麻豆传媒

SEX for the freshwater crustaceans called amphipods begins with an embrace.
Even now, among the muck and slime on the bottom of a pond near you, a male
amphipod is probably clasping his mate in a characteristic clinch. The murky
depths may not seem like a very salubrious setting, but this is the safest place
for such an intimate encounter. So why would an amphipod risk its life by coming
to the surface to embrace a floating twig? What could change this light-hating
bottom-dweller that dives to safety when under attack, into an exhibitionist
that scoots around the water surface and hugs flotsam in public?

The answer is a mind-bending worm known as Polymorphus paradoxus.
This parasite hijacks the brain of its host, forcing it to act recklessly when
attacked. It鈥檚 bad news for the amphipod鈥攚hich is likely to end up inside
a foraging muskrat or mallard鈥攂ut exactly what the parasite needs to
complete its life cycle and produce a new generation of worms. The key to what
is going on lies in that embrace. The way that infected amphipods grab onto
floating twigs or vegetation turns out to be remarkably similar to the male鈥檚
precopulatory mating hug. Simone Helluy and John Holmes of the University of
Alberta, who made this discovery, believe that P. paradoxus creates
crosswiring between an amphipod鈥檚 normal mating and escape behaviours, perhaps
by producing biochemicals that imitate their host鈥檚 neurotransmitters. So
instead of trying to escape to the depths, the hapless crustacean hugs a
floating twig.

Body snatchers used to belong to the realms of science fiction; now they are
the stuff of serious scientific journals. What is emerging is an impressive body
of evidence that some parasitic organisms, like P. paradoxus, are
exquisitely and grimly tailored by natural selection to take control of their
hosts. What distinguishes enslavers from other parasites is that they are not
merely along for the ride. Instead, they actively usurp the host鈥檚
evolutionarily honed machinery, forcing their victims to behave in ways that
promote the parasites鈥� survival at great peril to their own. Body snatchers
plague a wide range of hosts, including insects, fish, birds and
mammals鈥攑ossibly even ourselves.

Understanding how such parasites alter the behaviour of hosts to increase
their own chances of survival is not just the preserve of pond watchers and
curious Darwinians. Doctors and medical researchers also have a vested interest
in them. Parasites that cause human suffering on a grand scale use
body-snatching tactics to make a bigger impression on human lives than they
could otherwise. For example, the protozoans causing leishmaniasis manipulate
the sandfly host that passes them on to humans. By blocking its foregut so that
only the tiniest fraction of each blood meal reaches the digestive tract,
parasites force the hungry sandfly to probe human skin more often, transmitting
vastly greater numbers of protozoans than if the fly were easily satisfied.
Fleas carrying the plague bacillus also bite more, as do mosquitoes carrying La
Crosse virus, the most common mosquito-borne brain infection in the US.

Thirst for blood

Jacob Koella of the Pierre and Marie Curie University in Paris and Mike
Packer of Oxford University found similar behaviour in bloodsucking mosquitoes
infected with malaria. Healthy insects seek fewer meals as the evening
progresses, but those carrying the parasite feed heavily throughout the night.
What鈥檚 more, the malaria parasite has fine-tuned the biting tactics of the
mosquito to a surprisingly high degree. Koella and his colleagues Flemming
S rensen, Robert Anderson and Hilary Hurd have now discovered that when the
parasites are young, or 鈥渙ocytic鈥�, they reduce mosquito blood lust. But once
they mature into sporozoites that can survive in the bloodstreams of
vertebrates, the parasites make mosquitoes bite more hosts more often.
鈥淒ifferent stages of the parasite seem to have very specific manipulations
according to their interests in terms of transmission,鈥� says Koella.

He believes that one possible mechanism underlying this voraciousness is
reduced enzyme activity in infected mosquitoes. 鈥淪porozoite infection of the
salivary glands makes blood-feeding less efficient by decreasing the activity of
apryrase, an enzyme required by mosquitoes to locate blood when they are probing
their host,鈥� he says. Koella suspects that the parasite may also hijack other
chemical processes.

The blood flukes that cause schistosomiasis鈥攚hich cycle between
freshwater molluscs, such as African river snails, and unfortunate vertebrates,
including humans鈥攖ake manipulation of their host even further. Instead of
producing behaviour-altering chemicals themselves, these parasites trick their
snail hosts into doing the dirty work for them. In 1997 a team of researchers
led by Robert Hoek from the Free University in Amsterdam found that the blood
flukes hijack snail DNA, selectively activating genes for neuropeptides that are
probably involved in the animal鈥檚 growth, reproduction and metabolism.
鈥淧arasites are able to adjust vital brain functions through interference at the
level of gene expression in the brain, with neuropeptide genes as the prime
targets,鈥� the team reported. What鈥檚 more, 鈥渢he parasite constantly and
dynamically adjusts gene expression in the host brain in a parasitic
stage-specific manner鈥�. Like the malaria parasite, the flukes adjust their
environment鈥攖he host鈥攖o meet their changing needs.

As reports of host manipulation accumulate, it is easy to believe that
parasites are running the evolutionary show, with hosts just following
directives in zombie-like compliance. Compared to many parasites, hosts evolve
so slowly that evolutionary parasitologists have often assumed that host
countermeasures are, for all practical purposes, unchanging. But this is by no
means the full story. Koella points out that many parasites, including the ones
that cause malaria, do not have time to reproduce within the lifetime of their
hosts. 鈥淚n such cases,鈥� he says, 鈥渁 host may have enough time to evolve a
counter-response to the parasite鈥檚 pressure, so that coevolutionary
interactions鈥攚here a host and parasite evolve in response to each
other鈥攁re likely to become important.鈥� In other words, hosts may be a good
deal less helpless in the face of the parasitic juggernaut, than many
researchers have assumed.

Robert Poulin, a zoologist from the University of Otago in New Zealand, is an
outspoken critic of what he considers to be sloppy thinking. He believes that
host behaviour is best seen as the result of two competing, interacting sets of
genes. In his view, behavioural change is not 鈥渕anipulation鈥� unless the benefits
for the parasite outweigh the costs to the host. He also argues that the term
鈥渁daptation鈥� is often used too loosely. Behavioural changes in a host are only
adaptive to the parasite when they lead to its increased transmission, which is
often not the case. Even when the parasite clearly benefits, Poulin
distinguishes between incidental (passive, evolutionarily serendipitous) changes
and adaptive parasite manipulation of host defences honed through time by
natural selection. Determining who exactly is in charge can be
difficult鈥攏ot least in the case of the humble bumblebee.

Grave diggers

Conopid flies seize foraging bumblebees and lay eggs in their victim鈥檚
abdomen. Once parasitised, bumblebee workers often spend significantly less time
at their nest. This was originally interpreted as host manipulation by the fly
larvae because conopid larvae must pupate wherever their host dies, and the hive
contains bacteria that would harm them. But Poulin points out that a bumblebee鈥檚
nest is home to thousands of its close genetic kin, and that wandering bees
could be protecting relatives and ensuring that more copies of their genes
survive鈥攚hat evolutionary biologists call 鈥渋nclusive fitness鈥�. What鈥檚
more, Christine M眉ller and Regula Schmid-Hempel from the Zoological
Institute in Basel, Switzerland, have recently shown that by abandoning the nest
parasitised bees expose themselves to lower night-time temperatures, which
disrupts conopid larval development.

So it seems that wandering may be an adaptation by the host rather than
parasitic trickery, after all. But the arms race between bumblebees and conopid
flies is far from settled in the host鈥檚 favour. In subsequent research,
M眉ller found that some parasitised bees exhibit an extraordinary and unique
behaviour. Just before they die, they begin digging their own graves. By
entombing themselves in the soil they provide fly larvae with protected
hibernation sites for the winter. M眉ller found that flies developing in
buried bees were heavier and had fewer developmental malformations than flies
forced to hibernate on the surface.

鈥淎n absolute win by either side is unlikely in the coevolutionary arms race,鈥�
says Poulin. Careful analysis, however, should reveal who exactly benefits from
a given behavioural change and so can help clarify whether host adaptations or
parasite adaptations鈥攐r neither鈥攁re responsible. Take the killifish.
When parasitised by trematodes it swims to the water surface, gyrating wildly.
But who鈥檚 in charge? Is the trematode calling the shots so that it can increase
its chances of getting into an avian host? Or is the killifish looking for food
in a riskier environment to compensate for energy used to fight its parasitic
infection? Lab experiments reported in 1996 by Kevin Lafferty of the University
of California, Santa Barbara, and his colleagues showed that infected fish move
to the surface even when they are given unlimited food in the safety of the
depths. It seems like a clear case of body snatching.

Looking for winners and losers can yield some surprises, though.
Entomologists Richard Karban and Gregory English-Loeb from the University of
California at Davis have discovered that butterfly caterpillars infected by the
tachinid fly often switch from a diet of lupin leaves to eating poison hemlock.
Normally the caterpillars do not fare well on hemlock, but this diet increases
the chances of an infected caterpillar surviving to adulthood. Stranger still,
in 1997 Karban and English-Loeb reported that fly pupae hatching out of
hemlock-eating caterpillars are heavier than those coming from lupin eaters. So
a change in caterpillar tastes seems to benefit both host and parasite. Whose
genes carry the preference for hemlock is still a mystery, though Karban and
English-Loeb favour the idea that caterpillars are self-medicating and the
benefits to their parasite may be simply coincidental.

Another way to distinguish host defences from manipulation is to look at the
timing of behavioural changes. Terrestrial pill bugs infected by thorny-headed
worms move from their normal haunt in the shadows out into bright sunlight. The
host may be trying to raise its body temperature and harm the parasite by
inducing 鈥渂iological fever鈥�. But Janice Moore from Colorado State University has
found that the pill bugs only start behaving in this way once their parasite is
mature enough to infect its next host鈥攖he vertebrate predators that are
likely to eat exposed pill bugs. It鈥檚 a sure sign that the worm is in
control.

The feverish malaise that keeps malaria victims in bed with sweats and chills
makes them less vigilant in swatting away mosquitoes and so improves the
parasite鈥檚 chances of making the jump back into mosquitoes. Might not malaise,
then, constitute manipulation of the human host? Koella rejects the suggestion
on first principles: the timing is all wrong. 鈥淐linical symptoms are due to the
destruction of blood cells during the parasite鈥檚 replication, at a time when the
parasite cannot be transmitted,鈥� he says. 鈥淏y the time the transmissible stages
are formed and have finished their development, patients are already on their
way to getting better.鈥�

Poulin points out that this is often the soundest interpretation of
behavioural changes, because it is much easier for a parasite to disrupt the
workings of its host than to harness these to its own end. Such pathological
side effects of infection, he suggests, can be distinguished from genuine
manipulation using the comparative approach. 鈥淐omplex behavioural modification
is unlikely to have appeared independently twice by chance alone,鈥� says Poulin.
The fact that unrelated parasite species with similar needs cause the same
change in their hosts is strong evidence that evolutionary convergence鈥攁nd
so adaptation鈥攊s at play, rather than mere pathology.

Poulin鈥檚 favourite example of convergence is provided by the nematomorphs and
mermithid nematodes. Both induce such a thirst in their insect hosts that they
are compelled to seek out water. Horsehair worms, for example, lead Jerusalem
crickets to backwater haunts where they drown, leaving the worms exactly where
they want to be. 鈥淲hat strikes me as fascinating,鈥� says Poulin, 鈥渋s that
[nematomorphs and mermithid nematodes] have independently evolved similar life
cycles and an apparently similar manipulation of host behaviour.鈥� How, exactly,
do they cause this suicidal thirst in hosts? 鈥淎 chemical secretion, maybe,鈥�
Poulin speculates. 鈥淣obody knows. Looking for similarities in the mechanism used
by both groups to alter host behaviour would be really interesting, but it seems
that insect physiologists have not bothered.鈥�

鈥淪ummit disease鈥� is another example of convergent behaviour in manipulated
hosts. A fungal parasite infecting yellow dungflies and a baculovirus
parasitising moth larvae both cause their hosts to climb to the tops of trees or
bushes and perch in unusual, stereotyped positions before dying. Both the fungus
and virus parasites then rain spores down on new hosts. These strange behaviours
occur right before the host dies, which suggests that the unrelated parasites
have independently evolved the host manipulation.

Of course, parasites converging upon a particular transmission strategy need
not necessarily use the same tactics. Take the trematodes whose goal is for the
clams that carry them to be eaten by oystercatchers. Some infect the bivalve鈥檚
foot, reducing its growth and hence the clam鈥檚 ability to bury itself. Others
cause clams to leave telltale crawling tracks in the mud. Some trick clams into
burying themselves upside down, with the hinges of their shells pointing
downwards and the two valves open upwards in what Poulin describes as 鈥渁n
irresistible invitation for the probing bill of an oystercatcher鈥�. The methods
may vary, but the end result is the same.

So what do we humans have to fear from parasitic manipulation? Is the
increased libido inspired by syphilis, sometimes called 鈥渃upid鈥檚 disease鈥�, a
transmission tactic? Perhaps. Such studies in humans are rare. But recent
research hints that we may not be immune to body snatchers. Czech researchers
led by Jaroslav Flegr at Charles University in Prague found that the parasite
Toxoplasma gondii, which infects the liver and spleen, is associated
with guilt proneness and decreased self-sufficiency in infected humans. Whether
the parasite is to blame, however, is far from clear.

It may well prove impossible to unravel the knot of strands that make us act
as we do. But with much human behaviour so strangely inexplicable, body
snatchers don鈥檛 sound so far-fetched after all.