From Lab to Field: Context-Dependent Impacts of Pseudomonas-Produced 2,4-Diacetylphloroglucinol on Soil Microbial Ecology.

The secondary metabolite 2,4-diacetylphloroglucinol (2,4-DAPG), which is produced by Pseudomonas bacteria, is a potent antimicrobial agent with well-documented properties that suppress phytopathogens. However, its broader ecological impact on soil microbial communities is not understood. Through a combination of controlled microcosm and field trials, we have demonstrated that the effects of 2,4-DAPG are highly context-dependent. Laboratory exposure (10 mg kg-1) altered the abundance of 8.53% of bacterial and 6.91% of fungal amplicon sequence variants, and simplified the bacterial co-occurrence networks (reduced number of nodes and links). In contrast, field conditions amplified bacterial sensitivity (the Shannon index decreased from 4.77 to 4.17, p < 0.05) but maintained fungal stability (Shannon index varied from 3.93 to 3.97, p > 0.05); these conditions affected a smaller proportion of fungal ASVs (4.23%). Taxonomic analysis revealed consistent suppression of fungi of the Mucoromycota (e.g., Mortierella) and context-dependent shifts in bacteria, with an enrichment of Bacillota (e.g., Bacillus, Paenibacillus) in the laboratory but not in the field. Enzymatic responses revealed a dose-dependent activation of the C-cycle, with up to 7.4-fold increases in the laboratory and up to a 10.5-fold increase in the field. P- and N- cycles showed more complex dynamics, with acid phosphatase activity increasing 3.8-fold in laboratory conditions and recovering from initial suppression to an increase of 144% in field conditions, while N-acetylglucosaminidase activity increased and L-leucine aminopeptidase decreased under laboratory conditions. Our results suggest that the response of microorganisms to 2,4-DAPG in natural soils is reduced, probably due to functional redundancy and pre-adaptation to abiotic stresses. This difference between laboratory and field studies warns against extrapolating data from controlled experiments to predict outcomes in agricultural ecosystems, and emphasizes the need for a context-specific evaluation of biocontrol agents.