Faculty Bibliography

CD34+ stem cells play an important role during liver development and regeneration. Thus, we hypothesized that some human liver carcinomas (HLCs) might be derived from transformed CD34+ stem cells. Here we determined that a population of CD34+ cells isolated from PLC/PRF/5 hepatoma cells (PLC) appears to function as liver cancer stem cells (LCSC) by forming human liver carcinomas (HLC) in immunodeficient mice with as few as 100 cells. Moreover, the CD34+ PLC subpopulation cells had an advantage over CD34- PLCs at initiating tumors. Three types of HLCs were generated from CD34+ PLC: hepatocellular carcinomas, HCC; cholangiocarcinomas, CC; and combined hepatocellular cholangiocarcinomas, CHC. Tumors formed in mice transplanted with 12 subpopulations and 6 progeny subpopulations of CD34+ PLC cells. Interestingly, progenies with certain surface antigens (CD133, or CD44, or CD90, or EPCAM) predominantly yielded HCCs. CD34+PLCs that also expressed OV6 and their progeny OV6+ cells produced primarily CHC and CC.This represents the first experiment to demonstrate that the OV6+ antigen is associated with human CHC and CC. CD34+PLCs that also expressed CD31 and their progeny CD31+ cells formed CHCs. Gene expression patterns and tumor cell populations from all xenografts exhibited diverse patterns, indicating that tumor-initiating cells (TIC) with distinct antigenic profiles contribute to cancer cell heterogeneity. Therefore, we identified CD34+ PLC cells functioning as LCSCs generating three types of HLCs. Eighteen subpopulations from one origin had the capacity independently to initiate tumors, thus functioning as TICs. This finding has broad implications for better understanding of the multistep model of tumor initiation and progression. Our finding also indicates that CD34+ PLCs that also express OV6 or CD31 result in types of HLCs. This is the first report that PLC/PRF/5 subpopulations expressing CD34 in combination with particular antigens defines categories of HLCs, implicating a diversity of origins for HLC.

Epithelial cell adhesion molecule (EpCAM) is a transmembrane glycoprotein, which is frequently and highly expressed on carcinomas, tumor-initiating cells, selected tissue progenitors, embryonic and adult stem cells. During liver development, EpCAM demonstrates a dynamic expression since it can be detected in fetal liver, including cells of the parenchyma, whereas mature hepatocytes are devoid of EpCAM. Liver regeneration is associated with a population of EpCAM-positive cells within ductular reactions, which gradually lose the expression of EpCAM along with maturation into hepatocytes. EpCAM can be switched on and off through a wide panel of strategies to fine-tune EpCAM-dependent functional and differentiative traits. EpCAM-associated functions relate to cell-cell adhesion, proliferation, maintenance of a pluripotent state, regulation of differentiation, migration and invasion. These functions can be conferred either by the full-length protein and/or EpCAM-derived fragments, which are generated upon regulated intramembrane proteolysis. Control by EpCAM, therefore, not only occurs in the presence or lack of full-length EpCAM at cellular membranes, but also in varying rates of the formation of EpCAM-derived fragments that have their own regulatory properties, and in changes in the association of EpCAM with interaction partners. Thus, spatiotemporal localization of EpCAM in immature liver progenitors, transit-amplifying cells and mature liver cells will decisively impact on the regulation of EpCAM functions and might be one of the triggers that contribute to the adaptive processes in stem/progenitor cell lineages. This review will summarize EpCAM-related molecular events and how they relate to hepatobiliary differentiation and regeneration.

BACKGROUND: SK Hep-1 cells (SK cells) derived from a patient with liver adenocarcinoma have been considered a human hepatoma cell line with mesenchymal origin characteristics, however, SK cells do not express liver genes and exhibit liver function, thus, we hypothesized whether mesenchymal cells might contribute to human liver primary cancers. Here, we characterized SK cells and its tumourigenicity. METHODS AND PRINCIPAL FINDINGS: We found that classical mesenchymal stem cell (MSC) markers were presented on SK cells, but endothelial marker CD31, hematopoietic markers CD34 and CD45 were negative. SK cells are capable of differentiate into adipocytes and osteoblasts as adipose-derived MSC (Ad-MSC) and bone marrow-derived MSC (BM-MSC) do. Importantly, a single SK cell exhibited a substantial tumourigenicity and metastatic capacity in immunodefficient mice. Metastasis not only occurred in circulating organs such as lung, liver, and kidneys, but also in muscle, outer abdomen, and skin. SK cells presented greater in vitro invasive capacity than those of Ad-MSC and BM-MSC. The xenograft cells from subcutaneous and metastatic tumors exhibited a similar tumourigenicity and metastatic capacity, and showed the same relatively homogenous population with MSC characteristics when compared to parental SK cells. SK cells could unlimitedly expand in vitro without losing MSC characteristics, its tumuorigenicity and metastatic capacity, indicating that SK cells are oncogenic MSC with enhanced self-renewal capacity. We believe that this is the first report that human MSC appear to be transformed into cancer stem cells (CSC), and that their derivatives also function as CSCs. CONCLUSION: Our findings demonstrate that SK cells represent a transformation mechanism of normal MSC into an enhanced self-renewal CSC with metastasis capacity, SK cells and their xenografts represent a same relative homogeneity of CSC with substantial metastatic capacity. Thus, it represents a novel mechanism of tumor initiation, development and metastasis by CSCs of non-epithelial and endothelia origin.