High sensitive and accurate qualification and quantitation of intracellular metabolites is the core content of metabolomics research. This dissertation presents herein a highly selective and sensitive ultra-high pressure liquid chromatography-mass spectrometry (UPLC–MS) method for the reliable measurement of 55 water-soluble intracellular metabolites, including 20 amino acids, 16 non amino organic acids mainly from TCA (tricarboxylic acid) cycle, 9 cofactors and 10 sugars. The obtaining results are as follows:

       Firstly, the chromatographic conditions of RPLC (reversed-phase liquid chromatography), HILIC (hydrophilic interaction liquid chromatography) and ion exchange modes were investigated and five columns were chosen for the optimization. Satisfactory retention and peak shape, no excessive peak tailing, increased peak capacity and baseline separation of isobaric compounds were considered as optimization objectives. The results indicated that ACQUITY UPLC^® HSS T3 column operated under RPLC mode generated the best chromatographic resolution and peak shapes for 20 amino acids, while better chromatographic resolution and peak shapes for non amino organic acids, cofactors and sugars were obtained on the ACQUITY UPLC^® BEH Amide column operated in HILIC mode. Nine cofactors were separated under neutral conditions, which avoid the disruption of unstable compounds considering some of them were unstable under alkaline or acidic conditions. At the same time, MRM (multiple reaction monitoring) parameters of each metabolite were optimized. The final stable LC-MS/MS (liquid chromatography-tandem mass spectrometry) method for qualitative separation is constructed based on the optimization of the chromatographic conditions and MRM parameters.

       Finally, this UPLC-MS/MS (ultra-high pressure liquid chromatography-tandem mass spectrometry) method was applied to the quantitative analysis of intracellular metabolites extracted from wild-type and ethanol-adapted strains of Clostridium thermocellum cultivated with/without ethanol stress, and the results showed 34 different metabolites could be well-separated and quantitated. Further multivariate data analyses (PCA (principal component analysis), PLS-DA (partial least squares discriminant analysis)) were carried out using the quantitative data, the results showed significant metabolic differences between wild-type and ethanol-adapted strains cultured under three different conditions, which would provide in-depth information for illuminating the metabolic mechanism behind ethanol adaptation in C. thermocellum.