Discovery of fernane-type triterpenoids from Diaporthe discoidispora using genome mining and HSQC-based SMART technology

Endophytic fungi have garnered increasing attention due to the isolation of fascinating compounds with significant therapeutic value from these largely unexplored microbes. Fernane-type triterpenoids, a rare class of triterpenoids, have demonstrated a variety of biological activities. In this study, we employed a combination of genome mining and heteronuclear single quantum coherence (HSQC)-based small molecule accurate recognition technology (SMART) technology to search for fernane-type triterpenoids. Initially, potential endophytic fungi were identified through genome mining. Subsequently, fine fractions containing various fernane-type triterpenoids were selected using HSQC data collection and SMART prediction. These triterpenoids were then obtained through targeted isolation and identification. Finally, their antifungal activity was evaluated. As a result, one potential strain, Diaporthe discoidispora, was selected through genome mining. Fraction 3.4 was further isolated using various chromatographic methods, yielding three fernane-type triterpenoids, including two novel compounds, along with two new sesquiterpenes and four known compounds. Their structures were elucidated through analysis of high-resolution electrospray ionization mass spectrometry (HR-ESI-MS) and nuclear magnetic resonance (NMR) spectroscopic data. The absolute configurations were determined using single-crystal X-ray diffraction analysis and electron capture detector (ECD) analysis. The antifungal activities of the isolated compounds were evaluated, with compound 3 exhibiting moderate antifungal activity against Candida albicans CMCC 98001 and Aspergillus niger.