BackgroundCytotoxic efficacy of anticancer drugs has been widely studied with monolayer-cultured cancer cells. However, the efficacy of drugs under two-dimensional (2D) culture condition usually differs from that of three-dimensional (3D) one. In the present study, an in vitro tumor tissue model was constructed using alginate hydrogel, and in vitro cytotoxic efficacy of two anticancer drugs (cisplatin and DZNep) was investigated in chondrosarcomas, and compared to in vivo response.MethodsThree cell lines derived from human chondrosarcomas, CH2879, JJ012 and SW1353, were embedded in alginate hydrogel. Proliferation and survival were assayed by ATP measurement using Cell Titer-Glo luminescent cell viability assay kit, and by counting viable cells in beads. Collagen and COMP expression was determined by RT-PCR. Invasion/migration was estimated by counting cells leaving alginate beads and adhering to culture dish. Then, chondrosarcoma response to cisplatin and DZNep was compared between cells cultured in monolayer or embedded in alginate, and using chondrosarcoma xenografts in nude mice.ResultsChondrosarcomas survived at least for 8 weeks, after embedment in alginate. However, only CH2879 cells could proliferate. Also, this cell line is more invasive than SW1353 and JJ012, which was coherent with the grade of their respective primary tumors. Furthermore, the expression of type II collagen was higher in chondrosarcomas cultured in 3D than in 2D. Interestingly, this 3D culture system allows to validate the absence of response of chondrosarcomas to cisplatin, and to predict the efficiency of DZNep to reduce chondrosarcoma growth in vivo.ConclusionsThis study validates alginate beads as a relevant 3D model to study cancer biology and tumor responses to biological treatments.

No abstract

No abstract

Cell-surface proteoglycans (PGs) were extracted with Triton X-100 from rabbit articular chondrocytes cultured in monolayers and in alginate beads. They were first purified on DEAE-Trisacryl columns and the proportion of hydrophobic PGs was determined by both Octyl-Sepharose chromatography and partitioning in Triton X-114. These two methods revealed that the proportion of hydrophobic PGs was higher in monolayer culture system as compared to alginate beads (24 and 15%, respectively). Characterization of the PGs by Sepharose CL 6B gel filtration followed by electrophoresis indicated that the PGs isolated from monolayers were composed of three chondroitin sulfate (CS) PGs (core proteins of 180, 100 and 50 kDa) and a heparan sulfate (HS) PG (core protein of 60 kDa). In the alginate system. CSPGs with core proteins of 180, 45 and 32 kDa were observed, but no HSPG was present. In parallel, the effect of TGF-beta on the distribution of membrane-associated PGs was studied. The results showed that the synthesis of cell-surface PGs was stimulated by TGF-beta in monolayers whereas it was inhibited in alginate beads, but the amount of hydrophobic PGs was not altered by the growth factor. These data clearly indicate that TGF-beta induces a differential expression of the PG families present at the cell surface. Taken together, the results reveal the complex regulation of cell-surface PG distribution, which obviously depends on the culture method used and suggest that rabbit articular chondrocytes may differentially respond to extracellular ligands according to their morphological state and environment.