Pseudomonas fluorescens and Glomus mosseae trigger DMI3-dependent activation of genes related to a signal transduction pathway in roots of Medicago truncatula

Abstract

Plant genes induced during early root colonization of Medicago truncatula Gaertn. J5 by a growth-promoting strain of Pseudomonas fluorescens (C7R12) have been identified by suppressive subtractive hybridization. Ten M. truncatula genes, coding proteins associated with a putative signal transduction pathway, showed an early and transient activation during initial interactions between M. truncatula and P. fluorescens, up to 8 d after root inoculation. Gene expression was not significantly enhanced, except for one gene, in P. fluorescens-inoculated roots of a Myc(-)Nod(-) genotype (TRV25) of M. truncatula mutated for the DMI3 (syn. MtSYM13) gene. This gene codes a Ca(2+) and calmodulin-dependent protein kinase, indicating a possible role of calcium in the cellular interactions between M. truncatula and P. fluorescens. When expression of the 10 plant genes was compared in early stages of root colonization by mycorrhizal and rhizobial microsymbionts, Glomus mosseae activated all 10 genes, whereas Sinorhizobium meliloti only activated one and inhibited four others. None of the genes responded to inoculation by either microsymbiont in roots of the TRV25 mutant. The similar response of the M. truncatula genes to P. fluorescens and G. mosseae points to common molecular pathways in the perception of the microbial signals by plant roots.

Figure 1.

Quantification of P. fluorescens C7R12 colonizing the rhizoplane (A) and internal root tissues (B) of wild type (J5, ♦) and a DMI3 mutant genotype (TRV25, □) of Medicago truncatula 4 to 21 dai. Suspensions of root washings and of macerated surface sterilized roots were plated on King B medium and the number of cfu quantified.

Figure 2.

Light and electron micrographs of root colonization by P. fluorescens C7R12 (arrows) in wild-type (A–C) and DMI3 mutant (D–F) genotypes of M. truncatula. Sections of roots were fixed in glutaraldehyde and embedded in LR White resin, and observed directly (B, C, E, and F) or after reaction with a polyclonal P. fluorescens antibody (A and D). A and D, Immunolocalization of bacterial cells in the rhizoplane; B and E, bacteria within the root cortex; C and F, bacterial colonies between sloughing root tip cells. Bar = 10 μm.

Figure 3.

Effect of P. fluorescens inoculation on root fresh weight of M. truncatula wild-type J5 (A) and DMI3 mutant genotype (TRV25; B) observed between inoculated (▪) and noninoculated (⋄) roots.

Figure 4.

Expression analyses by RT-PCR of the Mtgap1 gene in M. truncatula roots. Amplification signals of cDNA from roots of wild-type (J5) and DMI3 mutant (TRV25) genotypes were quantified and compared between noninoculated roots (NI) and roots inoculated with P. fluorescens, G. mosseae, or S. meliloti.

Figure 5.

Relative abundance of transcripts of 10 M. truncatula genes from P. fluorescens-inoculated (black bars) and noninoculated (white bars) roots obtained by semiquantitative RT-PCR in wild-type J5 (A) and DMI3 mutant genotype TRV25 (B) from 4 to 21 dai.

Figure 6.

Relative abundance of transcripts of 10 M. truncatula genes from noninoculated (white bars) roots and roots inoculated with G. mosseae (A) or S. meliloti (B) obtained by semiquantitative RT-PCR in wild-type J5 (black bars) and DMI3 mutant genotype TRV25 (hatched bars).