Session G05: Physics of Bacterial Communities: Interaction and Dynamics

Interactions and structure of native gut microbiome ofC. elegans

Microbial communities are finite non-equilibrium systems, where the underlying interaction networks play out under environmental perturbation and drift due to migration and demographic stochasticity. The interactions among microbes (direct) and between microbes and the host (indirect) form a network, and efforts have been devoted to extracting these networks of interactions from co-occurrence data. However, how these interactions between microbes collectively form a microbiome is still not completely known. Here, eight bacterial species from a native C. elegans microbiome were cultivated in vitro and used to colonize sterile nematode gut. From in vitro data, we observed a range of inter-species interactions, and colonizing the gut with individual species revealed intestinal populations that increase over time, consistent with a simple Lotka-Volterra-type model for saturating growth. These bacterial populations are apparently multimodal across populations of homozygous, synchronized hosts. Using the experimental data to fit a simple mathematical model, we aim to understand the contribution of direct vs. indirect interactions in minimal native microbiome assembly, considering the contributions of noise and of heterogeneity across individual hosts.   

Compromising matrix structure to make biofilm bacteria easier to eat

Bacterial biofilms cause persistent and dangerous infections which are resistant to standard antibiotics and the immune system. A biofilm is a community of microbes that are embedded in a matrix of extracellular polymeric substances, or EPS, which protects the microbes from antibiotics and immune cells, making the eradication of biofilm infections difficult. Neutrophils are immune cells that use phagocytosis to engulf and digest microbes. Neutrophils have difficulty engulfing bacteria from the biofilms because of the embedding EPS. To remedy this, we use polymer-specific enzyme treatments matched to the dominant polymer produced by a bacterial strain—these treatments cause changes in the bulk mechanical properties of the biofilm, aiding phagocytosis and leading to higher engulfment rates. We utilize a protection-based assay, wherein bacteria from biofilms which have not been successfully engulfed by neutrophils are killed with antibiotics. The neutrophils can then be lysed and the “surviving” bacteria counted as a proxy for phagocytic success. Biofilms also may be stained with a pH-sensitive dye, which fluoresces in highly acidic environments, like inside a neutrophil, allowing discrimination between bacteria which are fully internalized and those which are bound to a neutrophil’s surface.   

Spatially localized adhesin sharing confers ecological stability in V. cholerae biofilms

Biofilm is an important life form of bacteria. Central to the evolutionary advantage of biofilm formation is the adhesion of biofilm to surfaces, which endows biofilm-dwelling cells with fitness advantages such as access to nutrient sources in their natural habitats and colonization in hosts. Adhesion is achieved by biofilm-specific adhesins secreted into the extracellular space, raising the question of whether the adhesins are exploitable public goods and how sharing and exploitation of adhesins may shape the ecology of biofilms. In this work, using Vibrio cholerae as the model biofilm-former, we show that biofilm adhesins are indeed susceptible to cheater exploitation, and the evolutionary stability of adhesin production depends on the spatial structure of biofilms and the flow conditions in the environment. We further show that exploitation of adhesins is localized within a well-defined spatial range around adhesin producers, which depends critically on the diffusion and advection of adhesin molecules. Based on the exploitation range and the spatial distancing between biofilm clusters, we construct a spatial model of exploitation and relate the Hamilton’s relatedness coefficient to these two length scales. Our results show that adhesin production is favored under conditions consistent with their natural habitats and host environments. Compared to other diffusive public goods in bacterial biofilms, adhesin production presents distinct ecological dynamics. We expect the mechanisms revealed in this study to be relevant to other matrix components as cooperative public goods in biofilm-forming microbes, and the concept of spatial range and the associated analysis tools to be generally applicable to other ecological traits.   

Irvin Oppenheim Award (2022): Microbial diversity and competition for resources in a model seasonal ecosystem

We share our world with communities of microbes. These microbes have co-evolved to live in every environment on our planet, comprising an essential part of every ecosystem, from the human body, to rainforest soil and coral reef. When inhabiting a host organism, they provide services essential for host health. A defining feature of microbial life is that wherever they are found, microbes compete fiercely for limited resources, a competition possibly as old as life itself. A second defining feature of microbial life is its astonishing diversity. Natural ecosystems typically display a hugely diverse array of coexisting microbial species.

How is microbial diversity maintained, in light of the aforementioned resource competition? Here, we focus on consumer-resource models subject to serial dilutions. Using a theoretical framework we developed, we probe the effects of resource competition, cross-feeding, mutation, and adaptation on diversity in microbial ecosystems. We find that diversity is influenced by these mechanisms in very different ways, suggesting that real ecosystems may not obey a universal nutrient-diversity relationship. Our results can be explained by an early-bird effect whereby species that grow quickly because their preferred nutrients are supplied, then leverage that early advantage - even after those preferred nutrients are depleted and the remaining nutrients are more efficiently metabolized by competitors.   

On the sizes of gut bacterial aggregates

The spatial organization of the gut microbiota influences microbial abundances, inter-microbial competition, and host-microbe interactions. The rules relating bacterial dynamics to large-scale structure remain unclear, however. We therefore studied experimentally and theoretically the formation of three-dimensional bacterial clusters, known to influence susceptibility to intestinal transport and access to the epithelium. Analyzing imaging data for eight different bacterial strains examined in isolation in the larval zebrafish gut, we find a common family of cluster size distributions that decay approximately as power laws with exponents close to -2, becoming shallower for large clusters in a strain-dependent manner. This type of distribution arises naturally from a Yule-Simon-type process in which bacteria grow within clusters and can escape from them, coupled to aggregation that drives the system toward a single massive cluster, reminiscent of gelation in soft-matter systems. We further show that power-law distributions can persist in multi-species gut communities. These results point to biophysical principles governing gut microbiome spatial organization that may be useful for inferring dynamics in experimentally intractable systems, such as humans.   

Oxygen Dynamics in a Vertically Stratified Microbial Community

Microbes in sediment naturally self-organize into distinct layers, each of which is composed of microbes that exhibit a common metabolism. This ubiquitous vertically stratified structure arises as microbes compete for oxidants and carbon sources, which diffuse from the surface. We study the dynamics by which the topmost stratum, corresponding to aerobic heterotrophy, emerges in homogenized salt marsh sediment. Using a fluorescent oxygen detector, we measure the 2D oxygen distribution with sub-millimeter spatial resolution at 1 min intervals over the course of several days. We vary the concentration of carbon sources in the sample. Assuming Michaelis-Menten kinetics for the metabolic rate of the constituent microbes, we invert the measured oxygen concentration for the rate of oxygen consumption. This analysis reveals how the oxygen penetration depth relaxes to steady state as microbes reproduce, die, and migrate through the sediment. These measurements give insight into the stability of microbial ecosystems in sediment.   

The physics of sessile droplets rediscovered in bacterial biofilms

Bacteria spend much of their life time in biofilms – surface-attached, densely packed microbial communities encapsulated in an extracellular polymeric matrix (EPS). This mode of living provides many benefits: Biofilms facilitate social interaction between cells and provide protection against external influences. Yet, surface-attached biofilm formation also involves many inevitable physical consequences. Frictional interaction between cells and substrate surface impacts the radial expansion of a growing colony. In addition, physical cell-cell interactions and their interaction with the encapsulating EPS controls biofilm growth and expansion. It remains often unclear how these microscopic interactions lead to the macroscopic properties of a biofilm, such as growth dynamics and macroscopic shape of a colony. In this talk, I will present our newest results showing that the growth of surface attached bacterial colonies can be captured by the physics of sessile droplets. In particular, we demonstrate that bacterial biofilms exhibit an effective surface tension. As colonies grow from a single founder cell we find that colonies achieve interfacial force balance at the biofilm boundary. These physical forces impact colony growth, expansion and interaction between merging colonies.   

Stochastic effects in quorum sensing mediated by extracellular vesicles

Quorum sensing (QS) allows bacterial cells to sense changes in local cell density, and facilitates a wide array of microbial processes, including biofilm formation, regulation of virulence, and horizontal gene transfer. While, traditionally, QS was thought to be mediated solely by freely diffusing autoinducer (AI) molecules, recent experiments have shown that a substantial fraction of AI molecules are packaged in extracellular vesicles. However, the precise role of extracellular vesicles in QS is not well understood. The prevailing model of QS is based on a mean-field reaction-diffusion framework that neglects fluctuations in AI concentration. We present here a fully stochastic reaction-diffusion model of QS that accounts for fluctuations in both freely diffusing and vesicle-bound AI molecules. We find that the packaging of AI molecules in extracellular vesicles amplifies local fluctuations in AI concentration, which can dramatically decrease the QS activation time of bacterial colonies. For systems with multiple bacterial colonies, we find that extracellular vesicles provide an alternate pathway for AI transport between colonies, and may be crucial for QS in extracellular environments with a high rate of AI degradation.   

High-throughput assays show timescale of phagocytic success depends on target toughness

A biofilm is a community of bacteria bound together in a matrix of extracellular polymeric substances (EPS) that can be produced by constituent bacteria or incorporated from the environment. The biofilm matrix protects constituent bacteria from external threats such as antibiotics and the immune system. Neutrophils are phagocytic white blood cells which clear infections by engulfing free-swimming bacteria, but are too small to engulf whole biofilms. Here we validate flow cytometry as high-throughput technique to measure phagocytic engulfment from populations of 20,000 neutrophils per experiment. Rapid data collection allows us to extend our study to neutrophils isolated from the whole blood of three healthy adult volunteers. We investigate phagocytic success using mechanically tunable polyacrylamide hydrogels infused with fluorescent microbeads as a model for bacterial biofilms. The tunability of polyacrylamide gels allows us to investigate the effects of toughness and yield strain independently from elastic modulus. We find that extended incubation times of up to 6 hours allow neutrophils to overcome low toughness of hydrogels and successfully phagocytose the imbedded beads, regardless of the gels’ elastic modulus. We begin extending these techniques to measure phagocytic success of neutrophils after incubation with live biofilms utilizing antibiotics to kill non-internalized bacteria, and pH-dependent stains which fluoresce brightly in acidic environments, e.g. inside the phagolysosome of a neutrophil.   

Bacterial foraging in patchy landscapes

Individual foraging behavior of Escherichia coli via chemotaxis and the emergence of traveling population waves are well understood. How these processes translate to the formation of biofilms and, more generally, the spatial distribution of (meta-) populations in patchy microbial environments remains unclear. To address this question, we fabricated microfluidic devices with 4 parallel arrays of 85 patches connected by corridors where the difference between arrays was the within-landscape corridor width variance (disorder). Using these crystal-like landscapes we followed the spatiotemporal colonization dynamics of E. coli and recorded single cell trajectories (~10⁶) over 24 hours. Interpreting trajectories using the ideal chain model of polymer physics allowed us to decipher the diversity of foraging behaviors in response to the topology of the habitat and local patch/corridor occupancy. Furthermore, we found that higher corridor disorder led to more jamming events and thus aggregation as well as a deviation from the spatial distribution of metapopulations in less disordered landscapes. These results highlight multiscale processes leading to biofilm development and emphasize the role of topology in microbial landscapes for predicting ecological dynamics.   

High dimensional geometry of fitness landscapes identifies master regulators of evolution and the microbiome

A longstanding goal of biology is to identify the key genes and species that critically impact evolution, ecology, and health. Yet biological interactions between genes, species, and different environmental contexts change the individual effects due to non-additive interactions, known as epistasis. In the fitness landscape concept, each gene/organism/environment is modeled as a separate biological dimension, yielding a high dimensional landscape, with epistasis adding local peaks and valleys to the landscape. Massive efforts have defined dense epistasis networks on a genome-wide scale, but these have mostly been limited to pairwise, or two-dimensional, interactions. Here we develop a geometric formalism that allows us to quantify interactions at high dimensionality in genetics and the microbiome. We then generate and also reanalyze combinatorically complete datasets (two genetic, two microbiome). In higher dimensions, we find that key genes (e.g. pykF) and species (e.g. Lactobacillus plantarum) distort the fitness landscape, changing the interactions for many other genes/species. These distortions can fracture a "smooth" landscape with one optimal fitness peak into a landscape with many local optima, regulating evolutionary or ecological diversification, which may explain how a probiotic bacterium can stabilize the gut microbiome.   

Coupling between spatial structure and heterogeneous antibiotic responses through self-generated nutrient gradients in microbial populations

In confined colonies, bacterial metabolism and environmental transport lead to self-generated nutrient gradients and emergent phenotypic structure with spatially heterogeneous cell growth and gene expression. Such structure strongly modulates the response to antibiotics, which, even in single cells, depends on a dynamic interplay between the drug’s effect and regulation of resistance gene expression. Here, we observe the response of spatially extended microcolonies of tetracycline-resistant E. coli to precisely defined dynamic drug regimens in a custom microfluidic device. We find complex and counter-intuitive responses, such as local growth rate increases upon antibiotic exposure and enhanced population-level resistance to subsequent exposures. A mathematical model incorporating direct regulation of resistance genes, metabolism-induced changes in expression, as well as nutrient diffusion across the colony captures these phenomena and thereby uncovers an intricate coupling between drug-induced growth modulation, resistance, and reorganization of the growth pattern through nutrient redistribution. Our results highlight the role of physical mechanisms such as nutrient transport for spatial structure and may inform the design of drug regimens that account for this heterogeneity.   

Active bulging promotes biofilm formation in a bacterial swarm

The interactions between microbes and their environment influence bacterial transport and colonization.  Here we studied cell-fluid interaction in bacterial colonies and the consequence of such interaction on biofilm development.  We found that an originally uniform swarming colony may spontaneously separate into different phases due to cell-fluid interaction, via a process that we called 'active bulging'.  Using genetically encoded fluorescent reporters, we show that the active bulging process promotes biofilm formation.  Our findings enrich the understanding of biofilm formation mechanisms and may inform the manipulation of self-assembly in bacterial active fluids.