Rendering the Metabolic Wiring Powering Wetland Soil Methane Production
Authors:
Mikayla A. Borton1*, Jared B. Ellenbogen1, Andrew P. Freiburger2, Jorge A. Villa3, Angela M. Oliverio1, Adrienne B. Narrowe1, Sophie Jurgensen1, David W. Hoyt4, Karl K. Weitz4, Nikola Tolic4, Bridget B. McGivern1, Emily K. Bechtold1, Katherine B. Louie5,6, Trent R. Northen5,6, Y. Chin7, E. Ward8, S. Bansal8, Christopher Henry2, Christopher S. Miller9, William Riley5, Timothy H. Morin10, Michael J. Wilkins1, Mary Lipton4, Gil Bohrer11, Kelly C. Wrighton1 ([email protected])
Institutions:
1Colorado State University–Fort Collins; 2Argonne National Laboratory; 3University of Louisiana–Lafayette; 4Environmental Molecular Sciences Laboratory; 5Lawrence Berkeley National Laboratory; 6DOE Joint Genome Institute; 7University of Delaware–Newark; 8United States Geologic Survey; 9University of Colorado–Denver; 10College of Environmental Sciences and Forestry, The State University of New York; 11The Ohio State University
Goals
Despite their vital roles in transforming nutrients and controlling greenhouse gas fluxes in wetlands, microbial knowledge is often limited to taxonomic identity alone and rarely includes cross-site comparisons. The team proposed to address this knowledge gap using coordinated, reproducible field measurements collected across a wetland-methane continuum spanning geographical locations. This project tests the overarching hypothesis that microbial genomic attributes are conserved across high methane-emitting wetlands. First, researchers will use a cross-wetland approach to define the microbial membership, physiology, and interactions directly contributing to wetland methane production. Next, the team will uncover the microbial decomposition network features that classify high methane emitting wetlands. Using this information, researchers will test the genomic resolution needed to make robust predictions of regional and global methane fluxes. These integrated field, laboratory, and modeling approaches will identify the unifying microbial properties governing soil carbon decomposition and methane fluxes, such that some level of biological representation in models will enhance predictions of soil methane fluxes.
Abstract
To illuminate the metabolic features of carbon decomposition in high methane-emitting wetlands, the project created the Multiomics for Understanding Climate Change (MUCC) database. This resource contains the identity, distribution, and functional information of 26,000 microbial genomes from wetlands, including over 500 methanogen and methanotroph genomes. The team coupled MUCC to 133 spatially and temporally resolved metatranscriptomes, along with paired amplicon, geochemical, and greenhouse gas data from over 700 samples collected from one of the most prolific methane-emitting wetlands in the United States. From this site, researchers unveiled previously unrecognized roles of archaea in carbon decomposition, resistance to redox changes exhibited by methane-cycling community members, and the delineation of metabolic networks contributing to methane production. This standardized sampling and sequencing design set the stage for future cross-site comparisons, enhancing the ability to generalize findings across wetlands.
Extending the sampling from a single wetland, coordinated field sampling in year 1 added paired greenhouse gas measurements, soil chemistry, and multiomics from 12 additional wetlands (three bogs, four fens, five marshes). Cross-wetland analysis of initial findings showed that in contrast to current paradigms, where acetolactic and hydrogenotrophic are the predominant forms of methanogenesis, methylotrophic methanogenesis is prevalent and active. Across sites genome-resolved metatranscriptomes showed that members of the Methanomassiliicoccales, Methanosarcinales, Methanomethyliales, and Methanobacteriales were expressing methylotrophic genes, while metabolomics revealed new methylotrophic substrates (e.g., syringate, trimethyllysine). Further corroborating these findings, the team established enrichments of wetland soils dosed with either carbon-13 (13C) dimethylsulfide or 13C acetate to demonstrate that field-relevant methylotrophic lineages were enriched in reactor 13C-labeled proteomes. Likewise, methylated substrates considerably contributed to methane production (on average 53%), demonstrating that methylotrophic methanogenesis is a likely contributor to methane across global freshwater wetlands and should be included in process-based models.
To integrate this new microbial knowledge into ecosystem scale models, researchers built genome-scale metabolic models (GEMs) from MUCC genomes. MUCC GEMs combine field-derived multiomics data, enrichment-derived isotopic evidence, reference genomes, and fundamental physical and thermodynamic principles to produce the most accurate metabolic representation of each strain observed in the MUCC samples. These GEMs utilize a novel probabilistic framework that captures the metabolic diversity of each metagenome assembled genome (MAG) by consolidating phylogenetically close references into a single GEM for simulation, allowing for the preferential simulation of common, conserved pathways while permitting niche pathways if conditions and constraints require it.
GEMs undergo initial testing for internal consistency, aligning with available experimental data. Then GEMs are merged into sample-level community models where GEMs are further refined to maximally recapitulate experimental data. This process tailors each sample to have a unique MAG model for each strain, whose metabolic behavior is described as how each strain transforms nutrients into biomass and byproducts. These whole-cell reactions constrained by relative abundance from multiomics data will ultimately be integrated into ecosystem-scale simulations. MUCC and the corresponding GEMs are publicly available on the DOE Systems Biology Knowledgebase (KBase), engendering collaborative enterprises with the goal to advance wetland climate-driven research. Ultimately, this research illuminates the metabolisms influencing the methane cycle, offering a direction for increased realism in predictive models of greenhouse gas emissions.
References
Ellenbogen, J. B., et al. 2023. “Methylotrophy in the Mire: Direct and Indirect Routes for Methane Production in Thawing Permafrost,” mSystems 9(1), e00698-23. DOI:10.1128/msystems.00698-23.
Oliverio, A. M., et al. 2024. “Rendering the Metabolic Wiring Powering Wetland Soil Methane Production,” bioRxiv. DOI:10.1101/2024.02.06.579222.
Shaffer, M., et al. 2023. “kb_DRAM: Annotation and Metabolic Profiling of Genomes with DRAM in KBase,” Bioinformatics 39(4), btad110.
Funding Information
This research was supported by the DOE Office of Science, BER program, grant no. DE-SC0023084. This program is supported by the U. S. DOE, Office of Science, through the GSP, BER program, under FWP ERKP123.