Is this the last thing that parasite-infected rodents see? Photo by Julie Falk via Creative Commons

The Great Puppet-Master Debate

ByCarl Zimmer
October 02, 2013
11 min read

Life is rough for parasites. Say you’re a tapeworm that only lives in the gut of one species of shark. You start out as an egg inside an adult tapeworm. Your parent releases you and a bunch of other eggs from its body, and its shark host shoots you out of its own body. Now you float in the vast ocean, stretching out on all sides. You are not made for the free-living world. If you don’t get into another host, you will never reach adulthood. Not just any host, but a fish. And not just any fish, but one species of shark. Chances are good, in other words, that you’ll die.

The miserable odds for individual parasites can potentially drive the evolution of something remarkable: the ability of parasites to manipulate their hosts. By controlling their hosts, the parasites can raise their odds of surviving and reproducing.

Scientists have gathered a number of examples of host manipulation over the years. I dedicated a chapter of my book Parasite Rex to some of them, and a group of scientists published a whole book on the topic last year. A lot of these parasites are cool just in and of themselves, whether they infect caterpillars or spiders, but inevitably people want to know whether parasites control us. I put it down to a combination of our species’s narcissism and love of zombie movies. There are a few potential cases of parasites influencing–if not outright manipulating–human behavior. And the most intriguing one of them all involves a single-celled parasite called Toxoplasma gondii.

Toxoplasma gondii forms cysts in people’s brains. Unless their host has a weak immune system, their cysts cause no apparent harm. But they make up in numbers what they may lack in deadliness. Perhaps a billion or more people carry Toxoplasma cysts in their brain. They pick up the parasites in the soil, undercooked meat, or cat litter.

It’s in cats that the Toxoplasma life cycle gets its start. The parasites mate in the intestines of cats and then produce egg-like offspring, which are passed out with cat droppings. The durable eggs can stay viable for months as they wait for their next host–which can be any species of mammal or bird. Those hosts swallow the parasites, which migrate out of their gut and wander their body; the ones that make it to the brain form protective cysts and play the waiting game again. Only if they can get back into a cat’s gut will they be able to take the next step in the Toxoplasma life cycle.

As other cases of parasite manipulation came to light, some scientists wondered whether Toxoplasma might have some tricks of its own. After all, there’s one obvious opportunity for increasing its odds of getting into cats: make its host easier for cats to catch.

Starting in the late 1990s, a number of researchers published evidence indicating that the parasite does, in fact, do this. Most of the work has been carried out on rats and mice. In a number of experiments, infection with Toxoplasma appears to make the rodents less frightened by the smell of cat urine. Some studies even hint at an attraction to the scent of their killer. (I’ve written about some of the research in recent years here and here.)

LIMITED TIME OFFER

Get a FREE tote featuring 1 of 7 ICONIC PLACES OF THE WORLD

Now, in the journal Trends in Parasitology, a team of Australian parasitologists have challenged the Toxoplasma puppet-master hypothesis, in a commentary entitled “Adaptive host manipulation by Toxoplasma gondii: fact or fiction?”
“Adaptive host manipulation by Toxoplasma gondii: fact or fiction?”
“Adaptive host manipulation by Toxoplasma gondii: fact or fiction?”
“Adaptive host manipulation by Toxoplasma gondii: fact or fiction?”

“In our opinion,” the authors declare, “the accepted dogma that T. gondii manipulates host behavior to increase transmission to cats, tells an appealing story but does not stand up to scrutiny.”

They make their way systematically through the published record of experiments on Toxoplasma. They observe that the parasites produce a range of effects on the behavior of their hosts. In some studies, animals become more active, while in others they become less active, and in others they experience no change at all. In one study, scientists found that Toxoplasma impaired a host’s ability to learn, and in another, it didn’t. The same split in results turns up in tests on memory, a preference for exploring new things, time spent near cat urine, and anxiety.

The scientists also attack the elements that make up the manipulation hypothesis. Just because a parasite does something that appears to make it easier prey does not mean that natural selection produced that change to improve its odds of completing its life cycle. Toxoplasma could be altering its hosts as a side effect of infection, not as an evolved adaptation. The authors note that another single-celled parasite called Eimeria also robs mice of their fear of cats, despite the fact that its life cycle takes it from mouse to mouse, not mouse to cat.

If natural selection really had been at work here, it would be necessary for manipulations to actually increase the number of offspring Toxoplasma produced. But no one has ever done a large-scale trapping experiment to see whether cats catch more Toxoplasma-infected prey than healthy ones. This sort of experiment has been done for other species. In an experiment on fluke-infected fish, researchers found they became easier targets for birds because they jumped around near the surface of the water.

On the other hand, some behaviors that seemed like they ought to make hosts more likely to be killed turned out not to. A tapeworm called Hymenolepis dimunata alternates between beetles and rats. When it infects beetles, they spend more time out in the open, where they ought to be easier for rats to find and eat. Despite such expectations, scientists found that infection didn’t actually raise the odds of a beetle getting killed. In the case of Toxoplasma, rodents may lose their fear of cat urine but might still avoid other scents from their predators, such as the smell of cat fur.

The Australian scientists also point out that Toxoplasma has a more flexible life cycle than it’s often given credit for. It has to get into cats in order to sexually reproduce. But it can also clone itself in other species. The parasite can even spread from mothers to their offspring. When scientists survey the DNA of Toxoplasma, they see evidence for a lot of cloning, and not a lot of mixing genes through sex. If the parasite isn’t moving much between cats and their prey, the natural selection for manipulation should be weak.

A lot’s been made of the fact that Toxoplasma winds up not just in rat brains, but in human brains. Some scientists have argued for a whole host of changes to behavior in people who carry the parasite. It’s tempting, for example, to see “cat ladies” as being in the thrall of cat-infecting parasites. The critics think all this speculation is completely unwarranted at this point.

“Given that research into human behavior is based at least partly on findings in rodents,” the authors conclude, “it is vital that we have a good understanding of how rodent behavior is affected by T. gondii, before we extrapolate to other species.”

I reached out to some of the scientists who have done the most prominent work advancing the manipulation hypothesis. While they all agreed that scientists should be careful to avoid assuming adaptations that may not actually exist, they dispute a lot of the claims made by the Australian critics.

“I firmly believe Toxoplasma is a clear case of actual manipulation and that their attempt to dismiss this is a little too naïve and simplistic,” says Joanne Webster of Imperial College. Webster herself is no “manipulation fundamentalist,” as it were–she was the one who did the beetle experiment that the Australian critics use as evidence against manipulations. But she thinks that the evidence for Toxoplasma is strong, and the criticism against its power to manipulate are weak.

Mathematical models of evolution, for example, show that it’s not necessary for there to be a huge boost in cat attacks in order for natural selection to favor manipulation. And while it’s true that Toxoplasma doesn’t need sex to reproduce, sexual reproduction in the long run has many advantages–such as mixing genes together into better combinations.

Webster agrees that it would be great to see whether cats are more likely to kill infects prey than uninfected ones, but that’s a hugely challenging undertaking for all sorts of reasons (such as the rules about animal welfare). Michael Eisen, a Berkeley researcher who has done experiments on Toxoplasma in mice in recent years, put it this way:

It’s an almost impossible experiment to do right. Are you going to infect mice and release them and a control group and see which get eaten? Where would you do that? How would you know the results wouldn’t be different in a different setting?

As for the different results from some studies on Toxoplasma infection, Webster argues that a lot of that may come down to the fact that scientists have run some experiments on lab rodents and others have studied wild ones. Lab animals have been bred for decades away from their natural threats–including cats. If these differences are accounted for, Webster still sees the evidence for manipulation as being strong.

Eisen, on the other hand, thinks that scientists have yet to do enough experiments to make any strong statements about Toxoplasma. “We’re still far away from having done truly definitive studies to characterize the behavior itself,” he said. “Let’s talk about this before we start calling Toxo the parasitic king of behavior manipulation.”

Still, he agrees with Webster that the critics make a weak case when they try to downplay the importance of cats to Toxoplasma’s long-term survival. If Toxoplasma can do so well without having sex inside of cats, he asks, why does it still carry so much genetic machinery for having sex and producing offspring? Eisen calls this argument from the critics “very poor evolutionary thinking.”

Finally, I got in touch with Ajai Vyas of Nanyang Technological University. He’s done some of the most detailed work on how Toxoplasma affects the brains of rodents, finding it zeroes in on emotional circuits. Vyas pointed out that scientists have debated for a long time how you can tell whether the effects of a parasite are an adaptation for manipulating their host. A few criteria have emerged. Is the effect complex? Is it something that well-fitted to a parasite? Does it turn up in different parasite species, suggesting natural selection has favored it repeatedly. By these standards, Vyas argues, it makes sense to look at Toxoplasma as a manipulator.

But it’s also important to bear in mind that the cats and rodents that we see today–even in the so-called wild–are living in a human-dominated world. Both predator and prey in this case have benefited from our company; we’ve moved them around the world, we’ve given them shelter, and we’ve given them–intentionally or not–an abundant supply of food. The manipulations that Toxoplasma originally evolved may have been less ambiguous before its hosts underwent so much rapid change.

“It could be a historical legacy rather than a present adaptation,” says Vyas. “I am not yet clear about this.”

Go Further