Proteotranscriptomic decoding of the surface-endocytome in primary glioblastoma cells identifies potential target antigens in the hypoxic tumor niche

Backgroung: Antibody and cell-based immunotherapies, e.g., antibody-drug-conjugates and CAR-T cells, targeted at cell-surface proteins, currently revolutionize clinical oncology. However, target selection warrants a better understanding of the surface-endocytome and how it is modulated by the tumor microenvironment. Here, we unravel the surface-endocytome landscape and its remodeling by hypoxia in primary cultures from glioblastoma (GBM) patients that currently lack targeted therapies. Methods. We employed a comprehensive approach for global and dynamic mapping of surfaceome and endocytosed (endocytome) proteins in primary GBM cultures at normoxia or hypoxia. Selected target candidates were validated by immunofluorescence analyses in patient tumor sections, spheroids, and 2D cultures, and were finally confronted by ADC cytotoxicity studies. Results: We reveal a heterogeneous surface-endocytome profile across GBM cultures from three patients representing different transcriptional subtypes. CD44 emerged as a commonly abundant surfaceome antigen across GBM cultures, and we established a direct correlation between CD44 endocytic activity and ADC cytotoxicity. We elucidate how hypoxia profoundly reshapes the global surface-endocytome, and identify several hypoxia-induced antigens (CXADR, CD47, BSG, CD81, FXYD6), unique to or shared among GBM cultures. Importantly, we highlight the limited correlation between transcriptomics and proteomics, emphasizing the critical role of membrane protein enrichment strategies and targeted mass spectometry. Conclusions: These studies provide a comperehensive map of the surface-endocytome and its remodeling by hypoxia in GBM as a resource for target discovery. As proof-of-concept, we validate several proteins, either abundantly expressed in normoxia or induced by hypoxia, for further exploration as potential targets of immunotherapeutic approaches in GBM. Overall design: To unravel the differences between normoxic and hypoxic genes expressed in primary GBM cell lines. The HGCC primary GBM cells U3034-MG, U3047-MG, and U3017-MG were subconfluently seeded, left overnight to attach, and subjected to normoxia (21% O2) or hypoxia (1% O2) for 24 hours. Then, cells were detached and RNA extraction was performed with AllPrep DNA/RNA quiagen mini kit as per the manufacturer's instructions. Each condition/sample was sent in triplicate experiments (3x Normoxia /cell line + 3xHypoxia/cell line) and after data receival, differentially expressed genes (DEGs) analysis was performed for each hypoxic cell line, compared to their respective normoxic samples.