HR-MAS-NMR of intact tissue biopsies is a well-established method resulting in one NMR spectrum per whole biopsy showing all detectable metabolites at once. The aim of this project was to explore the possibility and usefulness of monitoring specific locations within the biopsy using HR-MAS-NMR. Firstly, the method was applied to a classic toxicology situation. Many drug development compounds fail because of preclinical animal liver toxicity conventionally detected using histology. Usually, only one of the murine liver lobes is used for this and is assumed to be representative of the whole organ. In this work, a metabolic variation across
murine liver lobes has been investigated via a set of biopsies across all lobes. Using HR-MAS-NMR spectra analysed by various types of multivariate analysis, no lobe-specific metabolic variation could be found, confirming the general validity of the representative
lobe approach. To increase location specificity, a spatially-resolved NMR pulse sequence (slice local-
ized spectroscopy (SLS)) was modified and its respective effectiveness was explored. The pulse sequence was first validated using artificially created samples (phantoms), and practical examples were layered fruit separated by paraffin film and milled phantoms produced
from materials which were magnetic-susceptibility-matched to the HR-MAS rotor. The HR-MAS SLS sequence was then applied to a mixed mouse renal tissue biopsy, and renal cortex and medulla successfully assigned to individual slices from spatially-resolved spectra using pure cortex and medulla reference HR-MAS-NMR spectra and orthogonal projection to latent structures discriminant analysis (OPLS-DA) to establish metabolic
markers differentiating the two. Together, this work shows the potential of HR-MAS-NMR as applied to tissue biopsies.
Particularly, spatially-resolved methods hold potential for improved biochemical and mechanistic understanding and the methodology could be expanded to applications
in many areas of biomedical relevance.