Inability to find consistent bacterial biocontrol agents of Pythium aphanidermatum in cucumber using screens based on ecophysiological traits.

A collection of 821 rhizobacteria from cucumber, originating from different root locations and stages of plant development, was screened for potential biocontrol agents of Pythium aphanidermatum (Edson) Fitzp. The screening procedure exploited carbon source utilization profiles and growth rates of bacteria as indicators of a partial niche overlap with the pathogen. The bacteria were tested for growth on nine carbon sources (glucose, fucose, sucrose, maltose, asparagine, alanine, galacturonic acid, succinic acid, and linoleic acid), most of which are reported to be used by the zoospores of P. aphanidermatum in the infection process. The isolates were classified as fast- or slow-growing, depending on their growth rate in 1/10 strength TSB. By nonhierarchical cluster analysis, 20 clusters were generated of bacteria with similar profiles of carbon source utilization. Redundancy analysis showed that the type of root sample explained 47% of the variance found in the relative abundance of bacteria from the clusters. Bacteria from clusters using none or few of the carbon sources, e.g., maltose and linoleic acid, with many slow-growing isolates, showed a preference for plants in the vegetative or generative stage, or for old root regions (root base). Bacteria from clusters with fast-growing isolates, using many carbon sources, were relatively abundant in the seedling stage. A selection of 127 bacteria from the different clusters was tested for disease suppressive capabilities in bioassays on young cucumber plants in nutrient solution, inoculated with zoospores of P. aphanidermatum. Nine of these bacteria produced biosurfactants, and 27 showed antibiosis against mycelial growth in plate assays. For 31 isolates, significant positive effects on plant biomass were shown, as analyzed with a general linear regression model. For most isolates, these effects occurred only in one of two replicate assays and no reductions in the degree of root and crown rot were found. Of the isolates that used many of the tested carbon sources, only four had positive effects on plant biomass. The majority of the isolates that positively affected plant biomass used few to moderate numbers of carbon sources and did not produce antibiotics or biosurfactants. In conclusion, competition for the tested carbon sources with the zoospores did not play a decisive role in disease suppression, and no clear relation was found between ecophysiological traits and disease suppression. Only isolate 3.1T8, isolated from root tips in the generative stage of plant growth, significantly increased plant biomass and suppressed root and crown rot symptoms in five out of six bioassays. The isolate produced an antifungal substance in plate assays and showed biosurfactant production in several (cucumber-derived) media.