Phenanthrene toxicity and dissipation in rhizosphere of grassland plants (Lolium perenne L. and Trifolium pratense L.) in three spiked soils

Rhizodegradation is a technique involving plants that offers interesting potential to enhance biodegradation of persistent organic pollutants such as polycyclic aromatic hydrocarbons (PAHs). Nevertheless, the behaviour of PAHs in plant rhizosphere, including micro-organisms and the physico-chemical soil properties, still needs to be clarified. The present work proposes to study the toxicity and the dissipation of phenanthrene in three artificially contaminated soils (1 g kg(-1) DW). Experiments were carried out after 2 months of soil aging. They consisted in using different systems with two plant species (Ryegrass-Lolium perenne L. var. Prana and red clover-Trifolium pratense L. var. fourragere Caillard), three kinds of soils (a silty-clay-loam soil La Bouzule, a coarse sandy-loam soil Chenevieres and a fine sandy-loam soil Maconcourt). Phenanthrene was quantified by HPLC in the beginning (T (0)) and the end of the experiments (30 days). Plant biomass, microbial communities including mycorrhizal fungi, Rhizobium and PAH degraders were also recorded. Generally phenanthrene contamination did not affect plant biomass. Only the red clover biomass was enhanced in Chenevieres and La Bouzule polluted soils. A stimulation of Rhizobium red clover colonisation was quantified in spiked soils whereas a drastic negative phenanthrene effect on the mycorrhization of ryegrass and red clover was recorded. The number of PAH degraders was stimulated by the presence of phenanthrene in all tested soils. Both in ryegrass and red clover planted soils, the highest phenanthrene dissipation due to the rhizosphere was measured in La Bouzule soils. On the contrary, in non-planted soils, La Bouzule soils had also the lowest pollutant dissipation. Thus, in rhizospheric and non-rhizospheric soils the phenanthrene dissipation was found to depend on soil clay content.