Back

Partner fidelity and coevolution: Useful but not required for rapidly increased mutualistic benefits

Stoy, K. S.; Lac, D.; Ratcliff, W. C.

2025-09-08 evolutionary biology
10.1101/2025.09.07.674547 bioRxiv
Show abstract

Evolutionary theory predicts that specialization between mutualistic partners is beneficial, limiting conflict and increasing the opportunity to evolve cooperative benefits through sustained reciprocal selection. However, specialized mutualisms are relatively rare in nature. Few empirical studies have directly examined how multi-partner interactions affect mutualistic evolution, largely because tracking partner dynamics over evolutionary timescales is challenging in natural systems. We circumvent this constraint via experimental evolution with an engineered resource sharing mutualism in Bakers yeast (Saccharomyces cerevisiae), which allows precise control of partner fidelity across generations. We compared high partner fidelity (consistent pairings) with low partner fidelity (temporally rotating partners) across 54 rounds of selection. High partner fidelity produced the strongest mutualistic benefits (7.2% increased growth over controls), while low partner fidelity yielded only modest benefits (2.2% increase). Time-shift experiments confirmed that while coevolution enhanced benefit evolution, it was not strictly necessary: benefits also evolved through one-sided adaptation. Genomic analyses revealed parallel evolution in amino acid metabolism and starvation response genes, with the slower-growing genotype driving most evolutionary change. Surprisingly, high-fidelity lineages evolved as generalists rather than specialists, cooperating equally well with all partners tested. Overall, these results demonstrate that partner fidelity facilitates the rapid evolution of mutualistic benefits but is not strictly required for mutualistic evolution.

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.