference was in fact lower in stars than well-mixed metapopulations. We estimated the number and spectrum of mutations segregating at the end of our experiment by sequencing 24 evolved metapopulations (6 for each population size x topology combination). We achieved ~3000-fold sequencing depth and ~98.5% genome-wide coverage for each metapopulation which, after comparing the data with ancestral PA14 genomes (PA14 and PA14-LacZ), allowed us to detect all mutations segregating above ~5% frequency in each metapopulation. Figure 4 shows the average number of mutations segregating in each treatment. The results indicate that while all metapopulations in our experiment evolve in the clonal interference regime, large populations harbour more mutations on average than small populations (two-way ANOVA, main effect population size, p = 0.09), and well-mixed metapopulations support more mutations than star populations at both population sizes (two-way ANOVA, main effect of network treatment, p = 0.11, effect size of the difference between star and well-mixed mean averaged over both population sizes –0.678, 95% CI: [−1.56 to 0.203]). While these effects are not formally significant at the conventional level (p < 0.05), they reassure us that our experimental manipulations modulated the level of clonal interference in the expected directions.

Figure 4 with 2 supplements see all

Download asset Open asset

Reduced clonal interference could allow large effect beneficial mutants to increase in frequency more rapidly in stars relative to a well-mixed metapopulation. If these beneficial mutants are also less likely to be lost to drift during colonization of a new patch, large effect beneficial mutants should be enriched in low population size stars where we see accelerated adaptation. To test this prediction, we collected 48 clones from the last timepoint of each of the evolved metapopulations and assayed stationary phase density at 24 hr, K, as a proxy for fitness in the selection medium (Figure 5). Large effect clones are preferentially fixed in the well-mixed population compared to the star when population sizes are large (Figure 4, panels A, B, permutation K–S test, p < 10^–3 and p < 10^–3, for high and low migration rates, respectively, for 10,000 permutations), consistent with strong clonal interference biasing fixation towards large effect mutants. In small populations, by contrast, star metapopulations harboured more large effect clones compared to the well-mixed metapopulations across both high and low migration rates (Figure 4, panels C, D, permutation K–S test, p < 10^–3 and p < 0.05, for high and low migration rates, respectively, for 10,000 permutations). These results are further reinforced by inspecting the empirical cumulative distribution functions for K, which are right-shifted at small effective population sizes reflecting the predominance of larger effect mutants in stars relative to well-mixed populations (Figure 5—figure supplement 1).

Figure 5 with 1 supplement see all

Download asset Open asset

If the cause of acceleration is the preferential substitution of large effect mutations, then we should also expect to see higher levels of repeated, or parallel, evolution in small N stars than in any other treatment of our experiment. Parallel evolution is often taken to be an indicator of strong selection acting in a particular locus, as it is unlikely that the same genes or mutations would fix repeatedly by chance alone (Orr, 2005). Equally, it could be due to preferential substitution caused by the reduced probability of drift loss, and so higher fixation probability, among the rarer class of larger effect beneficial mutations available to selection (Orr, 2005; Chevin et al., 2010; Gillespie, 1984; Kassen and Bataillon, 2006; Fisher, 1930) together with the more rapid spread facilitated by reduced clonal interference. We tested this prediction by using our sequencing results to calculate the rate of repeated, or parallel, evolution for each mutated gene in our data set. We find that parallel evolution was higher in the small N star metapopulations compared with all others, an observation confirmed by the significant interaction between population size and network topology for three metrics of parallelism (p < 0.05; see methods). Closer inspection reveals star and well-mixed topologies do not differ in the extent of parallelism in large metapopulations but they do for small metapopulations, stars showing significantly higher parallelism than their well-mixed counterparts (Table 1), a result also reflected by higher frequency of all mutations (those appearing repeatedly and as singletons) in small stars compared to the small well-mixed metapopulations (Figure 4—figure supplement 2). As there is little reason to suspect the distribution of fitness effects among beneficial mutations available to selection differs systematically between star and well-mixed topologies for a given population size, the high levels of parallelism in the small N star must be associated with a higher fixation probability among the first mutations arising in the experiment. Given the over-representation of large effect beneficial mutants in small N star metapopulations, these results suggest accelerated adaptation is due to the combined effect of reduced clonal interference and large effect beneficial mutations having a higher probability of fixation because they avoid drift loss when rare in stars compared to well-mixed populations.

As expected from the strong selection pressure exerted by ciprofloxacin, we found many known resistance mutations circulating in moderate to high frequencies in these metapopulations (Figure 6). We identified frequent mutations (small indels and nonsynonymous SNPs) in the negative regulators and components of efflux pumps such as mexA and mexR (mexAB-oprM), mexS (MexEF-OprN), and nfxB (MexCD-OprJ) that are constitutively expressed when it is necessary to decrease intracellular concentration of ciprofloxacin (Poole, 2005; Richardot et al., 2016). We also uncovered mutations that prevent ciprofloxacin from binding to the DNA-modifying subunits of DNA gyrases (gyrA and gyrB) (Yoshida et al., 1990a; Yoshida et al., 1990b; Barnard and Maxwell, 2001). Not unexpectedly, our evolved metapopulations also harbour mutations in genes that do not confer resistance to ciprofloxacin but are global regulators of quorum sensing (lasR), c-di-GMP signalling and biofilm formation (wspA/F/R, morA), virulence and twitching motility (pilB/C/D/F) (Rumbaugh et al., 2000; Schuster and Greenberg, 2006; Williams and Cámara, 2009; Moradali et al., 2017; Chadha et al., 2022; Choy et al., 2004; Ha et al., 2014; Harrison et al., 2020; Katharios-Lanwermeyer et al., 2021; Ryder et al., 2007; Römling et al., 2013; Ha and O’Toole, 2015; Jenal et al., 2017; Ma et al., 2022). We and others commonly observe these same mutations in rich laboratory growth media, where their fitness advantage is thought to be associated with reducing the costs of metabolism (Mould et al., 2022; Schick et al., 2022).

Figure 6 with 2 supplements see all

Download asset Open asset

A last striking result from our genomic analyses is that metapopulations evolved under high and low mutation supply diverge in the repertoire of resistance mutations they accumulate over the evolutionary period. Notable examples include the enrichment of mexS and gyrA mutations exclusively in small and large metapopulations, respectively (binomial test, p < 0.005 and p < 0.05, respectively). We see the same pattern for putative non-resistance mutations as well: lasR and wspA being mutated in the large (binomial test, p < 0.005 and p < 0.05, for lasR and wspA, respectively) but not in the small metapopulations. Although we do not uncover any mutations that are network specific in the large metapopulations, there is a tendency for mutations in wspA and wspF to be enriched in the small star metapopulations (wspA and wspF mutations in six star compared to wspF mutations in two well-mixed metapopulations). Additionally, it is notable that mutations in nfxB are present in all (six out of six) small star but only 50% (three out of six) in the small well-mixed metapopulations sequenced, although the difference is not statistically significant (binomial test, p = 0.08). Furthermore, the average frequency of resistance-conferring mexS and nfxB mutations is modestly higher in small stars relative to small well-mixed metapopulations (Figure 4—figure supplement 1), although this effect is not formally significant (two-sample t-test p = 0.1585 and p = 0.2606, for nfxB and mexS, respectively, and p = 0.1729, Fisher’s method of combining multiple probabilities). This result, taken together with the observation that fewer mutations segregate in small stars compared to small well-mixed networks (Figure 4), lends further support to the idea that small stars lead to rapid enrichment of beneficial driver mutations.

Our genomic data is also consistent with the idea that mutation supply rate can bias the spectrum of mutations contributing to adaptation. Large populations with high mutation supply rates have access to a larger spectrum of mutations, and clonal interference during substitution can lead to preferential fixation of those with the largest fitness effects. Interestingly, however, there is little evidence that network topology itself biases the spectrum of mutations available to selection; rather, it is the preferential enrichment of large effect mutations that avoid drift loss in star metapopulations that governs the dynamics of molecular variation in these situations.

Our work provides direct, experimental evidence that the topology of connections in a metapopulation can alter the pace of adaptive evolution and the identity of the mutations responsible. Specifically, we have shown that small N star metapopulations, those characterized by satellite populations connected through a central hub, can accelerate adaptation over and above that observed in a well-mixed, fully connected setting. Acceleration is associated with an enrichment of large effect mutants and results in higher levels of gene-level parallelism compared to well-mixed populations of the same size. These results are consistent with an underlying mechanism involving the concentration of beneficial mutations in the hub by dispersal and facilitating their rapid spread to other leaves in the metapopulation. The contribution of reduced probability of drift loss and lower levels of clonal interference in accelerating adaptation is a topic that deserves further exploration.

That we observed acceleration in the small N stars is surprising because it does not fit with the two main models of reproduction used to track allele frequencies in metapopulations. The approach most common in population genetics, the Fisher–Wright model, assumes reproduction involves bulk replacement of parents by offspring in non-overlapping generations leading to global competition among individuals each generation. Metapopulation topology has little or no effect on fixation probabilities in this model, and the pace of adaptation is almost always fastest in well-mixed populations (Maruyama, 1970; Slatkin, 1981). An alternative approach, evolutionary graph theory, is based on the Moran model of individual reproduction and death, which leads to overlapping generations and local competition among neighbouring individuals. Moran-based models show that stars can increase the fixation probability of beneficial mutants relative to a well-mixed population, although at the cost of longer fixation times (Tkadlec et al., 2019; Frean et al., 2013; Lieberman et al., 2005). Our results show that both models fail to capture accurately the dynamics of adaptation on stars: in contrast to Fisher–Wright models, stars can adapt faster than well-mixed populations, whereas times to fixation appear to be much smaller than anticipated for the Moran process on graphs. Progress towards a general theory of adaptation in spatially structured populations will thus demand we rethink the central assumptions underpinning reproduction in both models.

Our results also bear on an old debate around how to appropriately conceptualize the role of spatial structure in adaptive evolution. One view, due to Fisher, 1930, effectively ignores spatial structure: selection across subpopulations is modelled as if it was occurring in a large, panmictic and freely recombining population, the fitness effect of any given allele being described by its average fitness improvement over the current wild type when tested against all possible genetic backgrounds from all subpopulations. An alternative view, attributed to Wright, 1932 and called the shifting balance model, treats spatial structure as an integral feature of adaptation: selection occurs independently in small, spatially separated subpopulations connected by migration. The fitness effect of an allele depends both on the genetic background in which it arises in a given subpopulation and on how effectively it outcompetes distinct beneficial mutants originating in other subpopulations. Our experimental results provide insight into the mechanisms driving this second phase of the shifting balance model, the competition among beneficial mutants derived from distinct subpopulations, by showing how, when mutation supply rates are low, the combined effects of reduced clonal interference and higher probability of avoiding drift loss can allow beneficial mutants to spread rapidly through a metapopulation.

How common is accelerated adaptation on star-like structures in more natural systems? This question remains difficult to answer. Many microbial communities occupy environments such as soil or hosts that are spatially structured (Dang and Lovell, 2016; Hall-Stoodley et al., 2004; Nadell et al., 2016; Kraemer and Boynton, 2017; García-Bayona and Comstock, 2018; Stacy et al., 2016; Hajishengallis et al., 2023; Werner et al., 2014), and some studies have shown that spatial structure following population bottlenecks imposed during host dissociation or biofilm dispersal can have profoundly different outcomes depending on the spatial structure of the population (Nadell et al., 2016; Abel et al., 2015; Steenackers et al., 2016). The fact that we only observed acceleration at the smallest population sizes in our experiment (N_e ~10⁵ individuals) suggests there could be an upper threshold of population size above which acceleration is unlikely to be observed. Nevertheless, the metapopulation structure of many large organisms may resemble that of a star, for example when range edge populations are sparse compared to the more abundant centre or due to habitat fragmentation (Sagarin et al., 2006; Pennington et al., 2021). If so, star topologies could be an explanation for the rapid spread of invasive species. Moreover, many scale-free networks, which often characterize contact networks in epidemiology, contain sub-structures resembling stars and are known to be weak amplifiers of selection (Leventhal et al., 2015; Lieberman et al., 2005). It is not inconceivable that metapopulation structures such as the star play a more important role in adaptive evolution than has been recognized up to now.

All sequence reads files are deposited in the National Center for Biotechnology Information (NCBI) Sequence Read Archive, BioProject PRJNA1401813.

1. 1. Chakraborty PP
   2. Kassen R

   (2026) Dryad Digital Repository

   Accelerated evolution in networked metapopulations of Pseudomonas aeruginosa.

   https://doi.org/10.5061/dryad.sf7m0cggv
2. 1. Chakraborty PP
   2. Kassen R

   (2026) NCBI BioProject

   ID PRJNA1401813. Accelerated evolution in networked metapopulations of Pseudomonas aeruginosa.

   https://www.ncbi.nlm.nih.gov/bioproject/PRJNA1401813

1. 1. Abel S
   2. Abel zur Wiesch P
   3. Davis BM
   4. Waldor MK

   (2015) Analysis of bottlenecks in experimental models of infection

   PLOS Pathogens 11:e1004823.

   https://doi.org/10.1371/journal.ppat.1004823

   • Google Scholar
2. Preprint

   1. Bailey SF

   (2021) Spatial structure affects evolutionary dynamics and drives genomic diversity in experimental populations of Pseudomonas fluorescens

   bioRxiv.

   https://doi.org/10.1101/2021.09.28.461808

   • Google Scholar
3. 1. Barnard FM
   2. Maxwell A

   (2001) Interaction between DNA gyrase and quinolones: Effects of alanine mutations at GyrA subunit residues Ser83and Asp87

   Antimicrobial Agents and Chemotherapy 45:1994–2000.

   https://doi.org/10.1128/AAC.45.7.1994-2000.2001

   • Google Scholar
4. 1. Bates D
   2. Mächler M
   3. Bolker B
   4. Walker S

   (2015) Fitting linear mixed-effects models using lme4

   Journal of Statistical Software 67:1–48.

   https://doi.org/10.18637/jss.v067.i01

   • Google Scholar
5. 1. Bolger AM
   2. Lohse M
   3. Usadel B

   (2014) Trimmomatic: a flexible trimmer for Illumina sequence data

   Bioinformatics 30:2114–2120.

   https://doi.org/10.1093/bioinformatics/btu170

   • Google Scholar
6. 1. Chadha J
   2. Harjai K
   3. Chhibber S

   (2022) Revisiting the virulence hallmarks of Pseudomonas aeruginosa: A chronicle through the perspective of quorum sensing

   Environmental Microbiology 24:2630–2656.

   https://doi.org/10.1111/1462-2920.15784

   • Google Scholar
7. 1. Chakraborty PP
   2. Nemzer LR
   3. Kassen R

   (2023) Experimental evidence that network topology can accelerate the spread of beneficial mutations

   Evolution Letters 7:447–456.

   https://doi.org/10.1093/evlett/qrad047

   • PubMed
   • Google Scholar
8. 1. Chevin LM
   2. Martin G
   3. Lenormand T

   (2010) Fisher’s model and the genomics of adaptation: Restricted pleiotropy, heterogenous mutation, and parallel evolution

   Evolution 64:3213–3231.

   https://doi.org/10.1111/j.1558-5646.2010.01058.x

   • PubMed
   • Google Scholar
9. 1. Choy WK
   2. Zhou L
   3. Syn CKC
   4. Zhang LH
   5. Swarup S

   (2004) MorA defines a new class of regulators affecting flagellar development and biofilm formation in diverse Pseudomonas species

   Journal of Bacteriology 186:7221–7228.

   https://doi.org/10.1128/JB.186.21.7221-7228.2004

   • PubMed
   • Google Scholar
10. 1. Dang H
    2. Lovell CR

    (2016) Microbial surface colonization and biofilm development in marine environments

    Microbiology and Molecular Biology Reviews 80:91–138.

    https://doi.org/10.1128/MMBR.00037-15

    • PubMed
    • Google Scholar
11. 1. Deatherage DE
    2. Barrick JE

    (2014) Identification of mutations in laboratory-evolved microbes from next-generation sequencing data using breseq

    Methods in Molecular Biology 1151:165–188.

    https://doi.org/10.1007/978-1-4939-0554-6_12

    • PubMed
    • Google Scholar
12. 1. Donker T
    2. Henderson KL
    3. Hopkins KL
    4. Dodgson AR
    5. Thomas S
    6. Crook DW
    7. Peto TEA
    8. Johnson AP
    9. Woodford N
    10. Walker AS
    11. Robotham JV

    (2017) The relative importance of large problems far away versus small problems closer to home: Insights into limiting the spread of antimicrobial resistance in England

    BMC Medicine 15:86.

    https://doi.org/10.1186/s12916-017-0844-2

    • PubMed
    • Google Scholar
13. Book

    1. Fisher RA

    (1930)

    The Genetical Theory of Natural Selection

    Oxford: At the Clarendon Press.

    • Google Scholar
14. 1. Frean M
    2. Rainey PB
    3. Traulsen A

    (2013) The effect of population structure on the rate of evolution

    Proceedings of the Royal Society B: Biological Sciences 280:20130211.

    https://doi.org/10.1098/rspb.2013.0211

    • PubMed
    • Google Scholar
15. 1. García-Bayona L
    2. Comstock LE

    (2018) Bacterial antagonism in host-associated microbial communities

    Science 361:eaat2456.

    https://doi.org/10.1126/science.aat2456

    • PubMed
    • Google Scholar
16. 1. Gillespie JH

    (1984) Molecular evolution over the mutational landscape

    Evolution 38:1116–1129.

    https://doi.org/10.1111/j.1558-5646.1984.tb00380.x

    • PubMed
    • Google Scholar
17. 1. Gordo I
    2. Campos PRA

    (2006) Adaptive evolution in a spatially structured asexual population

    Genetica 127:217–229.

    https://doi.org/10.1007/s10709-005-4012-9

    • PubMed
    • Google Scholar
18. 1. Guillaume F

    (2011) Migration-induced phenotypic divergence: The migration-selection balance of correlated traits

    Evolution 65:1723–1738.

    https://doi.org/10.1111/j.1558-5646.2011.01248.x

    • PubMed
    • Google Scholar
19. 1. Ha D-G
    2. Richman ME
    3. O’Toole GA

    (2014) Deletion mutant library for investigation of functional outputs of cyclic diguanylate metabolism in Pseudomonas aeruginosa PA14

    Applied and Environmental Microbiology 80:3384–3393.

    https://doi.org/10.1128/AEM.00299-14

    • PubMed
    • Google Scholar
20. Book

    1. Ha DG
    2. O’Toole GA

    (2015) C-di-GMP and its effects on biofilm formation and dispersion: a Pseudomonas aeruginosa review

    In: Ghannoum M, Parsek M, Whiteley M, Mukherjee PK, editors. Microbial Biofilms. John Wiley & Sons, Ltd. pp. 301–317.

    https://doi.org/10.1128/9781555817466.ch15

    • Google Scholar
21. 1. Habets MGJL
    2. Rozen DE
    3. Hoekstra RF
    4. de Visser JAGM

    (2006) The effect of population structure on the adaptive radiation of microbial populations evolving in spatially structured environments

    Ecology Letters 9:1041–1048.

    https://doi.org/10.1111/j.1461-0248.2006.00955.x

    • PubMed
    • Google Scholar
22. 1. Habets MGJL
    2. Czárán T
    3. Hoekstra RF
    4. de Visser JAGM

    (2007) Spatial structure inhibits the rate of invasion of beneficial mutations in asexual populations

    Proceedings of the Royal Society B 274:2139–2143.

    https://doi.org/10.1098/rspb.2007.0529

    • Google Scholar
23. 1. Hajishengallis G
    2. Lamont RJ
    3. Koo H

    (2023) Oral polymicrobial communities: Assembly, function, and impact on diseases

    Cell Host & Microbe 31:528–538.

    https://doi.org/10.1016/j.chom.2023.02.009

    • Google Scholar
24. 1. Hall-Stoodley L
    2. Costerton JW
    3. Stoodley P

    (2004) Bacterial biofilms: From the natural environment to infectious diseases

    Nature Reviews. Microbiology 2:95–108.

    https://doi.org/10.1038/nrmicro821

    • PubMed
    • Google Scholar
25. 1. Harrison JJ
    2. Almblad H
    3. Irie Y
    4. Wolter DJ
    5. Eggleston HC
    6. Randall TE
    7. Kitzman JO
    8. Stackhouse B
    9. Emerson JC
    10. Mcnamara S
    11. Larsen TJ
    12. Shendure J
    13. Hoffman LR
    14. Wozniak DJ
    15. Parsek MR

    (2020) Elevated exopolysaccharide levels in Pseudomonas aeruginosa flagellar mutants have implications for biofilm growth and chronic infections

    PLOS Genetics 16:e1008848.

    https://doi.org/10.1371/journal.pgen.1008848

    • PubMed
    • Google Scholar
26. 1. Herren CM
    2. Baym M

    (2022) Decreased thermal niche breadth as a trade-off of antibiotic resistance

    The ISME Journal 16:1843–1852.

    https://doi.org/10.1038/s41396-022-01235-6

    • PubMed
    • Google Scholar
27. 1. Hindersin L
    2. Traulsen A

    (2015) Most undirected random graphs are amplifiers of selection for birth-death dynamics, but suppressors of selection for death-birth dynamics

    PLOS Computational Biology 11:e1004437.

    https://doi.org/10.1371/journal.pcbi.1004437

    • PubMed
    • Google Scholar
28. 1. Jenal U
    2. Reinders A
    3. Lori C

    (2017) Cyclic di-GMP: Second messenger extraordinaire

    Nature Reviews. Microbiology 15:271–284.

    https://doi.org/10.1038/nrmicro.2016.190

    • PubMed
    • Google Scholar
29. 1. Kassen R
    2. Bataillon T

    (2006) Distribution of fitness effects among beneficial mutations before selection in experimental populations of bacteria

    Nature Genetics 38:484–488.

    https://doi.org/10.1038/ng1751

    • PubMed
    • Google Scholar
30. 1. Katharios-Lanwermeyer S
    2. Whitfield GB
    3. Howell PL
    4. O’Toole GA

    (2021) Pseudomonas aeruginosa uses c-di-GMP phosphodiesterases RmcA and MorA to regulate biofilm maintenance

    mBio 12:e03384-20.

    https://doi.org/10.1128/mBio.03384-20

    • PubMed
    • Google Scholar
31. 1. Kraemer SA
    2. Boynton PJ

    (2017) Evidence for microbial local adaptation in nature

    Molecular Ecology 26:1860–1876.

    https://doi.org/10.1111/mec.13958

    • PubMed
    • Google Scholar
32. 1. Kryazhimskiy S
    2. Rice DP
    3. Desai MM

    (2012) Population subdivision and adaptation in asexual populations of Saccharomyces cerevisiae

    Evolution 66:1931–1941.

    https://doi.org/10.1111/j.1558-5646.2011.01569.x

    • PubMed
    • Google Scholar
33. Preprint

    1. Kuo YP
    2. Hu J
    3. Carja O

    (2025) Clonal interference, genetic variation and the speed of evolution in structured populations

    bioRxiv.

    https://doi.org/10.1101/2025.02.22.639636

    • Google Scholar
34. Software

    1. Lenth RV

    (2020)

    Emmeans: estimated marginal means, aka least-squares means, version 2.0.1

    Package ‘emmeans.’.
35. 1. Leventhal GE
    2. Hill AL
    3. Nowak MA
    4. Bonhoeffer S

    (2015) Evolution and emergence of infectious diseases in theoretical and real-world networks

    Nature Communications 6:6101.

    https://doi.org/10.1038/ncomms7101

    • PubMed
    • Google Scholar
36. 1. Lieberman E
    2. Hauert C
    3. Nowak MA

    (2005) Evolutionary dynamics on graphs

    Nature 433:312–316.

    https://doi.org/10.1038/nature03204

    • PubMed
    • Google Scholar
37. 1. Ma LZ
    2. Wang D
    3. Liu Y
    4. Zhang Z
    5. Wozniak DJ

    (2022) Regulation of biofilm exopolysaccharide biosynthesis and degradation in Pseudomonas aeruginosa

    Annual Review of Microbiology 76:413–433.

    https://doi.org/10.1146/annurev-micro-041320-111355

    • PubMed
    • Google Scholar
38. 1. Maruyama T

    (1970) On the fixation probability of mutant genes in a subdivided population

    Genetical Research 15:221–225.

    https://doi.org/10.1017/S0016672300001543

    • Google Scholar
39. 1. Miralles R
    2. Moya A
    3. Elena SF

    (1999) Effect of population patchiness and migration rates on the adaptation and divergence of vesicular stomatitis virus quasispecies populations

    The Journal of General Virology 80 ( Pt 8):2051–2059.

    https://doi.org/10.1099/0022-1317-80-8-2051

    • PubMed
    • Google Scholar
40. 1. Moradali MF
    2. Ghods S
    3. Rehm BHA

    (2017) Pseudomonas aeruginosa Lifestyle: A paradigm for adaptation, survival, and persistence

    Frontiers in Cellular and Infection Microbiology 7:39.

    https://doi.org/10.3389/fcimb.2017.00039

    • PubMed
    • Google Scholar
41. 1. Mould DL
    2. Stevanovic M
    3. Ashare A
    4. Schultz D
    5. Hogan DA

    (2022) Metabolic basis for the evolution of a common pathogenic Pseudomonas aeruginosa variant

    eLife 11:e76555.

    https://doi.org/10.7554/eLife.76555

    • PubMed
    • Google Scholar
42. 1. Myall AC
    2. Peach RL
    3. Weiße AY
    4. Mookerjee S
    5. Davies F
    6. Holmes A
    7. Barahona M

    (2021) Network memory in the movement of hospital patients carrying antimicrobial-resistant bacteria

    Applied Network Science 6:1–23.

    https://doi.org/10.1007/s41109-021-00376-5

    • Google Scholar
43. 1. Nadell CD
    2. Drescher K
    3. Foster KR

    (2016) Spatial structure, cooperation and competition in biofilms

    Nature Reviews. Microbiology 14:589–600.

    https://doi.org/10.1038/nrmicro.2016.84

    • PubMed
    • Google Scholar
44. 1. Nahum JR
    2. Godfrey-Smith P
    3. Harding BN
    4. Marcus JH
    5. Carlson-Stevermer J
    6. Kerr B

    (2015) A tortoise–hare pattern seen in adapting structured and unstructured populations suggests a rugged fitness landscape in bacteria

    PNAS 112:7530–7535.

    https://doi.org/10.1073/pnas.1410631112

    • Google Scholar
45. 1. Nekkab N
    2. Crépey P
    3. Astagneau P
    4. Opatowski L
    5. Temime L

    (2020) Assessing the role of inter-facility patient transfer in the spread of carbapenemase-producing Enterobacteriaceae: the case of France between 2012 and 2015

    Scientific Reports 10:14910.

    https://doi.org/10.1038/s41598-020-71212-6

    • PubMed
    • Google Scholar
46. Software

    1. Oksanen J
    2. Simpson GL
    3. Blanchet GF
    4. Kindt R

    (2024)

    Vegan: community ecology package, version 2.7-2

    Package ‘vegan.’.
47. 1. Oliver A
    2. Mena A

    (2010) Bacterial hypermutation in cystic fibrosis, not only for antibiotic resistance

    Clinical Microbiology and Infection 16:798–808.

    https://doi.org/10.1111/j.1469-0691.2010.03250.x

    • PubMed
    • Google Scholar
48. 1. Orr HA

    (2000) The rate of adaptation in asexuals

    Genetics 155:961–968.

    https://doi.org/10.1093/genetics/155.2.961

    • Google Scholar
49. 1. Orr HA

    (2005)

    The probability of parallel evolution

    Evolution 59:216–220.

    • PubMed
    • Google Scholar
50. 1. Pavlogiannis A
    2. Tkadlec J
    3. Chatterjee K
    4. Nowak MA

    (2018) Construction of arbitrarily strong amplifiers of natural selection using evolutionary graph theory

    Communications Biology 1:71.

    https://doi.org/10.1038/s42003-018-0078-7

    • PubMed
    • Google Scholar
51. 1. Pennington LK
    2. Slatyer RA
    3. Ruiz-Ramos DV
    4. Veloz SD
    5. Sexton JP

    (2021) How is adaptive potential distributed within species ranges?

    Evolution 75:2152–2166.

    https://doi.org/10.1111/evo.14292

    • PubMed
    • Google Scholar
52. 1. Perfeito L
    2. Pereira MI
    3. Campos PRA
    4. Gordo I

    (2008) The effect of spatial structure on adaptation in Escherichia coli

    Biology Letters 4:57–59.

    https://doi.org/10.1098/rsbl.2007.0481

    • Google Scholar
53. 1. Poole K

    (2005) Efflux-mediated antimicrobial resistance

    The Journal of Antimicrobial Chemotherapy 56:20–51.

    https://doi.org/10.1093/jac/dki171

    • PubMed
    • Google Scholar
54. Book

    1. R Core Team

    (2022)

    R: A Language and Environment for Statistical Computing

    Vienna, Austria: R Foundation for Statistical Computing.

    • Google Scholar
55. 1. Richardot C
    2. Juarez P
    3. Jeannot K
    4. Patry I
    5. Plésiat P
    6. Llanes C

    (2016) Amino acid substitutions account for most mexs alterations in clinical nfxC mutants of Pseudomonas aeruginosa

    Antimicrobial Agents and Chemotherapy 60:2302–2310.

    https://doi.org/10.1128/AAC.02622-15

    • Google Scholar
56. 1. Römling U
    2. Galperin MY
    3. Gomelsky M

    (2013) Cyclic di-GMP: the first 25 years of a universal bacterial second messenger

    Microbiology and Molecular Biology Reviews 77:1–52.

    https://doi.org/10.1128/MMBR.00043-12

    • PubMed
    • Google Scholar
57. 1. Rumbaugh KP
    2. Griswold JA
    3. Hamood AN

    (2000) The role of quorum sensing in the in vivo virulence of Pseudomonas aeruginosa

    Microbes and Infection 2:1721–1731.

    https://doi.org/10.1016/s1286-4579(00)01327-7

    • PubMed
    • Google Scholar
58. 1. Ryder C
    2. Byrd M
    3. Wozniak DJ

    (2007) Role of polysaccharides in Pseudomonas aeruginosa biofilm development

    Current Opinion in Microbiology 10:644–648.

    https://doi.org/10.1016/j.mib.2007.09.010

    • PubMed
    • Google Scholar
59. 1. Sagarin RD
    2. Gaines SD
    3. Gaylord B

    (2006) Moving beyond assumptions to understand abundance distributions across the ranges of species

    Trends in Ecology & Evolution 21:524–530.

    https://doi.org/10.1016/j.tree.2006.06.008

    • Google Scholar
60. 1. Sanders LH
    2. Rockel A
    3. Lu H
    4. Wozniak DJ
    5. Sutton MD

    (2006) Role of Pseudomonas aeruginosa dinB-encoded DNA polymerase IV in mutagenesis

    Journal of Bacteriology 188:8573–8585.

    https://doi.org/10.1128/JB.01481-06

    • PubMed
    • Google Scholar
61. 1. Schick A
    2. Shewaramani S
    3. Kassen R

    (2022) Genomics of diversification of Pseudomonas aeruginosa in cystic fibrosis lung-like conditions

    Genome Biology and Evolution 14:evac074.

    https://doi.org/10.1093/gbe/evac074

    • PubMed
    • Google Scholar
62. 1. Schuster M
    2. Greenberg EP

    (2006) A network of networks: quorum-sensing gene regulation in Pseudomonas aeruginosa

    International Journal of Medical Microbiology 296:73–81.

    https://doi.org/10.1016/j.ijmm.2006.01.036

    • PubMed
    • Google Scholar
63. 1. Sharma N
    2. Yagoobi S
    3. Traulsen A

    (2023) Self-loops in evolutionary graph theory: Friends or foes?

    PLOS Computational Biology 19:e1011387.

    https://doi.org/10.1371/journal.pcbi.1011387

    • PubMed
    • Google Scholar
64. 1. Slatkin M

    (1981) Fixation probabilities and fixation times in a subdivided population

    Evolution 35:477–488.

    https://doi.org/10.1111/j.1558-5646.1981.tb04911.x

    • PubMed
    • Google Scholar
65. 1. Slatkin M

    (1985) Gene flow in natural populations

    Annual Review of Ecology and Systematics 16:393–430.

    https://doi.org/10.1146/annurev.es.16.110185.002141

    • Google Scholar
66. 1. Stacy A
    2. McNally L
    3. Darch SE
    4. Brown SP
    5. Whiteley M

    (2016) The biogeography of polymicrobial infection

    Nature Reviews. Microbiology 14:93–105.

    https://doi.org/10.1038/nrmicro.2015.8

    • PubMed
    • Google Scholar
67. 1. Steenackers HP
    2. Parijs I
    3. Dubey A
    4. Foster KR
    5. Vanderleyden J

    (2016) Experimental evolution in biofilm populations

    FEMS Microbiology Reviews 40:373–397.

    https://doi.org/10.1093/femsre/fuw002

    • PubMed
    • Google Scholar
68. 1. Tkadlec J
    2. Pavlogiannis A
    3. Chatterjee K
    4. Nowak MA

    (2019) Population structure determines the tradeoff between fixation probability and fixation time

    Communications Biology 2:138.

    https://doi.org/10.1038/s42003-019-0373-y

    • PubMed
    • Google Scholar
69. 1. Tkadlec J
    2. Pavlogiannis A
    3. Chatterjee K
    4. Nowak MA

    (2021) Fast and strong amplifiers of natural selection

    Nature Communications 12:4009.

    https://doi.org/10.1038/s41467-021-24271-w

    • PubMed
    • Google Scholar
70. 1. Werner GDA
    2. Strassmann JE
    3. Ivens ABF
    4. Engelmoer DJP
    5. Verbruggen E
    6. Queller DC
    7. Noë R
    8. Johnson NC
    9. Hammerstein P
    10. Kiers ET

    (2014) Evolution of microbial markets

    PNAS 111:1237–1244.

    https://doi.org/10.1073/pnas.1315980111

    • PubMed
    • Google Scholar
71. 1. Wiegand I
    2. Marr AK
    3. Breidenstein EBM
    4. Schurek KN
    5. Taylor P
    6. Hancock REW

    (2008) Mutator genes giving rise to decreased antibiotic susceptibility in Pseudomonas aeruginosa

    Antimicrobial Agents and Chemotherapy 52:3810–3813.

    https://doi.org/10.1128/AAC.00233-08

    • PubMed
    • Google Scholar
72. 1. Wilke CO

    (2004) The speed of adaptation in large asexual populations

    Genetics 167:2045–2053.

    https://doi.org/10.1534/genetics.104.027136

    • Google Scholar
73. 1. Williams P
    2. Cámara M

    (2009) Quorum sensing and environmental adaptation in Pseudomonas aeruginosa: A tale of regulatory networks and multifunctional signal molecules

    Current Opinion in Microbiology 12:182–191.

    https://doi.org/10.1016/j.mib.2009.01.005

    • PubMed
    • Google Scholar
74. 1. Winsor GL
    2. Griffiths EJ
    3. Lo R
    4. Dhillon BK
    5. Shay JA
    6. Brinkman FSL

    (2016) Enhanced annotations and features for comparing thousands of Pseudomonas genomes in the Pseudomonas genome database

    Nucleic Acids Research 44:D646–D653.

    https://doi.org/10.1093/nar/gkv1227

    • PubMed
    • Google Scholar
75. Conference

    1. Wright S

    (1932)

    The roles of mutation, inbreeding, crossbreeding and selection in evolution

    Proceedings of the XI International Congress of Genetics. pp. 209–222.

    • Google Scholar
76. 1. Yagoobi S
    2. Traulsen A

    (2021) Fixation probabilities in network structured meta-populations

    Scientific Reports 11:17979.

    https://doi.org/10.1038/s41598-021-97187-6

    • PubMed
    • Google Scholar
77. 1. Yeaman S
    2. Gerstein AC
    3. Hodgins KA
    4. Whitlock MC

    (2018) Quantifying how constraints limit the diversity of viable routes to adaptation

    PLOS Genetics 14:e1007717.

    https://doi.org/10.1371/journal.pgen.1007717

    • PubMed
    • Google Scholar
78. 1. Yoshida H
    2. Bogaki M
    3. Nakamura M
    4. Nakamura S

    (1990a) Quinolone resistance-determining region in the DNA gyrase gyrA gene of Escherichia coli

    Antimicrobial Agents and Chemotherapy 34:1271–1272.

    https://doi.org/10.1128/AAC.34.6.1271

    • Google Scholar
79. 1. Yoshida H
    2. Nakamura M
    3. Bogaki M
    4. Nakamura S

    (1990b) Proportion of DNA gyrase mutants among quinolone-resistant strains of Pseudomonas aeruginosa

    Antimicrobial Agents and Chemotherapy 34:1273–1275.

    https://doi.org/10.1128/AAC.34.6.1273

    • Google Scholar

Author details

1. Partha Pratim Chakraborty

   University of Ottawa, Ottawa, Canada

   Present address

   Gulbenkian Institute for Molecular Medicine, Rua da Quinta Grande, Oeiras, Portugal

   Contribution

   Conceptualization, Resources, Data curation, Formal analysis, Validation, Investigation, Visualization, Methodology, Writing – original draft, Writing – review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-6066-156X

2. Rees Kassen

   1. University of Ottawa, Ottawa, Canada
   2. McGill University, Montreal, Canada

   Contribution

   Conceptualization, Formal analysis, Supervision, Funding acquisition, Visualization, Writing – original draft, Project administration, Writing – review and editing

   For correspondence

   rees.kassen@mcgill.ca

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-5617-4259

• 0

  citations

Views, downloads and citations are aggregated across all versions of this paper published by eLife.