Mapping quorum sensing onto neural networks to understand collective decision making in heterogeneous microbial communities

Abstract

Microbial communities frequently communicate via quorum sensing (QS), where cells produce, secrete, and respond to a threshold level of an autoinducer (AI) molecule, thereby modulating gene expression. However, the biology of QS remains incompletely understood in heterogeneous communities, where variant bacterial strains possess distinct QS systems that produce chemically unique AIs. AI molecules bind to 'cognate' receptors, but also to 'non-cognate' receptors found in other strains, resulting in inter-strain crosstalk. Understanding these interactions is a prerequisite for deciphering the consequences of crosstalk in real ecosystems, where multiple AIs are regularly present in the same environment. As a step towards this goal, we map crosstalk in a heterogeneous community of variant QS strains onto an artificial neural network model. This formulation allows us to systematically analyze how crosstalk regulates the community's capacity for flexible decision making, as quantified by the Boltzmann entropy of all QS gene expression states of the system. In a mean-field limit of complete cross-inhibition between variant strains, the model is exactly solvable, allowing for an analytical formula for the number of variants that maximize capacity as a function of signal kinetics and activation parameters. An analysis of previous experimental results on the Staphylococcus aureus two-component Agr system indicates that the observed combination of variant numbers, gene expression rates and threshold concentrations lies near this critical regime of parameter space where capacity peaks. The results are suggestive of a potential evolutionary driving force for diversification in certain QS systems.Multicellular communities of microbes frequently coordinate changes in group behavior, which requires cellto-cell communication via the exchange of extracellular signaling molecules called autoinducers (AI), a process known as quorum sensing (QS) [1]. Microbes produce AI molecules at a default basal rate, and if the local concentration of AI molecules accumulates beyond a certain threshold, the AI production rate is amplified to an activated level via a positive-feedback loop, simultaneously regulating the expression of QS-controlled genes. This two-state QS regulatory network controls a wide array of collective behaviors in microbial communities, such as the formation of biofilms, the regulation of virulence, or lateral gene transfer [1][2][3][4][5].Although quorum sensing has traditionally been viewed as a process associated with homogeneous populations, several results in recent years have called this conventional wisdom into question [6][7][8]. In particular, it is by now well established that real microbial communities are frequently characterized by the stable coexistence of several variant QS systems in the population [9][10][11]. AI molecules produced by cells with one QS variant tend to activate QS in related kin cells that also express that same variant, with the corresponding native cognate receptor. However, when multiple ...

Burst statistics in an early biofilm quorum sensing model: the role of spatial colony-growth heterogeneity

“…While the modelling approach based on individual cells pursued here is very flexible and can accommodate many different features, approaches based on differential equations for chemical concentrations have the advantage that, at least in some limits, analytical solutions can be obtained 26,47 . In addition to these two strategies network-based approaches have been suggested 96 . We finally note that some of the results observed here may have relevance in morphogen-gradient based problems, in which the spatial distribution of the concentration of stimulating molecules can trigger a certain response of tissues.…”

Disruption of microbial communication yields a two-dimensional percolation transition

“…Rudge et al observed the emergence of striking fractal patterns in growing colonies of E. coli [48]. Physical models such as the Hopfield model have also been implemented to predict how the capacity for groups of cells to make decisions scales with the number of cell types [44]. In the spirit of these observations, here we have identified that disrupting the long-range communication of a quorum sensing microbe through interference by a degrader strain yields complex physical behavior characteristic of a percolation transition.…”

“…With the recent advances in systems and molecular biology, many new studies are being conducted to understand the complex behaviors of microbes that arise due to cellular and spatial heterogeneity [9,[26][27][28][41][42][43][44]. Several studies have shown that many cellular systems exhibit emergent phenomena previously studied in non-biological contexts and that can be explained using physical and statistical mechanical rules [45][46][47][48].…”

A neural network model predicts community-level signaling states in a diverse microbial community

“…To quantify the interaction weight between strains, we constructed a mathematical model. The model is based on a set of differential equations and accounts for signal crosstalk by introducing a crosstalk weight for each pair of receptor and signal, similar to models used previously [5,34]. Specifically, the expression of the QS-regulated gene lacZ follows: where the effective concentration of signal as the result of crosstalk is, …”

“…Interactions between network components influence state dynamics and are represented as weights, with the magnitude of the weight indicating the strength of the interaction and the sign of the weight indicating whether the interaction promotes or inhibits activation. We have previously implemented a neural network to theoretically analyze the information capacity within a QS networks composed to multiple Staphylococcus aureus strains [34]. Here we extend these ideas, combining both experimental and theoretical results, to test whether neural network models can be used to predict and control the activation of QS in communities of bacteria producing multiple signals.…”