Ruminant Nutrition Symposium: Metabolomics Applications in Dairy Cow Metabolism

An overview is given on mass spectrometry-based metabolomics: past, current, and future developments. Understanding the metabolome is challenging due to the enormous diversity of metabolite structures. Metabolomics uses a range of tools, established as service at the NIH West Coast Metabolomics Center at UC Davis. We here explore the use of ion mobility spectrometry (IMS) in combination with liquid chromatography (LC) and high resolution mass spectrometry (MS) for milk analysis by increasing total peak capacity, specifically to distinguish isobaric and isomeric species that are not resolved by LC-MS alone. As example, we used LC-IM-QTOF MS for unraveling the complex lipidome compositions in human, goat and cow milk. Measurements were performed on an Agilent 1290 Infinity II LC with 6560 drift tube-ion mobility-QTOF MS system using a Waters Acquity CSH C18 column with a 15 min-gradient of isopropanol/acetonitrile (90:10, B) varied from 15% to 99% and acetonitrile:water (60:40, A), ammonium formate/ formic acid buffered. Twenty μL of milk samples were extracted with MTBE and methanol. Dried extracts were reconstituted in methanol before LC-IM-MS analysis. Measured ion mobility drift times were corrected with external calibration for time spent outside the drift tube for precise determination of collision cross sections. We optimized the control over ion trapping in the drift cell. We analyzed lipid extract from 3 types of milk samples with untargeted LC-IM-MS. IM mode provided 6x more molecular features compared with regular LC-MS analysis without ion mobility. Using Agilent Mass Profiler to report IM peak feature, we developed a specific command line tool in R to characterize co-eluting isobaric species, followed by downstream identification by MS-DIAL. Our R command screened the IM features that exhibited distinct IM drift time (e.g., ?DT > 0.4 msec) by approximate m/z and LC retention time. We reduced the false discovery rate by evaluating DT intensities and m/z-specific DT after fitting a power-law regression. We discovered 73 isobaric species in milk lipidomic extracts that required ion mobility separation in addition to LC-QTOF MS.