doi: 10.1007/s00248-025-02618-w.
Evolution of One Species Increases Resistance to Invasion in a Simple Synthetic Community
Affiliations
- PMID: 41114853
- PMCID: PMC12537770
- DOI: 10.1007/s00248-025-02618-w
Item in Clipboard
Evolution of One Species Increases Resistance to Invasion in a Simple Synthetic Community
Microb Ecol.
.
Abstract
The species that make up a microbial community determine its potential function. A major goal of microbial ecology is to make assemblages of microbes - synthetic communities - with targeted applications. Replacing a dysfunctional community with a synthetic microbial community can have transformative impacts upon a host or ecosystem, yet the introduced community may be outcompeted by local species or communities, resulting in transient effects. Here, we study a simple synthetic community comprised of two species - E. coli and S. cerevisiae - that have coevolved for either 0, 1000 or 4000 generations, and evaluate the potential for 12 bacterial strains, from five species, to invade. We find that the dominant species (E. coli) in the community protects the less dominant species from being outcompeted during an invasion, and that this effect is strengthened by longer periods of coevolution. Using a mathematical model, we show how prolonged co-evolution leads to protective effects for a community member sensitive to displacement.
Keywords: Experimental evolution; Microbial community; Microbial evolution.
© 2025. The Author(s).
Conflict of interest statement
Declarations. Competing Interests: The authors declare no competing interests.
Figures
Co-evolved E. coli-yeast pairs. (A) Equilibrium frequencies for E. coli-yeast combinations that were co-cultured for either 0 generations, i.e. the ancestor (green), 1000 generations (blue) or 4000 generations (red). Each point represents an independent co-culture. (B) Population size at equilibrium, for E. coli and yeast in co-culture. (C) Maximum carrying capacity (Max OD, or stationary phase density) of E. coli and yeast in the spent media of their “partner” strain from the same generation. Y axis indicated growth in spent media divided by growth in unspent media, minus 1. For example, the rightmost column shows that 4000 generation evolved yeast grows significantly better in media that has already been spent by 4000 generation E. coli, than ancestral yeast grows in ancestral E. coli spent media (paired t-test t = − 14.35, p = 1.37 × 10−4, d.f. = 4). Error bars indicate standard error of the mean
Competitor strains spent media assays and pairwise growth assay outcomes. (A) Phylogenetic relationship of bacterial strains used in this study. Branch lengths do not indicate genetic distance. (B) Growth of E. coli and yeast in the spent media of the invading bacterial strains. E. coli showed varied growth in the spent media of “competitor” strains; while ancestral and 1000 generation E. coli grew poorly (dark blue), or showed intermediate growth (light shading) in the spent media of several “competitor” strains, 4000-generation E. coli showed improved growth (yellow) in Pseudomonas and some Acinetobacter spent media. In contrast, yeast showed intermediate (light shading) to improved growth (yellow) in the growth media spent by the competitor bacteria, with only a single case of poor growth (dark blue) in P. fluorescens spent media. (C) All competitor strains grew very poorly in E. coli spent media, while species and strains showed intermediate and some improved growth, in yeast spent media. Growth difference was determined by dividing growth in spent media by growth in unspent media, and subtracting 1
Three-way competitions are predicted by E. coli but not yeast pairwise assays. (A) Measurements of pairwise competitions between yeast (grey lines) and each competitor bacteria (coloured lines). The frequency of yeast at the final time point is indicated by inset text. (B) Measurements of pairwise competitions between E. coli (black lines) and each competitor bacteria (coloured lines). (C) Measurements of three-way competitions between yeast (grey lines), E. coli (black lines) and each competitor bacteria (coloured lines). The frequency of E. coli at the final time point is indicated by text. Each pair or set of lines indicates an independent 7-day experiment. Cultures for A. baumannii AB19978, ancestral S. cerevisiae, and ancestral E. coli failed to grow under the experimental conditions and are therefore not shown. (D) Predicted outcomes of pairwise competition assays, based on measurements of growth rates of each strain in monoculture. Growth rates of competitor bacteria were compared to growth rates of either E. coli or yeast (Fig. S2, Supplementary Methods). (E). Outcomes are either coexistence (green) or competitive exclusion of the competitor species (red), E. coli (blue) or yeast (yellow). For example, for (E), the three columns on left summarise the results of two-way assays between E. coli and the competitor strains and show that E. coli was excluded by only one of the competitor bacterial strains (E. coli EAEC042). (F) The three columns indicate 3-way assays between E. coli-yeast pairs and a competitor bacterial strain. The yellow-blue box indicates where both E. coli and yeast were excluded in the three-way growth assay. Two- and three-way competitions were carried out with five replicates, and the outcome designated as coexistence or extirpation based on the majority outcome, and extinctions confirmed with plate counts, where possible (Supplementary Methods)
Simulations with a three-species Lotka-Volterra model. Each panel shows the outcomes of three species interactions for a range of alpha values after 24 h of growth following altered classic Lotka-Volterra (LV) models (Supplementary Methods). Simulations were run using carrying capacity and growth rate values from the experiment. Outcomes of coexistence or exclusion are shown on the right of the figure. Purple indicates only the competitor is extinct, blue shows only E. coli is extinct, red shows only yeast is extinct, and green shows that all three species coexist. A, B, C, D) All interaction values are fixed at 0.01 (for example, the interaction coefficients between E. coli and yeast) except for those on the x and y axes. A) Competition outcomes for a range of interaction coefficients; the influence of E. coli on the competitor (x-axis) and the influence of the competitor on yeast (y-axis). For example, yeast goes extinct even when the influence of the competitor is very small, but the competitor goes extinct when the interaction with E. coli is largely negative. B) Competition outcomes for a range of interaction coefficients; the influence of the competitor on E. coli (y-axis) and the influence of E. coli on the competitor (x-axis). C) Competition outcomes for a range of interaction coefficients; the influence of the competitor on E. coli (x-axis) and the influence of yeast on the competitor. D) Competition outcomes for a range of interaction coefficients; the influence of the competitor on yeast (x-axis) and the influence of yeast on the competitor (y-axis), with an interaction of E. coli on the competitor of 0.01. E) The same interactions between Yeast and the competitor are shown, but with an interaction of E. coli on the competitor of 2
References
-
- Aggarwal S (2021) Digest: optimization of microbial community function through propagule strategies. Evolution 75:200–201 - PubMed
MeSH terms
Grants and funding
LinkOut - more resources
Full Text Sources
