Poster Presentation 23rd Annual Lorne Proteomics Symposium 2018

Spatial targeted metabolomics mapping in rat brain sections using DESI ion mobility mass spectrometry (#156)

Jose Castro-Perez 1 , Anthony Midey 1 , Hernando Olivos 1 , Emmanuelle Claude 1 , Bindesh Shrestha 1
  1. Waters Corp., Milford, MA, United States

Spatial mapping of small molecules, such as neurotransmitters, alongside lipids, can increase our understanding of biological functions of those molecules within the brain.  Mass spectrometry imaging (MSI) can be used to map distribution of molecules from any flat surfaces, including tissue sections. Desorption Electrospray Ionization (DESI) is an ambient ionization technique that can spatially profile the distribution of molecules in research tissue samples. DESI-MSI often provides complementary information to other techniques such as matrix-assisted laser desorption ionization (MALDI). Here we present the utility of DESI imaging to detect neurotransmitters, such as serotonin, adenosine, and glutamine directly in brain tissue samples. These results are for research use only and not for use in diagnostic procedures.

Rat brain was harvested and flash-frozen in liquid nitrogen before cryosectioning. Coronal tissue sections (8 microns thick) were mounted on regular glass microscope slides, vacuum dried, and analyzed without any further sample preparation. DESI-MSI data were collected and processed on a high definition mass spectrometer with ion mobility separation (SYNAPT HDMS G2-Si, QToF) using High Definition Imaging (HDI) 1.4 software with MassLynx 4.1 data acquisition control. DESI acquisitions were performed using methanol and water as a DESI spray solvent.

Small molecules such as amino acids (e.g., taurine, glutamine, arachidonic acid) and neurotransmitters (e.g., GABA, serotonin) were simultaneously detected along with lipids (e.g., phosphatidylcholine, lysophosphatidylcholine). Spatial correlation between detected metabolites and lipids were explored using analysis based on Pearson product-moment correlation coefficient and hierarchical clustering analysis. Molecular identification was aided using high mass accuracy (low PPM) database searches against publically curated databases, such as LipidMaps and Human Metabolome (HMDB). In addition to the accurate mass and high-fidelity isotopic distribution, collisional cross sections (CCS) or drift time data obtained during ion mobility separation was used to improve confidence in detected molecules.