Peng Xu, Lei Li, Yifan Zhang, Qinghong Jiang, Yuhang Zhang, Xiaoqing Wu, Yun Xu
To identify the ecological threshold and microbial mechanisms underlying organic amendment (OA)-mediated remediation of acidic red soils, pak choi was employed as a model plant, and a gradient of digestate-derived OA from 0% to 25% was established. This study integrated soil-plant assessment, community assembly modeling, metagenome-assembled genomes (MAGs), and metabolic network analysis. Both soil functioning and plant growth exhibited nonlinear responses. The optimal OA rate was found to be 15%, which increased the soil quality index and pak choi biomass by 92% and more than 13-fold compared to the control group. In contrast, a 25% OA rate elevated electrical conductivity, induced secondary salinity stress, and reduced productivity and economic returns. Under the 15% OA treatment, the microbial community showed the strongest deviation from neutral community model predictions, indicating that neutral processes had limited explanatory power for community assembly, while deterministic processes associated with altered soil conditions may have played a more important role in community reorganization. Accordingly, core functional MAGs (e.g., MAG302, MAG299, MAG321) were significantly enriched under this treatment, with a total relative abundance 17.3 times that of the control group. These MAGs harbored key genes involved in C, N, P, and S cycling (bglB, atoB, narG, nirK, nosZ, gcd, pst, sqr), suggesting functional complementarity in organic matter degradation, denitrification, phosphorus mobilization, and sulfide oxidation, thereby supporting efficient nutrient turnover and system function. Deviation from this threshold resulted in reduced core MAG enrichment and metabolic network synergy. Overall, this study provides genome-resolved targets for functional strain isolation and synthetic community construction, as well as a mechanistic basis for optimizing OA rates and developing microbiome-based precision remediation strategies.