Applying predictive models to decipher rhizobacterial modifications in common reed die-back affected populations.

The microbiota inhabiting the soil, as well as the rhizosphere, represents a key determinant of several plant functions. Like for humans, dysbiosis of the plant-associated microbiota may be a co-causal agent in disease with still obscure eziology. In the last decades, the common reed Phragmites australis has been deeply studied for its disappearance from natural stands, but no clear causative agents have been identified and no laboratory models of such "reed die-back syndrome" (RDBS) have been developed. In this study, we try to shed light on the RDBS, by comparing the rhizosphere microbiota of five Italian P. australis populations with different degrees of decline. Results obtained showed a biogeographical meaningful pattern of rhizosphere microbiota, coupled with an impact of RDBS. Obtained data allowed to construct a two-steps predictive model which enabled the prediction of the plant health status from the microbiota taxonomic composition, independently from their geographic location. In conclusion, this study represents one of the first overviews that statistically links RDBS to alteration of rhizosphere microbiota and suggests a model for the analysis of plant-bacteria relationships in nature.
(Copyright © 2018 Elsevier B.V. All rights reserved.)