Longitudinal flux balance analyses of a patient with episodic colonic inflammation reveals microbiome metabolic dynamics

ABSTRACT

We report the first use of constraint-based microbial community modeling on a single individual with episodic inflammation of the gastrointestinal tract, who has a well documented set of colonic inflammatory biomarkers, as well as metagenomically-sequenced fecal time series covering seven dates over 16 months. Between the first two time steps the individual was treated with both steroids and antibiotics. Our methodology enabled us to identify numerous time-correlated microbial species and metabolites. We found that the individual’s dynamical microbial ecology in the disease state led to time-varying in silico overproduction, compared to healthy controls, of more than 24 biologically important metabolites, including methane, thiamine, formaldehyde, trimethylamine N-oxide, folic acid, serotonin, histamine, and tryptamine. The microbe-metabolite contribution analysis revealed that some Dialister species changed metabolic pathways according to the inflammation phases. At the first time point, characterized by the highest levels of serum (complex reactive protein) and fecal (calprotectin) inflammation biomarkers, they produced L-serine or formate. The production of the compounds, through a cascade effect, was mediated by the interaction with pathogenic Escherichia coli strains and Desulfovibrio piger. We integrated the microbial community metabolic models of each time point with a male whole-body, organ-resolved model of human metabolism to track the metabolic consequences of dysbiosis at different body sites. The presence of D. piger in the gut microbiome influenced the sulfur metabolism with a domino effect affecting the liver. These results revealed large longitudinal variations in an individual’s gut microbiome ecology and metabolite production, potentially impacting other organs in the body. Future simulations with more time points from an individual could permit us to assess how external drivers, such as diet change or medical interventions, drive microbial community dynamics.

Graphical abstract

KEYWORDS:

• Flux balance analysis
• longitudinal data
• inflammatory bowel diseases
• colonic inflammation
• metagenomics
• microbiota
• metabolic modeling
• bioinformatics
• constraint-based modeling
• personalized medicine

Acknowledgments

We thank staff at the J Craig Venter Institute for performing the metagenomic sequencing of stool samples and the San Diego Supercomputer for providing the CPU hours for processing the metagenomic sequencing data. WL thanks the Center for Research in Biological Systems (CRBS) for support during part of the metagenomic analysis. A final acknowledgement to the Italian Consortium for Biotechnologies (CIB) for the support.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Author contributions

Arianna Basile: Funding acquisition, Conceptualisation, Investigation, Formal Analysis, Visualisation, Writing – Original Draft. Almut Heinken: Software, Methodology, Formal Analysis, Writing – view & Editing. Johannes Hertel: Supervision, Writing – Review & Editing. Larry Smarr: Funding acquisition, Formal Analysis, Writing – Review & Editing. Weizhong Li: Review & Editing. Laura Treu: Supervision, Writing – Review & Editing. Giorgio Valle: Writing – Review & Editing. Stefano Campanaro: Supervision, Conceptualisation, Writing – Review & Editing. Ines Thiele: Conceptualisation, Supervision, Funding acquisition, Software, Writing – Review & Editing.

Data availability statement

The raw abundance data have been submitted as part of Supplementary Table I.

Two later publications resequenced some of the LS1–7 samples, at a lower depth than reported herein, as part of research on a longer time series of LS fecal samples. The first publication²¹ resequenced LS 1–7 (12/28/2011 to 4/29/2013) as part of a longer time series of 27 LS samples (dates from 12/28/2011 to 12/07/2014 are listed in their Supplementary Table S1, Sheet Metadata) analyzing the metagenomics of E. coli strain dynamics. The metagenomics sequence of these 27 samples can be found in EBI under study PRJEB24161. The second publication³⁰ sequenced eight LS time series samples (dates from 12/28/2011 to 5/22/2016), including resequencing LS1–3, and added metaproteomic analysis for these eight time points. Metagenomic data are available through EBI under the study PRJEB28712 (ERP110957).

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/19490976.2023.2226921

Additional information

Funding

This work was financially supported by the “Budget Integrato della Ricerca Dipartimentale” (BIRD198423) PRID 2019 of the Department of Biology of the University of Padua, entitled “SyMMoBio: inspection of Syntrophies with Metabolic Modelling to optimise Biogas Production” to LT. Furthermore, this study was funded by grants from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (grant agreement No 757922), by the National Institute on Aging grants (1RF1AG058942 and 1U19AG063744), and from the Science Foundation Ireland under Grant number 12/RC/2273-P2 to IT. The Ph.D. fellowship of AB was supported by “Progetto di Eccellenza DiBio” of the University of Padua. AB was the recipient of the EMBO short-term fellowship 8720. Larry Smarr thanks the UC San Diego Calit2 Qualcomm Institute and the Center for Microbiome Innovation members for useful discussions and a private donor for financial support for this paper.