Global metabolome spatiotemporal dynamics within river corridors were driven by deterministic and stochastic processes

Abstract:

Rivers and streams receive substantial organic matter (OM) inputs, with much being transported to the ocean and some released into the atmosphere as CO₂. As this OM travels along watersheds, it undergoes biotic (i.e., microbial activity) and abiotic (i.e., photoactivity) transformation. Recent studies have shown that this OM plays a significant role in river corridor biogeochemistry, but the underlying processes constraining OM variability are not well understood. Here, we used broad geochemical characterization, FTICR-MS metabolomics, and genome-resolved metagenomics of surface and pore water samples collected from 7 hydrologically dynamic rivers over a 48 hour period to investigate OM dynamics at a global scale. Metabolomics analyses revealed divergent patterns in molecular formula properties across all rivers. For example, rivers occurred along a gradient with regard to nominal oxidation state of carbon but formed distinct groups based upon aromaticity index. To reveal the potential drivers of these patterns, we performed ecological null modeling according to meta-metabolome ecology, the recently proposed synthesis of environmental metabolomics and meta-community ecology. While global metabolome dynamics were driven primarily by variable selection (i.e., some force causing divergence), local scale processes were significantly varied: variable selection impacted 2 rivers, homogenous selection (i.e., some force causing convergence) affected 2 rivers, and 3 rivers were stochastically assembled. We determined that a range of environmental parameters were strongly related to observed assembly processes, including calcium, total nitrogen, and NPOC concentrations. To understand the biological impacts, we connected metagenomic functional potential to FTICR-MS data using the MetaCyc database finding that members of the Gammaproteobacteria and Actinobacteria consistently encoded genes corresponding to broad detected metabolite types across rivers. By understanding these mechanisms that drive metabolome variability, we can better predict global biogeochemical patterns within river corridors.