The analysis and imaging of lipids from complex samples by matrix-assisted laser desorption/ionisation mass spectrometry

摘要

Matrix assisted laser desorption/ionisation mass spectrometry (MALDI-MS) is a relatively new technique for imaging lipids directly in tissue, with useful applications in disease state profiling, bacterial typing and forensics. This thesis describes optimisation and evaluation of sample preparation procedures for MALDI-MS and MS imaging of lipids in complex samples.ududDirect analysis of fresh samples is shown to result in higher ion counts than analysis of desiccated tissues but thickness of tissue sections was shown to have a minimal effect on imaging results. Extracts of lipids from rat brain samples were used to optimise matrix selection for MALDI-MS of lipids. DHB was found to be the optimum matrix, in accordance with literature, for extract analysis; however, superior imaging results were obtained usingudαCHCA. An automated matrix deposition robot is compared to a manual airspray method. The robotic method demonstrated enhanced sensitivity, superior image resolution andudreduced variability between different sample plates compared to traditional manual methods. Optimised methods were applied to analysis of formalin fixed tissue. Successful imaging of phospholipids in fixed samples is demonstrated for the first time. Lipids in formalin fixed samples were found to be predominantly detected as sodium adducts (due to high levels of sodium in the buffered formalin). This was exploited to offer enhanced structural information afforded by collision induced dissociation (CID) of sodium rather than potassium adducts, which is well reported to increase the number of useful productudions detected.ududMethods for analysis and imaging of non-mammalian lipids by MALDI-MS were considered. Members of the Mycobacterium tuberculosis complex (MTBC) are important human pathogens. The cell wall of mycobacteria contains a number of lipids, which contribute largely to the virulence of pathogenic species. Preliminary research showed relatively poor sensitivity for these complex lipids but improved results were obtained by direct analysis of TLC plates. A binary matrix solvent system was developed offering considerably improved sensitivity. Successful detection of numerous lipids species involved in virulence, along with several previously unreported molecules is presented.