The capability to design microbiomes with predictable functions would enable new technologies for applications in health, agriculture, and bioprocessing. Towards this goal, we develop a model-guided approach to design synthetic human gut microbiomes for production of the health-relevant metabolite butyrate. Our data-driven model quantifies microbial interactions impacting growth and butyrate production separately, providing key insights into ecological mechanisms driving butyrate production. We use our model to explore a vast community design space using a design-test-learn cycle to identify high butyrate-producing communities. Our model can accurately predict community assembly and butyrate production across a wide range of species richness. Guided by the model, we identify constraints on butyrate production by high species richness and key molecular factors driving butyrate production, including hydrogen sulfide, environmental pH, and resource competition. In sum, our model-guided approach provides a flexible and generalizable framework for understanding and accurately predicting community assembly and metabolic functions.

No abstract

Highlights d Inference of microbial interaction networks in microfluidic droplets d Insight into synthetic microbial communities across different environments d Stochastic model of community assembly to study variability in community states d Elucidation of a complex web of interactions between antibiotics and a consortium

Understanding the principles of colonization resistance of the gut microbiome to the pathogen Clostridioides difficile will enable the design of defined bacterial therapeutics. We investigate the ecological principles of community resistance to C. difficile using a synthetic human gut microbiome. Using a dynamic computational model, we demonstrate that C. difficile receives the largest number and magnitude of incoming negative interactions. Our results show that C. difficile is in a unique class of species that display a strong negative dependence between growth and species richness. We identify molecular mechanisms of inhibition including acidification of the environment and competition over resources. We demonstrate that Clostridium hiranonis strongly inhibits C. difficile partially via resource competition. Increasing the initial density of C. difficile can increase its abundance in the assembled community, but community context determines the maximum achievable C. difficile abundance. Our work suggests that the C. difficile inhibitory potential of defined bacterial therapeutics can be optimized by designing communities featuring a combination of mechanisms including species richness, environment acidification, and resource competition.

Cyanobacteria are photosynthetic microorganisms whose metabolism can be modified through genetic engineering for production of a wide variety of molecules directly from CO, light, and nutrients. Diverse molecules have been produced in small quantities by engineered cyanobacteria to demonstrate the feasibility of photosynthetic biorefineries. Consequently, there is interest in engineering these microorganisms to increase titer and productivity to meet industrial metrics. Unfortunately, differing experimental conditions and cultivation techniques confound comparisons of strains and metabolic engineering strategies. In this work, we discuss the factors governing photoautotrophic growth and demonstrate nutritionally replete conditions in which a model cyanobacterium can be grown to stationary phase with light as the sole limiting substrate. We introduce a mathematical framework for understanding the dynamics of growth and product secretion in light-limited cyanobacterial cultures. Using this framework, we demonstrate how cyanobacterial growth in differing experimental systems can be easily scaled by the volumetric photon delivery rate using the model organisms Synechococcus sp. strain PCC7002 and Synechococcus elongatus strain UTEX2973. We use this framework to predict scaled up growth and product secretion in 1L photobioreactors of two strains of Synechococcus PCC7002 engineered for production of l-lactate or L-lysine. The analytical framework developed in this work serves as a guide for future metabolic engineering studies of cyanobacteria to allow better comparison of experiments performed in different experimental systems and to further investigate the dynamics of growth and product secretion.

Predicting the dynamics and functions of microbiomes constructed from the bottom-up is a key challenge in exploiting them to our benefit. Current models based on ecological theory fail to capture complex community behaviors due to higher order interactions, do not scale well with increasing complexity and in considering multiple functions. We develop and apply a long short-term memory (LSTM) framework to advance our understanding of community assembly and health-relevant metabolite production using a synthetic human gut community. A mainstay of recurrent neural networks, the LSTM learns a high dimensional data-driven non-linear dynamical system model. We show that the LSTM model can outperform the widely used generalized Lotka-Volterra model based on ecological theory. We build methods to decipher microbe-microbe and microbe-metabolite interactions from an otherwise black-box model. These methods highlight that Actinobacteria, Firmicutes and Proteobacteria are significant drivers of metabolite production whereas Bacteroides shape community dynamics. We use the LSTM model to navigate a large multidimensional functional landscape to design communities with unique health-relevant metabolite profiles and temporal behaviors. In sum, the accuracy of the LSTM model can be exploited for experimental planning and to guide the design of synthetic microbiomes with target dynamic functions.

Microbial conversion of renewable feedstocks to high-value chemicals is an attractive alternative to current petrochemical processes because it offers the potential to reduce net CO emissions and integrate with bioremediation objectives. Microbes have been genetically engineered to produce a growing number of high-value chemicals in sufficient titer, rate, and yield from renewable feedstocks. However, high-yield bioconversion is only one aspect of an economically viable process. Separation of biologically synthesized chemicals from process streams is a major challenge that can contribute to >70% of the total production costs. Thus, process feasibility is dependent upon the efficient selection of separation technologies. This selection is dependent on upstream processing or biological parameters, such as microbial species, product titer and yield, and localization. Our goal is to present a roadmap for selection of appropriate technologies and generation of separation schemes for efficient recovery of bio-based chemicals by utilizing information from upstream processing, separation science and commercial requirements. To achieve this, we use a separation system comprising of three stages: (I) cell and product isolation, (II) product concentration, and (III) product purification and refinement. In each stage, we review the technology alternatives available for different tasks in terms of separation principles, important operating conditions, performance parameters, advantages and disadvantages. We generate separation schemes based on product localization and its solubility in water, the two most distinguishing properties. Subsequently, we present ideas for simplification of these schemes based on additional properties, such as physical state, density, volatility, and intended use. This simplification selectively narrows down the technology options and can be used for systematic process synthesis and optimal recovery of bio-based chemicals.

L-lysine and other amino acids are commonly produced through fermentation using strains of heterotrophic bacteria such as Corynebacterium glutamicum. Given the large amount of sugar this process consumes, direct photosynthetic production is intriguing alternative. In this study, we report the development of a cyanobacterium, Synechococcus sp. strain PCC 7002, capable of producing L-lysine with CO2 as the sole carbon-source. We found that heterologous expression of a lysine transporter was required to excrete lysine and avoid intracellular accumulation that correlated with poor fitness. Simultaneous expression of a feedback inhibition resistant aspartate kinase and lysine transporter were sufficient for high productivities, but this was also met with a decreased chlorophyll content and reduced growth rates. Increasing the reductant supply by using NH4+, a more reduced nitrogen source relative to NO3−, resulted in a two-fold increase in productivity directing 18% of fixed carbon to lysine. Given this advantage, we demonstrated lysine production from media formulated with a municipal wastewater treatment sidestream as a nutrient source for increased economic and environmental sustainability. Based on our results, we project that Synechococcus sp. strain PCC 7002 could produce lysine at areal productivities approaching that of sugar cane to lysine via fermentation using non-agricultural lands and low-cost feedstocks.