09 April 2010

Mind controlling a cockroach

ResearchBlogging.orgThis post was chosen as an Editor's Selection for ResearchBlogging.orgMind control sounds impressive. But if you qualify that by saying, mind control of a cockroach, you might think, “Dude, it’s a roach. How hard can controlling its mind be?”

While roaches may not be among nature’s deepest thinkers, jewel wasps (Ampulex compressa, pictured below) pull off an impressive level of control by stinging the cockroach. The wasp walks the roach to the wasp’s nest, lays an egg on the roach, and the roach remains inside the nest and allows itself to be eaten by baby wasps. Importantly, the cockroach isn’t dying or immobilized or paralyzed – it just isn’t running away. (Video below, from supplementary material for this article.)

Obviously the wasp is injecting some sort of complex chemical cocktail in its venom, but what does it do to the roach’s nervous system? Gal and Libersat perform a series of experiments to pinpoint just how the wasp venom is working.

The wasp stings near the front end of the cockroach, where there are two major cluster of neurons: the brain (which gets called the supra-esophageal ganglion here) and the sub-esophageal ganglion (or, as some countries foppishly call it, the sub-oesophageal ganglion). The brain sits ahead of the “throat,” as it were, and the sub-esophageal ganglion sits just behind it.

Intuitively, you’d expect that the wasp would be targeting the roach’s brain. When we think of the place that actions are started, that’s what we think of. But this is a somewhat backbone-centric view of things, as most invertebrates have taken an “eggs in many baskets” approach to their nervous systems: they’re less concentrated in one place. It was already known that the wasps do sting the brain, but do they also work their hoodoo on the sub-esophageal ganglion?

Gal and Libersat first showed that when they removed the sub-esophageal ganglion from a cockroach, wasps given the opportunity to sting it took much, much longer compared to control. They interpret this as the wasp “looking” for its target, and it just keeps digging around in the roach’s body, looking for the missing neural tissue.

Then, they showed in intact roaches that placing a small drop of venom on the sub-esophageal ganglion significantly impaired the cockroach’s walking and escape running. When the venom was placed on the brain, though, there was no difference in behaviour compared to the control animals.

Not only did they see behaviour of whole animals depressed, when Gal and Libersat recorded from isolated sub-esophageal ganglia in a dish, they saw a significant decline in both spontaneous spiking, and spiking elicited by either wind or touch. And, as anyone who’s ever tried to hit a cockroach knows, these animals are very sensitive to wind and touch are normally very effective at running away from them. It looks like the initial movement away from these stimuli is not broken, but the normal running afterwards is messed up by the venom.

All of these experiments suggest that the sub-esophageal ganglion is really critical for the wasp’s mind control trick. This region is probably involved in controlling the initiation of voluntary locomotion (much like some of the command neurons in crayfish I wrote about a while ago).

But if so much of the wasp’s strange hypnotic power is exerted by controlling the sub-esophageal ganglion, the authors can only speculate on this: why does the wasp sting the roach’s brain?


Gal, R., & Libersat, F. (2010). A Wasp Manipulates Neuronal Activity in the Sub-Esophageal Ganglion to Decrease the Drive for Walking in Its Cockroach Prey PLoS ONE, 5 (4) DOI: 10.1371/journal.pone.0010019

Photo by TGIGreeny on Flickr, used under a Creative Commons license.


Geknitics said...

The guys at radiolab talked about the mind-controlling wasps in their show about parasites.

Daniel Bassett said...

There is something so wrong about this. I remember my old PhD supervisor telling me about this. It must have been presented at a neuroethology conference. He told me that the cockroach goes into continuous self preening mode? Unless one of its antennae gets pulled at in which case it just walks. Does this sound about right?

Zen said...

There's nothing mentioning preening in this particular article, though I haven't done a complete search. I did see Libersat present a talk on this once at one of the International Congress for Neuroethology meetings. Somehow, preening does sound familiar, but I would swear to it.

Daniel Bassett said...

No that was the conference my supervisor went to. I think it was in Germany. But yeah Im convinced they go into preening behaviour.