Comparative transcriptome analysis reveals the biocontrol mechanism of Bacillus velezensis E68 against Fusarium graminearum DAOMC 180378, the causal agent of Fusarium head blight

Abstract

Fusarium graminearum is the causal agent of Fusarium Head Blight, a serious disease affecting grain crops worldwide. Biological control involves the use of microorganisms to combat plant pathogens such as F. graminearum. Strains of Bacillus velezensis are common biological control candidates for use against F. graminearum and other plant pathogens, as they can secrete antifungal secondary metabolites. Here we study the interaction between B. velezensis E68 and F. graminearum DAOMC 180378 by employing a dual RNA-seq approach to assess the transcriptional changes in both organisms. In dual culture, B. velezensis up-regulated genes related to sporulation and phosphate stress and down-regulated genes related to secondary metabolism, biofilm formation and the tricarboxylic acid cycle. F. graminearum up-regulated genes encoding for killer protein 4-like proteins and genes relating to heavy metal tolerance, and down-regulated genes relating to trichothecene biosynthesis and phenol metabolism. This study provides insight into the molecular mechanisms involved in the interaction between a biocontrol bacterium and a phytopathogenic fungus.

Fig 1. Dual-culture interaction plate setup.

Fusarium graminearum DAOMC 180378 hyphae (dark grey) was grown from an agar plug placed at the centre of the plate on a cellulose membrane. Colonies of Bacillus velezensis E68 (light grey) were grown from aliquots of cell suspension spotted 2.5 cm from the centre of the plate. Striped areas were collected with a sterile loop or spatula for RNA extraction. Diagram is not to scale.

Fig 2. Schematic of experimental workflow of RNA-seq analysis of Bacillus velezensis and Fusarium graminearum.

Cultures of B. velezensis E68 and F. graminearum DAOMC 180378 were grown in single and dual culture. The striped areas were collected with a sterile spatula before RNA extraction. Total RNA was sent for library preparation and RNA sequencing. Bacterial RNA was depleted for ribosomal RNA, while fungal RNA was enriched with poly-A selection. Samples were sequenced on an Illumina HiSeq using a paired-end 150 bp protocol. Raw sequencing reads were pre-processed by removing adaptors and poor quality bases and reads. Filtered reads were aligned to their respective genomes and counted for each gene. Gene counts were normalized by the TMM method, low expression genes were removed from analysis and differential expression was calculated. In order to functionally annotate the genes, each gene sequence was used in a BLASTx search to compare to the UniprotKB protein database, annotating each gene with a description and gene ontology terms. The resulting GO terms were used in GO enrichment analysis. Known biosynthetic gene clusters were used in rotation gene set testing to determine their overall regulation pattern.

Fig 3. B. velezensis E68 inhibits the growth of F. graminearum DAOMC 180378 in dual culture.

(A) F. graminearum DAOMC 180378 grown in single culture on PDA, (B) Dual culture condition for B. velezensis E68 and F. graminearum DAOMC 180378 on PDA. (C) Hyphal morphology of F. graminearum in single culture. (D) and (E) Hyphal morphology of F. graminearum in dual culture with B. velezensis. Microscopy performed at 40x magnification.

Fig 4

Log₂ fold change of selected genes between single and dual culture based on RNA-seq and qPCR for (A) B. velezensis E68 and (B) F. graminearum DAOMC 180378. Asterisks (*) indicate significantly differentially expressed genes for RNA-seq and indicate p < 0.05 for qPCR.