Can the bacteria in our bodies control our behavior in the same way a puppetmaster pulls the strings of a marionette? I tremble to report that this wonderfully creepy possibility may be true.

The human body is, to some extent, just a luxury cruise liner for microbes. They board the SS Homo sapiens when we’re born and settle into their assigned quarters–the skin, the tongue, the nostrils, the throat, the stomach, the genitals, the gut–and then we carry them wherever we go. Some of microbes deboard when we shed our skin or use the restroom; others board at new ports when we shake someone’s hand or down a spoonful of yogurt. Just as on a luxury cruise liner, our passengers eat well. They feed on the food we eat, or on the compounds we produce. While the biggest luxury lines may be able to carry a few thousand people, we can handle many more passengers. Although the total mass of our microbes is just a few pounds, the tiny size of their cells means that we each carry about 100 trillion microbes–outnumbering our own cells by more than ten to one.

It’s important to bear in mind that you can carry this galaxy of microbes around and enjoy perfect health. These microbes, for reasons that are not entirely clear, behave like well-mannered passengers. They do not barge into the kitchen, take a cleaver to the cooks, and then eat all the food. Aboard the SS Homo sapiens, the crew includes a huge staff of security guards armed with lethal chemical sprays and other deadly weapons, ready to kill any dangerous stowaway (also known as the immune system). For some reason, the immune system does not unleash its deadly fury on the microbes–even when the microbes are fairly close relatives to truly dangerous pathogens.

In fact, our microbial passengers may actually help out the cruise liner’s crew. They can close up the ecological space in our bodies, so that invading pathogens can’t get a solid foothold. Some species in our guts can break down our food in ways that we can’t, and synthesize certain vitamins and other compounds beyond our biochemistry. The genes that the microbes carry–millions of them–expand our biochemical powers enormously.

To understand the human microbiome better, scientists have been cataloging the microbes in and on people’s bodies, and they’ve been sequencing their DNA. (Listen to my recent podcast with biologist Rob Knight for more.) Yesterday, Nature published a head-spinningly huge study on the microbiome from a team of European and Chinese researchers. Lurking in the stool of 124 volunteers, the scientists found, were 3.3 million microbial genes. The scientists identified a core of bacteria species carried in most people’s guts, as well as other species that varied from person to person.

As Ed Yong rightly points out, this study is most impressive as a titanic database. It is not the Theory of Everything for the human microbiome. That will take a lot longer to build, because the microbial ecosystem inside of us is so complex. Individual species don’t just sit in isolation, surviving in their own special way. Microbes cooperate with one another to get the food they need and produce the conditions in which they can thrive. In Microcosm, for example, I write about research suggesting that E. coli–a minor member of the gut ecosystem–may keep oxygen levels low enough for other species to invade and dominate. And it’s not as if there is some Platonic ideal of a microbiome that we all carry around with us from birth to death. The diversity of microbes I carry is different from the one you carry, and they both change over our lifetimes. Every time we take a dose of antibiotics, for example, the balance can change dramatically. And as the diversity of microbes changes, so do its ecological functions.

Which brings me, at last, to the possibility that the human microbiome can become our puppetmaster.

First some background. A lot of parasites have evolved the ability to manipulate their hosts for their own benefit. (I get into more detail about this in my book Parasite Rex and in this segment of the show Radio Lab.)

Very often, the parasites cause hosts to do things that help the parasites, instead of themselves. For example, a protozoan called Toxoplasma needs to get from rats to cats, and to help the process along, it makes rats lose their fear of cats. Parasites can also change the diet of their host as well as the way in which their hosts digest their food. Parasitic wasps living inside caterpillars, for example, cause catepillars to convert the plants they eat into compounds that supply quick energy (good for wasp larvae growing quickly) instead of storing them as fat for their own metamorphosis.

I was reminded of this sinister manipulation by a paper that was published in Science today by Rob Knight and his colleagues. They built on previous research that revealed that mice genetically engineered to be obese have different kinds of microbial diversity in their guts than normal mice. Scientists have found that if they transfer microbes from an obese mouse to a regular mouse that has had all its own germs stripped out, the recipient mouse will develop extra fat. In the case of these obese mice, it appears that the microbes become less efficient at helping the animals digest food, triggering a series of changes that leads the mice to be fat.

Knight and his colleagues discovered a different–and more disturbing–way that microbes can make mice fat. They started out by engineering mice so that they didn’t produce a protein normally found on the surface of gut cells, called TLR5. TLR5 can recognize bacteria, and some studies suggest that the cells can then pass along signals to the immune system, possibly sending a stand-down command so that the immune system doesn’t start trying to kill the microbes (and end up killing gut cells too).

Born without TLR5, mice got 20% fatter than normal. Not only that, but the mice had lots of other familiar symptoms that go along with being overweight, such as high levels of triglyceride, cholesterol, and blood pressure. Without TLR5 exerting its soothing influence, the mice suffered from chronic inflammation, probably thanks to the low-level war they were waging on their microbes. And things got worse for the mutant mice when they had to eat a high-fat diet. They gained more weight on a high-fat diet than regular mice, suffered even more inflammation, and even ended up diabetic.

The obesity of these TLR5-deficient mice was not the result of inefficiency, as in previous studies. Instead, the mice wanted to eat more–about 10 percent more than regular mice. Knight and his colleagues restricted the diet of the mutant to what the regular mice ate. A lot of their symptoms went away. So the change in their behavior was critical to their weight change.

The scientists also discovered that the make-up of the microbial diversity changed significantly in the mutant mice. Were the microbes giving the mice their symptoms? To find out, Knight and his colleagues knocked out the microbes with antibiotics. The mice ate less, put on less fat, and showed less diabetes-like symptoms.

To isolate the effects of the microbes even more, the scientists transferred them from mutant mice into the bodies of ordinary mice that had first had all their own germs stripped out. Remember–these mice have a normal set of TLR5 receptors. The scientists found that the microbes made the recipient mice hungry–and also made them obese, insulin resistant, and so on.

So here we are. Mice with a genetic make-up that alters the diversity of their gut microbes get hungry, and that hunger makes them eat more. They get obese and suffer lots of other symptoms. Get rid of that particular set of microbes, and the mice lose their hunger and start to recover. And that distinctive diversity of microbes can, on its own, make genetically normal mice hungry–and thus obese, diabetic, and so on.

When I first learned of this work, I asked Knight–with a mix of dread and delight–whether the microbes were manipulating their hosts, driving them to change their diet for the benefit of the microbes. He said he thinks the answer is yes.

This discovery doesn’t just have the potential to change the way we think about why we eat what we eat. (Am I really hungry? Or are my microbes making me hungry?) It also provides a new target in the fight against obesity, diabetes, and related disorders. What may be called for is some ecological engineering.

[Update: Links to papers fixed.]