Abstract 1531 Mass spectrometry-based proteomic analysis of glioblastoma-derived extracellular vesicles and healthy brain cells for biomarkers identification

Cellular communication within the brain plays a crucial role in maintaining physiological equilibrium as well as triggering the onset of diseases. Extracellular vesicles (EVs) are released by donor cells and encapsulate RNA, DNA, and proteins that influence gene and protein expression upon uptake by recipient cells. Transcriptomic and proteomic cell alterations have been associated with the development and progression of glioblastomas. This study aims to identify EV-derived glioblastoma biomarkers through a comparative mass spectrometry-based bottom-up proteomic analysis of glioblastoma LN-229 cell line and healthy human neurons, astrocytes, and endothelial brain cells (HEBC), therefore, advancing the understanding of the underlying mechanisms of disease progression and paving the way for the development of liquid-biopsy-based noninvasive diagnostic approaches. EVs were extracted from the conditioned media of LN-229, human neurons, astrocytes, and endothelial brain cell lines via polymer precipitation (Thermo Scientific, cat. # 4478359). Trifluoroacetic acid followed by neutralization with Tris buffer was used for rapid proteomic extraction. The protein samples were digested with trypsin and the peptides were purified using Pierce™ C18 Tips (Thermo Scientific, Cat. # 87782). Fifty nanograms of peptides were injected for mass spectrometry analysis on the timsTOF Pro 2 instrument. Each experimental group included three biological replicates. Data analysis was performed using Spectronaut software (version 18.3, Biognosys) for processing mass spectrometry data acquired using data-independent acquisition methods. The statistically significant proteins were annotated using the STRING database. Gene Ontology enrichment analysis was conducted to analyze the biological significance of the overexpressed proteins. Comparative analysis of astrocytes and LN-229 EVs showed that the glioblastoma fraction displayed an average of 1,499 proteins. The proteomic profile of human astrocyte-derived EVs featured 1,301 proteins. A total of 48 proteins showed statistically significant upregulation in glioblastoma EVs compared to human astrocytes. A subset of 13 proteins are associated with glioblastoma tumorigenesis and metastasis. Analysis of EVs derived from both glioblastoma and healthy human astrocytes, neurons, and HEBC was performed to identify biologically significant biomarkers. EVs released by the human brain include a mixture of EVs originating from neurons, astrocytes, and HEBC. Therefore, accurate biomarker identification requires a comparison of the proteomic expression profiles between glioblastoma and a healthy, diverse background. The analysis focused on identifying glioblastoma EV-derived proteins that were consistently overexpressed across at least two of the neurons, astrocytes, and HEBC experimental groups. A subset of 20 proteins was identified as potential biomarkers of glioblastoma. The STRING database annotation indicated that these proteins play a significant role in promoting cancer development and metastasis including increased protease activity, angiogenesis, and regulation of immune and redox response. This study sheds light on the role of EVs in glioblastoma progression and metastasis, and their biomarker's potential for early detection and targeted therapeutic interventions in glioblastoma patients. This research was funded by the National Science Foundation EPSCoR Research Infrastructure Improvement Track-4: EPSCoR Research Fellows, grant number 2228581.