Passive aggression: Parasitic wasp larvae interfere with each other via their host’s host plant

A large white butterfly caterpillar weaves a cocoon around the wasp larvae infesting its body. Photo by EntomoAgricola.

ResearchBlogging.orgI’m embarrassed to admit that I’ve only just gotten around to picking up Carl Zimmer’s book Parasite Rex. It’s turned out to be a wonderful compendium of all the peculiar ways parasites evade, confound, and resist the defenses of their hosts. Some of the wildest cases Zimmer examines, though, are parasites that manipulate their hosts’ behavior.

One grotesque and well-studied example is the wasp Cotesia glomerata. Female C. glomerata wasps inject their eggs into butterfly caterpillars, and when the eggs hatch, the wasp larvae eat the caterpillar from the inside, saving critical organs so the poor thing stays alive the whole time. Then, when the wasp larvae are ready to burrow out of the caterpillar and form pupae to complete their devlopment, they induce the half-dead caterpillar to spin a web around them and stand guard against predators. (In technical language, this life history makes the wasp a parasitoid, rather than a parasite.) Christie Wilcox has written up a fuller description of the whole grisly process, if you want more detail.

That sounds like a pretty incredible set of manipulations for one clutch of wormy-looking wasp larvae, but they’re not all that Cotesia glomerata can do. New evidence published in Ecology Letters suggests that C. glomerata can somehow make the plants that its host caterpillar feeds on less hospitable [$a] to the larvae of another caterpillar-infesting wasp. In other words, the wasp larvae may manipulate not just their host, but their host’s host.

First off, here’s video of Cotesia glomerata in action. Don’t watch this on your lunch break.

Now, the wasp’s plant manipulations. Lots of plants have what are called induced defenses against herbivores like the butterfly larvae that host C. glomerata larvae. Induced defenses are usually protective toxins that plants produce in response to herbivore damage [PDF]. Erik Poelman and his collaborators reasoned that, since C. glomerata can manipulate it’s host’s behavior, the parasites might change how plants respond to herbivory by infested caterpillars.

To test this, the team first had to induce plant responses. They grew Brassica oleracea—Brussels sprouts—plants in the greenhouse, then infested them with either un-parasitized caterpillars of the cabbage white butterfly Pieris rapae, cabbage white caterpillars infected with Cotesia glomerata, or cabbage white caterpillars infected with larvae of the related wasp C. rubecula. Once the caterpillars had nibbled on the plants enough to induce defensive responses, Poelman et al. removed the caterpillars in preparation for the experiment proper.

The team then introduced parasitoid-free caterpillars and caterpillars infested with one or the other parasitoid species onto host plants that had been through one of the three induction treatments, or that had never been exposed to herbivores. They then tracked the development of the caterpillars, and whether or not the wasp larvae inside them survived.

A healthy cabbage white butterfly caterpillar feeds on a piece of broccoli stem. Photo by Sam Fraser-Smith.

Larvae of C. rubecula fared more-or-less equally well no matter what kind of plant their host caterpillar fed on. But C. glomerata larvae had substantially higher mortality when their hosts fed on plants induced by caterpillars infested with the competitor species. While about 50 percent of C. glomerata larvae died if their hosts fed on plants induced by uninfested caterpillars or caterpillars infested with C. glomerata, almost 75 percent of C. glomerata larvae died when their hosts fed on plants that had previously been occupied by caterpillars infested with C. rubecula.

This impact isn’t because the host caterpillars fared poorly—in fact, caterpillars developed a little faster on plants induced by rubecula-infested caterpillars. So somehow, Cotesia rubecula seems to have influenced its hosts in a way that makes their host plants less hospitable to C. glomerata.

Poelman et al. are scrupulous to point out that this effect might not be anywhere nearly as strong in nature—host plants and host caterpillars might be plentiful enough that Cotesia glomerata can simply avoid the competitor species. On top of that, any natural selection that C. rubecula could be exerting on C. glomerata via induced responses in their shared hosts’ host plants is occurring at multiple removes. The effect Poelman et al. documented is probably not an adaptation for competition with C. glomerata so much as a side effect of C. rubecula‘s effect on its host.

So although this result shows that one parasitoid wasp can reach out and influence another through three other organisms—its own host, that host’s host plant, and the other wasp’s host—it’s not clear how strong that impact has been over the evolutionary history of these two Cotesia species. That said, this is a pretty nifty proof-of-concept.


Agrawal, A., Conner, J., Johnson, M., & Wallsgrove, R. (2002). Ecological genetics of an induced plant defense against herbivores: Additive genetic variance and costs of phenotypic plasticity. Evolution, 56 (11), 2206-2213 DOI: 10.1111/j.0014-3820.2002.tb00145.x

Poelman, E., Gols, R., Snoeren, T., Muru, D., Smid, H., & Dicke, M. (2011). Indirect plant-mediated interactions among parasitoid larvae. Ecology Letters DOI: 10.1111/j.1461-0248.2011.01629.x