Session M37: Microbial Communities I

Deciphering multi-scale interaction networks driving microbial community dynamics and functions

Envision a world where the trillions of bacteria inhabiting our bodies become the frontier of personalized medicine, having the ability to shape our health, performance and even influence behavior. When in harmony, this teeming world of bacteria offers numerous health benefits. However, a shift in this delicate balance can lead to substantial negative health effects due to contrasting evolutionary objectives. Precision engineering of the gut microbiome that can add, remove or modify functional capabilities of the system holds tremendous therapeutic potential for personalized and precision medicine. However, the complexity of this system that encompasses hundreds of species, unknown interaction networks and mechanisms driving these interactions have precluded our ability to effectively manipulate this system to our benefit. A detailed and quantitative understanding of this system would enable the discovery of molecular and ecological design principles of the system as well as novel control knobs for steering the gut microbiome to desired states. By integrating bottom-up construction of microbial communities with computational models, we reveal the networks of interactions driving microbial community assembly, health-relevant metabolite production and human gut pathogens. Our work provides a foundation for exploring and exploiting the interaction networks driving microbial communities for a wide range of potential applications in precision medicine, agriculture and bioprocessing.   

Visualizing bacterial dynamics in transparent soil mimics

Observing the dynamics of soil-dwelling bacteria in their environment is pivotal for our understanding of microbial survival and resource utilization. Soil presents a complex and heterogeneous environment which forces bacteria to adapt their growth and motility based on the morphology of their surroundings. Traditional methods of studying bacterial behavior in soils are inaccessible due to the opacity of natural soil matrices. To address this challenge, we create a transparent porous medium that emulates the structural heterogeneity found in natural soils. Unlike previous studies employing hydrogel-based porous media, our model incorporates cryolite, an impermeable mineral with a refractive index equivalent to water. This enables us to directly visualize bacteria using confocal microscopy in a structured environment that more accurately mimics soils by confining both the microbes and their nutrients. These cryolite-based porous media provide a minimalistic microcosm to test microbial responses to physicochemical changes such as varying soil texture and chemical environment. Our results provide insights into bacterial proliferation and chemotactic migration within soil-like environments, bypassing the issue of opacity of natural soils.   

Interplay of spatiotemporal dynamics and population dynamics of native and engineered bacteria in the gut

Colorectal cancer (CRC), marked by uncontrolled cell growth in the colon and rectum, ranks as the fourth most prevalent cancer and cause of related deaths in the US. Recent studies indicate a key role for the gut microbiome in CRC. As part of an interdisciplinary team, we seek to use genetically modified gut bacteria to detect early CRC and possibly influence its progression. For these engineered bacteria to be effective, they must survive, engraft, and proliferate. We use mathematical modeling and utilize the generalized Lotka-Volterra equations to model interactions between engineered and native bacteria and determine conditions for their survival and growth. Linear stability analysis further enhances our understanding of bacterial dynamics, the effects of diverse microbial interactions on bacterial populations, and potential system instabilities. By incorporating spatial variation, diffusion, and advection into our model, we explore the interplay of spatiotemporal dynamics and population dynamics of the native and engineered bacterial populations. Our results offer valuable insights into the survival, spatial distribution, and colonization potential of engineered bacteria, considering growth rate, motility, and interactions between populations.   

Bistability in dynamics of worm-microbe interactions

Microbial community composition across nominally identical hosts is highly variable. Sources of this variation are not well understood. To explore the variability, we let the roundworm C. elegans be colonized with single bacterial species from its native gut microbiome. We then observed the temporal dynamics of the bacterial population in many individual hosts. We observe a bistable distributions of population sizes across individual hosts. Our analysis of the temporal evolution of these distributions after perturbations suggests that demographic noise and stationary host heterogeneity alone cannot account for the observed variation. To account for this bistability, we suggest that the bacterial growth rate in the hosts should have multiple stable fixed points or alternatively the host has stochastic state switching.   

Diffusible Public Goods in Microbial Communities: Understanding Spatial Dynamics and Implications for Antibiotic Resistance

Microbial ecological interactions are often mediated by the production, secretion, and uptake of metabolites which act as ‘diffusible public goods. Neighboring organisms can potentially benefit from the conferred fitness advantage of the public good without incurring the cost of production. Determining the interaction range of diffusible public goods in spatially structured microbial communities can help us forecast evolutionary and ecological dynamics. With this perspective in hand, we study the outcome of diffusible public goods games in the context of a spatially structured resistant-susceptible microbial community exposed to an antibiotic gradient where the resistant strategy produces and secretes a diffusible public good capable of protecting the susceptible strategy. Motivated by recent studies that show a simple uptake and leak model leads to the emergence of a local interaction range, we use agent based simulations and microfluidic experiments to explore biologically relevant parameter space to predict dynamics as they apply to antibiotics and diffusible public goods. For instance, beta-lactam resistant bacteria are known to produce beta-lactamase enzymes that neutralize the beta-lactam. These enzymes, created internally, can diffuse outside of the cell, offering protection to nearby non-producers. The presence of such diffusible public goods in microbial systems can foster intricate relationships, even leading to co-dependent microbial consortia. Grasping these dynamics is crucial for a comprehensive understanding of antibiotic resistance.   

Flower petal patterns formed by dense colony of cross-feeding bacteria growing on hard agar

Metabolic crossfeeding is widely observed in multi-species bacterial communities. To probe the consequences of such crossfeeding in the spatiotemporal context, we studied a colony comprising of two E. coli strains exhibiting a commensal interaction involving a “producer” which can grow on lactose provided in the agar but excretes galactose, and a “consumer” which can grow on galactose but not on lactose. Contrary to expectations formulated based on batch culture measurements, we observed the emergence of a flower petal pattern formed by the two strains in the colony after initial inoculation as a uniform mixture. Through a combination of experiments and simulations, we elucidated key elements of cell growth and metabolic exchanges driving the observed patterns. Further, our results highlight an important role played by colony expansion along the vertical dimension, which is often neglected in the analysis of spatiotemporal dynamics in compact, dense bacterial communities.    

Controlling species coexistence through temporal niche engineering

The set of species that coexist in a community often determines its collective function, such as the biodegradation rate in wastewater treatment plants or the production of short-chain fatty acids in the gut. However, not all sets of species can stably coexist in any environment, and we lack general strategies to probe and control the coexistence of a given set of species (a "dream team"). Here, we develop and computationally test a general strategy to engineer a set of resources that allows a given set of species to stably coexist. We prescribe exact ratios of resource concentrations that result in any desired relative species abundances at the steady state. Our strategy relies on the engineering of temporal niches: time intervals during which the subset of resources present in the environment remains constant, relevant for laboratory serial dilution experiments or natural feast-famine cycles. By exploiting temporal variation, our algorithm can ensure coexistence despite using much fewer resources than species, thus vastly beating the competitive exclusion principle that is pervasive in ecology. Finally, our method also chooses the most structurally stable recipe among the ones possible, thus staying the most immune to the experimental error inherent in realizing any resource recipe. We believe that our community control strategy could be useful in a wide variety of cases, such as in mimicking natural ecosystems, e.g., isolated from guts, oceans, or soils, in the lab.   

Death And Chemotaxis: Dynamics of Phage-Bacteria Interactions in Crowded Environments

Bacteriophages (“phages”) are viruses that infect and kill bacteria; thus, phage-bacteria interactions shape microbiomes, with critical implications for agriculture, food, and medicine. However, laboratory studies typically use “well-mixed” cultures in test tubes or Petri dishes, which do not mimic nature’s complexities. In nature, bacteria and phages exist in crowded 3D environments, like soil and biological tissue, that impact behaviors. Furthermore, phage-bacteria interactions are typically studied at limited timepoints due to the experimental challenges of real-time analysis, further limiting understanding of how they influence microbiomes in practice.

Here, we address this gap in knowledge using direct visualization of phage-bacteria interactions in transparent crowded matrices composed of packed hydrogel microparticles. We use confocal microscopy to visualize the real-time dynamics of motile E. coli populations encountering lytic T4 phages as a function of the initial spatial distributions and concentrations. Additionally, we develop a theoretical framework that can predict our experimental observations, thereby helping to reveal the profound impact phages have on bacterial population dynamics.   

The spread of antibiotic resistance in multi-colony networks

Bacteria employ a variety of strategies to combat antibiotic drug therapies, from the molecular to the population scales. For instance, recent work has revealed that enzymatic drug degradation by resistant cells can lead to protection of otherwise sensitive bacteria on scales many times the size of a single cell. Understanding how antimicrobial resistance scales up from the molecular to community-level is of critical importance, especially since most bacteria live in spatially dense colonies that often form nearby one another. Despite many recent insights into ecological and evolutionary dynamics within a single colony, the interplay between multiple spatially separated colonies remains largely unexplored. Here, we investigate how intra-colony interactions shape inter-colony dynamics, and vice versa, in communities of E. faecalis, an opportunistic pathogen. Specifically, we study the spread of antibiotic resistance through spatially extended colony networks created on agar plates with an open-source pipetting robot.  We observe a rich set of spatial patterning effects that link cooperative dynamics at the intra- and inter-colony scales. Interestingly, we also find that inter-colony interactions in certain multi-colony networks lead to a reduction in the number of resistant cells not seen when identical colonies are grown separately.   

New sector morphologies emerge from anisotropic colony growth

Expanding populations are of central interest in population genetics because range expansion can have lasting consequences on genetic diversity. Bacterial colonies in a Petri dish are a common model system of range expansion, and the enhanced effect of drift at the expanding front generates spatially segregated domains known as sectors. Typically, these colonies are grown on a uniform substrate which naturally produces an isotropic growth pattern. The effects of anisotropy on colony growth and sector establishment remain understudied. In this work, we use a simple model to analyze the influence of anisotropy on bacterial range expansion, with specific focus on sector morphology. Our results indicate that pronounced anisotropy introduces a unique sector morphology not achievable in standard isotropic conditions.   

Range expansion and localization in patchy landscapes

Identifying ecological processes governing the spatial distribution and assembly of microbial populations and communities is an ongoing challenge in microbial ecology. Linking colonization dynamics of traveling bacterial waves preceding stationary population development, including biofilm formation, is one microbial system of significant importance. How and where stationary populations are distributed across the landscape following range expansion is impacted by cell transitions in lifestyle from planktonic –foraging up chemical gradients– to sessile –surface attachment and recruitment–  which is influenced by environmental cues and spatial confinement. In extreme cases of confinement, attachment events can fragment the landscape localizing otherwise planktonic chemotactic bacterial waves forcing an early onset to stationary development. Coupling microfluidics experiments of bacterial colonization dynamics in patchy landscapes with a simple lattice model, this work aims to connect key processes at play in planktonic-to-sessile transitions to understand observed patterns in metapopulation distribution including localization events.   

Cooperative antibiotic resistance in spatially structured bacterial colonies

Spatial organization is a pervasive feature of life, from the molecular to the community scale. In the context of population- and community- scale behaviors of bacteria, spatial structure is often ignored for simplicity, but can have profound impacts on functional outcomes and collective properties. In this project, we investigated how antibiotic exposure affects the spatial organization of a bacterial community in which antibiotic resistance is a social trait. We use a combination of numerical modeling and experimental E. faecalis colonies on agar to understand how protection of antibiotic-sensitive bacteria by resistant bacteria via antibiotic degradation plays out across a spatially extended system.

We found that (immotile) mixed resistant and sensitive bacterial populations within a colony grow to form complex spatial arrangements at the scale of hundreds to thousands of cells, which are highly sensitive to initial antibiotic concentrations and community composition. This spatial pattern formation process can be explained by the interplay of local competition and cooperative antibiotic protection at a dynamic length scale, increasing from a few cell lengths to the scale of the whole colony, as resistant subpopulations expand.   

Single-cell morphology dictates bacterial growth dynamics under 3D confinement

Can physical properties of the microenvironment act as selection pressures at the population-level? Our current understanding of factors that exert such effect on population growth dynamics implicates genetic mutations and chemical cues, based on experimental assays performed using homogeneous liquid or 2D cultures. However, in their natural niche, bacteria inhabit complex and disordered 3D microenvironments with diverse mechanical properties. Here, to test if the physical microenvironment can selectively favor the collective growth of certain microbial strains under 3D confinement, we design transparent porous 3D growth media that match the viscoelastic properties of natural microbial habitats. Combining optical density-based growth measurements, 3D confocal microscopy, and agent-based simulations, we find that the shape anisotropy of high-aspect-ratio bacteria provides them with a selective advantage to grow more efficiently under increased 3D confinement as opposed to spherical bacteria. More precisely, under 3D confinement, high aspect ratio bacteria produce elongated colonies with larger surface areas allowing them to access nutrients more effectively. Our work provides an example of how the alteration in the physical of the microenvironment can dictate the microbiome composition in 3D disordered materials. This will help in understanding and modeling population dynamics within microbial communities inhabiting diverse biological niches using elementary physical principles.