Ecotoxicological effects of rare earth mining drainage: Unraveling dual antibiotic resistance gene regulation and risk-abundance decoupling through microbial community restructuring

Large-scale mining of ion-adsorption rare earth elements (REEs) generates acidic mine drainage (AMD) laden with REEs and heavy metals (HMs), yet its cascading impacts on microbial community assembly and antibiotic resistance genes (ARGs) dissemination across multi-matrices remain poorly characterized. By integrating high-throughput sequencing, co-occurrence network analysis, and partial least squares path modeling (PLS-PM), we unraveled that mining-induced geochemical divergence between mining-impacted areas and adjacent watersheds acts as a dominant environmental filter, reshaping the microbial community assembly and ARGs endowment. Key findings were in three areas: (1) Pollution-driven microbial adaptation. Significant shifts in microbial composition fostered "alternative steady states" without altering richness and induced network polarization-simplification in mining water (degree/density reduced by about 50 %) versus complexification in mining soil (degree/connections increased by about 3-fold), with enhanced mutualistic interactions and a 3-fold reduction in keystone species complexity. (2) ARGs risk-abundance decoupling. Mining areas exhibited 2.6-fold higher ARGs health risks than watersheds (p = 0.019), despite comparable abundance levels (p > 0.05), necessitating a paradigm shift from quantitative surveillance to health risks monitoring. (3) Dual ARGs regulation. While acidic REEs co-contamination directly promoted ARGs proliferation via co-selection (pathway coefficient = 0.254), it concurrently mitigated overall ecological risks through host community restructuring and potential horizontal gene transfer suppression (total effects = -0.186). These findings elucidate the ecological trade-offs between microbial adaptive resilience and ARGs dissemination in mining-impacted ecosystems, while establishing a mechanistic framework for optimizing targeted remediation strategies and sustainable resource extraction protocols.