Dynamics of Bacterial and Fungal Communities Related to Mycotoxin Production in Stored Wheat Grain

Wheat grain deterioration during storage poses a critical threat to food security, primarily driven by fungal succession and mycotoxin contamination. In this study, we simulated high-humidity (97%) and warm (28°C) storage conditions to characterize the dynamics of bacterial and fungal communities and their correlation with mycotoxin production. We identified three distinct phases of mycotoxin accumulation: a non-toxigenic early stage (D0-D6), an AFB1-only middle stage (D15-D25), and a multi-mycotoxin late stage (D35-D50) characterized by the co-occurrence of AFB1, ZEN, and DON. High-throughput sequencing revealed a pronounced simplification of the microbial ecosystem. Bacterial communities shifted from diverse Pseudomonadota (formerly Proteobacteria) and Firmicutes to a near-monoculture of Actinobacteriota (reaching 99.96% dominance), within the phylum Actinobacteriota, a succession from Saccharopolyspora to Streptomyces also occurred. Fungal communities transitioned from field fungi (Alternaria) to storage molds (Aspergillus), followed by xerophilic taxa (Pithoascus and Microascus). Notably, the emergence of DON at D35 coincided with a transient, specific increase in Fusarium relative abundance (1.37%). Co-occurrence network analysis demonstrated a sharp collapse in ecological connectivity, distinguishing the complex networks of mycotoxin-negative samples from the fragmented structures observed in mycotoxin-positive stages. Functional prediction further indicated a convergence towards saprotrophic lifestyles. These findings demonstrate that mycotoxin co-contamination is a system-level outcome of specific bacterial and fungal succession and ecological network destabilization, providing valuable biomarkers for early warning of grain spoilage.