Development of an innovative 3D cell culture system to study tumour--stroma interactions in non-small cell lung cancer cells.

ABSTRACT

INTRODUCTION: We describe a novel 3D co-culture model using non-small cell lung cancer (NSCLC) cell lines in combination with lung fibroblasts. This model allows the investigation of tumour-stroma interactions and addresses the importance of having a more in vivo like cell culture model. METHODS: Automation-compatible multi-well hanging drop microtiter plates were used for the production of 3D mono- and co-cultures. In these hanging drops the two NSCLC cell lines A549 and Colo699 were cultivated either alone or co-cultured with lung fibroblasts. The viability of tumour spheroids was confirmed after five and ten days by using Annexin V/Propidium Iodide staining for flow-cytometry. Tumour fibroblast spheroid formation was characterized by scanning electron microscope (SEM), semi-thin sections, fluorescence microscope and immunohistochemistry (IHC). In addition to conventional histology, protein expression of E-Cadherin, vimentin, Ki67, fibronectin, cytokeratin 7 and α-smooth muscle actin (α-SMA) was investigated by IHC. RESULTS: Lower viability was observed in A549 monocultures compared to co-cultures, whereas Colo699 monocultures showed better viability compared to co-cultures. Ki67 expression varied significantly between mono- and co-cultures in both tumour cell lines. An increase of vimentin and decreased E-Cadherin expression could be detected during the course of the cultivation suggesting a transition to a more mesenchymal phenotype. Furthermore, the fibroblast cell line showed an expression of α-SMA only in co-culture with the cancer cell line A549, thereby indicating a mesenchymal to mesenchymal shift to an even more myofibroblast phenotype. CONCLUSION: We demonstrate that our method is a promising tool for the generation of tumour spheroid co-cultures. Furthermore, these spheroids allow the investigation of tumour-stroma interactions and a better reflection of in vivo conditions of cancer cells in their microenvironment. Our method holds potential to contribute to the development of anti-cancer agents and support the search for biomarkers.