Abstract
The emergence of multidrug-resistant fungi Candida auris is a worldwide health crisis connected with high rates of mortality. There is a critical need to find novel and unique antifungal compounds for treating infections of multidrug-resistant fungi such as C. auris. This study aimed to illustrate that biosynthetic gene clusters in native bacterial isolates are able to produce antifungal compounds against the multidrug-resistant fungus C. auris. It was successfully achieved using large-scale antifungal activity screening, cytotoxicity analysis, and whole genome sequencing integrated with genome mining-guided analysis and liquid chromatography-mass spectrometry (LC/MS). A list of possible gene candidates was initially identified with genome mining methods to predict secondary metabolite gene clusters of antifungal-compound-producing bacteria. Then, gene clusters present in the antifungal-compound-producing bacteria were identified and aligned with the reference genome using comparative genomic approaches. Bacillus halotolerans AQ11M9 was identified through large-scale antifungal activity screening as a natural compound-producer against multidrug-resistant C. auris, while it was nontoxic to normal human skin fibroblast cells (confirmed using a cell viability assay). The genome (4,197,347 bp) of B. halotolerans AQ11M9 with 2931 predicted genes was first mined for detecting and characterizing biosynthetic gene clusters, which revealed 10 candidate regions with antifungal activity. Clusters of AQ11M9 encoded non-ribosomal peptide synthase (NRPS) (bacilysin, bacillibactin, paenibactin, surfactin, plipastin, and fengycin) and polyketide (macrobrevin). The presence of gene clusters with anti-C. auris activity, and surfactin identified through LC/MS, from AQ11M9 suggests the potential of utilizing it as a source for a novel and powerful anti-C. auris compound.