Development and validation of a systematic platform for broad-scale profiling of microbial metabolites

Liquid chromatography-mass spectrometry based profiling of microbial metabolites has been a challenging task due to their diverse physicochemical properties and wide concentration ranges. This study is aimed to develop a systematic platform for the broad-scale profiling of microbial metabolites by integrating aqueous-lipophilic biphasic extractions and chemical derivatizations with a data-dependent automatable metabolite annotation algorithm. This complementary strategy of detection will not only largely expand the metabolite coverage, but also facilitate the drawing out of interested submetabolome using designed chemical derivatizations. Then, the data-dependent metabolite annotation algorithm is able to automatically match the raw MS/MS data with those of compounds in the self-collected databases. The performance of this platform is illustrated through the analysis of two representative bacteria (Escherichia coli and Pseudomonas aeruginosa) and intestinal contents samples from experimental colitis mice. As a result, 292 metabolites corresponding to 875 annotated features distributing over 25 chemical families were putatively annotated in a short time. Of these metabolites, 197 and 218 are respectively from the bacteria and intestinal contents, and 107 are identified in all three biological samples. This systematic platform could be used to accomplete high-coverage detection and high-quality data processing of microbial metabolites. At the same time, chemical derivatization design and the establishment of self-collected databases will facilitate self-driven untargeted analysis.