High-throughput profiling of microbial extracts.

Microbial culture extracts are used for natural products screening in the drug lead discovery process. An extract of a microbial culture is a complex mixture of organic compounds, making it difficult to evaluate the diversity and redundancy of the compounds. However, having a diverse extract library is a key to success in generating a lead for drug discovery. We have developed a high-throughput and robust LC-MS analysis and data processing method for visualizing sample profile and diversity. In the LC-MS analysis of 16 025 microbial culture extracts, positive and negative ions were acquired simultaneously with an electrospray ionization source. The raw data were processed using ACD IntelliXtract, and peak lists of each extract were generated and stored in the database. The peak list data were binned by nominal mass (m/z 150-1350) and retention time. How frequently the binned peaks, termed peak identifiers, appeared in the extracts library was calculated, and these data were visualized by scattered plots in Spotfire. From 4 to 115 peaks were observed in each sample. As it is easy to eliminate the ubiquitous peaks shown in most samples and simplify the plots, the unique or redundant compounds could be detected.