Gibbs Energy Dynamic Yield Method (GEDYM): Predicting microbial growth yields under energy-limiting conditions

摘要

Biomass-explicit biogeochemical models assign microbial growth yields (Y) using values measured in the laboratory or predicted using thermodynamics-based methods. However, Y values are rarely measured under the low energy delivery conditions that often prevail in the subsurface, and existing predictive methods for calculating Y values when the catabolic energy supply rate is limited remain poorly tested. Here, we derive and validate a new semi-theoretical method for calculating Y values: the Gibbs Energy Dynamic Yield Method (GEDYM). Method validation relies on a compilation of 132 geochemically relevant literature Y values comprising predominantly (60%) low energy ( - 25 kJ (mol e(-))(-1)) metabolisms. GEDYM is based on estimating the Gibbs energy change of the metabolic reaction (Delta G(met)), which links the Gibbs energy changes of the catabolic (Delta G(cat)) and anabolic (Delta G(an)) reactions of a microorganism through its growth yield. Given that the values of Delta G(met), Delta G(cat) and Delta G(an) all depend on their respective reaction quotients, the resulting Y values account for changes in the chemical environment surrounding the cells. GEDYM incorporates an empirical relationship that accurately estimates the extent to which Delta G(met) deviates from its standard state value from the relative difference between Delta G(cat) and its corresponding standard state value. GEDYM yields Y values with lower relative errors and statistical bias than the existing Gibbs energy dissipation method (GEDM). Using dissimilatory iron reduction, sulfate reduction and methanogenesis as examples, we illustrate the importance of considering variations in Delta G(cat) and Delta G(an) when predicting Y values for individual metabolisms. Because of its ability to dynamically adjust the values of Delta G(met) and Y to variable geochemical conditions, GEDYM yields a more realistic representation of geomicrobial activity in predictive reactive transport models. (C) 2018 The Author(s). Published by Elsevier Ltd.