Final Discussion about infested caterpillar behaveiour – part 04

27 martie 2009

The behaviour of
parasitized hosts changed dramatically after the egression and pupation
of parasitoid larvae. Hosts stopped walking and feeding and remained
near parasitoid pupae. In addition, they performed 10 times more
head-swings than unparasitized hosts during encounters with predators.
As a result, predators were deterred in 58% of the encounters with
parasitized hosts, but gave up in only 15% of the encounters with
unparasitized hosts. It could be argued that this behavioural change
serves the parasitoids as well as the host, because both would suffer
less predation. However, the guarding caterpillar always died shortly
after the adult parasitoids emerged from their pupae. Thus increased
caterpillar survival during the period in which parasitoids pupate does
not result in increased host fitness. Hence, the hosts appear to behave
as a bodyguard of the parasitoid pupae.

The field experiment
further confirmed that parasitoid pupae indeed suffered less predation
in presence of their host. Host defence of parasitoid pupae was
ineffective against hyperparasitoids, but this did not appear to
represent an important parasitoid mortality factor. Possibly, these
specialized natural enemies have adapted to the defending host. We
conclude that the parasitoids, and not the hosts, benefited from the
behavioural changes of the host that appear to be induced by the

It is unlikely that
parasitoids select hosts that showed atypical behaviour at the time of
parasitism as we used unparasitized and parasitized caterpillars
emerging from the same batches of eggs. The sudden cessation of
movement and feeding of parasitized caterpillars upon parasitoid
egression, the increased number of head-swings, and the total lack of
such behavioural changes in unparasitized caterpillars further confirms
this. Hence, the behavioural changes described here are consistent with
the hypothesis that they are induced by the parasites. This begs for an
explanation of how the parasitoid induces behaviour changes in its host
and which stage induces it. Given the long time (2 weeks) between
parasitism and the behavioural change, the adult parasitoid is not
likely to be the inducer. Furthermore, the changes in host locomotion
behaviour were not induced by stimuli from the parasitoid pupae,
because removal of the pupae from parasitized hosts or adding pupae to
unparasitized host did not alter or induce the behavioural changes.
Moreover, the mechanical damage caused by egressing parasitoid larvae
is probably not the cause of the behavioural change. In pilot
experiments, artificially damaging unparasitized hosts did not induce
modified behaviour (F. Colares pers. obs.).

Parasitoid larvae
are known to interfere with host endocrine functions, causing the host
to stop feeding before parasitoid larvae egress [10], [28], [31][35].
Levels of juvenile hormone, ecdysteroids and neurotransmitters (e.g.
octopamine) have been found to increase shortly before parasitoid
egression [33][35]. However, it is not clear whether parasitoid larvae produce these substances in sufficient quantity to change host behaviour [10], [34].
Moreover, the most important behavioural changes in the present study
occur only after the parasitoids have egressed. The egression usually
takes about 1 hour, and the caterpillars do not respond strongly to
disturbance during egression, but only 12 hours after the event. This
casts doubt on the role of the parasitoid larvae in the behavioural
changes. However, when we dissected caterpillars from which parasitoids
had egressed 34 days before, we found 12 active parasitoid larvae
that had remained behind in the host, as has been found in another
system [36].
We hypothesise that these parasitoid larvae are responsible for the
changes in host behaviour. A similar mechanism has been described for
the trematode D. dendriticum [37] and the liver fluke Brachylecithum mosquensis [23],
which both use ants as an intermediate host. One or two of the
parasites migrate to the ant’s brain, where they encyst and are
believed to affect the ant’s behaviour. These so-called brainworms are
not transmitted, and appear to be sacrificed to enable transmission of
their kin [38].
If the parasitoid larvae of the system described here also stay behind
to manipulate the host and do not pupate later, this would represent a
cost of host manipulation: some offspring are sacrificed for higher
survival of their kin [39]. This hypothesis needs further investigation.

There has been
considerable debate on behavioural changes of hosts being true
manipulations by the parasitoid or by-products of infection [2], [4].
Although we do not yet know the mechanisms that induce behavioural
changes in our system, it is clear that the modified behaviour is
beneficial to the parasitoid. Hence, even if behavioural changes were
initially by-products of infection, parasitoids would be strongly
selected to induce these by-products more effectively, and it would be
currently impossible to distinguish between coincidentally beneficial
by-products and parasitoid adaptation [1].


article possible thanks to this team

part 04 – final discussion is soon to come up


Leave a Reply

(insereaza codul din stanga)

Toate drepturile rezervate