Abstract
The diversity of plant-associated bacteria is vast and can be determined by 16S rRNA gene metabarcoding. Fewer of them have plant-beneficial properties. To harness their benefits for plants, we must isolate them. This study aimed to check whether 16S rRNA gene metabarcoding has predictive power in identifying the majority of known bacteria with plant-beneficial traits that can be isolated from the sugar beet (Beta vulgaris L.) microbiome. Rhizosphere and phyllosphere samples collected during one season at different stages of plant development were analyzed. Bacteria were isolated on rich unselective media and plant-based media enriched with sugar beet leaves or rhizosphere extracts. The isolates were identified by sequencing the 16S rRNA gene and tested in vitro for their plant-beneficial properties (stimulation of germination; exopolysaccharide, siderophore, and HCN production; phosphate solubilization; and activity against sugar beet pathogens). The highest number of co-occurring beneficial traits was eight, found in isolates of five species: Acinetobacter calcoaceticus, Bacillus australimaris, B. pumilus, Enterobacter ludwiigi, and Pantoea ananatis. These species were not detected by metabarcoding and have not previously been described as plant-beneficial inhabitants of sugar beets. Thus, our findings point out the necessity of a culture-dependent microbiome analysis and advocate for low-nutrient plant-based media for high-yield isolation of plant-beneficial taxa with multiple beneficial traits. A culture-dependent and -independent approach is required for community diversity assessment. Still, isolation on plant-based media is the best approach to select isolates for potential use as biofertilizers and biopesticides in sugar beet cultivation.