In nature, microorganisms are inhabiting nearly all ecological niches on Earth, where they live in complex communities and interact with other organisms and their environment. These intra- and interspecies or even interkingdom interactions can be beneficial, neutral or antagonistic for each of the partners. It is believed that a variety of these interactions are mediated by secondary metabolites. These small molecules are unlike primary metabolites not directly involved in growth and proliferation of the producing organism but might fulfil tasks, which improve their fitness, such as communication, nutrient acquisition or competitive inhibition, in the given environment. Under laboratory conditions, it has been shown that secondary metabolites greatly impact microbial interactions. Noteworthy, humans have already benefitted from several applications of them since their discovery. The first described microbial secondary metabolites, such as antibiotics, have revolutionised medicine and disease treatments and consequently tremendously increased life expectancy of both humans and farm animals. Additionally, the use of secondary metabolites-producing bacteria as biocontrol agents or probiotics have improved plant, livestock as well as human health. However, besides their primarily in vitro observed antimicrobial effects, our understanding of the true ecological role of these molecules is still limited.

Members of the genus Bacillus have beneficial properties by promoting plant growth and improving plant health by antagonising plant pathogenic fungi and bacteria. This Ph.D. project investigated in six separate publications the influence of secondary metabolites from Bacillus species, especially Bacillus subtilis, on plant pathogenic fungi, bacterial communities and cell differentiation.

Study 1 revealed that the antifungal potential of 23 B. subtilis strains, isolated from 11 different environments, varied among co-isolated strains due to differences in the production of nonribosomal peptides. A mutant-based screening highlighted that the biocontrol properties of B. subtilis depend on both the tested bacterial strain and the targeted fungal plant pathogen. We demonstrated that the nonribosomal peptide plipastatin is sufficient to inhibit Fusarium species (spp.), whereas, a combination of plipastatin and surfactin was essential in the majority of B. subtilis strains for a strong inhibition of Botrytis cinerea. Genomic data of selected B. subtilis strains, obtained from study 2, allowed us to identify either missing core genes, a nonsense mutation, or potentially altered gene regulation as the cause for interrupted or altered nonribosomal peptide production. Furthermore, we could demonstrate that these B. subtilis isolates harboured core biosynthetic gene clusters present in nearly all isolates and distinct accessory biosynthetic gene clusters present in only some of them.

Study 3 highlighted that nonribosomal peptides of B. subtilis had only minor effects on soil-derived semi-synthetic bacterial mock communities but influenced the frequency of the closely related genera Lysinibacillus and Viridibacillus. Moreover, we could determine the susceptibility of Lysinibacillus fusiformis M5 towards surfactin.

Study 4 exhibited that the excellent antifungal properties of Bacillus velezensis DTU001, depend primarily on the produced nonribosomal peptide iturin, but a crude extract containing at least iturins, fengycins and surfactins was most bioactive against the targeted human and plant pathogenic fungi.

Study 5 demonstrated that in the laboratory strain B. subtilis NCIB 3610, genes encoding for matrix components and that are necessary for biofilm formation and plant root colonisation are as well crucial for colonisation of fungal hyphae of the Ascomycota Aspergillus niger and the Basidiomycota Agaricus bisporus.

Study 6 revealed that the presence of surfactin is not necessary for biofilm formation and plant root colonisation in B. subtilis NCBI 3610 and various B. subtilis environmental
strains, as claimed in previous studies.

In conclusion, this Ph.D. study highlights the coexistence of B. subtilis nonribosomal peptide producer and non-producer strains in ecological niches and the associated diversity in antifungal potential. Additionally, it reveals the impact of nonribosomal peptide mixtures on both bacterial communities and pathogenic fungi. Furthermore, surfactin production and biofilm formation are both essential for biocontrol properties but should be considered as two independent processes in B. subtilis.