« Intelligent design falls into chaos? | Main | Celebrating Science Seasonally »

Spatial structure and the evolution of cooperation between microbes and plants

Evolutionary theory suggests that cooperation should often be unstable, even when it benefits all concerned. Toby Kiers and I discussed this problem and some possible solutions in a recent review article in Annual Review of Ecology and Evolution. Meanwhile, Jim Bever and colleagues have published some important experimental data in Ecology Letters, helping to explain cooperation between plants and fungi, a little too late to include in our review. Their paper is titled: "Preferential allocation to beneficial symbiont with spatial structure maintains mycorrhizal mutualism."

Many bacteria and fungi associated with plant roots benefit more from healthy plants than from dead or dying ones. If each individual plant were colonized by only one strain of bacteria or fungus, then strains that helped their host plants (by providing them with nitrogen or phosphorus, for example) would indirectly help themselves, gaining an evolutionary edge over their competitors of the same species. These beneficial strains would become more common in each generation. In other words, cooperation would evolve.

The problem is that each plant is typically associated with several strains of each species of bacteria or fungus. Strains that invest less in helping the plant have more resources to spend on their own reproduction. If less-generous strains benefit equally from the contributions of more-generous strains on the same plant, then less-generous strains will become more common over generations. In other words, "cheating" will evolve.

As a Ph.D. student in my lab, Toby showed how legume plants that depend on root-nodule bacteria for nitrogen solve this problem. They monitor individual nodules and cut off oxygen supply (and perhaps other resources) to nodules that provide them with less nitrogen. This reduces the reproduction of rhizobium bacteria in nodules that provide no nitrogen to less than half that of those that provide lots of nitrogen. So cheaters don't prosper.

Our review suggested that something similar could explain cooperation between plants and the mycorrhizal fungi that provide them with phosphorus and other benefits. The mathematical models that Stuart West developed in our lab to explain cooperation between plants and rhizobia could probably be extended to apply to these fungi as well. But what about actual data?

Previously, Bever showed that, when fungal strains were thoroughly mixed together, less-beneficial strains became more common. (This is the same result we would expect with rhizobia, if individual root nodules each contained many different strains: the plant wouldn't be able to help the good strains without also helping the bad ones.) But what if different fungal strains are found in different patches of soil around a plant? Do plants send more resources to fungi that supply them with more phosphorus? If so, does that let these more-beneficial fungi reproduce more?

To answer this question, they used wild onion plants with their roots growing into two separate containers of soil. (One of Toby's experiments also used this split-root approach.) The two containers were inoculated with either the same or different fungal strains, using strains expected to differ in benefits to the plant. The main resource plants provide to their fungal partners is carbon compounds. Plants were exposed to radioactive carbon dioxide, so that carbon compounds produced by photosynthesis, including those the plant provided to the fungi, would be radioactive.

When both sides of a plant's root system were inoculated with the same fungal strain, one strain provided essentially no benefit: plant growth was the same as an uninoculated control. The other strain helped the plant a lot, leading to almost twice as much growth as the control. Surprisingly, perhaps, plants inoculated with both strains, on different sides, grew about as well as plants inoculated with only the good strain. (Similarly, Toby found that three old soybean varieties did as well with a mix of good and bad rhizobia as with good rhizobia alone, although this was not true of three modern varieties.) One way that a plant could do well with a mixture of good and bad symbionts is to preferentially allocate resources to the good ones and that is what Jim Bever's plants did: the good strain got more than an equal share of the carbon compounds from the plant's photosynthesis.

But did the good strain actually benefit from getting more carbon? "Benefit" to an evolutionary biologist, means reproducing more. As Darwin wrote:

"I use the term Struggle for Existence in a large and metaphorical sense, including dependence of one being on another, and including (which is more important) not only the life of the individual, but success in leaving progeny."

The relative success of the two strains depended on spatial structure. When the two strains were mixed together on both sides of the root system, the less-beneficial strain produced more spores. This suggests that this strain was not simply defective, but rather a "cheater", benefiting by investing less in helping the host and more in its own reproduction. When the two strains were separated, however, the host's preferential allocation of carbon compounds resulted in higher spore production by the better symbiont.

A key question, identified by Bever et al., is whether spatial structure in the field is more similar to inoculating two sides of the root system with different fungal strains, or mixing the two strains together on each side. They cite previous studies showing that field soils have patchy distribution of different fungal strains, perhaps sufficient to let plants favor better fungal strains. Plowing, however, would result in more uniform mixing of fungal strains. This could give cheating strains an evolutionary edge.

As we noted in an earlier paper, however, soil mixing from plowing could also have the opposite effect. If soils are mixed up regularly, the descendants of symbionts that share an individual host plant this year are unlikely to encounter each other in the future. One strain on a plant doesn't benefit from helping the others, by providing their shared host with phosphorus, but the only cost is the direct cost of providing the phosphorus. In unmixed soils, however, this year's neighbor is next year's competitor: the descendants of fungi that share a plant this year are likely to compete in the future. In unmixed soils, therefore, benefiting other strains on the same plant is negative -- given the option, killing them might be better -- rather than neutral. Figuring out the balance between positive and negative effects of plowing on the evolution of mutualism could be quite a challenge.

Post a comment

(If you haven't left a comment here before, you may need to be approved by the site owner before your comment will appear. Until then, it won't appear on the entry. Thanks for waiting.)

Type the characters you see in the picture above.