Comparative Analysis of Rice Response to Rhizobacteria Highlights Transcriptional Signatures to Detrimental and Beneficial Interactions

The microbiota of plants is composed of a large diversity of microbes, engaging with their host via various types of interactions, from beneficial to detrimental. Consequently, how plants perceive mutualistic from pathogenic microbes and which molecular pathways drive the outcome of the interaction (partnership or defense) are critical questions for the understanding of plant–microbe interactions. In this study, we evaluated the rice transcriptional response to a diversity of bacteria spanning from root pathogens to plant growth-promoting rhizobacteria (PGPRs) in hydroponic conditions. A transcriptomic analysis of rice leaf and root tissues was performed to identify common and specific transcriptional signatures during interaction with non-plant-isolated (Escherichia coli and Stenotrophomonas maltophilia), plant-isolated (Pseudomonas defensor and Bradyrhizobium sp. ORS278), or rice-isolated bacteria (Azospirillum sp. B510 and Burkholderia plantarii). Among all tested groups, the pools of differentially expressed genes (DEGs) modulated by all bacteria or by the PGPRs represented only a fraction of the total DEGs, highlighting the specificity of response to each bacterium. However, plant-isolated bacteria induced a stronger response than non-plant-isolated bacteria. The rice response to PGPRs, pertaining to growth induction and stress mitigation, allowed us to identify putative marker genes of beneficial interaction. Ten genes (such as OsPR1b and rNBS41, related to defense response) were oppositely regulated: repressed upon pathogenic and induced upon PGPR interactions. These genes could be pivotal to triggering mutualism or immunity signals. These transcriptomic clues to bacterial mitigation of rice health were confirmed in a greenhouse experiment, in which PGPRs demonstrated strong biocontrol potential against a foliar phytopathogen. [Figure: see text] Copyright © 2026 The Author(s). This is an open access article distributed under the CC BY 4.0 International license.