Ecological composite fertilizer application enhances wheat yield and optimizes rhizosphere microbial community under reduced fertilization

Excessive fertilization poses a major threat to sustainable agriculture, resulting in resource waste and environmental degradation. The ecological composite fertilizer (ECF) combined with fertilizer reduction represents a promising strategy to improve rhizosphere microbial diversity in wheat systems. A field experiment, containing six treatments, namely traditional compound fertilizer (TF, applied at the conventional rate) with a 10% reduction (TF90), TF90 plus ECF application (TF90+ECF), TF with a 15% reduction (TF85), TF85 plus ECF application (TF85+ECF), TF with a 20% reduction (TF80), and TF80 plus ECF application (TF80+ECF), was conducted to explore the influences of fertilizer reduction combined with ECF application on wheat yield and rhizosphere soil microbial diversity. Results showed that the TF85+ECF treatment achieved the highest wheat yield at 8,717.33 kg ha−1, which was significantly greater than all other treatments and represented a 30.63% increase over the TF85 treatment. The TF85+ECF group significantly enhanced the activities of the carbon and nitrogen cycling enzymes β-1, 4-glucosidase glucosidase (BG) and urease (UE), and increased the abundances of the functional genes cbbLR and amoA. In the +ECF treatment groups (TF90+ECF, TF85+ECF, and TF80+ECF), linear discriminant analysis effect size (LEfSe) and specialization-occupancy (SPEC-OCCU) analyses identified keystone microbial taxa, including positively correlated taxa with biocontrol and metabolic versatility (e.g., Trichoderma, Solicoccozyma) and negatively correlated potential pathogens (e.g., Alternaria). Co-occurrence network analysis revealed that the TF85+ECF group streamlined bacterial network architecture while enhanced fungal network complexity and connectivity. Mantel tests and correlation analyses indicated that soil organic carbon, BG activity, and cbbLR gene abundance were significantly linked to microbial community structure, and keystone taxa were strongly correlated with soil nutrient cycling functions. Our findings provide a microbiome-based strategy and a novel perspective for sustainable wheat production and targeted microbial management in agriculture.