Spheroid Culture Differentially Affects Cancer Cell Sensitivity to Drugs in Melanoma and RCC Models

Abstract

2D culture as a model for drug testing often turns to be clinically futile. Therefore, 3D cultures (3Ds) show potential to better model responses to drugs observed in vivo. In preliminary studies, using melanoma (B16F10) and renal (RenCa) cancer, we confirmed that 3Ds better mimics the tumor microenvironment. Here, we evaluated how the proposed 3D mode of culture affects tumor cell susceptibility to anti-cancer drugs, which have distinct mechanisms of action (everolimus, doxorubicin, cisplatin). Melanoma spheroids showed higher resistance to all used drugs, as compared to 2D. In an RCC model, such modulation was only observed for doxorubicin treatment. As drug distribution was not affected by the 3D shape, we assessed the expression of MDR1 and mTor. Upregulation of MDR1 in RCC spheroids was observed, in contrast to melanoma. In both models, mTor expression was not affected by the 3D cultures. By NGS, 10 genes related with metabolism of xenobiotics by cytochrome p450 were deregulated in renal cancer spheroids; 9 of them were later confirmed in the melanoma model. The differences between 3D models and classical 2D cultures point to the potential to uncover new non-canonical mechanisms to explain drug resistance set by the tumor in its microenvironment.

Keywords: 3D; RCC; cisplatin; cytochrome; doxorubicin; drug-resistance; everolimus; glutathione-s-transferases; melanoma; spheroids.

Figure 1

Spheroid formation for 7 days. B16F10 and RenCa cell lines were seeded 500 of cells per drop in RPMI medium supplemented with methylcellulose; 3 days after cultivation in hanging drops, cell aggregates were transferred in agarose coated 96-well plates, and cultivated for another 4 days.

Figure 2

Sensitivity of cells in 2D and 3D culture. (A)—sensitivity to everolimus, doxorubicin, and cisplatin of B16F10 cell line in 2D and 3D culture conditions as assessed by the use of the alamarBlue method. (B)—sensitivity to everolimus, doxorubicin, and cisplatin of RenCa cell lines in 2D and 3D cultures assessed using alamarBlue method. Statistical analysis was performed by One-Way ANOVA/Tukey test or Kruskal–Wallis/Dunn’s test—* p < 0.05, ** p < 0.0021, **** p < 0.0001, N ≥ 3 (2D—control).

Figure 3

Drug distribution in the spheres. Immunofluorescent picture of doxorubicin (0.001 µM) treated melanoma and renal cancer spheres after 48 h treatment—(A). Histograms representing drug distribution in the cross-section of the sphere of melanoma and renal cancer model—(B).

Figure 4

Regulation of MDR1 and mTor expression. (A)—Relative to 2D protein expression of MDR1 of melanoma and renal cancer cells, Vinculin served as loading control. (B)—Relative to 2D protein expression of mTor of melanoma and renal cancer cells, Vinculin served as loading control. (C)—Detection of MDR1 and mTor by western blot. (D)—The gene expression of mtor was determined by quantitative RT-PCR (qRT-PCR); β-actin served as an internal control. (E)—Sensitivity to tariquidar in various concentrations estimated by Alamar blue assay—(D). Statistical analyses were performed by One-Way NOVA/Tukey test or Kruskal–Wallis/ Dunn’s test—* p < 0.05, ** p < 0.0021, **** p < 0.0001, N ≥ 3 (2D—control).

Figure 5

Modulation of drug related genes by 3D cultures. (A)—Enrichment analysis based on Gene Ontology pathway shows the top 4 activated processes in both cell types (web-based gene set analysis toolkit enrichment method: ORA; organism: mus musculus). (B)—screening of genes related with metabolism of xenobiotics. (D)—screening of genes related with metabolism of drugs. (C,E)—Functional enrichment protein network performed in Cytoscape software (v.3.8.0). (F)—The gene expression of Cyp2s1, Cyp2f2, Gstm6, Gsto2, Gsta4, Hpgds, Gsta1, Gsta2, Gstm7, Aldh31a for RenCa cells was determined by quantitative RT-PCR (qRT-PCR); β-actin served as a quantitative internal control. (G)—The gene expression of Cyp2s1, Cyp2f2, Gstm6, Gsto2, Gsta4, Hpgds, Gsta1, Gsta2, Gstm7, Aldh31a for B16F10 cells was determined by quantitative RT-PCR (qRT-PCR); β-actin served as a quantitative internal control. Statistical analysis was performed by One-Way ANOVA/Tukey test or Kruskal–Wallis/ Dunn’s test—* p < 0.05, ** p < 0.0021, *** p < 0.0002, N ≥ 3 (2D—control).