P02.06.A MASS SPECTROMETRY BASED SPATIAL PROTEOMICS REVEALS DIFFERENTIALLY EXPRESSED PROTEINS IN THE CORE AND PERIPHERY OF GLIOBLASTOMA

Abstract BACKGROUND Glioblastoma is the most frequent and malignant tumor of the brain, and due to its infiltrative behavior, tumor recurrence is inevitable. Previous studies suggest various genes and signaling pathways as regulators of glioblastoma brain infiltration but many of these use transcriptomics-based discovery approaches and only few are based on human tissue. Although multiple biomarkers have been identified at the transcript level, it has been challenging to translate these findings in a clinical setting. A significant body of research has documented the discordance between mRNA and protein levels at different scales, potentially explaining the limited success. Accordingly, here we take a protein-centric view to investigate the tumor core, transition zone, and periphery of glioblastoma in order to identify potential therapeutic targets and signaling pathways responsible for the infiltration. MATERIAL AND METHODS Thirteen formalin-fixed paraffin-embedded (FFPE) glioblastoma, IDH wildtype patient tissue samples underwent Zeiss Palm MicroBeam-based laser capture microdissection (LCM) for isolation of 50.000 μm2 areas from core, transition zone, and periphery. Proteome quantifications were performed using the EvoSep One liquid chromatography platform in combination with the Orbitrap Astral Mass Spectrometer. RESULTS We obtained over 7500 protein groups from 50,000 μm2 sections and found 2347 spatially differentially regulated proteins, demonstrating the technological advancement in capturing regions-specific proteome profiles. We observed that the tumor core is enriched with tumor cells of mesenchymal subtype, while the periphery is enriched with tumor cells of classical, proneural, and neural-like subtype. Quantitative differential analysis revealed an increase in netrin and hedgehog signaling-related protein expression going from the tumor core to the periphery, suggesting potential signaling pathways that may play a role in the infiltration of tumor cells. CONCLUSION This study demonstrates our ability to perform low-input-based proteomics with high sensitivity. Our analysis identified potential therapeutic targets and their corresponding signaling pathways in the glioblastoma periphery. These targets are currently being functionally investigated and tested.