Developing Next Generation Technologies for Spatially Targeted Proteomics

Abstract

Matrix-assisted laser desorption/ionization (MALDI) imaging mass spectrometry (IMS) is a technology that allows for the desorption of ions into the gas phase, generating 2 dimensional molecular maps of a tissue surface, plotting both the ion location and relative intensity. The ability to identify proteins and map their distributions in a tissue sample is a powerful way to determine and track proteomic changes associated with health and disease. The molecular identification of proteins of interest is an important part of an IMS experiment but can be a challenge from tissue due to limitations in manipulating large, MALDI generated protein ions in the mass spectrometer. Due to these challenges, proteomic workflows for MALDI IMS can incorporate both high mass accuracy platforms and secondary, offline experiments utilizing liquid chromatography tandem mass spectrometry (LC-MS/MS) to generate protein identifications that can be correlated to the IMS data through accurate mass matching. Recently, the high mass resolution, accurate mass capabilities of Fourier transform ion cyclotron resonance mass spectrometry (FT-ICR MS) have been shown to facilitate the identification of proteins in matrix-assisted laser desorption/ionization (MALDI) IMS. However, these experiments are typically limited to proteins of relatively low m/z due to difficulties transmitting, measuring, and fragmenting large molecular weight ions of low charge states (< 3). To address transmission challenges of large protein ions generated using MALDI, we modified the source region of a commercial 15T FT-ICR to regulate and optimize gas flow through the ion funnel of the mass spectrometer, effectively doubling the transmittable mass range allowing for the detection and isotopic resolution of proteins up to m/z 24,000. Although useful, accurate mass measurements do not provide enough confidence alone to identify a protein from tissue. Limitations in efficient gas phase fragmentation of large protein ions generated by MALDI has led to the development of a new offline spatially targeted protein identification strategy known as liquid extraction surface analysis (LESA). This robotic manipulation of small volumes of liquid (0.5-2.0 μL) across the surface of a tissue enables the identification of upwards of 1400 proteins which can be associated to discrete foci on the tissue surface. Limitations in the effective droplet diameter of a LESA extraction on tissue has been overcome by the method’s incorporation of autofluorescence microscopy to identify target foci and robotic deposition of liquid micro-enzymatic digestions in a spatially targeted manner. Currently, tissue foci as small as 110 μm can be targeted for tryptic digestion and subjected to proteomic analyses. Through a synergistic combination of these analytical technologies and methods, the work described herein enables visualization and identification of a greater fraction of the proteome from tissue and ultimately a deeper understanding of pathologically relevant processes.