Effect of Two Plant Species, Flax (Linum usitatissinum L.) and Tomato (Lycopersicon esculentum Mill.), on the Diversity of Soilborne Populations of Fluorescent Pseudomonads

Suppression of soilborne disease by fluorescent pseudomonads may be inconsistent. Inefficient root colonization by the introduced bacteria is often responsible for this inconsistency. To better understand the bacterial traits involved in root colonization, the effect of two plant species, flax (Linum usitatissinum L.) and tomato (Lycopersicon esculentum Mill.), on the diversity of soilborne populations was assessed. Fluorescent pseudomonads were isolated from an uncultivated soil and from rhizosphere, rhizoplane, and root tissue of flax and tomato cultivated in the same soil. Species and biovars were identified by classical biochemical and physiological tests. The ability of bacterial isolates to assimilate 147 different organic compounds and to show three different enzyme activities was assessed to determine their intraspecific phenotypic diversity. Numerical analysis of these characteristics allowed the clustering of isolates showing a high level (87.8%) of similarity. On the whole, the populations isolated from soil were different from those isolated from plants with respect to their phenotypic characteristics. The difference in bacteria isolated from uncultivated soil and from root tissue of flax was particularly marked. The intensity of plant selection was more strongly expressed with flax than with tomato plants. The selection was, at least partly, plant specific. The use of 10 different substrates allowed us to discriminate between flax and tomato isolates. Pseudomonas fluorescens biovars II, III, and V and Pseudomonas putida biovar A and intermediate type were well distributed among the isolates from soil, rhizosphere, and rhizoplane. Most isolates from root tissue of flax and tomato belonged to P. putida bv. A and to P. fluorescens bv. II, respectively. Phenotypic characterization of bacterial isolates was well correlated with genotypic characterization based on repetitive extragenic palindromic PCR fingerprinting.