Development of an in-House Software Platform and its Application in Untargeted Ultraperformance Liquid Chromatography Coupled to Time-of-Flight Mass Spectrometry Analysis in Metabolomics and Traditional Chinese Medicine Studies

The recent development in high-resolution mass spectrometry (MS) has afforded powerful analytical equipments, e.g., ultra performance liquid chromatography coupled to time-offlight mass spectrometry (UPLC TOFMS), for rapid qualitative analysis of samples as complex mixtures in biomedical research. This is especially beneficial for the research fields as untargeted metabolomics and pharmaceutical analysis of natural product, e.g., the traditional Chinese medicine. However, a significant bottleneck in deriving biological knowledge from the studies is the process of identification of the samples’ components, which is a labor-intensive step that follows data acquisition and analysis and must occur before biological interpretation is possible [1]. To address this bottleneck, several computer software platforms have been developed in recent years for automated data processing [1,2]. In our laboratory, we developed our own software platform in 2012, the operation of which is very simple [3]. Unlike the reported software platforms that address raw LC-MS data and require the Taverna or R environment, our software platform deals only with the dataset of LC-MS features produced after the pre-processing step and requires nothing more than a Microsoft Windows operating system and Microsoft Excel software. One component of our platform is a software named “Searcher”, which works as an Excel “add-in” component and can make automatic putative batch identifications by matching the measured m/z data list with a reference m/z list. Another component is a series of in-house, localized, Excel-format databases that contains accurate theoretical m/z values for multiple types of ions commonly observed in the electrospray ionization (ESI). Currently the series includes a HMDB (Human Metabolome Database, http://www. hmdb.ca/[4-6])-derived database (HMDBDDB), a database of observed metabolite signals in UPLC TOFMS analysis of serum of Sprague-Dawley (SDRSDB), a database of observed metabolite signals in UPLC TOFMS analysis of urine of Sprague-Dawley (SDRUDB) and a literature-derived chemical database of the plants of Rehmannia, Angelica, Paeonia and Ligusticum (RAPLDB). The SDRSDB and SDRUDB can be used for distinguishing the signals of endogenous metabolites from exogenous metabolites in serum or urine samples of SD rats. The RAPLDB can be used to detect the signals of reported phytochemicals of the four medicinal plant families. With these two components (i.e., the Searcher software and localized database series), rapid automatic putative batch identification of high-resolution LC-MS data can be achieved with very simple operations, and the obstacle with data analysis in our laboratory was successfully overcome. Below are examples of some research works in our lab using this platform.