羊肚菌稳产与绝收地块根际土壤细菌 / 真菌群落结构差异及其与出菇率的关联

Abstract

Morel (Morchella spp.) is a high-value edible and medicinal fungus widely cultivated in Southwest China. Large-scale field cultivation has expanded rapidly in recent years, yet yield divergence between adjacent plots under identical field management remains a major industrial bottleneck. Some plots achieve stable and high yield, while neighboring plots show normal mycelial colonization but fail to form primordia and end up with total yield loss. Traditional soil physicochemical tests cannot reliably predict such failure. To reveal the differences in rhizosphere bacterial and fungal communities between stable-yield and total-failure plots, and clarify the linkage between microbiome and primordium emergence rate, paired field plots were selected in Chengdu Plain (Sichuan Province). Rhizosphere soil samples were collected at the critical primordium differentiation stage. High-throughput sequencing of 16S rRNA and ITS was applied to characterize bacterial and fungal communities. Alpha diversity, Beta diversity, taxonomic composition, biomarker species, soil physicochemical properties and field fruiting data were analyzed with Mantel test, random forest and co-occurrence network analysis. Results showed that soil pH, organic matter, available nitrogen, phosphorus and potassium had no significant difference between stable and failed plots (P > 0.05), indicating soil physicochemical properties were not the decisive factor for total failure. The Alpha diversity indices (Shannon, Chao1, Pielou evenness) of bacteria and fungi in failed plots were significantly lower than those in stable plots (P < 0.05). Principal coordinate analysis demonstrated significant separation of bacterial and fungal community structure between two groups (bacteria R² = 0.213, P = 0.021; fungi R² = 0.267, P = 0.001). At phylum level, stable plots harbored higher relative abundance of Proteobacteria and Actinobacteria, while failed plots were enriched in Mucoromycota and pathogenic Ascomycota. LEfSe analysis identified Pseudarthrobacter, Flavobacterium as beneficial biomarkers in stable plots; Fusarium, Paecilomyces and Acremonium were indicator pathogens in failed plots. Mantel test showed fungal community structure correlated more strongly with fruiting rate than bacterial community. Random forest model proved fungal Inverse Simpson index and Pielou evenness could serve as predictive indicators for morel fruiting potential. Co-occurrence network revealed that microbial network in stable plots contained more nodes and stronger modularity with stable microbial interactions, whereas failed plots formed simplified networks dominated by pathogenic fungi that disrupted microecological balance. On this basis, practical field strategies including pre-planting soil microbial diagnosis, targeted soil microecology reconstruction, plot screening, soil pre-treatment, functional microbial inoculant application and in-season microbial monitoring were proposed. This study provides theoretical basis and operable technical framework to mitigate sudden total failure and stabilize morel yield.