Development of a FRET-based 3D in vitro breast tumor model and its application in evaluating drug-induced apoptosis
Date of Issue2017-05-02
School of Chemical and Biomedical Engineering
Tumorigenesis is a dynamic process involving growth, proliferation, heterotypic cell-cell and cell-ECM interactions and migration. Having traditionally relied on two-dimensional (2D) cultures for cancer research, it has been well-proven that these systems are unable to recapitulate the complexity and heterogeneity of in vivo tumors effectively. Compelling evidence from various studies have shown that three-dimensional (3D) systems can overcome these limitations and can bridge the gap between 2D and animal models. These 3D models would be able to provide a deeper understanding of cancer pathogenesis and drug responses. Therefore, 3D spheroids of breast cancer were developed in less than 24 h in this thesis using the non-adhesive coating method. Expectedly, these breast cancer cells were able to form tight cluster of cells that simulated in vivo tumors in its architecture (the necrotic core) and expression of tumorigenic markers better than 2D monolayers. Three-dimensional models have been extensively used to determine in vivolike responses of investigational drugs and provide additional information on the efficacy of the drug in a more physiological system than 2D monolayers. However, many of the currently available techniques to assess drug-induced effects require dissociation of spheroids, large number of samples or complex technologies to visualize the drug responses making them destructive, time-consuming, and lacking real-time analysis. This has created a need for a physiologically relevant model that at the same time provides rapid, non invasive and longitudinal analysis of druginduced effects. Therefore, in this thesis, a novel fluorescence resonance energy transfer (FRET)-based 3D in vitro model has been developed to evaluate drug-induced apoptotic responses in a real-time manner. In this work, the FRET-based C3 biosensor, which detects the activation of caspase-3 enzyme during apoptosis, was employed. Therefore, when breast cancer cells were transfected with this biosensor (MCF7-C3), real-time information about the cell status under FRET imaging was achievable. Subsequently, on subjecting these spheroids to chemotherapeutic agent paclitaxel, cells at the rim of the spheroid displayed a reduced FRET effect (47 % reduction for 2000 nmol/L paclitaxel treatment), in a dose- and time-dependent manner. These results provide a proof-of-concept of using FRET-based biosensors in 3D models to detect and evaluate drug response in a non-invasive and time-series manner. Subsequently, this biomimetic 3D platform of MCF7-C3 cells was employed to screen various Au (I) N-heterocyclic carbene (NHC) complexes to assess their anti-cancer properties. The dose response curve and mitochondrial membrane potential analyses suggested that gold-complexed compounds had anti-cancer properties and the monomer compounds acted as controls not having any apoptosisinducing activity. Furthermore, it was confirmed by FRET imaging that these five gold complexes were able to activate caspase-3 enzyme, thus triggering apoptosis, in less than 72 h at a concentration of 5 μM on 2D. However, when these five compounds were tested in our 3D biomimetic model, the complexes were unable to elicit a response equivalent of their 2D response. This was further substantiated with the evaluation of FRET ratio of gold-complexed compounds and their corresponding monomer. This result demonstrates the applicability of using our 3D biomimetic system as an intermediate drug screening platform to bridge the gap between 2D and animal models for faster and efficient drug efficacy analysis. It is well-documented that cancer cells are not the only entities that constitute in vivo tumor. Cell types like fibroblasts, macrophages etc., interact with the cancer cells, in addition to providing physical support (dense ECM), thus contributing to the tumor microenvironment. As a step towards recreating an accurate tumor model in vitro, we have incorporated fibroblasts into our model system. In 24 h, a 3D coculture system comprising of FRET-based breast cancer cells and fibroblasts was formed. Further histological analysis have shown that this 3D co-culture system grown for a week resembled in vivo tumors more closely than their 3D monospheroid. It is believed that the additional involvement of fibroblasts in this FRETbased 3D co-culture model would enable better accuracy in screening and evaluating responses of investigational compounds. In conclusion, this thesis demonstrates that FRET-based 3D in vitro breast tumor models would serve as a novel platform for non-invasive imaging and longitudinal monitoring of rug-induced effects. Screening of NHC complexes as potential anti-cancer agents stand as a culmination of development of this 3D in vitro model. Additionally, the role of fibroblasts in tumor progression has been explored by incorporating them in our previously established 3D mono-spheroid model.