The wasp that walks cockroaches

Blogging on Peer-Reviewed Research
On Wednesday, I posted about a parasitic wasp that turns a caterpillar into both a living incubator and a zombie-like bodyguard for its larvae. Well, it seems to be a bumper week for wasp research; today, we have yet another demonstration of the amazing tactics used by these macabre parents to provision their young with food.

i-fa9a636fa501ad99f6a79bff831fc0c1-Waspcockroach.jpgThe stunning colours of the jewel wasp (Ampulex compressa) belie its gruesome habits. Its grubs feed on the bodies of cockroaches supplied by their mother. When a female wasp finds a roach, she stings it twice - once in its mid-section to immobilise its front legs, and the second directly into its brain. There, she pumps in a venom that  stupefies the roach and changes its behaviour.

It's not paralysed but it moves sluggishly and shows no desire to flee from danger. In this befuddled state, the jewel wasp can grab the roach by its antennae and walk it around like a dog on a leash.  The wasp leads its roach to its nest, where it seals it up and lays an egg on its belly. Even as the larva hatches and starts to eat the roach alive, the hapless insect doesn't struggle or fight.

Now, Ram Gal and Frederic Libersat from Ben-Gurion University in Israel have discovered how the wasp's venom keeps its victim so sedate but otherwise mobile and healthy - it's an incredibly precise tool that specifically reduces the cockroach's motivation to walk.   

Shock and oar

i-1b7d6acd1f01bfb91d5529c39765a9e8-Jewel-wasp.jpgGal and Libersat placed stung cockroaches into a chamber where half the floor could be electrified. For a normal cockroach, a shock that's strong enough to make its leg muscles twitch is usually strong enough to make it walk to safety. When the roaches are first stung, they behave in much the same way, but their responses soon change.

An hour later, and it takes almost 5 times the voltage to make them scamper even though their leg muscles contract in a normal way. After 4 hours, the shocks need to be 10 times stronger. It also took as many as four shocks in a row to reliably get them moving. In the experiment, their responses only return to normal after 72 hours when the venom's effects started to wear off. Of course, in the wild, it would be far too late for them by that  point.

These results show that it takes a lot more to get the stung roaches to move on their own accord. As Gal and Libersat put it, the stung cockroaches have "a deficit in 'reaching the decision' to walk". They're also less likely to keep on moving once they've started. Gal and Libersat demonstrated this by placing the roaches in a water-filled cylinder, a potentially fatal predicament that they need to escape within minutes.

Normally, cockroaches start swimming as soon as they hit the water, and spend about 90% of their time trying to escape from drowning. And while the vast majority of the stung cockroaches also started to swim, they didn't keep at it for long, kicking for just 10 seconds in every minute. It wasn't that they were more tired. When Gal and Libersat removed the slackers from the cylinder and turned them upside-down, they tried to right themselves by kicking out with their legs, as vigorously as normal cockroaches do.

Targeted weapon

The fact that stung cockroaches could still flip themselves back on their feet shows that the jewel wasp's venom doesn't affect their general motor skills. When they try to right themselves, their muscles show the same degree of activity as those of normal cockroaches in the same dilemma. Nor did the venom affect the roaches' ability to fly. When Gal and Libersat blew air over the wind-sensitive hairs on the insects' back ends, their wing muscles contracted at normal strength and frequency.

Gal and Libersat suggest that the wasp's  venom could affect certain signalling chemicals such as octopamine and dopamine, that are known to affect the movements  and motivations of insects. Indeed, the group has previously shown that the zombie cockroaches can be restored to their active ways by injecting them with octopamine (see video below).

Regardless of the method, it's clear that the wasp's venom is a precision weapon. It doesn't just indiscriminately target the roach's responsiveness or its ability to move. Instead, it's so well matched to the brain of its prey that it only affects  the specific neural circuits that are involved in walking.

Reference: Current Biology 10.1016/j.cub.2008.04.076

Images from Frederic Libersat and Tuan Cao

More like this

Many parasites alter their hosts behavior by small manipulations in the brain that go largely unnoticed by the host until its eaten by another. But not all use so much finesse. This week's parasite is probably the rudest as far as mechanisms go - luckily, their hosts are cockroaches, so we don't…
As this is a Zombie Day on, here is a re-post of one of my old post about one of the coolest parasites ever (from February 04, 2006): I am quite surprised that Carl Zimmer, in research for his book Parasite Rex, did not encounter the fascinating case of the Ampulex compressa (…
One of the coolest parasites ever (from February 04, 2006): I am quite surprised that Carl Zimmer, in research for his book Parasite Rex, did not encounter the fascinating case of the Ampulex compressa (Emerald Cockroach Wasp) and its prey/host the American Cockroach (Periplaneta americana, see…
Humans are now one large step closer to producing armies of undead, zombie hordes who live only to feast on the brains of the living and reproduce by infecting non-zombies with their "zombie virus" through their saliva or other bodily fluids. This is thanks to the work of a team of scientists at…

I was at a small meeting with Frederic recently, and heard him talk and had a discussion with him. He is quite an entertaining speaker and a nice guy!

A compound that affects movement and motivation but not the ability of the muscles to function.

Applicable to Parkinson's?

So if I introduce a nest of these into my apartment, will it take care of my nasty NYC roach problem?

Only if you train the wasps to lead the roaches back onto the street ;)

I thought I remember reading somewhere that Jewel wasps have the the most painful sting of any insect group.

I think the evolutionary angle is that the wasp makes the roach use its energy to come back to the nest instead of the wasp's. And it is more efficient for the roach to use the energy directly, than for the wasp to eat that energy.

This Emerald Wasp looks similar to the domestic Ensign Wasp, which is supposed to be a cockroach parasite. I wonder if it operates similarly? I have found ensign wasps in rooms in Houston that have large cockroaches.

Call me when there's a wasp that can ride the roach back to it's lair.

For an evolutionary angle, perhaps this wasps ancestors killed cockroaches and had to drag them back themselves, a wasp whose sting zombified the roach instead would have a clear advantage.


Ensign, or evaniid wasps, are parasitoids of the cockroach oothecae (egg case) not the adult cockroach.

Totally unrelated biologies, both very cool.