UCLA

Mass spectrometry (MS), especially tandem mass spectrometry (MS/MS), is a powerful tool for proteomic and metabolomics applications. Untargeted metabolomics results can be well visualized and interpreted by using the cloud plot with XCMS Online software. The first objective of this study is to perform a comprehensive metabolomics analysis of oral cancer cells and identify metabolites altered by the knockdown of either adenylate kinase 2 (AK2) or phosphorylate glycerol kinase 1 (PGK1). UM1 and UM2 oral cancer cells were treated with siRNA to knockdown AK2 or PGK1. MS/MS and XCMS were performed to compare the metabolite profiles between the cells with siRNA knockdown and with scrambled siRNA control. Our studies confirmed the utility of XCMS to interpret the metabolomic results from oral cancer cells. When AK2 or PGK1 was knocked down in the UM1 or UM2 cells, more metabolites were found to be down-regulated than up-regulated. Heat map analysis indicates that a common group of metabolites were altered by AK2 knockdown between the UM1 and UM2 cells, and similar finding was observed for the PGK1 knockdown study.

Tracer-based metabolomics, a subset of metabolomics with a labeled substrate, is a new platform that would help researchers understand the metabolic phenotype of cancer cells. The second objective of this study is to develop the novel methodology which combines the tracer-based metabolomics, immunoprecipitation (IP), and MS-based proteomics to detect the metabolic labeling of a specific protein from the entire protein complex in oral cancer cells. [U-13C6]-glucose was introduced into the UM1 and UM2 cells, and the labeled proteins were analyzed by liquid chromatography (LC) with MS/MS. We found that UM1 and UM2 cells displayed different types of 13C labeled peptide mass isotopomer distribution patterns. Mass isotopomer distribution pattern decayed faster and the intensities of each isotopic peak were lower for the UM2 cells than those for the UM1 cells. We also demonstrated that a specific labeled protein, e.g., 78kDa glucose-regulated protein (GRP 78), can be pulled down with IP and analyzed by LC-MS/MS. Our results indicated that the UM1 cells utilize more glucose than the UM2 cells possibly to maintain their invasive and metastatic phenotypes. Also, the methodologies were able to identify any single 13C-labeled protein from the whole cell lysate if antibody is commercially available. Therefore, using XCMS and our newly developed tracer-based metabolomics, we may have an improved understanding of the metabolic phenotype of oral cancer cells.