Modeling Primary Liver Cancer using Tumor-Derived Extracellular Matrix

Abstract

Due to the global health impact of primary livercancer, resourceful models are required to develop new therapies and to studychemo-resistance. Recentadvantages in understanding cancer complexity have shifted research to theimportance of the niche of the tumor cells. In cholangiocarcinoma (CCA), anaggressive malignancy of the liver, the extracellular matrix (ECM) surroundingthe cancer cells plays an essential role in tumor progression andchemo-resistance. To research potential mechanisms involved indrug-resistance it is necessary to grow CCA-derived cells in the vicinity oftheir native tumor micro-environment. Cell-free scaffolds can be created bydecellularizing liver tissue contributing to the availability of new tissueengineering culture platforms. The resulting extracellular matrix can serve asa scaffold resembling the native non-cellular component of the tumorenvironment. However, a protocol to decellularize CCA tumor tissue is not yetestablished. It is not possible to use the traditional perfusion-baseddecellularization protocol for tumors, due to the lack of an intact capsule andvascular system to cannulate. The aim of this study is to develop a tumorscaffold retaining the essential characteristics of the extracellular matrixand to investigate its feasibility to serve as a scaffold for CCA-derivedorganoids. To do this, CCA tumors were decellularized using an adaptedprotocol. Efficiency of removal of cells was determined by histology andanalysis of DNA content and the preservation of collagen and tissue stiffnesswas assessed. Architecture of decellularized tumor was compared to architecture of decellularized normal liver tissue. A-cellular scaffolds wererepopulated with CCA-derived organoids and ingrowth and viability of cells wereevaluated by histology and confocal imaging. Metabolic activity and geneexpression levels were compared to cells grown in standard basement membranematrix (BME) gel and grown in normal liver a-cellular scaffolds. Aftercompleting the protocol, tumor tissues were white in color and histologicalexamination revealed no cells could be detected indicating completedecellularization. DNA content was slightly above the critical thresholdindicating not all nuclear debris was removed. Characterization of the tumortissue and normal liver matrices revealed markedly higher collagen content andhigher stiffness in tumor tissue underlining the importance to differentiatebetween tumor and normal scaffolds. Recellularization of the scaffoldsrevealed that cells did not only attach to the surface, but also grow insidethe scaffold. Comparable metabolic activity was observed between cells grown intumor scaffold and in BME. We observed that cells seeded in tumor scaffoldsshowed different gene expression levels than those seeded in standard BMEculture. The resulting combination of CCA-derived organoids and tumor ECMprovides an innovative basis for a model in which cell-extracellular matrixinteractions can be studied. To make a complete culture platform, futurestudies should focus on including the cellular component of the tumor micro-environment.This will enable the identification of patient-specific drug sensitivity and tostudy mechanisms involved in drug-resistance.