Protection of Arabidopsis thaliana against leaf-pathogenic Pseudomonas syringae by Sphingomonas strains in a controlled model system

Abstract

Diverse bacterial taxa live in association with plants without causing deleterious effects. Previous analyses of phyllosphere communities revealed the predominance of few bacterial genera on healthy dicotyl plants, provoking the question of whether these commensals play a particular role in plant protection. Here, we tested two of them, Methylobacterium and Sphingomonas, with respect to their ability to diminish disease symptom formation and the proliferation of the foliar plant pathogen Pseudomonas syringae pv. tomato DC3000 on Arabidopsis thaliana. Plants were grown under gnotobiotic conditions in the absence or presence of the potential antagonists and then challenged with the pathogen. No effect of Methylobacterium strains on disease development was observed. However, members of the genus Sphingomonas showed a striking plant-protective effect by suppressing disease symptoms and diminishing pathogen growth. A survey of different Sphingomonas strains revealed that most plant isolates protected A. thaliana plants from developing severe disease symptoms. This was not true for Sphingomonas strains isolated from air, dust, or water, even when they reached cell densities in the phyllosphere comparable to those of the plant isolates. This suggests that plant protection is common among plant-colonizing Sphingomonas spp. but is not a general trait conserved within the genus Sphingomonas. The carbon source profiling of representative isolates revealed differences between protecting and nonprotecting strains, suggesting that substrate competition plays a role in plant protection by Sphingomonas. However, other mechanisms cannot be excluded at this time. In conclusion, the ability to protect plants as shown here in a model system may be an unexplored, common trait of indigenous Sphingomonas spp. and may be of relevance under natural conditions.

Fig. 1.

(A) Time course of axenic and Methylobacterium-inoculated A. thaliana plants. A mixture of five Methylobacterium strains (Table 1) was applied by seed inoculation. Nineteen-day-old plants were mock treated or infected by spraying the plant pathogen P. syringae pv. tomato DC3000 (Pst) onto their leaves and imaged at different days postinfection (dpi). (B) Population dynamics of P. syringae pv. tomato DC3000 in the phyllosphere. Each data point represents the log-transformed mean of 12 plant individuals (CFU per gram of leaf fresh weight). Error bars indicate the standard errors of the means (SEM). Comparable results were obtained when the experiment was repeated independently.

Fig. 2.

(A) Time course of axenic and Sphingomonas-inoculated A. thaliana plants. A mixture of five Sphingomonas strains (Table 1) was applied by seed inoculation. Twenty-one-day-old plants were mock treated or infected by spraying the plant pathogen P. syringae pv. tomato DC3000 (Pst) onto their leaves and imaged at different days postinfection (dpi). (B) Population dynamics of P. syringae pv. tomato DC3000 in the phyllosphere. Each data point represents the log-transformed mean of 12 plant individuals (CFU per gram of leaf fresh weight). Error bars indicate the standard errors of the means (SEM). Comparable results were obtained when the experiment was repeated independently.

Fig. 3.

Snapshot of axenic and Sphingomonas-inoculated A. thaliana plants at 19 days postinfection (dpi). A mixture of five Sphingomonas strains was applied by leaf inoculation 1 week before infection with P. syringae pv. tomato DC3000 (Pst). The plot shows the pathogen population of individual plants at 19 dpi (log-transformed CFU per gram of leaf fresh weight). Means and statistically significant differences are indicated (***, P < 0.001).

Fig. 4.

Snapshot of axenic and Sphingomonas-inoculated A. thaliana plants at 32 days postinfection (dpi). Plants were seed inoculated with a mixture of five Sphingomonas strains and infected with X. campestris pv. campestris (Xcc). The plot shows the pathogen population of individual plants at 32 dpi (log-transformed CFU per gram of leaf fresh weight). Means and statistically significant differences are indicated (***, P < 0.001).

Fig. 5.

Population size of 18 different Sphingomonas (in sensu lato) isolates on A. thaliana plants (top), the corresponding population size of the plant pathogen P. syringae pv. tomato DC3000 (center), and the plant disease index scored from 1 to 5 (bottom). Plants were leaf inoculated with the individual Sphingomonas strains 1 week before infection, and data were collected at 21 dpi. Each data point in the population-size plots represents the means ± SEM of 10 plant individuals (except for strains Fr1 and C3, for which n = 6), whereas means and standard errors of the disease index are based on 24 to 32 individual plants. The gray horizontal bars indicate the pathogen population on and the disease index of axenic plants. The Sphingomonas isolates are sorted according to the disease index and are grouped into fully protective (+), intermediately protective (±), and nonprotective (−) phenotypes.

Fig. 6.

Biolog substrate utilization profiles of the plant pathogen P. syringae pv. tomato DC3000 (Pst) and representative Sphingomonas and Methylobacterium strains. Sphingomonas sp. Fr1 and S. phyllosphaerae showed a plant-protective effect, while S. aerolata, S. wittichii, M. extorquens PA1, and Methylobacterium sp. 32 did not. Black boxes indicate positive values, and gray boxes indicate weakly positive values.