Scialog: Molecules Come to Life—Understanding the Web of Life
The complex interplay of ecological interactions that leads to stable ecological communities – the dynamic web of life – is challenging to explain, even at the level of single-celled bacteria. However, a collaboration of nine researchers, led by Pankaj Mehta, Associate Professor of Physics, Boston University, and Alvaro Sanchez, Assistant Professor of Ecology and Evolutionary Biology, Yale University, has had recent success finding fundamental, quantitative principles to explain the structure of bacterial communities (or microbiomes) that stably assemble around a single food source despite the organisms’ competition for nutrients. The researchers’ results recently appeared in an article published in Science. (See: http://science.sciencemag.org/content/361/6401/469)
Mehta and Sanchez’s research is supported by a rare, direct collaboration among three prominent science philanthropies. The duo came together and proposed their project on microbial ecology in 2016 at the Scialog: Molecules Come to Life conference in Tucson, AZ, organized by Research Corporation for Science Advancement and cosponsored by the Gordon and Betty Moore Foundation, with additional support from the Simons Foundation. Scialog (science and dialog) is designed to rapidly catalyze new teams of researchers to tackle innovative, interdisciplinary collaborative basic research on globally important problems. Funding for Mehta and Sanchez’s Scialog project was provided by the Moore and Simons Foundations.
Surveys of microbiome composition across a wide range of settings, from the oceans to the human body, have revealed intriguing patterns of organization, Mehta and Sanchez note in their Science article, adding these include high microbial diversity, the coexistence of multiple closely related species with the same function group, and functional stability despite large species turnover.
“However, the lack of a theory of microbiome assembly is hindering progress towards explaining and interpreting these empirical findings,” they write, “and it remains unknown which of the function and structural features exhibited by microbiomes reflect specific local adaptations at the host or microbiome level, and which are generic properties of complex, self-assembled microbial communities.”
With the goal of connecting statistical patterns of microbiome assembly with ecological theory, Mehta and Sanchez and their associates collected dozens of microbiomes found in soil and on plants. All of these naturally occurring ecosystems of bacteria contained substantial species variability, the researchers noted, and recent advances in DNA sequencing enabled them to map the composition of these microbial communities with a high resolution. They then placed these highly diverse microbiomes into tightly controlled conditions in minimal media with glucose added as the only carbon source, i.e., the single food source for the bacteria. They monitored changes in the microbiomes over 80 generations and found that, in all cases, stable coexistence is the norm, and large multi-species bacterial communities would always arise through the production and sharing of new resources.
Moreover, although these communities assembled in identical environments would contain different species, reflecting their different origins, “A similar family-level type composition arose in all communities despite their very different starting points,” the researchers noted. In the initial rounds of experiments these family-level community structures were dominated by members of two families, Enterobacteriaceae (which include pathogens such as Salmonella and E. Coli) and Pseudomonadaceae (which include some disease-causing organisms in plants).
Subsequently, however, microbiomes that were provided with alternate food sources of either citrate or leucine produced communities dominated by different bacterial families. In other words, as Mehta and Sanchez note, “the supplied source of carbon governs community assembly.” Eventually, using machine-learning techniques, they were able to predict with 97 percent accuracy which communities had been fed a specific food source.
“Rather than selecting for the most-fit single species,” they write in the Science article, “our environments select complex communities that contain fixed fractions of multiple coexisting families whose identity is determined by the carbon source [food] in a strong and predictable manner.”
Their results also led them to hypothesize that competitor species can coexist on a single food source, in contrast with classic consumer-resource models which predict that when multiple species compete for a single externally supplied growth-limiting resource, the only possible outcome is competitive exclusion in which a single species dominates. Mehta and Sanchez posit multiple species are able to coexist because the microbes “secrete metabolic byproducts into the environment, which could then be used by other community members.” Further experimentation supported this idea: They allowed microbiomes to exhaust their initial food sources and observed they continued growing by as much as 25 percent, on secretions alone.
Mehta and Sanchez have concluded that microbiomes consisting of “metabolic generalists,” meaning bacteria that can consume both the basic food source as well as the secretions of competing bacteria, tend to create stable communities. And while they note their research did not delve into how pH changes, phage activity or more complex food sources might affect microbiome stability, they suggest that their approach to the mysteries of bacterial communities, when coupled with mathematical modeling, “represent an underutilized but highly promising avenue to reveal the existence of generic mechanisms and statistical rules of microbiome assembly, as well as a stepping stone towards developing a quantitative theory of the microbiome.”
