He, Wenshi
(2023)
Surface Mass Spectrometry for in situ Metabolomic Analysis of Human Brain Tumours.
PhD thesis, University of Nottingham.
Abstract
Malignant brain tumours represent one of the most devastating diagnoses a clinician may ever deliver to a patient. Due to their aggressive, invasive, and destructive nature, these tumours are considered among the deadliest forms of human cancers in both adult and paediatric patients. The current multimodal treatment is insufficient to cure these patients, but merely extends their survival. Therefore, there is an urgent need for effective anti-cancer therapeutic products.
To identify novel therapeutic agents, we need to have improved understanding of the genetic landscape and underlying molecular characteristics of brain tumours, especially within the context of tumour heterogeneity, which likely accounts for therapy failure as certain cancer cells can escape from immune surveillance and therapeutic threats. Metabolomics has become an increasingly popular “omics” approach to understanding the underlying biological mechanisms related to a disease and identifying important metabolic pathways and biomarkers.
Surface mass spectrometry (MS) for in situ metabolic characterisation has great potential for high-throughput untargeted metabolomic analysis of brain tumour tissue. It allows rapid analysis of a large number of samples (as fast as 1 – 2 min per sample), which is challenging with the conventional liquid chromatography-mass spectrometry (LC-MS). Also, surface MS has widened the range of samples for metabolomics analysis to include those less incompatible with LC-MS, such as the formalin-fixed paraffin-embedded (FFPE) tissue sections. FFPE tissue archives are conventionally used in histopathological settings but gained increasing popularity in omics research because of their larger availability compared to the fresh frozen tissue and the fact that they can be stored at ambient environment for many years while maintaining vast resource for biomolecular investigation. However, the pre-extraction needed for LC-MS from these samples is extremely difficult as the region of interest may have a diameter of only a few hundred micrometre and the tissue sections are typically as thin as 4 µm. A solution to this is the surface MS that requires minimal amount of tissue material with minimal sample preparation.
In the recent years, rapid development and technical innovations have been made in the field of surface MS, including the OrbiSIMS and liquid extraction surface analysis (LESA)-MS. In this thesis, untargeted metabolic profiling methods using these two techniques were developed. In OrbiSIMS, we focused on the evaluation of sample substrates (conductive ITO slides and regular non-conductive slides) and analysis temperature (room-temperature and cryogenic condition), and their effects on untargeted metabolic profiling. It was discovered that room-temperature analysis on freeze-dried samples and cryogenic analysis of fresh frozen samples allowed the identification of many exclusive metabolite ions. In LESA-MS, we compared different solvent extraction systems, sampling methods (i.e., liquid microjunction sampling and "contact” LESA), MS/MS parameters and other experimental conditions, which enabled us to choose the most suitable settings to use in the subsequent experiments.
The developed method was first applied to study glioblastoma (GBM) intra-tumour heterogeneity. We showed that GBM cell sub-populations (necrotic, viable, and non-cancerous) within single tumours displayed distinct metabolic profiles. Therefore, it is possible that this analytical approach can be employed to predict different histological features in GBMs, such as the clinically relevant infiltrative margins which harbour molecular signatures associated with residual disease. Also, by mapping ubiquitous metabolites across necrotic and viable regions into pathways, we discovered metabolic activities (i.e., tryptophan metabolism) that were potentially essential for not only cancer cells with rapid proliferation (i.e., viable GBM cells) but also necrotic cells that often indicate poor prognosis and tumour recurrence, which makes it a potential therapeutic target.
Furthermore, we extended and tailored the developed metabolomic method to investigate metabolic alterations associated with paediatric high-grade gliomas. In this pilot study, we discovered that high-grade and low-grade gliomas exhibited distinct metabolic signatures characterised by phosphate, 5,6-dihydrouracil, imidazole-4-acetaldehyde, indol-3-yacetaldehyde, and L-3-cyanoalanine. We also discussed the current limitations with using tissue microarrays (TMAs) for LESA-MS and OrbiSIMS analysis, and provided practical solutions for future studies.
Further interest of LESA-MS/MS metabolomics comes from its application as a direct infusion MS technique for the analysis of liquid biopsy. A fast and robust workflow was developed using milk samples for the investigation of cattle lameness. In this study, we also discussed novel statistical strategies such as model triangulation that helped increase confidence in biomarker discovery.
Actions (Archive Staff Only)
 |
Edit View |
Tools
Tools
![[thumbnail of Corrections]](https://eprints.nottingham.ac.uk/style/images/fileicons/application_pdf.png "Corrections")