Studying aquatic
biotransformation of chemicals in laboratory experiments,
i.e., OECD 308 and OECD 309 studies, is required by international
regulatory frameworks to prevent the release of persistent chemicals
into natural water bodies. Here, we aimed to address several previously
described shortcomings of OECD 308/309 studies regarding their variable
outcomes and questionable environmental relevance by broadly testing
and characterizing a modified biotransformation test system in which
an aerated water column covers a thin sediment layer. Compared to
standard OECD 308/309 studies, the modified system showed little inter-replicate
variability, improved observability of biotransformation, and consistency
with first-order biotransformation kinetics for the majority of 43
test compounds, including pharmaceuticals, pesticides, and artificial
sweeteners. To elucidate the factors underlying the decreased inter-replicate
variability compared to OECD 309 outcomes, we used multidimensional
flow cytometry data and a machine learning-based cell type assignment
pipeline to study cell densities and cell type diversities in the
sediment and water compartments. Our here presented data on cell type
composition in both water and sediment allows, for the first time,
to study the behavior of microbial test communities throughout different
biotransformation simulation studies. We found that sediment-associated
microbial communities were generally more stable throughout the experiments
and exhibited higher cell type diversity than the water column-associated
communities. Consistently, our data indicate that aquatic biotransformation
of chemicals can be most robustly studied in test systems providing
a sufficient amount of sediment-borne biomass. While these findings
favor OECD 308-type systems over OECD 309-type systems to study biotransformation
at the water–sediment interface, our results suggest that the
former should be modified toward lower sediment–water ratios
to improve observability and interpretability of biotransformation.