The role of cancer stem cells in liver cancer heterogeneity and subtypes

在一项比较慢性肝病和肝细胞癌的基因表达阵列研究中，我们揭示了一个分子签名，该签名区分了这些患者发展为晚期疾病的风险。上皮细胞黏附分子被确定为该信号中的主导基因，该基因的沉默导致肝癌细胞的生长抑制。我们的结果支持这样的观点，即在单个肿瘤形成之前，可以获得大量的癌前基因，并表明肝细胞癌存在另一种机制，即多个肿瘤启动事件可以在肿瘤开始时同时发生。在癌前组织中识别这些癌前基因及其激活或失活的机制可能会改善未来对高危人群的诊断，并可能指导新的化学预防策略。我们还利用全球分子图谱分析了EpCAM阳性和阴性肝细胞癌之间的基因表达差异。我们发现EpCAM可以显着地将肝细胞癌区分为两个类似于肝脏谱系的亚型。特别是，EpCAM阳性的肝细胞癌表现出明显的肝干/祖细胞特征，包括WNT-β连环蛋白途径的激活。同时，EpCAM阴性的肝细胞癌具有成熟肝细胞的特征。我们还发现，通过对甲胎蛋白状态的额外评估，肝细胞癌可以进一步分为四种不同的亚型。这四种亚型与肝细胞癌的预后有关，EPCAM双阳性和甲胎蛋白阳性细胞预后最差。此外，这些亚型类似于肝脏谱系的某些阶段，EPCAM/甲胎蛋白阳性细胞显示出明显的肝干细胞/祖细胞特征的分子特征。此外，这些细胞具有预后不良的肝细胞癌亚型的特点，能够在体外和体内模型中启动高侵袭性的肝细胞癌。提示EPCAM和甲胎蛋白是诊断肝细胞癌的有用标记物，可作为一种方便的预后分类系统。此外，EpCAM和甲胎蛋白可能作为下游分子维持肝癌的干性，并作为肝癌起始细胞的良好标记物。我们还探讨了EPCAM在具有干细胞/祖细胞特征的肝细胞癌亚型中升高的机制。我们发现WNT-β-连环蛋白途径的激活调节了EpCAM的表达。我们证明EpCAM是WNT-β-连环蛋白信号的生物传感器，并通过直接的Tcf结合元件相互作用被这一途径转录上调。我们的数据表明，EpCAM的表达和WNT-β-catenin信号的融合在维持肝细胞癌细胞的生长方面发挥了作用。抑制肝癌细胞生长可通过阻断EpCAM/WNT-β-catenin信号通路实现。根据这些发现，我们认为EpCAM/WNT-β-catenin信号转导功能维持了肝细胞癌干细胞的生长，基于EpCAM表达的肝细胞癌分类在临床环境中可能有助于对可能受益于β-catenin/EpCAM辅助治疗的肝细胞癌患者进行分层。MicroRNAs是一种内源性非编码小RNA，通过与肿瘤发生的功能联系来调节基因的表达。使用全球microRNA微阵列方法来探索某些microRNA是否与肝细胞癌干细胞相关。我们发现，在肝癌干细胞中，保守的microRNA-181家族成员表达上调。抑制microRNA-181导致肝细胞癌干细胞数量减少和致瘤活性降低，而加入microRNA-181则导致该细胞类型的丰富。在进一步的研究中，我们发现microRNA-181可以直接靶向肝脏中分化的转录调节因子和WNT-β-连环蛋白信号的抑制物。此外，我们最近还发现，Wnt/β-catenin信号在转录水平上激活了肝癌中的microRNA-181s。这些结果提示了microRNA-181家族成员、Wnt/β连环蛋白信号与肝癌干细胞之间的新的调控联系，并暗示分子靶向microRNA-181或Wnt/β-连环蛋白信号转导可能根除肝细胞癌。我们最近评估了与肝内胆管细胞癌(ICC)的肝细胞癌干细胞和预后相关的分子特征。利用Affymetrix mRNA和NanoStringmicroRNA微阵列，亚洲ICC病例可以分为两个主要亚组，其中一个亚组与先前发现的具有干细胞基因表达特征的肝细胞癌有相同的基因表达特征。对ICC特异的mRNA和microRNA表达谱的综合分析表明，将miR-200C信号与上皮-间充质转化(EMT)联系起来的共同信号通路在ICC中优先激活，与干细胞的基因表达特征有关。MiR-200C的失活导致EMT的诱导，而miR-200C的激活导致EMT的减少，包括减少细胞的迁移和侵袭。我们还发现，已知的肝脏干/祖细胞标记物NCAM1是miR-200C的直接靶标。我们的研究结果表明，ICC和HCC具有共同的干细胞型分子特征，预后较差。我们的工作表明，EMT的特定成分可能被用作关键的生物标志物和临床相关的治疗靶点，用于侵袭性的干细胞样ICC。

In a gene expression array study comparing signatures of chronic liver diseases with hepatocellular carcinoma, we revealed a molecular signature that separates these patients for their risk of developing advanced disease. Epithelial cell adhesion molecule was identified as the lead gene in this signature and silencing of this gene resulted in growth suppression of hepatocellular carcinoma cells. Our results support the notion that large numbers of procarcinogenic genes can be acquired before a solitary tumor is formed, and suggest that an alternative mechanism exists for hepatocellular carcinoma, whereby multiple tumor-initiating events can simultaneously occur at the beginning of tumor initiation. The identification of these procarcinogenic genes in premalignant tissues and the mechanisms for their activation or inactivation may improve the future diagnosis of high-risk populations and may guide new strategies for chemoprevention. We have also used global molecular profiling to analyze the gene expression differences among EpCAM positive or negative hepatocellular carcinoma. We found that EpCAM could significantly differentiate hepatocellular carcinoma into two subtypes that resembled liver lineages. In particular, EpCAM-positive hepatocellular carcinoma displayed distinct features of hepatic stem/progenitor cells including an activation of the wnt-beta catenin pathway. Meanwhile, EpCAM-negative hepatocellular carcinoma displayed features of mature hepatocytes. We also found that hepatocellular carcinomas could be further stratified into four distinct subtypes with the additional assessment of alpha-fetoprotein status. These four subtypes were associated with prognostic outcome of hepatocellular carcinoma and cells double positive for EpCAM and alpha-fetoprotein had the worst prognosis. Furthermore, these subtypes resembled certain stages of liver lineages and EpCAM/alpha-fetoprotein-positive cells displayed a distinct molecular signature with features of hepatic stem/progenitor cells. Moreover, these cells, characterizing a poor prognostic hepatocellular carcinoma subtype, were capable of initiating highly invasive hepatocellular carcinoma in in-vitro and in-vivo models. This work suggests that EpCAM and alpha-fetoprotein are useful diagnostic markers for hepatocellular carcinoma which can be used as a convenient classification system for prognosis. Furthermore, EpCAM and alpha-fetoprotein may act as downstream molecules to maintain HCC stemness and serve as good markers for HCC initiating cells. We also explored the mechanism by which EpCAM is elevated in hepatocellular carcinoma subtypes with stem/progenitor cell features. We found that the activation of wnt-beta-catenin pathway regulates EpCAM expression. We demonstrate that EpCAM is a biosensor for wnt-beta-catenin signaling and is transcriptionally up-regulated by this pathway through direct Tcf binding element interactions. Our data suggest that the convergence of EpCAM expression and wnt-beta-catenin signaling functions to maintain hepatocellular carcinoma cell growth. Inhibition of hepatocellular carcinoma cell growth could be achieved through blockade of EpCAM/wnt-beta-catenin signaling in EpCAM-positive hepatocellular carcinoma cells. With these findings, we propose that EpCAM/wnt-beta-catenin signaling functions to maintain hepatocellular carcinoma stem cell growth and that EpCAM expression-based classification of hepatocellular carcinoma could be useful in clinical settings to stratify hepatocellular carcinoma patients who may benefit from beta-catenin/EpCAM adjuvant therapies. MicroRNAs are endogenous small noncoding RNAs that regulate gene expression with functional links to tumorigenesis. A global microRNA microarray approach was used to explore whether certain microRNAs were associated with hepatocellular carcinoma stem cells. We found that the conserved microRNA-181 family members were up-regulated in hepatoceullar carcinoma stem cells. Inhibitionof microRNA-181 led to a reduction in number and tumor initiating activity of hepatocellular carcinoma stem cells while addition of microRNA-181 led to an enrichment of this cell type. In further studies, we showed that microRNA-181 could directly target transcriptional regulators of differentiation in the liver and an inhibitor of wnt-beta-catenin signaling. In addition, we have recently shown that Wnt/beta-catenin signaling transcriptionally activates microRNA-181s in HCC. These results suggest a novel regulatory link between microRNA-181 family members, Wnt/beta catenin signaling and liver cancer stem cells and implies that molecular targeting of microRNA-181 or Wnt/beta-catenin signaling may eradicate hepatocellular carcinoma. We have recently assessed molecular signatures related to HCC stemness and outcome in intrahepatic cholangiocarcinoma (ICC). Using Affymetrix mRNA and Nanostring microRNA microarrays, Asian ICC cases could be segregated into two main subgroups, one of which shared gene expression signatures with previously identified HCC with stem cell gene expression traits. Integrative analyses of the ICC-specific mRNA and microRNA expression profiles revealed that a common signaling pathway linking miR-200c signaling to epithelial-mesenchymal transition (EMT) was preferentially activated in ICC with stem cell gene expression traits. Inactivation of miR-200c resulted in an induction of EMT while activation of miR-200c led to a reduction of EMT including a reduction of cell migration and invasion. We also found that NCAM1, a known hepatic stem/progenitor cell marker, was a direct target of miR-200c. Our results indicate that ICC and HCC share common stem-like molecular characteristics and poor prognosis. Our work suggests that specific components of EMT may be exploited as critical biomarkers and clinically relevant therapeutic targets for an aggressive form of stem cell-like ICC.