Formation, characterization and modeling of emergent synthetic microbial communities

Abstract

Microbial communities colonize plant tissues and contribute to host function. How these communities form and how individual members contribute to shaping the microbial community are not well understood. Synthetic microbial communities, where defined individual isolates are combined, can serve as valuable model systems for uncovering the organizational principles of communities. Using genome-defined organisms, systematic analysis by computationally-based network reconstruction can lead to mechanistic insights and the metabolic interactions between species. In this study, 10 bacterial strains isolated from the Populus deltoides rhizosphere were combined and passaged in two different media environments to form stable microbial communities. The membership and relative abundances of the strains stabilized after around 5 growth cycles and resulted in just a few dominant strains that depended on the medium. To unravel the underlying metabolic interactions, flux balance analysis was used to model microbial growth and identify potential metabolic exchanges involved in shaping the microbial communities. These analyses were complemented by growth curves of the individual isolates, pairwise interaction screens, and metaproteomics of the community. A fast growth rate is identified as one factor that can provide an advantage for maintaining presence in the community. Final community selection can also depend on selective antagonistic relationships and metabolic exchanges. Revealing the mechanisms of interaction among plant-associated microorganisms provides insights into strategies for engineering microbial communities that can potentially increase plant growth and disease resistance. Further, deciphering the membership and metabolic potentials of a bacterial community will enable the design of synthetic communities with desired biological functions.

Keywords: Flux balance analysis; Genome-scale model; Metabolic interaction; Metaproteomics; Microbial community; Rhizosphere bacteria.

Graphical abstract

Fig. 1

Overview of experimental design for the bottom-up assembly of stable communities utilizing defined bacterial isolates.

Fig. 2

Analysis of the relative abundances of the 10 bacterial strains after sequential passages in MOPS minimal and R2A complex media. The relative abundances of each bacterial strain in the community are based on A) 16S rRNA gene amplicon sequencing results and B) metaproteomic results. Passage 0 represents the end of the first growth cycle after the initial inoculation. Each bar is a replicate, with three replicates per passage. The numbers shown on the bottom represent the passage number for those samples.

Fig. 3

Comparison of experimentally determined growth parameters with FBA model predicted growth. A) Relative growth rates and lag times for the individual strains in MOPS medium; B) relative growth rates and lag times for the strains in R2A medium. Each data column of experimental results represents the mean and error bars are the standard deviation over three parallel experiments.

Fig. 4

Predicted metabolite exchange among the four dominant members of the microbial community formed in MOPS medium as proposed by the community FBA model. The percentage of detected enzymes by metaproteomics analyses is shown in parentheses.

Fig. 5

Predicted metabolites exchange among the three dominant members of the microbial community formed in R2A medium as proposed by the community FBA model. The percentage of detected enzymes by metaproteomics analyses is shown in parentheses.