Cooperative interactions between seed-borne bacterial and air-borne fungal pathogens on rice

Abstract

Bacterial-fungal interactions are widely found in distinct environments and contribute to ecosystem processes. Previous studies of these interactions have mostly been performed in soil, and only limited studies of aerial plant tissues have been conducted. Here we show that a seed-borne plant pathogenic bacterium, Burkholderia glumae (Bg), and an air-borne plant pathogenic fungus, Fusarium graminearum (Fg), interact to promote bacterial survival, bacterial and fungal dispersal, and disease progression on rice plants, despite the production of antifungal toxoflavin by Bg. We perform assays of toxoflavin sensitivity, RNA-seq analyses, lipid staining and measures of triacylglyceride content to show that triacylglycerides containing linolenic acid mediate resistance to reactive oxygen species that are generated in response to toxoflavin in Fg. As a result, Bg is able to physically attach to Fg to achieve rapid and expansive dispersal to enhance disease severity.

Fig. 1

Co-detection of Bukholderia glumae (Bg) and Fusarium graminearum (Fg) in rice fields. a Both Bg and Fg were isolated from surface-sterilised rice grains that were placed on PDA for 3 d. b Bg and Fg were detected with specific primer pairs from diseased rice grains. Lane M, 100 bp marker; Lane C, positive control (Fg and Bg mixed DNA). c Coexistence frequencies of Bg and Fg in 50 rice grains collected at 7-d intervals beginning with the flowering season from field-grown black-type rice (BR) and glutinous-type (GR) rice were calculated. Bg and Fg were detected with specific primers

Fig. 2

Resistance of Fg to toxoflavin. a Toxoflavin resistance was examined for Fg WT strain, GZ03639, for a toxoflavin-sensitive mutant derived from GZ03639 (ΔGzZC190), and for a GzZC190-complemented mutant (GzZC190c). Each strain was grown on MM supplemented with toxoflavin as indicated for 4 d. b Lipid staining was performed for GZ03639, ΔGzZC190, GzZC190c, Colletotrichum gloeosporioides (Cg) and Magnaporthe oryzae (Mo) cells. Mycelia were incubated for 24 h in MM containing toxoflavin and/or linolenic acid (C18:3) and stained with Nile Red. Scale bar, 10 μm. c Effect of linolenic acid, toxoflavin, and H₂O₂, in combination or individually, on the strains indicated that were grown for 4 d on supplemented MM

Fig. 3

Quantification of mycelia TAGs by LC–MS/MS and expression patterns of genes involved in the biosynthesis of linolenic acid (18:3). a Dry mycelia of GZ03639 and ΔGzZC190 (n = 3 each) were incubated for 24 h in MM with and without toxoflavin before TAGs were detected with LC–MS/MS. The data presented are the mean ± s.d. (ANOVA; *P < 0.05, **P < 0.01). b RNA-seq analyses were performed to obtain expression profiles of genes involved in the production of linolenic acid (18:3) from palmitic acid (16:0). The KEGG database (http://www.kegg.jp/) and manual annotation based on orthologs of other species were used to annotate genes involved in the production of linolenic acid

Fig. 4

Effects of toxoflavin on the generation of spores and DON production. a Effect of bacterial filtrates on Fg spore production. WT strain GZ03639 was incubated with bacterial culture filtrates for 24 h. BGR1: Bg, DC3000: Pseudomonas syringae, CH67: Burkholderia pyrrocinia. b Effect of toxoflavin on Fg spore production. GZ03639 was incubated for 24 h in CMC or LB medium supplemented with toxoflavin at the concentrations indicated. c Effect of toxoflavin on transcription of the trichothecene biosynthesis genes, Tri5 and Tri6, in GZ03639 grown in GYEP medium (control) and in the presence of toxoflavin (80 mg l⁻¹) or H₂O₂ (1 mM). d Effect of toxoflavin on Fg DON production in GYEP medium (control) and in the presence of toxoflavin (80 mg l⁻¹) or H₂O₂ (1 mM). Bars not sharing a letter are significantly different according to Tukey’s test (P < 0.05, n = 3) and data presented are the mean ± s.d

Fig. 5

Physical association between Bg and Fg. a Bacteria and fungi were co-inoculated inside a culture plate (35 × 10 mm) containing LB agar for up to 10 d in the dark at 30 °C. A field emission scanning electron microscope detected spores in the zone where bacterial cells were present (A) and not in the zone where bacterial cells were absent (B). Scale bar, 10 μm. b Physical association between Fg and Bg. The negative control DC3000 did not show transported hyphae outside of the ring. c Capillary tubes filled with PDB or GZ03639 filtrate were placed into the wells of 96-well plates containing BGR1 or DC3000 bacterial cells. After incubating the plates at 30 °C for 3 h, liquid in the capillaries were spread onto LB agar plates and colonies formed after 24 h were counted. The experiment was repeated twice with three replicates per sample and data presented are the mean ± s.d. (ANOVA; ***P < 0.001)

Fig. 6

Interactions between Bg and Fg on rice plants. a Disease severity of rice plants inoculated with either Bg, Fg or a combination of both. BGR1 and GZ03639 indicate single inoculations with Bg and Fg, respectively. GZ03639 + BGR1 indicates a co-inoculation of both pathogens. b DON was quantified in rice heads inoculated with either Bg, Fg, or a combination of both 7 d and 14 d after inoculation. The experiment was repeated twice with three replicates per sample and data presented are the mean ± s.d. (ANOVA; *P < 0.05, ***P < 0.001). c Images of Fg and Bg in rice husks as observed with a field emission scanning electron microscope. Scale bar, 10 μm