Prostate cancer stem cells and their involvement in metastasis
The recently resurrected cancer stem cell (CSC) theory sheds new light on understanding tumor biology. Most solid tumors have now been shown to contain CSCs, i.e., stem cell-like cancer cells. These cells, although generally rare, appear to be highly tumorigenic and may be the cells that drive tumor formation, maintain tumor homeostasis, and mediate tumor metastasis. In order to test whether any given human tumor cell population has CSC properties, the relatively enriched single tumor cells have to be put into a foreign microenvironment in a recipient animal to test their tumorigenic potential. Furthermore, various in vitro assays can be performed to demonstrate that the presumed CSCs have certain biological properties normally associated with the stem cells (SCs). Herein, I first present a comprehensive review of the experimental methodologies that our lab has been using in assaying putative prostate cancer (PCa) SCs in culture, xenograft tumors, and primary tumor samples. Clonal morphology is one of the critical properties of cultured cancer cells that has been largely ignored. Interestingly, long term-cultured human epithelial cancer cells form holoclones, meroclones, and paraclones, and tumor cell holoclones have been hypothesized to harbor stem-like cells. Using PC3 human prostate carcinoma cells as a model, we provide direct experimental evidence that tumor cell holoclones contain stem-like cells that can initiate serially transplantable tumors. Importantly, holoclones derived from either cultured PC3 cells or holoclone-initiated tumors can be serially passaged and regenerate all three types of clones. In contrast, meroclones and paraclones cannot be continuously propagated and fail to initiate tumor development. Phenotypic characterizations reveal high levels of CD44, [alpha]2[beta]1 integrin, and [beta]-catenin expression in holoclones, whereas meroclones and paraclones show markedly reduced expression of these markers. These observations have important implications in understanding morphologic heterogeneities and tumorigenic hierarchies in human epithelial cancer cells. PCa metastasis represents the worst outcome, and, if unchecked, will eventually kill the patient. Although many PCa cell-intrinsic molecules and end-organ factors have been implicated in the metastatic dissemination of PCa cells, the role of primary tumor microenvironment and the nature of the metastatic PCa cells remain poorly defined. By establishing a reliable and quantifiable experimental PCa metastasis model in NOD/SCID mice, we show that PCa cells implanted orthotopically (i.e., in the prostate) metastasize much more extensively and widely than those implanted ectopically (i.e., subcutaneously or s.c). Microarray-based gene expression profiling reveals that the orthotopically implanted human PCa cells prominently overexpress not only several classes of molecules involved in proteolysis/invasion/angiogenesis and inflammation, but also numerous developmental and SC regulating genes. These latter observations suggest that the orthotopic microenvironment (i.e. mouse prostate) appears to be promoting the manifestation of CSC phenotypes and these CSCs might be involved in enhanced metastasis in the orthotopic microenvironment and later distant organ metastasis. In support, shRNA-mediated knockdown in many metastatic and CSC genes greatly inhibits PCa cell metastasis. Importantly, PCa cells that express high levels of osteopontin (OPN) or CD24, when prospectively purified out and used in spontaneous metastasis assays, demonstrate high metastatic capacities characteristic of metastatic CSCs. In sharp contrast, PCa cells negative for OPN and CD24 expression show little metastatic property. Finally, we provide multiple pieces of additional evidence that metastatic/metastasizing PCa cells possess CSC properties.