Comparative genomics of Bacillus cereus sensu lato spp. biocontrol strains in correlation to in-vitro phenotypes and plant pathogen antagonistic capacity

Abstract

Bacillus cereus sensu lato (Bcsl) strains are widely explored due to their capacity to antagonize a broad range of plant pathogens. These include B. cereus sp. UW85, whose antagonistic capacity is attributed to the secondary metabolite Zwittermicin A (ZwA). We recently isolated four soil and root-associated Bcsl strains (MO2, S-10, S-25, LSTW-24) that displayed different growth profiles and in-vitro antagonistic effects against three soilborne plant pathogens models: Pythium aphanidermatum (oomycete) Rhizoctonia solani (basidiomycete), and Fusarium oxysporum (ascomycete). To identify genetic mechanisms potentially responsible for the differences in growth and antagonistic phenotypes of these Bcsl strains, we sequenced and compared their genomes, and that of strain UW85 using a hybrid sequencing pipeline. Despite similarities, specific Bcsl strains had unique secondary metabolite and chitinase-encoding genes that could potentially explain observed differences in in-vitro chitinolytic potential and anti-fungal activity. Strains UW85, S-10 and S-25 contained a (~500 Kbp) mega-plasmid that harbored the ZwA biosynthetic gene cluster. The UW85 mega-plasmid contained more ABC transporters than the other two strains, whereas the S-25 mega-plasmid carried a unique cluster containing cellulose and chitin degrading genes. Collectively, comparative genomics revealed several mechanisms that can potentially explain differences in in-vitro antagonism of Bcsl strains toward fungal plant pathogens.

Keywords: biocontrol agent; chitinase; comparative genomics; phytopathogen; secondary metabolites; zwittermicin.

Figure 1

Dual culture inhibition assay of the five Bcsl strains against soilborne phytopathogens. Images of dual culture experiments (A); and inhibition zones measurements in dual cultures of 3 days old Bcsl strains with Pythium aphanidermatum (B); Rhizoctonia solani (C); and Fusarium oxysporum (D) on 50% LB agar at 25°C. Images are shown following 3 days of incubation for P. aphanidermatum and R. solani and 16 days of incubation for F. oxysporum. Box plots represent data from three independent experiments with five replicates each. The lines in the box plot represent the median while the x symbols represent the mean. Different letters indicate statistically significant differences (p < 0.05) based on the ANOVA Tukey–Kramer post hoc test (α = 0.05).

Figure 2

Antagonistic effect of cell-free supernatant (CFS) of the five Bcsl strains against the three model phytopathogens. Images showing antagonistic effect on mycelial growth of the three phytopathogen, F. oxysporum, R. solani, and P. aphanidermatum inoculated in the center of 9 cm PDA plates, mediated by CFS of 5 days old Bcsl strains grown on 50% LB medium (A); Inhibition zone measurments of R. solani and P. aphanidermatum mycelia by the CFS of the five Bcsl strains (B). The graph represents data from three independent experiments, and the vertical line shows standard deviations between replicates.

Figure 3

Heatmap and phylogenomic tree (A) showing similarity of the five-targeted Bcsl strain genomes relative to other species of Bacillus, based on average nucleotide identity (ANI) values, calculated using the Orthologous Average Nucleotide Identity Tool (OAT); Venn diagrams constructed using the OrthoVenn 2 online service displaying the distribution of shared and unique orthologous clusters among the five Bcsl strains (B).

Figure 4

Chitinolitic activity of five Bcsl strains. Area of clearing zones indicating the different chitinolytic activity of the five Bcsl strains on M9 minimal medium containing colloidal chitin as a sole carbon source. The graph showing the mean and standard deviations of 10 independent measurements (left); and image illustrating the extracellular chitinolytic activity of the five isolates (right). Different letters indicate statistically significant differences (p < 0.05) based on the ANOVA Tukey-HSD test (α = 0.05).

Figure 5

Occurrence of secondary metabolite encoding biosynthetic gene clusters with putative antifungal activity in the five analyzed Bcsl strains based on Antismash predictions. Heatmap shades represent the percentage of genes in the queried cluster that are similar to the known BGC. The darker the color the higher the percentage of similar genes in the cluster to the known BGC. ^aCDPS: cyclic dipeptides. ^bRiPPs: Ribosomally synthesized and post-translationally modified peptides.

Figure 6

Comparative analysis of plasmids containing ZwA BGC homologues in Bacillus cereus spp. UW85, S-10 and S-25. (A) Venn-diagram showing similar and unique RAST annotated genes in the three plasmids. (B) Genetic map of the S-10 (green) and S-25 (blue) plasmids aligned against the UW85 reference plasmid (red) using the BRIG software package. The annotations of relevant encoded proteins from three sequenced plasmids appear in the outer black ring, and the GC content of the reference plasmid is displayed between the inner black and red rings.

Figure 7

Synteny of ZmA homolog BGCs. The reference ZwA BGC of UW-85 (middle), is flanked by S-10 (top), and S-25 (bottom) homologues. The annotations of relevant encoded proteins from three BGCs are indicated by arrows. ZmA BGC genes are shown in black capital letters A–V, and kanosamine encoding genes in red capital letters. The figure was generated using Easyfig software with a cut-off of 90% gene identity.