This thesis presents efforts in the advancement and application of high-spatial resolution matrix-assisted laser desorption ionization-mass spectrometry imaging (MALDI-MSI) for the mapping of small metabolites and lipids at the cellular and sub-cellular level. The following work presents a number of advances, using both 2- and 3-dimensional MALDI-MSI to enable visualization at the sub-cellular level. The first chapter consists of a general introduction to the technique of MALDI-MSI, and the seventh and final chapter provides a brief summary of the presented work and possible future directions.

The second chapter presents a technology development for the optimization and application of matrix recrystallization to improve lipid ion signals in maize embryos and leaves. Using the optimized recrystallization conditions, the ion signals were improved three times, enhancing the image quality of lipid species with no apparent changes in their localization. Additionally, when methanol was used as a recrystallization solvent, unexpected side reactions were observed between phosphatidic acid and methanol vapor, suggesting recrystallization solvent should be carefully selected to avoid side reactions.

The third chapter presents an application using 5- and 10-um high spatial resolution MALDI-MSI to explore quantitative fatty acyl distributions of two classes of thylakoid membrane lipids along the developmental gradient of maize leaves in two inbred lines, B73 and Mo17, and the reciprocal hybrid lines, B73xMo17 and Mo17xB73. This study demonstrated that high-resolution MALDI-MSI analysis can be directly applied to multicellular plant tissues to uncover cell-specific metabolic biology that has not been possible using traditional metabolomics methodologies. For example, certain thylakoid membrane lipids (e.g. phosphatidylglycerol (PG) 32:0) show genotype-specific differences in cellular distributions. Inbred B73 shows preferential localization of PG 32:0 in bundle sheath cells, while a more uniform distribution between bundle sheath and mesophyll cells in inbred Mo17.

The fourth chapter present the first time MALDI-MSI has been applied for three dimensional chemical imaging of a single cell using newly fertilized individual zebrafish embryos as a model system. High-spatial resolution MALDI-MSI was used to map and visualize the three-dimensional spatial distribution of phospholipid classes, phosphatidylcholines (PC), phosphatidylethanolamines (PE), and phosphatidylinositols (PI), in the zebrafish embryo. The 3D MALDI-MSI volumetric reconstructions were then used to compare four different normalization approaches to find reliable relative quantification in 2D- and 3D- MALDI MSI data sets. Furthermore, two-dimensional MSI was studied for embryos at different cell developmental stages (1-, 2-, 4-, 8-, and 16-cell stage) to investigate the localization changes of some lipids, revealing heterogeneous localizations of different classes of lipids in the embryo.

The fifth chapter discusses the development of a high-throughput MALDI-MS based metabolomics platform using a microarray of nanoparticles and organic matrices. Five matrices that provide broad metabolite coverage were selected and used to analyze turkey gut microbiome samples. Over two thousand unique metabolite features were reproducibly detected across intestinal samples from turkeys fed a diet amended with therapeutic or sub-therapeutic antibiotics, or non-amended feed. This protocol was applied to fifty two turkey cecal samples at three different time points from the antibiotic feed trial, which allowed distinct metabolite profiles to be discovered.

The sixth chapter presents an on-tissue chemical modification strategy for high-spatial resolution MALDI-MSI. A mass spectrometry imaging methodology was use to selectively enhance the metabolite signals for a sub-metabolome at a time by performing on tissue derivatizations. Three well-known on-tissue derivatization methods were used: coniferyl aldehyde for primary amines, Girard’s reagent T for carbonyl groups, and 2-picolylamine for carboxylic acids. This proof of concept experiment was applied to cross-sections of maize leaves and roots, and enabled the identification of over five hundred new unique metabolite features. Combined, this approach facilitated the visualization of various classes of compounds, which can eventually allow high-spatial resolution MSI in the metabolomics scale.