Related Experiment Video
Updated: Dec 11, 2025

Therapy Testing in a Spheroid-based 3D Cell Culture Model for Head and Neck Squamous Cell Carcinoma
Published on: April 20, 2018
Central nervous system (CNS) tumor cell heterogeneity contributes to differential platinum-based response in an in
Bryan Ôrtero Perez Gonçalves1, Sílvia Ligório Fialho2, Bárbara Reis Silvestrini2
1Cellular Biology, Research and Development Department, Ezequiel Dias Foundation, Belo Horizonte, Minas Gerais 30510-010, Brazil; Department of Biochemistry and Immunology, Federal University of Minas Gerais, Belo Horizonte, Minas Gerais 31270-901, Brazil.
Abstract:
One of the models that best explains the cellular heterogeneity observed in central nervous system (CNS) tumors is the presence of cancer stem cells (CSCs). CSCs can originate from differentiated adult cells that return to an undifferentiated stage through the mechanism known as epithelial-mesenchymal transition (EMT). In this paper, we evaluated cellular and molecular heterogeneity and the participation of the epithelial-mesenchymal transition (EMT) in glioblastoma (U-87 MG and LN-18) and neuroblastoma (KELLY and IMR-32) cell lines cultured in monolayer (2D) and neurosphere (CSC enrichment- 3D) models. For this, after treatment with cisplatin, we studied different cell subpopulations by immunophenotyping using neural stem cell/progenitor markers (ALDH, CD24, CD56, and CD133), mesenchymal stem cell markers (CD73, CD90, CD105, and CD146) and hematopoietic markers (CD14, CD19, CD34, CD45, and HLA-DR) and mRNA expression profiles of genes related to EMT, such as ZEB1, TWIST1, TGFB1, STAT3, and lncRNA HOTAIR. In addition, we evaluated the growth capacity of residual cells when treated with cisplatin using the chorioallantoic membrane (CAM) model to study disease relapse. After treatment with cisplatin, we found that the expression of STAT3 and TGFB1 genes markedly increased in the neurosphere of the IMR-32 cell line, and TWIST1 was upregulated in the neurosphere of LN-18. Only the nontreated monolayer of LN-18, KELLY, and IMR-32 amplified the lncRNA HOTAIR. The IMR-32 cell line exhibited an enrichment of CD24+/ALDH+ and this cell subset decreased after cisplatin treatment. We observed the loss of CD146+/CD73+ cell subpopulations in U-87 MG monolayer and neurosphere models, after cisplatin treatment, while in LN-18 monolayer cisplatin-treated cells, CD73+/CD90+ cell subpopulations increased. Neuroblastoma cell lines showed CD14+/HLA-DR- cell subpopulations representative of myeloid-derived suppressor cells (MDSCs). Tumors generated from residual cells, after exposure to cisplatin, grafted on CAM showed patterns of organization different from those of the controls. Thus, our findings strongly supported the idea that definitions of tumor phenotypic characteristics may help to establish better therapeutic strategies for the development of new drug targets.

