Indirect reduction of Ralstonia solanacearum via pathogen helper inhibition

Abstract

The rhizosphere microbiome forms a first line of defense against soilborne pathogens. To date, most microbiome enhancement strategies have relied on bioaugmentation with antagonistic microorganisms that directly inhibit pathogens. Previous studies have shown that some root-associated bacteria are able to facilitate pathogen growth. We therefore hypothesized that inhibiting such pathogen helpers may help reduce pathogen densities. We examined tripartite interactions between a model pathogen, Ralstonia solanacearum, two model helper strains and a collection of 46 bacterial isolates recovered from the tomato rhizosphere. This system allowed us to examine the importance of direct (effects of rhizobacteria on pathogen growth) and indirect (effects of rhizobacteria on helper growth) pathways affecting pathogen growth. We found that the interaction between rhizosphere isolates and the helper strains was the major determinant of pathogen suppression both in vitro and in vivo. We therefore propose that controlling microbiome composition to prevent the growth of pathogen helpers may become part of sustainable strategies for pathogen control.

Fig. 1. Conceptual overview of direct and indirect effects of rhizobacterial strains on pathogen growth.

In this work, we subdivided the net, apparent effect of single rhizosphere bacterial isolates (R) on pathogen density into direct effects on the pathogen (P) and indirect effects mediated by interactions with helper bacteria (H).

Fig. 2. Taxonomic characterization of rhizobacterial isolates that inhibited or helped the growth of Ralstonia solanacearum.

A Distribution of in vitro effects of 160 rhizobacterial supernatants on R. solanacearum growth. The red vertical line represents no effect on R. solanacearum growth. B Cladogram depicting the phylogenetic relationship among the 160 isolates based on their full-length 16 S rRNA gene sequences. The inner ring depicts the different effect of isolates supernatant on R. solanacearum growth: positive effect (blue), negative effect (red) and no significant effect (gray). The outer ring shows the four phyla to which the isolates belong. C The proportion of rhizobacterial isolates per phylum whose supernatant showed inhibitory, stimulatory or no effect on R. solanacearum growth. The size of the circles represents the number of rhizobacterial isolates in the given phylum. The thickness of lines represents the percentage of rhizobacterial isolates that have the indicated effect on R. solanacearum growth in each phylum.

Fig. 3. Effect of helper strains on Ralstonia solanacearum growth and plant disease severity.

Effects of the two helper strains Phyllobacterium ifriqiyense (Pi) and Microbacterium paraoxydans (Mp) on Ralstonia solanacearum (Rs) growth in vitro (A) and in vivo (B) and on plant disease severity (C). Different letters indicate significant differences based on Tukey post hoc test. Error bars show ±1 SE (n = 3 for in vitro, n = 4 for in vivo). D Effects of 46 rhizobacterial strains on the growth of R. solanacearum and the two model helper strains in vitro. The x-axis shows the direct effect of each rhizobacterial strain on R. solanacearum growth (data from the experiment in which R. solanacearum was grown in the presence of supernatant from each of the 46 rhizobacterial strains—the same data is presented on the x axis of Fig. 4A). The y-axis shows the effect of each rhizobacterial strain on each of the two helper strains (data from the experiment in which each helper was grown in the presence of supernatant from each of the 46 rhizobacterial strains—the same data is presented on the x axis of Fig. 4C). In (C), “−1”, “0” and “1” on the x-axis denote that R. solanacearum growth is completely inhibited, not influenced or increased 2× by supernatant from the rhizobacteria, respectively. Similarly, “−1”, “0” and “1” on the y-axis denote the same growth effects with reference to growth of the helper strains. Black dots indicate results involving interactions with Pi, and red dots indicate results involving interactions with Mp.

Fig. 4. The importance of direct versus indirect effects on Ralstonia solanacearum density and disease severity in the presence of helper strains.

In the presence of helper Phyllobacterium ifriqiyense (Pi) or Microbacterium paraoxydans (Mp), respectively, the importance of direct effects on the density of R. solanacearum both (A) in vitro and (B) in vivo, and (C) disease severity (the data on the x axis of (A) are the same data which was presented on the x axis of Fig. 3C, the data on x axis of (B) and (C) are part of the data on x axis of (A)); the importance of indirect effects on the density of R. solanacearum both (D) in vitro and (E) in vivo, and (F) disease severity (the data on the x axis of (D) are the same data which was presented on the y axis of Fig. 3C, the data on x axis of (E) and (F) are part of the data on x axis of (D)). In all panels, “−1”, “0” and “1” on the x-axis denote that R. solanacearum growth (A, B, and C) or helper growth (D, E, and F) is completely inhibited, not influenced or increased 2× by supernatant from the rhizobacteria, respectively.

Fig. 5. The relative importance of direct versus indirect effects on Ralstonia solanacearum density and disease severity in the presence of helper strains.

Relative importance of direct versus indirect effects on Ralstonia solanacearum density both in vitro (A) and in vivo (B), and disease severity (C) in presence of helper strains on the interaction scenario where rhizobacterial strains inhibited both the pathogen and its helpers (quadrant “H⁻P⁻” in Fig. 3C). This shows the predicted R. solanacearum densities and disease incidence for different values of the inhibition via pathogen (Direct) or helper (Indirect) as estimated from the statistical model (Table 1) which with direct effects, indirect effects, and an interaction between helper strains and indirect effects as fixed factors. For the Direct line, the indirect effect was set to zero, while for the indirect line, the direct effect was set to zero.