Fast evolution unlocks forbidden communities

When ecological and evolutionary dynamics occur on similar timescales, otherwise inaccessible stable and diverse communities can spontaneously assemble.

Explaining how a large diversity of species can coexist within a community — from animals in a savanna to oceanic plankton or soil microbiomes — is an enduring challenge in ecology. Mechanisms that promote coexistence continue to be uncovered both theoretically and empirically1,2,3,4. But even when a current state of coexistence can be theoretically justified, a further challenge lies in understanding how such diverse communities were originally assembled5. Thus, a fundamental but insufficiently tackled question is which of the communities that are theoretically feasible could be realized and under what conditions. Writing in Nature Ecology & Evolution, Kotil and Vetsigian6 uncover an assembly mechanism that can lead to the spontaneous formation of microbial communities.

In ecology, a community is assumed to assemble through sequential, repeated immigration of species from the available regional species pool, and it is said to be ecologically stable if it is resistant to colonization by any additional species in that pool5. Although evolution can further shape this process — for example, by adaptation of immigrant species to the new habitat or by speciation — it is typically less investigated owing to the assumption that ecological processes occur on a much faster timescale than evolutionary ones. Such a separation of timescales has also classically been assumed and successfully employed by evolutionary biologists. For example, the adaptive dynamics framework considers mutations to be rare, sequential events, such that the ecological dynamics between the resident population and the invading mutant equilibrate before a new mutant arises. A resident population that cannot be invaded by any rare mutant is said to be in an evolutionarily stable state7 (Fig. 1a).

Fig. 1: Frequent mutations allow the assembly of otherwise inaccessible diverse microbial communities.

a, In the classical world of rare mutations, ecological dynamics between the resident and a mutant equilibrate before any new mutant can arise (an example of competitive exclusion is shown). Therefore, only certain communities are accessible via this process. b, When mutations are more frequent, new mutants can arise before the ecological dynamics in the resident population equilibrate. Kotil and Vetsigian6 show that this allows the right mutant to arise at the right time (in other words, when there is the right combination of interacting strains) and facilitates the emergence of previously inaccessible eco-evolutionarily stable communities.

However, this separation of timescales fails when one considers microbial communities: because their evolutionary dynamics are more rapid owing to high mutation rates and extensive horizontal gene transfer8, ecological and evolutionary processes occur on similar timescales. In this case, mutation potentially becomes as important a driving force for community assembly as immigration, and the eco-evolutionary stability of communities must be assessed. In this context, it is interesting to ask whether moving away from the classical world of very rare mutations can lead to fundamentally and qualitatively different phenomena.

Kotil and Vetsigian build on previous work demonstrating that the three-way interaction between an antibiotic-producing strain, an antibiotic-sensitive one and an antibiotic-degrading strain that attenuates the inhibitory interaction between the first two can lead to ecologically stable communities4. That study offered an elegant — and empirically grounded — example of higher-order interactions, which have been proposed to play a crucial role in mediating the ecologically stable coexistence of diverse species1,9. However, that research did not explore whether and how these communities could be assembled de novo. Following up on that work, Vetsigian previously confirmed10 that such communities were in fact attainable via evolutionary assembly processes, but stopped short of exploring the underlying assembly mechanism that allowed them to emerge. This is where the current study comes in and brings the story full circle. Not only do the authors uncover the responsible mechanism, but they further reveal it to be a qualitatively different one that could not have operated in the classical world of rare mutations. In an inspired spin on quantum tunnelling — the quantum-mechanical phenomenon in which a particle can pass through a barrier that it cannot surmount within the framework of classical mechanics — Kotil and Vetsigian term their uncovered mechanism ‘eco-evolutionary tunnelling’.

This tunnelling leads to communities that are evolutionarily stable but inaccessible via the step-by-step assembly approach of adaptive dynamics: their emergence requires ‘well-timed’ mutants to arrive during a specific window of opportunity, which is impossible when mutations are rare, but becomes possible as soon as mutations are more frequent (Fig. 1b). Moreover, the rate of community emergence is an increasing, albeit eventually saturating, function of the mutation rate. By exploring this process in a metacommunity model that allows for migration between patches, the authors reveal that the emergent communities are in fact eco-evolutionarily stable and that an intermediate migration rate — such that the dynamics on the different patches are least synchronized — maximizes their rate of assembly.

Thus, consistent with explorations in evolutionary game theory11, Kotil and Vetsigian respond affirmatively to the question of whether moving away from the classical world of rare mutations can lead to qualitatively different phenomena. As we become increasingly aware of the importance of studying microbial community dynamics, their finding reinforces the need to treat microbes in a properly eco-evolutionary framework and makes the further investigation of eco-evolutionary phenomena even more imperative.

Interestingly, it appears that the mutants that catalyse community emergence are ‘well timed’ precisely when their appearance leads to higher-order interactions, that is when their presence modifies an existing interaction, or set of interactions, in the resident population. This interpretation adds to the excitement produced by recent studies proposing that the existence of higher-order interactions among co-occurring species can stabilize their dynamics and facilitate their coexistence1,4,9,12. While that may be theoretically true, it has been less clear — until now — how such higher-order interactions would arise and persist during the community assembly process. Kotil and Vetsigian’s findings suggest that higher-order interactions might in fact be what allows the assembly of diverse communities to begin with, and their results leave us with a fascinating question: are frequent mutation and/or migration events necessary to allow the emergence of higher-order interactions and the subsequent assembly of stable and diverse communities?


  1. 1.

    Levine, J. M., Bascompte, J., Adler, P. B. & Allesina, S. Nature 546, 56–64 (2017).

    CAS  Article  Google Scholar 

  2. 2.

    Kartzinel, T. R. et al. Proc. Natl Acad. Sci. USA 112, 8019–8024 (2015).

    CAS  Article  Google Scholar 

  3. 3.

    Goldford, J. E. et al. Science 361, 469–474 (2018).

    CAS  Article  Google Scholar 

  4. 4.

    Kelsic, E. D., Zhao, J., Vetsigian, K. & Kishony, R. Nature 521, 516–519 (2015).

    CAS  Article  Google Scholar 

  5. 5.

    Fukami, T. Annu. Rev. Ecol. Evol. Syst. 46, 1–23 (2015).

    Article  Google Scholar 

  6. 6.

    Kotil, S. E. & Vetsigian, K. Nat. Ecol. Evol. (2018).

  7. 7.

    Nowak, M. A. Evolutionary Dynamics (Harvard Univ. Press, Cambridge, 2006).

  8. 8.

    Bonham, K. S., Wolfe, B. E. & Dutton, R. J. eLife 6, e22144 (2017).

    Article  Google Scholar 

  9. 9.

    Grilli, J., Barabás, G., Michalska-Smith, M. J. & Allesina, S. Nature 548, 210–213 (2017).

    CAS  Article  Google Scholar 

  10. 10.

    Vetsigian, K. Nat. Ecol. Evol. 1, 0189 (2017).

    Article  Google Scholar 

  11. 11.

    Antal, T., Traulsen, A., Ohtsuki, H., Tarnita, C. E. & Nowak, M. A. J. Theor. Biol. 258, 614–622 (2009).

    Article  Google Scholar 

  12. 12.

    O’Dwyer, J. P. Nature 548, 166–167 (2017).

    Article  Google Scholar 

Download references

Author information



Corresponding author

Correspondence to Corina E. Tarnita.

Ethics declarations

Competing interests

The author declares no competing interests.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Tarnita, C.E. Fast evolution unlocks forbidden communities. Nat Ecol Evol 2, 1525–1526 (2018).

Download citation

Further reading


Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing