Microbial diversity inside pumpkins: microhabitat-specific communities display a high antagonistic potential against phytopathogens.

Recent and substantial yield losses of Styrian oil pumpkin (Cucurbita pepo L. subsp. pepo var. styriaca Greb.) are primarily caused by the ascomycetous fungus Didymella bryoniae but bacterial pathogens are frequently involved as well. The diversity of endophytic microbial communities from seeds (spermosphere), roots (endorhiza), flowers (anthosphere), and fruits (carposphere) of three different pumpkin cultivars was studied to develop a biocontrol strategy. A multiphasic approach combining molecular, microscopic, and cultivation techniques was applied to select a consortium of endophytes for biocontrol. Specific community structures for Pseudomonas and Bacillus, two important plant-associated genera, were found for each microenvironment by fingerprinting of 16S ribosomal RNA genes. All microenvironments were dominated by bacteria; fungi were less abundant. Of the 2,320 microbial isolates analyzed in dual culture assays, 165 (7%) were tested positively for in vitro antagonism against D. bryoniae. Out of these, 43 isolates inhibited the growth of bacterial pumpkin pathogens (Pectobacterium carotovorum, Pseudomonas viridiflava, Xanthomonas cucurbitae); here only bacteria were selected. Microenvironment-specific antagonists were found, and the spermosphere and anthosphere were revealed as underexplored reservoirs for antagonists. In the latter, a potential role of pollen grains as bacterial vectors between flowers was recognized. Six broad spectrum antagonists selected according to their activity, genotypic diversity, and occurrence were evaluated under greenhouse conditions. Disease severity on pumpkins of D. bryoniae was significantly reduced by Pseudomonas chlororaphis treatment and by a combined treatment of strains (Lysobacter gummosus, P. chlororaphis, Paenibacillus polymyxa, and Serratia plymuthica). This result provides a promising prospect to biologically control pumpkin diseases.