Date Published: January 24, 2019
Publisher: Public Library of Science
Author(s): Fereshtehsadat Mirab, You Jung Kang, Sheereen Majd, Neil Cameron.
Treatment of glioblastoma, the most common and aggressive type of primary brain tumors, is a major medical challenge and the development of new alternatives requires simple yet realistic models for these tumors. In vitro spheroid models offer attractive platforms to mimic the tumor behavior in vivo and have thus, been increasingly applied for assessment of drug efficacy in various tumors. The aim of this study was to produce and characterize size-controlled U251 glioma spheroids towards application in glioma drug evaluation studies. To this end, we fabricated agarose hydrogel microwells with cylindrical shape and diameters of 70–700 μm and applied these wells without any surface modification for glioma spheroid formation. The resultant spheroids were homogeneous in size and shape, exhibited high cell viability (> 90%), and had a similar growth rate to that of natural brain tumors. The final size of spheroids depended on cell seeding density and microwell size. The spheroids’ volume increased linearly with the cell seeding density and the rate of this change increased with the well size. Lastly, we tested the therapeutic effect of an anti-cancer drug, Di-2-pyridylketone-4,4-dimethyl-3-thiosemicarbazone (Dp44mT) on the resultant glioma spheroids and demonstrated the applicability of this spheroid model for drug efficacy studies.
Glioblastoma multiforme (GBM, grade IV glioma) is the most common and aggressive type of primary brain tumor and has the worst prognosis among these tumors . Traditional treatment strategies that include surgery, radiation therapy, and chemotherapy have limited success in eliminating these tumors, often leading to their fatal recurrence [2, 3]. Development of more effective treatments for this deadly cancer requires simple yet realistic tumor models that enable proper evaluation of their therapeutic efficacy prior to the clinical use. For several decades, drug efficacy evaluations have been performed on two-dimensional (2D) in vitro disease models formed on flat and rigid substrates. Despite the ease of use, these 2D cultures often fail to predict the in vivo tissue response to candidate drugs as they do not resemble the natural tumor stromal heterogeneity, cell-cell interactions, gradient in nutrient concentrations, and extracellular matrix (ECM) compounds . Moreover, culturing cells in a monolayer format leads to a default apical-basal polarity that is different from the spatial organization of cells in natural tissues, making cells more sensitive to anticancer drugs . Recent efforts have thus, been focused on the development of more physiologically relevant models that can better predict the response of natural tumors to candidate anti-cancer compounds .
This study describes the preparation and characterization of size-controlled glioma spheroids from a commonly used glioma cell line, U251. Agarose gel microwells with no surface treatment were applied to produce highly viable (> 90%) tumor spheroids. We demonstrated that the design of microwells and the initial cell seeding density provided great control over the final size of spheroids. The rate of growth in the glioma spheroids (~2.3% volume increase per day) was comparable to that reported for natural brain glioams (~2.2%) , demonstrating U251 spheroids produced in agarose microwells can effectively mimic the growth of natural brain tumors for at least one week. Moreover, we explored the applicability of these glioma spheroids for drug evaluation studies by testing the effect of an anti-tumor chelator Dp44mT on these spheroids. The results demonstrated that Dp44mT treatment effectively reduced the tumor size (~11.4% per day), which is comparable with prior reports on Dp44mT-induced size reduction of other tumor types in animal models (~8%) . Moreover, flow cytometric analysis confirmed that Dp44mT induced significant apoptotic cell death in drug treated glioma spheroids, supporting the prior reports on the apoptosis inducing effect of Dp44mT on other tumor types [23, 34]. Together, these results demonstrate the great potential of the present glioma model for brain glioma studies particularly for drug performance evaluations. This model can thus, contribute to the development of more effective therapeutics for brain gliomas.