Project Summary/Abstract
Natural products (NP) are a mainstay of drug discovery, accounting for up to 50% of approved drugs either
as direct natural molecules or as inspiration for synthetic molecules. High-throughput screening of compound
libraries is a common starting point for drug development campaigns. The quality of these libraries is
therefore critical to high-throughput screening campaign success. NP compound library design is
particularly challenging given redundancies in NP production between isolates and greater costs of compound
production and isolation. Evidence-based and scientifically rigorous methods to optimize NP library
design are therefore urgently required. MPIs previously demonstrated using the fungus Alternaria that liquid
chromatography-tandem mass spectrometry (LC-MS/MS)-based analysis of fungal extracts can reveal the
minimal number of extracts to include in a chemical library, to achieve saturation of chemical diversity.
Strikingly, this could be as few as 39 isolate extracts, depending on Alternaria clades. The overall objective of
this proposal is to demonstrate the broader utility of this bioanalytical approach, to the significant biological
problem of high-throughput screening NP chemical library design. The central hypothesis of this proposal is
that untargeted metabolomics will enable rational NP library generation and provide quantitative evidence as to
whether current fungal culture and library generation approaches improve library diversity or lead to
unnecessary redundancy. This proposal builds on MPI’s extensive expertise in metabolomics, NP analysis and
drug development. In addition, it is enabled by the MPI’s access to the large collection of fungal isolates from
the University of Oklahoma Citizen Science Soil Collection Program. This collection currently totals >88,000
isolates from 893 fungal genera. We will focus on three common library design approaches, in three
independent aims. Aim 1 will assess whether comparable chemical diversity and improved bioassay hit rates
can be obtained from focused, rationally-designed NP libraries, compared to current large-scale library build
approaches (historical sampling, cosmopolitan collections, morphological diversity, deep local collections, etc.).
Aim 2 will systematically assess the impact of co-culture on elicited chemical diversity and hit rates, comparing
sympatric vs allopatric co-culture systems. Aim 3 will systematically quantify the impact of environment-
mimicking culture conditions such as soil or bacterial-derived signals, on elicited chemical diversity and hit
rates. The proposed research is innovative because it applies a bioanalytical approach to NP library design
and because it challenges existing dogmas in this field. The proposed research is significant because it will
lead to an improved understanding of the determinants of NP chemical diversity and enable improved NP
library design. Overall, our results will lead to validation of a new approach for rational NP library
design, with major implications for drug development.