Development of robust DESI imaging MS instrumentation for analysis of tissue samples

Abstract

Desorption electrospray ionisation (DESI) is an ambient ionisation method that can be used for mass spectrometric imaging. Due to its ability to ionise lipids, its non-destructive nature and its minimal sample preparation, it is particularly suitable for biological tissue imaging and it can be combined directly with classical histopathological staining methods. However, despite having been one of the earliest ambient ionisation techniques, first published more than ten years ago, DESI still suffers from repeatability and reproducibility issues. The aim of this project was to identify and eliminate the primary sources of variability in DESI. One major source of variability was found to be solvent capillary positioning in the DESI sprayer. The ideal positioning of the capillary was hypothesised to be perfect centering, although this could not be achieved in practice. However, a fixed capillary position as close to the central position as possible was successfully implemented. This eliminated movement and vibration of the capillary, improving repeatability. By using a tapered, machine-cut, small inner diameter capillary the operational parameters could be optimised for improved spatial resolution. The improved sprayer was combined with a fast-scanning QToF MS for fast, high spatial resolution DESI-MS imaging. Tests on rat brain sections showed that DESI was able to distinguish between different tissue types even with a more than tenfold increase of scan speed. The improved DESI source was also used to analyse mouse and human brain tissue sections as part of a larger study on remyelination in multiple sclerosis. It was shown that DESI can be combined with Raman spectroscopy to provide complementary imaging information, although the two methods could not be performed on the same sections. An alternative, closely related ambient ionisation method, desorption electro-flow focusing ionisation (DEFFI) was tested for tissue imaging performance and repeatability. DEFFI uses a co-flowing gas stream to focus a charged solvent into a jet, making the primary spray inherently concentric. Repeatability was similar to a carefully optimised DESI sprayer and after adjustment of operational parameters, its imaging resolution was comparable. The comparison of DESI and DEFFI data suggested that the data was sufficiently similar to allow integration of DEFFI into existing DESI workflows. Finally, the impact of MS inlet capillary dimensions and heating was investigated. These experiments suggested that ion production in DESI partially occurs in the inlet capillary. A small capillary inner diameter was found to be crucial for dissociation of ion clusters. Capillary heating was shown to improve overall sensitivity and also to make DESI less sensitive to geometrical changes. This supports the hypothesis that some desolvation and ion formation occurs during droplet transfer into the MS. Overall, the work presented here brings DESIMS imaging closer to becoming a routine tool in clinical diagnostics.