What puts the “co” in coevolution?

ResearchBlogging.orgCoevolution is like the opposite of pornography – lots of scientists can define it with nicety, but most of us have trouble saying for sure whether any given pair of species are actually coevolving. “Coevolution” literally means “evolving together” – more formally, that an evolutionary change in one species causes a reciprocal change in another [$$]. But this process can be quite complicated to demonstrate in practice. In the latest example of this conundrum, a paper in this month’s Evolution suggests that one relationship we thought was coevolutionary maybe isn’t [$-a].

Leafcutter ants at work
Photo by rofanator.

Leafcutter ants have been thought to be involved in clear-cut coevolutionary relationships with a number of microbial species. Leafcutters, as is pretty well known, harvest leaf fragments to feed fungal gardens, which the ants use as a food source. That’s one relationship: ant-fungus. Less well-known are the ants’ relationships with bacteria – bacteria that fight off the fungus-garden-killing diseases, in the genus Pseudonocardia. Pseudonocardia grow on leafcutter ants’ exoskeletons [$-a], and the ants seem to regulate the bacteria’s growth depending on how much they need the antibiotics it produces. This seems like an obvious case of coevolution – the ants and their bacteria symbiotes live in close proximity to each other, and seem to rely on each other for their respective ways of life.

But the new paper, by Mueller et al., indicates that appearances can be deceiving. The authors present a new phylogeny of the bacterial family containing Pseudonocardia, which suggests that leafcutter ants frequently “recruit” new strains of Pseudonocardia from their environment. Coevolution, the authors argue, would mean that a single lineage of bacteria has been associated with the ants from the beginning of the relationship – but the phylogeny shows, instead, that ant-associated Pseudonocardia are often more closely related to free-living bacteria than to other strains of ant-associated Pseudonocardia. That is, the ant-associated strains are not monophyletic. Furthermore, individual ant species often carry Pseudonocardia from multiple different evolutionary lineages, which suggests that “recruitment” events don’t happen one after another, but continuously.

This result makes good biological sense. Ants using Pseudonocardia to control disease probably stimulate the evolution of resistant disease organisms, just the same problem that humans have found after less than a century of antibiotic use. Recruiting new strains of antibiotic-producing bacteria is just the way to deal with resistant disease organisms. It therefore makes about as much sense to say that ants are coevolving with Pseudonocardia as it would to say that humans are coevolving with penicillin – both the ant-associated bacterium and the mass-produced antibiotic are tools to be cast away when no longer useful.


C.R. Currie, J.A. Scott, R.C. Summerbell, D. Malloch (1999). Fungus-growing ants use antibiotic-producing bacteria to control garden parasites. Nature, 398 (6729), 701-4 DOI: 10.1038/19519

DH Janzen (1980). When is it coevolution? Evolution, 34 (3), 611-2 http://www.jstor.org/pss/2408229

U.G. Mueller, D. Dash, C. Rabeling, A. Rodrigues (2008). Coevolution between Attine ants and Actinomycete bacteria: A reevaluation. Evolution, 62 (11), 2894-912 DOI: 10.1111/j.1558-5646.2008.00501.x

1 thought on “What puts the “co” in coevolution?

  1. It’s amazing how often your blog shows up in my Google results. Thanks for summarizing this paper so I don’t have to read it.

Comments are closed.