Effect of antibiosis on antagonist dose-plant disease response relationships for the biological control of crown gall of tomato and cherry.

ABSTRACT The crown gall pathosystem was used to evaluate a model that describes the dose-response relationship between biological control agents and plant pathogens. The model predicts that this relationship can become asymptotic, such that increased antagonist doses cannot compensate for deficiencies in disease suppression. Wounded roots of tomato (Lycopersicon esculentum) and cherry (Prunus mahaleb) plants were dipped into different concentrations of the biological control organism Agrobacterium radiobacter strain K84 prior to inoculation with the pathogen A. tumefaciens. Pathogen strains sensitive or resistant to the antibiotic agrocin 84 were used, and for tomato experiments, a derivative of A. radiobacter strain K84 that does not produce agrocin 84 also was included as an experimental treatment. As predicted by the dose-response model, the amount of disease suppression per unit of antagonist decreased with increasing antagonist dose and became asymptotic at high antagonist densities. Control of crown gall of tomato was nearly complete with the combination of A. radiobacter K84 and an agrocin 84-sensitive strain of A. tumefaciens. Pathogen resistance to agrocin 84 or lack of agrocin 84 production by A. radiobacter resulted in antagonist dose-crown gall incidence relationships that were apparently asymptotic at levels of control significantly less than 100%. For field-grown cherry, similar dose-response relationships were observed with higher asymptotic levels of disease suppression obtained when trees were inoculated with an agrocin 84-sensitive A. tumefaciens strain compared with an agrocin 84-resistant pathogen strain. The differences among bacterial strain combinations in the magnitude of the asymptote defined by the dose-response relationships suggest that A. radiobacter impacts a smaller proportion of the pathogen population when the activity of agrocin 84 is muted.