Bacillus velezensis TB918 mitigates garlic dry rot disease by forming consortia with Pseudomonas in the rhizosphere and bulb

Abstract

Garlic dry rot (GDR), primarily caused by Fusarium proliferatum, is a significant postharvest disease that leads to substantial economic losses. Our previous research demonstrated that supplementing Bacillus-based biocontrol formulations with sucrose could boost its efficiency in protecting plants by building a hostile rhizomicrobiome for destructive soil-borne pathogens. B. velezensis TB918, previously isolated from pepper rhizosphere soil, exhibited a strong in vitro antifungal effect on Fusarium. In this study, we conducted a field experiment to investigate the efficacy of B. velezensis TB918 in controlling GDR, and explored the changes in microbial communities in garlic plants and rhizosphere soil following the application of TB918 with or without sucrose supplementation. Using 16S rRNA and ITS amplicon sequencing, we found that the introduction of TB918 significantly increased the abundance of Pseudomonas in garlic rhizosphere, especially when combined with sucrose. Three Pseudomonas strains were isolated from garlic tissues and rhizosphere soil treated with TB918 and sucrose, among which the GP2 strain exhibited antagonistic effects against pathogen ad planta. Co-culture and colonization assays showed that TB918 facilitated the biofilm formation of Pseudomonas strain by forming consortia. Interestingly, the abundance of potentially non-pathogenic Fusarium concentricum also increased, suggesting a potential niche exclusion effect. Our results demonstrated that TB918 in combination with sucrose effectively reduced the incidence of GDR during storage. This study provides valuable insights into the use of biocontrol agents and sucrose to modulate the garlic microbial community and suppress soil-borne pathogens.

Keywords: Bacillus velezensis TB918; Fusarium; biocontrol; garlic dry rot; rhizosphere microbiota.

Figure 1

Garlic dry rot (GDR) typical symptoms on cloves and effects of biocontrol agents against GDR. (A) Symptoms on garlic cloves were observed 6 months after harvest. The GDR disease severity was graded into six classes: 0 = asymptomatic; 10% = small lesions on cloves; 30% = small depressed brown spots; 60% = one or more extended depressed brown spots; 85% = extended lesions or spots on one or more surfaces on cloves with dehydrated appearance; 100% = completely rotted cloves with dehydrated appearance and visible white mycelium. The bio-agent B. velezensis TB918 and paired with sucrose significantly reduced the incidence rate (B) and disease severity (C) of GDR. CK: irrigation water, P: commercial biocontrol agent P. polymyxa, B: B. velezensis TB918, BS: B. velezensis TB918 and sucrose. Both indexes were calculated as the mean of 10 cloves of each treatment and five test replicates. The bars represent the standard error. The letters above the columns indicate statistically significant differences based on the index using the Student’s t-test (p < 0.05).

Figure 2

Diversity of microbiota associated with garlic bulb inoculating biocontrol agents. Alpha diversity of bacteria (A) and fungi (B) in different garlic bulb tissues and rhizosphere soil after inoculating biocontrol agents. CK (irrigation water), B (B. velezensis TB918), BS (B. velezensis TB918 and sucrose) and P (commercial biocontrol agent P. polymyxa) depict rhizosphere soil samples and different treatments. M (margin soil), G (garlic cloves), I (inner sheaths) and O (outer sheaths) depict different sampling locations. PCoA analysis of bacterial (C) and fungal (D) community structure in different garlic bulb tissues and rhizosphere soil, based on unweighted UniFrac distance. Samples were clustered into five groups: M (purple circle): margin soil group, R (orange triangle): rhizosphere soil group, G (cyan diamond): garlic cloves group, I (yellow square): inner sheaths group, O (blue cross): outer sheaths group.

Figure 3

The bacterial community composition of garlic tissues and rhizosphere. Composition of different bacterial communities at the phylum (A) and genus (B) level. The relative abundance of different bacterial phylum or genus within the different communities is shown in different colors. Others: the merging of taxa with the relative abundance less than 0.01. The difference in the abundance of Bacillus (C) and Pseudomonas (D) based on Kruskal-Wallis H test (p < 0.05). Bars with different letters are significantly different (p < 0.05). CK (irrigation water), B (B. velezensis TB918), BS (B. velezensis TB918 and sucrose) and P (commercial biocontrol agent P. polymyxa) depict rhizosphere soil samples and different treatments. M (margin soil), G (garlic cloves), I (inner sheaths), and O (outer sheaths) depict different sampling locations.

Figure 4

The fungal community composition of garlic tissues and rhizosphere. Composition of different fungal communities at the phylum (A) and genus (B) level. The relative abundance of different fungal phylum or genus within the different communities is shown in different colors. Others: the merging of taxa with the relative abundance less than 0.01. The difference in the relative abundance of OTU1289 (C) and OTU1790 (D) based on Kruskal-Wallis H test (p < 0.05). Bars with different letters are significantly different (p < 0.05). CK (irrigation water), B (B. velezensis TB918), BS (B. velezensis TB918 and sucrose) and P (commercial biocontrol agent P. polymyxa) depict rhizosphere soil samples and different treatments. M (margin soil), G (garlic cloves), I (inner sheaths), and O (outer sheaths) depict different sampling locations.

Figure 5

Microbial clustering and distribution of garlic tissues and rhizosphere. UPGMA-Tree Clustering analysis on genus level of soil bacterial (A) and fungal (B) community based on the Unweighted Pair group Method with Arithmetic Mean (UPGMA) in garlic with different treatments. Samples were clustered into five groups: M: margin soil group, R: rhizosphere soil group, G: garlic cloves group, I: inner sheaths group, O: outer sheaths group. The Venn diagram of bacterial (C) and fungal (D) OTUs of garlic tissues and rhizosphere. CK (irrigation water), B (B. velezensis TB918), BS (B. velezensis TB918 and sucrose) and P (commercial biocontrol agent P. polymyxa) represent rhizosphere soil samples and different treatments. M (margin soil), G (garlic cloves), I (inner sheaths) and O (outer sheaths) represent different sampling locations. Flower plot showing numbers of specific OTUs found in each treatment (in the petals), and common OTUs for different treatments (in the center).

Figure 6

The co-occurrence network diagram of garlic rhizosphere inoculating biocontrol agents based on Spearman’s correlation analysis of the abundances for the 50 most abundant microbial genera. (A) Network diagram based on the bacterial genus level. (B) Network diagram based on fungal genus level. The red line indicates a positive correlation and the green line indicates a negative correlation. The size of the dots represents the average abundance of the genus. CK (irrigation water), B (B. velezensis TB918), BS (B. velezensis TB918 and sucrose), and P (commercial biocontrol agent P. polymyxa) represent rhizosphere soil samples with different treatments.

Figure 7

Influence of Bacillus on biofilm phenotype and colonization of isolated Pseudomonas. (A) Biofilm phenotype of the isolated Pseudomonas strains. GP1-3 represents different Pseudomonas strains. Well diameter is 16.5 mm. (B) Pellicle biomass quantified by dry weight. Pellicles were cultivated in MSgg medium for 24 h. TB918 represents B. velezensis TB918, GP1-3 represent different isolates. Gray and blue bars represent mono-culture of TB918 and GP1-3, respectively. Co-culture means mixed cultivation of TB918 and GP1-3 either. The bars represent the standard error. The letters above the columns indicate statistically significant differences based on the index using the Student’s t-test (p < 0.05). (C) The roots of 14-day-old garlic were colonized by B. subtilis 3610 (mKate2, red arrows), P. fluorescens GP2 (GFP, green arrows), and both. LSCM pictures are representative of at least 10 independent garlic roots. Scale bar represents 20 μm. (D) The difference in the garlic root colonization between Bacillus and Pseudomonas in mono- or co-inoculation condition was determined by counting colony forming unit (CFU) per mm root length. The bars represent the standard error. The letters above the columns indicate statistically significant differences based on the index using the Student’s t-test (p < 0.05).