A Fluke of Nature

The Amazing Story of the "Brainworm"

by Panini

Life Cycle

The Lancet Liver Fluke (species: Dicrocoelium dendriticum , class Trematoda ) is so called because of the lancet shape of its adult form, and the fact that it resides in the liver of its primary host (also called the "definitive host"). As alluded to earlier, its life cycle is complex and multiphasic. The basics of the life cycle were worked out nearly 40 years ago in [1] , and further refined in [2] . [ Life Cycle Diagram]

The lancet fluke infects its primary mammalian host (e.g., a sheep), when the host ingests infected ants along with the grass it is grazing. The infected ants contain fluke metacercariae (cysts with the pre-larval fluke) inside.

Once inside the gastrointestinal system of primary host, the metacercariae "excyst", becoming baby worms. They then locate and penetrate into a vein, using the blood transport system of the host as a sort of public transit system, and eventually reach the capillaries of the liver. They bore through the walls of these and migrate to the bile duct in the liver, where they take up permanent residence soaking up nutrients from the host for the rest of their lives.

When the worms are sexually mature, they produce eggs which are passed through in the host's feces. Note that the feces likely contains eggs from several different individual flukes. The feces from the primary host containing the eggs are then eaten by land snails (e.g., genera Zebrina and Cionella ) which are the first intermediate hosts.

Once inside the snail, the ciliated larvae (called miracadia) emerge from the eggs. The miracadia grow into mother sporocysts which reproduce asexually into daughter sporocysts. From inside these sporocysts the cercariae larvae emerge. The cercariae do not appear to harm the snail, which is simply being used as a sort of "snail taxi" by the fluke!

These cercariae exit the snails inside of "slime balls" (mucus masses). Each slime ball contains hundreds of cercariae. These slime balls are then eaten by ants species that nest in open, sunny areas (e.g., Formica fusca or Serviformica rufibarbis ). These ants serve as the second intermediate host for the trematode.

Inside the ant, the cercariae bore through the crop 2 wall and enter the gaster 3 . They then close the bore holes with a special material that leaves a telltale dark spot in the crop wall. Typically, a single ant may take in about 50 cercariae; therefore, there will be about 50 bore holes. Note that had the cercariae not repaired the bore holes, it would surely have resulted in the quick death of the ant. With the repairs, even severely infested ants still continue to live and feed normally for at least one year.

The cercariae then develop into metacercariae and cysts within the ant.

The Brainworm

When an infested ant is examined and the bore holes carefully counted, there is always one fewer cyst than bore holes. Further examination reveals the mystery; the missing cyst is located in an unexpected place -- the lower pharyngeal ganglion (LPG) -- where it forms a special, thin-walled cyst know as a "brainworm".

This brainworm has two very peculiar features:

1. It modifies the ant's nesting behavior, so that when the temperature drops (e.g., at night) it does not return to its nest as usual, but instead climbs to the top of a blade of grass or other vegetation, and fastens itself there with its mandibles. Thus exposed, this greatly increases the likelihood of the ants to be eaten by grazing cattle or sheep, and therefore the likelihood of the trematode infecting a primary host.
2. The brainworm itself loses the ability to reproduce or infect the mammal host. Therefore, it is sacrificing itself for the benefit of the other cercariae in the infested ant. Carefully designed experiments show that always exactly one of the cercariae turns into a brainworm, even though all of them are capable of doing so.

So the brainworm increases the probability of being picked up by grazing sheep or cattle, allowing the metacercariae to infect that host and repeat the cycle all over again.

Further Questions

When I first learned about the brainworm in [8] I was utterly fascinated. Three main questions immediately popped up in my mind, and I could not rest until I had the answers. So I spent a couple of days in the Stanford Biology Library researching this topic and tracking down all the relevant scientific papers I could find to try to answer these questions.

My first question was, how does the cercaria get selected to become the brainworm? And how do they ensure that there is always exactly one brainworm, no more, no less. Do they get together and draw straws? Or play a microscopic version of rock-paper-scissors? Is there a mixture of brainworm-capable cercariae, and free-loaders?

The answers to these questions were worked out in [2] :

After having penetrated the crop wall, the parasites inside the gastric cavity of the host go through an inactive stage, which depending on the temperature can last 16 to 25 hours. Then they start migrating through the thorax to reach the head of the host. This cranial migration ends with the formation of the "brain worm", which encysts itself immediately in the LPG. At this point in time, 30% of all parasites on the average have left the hind abdomen, to which they will return in 1.5 - 2 days (caudal migration). This is where they will also envelop themselves with a cyst covering, but much later than the "brain worm". [2]

So, it turns out that all of the cercariae are capable of becoming brainworms. If they are extracted from an infested ant and each one implanted into a separate ant, every one of them develops into a brainworm. Based on certain environmental triggers including temperature, all of the cercariae start migrating to the LPG. The first one to reach the LPG encysts itself there, becoming the brainworm, and this causes all the rest of the cercariae to reverse their migration back to the lower abdomen:

The decisive environmental control factors proved to be the LPG and the "brain worm". It is very probable that the LPG attracts the cercaria, thus controlling the cranial migration. Responsible for the caudal migration is the "brain worm" (possibly working together with the LPG). If one experimentally prevents the formation of a "brain worm", the parasites don't return to the hind abdomen. [2]

My second question was, what are the actual causal mechanisms for directing and modifying the behavior of the ant by the brainworm? Does it sit there in the LPG, pulling strings like some microscopic puppetmaster [10] ? Or push buttons and rewire neurons like some master computer engineer? It turns out the specific mechanisms are not known as yet [Hint: possible thesis topic for an enterprising young biology student!]. However it has been observed that the modified behavior, climbing to the top of grass and clamping down there, is a phylogenetically old sleep behavior [4] . That is, some phylogenetically older species of non-social Hymenoptera (e.g., Ammophila ) typically exhibit this behavior.

The phylogenetically newer Hymenoptera (e.g., Formica ) have developed new sleeping behaviors due to their different nesting patterns. However it is very probable that the "wiring" for the old behavior remains underneath, but is being suppressed and supplanted by new behaviors. In other words, the old subroutines are still there underneath the covers! The brainworm has apparently found a way to unsuppress and reinstantiate the old behavior to its advantage.

My third and final question was, what is the correct evolutionary explanation for the altruistic brainworm.

Evolutionary altruism 4 in nature, defined as any behavior which benefits (increases the chances of survival of) others at the expense of the individual, has always been somewhat troublesome to explain from an evolutionary standpoint. This is because a simple-minded interpretation of Darwinism holds that each individual always seeks to maximize its own survival. However this is too simplistic and not actually implied by Darwinism at all. And it isn't difficult to see that the simplistic interpretaion cannot be true. Numerous counterexamples abound in nature. For example when a bee stings an intruder it subsequently dies, thereby sacrificing itself for the good of the hive.

To explain such cases of altruism, W. Hamilton developed the theory of inclusive fitness [3] . Inclusive fitness says that to evaluate the fitness of an individual, one should include not only its own survival, but also the survival of its genetic relatives, properly weighted according to its genetic distance from the individual. For example, siblings share half of their genetic material. So in the sense of inclusive fitness, one individual is worth two siblings. Similarly, an individual is worth 2 children, or 4 grandchildren, or 8 cousins. This means from an evolutionary standpoint, it would be worthwhile to sacrifice ones own survival, if such behavior ensures the survival of an appropriate number of genetic relatives. This type of mechanism is called kin selection . The kin selection hypothesis holds that whenever such cases of altruistic behavior is observed in nature, the explanation must be that the behavior in question increases the survival of a sufficient number of genetic relatives. Even if it appears to reduce the survival of the individual, it is increasing the survival of its genes, so that its inclusive fitness still rises.

So Darwinism, correctly interpreted to include kin selection, successfully explains most observed cases of altruism. But does it explain all of them? Only if the altruism is between related individuals. If a case was ever observed of altruism between unrelated or distantly related individuals, then kin-selection would fail to explain it. Such cases have been observed, and are described in detail in [8] . (However these still remain controversial; for example see Dawkins' reply [9] to [7] ). The mechanism proposed to explain these cases of unrelated individuals is called group selection . The group-selection hypothesis holds that altruism can in fact evolve between unrelated individuals, under the right set of circumstances. These circumstances can be rigorously defined mathematically.

So, is the brainworm a case of kin selection or group selection? In 1976, Professor Wolfgang Wickler argued for a kin selection explanation:

Such altruism can only be favoured by selection if there is a high degree of genetic similarity between the brain worm and the cercariae that benefit from its death. [...] One need only postulate that the cercariae contained in one mucus ball derive from one sporocyst, that is from one egg of the adult fluke. It seems worthwhile to check this assumption. [4]

I agree wholeheartedly with the last sentence!

Shortly after this paper appeared, D. S. Wilson wrote another paper [5] in response to this one, where he argued against the Wickler position, and demonstrated the theoretical possibility at least of this being a case of group selection. However, in order to prove it one way or the other, we need to experimentally check the degree of relatedness among the cercariae taken in by an ant. The original research done by Hohorst, et al [1] and [2] did not check this. I searched the biology collections at the Stanford library for subsequent research on this topic, but could not find any. Finally, I located Professor Wickler himself -- currently department head at the Max Planck Institut fuer Verhaltensphysiologie -- and asked him two questions: 1. What is his current view, 25 years later, of the 1977 Wilson paper [5] , and 2. Was he aware of any further work in this area, and specifically had anyone checked the assumption as Wickler had suggested in his paper? First, I want to express my gratitude that Professor Wickler took the time out of his busy schedule to respond to my questions. Wickler's reply, in part:

With regard to the trematode: I felt quite happy with D.S.Wilson's 1977 suggestion. A group selection argument might do as well, but I think one will have to check in either case the real relationships among co-occurring trematodes in a given host. I do not know of any study that has tried this. I myself have not continued studies in this direction. [11]

So, the question is still open! Wickler is quite willing to accept a group selection explanation, provided that the actual degree of relatedness is verified experimentally. [Hint: another possible research topic for biology students!] I also am anxious to find the answer to this question because if the cercariae within a single ant do not have a high degree of kinship then the life cycle of this trematode can serve as the definitive model for the group selection argument.

So the story of the brainworm, amazing as it is, still needs to finished; some parts of the story still remain a mystery.

References and Recommendations for Further Reading

[1] Hohorst, W., and Graefe, G., "Ameisen - obligatorische Zwischenwirte des Lanzettegels ( Dicrocoelium dendriticum )", Naturwissenschaften, 48: 229-230 (1961)

The original paper describing the life cycle of this trematode.

[2] Schneider, G., and Hohorst, W., "Wanderung der Metacercariae des Lanzett-Egels in Ameisen", Naturwissenschaften, 58: 327-328 (1971)

A follow-up paper primarily discussing the migration of the metacercariae in the ants and formation of the brainworm.

[3] Hamilton, W.D., "The evolution of social behavior", Journal of Theoretical Biology 7:1-52 (1964)

Classic paper which introduced the concept of inclusive fitness .

[4] Wickler, W., "Evolution-oriented ethology, kin-selection, and altruistic parasites", Z. Tierpsychol. 42: 205-214 (1976)

An excellent ethology paper discussing many examples of altruistic parasites in the context of kin selection. Includes specifically the case of the trematode brainworm.

[5] Wilson, D. S., "How nepotistic is the Brain Worm?", Behavioral Ecology and Sociobiology 2: 421-425 (1977)

An interesting paper partially disputing the kin-selection aspect of [4] and proposing a group-selection explanation instead.

[6] Wilson, D. S., "Structured demes and the evolution of group-advantageous traits", American Nat. 111: 157-185 (1977)

[7] Wilson, D.S. & Sober, E., "Reintroducing group selection to the human behavioral sciences.", Behavioral and Brain Sciences 17 (4): 585-654 (1994).

Paper that introduced many of the ideas later expanded in [8] . Includes the enchanting story of the rowing team crickets.

[8] Sober, E., and Wilson, D. S., Unto Others , 1998, Harvard University Press, Cambridge, MA

An excellent new book discussing altruism -- both evolutionary altruism and psychological altruism -- in great depth.

[9] Dawkins, Richard, "Burying the Vehicle", 1998?

http://www.world-of-dawkins.com/burying_the_vehicle.htm

Dawkins' rebuttal to [7] and [8] .

[10] Heinlein, Robert, The Puppet Masters, 1951, Doubleday & Co., New York, NY

Classic science fiction about a race of alien beings that act as brainworms to humans.

A classic science fiction story about a race of aliens beings that attach themselves to a human being behind the neck, and control their behavior -- brainworms of sorts.

[11] Wickler, W., Personal communication, 5/2000

[12] Dean G. McCurdy, Mark R. Forbes, and J. Sherman Boates, "Evidence that the parasitic nematode Skrjabinoclava manipulates host Corophium behavior to increase transmission to the sandpiper, Calidris pusilla ", Behav. Ecol. 10: 351-357 (1999)

Evidence of another good example of host behavior manipulation by a trematode parasite. In this case a parasite causes a crab (intermediate host) to spend more time near the surface where it is more likely to be eaten by a bird (primary host).

[13] Ohio State University College of Biological Sciences web site:

http://www.biosci.ohio-state.edu/~parasite/home.html

An excellent web site for Parasitology resources.