Skip to Main Content

Critter Chemistry

June 30, 2001

Plants to Bugs: Buzz Off!

Plants Use Volatile Signaling Compounds to Fend Off Attack and Possibly Warn Nearby Plants

Sophie Wilkinson

Plants may seem passive in the face of an attack by insects, but they aren't. In fact, plants can marshal elegant defenses in order to do battle with their enemies. And they just might be able to inform their neighbors that they're in danger.

Plants recognize chemicals in herbivore oral secretions and in that way can discriminate between pruning shears and a herbivore.

Plants' security measures fall into two classes. Direct defense entails the expression of defense genes, leading to the production of chemicals such as nicotine or protease inhibitors that are unpalatable or harmful to insects. Alternatively, a plant under attack can rely on an indirect defense. In this scenario, the plant emits volatile chemicals such as terpenes that attract predatory or parasitic insects. If the insects are close at hand, they can turn up in less than 24 hours to take on the organism that is munching on the plant. The plant's protectors might be parasitic wasps that lay eggs in plant-munching caterpillars; when the eggs hatch, the wasp larvae eat the caterpillars.

The alarm call issued by a plant also "tells potential herbivores that they have been discovered, so it has the effect of deterring other herbivores from laying eggs," says Ian T. Baldwin, founding director of the Max Planck Institute for Chemical Ecology in Jena, Germany [Science, 291, 2141 (2001); Nature, 410, 577 (2001)]. "Moths avoid laying eggs on plants that are giving off these volatile signals, either because they want to avoid competition for their babies or because they don't want to put eggs on a plant that is going to attract predators."

Researchers have been unraveling these complex interactions between plants and insects since the 1980s, when Marcel Dicke, professor of insect-plant interactions at Wageningen University in the Netherlands, says he was "the first to show that plants communicate with the enemies of their enemies [Neth. J. Zool., 38, 148 (1988)]. We know that terpenes are involved and also methyl salicylate."

PLANTS HAVE LEARNED not to use such signals without cause. In many species, the hormone methyl salicylate is emitted only when the plant is attacked by insects but not when other types of damage occur, Dicke notes. Apparently, plants recognize chemicals in herbivore oral secretions and in that way can discriminate between pruning shears and a herbivore, he says. Some of the compounds found in the secretions include volicitin, fatty acid derivatives and conjugates, and -glucosidase.

In addition to the communication that passes from plants to insects by means of volatile chemicals, messages may be carried via volatile signals through the intracellular spaces within individual plants. A classic example is the plant hormone ethylene. "There could be a lot of information being trafficked as gas," says Edward E. Farmer, professor of plant molecular biology at the University of Lausanne, Switzerland, who recently authored a minireview about volatile plant signals [Nature, 411, 854 (2001)].

BIOLOGICAL WARFARE Plants emit volatiles such as these when they battle foraging insects.
WITHOUT A PRAYER Dicke studies predatory mites (bottom) that prey on herbivorous two-spotted spider mites.

Clarence A. (Bud) Ryan Jr., professor of biochemistry at Washington State University, says his team discovered this intraplant or systemic signaling process in 1972 [Science, 175, 776]. "We showed that when you wound a tomato plant or potato plant on its lower leaves, the upper leaves start making defense compounds within a couple of hours," he says. "A signal goes from one leaf to another. Later, we identified the signal as a polypeptide, a wound hormone that moves through the plant to turn on genes."

<RYAN'S TEAMis now trying to determine whether signaling with such polypeptides, which he dubbed systemins, is common. "We found it in potato and tomato, and we have found a new set of polypeptide signals in tobacco [Nature, 411, 817 (2001)]. Now we're looking for it in alfalfa and Arabidopsis," a small flowering plant. To date, more than 100 examples of systemic signaling responses have been found in plants, Ryan says, "so it seems to be everywhere."

The systemins bring about production of protease inhibitors that harm the digestive system of insects that chew on the plants. "They get sick and quit eating. It takes them much longer to develop, and during that time, the natural enemies of insects--like birds and other small animals--can come along and eat the insects," Ryan says. "It's a complex plant-animal-insect interaction that's going on there."

Intraplant signals may be transmitted by other types of chemicals, such as methyl jasmonate. Recent work by Yang Do Choi of South Korea's Seoul National University [Proc. Natl. Acad. Sci. USA, 98, 4788 (2001)] focuses on jasmonic acid and its methylation by plants such as Arabidopsis. "Jasmonic acid could be a nonvolatile, standard signal in the plant, and methyl jasmonate could be trafficked as a volatile signal," Farmer explains. "The same hormone could behave in two different ways."

A large number of the volatiles produced by plants possess a chemically reactive ,-unsaturated carbonyl feature, Farmer notes. "Without identifying the significance of this feature, at least two different labs have shown that these molecules turn on defense-related phenomena such as phytoalexin synthesis and the activation of genes typical of cell stress responses." Phytoalexins are antibiotics. Farmer's lab is checking which genes these volatile electrophiles target. In the meantime, he is pondering whether "these molecules are candidates for intraplant volatile signaling and potentially even interplant signaling."

Once scientists better understand plant communication, they may be able to put it to work. In a world leery of synthetic pesticides, for example, the concept of enlisting plants' natural defenses in aid of agriculture is an attractive one. To mimic plants' indirect defense methods, farmers could "spray plants with a compound that triggers the plant to release its bouquet of distress signals, which will attract parasitic or predatory insects," Farmer says.

Some plants emit smaller amounts of volatiles than others, and that can make it hard for parasitoids or predatory insects to zero in on the plants and find their prey in a field, Dicke notes. "If you by chance have a plant cultivar that emits very little, then these laborers are blindfolded and still have to find their victims in this crop," he says, "whereas if you have plants that really cry out loudly and provide clear signals to the biocontrol agents that you release, then that's a big help for these little creatures"--which measure a few millimeters up to a centimeter. "If you can select cultivars that produce larger amounts in an effective way, then that would be a big help in agriculture for providing a durable strategy of combating pests," Dicke says. "And that's something we're aiming at in our program."


Alternatively, genes from plants that can successfully defend themselves against herbivores could be transferred to more vulnerable crops, Ryan suggests.

Farmers could also utilize plants' direct defense protocols. For instance, when a pest attack is anticipated, farmers could spray their crops with a volatile that activates the plants' defense genes. Methyl jasmonate could be used to activate antimicrobial defensin genes in plants such as cabbages, Farmer says. Or, in plants such as oilseed rape, it could activate synthesis of glucosinolates, which break down into poisonous and distasteful compounds.

"IT'S LIKE VACCINATING a person against a disease, except it's less specific," Farmer says. "The plant will protect itself, and it will be difficult for the pest to develop a resistance. The plant turns on a number of genes, and the insect or pathogen that attacks it cannot undergo one mutation that makes it resistant to all those different defense gene products."

That contrasts with insects' response to many conventional pesticides as well as more modern ones such as the toxins produced by a gene borrowed from Bacillus thuringiensis (Bt). A Bt toxin engineered into a crop such as corn doesn't come with an on/off switch. "So all the corn plants are expressing Bt at all times," Baldwin says. "That gives you great resistance, but you won't have resistance for very long, because you are selecting continuously for insects that are able to become resistant to Bt toxin. So what you want is a system that only expresses these Bt toxins--or whatever toxin you use--when the plant is attacked."

Baldwin proposes that the Bt and natural plant defense systems could be combined: Plants could be engineered with genetic regulators known as inducible promoters that react to the volatile signals emitted by plants under attack. Baldwin says his institute is also "looking at promoters that respond to spit factors in caterpillars."

Whether or not such techniques take hold, they are based on firm scientific ground. But there is another sector of the plant communication field that is on somewhat shakier footing. The central question in this arena: Can a plant under attack alert other nearby plants so that they can take steps to defend themselves?

It's an appealing but controversial notion. "I'm skeptical about it," Farmer declares. "But it's a difficult subject. It's in that interim period where we don't know if it's going to turn out to be something fascinating or not."

It's not that Farmer hasn't tried to find out. When he was a postdoc, he and Ryan conducted a lab experiment that showed that a sprig of sagebrush (Artemisia tridentata) placed in a box with a tomato plant turned on defensive proteinase inhibitor genes in the tomato plant [Proc. Natl. Acad. Sci. USA, 87, 7713 (1990)]. The sagebrush was emitting the volatile chemical methyl jasmonate, which controls plant growth.

Farmer interprets this experiment to mean "that sagebrush produces and releases methyl jasmonate to stop other plants from growing underneath or among the branches of the sagebrush."

Ryan, on the other hand, describes the experiment as "the first real documentation of interplant communication." Ryan says more recent evidence backs up his interpretation. He cites a study in which Jennifer S. Thaler, then a postdoc at the University of California, Davis, entomology department, sprayed tomato plants in a field with jasmonic acid and found that wasps killed twice as many caterpillars on treated plants as on those that weren't sprayed [Nature, 399, 686 (1999)]. "It's clear that these signals can be in the air and turn on defense responses," Ryan states.

Another example involves salicylic acid, which increases in concentration when a plant is attacked by pathogens, Ryan says. When salicylic acid is methylated, it becomes volatile. In 1997, Rutgers plant biology professor Ilya Raskin showed that methyl salicylate given off by one tobacco plant could protect another, turning on its defense mechanisms including enzymes that kill viruses, bacteria, and fungi [Nature, 385, 718 (1997)].

BUG SPIT Chemicals such as volicitin tell a plant when it's under attack by an insect.

Researchers have also studied such phenomena in trees. The University of Washington's David F. Rhoades published a paper in the early 1980s describing work with alder and willow trees and was thus the first to report on the so-called talking tree phenomenon, Baldwin says. "He was looking at larvae performance on trees he had infested with lepidopteran larvae and noticed that trees next to them--his controls--also changed in their ability to sustain larvae," Baldwin says. Rhoades found that insects that were fed leaves from the attacked trees--as well as from nearby unattacked trees--grew more slowly than those fed from trees that weren't under attack and were farther away. "He ruled out root connections and concluded that something was being transmitted from wounded trees to nearby unwounded ones and altering their suitability for insects," Baldwin says.

Also in the early 1980s, Baldwin--at the time a Dartmouth undergraduate in the biological sciences lab of Jack C. Schultz--helped carry out work on oaks and sugar maple seedlings. Schultz and Baldwin placed damaged seedlings in a container with undamaged ones and found that concentrations of defensive phenolics and hydrolyzable and condensed tannins increased within 36 hours in the leaves of undamaged trees [Science, 221, 277 (1983)].

Baldwin acknowledges that "there were some problems with our study," and the work "was met with resounding disbelief." As a consequence, "it was almost impossible to work on the topic for a good 15 years," he says.

Research is picking up again, however. Last year, Baldwin and UC Davis entomology professor Richard Karban reported on a study with sagebrush and wild tobacco. They determined that methyl jasmonate released by clipped sagebrush induced increased production of the defensive agent polyphenol oxidase in nearby tobacco plants. And the tobacco plants experienced less than half the leaf damage from grasshoppers and cutworms than control plants [Oecologia, 125, 66 (2000)].

"But sagebrush is the only plant we know that produces this volatile in response to damage," Dicke says. Other compounds may serve the same purpose in other plants, however.

For instance, Junji Takabayashi in the Laboratory of Ecological Information at Kyoto University in collaboration with molecular biologist Gen-Ichiro Arimura; Wilhelm Boland, one of the directors of the Max Planck Institute for Chemical Ecology; and others looked at the effect of volatiles from lima bean leaves infested with spider mites on leaves from healthy lima bean plants [Nature, 406, 512 (2000)]. The experiments, which were conducted in a bell jar, indicate that homoterpene signals may pass between the plants, activating defense genes in the healthy leaves.

IN ANOTHER PROJECT, Dicke studied what might happen to plants next to those damaged by herbivores. "We now have evidence that plants neighboring a damaged plant not only emit airborne volatiles but also emit something from their roots that can be picked up by a neighboring plant," Dicke says. "And then the neighboring plant starts producing volatiles that are picked up by carnivores, and it becomes more attractive to them."

Dicke will publish this work in a special issue of Biochemical Systematics & Ecology devoted to the interplant communication debate. He has just finished editing the issue, which will be published in late August. Dicke says the issue also includes a paper from a British group working with different plants and herbivores that shows a similar phenomenon.

Although Farmer is skeptical about airborne interplant communication, he is intrigued by the possibility of root-to-root communication. But he cautions that such signaling may be related not to attack, but to growth and patterning of plant populations.

Farmer's reservations about interplant communication center on the difficulty of demonstrating the phenomenon in the field as opposed to the controlled environment of a lab. In the natural environment, for instance, one leaf on a plant gets more sun than another, and as a result the leaves have different characteristics. "The background noise of gene expression being different in different individual leaves in the same plant is going to be an obstacle," Farmer says.

A second obstacle, he says, is the design of a suitable assay. "If you are studying plants communicating to insects, you can look at how the insects behave. They will fly to a plant or fly away from it. But when you're looking at potential signaling between plants, it's very hard to make a simple assay."

Baldwin concurs. Until recently, he says, researchers had to rely on evidence such as changes in defense metabolites in plants that are downwind from plants that are being attacked. "That's not the most sensitive way to look for a plant response," he says. But with the advent of the genomics revolution, researchers can now call on microarray studies, which reveal changes in a plant's gene expression.

Farmer also notes that volatiles are diluted in air. That's not an obstacle for insects, which can be "extremely sensitive to pheromones and can track down their mates over large distances. But it's an open question whether plants are going to be able to do the same thing with chemicals that we haven't identified yet."

Baldwin agrees that concentration is the basic problem with the between-plant communication concept. A wounded plant would "have to be able to give off a signal molecule at concentrations that are high enough to effect a change in a receiver plant," he says. Sagebrush is unique in that it produces "three to four orders of magnitude higher amounts of jasmonate than any other plant does." Even at that relatively high level, a neighboring tobacco plant must be no farther than 10 cm away to react to jasmonate cues released by sagebrush.

Dicke is unfazed. "We know that plants can perceive many chemicals from the environment, that they can respond to many environmental cues," he says. He would be surprised if plants weren't "able to eavesdrop on their neighbors and respond to their cries for help." We simply need to find out how, he adds.

<Back To Critter Chemistry


What It's Like To Start Up A Max Planck Institute

Imagine having a scientific society offer to establish a new institute that you can shape as you wish and run for as long as you like. This isn't some overworked professor's fantasy. Instead, it's the model the Max Planck Society has used to set up 80 institutes and five research groups in Germany.

7931baldwin 2Juli01_16
TURNED LOOSE Max Planck Society gave Baldwin free rein to map out the rest of his career, which may well be spent in his institute's new building in Jena.

The society selects an emerging field, finds one of the field's leading lights, and turns him or her loose. "The field must be something that is not replicated anywhere else in Germany or, preferably, in the world," says Ian T. Baldwin, founding director of the Max Planck Institute for Chemical Ecology in Jena, in eastern Germany. "It must be an interface between two or three disciplines or a whole new area by itself. Then the society identifies a person who has pioneered that area, they recruit that individual, and then everything else is up to that individual."

Openings for directors aren't advertised, notes Baldwin, who holds a B.A. in biology and a Ph.D. in neurobiology. He was tapped six years ago, when he was a biology professor at State University of New York, Buffalo.

Directors tend to stay put once they sign on. "It's pretty hard to leave a position like this," Baldwin says. "At SUNY Buffalo, my research vision was pretty much limited to a three-year granting period from the National Science Foundation, or five years if I was lucky." At the society's institutes, on the other hand, the horizon stretches much farther. "Where do you want the field to be at the end of your career? That's the depth of vision that you're allowed to pursue, and that's completely unique in the world," he says.

With such freedom comes the opportunity for a smashup. "It's terribly scary, because you could head off into a totally bogus area," Baldwin admits. But the society limits that likelihood with the help of an advisory board. Directors must retire at 65, at which point it's decided whether the institute continues or gets scrapped, he says. Baldwin works with three codirectors: Wilhelm Boland, who heads the bioorganics department; Jonathan Gershenzon, plant biochemistry and molecular biology; and Thomas Mitchell-Olds, genetics and evolution.

The chemical ecology institute--which investigates chemical signals that mediate the interactions between plants, animals, and their environment--is getting ready to move into a new building this fall. Its design reflects not only the institute's scientific needs but also its geographic location. For instance, Germany "has rules that say that employees may not work in a room for more than two hours where they cannot see daylight," Baldwin says. "That means that when you design a building to house 250 people, it has to have an enormous amount of surface area. That creates buildings that look like microvilli. The building looks like a hand--five big fingers that contain the labs--with some communal areas in the palm."

As a result, the staff isn't able to mingle easily. "So one has to design watering holes that will bring people out of those microvilli to talk to each other," Baldwin says. The communal areas include a large open space for butterfly colonies.

Baldwin has also encountered more fundamental cultural remnants of the former East Germany. He has had to "teach people how to deal with a capitalistic economy. The whole notion that the customer is king is completely foreign."

Other cultural divides Baldwin is trying to bridge have nothing to do with geography and everything to do with scholarly mores. Bringing together researchers including analytical chemists, ecologists, and molecular biologists has been a "remarkable challenge," he says. "They come from very different scientific traditions with very different notions about what a good experiment is." In addition, they speak "completely different languages," Baldwin says. "They almost need a decoder ring to understand each other."

Despite the challenges, Baldwin clearly wants to be nowhere else. But it wouldn't hurt, he concedes, to "have a degree in architecture, engineering, management, business administration, and psychology in addition to being a scientist, because all these skills are very much needed to be successful in starting a new institute."

Chemical & Engineering News
ISSN 0009-2347
Copyright © 2010 American Chemical Society