AbstractBackground & AimsIn liver cancer, leucine‐rich repeat‐containing G‐protein coupled receptor 5 (LGR5) compartment represents an important tumor‐initiating cell (TIC) population and served as a potential therapeutic target. Cancer‐associated fibroblasts (CAFs) is a critical part of the tumor microenvironment, heavily influenced TIC function and fate. However, deeply investigations have been hindered by the lack of accurate preclinical models to investigate the interaction between CAFs and TIC. Organoids model have achieved major advancements as a precious research model for recapitulating the morphological aspects of organs, and thus also serving as a candidate model to investigate the mutual interaction between different cell types. Consequently, this study aimed to construct a three‐dimensional (3D) co‐culture organoid model of primary LGR5‐expressing tumor stem cells from primary murine liver tumors with CAFs to investigate the impact of CAFs on LGR5 marked TICs in liver cancer.Materials and MethodsFirst, both of the transgenic LGR5‐diphtheria toxin receptor (DTR)‐GFP knock‐in mice and transgenic Rosa26‐mT mice developed primary liver tumors by diethylnitrosamine (DEN) administration. Tumor organoids and CAFs were generated from those primary liver cancer separately. Second, LGR5‐expressing TICs organoid with CAFs were established ex vivo based on cell–cell contact or trans‐well co‐culture system, and the mutual influence between those two types of cells was further investigated. Subsequently, immunodeficient mouse‐based xenograft model was further adopted to evaluate the influence of CAFs to LGR5 tumor stem cell, tumor formation, and metastasis.ResultsThe co‐culture organoid model composed of murine liver tumor LGR5+ tumor‐initiating cells and CAFs in 3D co‐culture was successfully established, with the intention to investigate their mutual interaction. The existence of CAFs upon engrafting tumor organoids resulted in dramatic higher number of LGR5+ cells in the neoplasia when compared with engrafting tumor organoids alone. Furthermore, ex vivo culture of isolated LGR5+ cells from tumors of co‐engrafted mice formed significantly larger size of organoids than mono‐engrafted. Our results also indicated significantly larger size and number of formed organoids, when LGR5+ cells co‐cultured with CAF in both cell–cell contact and paracrine signaling in vitro, comparing to LGR5+ cells alone. Furthermore, we found that specific knockout of LGR5 expressing cells suppressed CAF‐mediated promotion of tumor formation, growth, and metastasis in the experimental mice model.ConclusionsAltogether, in a 3D co‐culture type of murine liver LGR5+ cells and cancer‐associated fibroblasts, we have demonstrated robust effects of CAFs in the promotion of LGR5 marked liver TICs. We also further revealed the influence of tumor microenvironment on stem cell‐related therapy, suggesting the possibility of combing CAF‐targeted and tumor stem cell targeted therapy in treating liver cancer.