A Closer Look at Ampulex Compressa

 Ampulex compressa and its host Periplaneta americana


The parasitoid* wasp Ampulex compressa uses a cockroach to provide shelter and a food source and  for its eggs and larvae. The wasp temporarily immobilizes the cockroach Periplaneta americana with a venom-laden sting, which is followed by a second injection of venom directly into the nervous system of the cockroach.  This second sting dampens the cockroach escape responses, rendering it docile.  The wasp leads the “zombified” cockroach into a burrow and lays its egg on the host’s abdomen. The wasp leaves the burrow, sealing the cockroach and her offspring inside.  When the larva hatches from the egg, it feeds on the cockroach hemolymph for several days, later penetrating and crawling into the still living body of the cockroach to finish its maturation process. Once fully matured, the wasp exits the lifeless shell of its host and goes on to perpetuate the cycle.

*A parasitoid is a parasite that uses its host for the development process, living inside it or on it until mature, eventually killing the host (Hoffman et al., 1993).

An Effective and Precise Sting

Through many generations of evolutionary success, Ampulex compressa’s stinger has been refined into a complex and precise weapon. Successfully paralyzing a host is no small feat, let alone keeping that host alive for an extended period while the larva pupates. The success of this unique reproductive strategy relies upon a perfectly executed pair of stings utilizing a potent neruromodulating venom.

Once a host is identified, the female jewel wasp will sting it in the pro-thoracic ganglion, which induces a transient paralysis of the front legs (Gal 2005). The active chemical components of this venom are GABA, beta-alanine, and taurine (Moore 2006). GABA is an inhibitory neurotransmitter, while beta-alanine and taurine act as agonists on GABA receptor chloride channels, which means that they activate GABA receptors (Moore 2006). The combination of these three chemicals mediates inhibitory post-synaptic potentials in the pro-thoracic ganglion, temporarily blocking efferent motor output (Moore 2006). The venom cocktail seems to be perfectly calibrated for this task: experiments have shown that while each of these constituents is able to block action potentials to some degree, the combination yields the most complete block (Moore 2006).

During the brief 2-3 minute window, the wasp is able to place a second more accurate sting directly in both the brain and the sub-esophageal ganglion (Haspel 2003). Experiments have shown that this sting is extremely precise; venom is concentrated almost exclusively in the brain and sub-esophageal ganglion (Haspel 2003). This implies that the wasp is able to discriminate between nervous and non-nervous tissue, probably by the use of specialized sensors on its stinger. Sensilla-like structures have been identified around the stinger, which might enable the wasps to detect nervous tissue (Gal 2005).

This scanning electron micrograph shows the sensilla-like structures, densely concentrated around the distal end of the stinger. The arrow in the final picture points to the structures, which are less than 1µm in diameter (Gal 2005).

The Second Sting

The second sting occurs during the brief, 2-3 minute interval in which the cockroach is temporarily paralyzed.  The wasp injects a cocktail of neuromodulators directly into the sub-esophageal ganglion of the host cockroach which induces a series of behavioral changes serving the needs of the wasp.  The ability of the wasp to precisely locate the cockroach brain has been shown experimentally to be the result of the evolution of specialized sensors on the wasp stinger which allow it to locate nervous tissue (Gal 2005).  There was some debate as to whether the sting occurred in the general vicinity of the sub-esophageal ganglion or directly into it.  It was known by 1997 that the venom consists of a mixture of peptides and proteins which would have difficulty crossing the sheath which surrounds the ganglia, and the specificity of the sting location was shown by Leibersat et al. in 2003.  Wasps were injected with C14 radio-labeled amino acids, which in turn were incorporated in the venom. They observed that wasps stung a brainless cockroach 15 times longer than the control, while using minimal venom, demonstrating a “targeted” delivery of venom directly to the cockroach central nervous system.

This image shows the precise second sting directly into the prey’s nervous tissue (Haspel 2003)

Haspel 2003

This second, targeted sting induces a period of intense grooming by the cockroach for roughly 30 minutes, followed by a long period of hypokinesia, or extreme lethargy.  Grooming in insects serves a variety of functions, from cleaning the outer surface of the body of dirt and ectoparasites to social and reproductive functions.  The venom of Ampulex compressa elicits the full range of grooming behaviors in the cockroach, and those behaviors, which involve complex and coordinated movements and well as systematic cleaning rituals, does not differ from normal cockroach grooming except in its long duration (Weisel-Eichler 1999).

There are several theories on the purpose of this grooming period; however there is no evidence to suggest that the grooming is of adaptive significance.  While it is tempting to assume that the excessive grooming is designed to clean and prepare the cockroach body for oviposition, it is equally likely that it is a coincidental side effect of the venom, or possibly serves to ensure that the cockroach stays stationary while the wasp searches for a burrow in which to entomb it (Leibersat 2009).

Studies have demonstrated that the excessive grooming is not the result of the stress of the attack, mechanical irritation, or by venom injected into parts of the cockroach body other than the head.  Therefore, it has been concluded that the wasp venom acts directly on the neural network responsible for moderating grooming.  Specifically, it is thought that the venom contains a monoamine (dopamine, octopamine, or a related, possibly more potent neurotransmitter) which is responsible for modulating grooming behaviors (Weisel-Eichler 1999).  The vastly different effects precipitated by the first and second sting are a particularly intriguing feature of this interaction.  The same venom that induces transient paralysis when injected into the pro-thoracic ganglion causes a drastic shift in the personality of the insect when injected into the brain.

Following this grooming phase, the wasp returns to the cockroach, cuts off its antennae, and feeds on its hemolymph.  The wasp then leads the cockroach, which puts up no resistance, to the burrow, where it is entombed in this hypokinetic state for the remainder of the wasp offspring’s development.  This hypokinesis, which is the result of changes in the post-synaptic potentials for initiating certain motor functions, also lowers the metabolism of the cockroach.  While the host does not eat or drink for the remainder of its life (roughly 4 weeks from the time the egg is laid), it is able to serve as a ready source of nutrition for the developing wasp larva.

Interestingly, and in contrast to the effects of other examples of neurotoxic venoms, only a specific subset of host cockroach behavior is modulated by Ampulex compressa venom.  In the laboratory, stung cockroaches demonstrate a five-fold decrease in the firing rate of octopaminergic neurons, which play a large role in escape motor response.  Specifically, thoracic premotor neurons exhibit drastically different behavior than in non-stung experimental groups.  For example, they show decreased resting potential, a decrease in spontaneous firing, larger amplitude action potentials, and a dramatic decrease in response to physical stimuli (Leibersat 2009). However, while escape responses are greatly dampened, stung cockroaches appear to maintain their ability to fly and groom normally (Fouad 1995).

In essence, the second sting seems to mediate the interface between sensory input and motor output by increasing the threshold potential for initiating flight responses.

Larval Development

After stinging the cockroach, the wasp lays an egg and places it on the cuticle of the coxal segment of the metathoracic leg (see image below, box A). The wasp larva hatches within 2-3 days and excavates a shallow hole into the cockroach exoskeleton and feeds off of its hemolymph for 2-3 days. After this feeding period the larva then bites a large hole along the soft cuticular joint of the abdomen and moves into the body, feeding on the host’s organs for 3 days (Gal 2009). During this larval development, the hypokinetic cockroach’s metabolism is depressed, as indicated by lower than normal oxygen consumption (Haspel 2005), ensuring that the host will be able to survive long enough to provide the larva with a viable food source.  In fact, the wasp larva consumes the cockroach internal organs strategically in order to prolong the host’s life (Haspel, 2005).  The mechanism by which the larva engages in this selective feeding is not known at this point.  After 8 days of larval development, the wasp begins its 5 week pupation in a cocoon consisting of the now empty cockroach exoskeleton. When the wasp reaches maturity, it breaks through the shell and exits the burrow.

The Larval Stages of Ampulex Compressa

One Response to “A Closer Look at Ampulex Compressa”

  1. Hello again,

    This page is also very well done! The only suggestion I have is that maybe you could find a video of the wasp in action because I think it would be a really cool way for us to get a picture of how quickly and accurately the wasp attacks. If there isn’t one out there though, it’s not a big deal; the page still looks awesome!

    – Madelyn

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: