by Mark O. Martin
An adult hoverfly, also known as a flower fly. Source.
A few weeks ago, I wrote a short Small Things Considered essay describing the diverse roles that odors can play in microbiology. Articles here and there written by others attest to a growing interest in sociomicrobiology. As for myself, I have long suspected that microbes are constantly sending and responding to a wide communication ‘bandwidth’ of rich chemical chatter, and that other organisms can use or eavesdrop on those signals. Such communication is common in the eukaryotic world, conveyed by a variety of chemical messages or semiochemicals with myriad effects.
Thus, I was charmed by a recent article by Pascal Leroy and coworkers, as well as the accompanying Research Highlights essay. The researchers report how it is that the hoverfly Episyrphus balteatus is able to locate its prey, the pea aphid Ayrthrosiphon pisum. As you might guess, reading this blog, there is indeed a microbial component to this story.
Pea aphid. Source.
Aphids, as even the most desultory of gardeners know, are a serious problem in agriculture, infesting everything from houseplants to large stands of crops. It is estimated that these pests are responsible for hundreds of millions of dollars in crop damage annually. Finding a biological solution to this pest is a laudable and evolving (!) goal, one prompting many researchers to investigate the potential of aphid-munching predators, such as the voracious hoverfly, as possible biocontrol agents.
How do hoverflies find the aphids on which they so avidly feed? In the semiochemical lexicon, there must be an ‘eavesdropping chemical’ or kairomone that alerts the predator to the presence of its sap-sucking prey. Most aphids, which feed upon plant sap, produce honeydew, a mixture of amino acids, lipids, and sugars. (Incidentally, since sap is poor in nutrients such as some amino acids, many aphids possess endosymbiotic bacteria that provide them with the missing ones.) This honeydew is a sweet treat to many other organisms, including the famous aphid farming ants, which tend these tiny insect ‘cows’ for their excretions.
It was clear from earlier agricultural research that honeydew contains volatile chemicals that indeed attract predators on the lookout for aphid prey; the nature of these chemicals had remained elusive until recently. Some wondered if these volatile chemicals might be the metabolic products of microbes growing in the honeydew itself.
Hoverfly larva feeding upon an aphid. Source.
Leroy and coworkers not only wondered, but investigated, finding two culturable types of bacteria in pea aphid honeydew: Acinetobacter calcoaceticus and Staphylococcus sciuri, according to 16S rRNA analysis. These bacteria, they showed, are present in wild populations of aphids and not an artifact of laboratory cultivation.
They used solid phase microextraction (SPME) and gas chromatography-mass spectrometry (GC-MS) to identify the volatile compounds in honeydew, finding fifteen including limonene, butanoic acid, and propanone. Any one or more of these compounds might be the kairomone responsible for attracting hoverflies to aphid-infested plants. Filter sterilized honeydew incubated for 24 hours had only a few volatiles, further suggesting the possible involvement of microbial metabolism.
Figure 1: Hoverfly behavior in response to various treatments in a wind tunnel assay. (a) Percentage of individuals recovered on target plants; (b) Number of eggs recovered from target plants. Source.
The next step was to link the volatiles to the bacteria found in the honeydew, and to demonstrate their effectiveness as hoverfly attractants. To that end, the investigators adapted a simple wind tunnel apparatus that allowed them to monitor the movement of hoverflies and how often they deposited their eggs on target plants. The number of eggs laid is a more significant parameter, but both classes of data tell a mutually reinforcing story.
What was gratifying to a microbial supremacist like myself is what came next. Leroy and coworkers showed that hoverflies were attracted to and laid eggs on plants sprayed with aphid honeydew, but not if the honeydew had been filter sterilized (Figure 1). Inoculating the sterilized honeydew with A. calcoaceticus had little effect, but S. sciuri inoculated honeydew attracted hoverflies to the sprayed plants nearly as effectively as honeydew or the aphids themselves! What is more, a lab medium inoculated with S. sciuri and sprayed on the pea plants also attracted hoverflies effectively. Thus, it was clear that the metabolic activity of S. sciuri produced at least one volatile chemical compound capable of acting as a kairomone to hoverflies. Identification of the individual volatile chemicals by SPME and GC-MS revealed compounds readily generated by modification of amino acid-related catabolites and fermentation-associated products. Testing candidate compounds one at a time showed that 3-methyl-2-butanal had the most pronounced effect as assessed by elicited egg laying; some related compounds had the ability to attract the hoverflies in the wind tunnel apparatus.
Finally, the researchers confirmed their results in both greenhouse and preliminary field studies. Nutrient medium inoculated with S. sciuri and sprayed on a potato field attracted hoverflies and led to a higher number of eggs laid by the predators; nutrient medium alone did not have this effect (Figure 2).
Figure 2: Field test of hoverfly behavior in response to nutrient media with and without Staphylococcus sciuri. (a) Number of hoverflies trapped per square meter of target area; (b) Number of hoverfly eggs found per square meter of target area. Source.
Some questions remain, of course. Although successful, the focus of the study was on culturable bacteria, rather than the total microbiota associated with the aphid. It could well be that other nonculturable aphid-associated microbes have a modulating or even different semiochemical effect than S. sciuri alone. In addition, how widespread might this kind of association be in nature? Do other aphid-munching predators also respond to this particular chemical cue? I also wonder if S. sciuri ‘s association with hoverflies isn't a sneaky microbial trick to promote transmission of the bacterium far and wide. Regardless, the lesson of this fine article remains clear: in this particular predator-prey system, a type of bacterium associated with the aphid prey generates a volatile chemical that can act as a kairomone, attracting the predator hoverfly. Understanding the association of this bacterium with the pea aphid in greater detail may provide new and environmentally friendly approaches to control of this pest.
This study did make me ponder the effects of microbial metabolites on more complex behaviors than those found in insects. There is some evidence that microbial metabolism is involved in territoriality and social recognition of hyenas. And it is well known that aspects of human body odor are due to microbial action on human sweat. Consider H.G. Wells, the famous science fiction writer and reportedly something of a ladies’ man. A woman, when asked why the relatively modest looking Mr. Wells was so attractive, replied that he “…had a really interesting smell…” Which brings us back to the possibly disquieting role that microbial metabolic products may have on behavior—even our own. We may find that we are, from time to time, marionettes dancing on microbially manipulated strings!
Reference
Leroy PD, Sabri A, Heuskin S, Thonart P, Lognay G, Verheggen FJ, Francis F, Brostaux Y, Felton GW, Haubruge E. (2011). Microorganisms from aphid honeydew attract and enhance the efficacy of natural enemies. Nature communications, 2.. PMID 21673669
Mark is associate professor in the Department of Biology, University of Puget Sound, an Associate Blogger for STC, and a passionate advocate for the Small Things.
Dear Dr. Leroy---I could only discuss a tiny slice of your research, but I am so delighted you seem to think well of the essay! I actually used your article in my microbiology course---several of your figures are *ideal* for teaching students about interpreting data! Best wishes...-MM
Posted by: Mark O. Martin | December 04, 2011 at 12:14 PM