Evaluation of plant sources for antiinfective lead compound discovery by correlating phylogenetic, spatial, and bioactivity data.

Antibiotic resistance and viral diseases are rising around the world and are becoming major threats to global health, food security, and development. One measure that has been suggested to mitigate this crisis is the development of new antibiotics. Here, we provide a comprehensive evaluation of the phylogenetic and biogeographic patterns of antiinfective compounds from seed plants in one of the most species-rich regions on Earth and identify clades with naturally occurring substances potentially suitable for the development of new pharmaceutical compounds. Specifically, we combine taxonomic and phylogenetic data for >7,500 seed plant species from the flora of Java with >16,500 secondary metabolites and 6,255 georeferenced occurrence records to 1) identify clades in the phylogeny that are characterized by either an overrepresentation ("hot clades") or an underrepresentation ("cold clades") of antiinfective compounds and 2) assess the spatial patterns of plants with antiinfective compounds relative to total plant diversity across the region. Across the flora of Java, we identify 26 "hot clades" with plant species providing a high probability of finding antibiotic constituents. In addition, 24 "cold clades" constitute lineages with low numbers of reported activities but which have the potential to yield novel compounds. Spatial patterns of plant species and metabolite diversity are strongly correlated across Java, indicating that regions of highest species diversity afford the highest potential to discover novel natural products. Our results indicate that the combination of phylogenetic, spatial, and phytochemical information is a useful tool to guide the selection of taxa for efforts aimed at lead compound discovery.