Microbial mutualism generates multistable and oscillatory growth dynamics

Abstract

Microbial communities typically comprise multiple different species with an intricate network of interactions, ranging from competitive to cooperative, between them. How does the nature of these inter-species interactions impact overall community behavior? While the influence of purely competitive interactions is well-studied, the opposite case of mutualistic interactions—which are also prevalent in many naturally-occurring communities—is poorly understood. Here, we address this gap in knowledge by mathematically modeling a well-mixed two-species community of aerobes and anaerobes having mutualistic metabolic interactions between them. Despite the simplicity of the model, we find that it reproduces three characteristic experimental findings. In particular, in response to changes in the fluxes of exogenously-supplied carbon and oxygen, the community adopts two distinct stable states with differing fractions of aerobes and anaerobes. These states are bistable, capable of arising under identical environmental conditions; transitions between the two are therefore history-dependent and can give rise to oscillations in the bacterial and chemical concentrations. Moreover, using the model, we establish biophysical principles describing how oxygen depletion and nutrient sharing jointly dictate the characteristics of the different states as well as the transitions between them. Altogether, this work thus helps disentangle and highlight the pivotal role of mutualism in governing the overall stability and functioning of microbial communities. Moreover, our model provides a foundation for future studies of more complex communities that play important roles in agriculture, environment, industry, and medicine.

Sparsity of higher-order landscape interactions enables learning and prediction for microbiomes

“…While our approach works well in the simulated and experimental data we applied it to, this method hinges on the assumption that there is a unique steady-state set of community abundances, rather than the potential for multiple equilibria, or more complex late-time dynamics [35][36][37][38]63]. There is therefore scope to consider further generalizations in cases where experimental data suggests that these outcomes are possible.…”

“…We first explicitly state the problem to address: given a set of S species to draw from, there are 2 S − 1 possible combinations, or seed communities, that can be formed, based on the initial presence-absence of species. This initial condition naturally does not capture the full subsequent behavior of the community, which could be extremely complex [35][36][37][38]. Here, we will make a simplifying assumption that, from a given initial condition, all species will approach an equilibrium at late times, such that each species ends up with a relative abundance that only depends on the seed community composition.…”