A statistical analysis-based method for finding bioactive natural products

<p>Natural products with biological activities are promising drug seeds. Functional foods and herbal medicines are useful screening sources for natural products. Aggregation of amyloid β protein (Aβ42) induced neurotoxicity in the pathogenesis of Alzheimer's disease (AD). Therefore, reagents that inhibit the aggregation of Aβ42 are useful to develop anti-AD drugs. Fruitful anti-aggregative natural products have been ever reported by many research groups including us. In general, it is necessary to repeat the purification of the active components from extracts. We have developed an efficient method, using LC-MS combined with principal component analysis (PCA), to search for activity-dependent compounds that prevent the aggregation of Aβ42 from 46 herbal medicine extracts originating from 18 plants (380 extracts in total). Only lotus-derived 5 extracts (Kakou, Kayou, Gusetsu, Rensu, and Renbou) showed differentially inhibitory activities depending on the part of the plant from which they are derived (e.g. petiole, leaf, root node, stamen, and receptacle, respectively). To compare the anti-aggregative properties of compounds of active crude drugs with those of inactive crude drugs, these extracts were subjected to LC-MS, followed by PCA. From 22 candidate compound lists identified from the analysis, glucuronized and glucosidized quercetin, as well as 6 flavonoids (datiscetin, kaempferol, morin, robinetin, quercetin, and myricitrin), including catechol and/or flatness moiety suppressed Aβ42 aggregation, whereas curcumol, a sesquiterpene, did not. These methods offer an activity-differential methodology to identify bioactive compounds with anti-aggregation activity. This review focuses on this cutting-edge knowledge on the exploration of natural products that could be applied to future disease therapeutics and prevention.</p>