Extracellular vesicle cargo change 2D/3D cell culture conditions in glioblastoma. Extracellular vesicle cargo change in 3D tumor organoids as compared to conventional 2D glioblastoma cell systems

Abstract Background: Newer 3D culturing approaches are a promising way to better mimic the in vivo tumor microenvironment and to study the interactions between the heterogeneous cell populations of glioblastoma multiforme. Like many other tumors, glioblastoma uses extracellular vesicles as an intercellular communication system to prepare surrounding tissue for invasive tumor growth. However, little is known about the effects of 3D culture on extracellular vesicles. The aim of this study was to comprehensively characterize extracellular vesicles in 3D organoid models compared with conventional 2D cell culture systems, using two different methods of vesicle isolation. Methods: Primary glioblastoma cells were cultured in 2D and 3D organoid models. Extracellular vesicles were obtained by precipitation and immunoaffinity, with the latter allowing targeted isolation of the CD9/CD63/CD81 vesicle subpopulation. Comprehensive vesicle characterization was performed and miRNA expression profiles were generated by smallRNA-sequencing. In silico analysis of differentially regulated miRNAs was performed to identify mRNA targets and corresponding signaling pathways. The tumor cell media and extracellular vesicle proteome were analysed by high-resolution mass spectrometry. Results: We observed an increased concentration of extracellular vesicles in 3D organoids. Differential gene expression analysis revealed twelve significantly regulated miRNAs (nine downregulated and three upregulated in 3D organoids compared with 2D culture). MiR-23a-3p, known to be involved in glioblastoma invasion, was significantly increased in 3D. MiR-7-5p, which counteracts GBM malignancy, was significantly decreased. Moreover, we identified four miRNAs (miR-323a-3p, miR-382-5p, miR-370-3p, miR-134-5p) located within the DLK1-DIO3 domain, a cancer-associated genomic region, suggesting a possible importance of this region in glioblastoma progression. Overrepresentation analysis identified alterations of EV cargo in 3D organoids, including representation of several miRNA targets and proteins primarily implicated in the immune response. Conclusion: Our results show that extracellular vesicles secreted by glioblastoma cells in 3D organoid models harbour different miRNA and protein profiles compared with the conventional 2D model. Thus, culture conditions alter communication through EVs, evidently toward immune signaling, suggesting the occurrence of an immunosuppressive phenotype that appears to be better reflected in 3D organoids, even at the EV level. These findings may provide the basis for clinical biomarker studies and new therapeutic options in GBM.