How Powerful physics tool could help scientists understand complex ecosystems

Your gut is home to microbial madness. Hundreds of trillions of bacteria belonging to countless species interact with one another in complex ways that can both keep you healthy and cause disease. Teasing out these interactions would seem an impossible task. Now, microbiologists have found help from an unlikely source: physics. A new experiment suggests a powerful concept known as a phase transition can predict how complex ecosystems—like those composed of the bacteria in your belly—behave. The finding could help us keep our guts healthy and even protect other complex ecosystems such as rainforests and coral reefs. “It’s a beautiful piece of work,” says Fernanda Pinheiro, a physicist who studies bacterial ecology and physiology at the Human Techno pole, who was not involved with the work. A phase transition in physics works like this: Everything you really need to know about huge collections of particles—whether the 1023 molecules in a glass of water are liquid or solid, for example, or whether the countless atoms in a metal will arrange themselves into a magnet—is often controlled by a few simple factors, such as temperature and pressure. Theorists as far back as the 1970s have similarly suggested two factors—the total number of species and the strength of interactions between species—could predict whether complex living systems, such as the thousands of species in a rainforest, will remain stable. But testing such theories has proved challenging. That’s because there’s usually no way to experimentally manipulate population sizes or interspecies interactions in natural ecosystems. “You don’t have a knob you can turn that causes lions to eat zebras better,” says Steppe Kuhn, an ecologist at the University of Chicago who was not involved with the study. To get at the problem, Massachusetts Institute of Technology (MIT) physicist Jeff Gore and his colleagues created bespoke ecosystems in the lab. They scooped up 24 bacterial species from the soil of a Boston-area nature preserve and swiped another 24 from nematode guts. They grew the microbes together in plastic wells and increased and decreased the concentration of nutrients to manipulate how strongly the different species interacted with each other. The more nutrients, the more the different species competed. The experimental ecosystems went through three distinct phases as the number of species in the mix or the intensity of interactions between species increased. At first, every species’ population remained stable. Then, when the number of species or the interactions between species crossed a certain threshold, the system abruptly entered a new phase in which some species started to die out. As the experimenters kept adding species and ratcheting up nutrient levels, the system crossed into a third phase: The remaining species’ populations began to fluctuate wildly, indicating the ecosystem as a whole had lost stability. The upshot: Just two variables—the number of species and the average interaction strength—determined whether a mishmash of different microbes would be stable or chaotic, says study author HU, a mechanical engineering graduate student at MIT. The paper, published today in Science, is the first to report replicable phase transitions based on species interactions and diversity in communities with more than a tiny handful of species, Kuhn says. Theorists have long suspected that fluctuations like those Gore’s team found can allow large numbers of species to coexist, because as one species’ population crashes to a low level, it could create room for another’s to grow. The study “gives reason to hope that such a phase could also exist in natural communities,” which could help explain why so many species are able to coexist in real-world ecosystems, says Daniel Fisher, a physicist at Stanford University, who was not involved with the work. But in nature, organisms live in environments with complicated spatial structures and outside influences that Gore’s team did not examine, Fisher notes. For example, the gut is divided into different regions and is constantly being flooded with nutrients, chemicals, and water. Because of such complexity, Fisher says, “whether [the finding] is relevant for anything in the real world is very much up in the air.” Nevertheless, the work is “a very important step,” says Ophelia Venturelli, a biochemist at the University of Wisconsin, Madison, who was not involved in the study. The advance could, for example, help researchers design mixes of gut bacteria that will remain healthy and resist takeover by pathogens like Clostridium difficile, which can cause severe diarrhea and pain and even death, she says. As a next step, Venturelli hopes to see researchers document microbial phase transitions in the guts of lab mice or other less artificial ecosystems. “I would be very excited to test some of these ideas that Jeff’s team has discovered in more realistic environments.” The study could also provide a foundation for testing how likely a bacterial community is to develop antibiotic resistance, Pinheiro says. “The fact that they find patterns” in microbial ecosystems, she says, “will inspire a lot of work.”