Stenotrophomonas maltophilia impedes Bacillus biocontrol of tomato wilt disease by degrading its lipopeptide antibiotics

Abstract Harnessing antibiotic-producing microorganisms that antagonize pathogens represents a sustainable approach for plant disease management. However, biocontrol agents that are effective in the laboratory often have diminished or variable performance in the field. It is often assumed that microbial interactions within the plant rhizosphere can influence the performance of biocontrol agents. To validate this hypothesis, we established a tripartite bacterial model system based on field investigations, involving antibiotic producers (Bacillus amyloliquefaciens P224, Bacillus subtilis P165, and Bacillus velezensis P63), an antibiotic degrader (Stenotrophomonas maltophilia P373), and a bacterial plant pathogen (Ralstonia solanacearum PA1). The selected Bacillus species antagonize R. solanacearum and act as biocontrol agents of the bacterial wilt of tomatoes caused by this pathogen. We demonstrated that S. maltophilia diminished this biocontrol effect by degrading the lipopeptide antibiotics iturin, fengycin, and surfactin secreted by Bacillus spp., thereby serving as a “pathogen helper” that indirectly facilitated pathogen invasion. Further transcriptomic and proteomic analyses revealed that the lipopeptide inactivation mechanism in S. maltophilia involved multi-drug efflux systems, ribosomal adaptation, and enzymatic hydrolysis. Additionally, the interspecies interactions in our model system are modulated by nutrient availability, with elevated carbon sources enhancing the interference competitive ability of Bacillus spp. against S. maltophilia, thereby mitigating its negative impact on the biocontrol of R. solanacearum. Our study sheds light on the complex interactions among plant pathogens, biocontrol agents, and the indigenous microbial community, underscoring the necessity to account for native antibiotic-degrading organisms when applying biocontrol strategies for effective disease management.