Poster Presentation 24th Annual Lorne Proteomics Symposium 2019

Optimised Desorption Electrospray Ionisation Mass Spectrometry Imaging (DESI-MSI) method for the analysis of proteins/peptides directly from tissue sections (#114)

Emmanuelle Claude 1 , Mark Towers 1 , James Hughes 2 , Patricia Lalor 2 , Helen Cooper 2 , David Heywood 1
  1. Waters Corporation, Wilmslow, United Kingdom
  2. School of Biosciences , University of Birmingham, Birmingham, United Kingdom

DESI-MSI is typically known for the mapping of small molecules such as lipids, directly from tissue sections. Whilst the detection of spotted protein standards from target plates has previously been shown, the detection of large biomolecules directly from tissue sections has presented difficulty. Here we describe a newly developed method combining a series of optimized parameters and conditions which allow the extraction of large biomolecules from the tissue in droplet form. Detection of these multiply charged ions is enhanced using ion mobility which enables separation from the intense endogenous ion species and chemical background aiding in visualisation.
All experiments were carried out on a SYNAPT HDMS G2-Si Q-ToF (Waters) with a Prosolia2D DESI stage. A modified spray head assembly and an inlet capillary with a heated sheath connected to an adjustable power supply were used; voltages between 0 and 12V were applied to generate a variable heating effect.
To remove lipids and salts, the tissues were subjected to a series of ethanol based washes followed by chloroform. The heated inlet capillary was particularly beneficial giving an enhanced ionisation of proteins and peptides from the tissue with further optimisation in terms of sprayer inlet geometry, solvent composition, nebulising gas pressure and solvent flow rate. The orthogonal separation of ion mobility was found to be essential to observe the most abundant charge series in the highly complex data set. The most abundant charge series observed were those belonging to Haemoglobin with charge states from 9+ to 21+ observed. Extraction of the trendlines via the Driftscope software and subsequent use of MaxEnt 1 algorithm allowed a putative assignment of four Haemoglobin subunits from rat liver tissue. Additional charge series putatively assigned to FABP , HSP10 and COX8A for the human liver tissue section and MBP for the brain tissue section were observed. Overall further trend lines were found relating to background solvent peaks, residual lipids and more importantly lower abundant small proteins / endogenous peptides with charge states from +2 to +8.