Linking Microbial Gene-Expression and Denitrification Rates in Aquifers: Insights From an Intermediate-Scale In Situ Injection-Extraction Experiment and Reaction Model

Microorganisms catalyze the turnover of reactants in the subsurface, determining the fate of major groundwater constituents and contaminants. Yet our quantitative understanding of microbial dynamics lags behind its importance. Functional molecular-biological methods offer a powerful tool (box) to quantify microbial dynamics in the environment. Their quantitative potential has only recently been probed via the integration of DNA-synthesis and RNA-transcription processes into reaction models. Here, we present the results of an experiment where we actively stimulated the microbial community via nitrate injection into an anaerobic aquifer and monitored concentrations in an extraction well 1.6 m apart. We installed microbial trapping devices (MTDs) to monitor the sediment-associated microbial community (16S rRNA gene sequencing and qPCR) over time. A nitrate-driven exponential increase in napA and narG gene copies and a shift in the relative abundances of the microbial community toward known denitrifying taxa in the trapping devices highlight the fast response of the microbial community throughout the 17-day experiment. We developed a gene-explicit reaction model that simulates the concentration dynamics measured at the extraction well and considers an additional electron-donor limitation that explains the observed drop in gene abundances despite nitrate availability. Our model also yielded a relationship between measured denitrification genes and computed rates, confirming previous predictions that the gene-rate relationship is non-linear and hysteretic. While the MTDs provided easy access to microbial biomass, our model shows that the kinetics therein differed from those in the aquifer. Thus, MTDs may not be truly representative of the conditions experienced by microbes in the aquifer itself.