Transcriptomic analysis of human glioblastoma U87 cells in monolayer and multi-cellular spheroid cell culture models

Abstract

The aim of this study was to evaluate the changes in genome-wide gene and miRNA expression in human glioblastoma U87 cells after the transition from standard monolayer to multi-cellular spheroid cell culture conditions. Microarray analysis revealed differentially expressed genes between 2D and 3D cultured cells. Next generation short non-coding RNA sequencing identified miRNAs with a significant expression change in cell spheroids. Data of these genome-wide expression analyses were validated using RT-qPCR. Bioinformatical enrichment analysis of microarray data identified 25 significantly enriched KEGG functional gene groups which were divided into 4 major categories: immune response related; cell adhesion, metabolism, and other functional groups. Metabolic mapping analysis showed that main expression profile differences between 2D and 3D MCS model systems arise from upregulation of genes involved in autoimmune response, angiogenesis and cell adhesion mechanisms. Moreover, investigation of KEGG pathway maps led to the hypothesis that transcription in MCS is hypoxia-induced and regulated through a nuclear hormone receptor RORA, which is activated by cholesterol-derived ligands via activation of steroid biosynthesis pathway. For the purpose of determining whether differentially expressed miRNA is responsible for altered gene expression, we generated correlation matrices of miRNA and their experimentally validated or in silico predicted target genes. Interestingly, we have shown that there is no significant inverse correlation of miRNA and their respective target gene expression profiles between 2D and 3D MCS cultured U87 cells.