Volatile-mediated antagonism of soil bacterial communities against fungi

Abstract

Competition is a major type of interaction between fungi and bacteria in soil and is also an important factor in suppression of plant diseases caused by soil-borne fungal pathogens. There is increasing attention for the possible role of volatiles in competitive interactions between bacteria and fungi. However, knowledge on the actual role of bacterial volatiles in interactions with fungi within soil microbial communities is lacking. Here, we examined colonization of sterile agricultural soils by fungi and bacteria from non-sterile soil inoculums during exposure to volatiles emitted by soil-derived bacterial communities. We found that colonization of soil by fungi was negatively affected by exposure to volatiles emitted by bacterial communities whereas that of bacteria was barely changed. Furthermore, there were strong effects of bacterial community volatiles on the assembly of fungal soil colonizers. Identification of volatile composition produced by bacterial communities revealed several compounds with known fungistatic activity. Our results are the first to reveal a collective volatile-mediated antagonism of soil bacteria against fungi. Given the better exploration abilities of filamentous fungi in unsaturated soils, this may be an important strategy for bacteria to defend occupied nutrient patches against invading fungi. Another implication of our research is that bacterial volatiles in soil atmospheres can have a major contribution to soil fungistasis.

Figure 1

Flow diagram of the key experimental procedure in this study. The sandy soils #1, #2, and #3 were collected from three agricultural fields in the Netherlands. The left small Petri‐dish served as bacterial community volatile‐producing compartment. Two growth media, namely REA and 1/10 TSB agar (TSBA) containing the fungal inhibitors cycloheximide and thiabendazole, were used to grow bacteria extracted from the three soils. The other small Petri‐dish containing 20 g sterilized soil was used as the volatiles receiving compartment. After closing the big Petri‐dish (9 cm dia) and pre‐exposure for 3 days to volatiles originating from the producer compartment, 0.5 g of fresh soil was added to the centre of the receiver compartment. Next, the big Petri‐dish was closed again and development of fungal and bacterial biomass and communities was determined after 14 days of incubation. As controls, producer compartments (n = 18) containing the respective growth media without bacteria inoculums were used.

Figure 2

Effect of volatiles produced by soil bacterial communities growing on two agar media on fungal biomass (A), and bacterial 16S gene copy numbers (B) developing in sterile soil receiver compartments. Colonization of sterile soil by fungi and bacteria in volatile receiving compartments started with addition of a small amount of non‐sterile soil and growth was determined after 14 days of exposure to volatiles produced by bacteria growing on/in ‘REA’: root‐exudate agar, and ‘TSBA’: 1/10 TSB agar. Controls consisted of the same growth media without inoculation of bacteria. Mean values and standard deviations are presented (n = 3). The asterisks indicate that differences between soils exposed to bacterial volatiles and controls are statistically significant (*p < 0.05, **p < 0.01) as determined by Fisher's LSD test.

Figure 3

A. Principal component analysis based on Bray–Curtis results of the composition of fungal communities in receiver compartments (sterile soils inoculated with fresh soil) that were exposed (VCs) or not (CK) to volatiles produced by soil bacterial communities growing on two agar media (REA, 1/10 TSBA). The percentages on the axis labels represent the percentages of variation explained by the principal coordinates. B. Mean proportion (n = 6) of the most abundant fungal genera (ITS sequences >1%) in soils that were exposed (VCs) or not (CK) to volatiles produced by soil bacterial communities growing on two agar media (REA, 1/10 TSBA). *p < 0.05, **p < 0.01.

Figure 4

A. Heatmaps showing the composition of bacteria OTUs extracted from three agricultural soils and growing on root‐exudate agar (REA). The x‐axis corresponds to individual samples with three replicates per soil bacterial community. The y‐axis corresponds to the relative abundance of each bacterial OTU. Hierarchical cluster analysis was based on the Bray–Curtis dissimilarity with complete‐linkage method. B. Partial least square‐discriminant analysis (PLS‐DA) 2D‐score plots of volatiles profiles emitted by bacterial communities extracted from three agricultural soils (S1, S2 and S3) and growing on REA, including 95% confidence regions. The explained variances are shown in brackets.