Abstract
Biochemometric approaches, which integrate bioactivity data with spectroscopic or spectrometric data, offer significant potential to streamline the discovery of bioactive compounds in targeted isolation strategies. However, the complexity of natural extracts and the presence of structurally similar analogs make this process time-consuming and resource intensive. This study introduces a 2D nuclear magnetic resonance (NMR)-based heterocovariance analysis (HetCA) workflow to identify chemical features that correlate positively or negatively with bioactivity in complex mixtures. As a proof-of-concept, the workflow was established using artificially mixed samples of pentacyclic triterpenes which were screened for modulatory activities of the retinoic acid receptor-related orphan receptor gamma (RORγ) and the G protein-coupled bile acid receptor (TGR5). The validated concept was then exemplified using a triterpene-rich Eriobotrya japonica leaf extract. The applied workflow enabled the targeted and accurate identification of bioactive constituents from E. japonica that modulate RORγ and/or TGR5 using this newly developed biochemometric 2D NMR HetCA approach.