Abstract
The field of secondary metabolism has greatly benefitted from computational advances in recent years. This has been particularly true for fungal natural product studies. Strides in genome mining have led to the identification of an extraordinary number of secondary metabolite biosynthetic gene clusters (BGCs) across the fungal Kingdom and metabologenomic platforms can group BGCs into gene cluster families and link them to initial chemical structures. Missing are computational applications focused on identifying BGC regulatory networks. Here, we applied the new online gene regulatory network resource, GRAsp (Gene Regulation of Aspergillus fumigatus), to identify unknown and unpredictable BGC trans-acting transcriptional/metabolite production modules. GRAsp correctly predicted a two-component regulatory module composed of the transcription factors (TFs), RogA (regulation of gliotoxin) and HsfA, which negatively regulate the gliotoxin BGC and are also involved in gliotoxin self-protection. RogA functions through the repression of gliZ, the pathway-specific gliotoxin TF, and HsfA functions by activating rogA expression. Furthermore, GRAsp identified TFs that regulate the production of two BGCs lacking pathway-specific TFs, the helvolic acid and fumitremorgin BGCs, respectively. Finally, the known TF, NsdD, was predicted and found to regulate the hexadehydroastechrome BGC. These advances highlight the power of inference algorithms to uncover unpredictable networks in specialized metabolite synthesis.IMPORTANCEToxic secondary metabolites are virulence factors of the opportunistic fungal pathogen Aspergillus fumigatus, yet the transcriptional networks regulating secondary metabolite production remain elusive. Uncovering novel regulators without any prior information is challenging. Computational programs have gained prominence in the field of secondary metabolite research due to their accuracy and ability to handle vast amounts of data, including DNA, RNA, and protein data. In this study, a newly developed online computer platform, Gene Regulation of A. fumigatus, was used to identify five regulators involved in the production of several A. fumigatus toxins, including gliotoxin, helvolic acid, fumitremorgin, and hexadehydroastechrome. This work illustrates the potential for discovering new trans-acting regulators and mechanisms of secondary metabolite regulation through the examination of computational gene regulatory networks.