Neuroethology of Parasitoid Wasps

In a few species of predatory wasps, venoms appear to act centrally to manipulate various behaviors of the prey. With the neuronal chemical manipulation of its host behavior, wasp venom opens a window to look for the neuronal underpinnings of animal behavior, a central theme of the field of Neuroethology. In this article, I focus on one case study where a wasp hijacks the brain of its host to control its motivation to perform specific behaviors.

[edit] Introduction

Most predators kill their prey and consume them immediately, and most venomous predators are no exception to this rule. In contrast, parasitic animals do not necessarily kill their host/prey. They often alter the behavior of their hosts in many ways, including phototaxis, locomotion, behavioral fevers, foraging behavior, reproduction and a variety of social interactions, to name a few. Although the alteration of host behavior by parasites is a widespread phenomenon, underlying mechanisms are only beginning to be deciphered. Among the most fascinating examples of behavioral manipulation are those seen in arthropods parasitized by various species of parasitoid wasps. These wasps manufacture venoms to manipulate the host nervous system in ways that are tailored to the developmental needs of their offspring. The direct manipulation of the host nervous system and behavior may take several forms. In some instances the venom is purely paralytic, affecting either the peripheral or central nervous system to induce true paralysis, which can be transient or long-lasting (Piek T. 1990, Libersat and Gal 2007). In other instances, the venom induces a neuro-chemical manipulation of specific behaviors of the host (Adamo 2002, Libersat et al. 2009). Although the alteration of host behavior by parasitoids is a widespread phenomenon, the underlying neuronal mechanisms are only now beginning to be deciphered. As of today, only a few behavioral alterations can be unambiguously linked to alterations in the nervous system.

[edit] Remarkable examples of behavioral manipulation by parasitoid wasps

The most striking manipulation of the host nervous system and behavior is when the venom affects behavioral subroutines to produce finer manipulations of the host behavior. A remarkable example of such manipulation is that of the braconid parasitoid wasp (Glyptapanteles sp.) that induces a caterpillar (Thyrinteina leucocerae) to behave as a bodyguard of its offspring (Grosman et al. 2008). After parasitoid larvae exit from the host to pupate, the host remains alive but displays stunning modifications in its behavior: it stops feeding and remains close to the parasitoid pupae to defend these against predators with violent head swings. The parasitized caterpillar dies soon after while unparasitized caterpillars do not show any of these behavioral changes.

But the most formidable alteration of behavior ever attributed to a parasitoid wasp is probably the Ichneumonid wasp Hymenoepimecis’s manipulation of its spider host (Eberhard 2000). In this exceptional example of host behavioral manipulation, the parasitoid wasp takes advantage of the natural behavior of web waving of its prey to provide the larva with a shelter. Instead of paralyzing and then burrowing the host, this wasp literately coerces the host to build the shelter for its future larva. The wasp stings its spider host, Plesiometa argyra (Araneidae) on the spider's web. The sting evokes a total transient paralysis during which the wasp lays its egg on the paralyzed spider and flies away. Soon after the sting the spider recovers to resume apparently normal activity. It builds normal orb webs to catch prey, while the wasp's egg hatches and the larva grows by feeding on the spider's hemolymph. The larva feeds for about two weeks and just before it is about to kill the spider, a dramatic behavioral change occurs in the spider. The prey, driven by an unknown cause, starts weaving a very unique web with a design apparently tailored to fit the needs of the larva for its next stage in development, the metamorphosis. The new web is very different from the normal orb-shaped web of P. argyra, and is designed to support the larva’s cocoon so that the cocoon is suspended in the air, rather than lying on the ground. In this safe net, the wasp larva consumes the spider, ultimately killing it, and then pupates in the suspended net. Interestingly, if the wasp larva is removed just prior to the execution of the death sentence, the spider continues to build the specialized cocoon web. Hence, the changes in the spider's behavior must be induced chemically rather than by direct physical interference of the wasp larva. The wasp larva must secrete chemicals to manipulate the spider's nervous system to cause the execution of only one subroutine of the full orb web construction program while repressing all other routines. The nature of the chemicals involved in this extreme alteration of spider's behavior is unfortunately unknown.

[edit] A case study in the neural mechanisms of host manipulation

[edit] Ampulex's hunting strategy and offspring development

Figure 1: A. The wasp stings a cockroach in the head. B. A wasp emerges from the cockroach abdomen after pupation. C. Schematic representation of a dorsal view of a cockroach head shows the relative positions of the head ganglia in the head capsule. The brain (br) and sub-esophageal ganglion (SEG) are shown in yellow. The major structures of the brain include the Central Complex (cc, blue), the Mushroom Bodies (mb, green) and the Antennal Lobes (al, red). D. After a radiolabeled wasp stings a cockroach into the head, radiolabeled venom is found in the cockroach posterior to the central complex and around the mushroom bodies of the brain (D) and around the center of the SEG (E). Scale Bar: 0.25 mm.

The best understood manipulation of host nervous system and behavior by a parasite is the case of the Sphecid cockroach-hunter Ampulex compressa (Libersat 2003). After grabbing its cockroach prey (usually Periplaneta americana) at the pronotum or the base of the wing, the wasp inflicts a first sting into the thorax. This sting renders the prothoracic legs transiently (1-2 min) paralyzed and presumably facilitates the second sting into the neck, which is much more precise and time-consuming (Fig. 1A). After the neck-sting is complete, the wasp leaves the cockroach for about 30 minutes and searches for a burrow. During this period, the cockroach is far from being paralyzed but grooms frenetically for about 20 minutes. When this period is almost over, the wasp returns to the cockroach and pushes him around with its mandibles as if to evaluate the success of the sting. This is when another effect of the venom begins to take place, as the cockroach becomes a submissive 'zombie' capable of performing, but not initiating, locomotion. The wasp then cuts the cockroach’s antennae with the mandibles and sucks up hemolymph from the cut end. It then grabs one of the cockroach’s antennal stumps and leads the host to the pre-selected burrow for oviposition, walking backwards facing the prey. The stung cockroach follows the wasp in a docile manner, like a dog on a leash, all the way to the burrow. Then, the wasp lays an egg and affixes it on the cuticle of the coxal segment of the middle cockroach leg. Having its egg glued on the live food source, the wasp exits the burrow and blocks the entrance with small pebbles collected nearby, sealing the lethargic host inside. The larva hatches within two to three days and perforates the cuticle of the cockroach coxa to feed on hemolymph for the next few days. About five days after the egg was laid, the larva moves to the thoracic-coxal junction of the metathoracic leg and bites a large hole along the soft cuticular joint, through which it then penetrates the cockroach. The larva feeds on the internal organs of the cockroach until, two days after entering the host it occupies the entire cockroach abdominal cavity. Pupation occurs inside the cockroach abdomen (Fig. 1B), roughly eight days after the egg was laid. The two stings by A. compressa induce a total transient paralysis of the front legs followed by grooming behavior and then by a long-term hypokinesia of the cockroach prey. In this state, the cockroach remains alive but immobile and unresponsive, and serves to nourish the wasp larva. The long lasting lethargic state occurs when the venom is injected into the head but not when it is injected only into the thorax. Under laboratory conditions, and if not parasitized by the wasp larva, cockroaches gradually recover from this lethargic state within one or two weeks, demonstrating a partial long-term paralysis of the cockroach. In nature, cockroaches probably rarely reach recovery as the A. compressa larva consumes them before the end of this convalescent time.

Ampulex slips her stinger through the roach's neck to inject venom directly into its brain. The unique effects of Ampulex’s venom on prey behavior and the site of venom injection both suggest that the venom targets the prey's central nervous system. To obtain a direct proof of central injection of the venom, so-called "hot" wasps were produced by injecting them with a mixture of C¹⁴ radiolabeled amino acids which were incorporated into the venom. In cockroaches stung by "hot" wasps, most of the radioactive signal was found in the thoracic ganglion and inside the two head ganglia: the supra- and the sub-esophageal ganglia (Fig. 1C, D).

[edit] The transient paralysis of the front legs

The Ampulex venom, is a complex cocktail of proteins, peptides and sub-peptidic components. Of this cocktail, only low molecular weight fractions seem to be responsible for the short lived paralysis of the front legs. Biochemical screening of the active fractions revealed that the venom contains high levels of the inhibitory neurotransmitter GABA, and a GABA receptor agonist β-alanine (Libersat et al. 2009). Another component in these fractions was identified as taurine, which is known to impair the reuptake of GABA by the GABA transporter from the synaptic cleft. These constituents mimic the transient action of whole venom, synergistically causing a total transient block of synaptic transmission at the central synapse through GABA inhibition. The natural release or artificial injection of GABA at a synapse causes opening of chloride channels in postsynaptic membrane. If the sodium channels in the postsynaptic membrane are opened by synaptic release of ACh when GABA-gated chloride channels are opened by venom components, then for each sodium ion entering the cell, a chloride ion will accompany it. The simultaneous entry of a negative ion and positive ion will produce no net change in membrane potential, thus limiting the production of postsynaptic action potentials. The utilization of venom cocktails containing multiple toxins with distinct but joint pharmacological actions has been previously described in venoms of spiders and marine cone snails]. To conclude, for the total transient paralysis, the study of Ampulex venom has demonstrated a novel strategy for venom-induced synaptic block through chloride channel activation.

[edit] The long term hypokinesia

Figure 2: Schematic and simplified drawing of a cockroach nervous system depicting circuitries that affect walking related behaviors. The walking pattern generator that orchestrates leg movements is located in the thorax. It consists of motor neurons innervating leg muscles (6), sensory neurons associated with sensory structures on the legs (not shown) and thoracic interneurons (TIAs; 4), which synapse onto the motor neurons directly and indirectly via local interneurons (5). The TIAs receive inputs from several interneurons. For example, sensory neurons in the antennae or cerci (1) recruit ascending (2) or descending (3 ) interneurons, which converge directly onto the TIAs to ultimately evoke escape responses. In addition, neurons of the pattern generator receive input from thoracic neuromodulatory cells (8). One example of these is the dorsal unpaired median (DUM) neurons, which secrete octopamine and modulate the efficacy of premotor-to-motor (4-to-6) synapses. The neuromodulatory cells, in turn, receive tonic input through interneurons descending from the brain (7) and subesophageal ganglion (SEG) (not shown). This tonic input affects the probability of the occurrence of specific motor behaviors by modulating the different thoracic pattern generators. The wasp A. compressa injects its venom cocktail directly into both cerebral ganglia to modulate some specific yet unidentified cerebral circuitries. Our current hypothesis states that in the SEG the venom suppresses the activity of brain projecting DUM neurons (9), which control the activity of brain descending interneurons (7) that modulate, either directly (not shown) or indirectly via the neuromodulatory cells (8), the walking pattern generator. Hence, the venom injected into the cerebral ganglia decreases the overall excitatory input to the thoracic walking pattern generator. As a result, walking-related behaviors are specifically inhibited and stimuli to the antennae or cerci fail to evoke normal escape responses.

The long term hypokinesia of a stung cockroach is probably the most interesting in terms of host behavioral manipulation, because a stung cockroach becomes a submissive ‘robot', and fails to initiate spontaneous or evoked locomotion. It is almost certain of adaptive value to the wasp, since it enables resistance-free host feeding, transportation to the burrow and oviposition. Such a behavioral manipulation is induced only if A. compressa stings the cockroach in the head ganglia. Hence, the inability of stung cockroaches to start walking cannot be accounted for by a direct effect of the venom on locomotory centers in the thoracic ganglia of insects.

The hypokinetic state is characterized by very little spontaneous or provoked activity, an important hallmark of which being the inability of stung cockroaches to produce normal escape responses. Wind stimuli directed at the cerci, which normally produce strong escape responses, are no longer effective in stung cockroaches. Normally, wind-sensitive hairs on the cerci detect the minute air movements produced by a predator's strike and excite giant interneurons (GIs) in the terminal abdominal ganglion (TAG) to mediate escape running behavior (Fig 2). The GIs activate various thoracic interneurons in the thoracic locomotory centers which, in turn, excite various local interneurons or motoneurons associated with escape running. In addition, escape running can be triggered by tactile stimuli applied to the antennae or other cuticular regions. Antennal and wind information is carried by two distinct populations of interneurons located each at the far and opposite ends of the nervous ganglionic chain to converge on the same thoracic premotor circuitry which controls similar escape leg movements. Studies on stung cockroaches show that the sting affects neither the response of the sensory neurons and associated ascending GIs nor that of the brain interneurons descending to the thorax. Moreover, thoracic interneurons receive comparable synaptic drive from the GIs in control and stung animals. Thus, the ultimate effect of the venom injected into the head ganglia must take place at the connection between the thoracic interneurons and specific motorneurons.

Unlike normal cockroaches which use both fast and slow motoneurons for producing rapid escape movements, stung cockroaches activate only slow motoneurons, which are also important to maintain posture, and do not produce rapid movements. This lack of response of fast motoneurons appears to be due to a reduction in the synaptic drive they receive from pre-motor interneurons. Such reduction could be due to a modulation of a particular neuromodulatory system that controls a specific subset of behaviors. In this case, the venom would chemically manipulate specific pathways in the head ganglia which themselves regulate neuromodulatory systems involved in the initiation and/or execution of movement (Fig 2). Monoaminergic systems are again probable candidates, as alterations in these systems are known to affect specific subsets of behaviors (Libersat and Pflueger 2004). For instance, depletion of the synaptic content of monoaminergic neurons, and especially of dopaminergic or octopaminergic neurons, induces impairment in the ability of cockroaches and crickets to generate escape behavior. Recent studies on identified octopaminergic neurons known to modulate the excitability of specific thoracic premotor neurons in the cockroach thorax have demonstrated that DUM neurons' activity is compromised in stung cockroaches. The alteration in the activity of octopamine neurons could be part of the mechanism by which the wasp induces a change in the excitability of thoracic premotor circuitries.

[edit] Ampulex manipulates the motivation for walking of its cockroach prey

Unlike most paralyzing venoms, Ampulex’s venom seems to affect the ‘motivation’ of its host to initiate movement, rather than affecting the motor centers. Indeed, the wasp injects its venom directly into brain areas considered ‘higher’ neuronal centers that modulate, among other things, the initiation of movement (Fig. 1C, D). Thus, the venom-induced hypokinesia could result from an overall decrease in arousal or, alternatively, a specific decrease in the drive to initiate or maintain walking (Gal and Libersat 2008). In fact, the venom specifically increased thresholds for the initiation of walking-related behaviors and, once such behaviors are initiated, affects the maintenance of walking (Gal and Libersat 2008). Nevertheless, the walking pattern generator itself appears to be intact. Thus, the venom, rather than decreasing overall arousal, manipulates neuronal centers within the cerebral ganglia that are specifically involved in the initiation and maintenance of walking. Furthermore, stung hypokinetic cockroaches show no deficits in spontaneous or provoked grooming, righting behavior, or ability to fly in a wind tunnel. Hence, the head sting affects specific subsets of motor behaviors, rather than affecting behavior in general. How this comes about is not completely worked out but we have uncovered some important pieces of the puzzle. The wasp injects its venom directly into the SEG and in and around the central complex in the brain. We have shown that in stung cockroaches focal injection of a potent octopaminergic receptor agonist around the central complex area partially restores walking. Conversely, in controls, focal injection of a selective OA receptor antagonist into the same area reduces walking. However, it appears that the relevant neurons that modulate walking reside in the SEG and send axons to innervate motor centers, such as that associated with the central complex. Within this group of ascending interneurons, some octopaminergic ascending SEG neurons provide dense innervation in the central portion of the brain. Thus, the SEG sting might be affecting the activity of SEG octopaminergic ascending neurons to cause reduced OA levels in the walking centers of the brain (Fig 2). To summarize, the current hypothesis is that venom injection into the head ganglia selectively depresses the initiation and maintenance of walking by modifying the release of OA as a neuromodulator in restricted regions of the cockroach brain.

[edit] References

Adamo SA. (2002) Modulating the modulators: parasites, neuromodulators and host behavioral change. Brain Behav Evol.;60(6):370-7.
Eberhard WG. (2000) Spider manipulation by a wasp larva. Nature. 406(6793):255-6. PubMedID:10917517
Gal R, Libersat F. (2008) A parasitoid wasp manipulates the drive for walking of its cockroach prey. Curr Biol. Jun 24;18(12):877-82. Pubmed ID:18538568
Grosman AH, Janssen A, de Brito EF, Cordeiro EG, Colares F, Fonseca JO, Lima ER, Pallini A, Sabelis MW. (2008). Parasitoid Increases Survival of Its Pupae by Inducing Hosts to Fight Predators. PLoS ONE, 2008; 3 (6):e2276.
Libersat F (2003) Wasp uses venom cocktail to manipulate the behavior of its cockroach prey. J Comp Physiol [A], 189:497-508.
Libersat F and Pflueger HJ (2004) Monoamines and the orchestration of behavior. Bioscience, 54 (1) 17-25.
Libersat, F. and Gal, R. (2007). Neuro-manipulation of hosts by parasitoid wasps. In Recent Advances in the Biochemistry, Toxicity and Mode of Action of Parasitic Wasp Venoms (ed. J. Yoder and D. Rivers), pp 96-114. Research Signpost, Kerala, India
Libersat F, Delago A and Gal R. (2008) Manipulation of host behavior by parasitic insects and insect parasites. Annu. Rev. Entomol. 54:189–207
Piek T. (1990) Neurotoxins from venoms of the Hymenoptera--twenty-five years of research in Amsterdam. Comp Biochem Physiol C 96: 223-33

Internal references

Valentino Braitenberg (2007) Brain. Scholarpedia, 2(11):2918.

Eugene Roberts (2007) Gamma-aminobutyric acid. Scholarpedia, 2(10):3356.

Tamas Freund and Szabolcs Kali (2008) Interneurons. Scholarpedia, 3(9):4720.

Peter Jonas and Gyorgy Buzsaki (2007) Neural inhibition. Scholarpedia, 2(9):3286.

Rodolfo Llinas (2008) Neuron. Scholarpedia, 3(8):1490.

Frank Hoppensteadt (2006) Predator-prey model. Scholarpedia, 1(10):1563.

[edit] Further Reading

Quicke DLJ., (1997). Parasitic Wasps. Chapman and Hall, London. pp 1–470
O’Neill KM., (2001). Solitary wasps: Behavior and Natural History. Comstock Pub., Cornell University, Ithaca and London,. pp 1-406.
Moore J. (2002) Parasites and the Behavior of Animals Oxford University Press, U.S.A. pp 1-338 pages
Zimmer C. (2000) Parasite Rex: Inside the Bizarre World of Nature's Most Dangerous Creatures. Free Press / Simon & Schuster, pp 1-320