Plants may seem defenseless against insects, having neither hands nor tails to brush them away. But many produce potent repellent chemicals, ranging from ones that just taste or smell bad to ones that can kill.
These disgusting compounds work well against nibbling and sap-sucking insects that feed on a wide range of plant types, as well as grazing and browsing mammals. Yet inevitably, over the course of evolution, certain animals specialized to the point that they’re now attracted to even the most repulsive stuff that plants have come up with. In fact, many of the crops our own species grow for consumption, from tobacco to coffee to coriander, appeal to us precisely because of the compounds they produce to discourage herbivores. So, what’s a poor plant to do?
Some, through natural selection, have evolved a different strategy aimed at dispatching unwelcome visitors: They send out odor signals to attract their enemies’ enemies. Depending on the signal, various members of a motley crew of self-interested creatures may respond. Some devour the plant eaters, while some simply deposit their eggs in them and let their larvae finish the job. Others bring in deadly bacteria.
Chemical ecologist Ted Turlings of the University of Neuchâtel in Switzerland has been studying this bag of tricks for over three decades in various spots around the world. A better understanding of these natural solutions, he believes, may help us to reduce our use of chemical pesticides, which are a threat not only to many potential insect allies but also to human health.
Turlings and coauthor Matthias Erb of the University of Bern described the intriguing chemistry between plants and the organisms that protect them from pests and how we might use it to our advantage in the Annual Review of Entomology. This conversation has been edited for length and clarity.
How did you first get interested in these kinds of rather unexpected interactions?
After finishing my studies in the Netherlands, I did a Ph.D. in the United States at the University of Florida — I actually worked at the facilities of the United States Department of Agriculture. My supervisor at the time asked me to find out how specific parasitic wasps find their hosts. These are tiny wasps that lay their eggs in other insects, and out of each egg comes a larva that will then eat the host insect.
I was specifically studying wasps that attack caterpillars that feed on corn plants, and we initially thought the wasps must be using odors and visual cues that come from the caterpillars. But we found out rather quickly that it was the plants that actively emit volatiles that attracts these wasps. And it turned out that they were producing a blend of very specific compounds in response to an attack by caterpillars, ones that they normally do not emit.
Just damaging the plant does not induce these same emissions. But if you damage the plant and then smear some caterpillar spit on the damaged site, you get the same reaction. That was very intriguing to us, as it suggested that the plants can recognize what is attacking them and respond accordingly.
Some compounds are emitted immediately from the damaged site — we call those green leaf volatiles; they’re what you smell when you mow the lawn. But in addition to that, the plant will synthesize these new compounds in response to the spit, ones it only produces when attacked by caterpillars. And these aren’t just emitted from the damaged site but also from the undamaged leaves of a plant that is under attack.
Much later, it was found that it’s not just parasitic wasps that can pick up these odors but neighboring plants as well. And other researchers have discovered that the response is stronger in plants that are more closely related to each other. Scientists are still debating whether other plants are just eavesdropping on this signal or whether plants under attack are communicating and may benefit from alerting the neighbors. I was actually discussing this with a colleague yesterday.
How does this work from the wasp’s point of view? Is there a wasp species for every combination of plant and pest?
Not quite. The wasps we work with are generalists, so they can attack many different caterpillar species on many different kinds of plants. These plants all emit different volatiles. It has been shown that wasps have to learn to associate these odors with the presence of insect hosts for them to lay their eggs.
So the wasps are confronted with different signals all the time, and it’s not always predictable which signal will be the most beneficial. You can imagine that a naive wasp that has just emerged from its cocoon will start off using general signals, like the green leafy volatiles that I’ve mentioned before, which are emitted by plants when attacked.
This way, they may find a plant under attack from beetles, and since they can definitely not lay their eggs in those, they will learn to avoid the specific odors associated with that situation. But as soon as they find a plant under attack from a caterpillar and identify it as the right host, they will remember the surrounding odors as something to focus on.
Would it be possible to train wasps to be attracted to a certain smell, say strawberry, and spray it on plants we’d like them to protect against herbivorous insects?
It certainly makes sense, if you are trying to fight one specific pest, to train wasps on the odors associated with that pest. But I wouldn’t go with the strawberry approach you propose because what would happen, I think, is they will fly around, look for hosts wherever it smells like strawberry, and not find anything. And eventually, they’ll start avoiding that odor. It might be a good way to keep them in your field for a while, but it may not work for very long.
It seems to me it would be more useful to train the wasps, before you release them, on an odor that is likely to get them to the pest you want them to kill. This might be done by combining an odor typically emitted by a plant under attack with caterpillar feces, which the wasps get very excited over, but can only smell when they get very close.
Caterpillar feces is almost odorless, which makes sense: The caterpillars have evolved to minimize any possible signal for their enemies. For the plants, it is the opposite: They release very large amounts of the volatiles they produce to attract the wasps and other natural enemies of the caterpillars.
The widespread use of pesticides in agriculture to get rid of crop pests has contributed to drastic declines in insect populations. Might these pesticides have reduced the numbers of natural enemies available to help us control crop pests, and do you think there is a way to bring them back?
Yes. In Switzerland, there are efforts to have flower strips alongside cropping fields, which already helps to keep natural enemies around.
But if you think about how corn is grown in the US, that will be quite tricky. In much of the Corn Belt, where 40 percent of all the corn in the world is produced, there is nothing but cornfields. Well — and plenty of western corn rootworm, a beetle larva that feeds on the nutrient-rich crown roots that support the stem.
There currently are relatively effective ways to control that pest by using pesticides and genetically modified plants, the Bt plants, that produce a toxin that kills the larvae. But in all cases, I’m afraid the beetles will eventually develop resistance.
The only way to avoid that, I think, is to use living organisms to control the pests.
Given that these larvae are underground, I imagine wasps will not be very helpful.
That’s right. But in fact, there are some very similar mechanisms operating underground. When under attack, the roots of most corn varieties outside the US produce a compound called caryophyllene, which attracts a number of species of tiny worms called nematodes that have a really intriguing life cycle.
Juvenile nematodes use signals like caryophyllene to find beetle larvae. Then they penetrate into the larvae through any opening they can find. There they release the bacteria they always carry with them. These bacteria produce toxins that kill and digest the insects very quickly, and then they rapidly multiply. The nematodes then feed on these bacteria and the digested insides of the insect.
They go through several life cycles, producing thousands of new nematodes that will eventually burst out of the cadaver and go looking for a new host.
Intriguingly, though caryophyllene repels many herbivorous insects, the beetle larvae referred to as corn rootworms are, in fact, attracted to low amounts of it. Perhaps for this reason, in the US, the ability to produce caryophyllene has been selected out of the local corn varieties.
We have shown in a field trial that if we restore this ability by inserting a gene from oregano, a plant that produces plenty of caryophyllenes, we can reduce the damage from corn rootworms significantly because their roots were highly attractive to the nematode. Yet because the role of caryophyllene is twofold — it attracts the rootworm-killing nematode but also the corn rootworm itself — I am not sure this will ever be commercialized.
We are now exploring methods to apply the nematodes more directly to pest insects. These efforts are focused on the fall armyworm, which is actually a caterpillar, a major pest in Africa and Asia that is also very likely to show up in Europe in the next few years. Since it attacks the plant aboveground, it would not normally come into contact with nematodes that attack insect larvae belowground. So it has no defenses against nematodes.
We have found that the nematodes are very effective at killing the caterpillars if we inject them in a gel into the center of the plant or apply them to its surface. We have done field trials in Rwanda, showing this is just as effective as using a pesticide, and we are running a field trial in Vietnam this year to test it there as well.
And in each case, we use locally occurring nematodes to avoid the risks associated with the introduction of new species.
Where do these rapidly spreading pest species like the fall armyworm suddenly come from?
From the Americas, possibly from Mexico, where many of our crops originate. In its native range, the fall armyworm is less of a problem, probably because it is quite effectively eliminated by parasitic and predatory insects. But elsewhere in the world, it appears that such natural enemies are far less effective.
The fall armyworm was first detected in Africa in 2016, and within three years, it spread all across sub-Saharan Africa, causing billions of dollars of damage. Then, in 2018, it was reported in Asia, and now it’s spread all over Asia and even Australia, creating incredible damage. That has resulted in an enormous increase in pesticide use, which has extremely negative effects not just on the environment but also on the health of the people that are applying it, who often don’t have the means to protect themselves.
So it is very important to look for alternatives with local partners, which is where the nematodes come into play.
What do you think is the chance that pests will become resistant to the nematodes as well?
Never say never, but I think it is unlikely because, unlike pesticides or toxins, the nematodes and the bacteria they carry can adapt to any changes the pests might evolve to avoid being killed. There are so many nematodes that there will likely always be a few with a genetic mutation to overcome the pest species’ resistance. And if that nematode doesn’t evolve by itself, we can also give it a hand by selecting the nematodes that continue to be effective in the lab, even switching to a different species. That is the advantage of working with living organisms: We can select, and they can evolve. And the better we understand them, the more possibilities will open up.
We are working on sensors now that can detect which plants are under attack and even distinguish between different caterpillar species on the basis of the odors the plants produce. We hope that in a couple of years, we could have a robot going through a field, smelling the plants, and not only giving the farmer real-time feedback on which plants are under attack but perhaps also spraying some nematodes on them.
Robots that can detect weeds and spray them with herbicide already exist, and in August, we will be testing if these robots can also be used to spray the insect-killing nematodes.
This article originally appeared in Knowable Magazine, an independent journalistic endeavor from Annual Reviews. Sign up for the newsletter.
