Using 13CO2 to Track Carbon Flow Mediated by Fungal-Bacterial Interactions in Grassland Soils
Authors:
Mengting Yuan1* (maggieyuan@berkeley.edu), Katerina Estera-Molina1,2, Javier A. Ceja-Navarro3, Christina Fossum1, Anne Kakouridis1, Erin Nuccio2, Giovana Simao Slanzon4, Siyang Jian5, Zhifeng Yang 5, Jizhong Zhou5, Jennifer Pett-Ridge2, Mary K. Firestone1, and Nhu Nguyen4
Institutions:
1University of California–Berkeley; 2Lawrence Livermore National Laboratory; 3Northern Arizona University; 4University of Hawaiʻi–Mānoa; and 5University of Oklahoma
Goals
As the dominant groups of soil microbes, fungi and bacteria together drive essential biogeochemical cycles belowground. However, the dynamics, mechanisms, and ecological implications of bacterial-fungal interactions (BFIs) are not well-understood, especially on the community level and under abiotic stress. The broader goal is to build a quantitative and mechanistic framework to address how BFIs determine the availability and fate of C and N across the complexity of soil niches. The three interrelated objectives each addresses a critical factor determining the behavior of BFIs across soil C source, mineralogy, and water availability: (1) measure how grassland BFIs are shaped by the availability of different C source, and in turn mediate soil C and N mineralization; (2) determine how BFIs may mediate C stabilization and mineralization via aggregation and mineral surface interaction across soils of different mineralogies; and (3) quantify how reduced water availability interplays with C source, C availability, and soil mineralogy in structuring BFIs and BFI-mediated soil processes.
Abstract
Fungi and bacteria are the two dominant groups of soil organisms that consume, process, and translocate plant-derived soil organic matter (SOM) and thus are critical to global biogeochemical cycling. While fungi and bacteria’s separate stereotypical processing of SOM are relatively well documented, there is an increasing recognition that niche-bridging fungal network and their interacting bacteria commonly co-mediate the flow and fates of plant-derived carbon (C). Less understood is how bacteria and fungi interact, and how these interactions change in response to environmental perturbations to influence the rate of C processes and fates of C in soil. The hyphosphere, soil explored by fungal hyphae, represent a potential hotspot niche of microbial activity that is just starting to be understood.
To characterize the interactions between fungi and bacteria in driving the flow of photosynthetic C belowground, researchers developed the Dynamic Ecosystem Labeling (DEL) system to deliver 13CO2 to live plants, which assimilated the 13C and transferred it belowground via rhizodeposition. In a large-scale field manipulation study on the impacts of reduced precipitation on soil microbial interactions in a Mediterranean grassland, the team conducted four separate labeling events. Each labeling event differed in duration, between 5 to 14 days, and covered different phenological timepoints of the dominant grass Avena spp.: seedling, exponential growth, and peak biomass. During labeling, the DEL system tightly controlled the chamber headspace CO2 concentration, driving the headspace average 13C atom% to 42.9 ± 13.3% and a maximum of 70.3%. A part of the 13C incorporated into the ecosystem was quickly mineralized and detected as CO2 over at least four weeks after labeling ended. Incredibly, 23% of the 13C entered soil was still detectable after 2 years, three quarters of which were found to be associated with soil minerals. Sufficient stable isotope was incorporated into the belowground microbial communities such that labeled C could be followed into the DNA through quantitative Stable Isotope Probing (qSIP). Using amplicon sequencing of 16S and ITS genes coupled with qSIP, the DNA of nearly 150 taxa of bacteria and fungi were found to be significantly enriched with 13C after 5 days of labeling, indicative of a potential food web that facilitates the bioprocessing and flow of rhizodeposits. Although the fungal and bacterial community composition and co-occurrence networks changed most profoundly over time, reduced precipitation significantly reduced the number of taxa that were 13C enriched. This suggests that bacteria and fungi under reduced precipitation might have less access to newly fixed plant C. 13C provides a powerful and sensitive tracer to follow and quantify the flow of C mediated by soil bacteria and fungi.
In a greenhouse experiment, the DEL system was used to label Avena barbata growing in mesocosms that had an air gap to separate the hyphosphere of arbuscular mycorrhizal fungi (AMF) Rhizophagus intraradices from the rhizosphere. The team quantified the amount of C being transported away from the rhizosphere by AMF hyphae crossing the air gap, and studied the microbial composition associated with AMF hyphae and their 13C enrichment. In six weeks, over 1% of the total soil C in the hyphal compartment was 13C labeled, and a quarter of which was found associated with soil minerals. Amplicon sequencing indicated that AMF significantly modified the soil prokaryote community composition, but not diversity; nineteen amplicon sequence variants significantly increased in the presence of AMF, including Arthrobacter sp., Caulobacter sp., Rhizobium sp., Dongia sp., and Verrucomicrobia. Identification of the 13C enriched taxa, which could be the primary consumers of 13C imported via AMF hyphae, is underway.
Incorporating 13C-informed C pool sizes and microbial activity into the Microbial ENzyme Decomposition (MEND) model improved the prediction of the decomposition rates for different C pools. The next steps include (1) developing a BFI module for the MEND model; (2) using field-based fungal ingrowth core experiment and simplified soil interactions microcosm (SIM) experiments to derive the essential parameters that indicate the types and strengths of BFIs; and (3) assessing the soil C stability and storage mediated by the BFIs in different soil mineralogies and under reduced water availability.
Funding Information
This research was supported by the DOE Office of Science, Office of Biological and Environmental Research (BER), grant no. DE-SC0023106. Work at LLNL is performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344.