Threshold-like transitions of microbial community states and functional reorganization in petroleum hydrocarbon-contaminated soils.

Petroleum hydrocarbon (PH) contamination alters soil microbial communities, yet whether microbial responses to PH contamination occur gradually or exhibit threshold-like transitions remains poorly understood. Here, soils collected along a PH contamination gradient surrounding an oil well were analyzed to identify microbial community states and their associations with environmental factors, community assembly, interactions, and functional potential. Bacterial communities exhibited a bimodal distribution, indicating two distinct states: State A was predominantly associated with higher PH contamination, whereas State B was associated with lower contamination. Random forest analysis identified PH contamination, particularly n-alkanes and polycyclic aromatic hydrocarbons (PAHs), together with inorganic nitrogen (N), as the environmental factors most strongly associated with state differentiation. Potential landscape analysis identified intermediate contamination ranges where both states occurred (2.50 ≤lg(n-alkanes)≤ 3.21; 1.61 ≤lg(PAHs)≤ 1.69), indicating threshold-like differentiation. State A exhibited lower diversity, stronger deterministic assembly, more modular co-occurrence networks, enrichment of hydrocarbon-degrading taxa, and greater predicted functional potential for hydrocarbon activation, aromatic ring cleavage, denitrification, and N assimilation. State B showed greater network connectivity and stronger predicted potential for downstream carbon assimilation, nitrification, and more balanced N transformation. These findings demonstrate that PH contamination is associated with nonlinear reorganization of microbial community structure and functional potential across different community states.