Integrating Enzyme-Based Kinetics in Reactive Transport Models to Simulate Spatiotemporal Dynamics of Biomarkers during Chlorinated Ethene Degradation

Abstract

Biomarkers such as functional gene mRNA (transcripts) and proteins (enzymes) provide direct proof of metabolic regulation during the reductive dechlorination (RD) of chlorinated ethenes (CEs). Yet, current models to simulate their spatiotemporal variability are not flexible enough to mimic the homologous behavior of RDase functional genes. To this end, we developed new enzyme-based kinetics to model the concentrations of CEs together with the transcript and enzyme levels during RD. First, the model was calibrated to existing microcosm data on RD of cis-DCE. The model mirrored the tceA and vcrA gene expression and the production of their enzymes in Dehalococcoides spp. Considering tceA and vcrA as homologous instead of nonhomologous improved fitting of the mRNA time series. Second, CEs and biomarker patterns were explored as a proof of concept under groundwater flow conditions, considering degraders occurring in immobile and mobile states. Under both microcosm and flow conditions, biomarker-rate relationships were nonlinear hysteretic because tceA and vcrA acted as homologous genes. The mobile biomarkers additionally undergo advective-dispersive transport, which increases the nonlinearity and makes the observed patterns even more challenging to interpret. The model offers a thorough mechanistic description of RD while also allowing simulation of spatiotemporal dynamic patterns of various key biomarkers in aquifers.

Keywords: biodegradation; contaminated groundwater; enzyme-based kinetics; organohalide respiration; reactive transport modeling; reductive dechlorination.

Figure 1

Conceptual model of the metabolic pathways responsible for the complete biodegradation of cis-1,2-DCE to ethene in the microcosm experiment from Kranzioch et al. considered in this study. The light gray polygons represent the concentration-dependent activation of the generic transcription factors by cis-1,2-DCE and VC. Numbers in the white circles refer to the equations described in the following section, representing specific reactions of the considered metabolic pathways.

Figure 2

Temporal patterns of chemical concentrations (a), biomarker levels (b), degradation rates (c), and transcription factors (d) in the microcosm experiment of Kranzioch et al., obtained through the enzyme-based kinetics and considering the metabolic pathways in Figure 1 (solid and dashed lines). The CE, ethene, and chloride concentrations, as well as the Dehalococcoides level and the degradation rates obtained through Monod kinetics, are shown as dotted lines. In the plots, the concentration of 16S rRNA gene copies corresponds to the number of Dehalococcoides cells. The black and gray dashed lines represent the quantification limits of the Dehalococcoides 16S rRNA gene and the tceA and vcrA transcripts, respectively. Refer to the original paper for the error bars related to the standard deviation of all of the chemical and biomarker measurements.

Figure 3

1D scenario-based reactive transport model simulating two relevant stages of the evolution of a cis-DCE plume in groundwater: the elongation phase at 10 years, when the plume is in a transient state, and the steady-state condition at 40 years. In addition to chemicals (a, b), the plot includes the immobile (c, d) and mobile (e, f) biomarkers and the degradation rates. The solid and dashed lines refer to the enzyme-based 1D flow model, the dotted to the Monod-based, while the long-dashed lines refer to the conservative transport of cis-DCE. In the plots, the concentration of 16S rRNA gene copies corresponds to the number of Dehalococcoides cells.

Figure 4

Relationships between biomarker levels and degradation rates of the two transformation steps of cis-DCE, via VC, to ethene, as shown for the batch model (a-h) and the 1D flow model at steady-state (i-p). Each plot describes how the biomarker-rate relationship varies over time or distance (color scales) for the batch model and 1D flow model, respectively. The linear correlation and degree of hysteresis (i.e., how different the ascending and descending parts of each curve are) are quantified by the Pearson coefficients (r) and the hysteresis loop areas (HLAs) in Tables S4 and S5, respectively.