TMET-36. A spatially resolved multi-omic pipeline to study metabolic therapy response in glioblastoma

Abstract Glioblastoma presents a significant therapeutic challenge due to its molecular heterogeneity and resistance to standard treatments. Understanding the spatial distribution of proteins and metabolites within the tumour is essential for developing effective therapies, including metabolic strategies such as arginine deprivation using ADI-PEG20 (pegylated arginine deiminase). To address this spatial complexity, we developed a spatially resolved multi-omic pipeline that integrates desorption electrospray ionisation mass spectrometry imaging (DESI-MSI) for metabolomic profiling with LC-MS/MS-based quantitative proteomics on a single tissue section to preserve spatial context. Using fresh-frozen healthy mouse brain sections (8 μm thick), we performed high-resolution spatial metabolomic imaging with DESI-MSI and excised regions of interest (ROIs) by laser microdissection. Proteins from these regions were extracted, digested, and analysed by LC-MS/MS, enabling identification of over 1,100 proteins within 10 × 10 μm ROIs. This approach preserves spatial information, maximises protein recovery, and improves reproducibility. We then applied DESI-MSI to glioblastoma mouse tumours treated with two doses of ADI-PEG20. DESI-MSI revealed extensive arginine depletion across the tumour mass, demonstrating the drug’s effective metabolic impact, along with variable metabolic distributions between the tumour core and periphery. These findings highlight the spatial complexity of metabolic responses in glioblastoma and emphasise the importance of preserving spatial context to better understand tumour biology. Our spatially resolved multi-omic workflow supports high-resolution metabolite mapping for hypothesis-driven spatial analysis of proteins from a single tissue section, making it particularly advantageous for studies involving rare or limited clinical samples. It holds strong potential for widespread application in research across diverse tissue types and disease models.