Deciphering pharmaceutical resistance in sulfur-driven autotrophic denitrification: an integrated multi-omics...

Sulfur-driven autotrophic denitrification (SdAD) is a promising low-carbon technology for nitrogen removal; however, its stability and adaptive mechanisms under pharmaceutical stress remain poorly understood. In this study, ibuprofen (IBU) was used as a representative pharmaceutical to investigate the response of an SdAD system. Throughout the 210-day operational period, the system demonstrated exceptional functional robustness, maintaining high sulfide (97.46 ± 3.18%) and inorganic nitrogen (99.17 ± 4.34%) removal efficiencies across IBU concentrations ranging from environmentally relevant levels to elevated shock loads (100-2000 μg/L). Underpinning this macroscopic stability, community-level analyses revealed that instead of succumbing to inhibition, the SdAD microbiome actively reorganized its composition and topological structure to accommodate the selective pressure. This adaptation was characterized by enhanced microbial diversity and stress-induced network modularity (particularly at 100-500 μg/L), alongside strengthened cooperative interactions between sulfur-oxidizing bacteria and denitrifiers. To unravel the specific molecular drivers of this resilience, we integrated AlphaFold-based structural modeling with machine learning-coupled molecular docking. This enabled us to resolve the three-dimensional structure of sulfide: quinone oxidoreductase (SQR) and perform the first structure-function analysis of SQR within an SdAD context under pharmaceutical stress, revealing that arginine residues serve as key interaction hotspots for IBU binding. Consistent with this binding mechanism, multi-omics data further corroborated a systemic adjustment involving the coordinated regulation of sulfur oxidation genes and the transcriptional upregulation of arginine biosynthesis pathways. Overall, these findings shed light on how the SdAD community mitigates pharmaceutical toxicity through a multi-tiered strategy involving ecological network reorganization and metabolic compensation. Methodologically, this work highlights the value of integrating artificial intelligence-driven structural biology with multi-omics analyses to decode the mechanisms of contaminant resistance in biological wastewater treatment systems.