A Metabolic modelling-based framework for predicting trophic dependencies in native rhizobiomes of crop plants

Authors: GINATT, ALON - Newe Ya'Ar Research Center; BERIHU, MARIA - Newe Ya'Ar Research Center; CASTEL, EINAM - Newe Ya'Ar Research Center; MEDINA, SHLOMIT - Newe Ya'Ar Research Center; CARMI, GON - Newe Ya'Ar Research Center; FAIGENBOIM-DORON, ADI - Newe Ya'Ar Research Center; SHARON, ITAI - Migal Galilee Research Institute; TAL, OFIR - Kinneret Limnological Laboratory, Israel Oceanographic And Limnological Research; DROBY, SAMIR - Newe Ya'Ar Research Center; Somera, Tracey; MAZZOLA, MARK - Stellenbosch University; EIZENBERG, HANAN - Newe Ya'Ar Research Center; FREILICH, SHIRI - Newe Ya'Ar Research Center

Submitted to: eLife
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 3/15/2024
Publication Date: 9/10/2024
Citation: Ginatt, A., Berihu, M., Castel, E., Medina, S., Carmi, G., Faigenboim-Doron, A., Sharon, I., Tal, O., Droby, S., Somera, T.S., Mazzola, M., Eizenberg, H., Freilich, S. 2024. A Metabolic modelling-based framework for predicting trophic dependencies in native rhizobiomes of crop plants. eLife. https://doi.org/10.7554/eLife.94558.2.
DOI: https://doi.org/10.7554/eLife.94558.2

Interpretive Summary: Plant-microbe and microbe-microbe interactions taking place in the rhizosphere impact plant health in many different ways. The harnessing of specific interactions which affect key biological activities could allow for the optimization of selected plant-growth supporting functions. In this study, we present an exploratory computational framework which aims to illuminate the black box of interactions occurring in the rhizospheres of crop plants. To the best of our knowledge, this is the first attempt to generate GSMMs (genome scale metabolic models) en masse (~400) based on high-quality MAGs (metagenome assembled genomes) derived from a specific ecosystem. This work represents a significant scientific advancement because the generic simulation platform developed not only enables the analysis of interactions between microbes but between microbes and their hosts in their natural environment as well. This computational framework now makes it possible to begin untangling the vast, complex network of plant-bacterial and bacterial-bacterial interactions occurring in the apple rhizosphere. Our network model identifies over 600,000 pathways. The study also provides new data about bacteria in healthy vs. replant-diseased orchard soil systems (including novel information on the putative functions they perform). Finally, this framework is applicable to a wide and diverse range of ecosystems and has the potential to stimulate additional microbiome research in other fields.

Technical Abstract: Plant-microbe and microbe-microbe interactions taking place in the rhizosphere impact plant health in many different ways. The harnessing of specific interactions which affect key biological activities could allow for the optimization of selected plant-growth supporting functions. Mechanistic knowledge regarding the elucidation of these dynamics, however, is limited. The framework presented in this study enables the characterization of all trophic interactions occurring within a microbial community, along with its environment. A total of 243 genome-scale metabolic models of bacteria associated with the apple rhizosphere were curated using genome resolved metagenomics. Using the iterative microbial community growth module in combination with environment-specific metabolomic inputs, the resulting trophic successions and metabolic interactions were gathered to form a network of communal trophic dependencies. Then, based on the interactions network, the metabolic profiles of differentially abundant bacteria from that environment were identified. The framework presented here provides a snapshot of the metabolic dynamics occurring within a microbial community with respect to their natural environment and generates various hypotheses regarding the metabolic capabilities of the bacteria in it.