Applying molecular networking to natural products chemistry

Abstract

Advancements in analytical, chemical and spectroscopic technologies, and rapid and robust dereplication and prioritization methodology, have greatly enabled the discovery of new natural products (NPs) from complex natural extracts. This thesis reviews these advances and seeks to apply best practice to explore a library of ~1,000 southern Australian and Antarctic marine sponges, algae and tunicates. In particular, this thesis explores the application of advanced mass spectrometry (UPLC-QTOF-MS/MS) and Global Natural Products Social Molecular Networking (GNPS) guided prioritization. Selected prioritized extracts were then subjected to detailed chemical investigation using a sequence of efficient fractionation techniques, including solvent extraction, partitioning and trituration, followed by analytical/semi-preparative and preparative reversed phase HPLC, and size exclusion Gel chromatography. Structure elucidation of purified metabolites was achieved by detailed spectroscopic analysis, acquiring, and interpreting UV-Vis, [α]D, HRMS, and 1D and 2D NMR data, and by searching online database, existing literature, and where possible comparison with authentic standards. In addition to expanding knowledge of marine natural products, this study demonstrates the value of GNPS molecular networking to map and prioritize chemical diversity, and detailed spectroscopic analysis to solve complex structures. Chapter one highlights the advances in natural product discovery technology, and growing trend toward dereplication and prioritization. Recent advances are put in context by a brief discussion on the evolution of dereplication and prioritization strategies over the last 50+ years. Chapter two documents UPLC-QTOF-GNPS based prioritization of potential marine sponge extracts containing new and/or novel compounds from Capon Group marine extracts library. It outlines preparation of new n-BuOH soluble extracts for 960 marine specimens and MS/MS data acquisition using UPLC-QTOF. It also describes a GNPS molecular networking analysis of 960 n-BuOH soluble extracts together with 95 authentic standards of marine metabolites to prioritize promising extracts for detailed chemical investigation.Chapter three focuses on the isolation and characterization of a new class of sesterterpene butenolides from a southern Australian marine sponge, Cacospongia sp. (CMB-03404). Preliminary HPLC-DAD-MS and 1H NMR analysis of CMB-03404 extract, suggested the presence of sesterterpene tetronic acids. However, GNPS analysis revealed that “sesterterpenes” from CMB-03404 failed to cluster with known sesterterpene tetronic acids. Subsequent fractionation and structure elucidation, using detailed 1D and 2D NMR spectroscopic analysis, enabled the isolation and characterization of fifteen new sesterterpene butenolides, cacolides A-L and cacolic acids A-C – with unprecedented “non-tetronic acid” moieties. To emphasize this point we provide a comparative GNPS analysis of 5 × Psammociniaspp., 5 × Sarcotragus spp. and 5 × Ircinia spp. known to produce sesterterpene tetronic acids, with the cacolides found in Cacospongia sp. CMB-03404.Chapter four describes a GNPS molecular networking guided approach to detect, isolate and identify new analogues of trachycladindoles, a class of exceptionally rare indole-2-carboxylic acid aminoimidazoles. Comparative GNPS analysis of 960 marine extracts with authentic standards of trachycladindoles A-G, identified Geodia sp. (CMB-01063) as a source of new trachycladindole analogues. Subsequent extraction, fractionation and spectroscopic analysis identified six new analogues, trachycladindoles H-M. Chapter five reports a GNPS based prioritization and isolation of a new class of sesquiterpenes from Dysidea sp. (CMB-01171). Although sponges of the genus Dysidea are well known to produce interesting new natural products, prior to our study the number of published accounts encouraged the view that the chemistry of Dysidea had been exhaustively studied, and any future investment would be unlikely to yield significant new discoveries. Our GNPS guided approach disproved this view and led to the discovery of ten unprecedented sesquiterpenes, including dysidealactams A–F and dysidealactones A and B, featuring both unprecedented and very rare functional groups. Chapter six documents the GNPS guided discovery of a novel alkaloid, thorectandrin A, from a southern Australian marine sponge, Thorectandra choanoides (CMB-01889). GNPS cluster analysis along with HR-MS data provided the molecular formula (MF) of many biosynthetically related minor co-metabolites, and demonstrated the power of GNPS to interrogate the halo of co-metabolites present at concentrations far too low for traditional approaches.