Worldwide primary liver cancer, which includes primary hepatocellular carcinoma (HCC), intrahepatic cholangiocarcinoma (ICC) and fibrolamellar HCC, is the fifth most common cancer and the third leading cause of cancer-associated mortality (1). Currently, surgery and liver transplantation are considered the optimal curative treatments for the disease (2); however, there is a significant shortage of organ donors, and surgical complications, recurrence and metastasis are common (3). Although epidemiological risk factors, including hepatitis B virus (HBV) and hepatitis C virus infection, have been identified, the molecular mechanisms underlying primary liver cancer remain unclear (4). Therefore, elucidation of the molecular pathogenesis of liver cancer and the development of effective targeted therapies is urgently required (5).

The Notch signaling pathway is evolutionarily conserved and controls numerous developmental processes, such as cell fate determination, terminal differentiation and proliferation (6). During embryonic development and adulthood, intracellular Notch signaling is required for cell specification, lineage commitment and maintenance of progenitor cells (7), in particular in the control of endothelial cell differentiation, arteriovenous specification and vascular development (8).

In mammals, the canonical Notch pathway includes four receptors (Notch1, 2, 3 and 4) and two ligand families [Jagged (JAG) 1 and 2 and Delta-like-ligand (Dll) 1, 3 and 4]. The Notch signaling pathway consists of ligand-induced activation of receptors, proteolytic cleavage and subsequent translocation of the notch intracellular domain (NICD) to the nucleus, where it functions as a transcriptional regulator (9). Activation of Notch signaling requires direct or indirect contact between cells expressing Notch ligands or receptors, and transmitting and receiving cells are subsequently modified by the interaction. Initially, cells express Notch receptors and ligands, and as the interaction continues, one cell becomes a transmitting cell by upregulating the expression of ligands and downregulating of receptors, while the receiving cells follow the opposite pattern (10). Prior to the NICD being transported to the nucleus, it is cleaved by the γ-secretase complex. In the nucleus, NICD interacts with C-repeat/DRE binding factor 1, a DNA-binding transcriptional repressor also known as the recombination signal binding protein for immunoglobulin Kappa J region (RBPJ), and converts it into a transcriptional activator that induces the transcription of target genes, including the family of Hes and Hey-associated transcription factors. Furthermore, in the liver, Notch partially controls the expression of Sox9, HNF1 Homeobox B and transforming growth factor-β, which are key regulators in hepatic lineage commitment (11,12) (Fig. 1).

RBPJ is a DNA-binding protein, also known as CSL, which is a member of the Suppressor of Hairless (Drosophila melanogaster) family of transcription factors that recognizes the consensus sequence C(T)GTGGGAA. RBPJ predominantly acts as a transcriptional repressor of promoters that possess RBPJ binding sites via the recruitment of other co-repressors. The most important function of RBPJ is to mediate signals from Notch receptors. RBPJ is a common downstream transcription factor of Notch receptors and its absence indicates a complete block of the Notch signaling pathway (13).

The present review discusses the findings of recent studies regarding Notch expression and summarizes Notch signaling during HCC and ICC development.

Notch is a highly regulated signaling mechanism with numerous specific features. In humans, mutations in Notch ligands or receptors are associated with various diseases; for example, JAG1 and Notch2 mutations are associated with Alagille syndrome (14), and Notch3 mutations with cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (15). Human genetic diseases and mutant mouse models have demonstrated the importance of Notch signaling in the development and remodeling of intrahepatic bile ducts (IHBD). Alagille syndrome (AGS) is a human autosomal dominant disorder that is caused by mutations in the Notch ligand JAG1, and less commonly in the Notch2 receptor (16). The estimated prevalence of AGS is 1 case per 70,000 live births worldwide (17). The disease is a multi-organ disorder most commonly diagnosed by liver abnormalities, which lead to hepatic bile duct paucity and cholestasis at birth (18). Cardiac, skeletal and ophthalmological abnormalities, and less frequently renal or vascular deficiencies, are also observed in patients with AGS (19). Numerous renal abnormalities are observed in patients with AGS, including renovascular disease, renal tubular acidosis, tubulointerstitial nephritis and renal dysplasia/hypoplasia (20). Notably, mice with haploinsufficiency for JAG1 exhibit no significant phenotypic abnormalities, suggesting that additional modifier genes contribute to the AGS phenotype observed in humans (21). Additionally, mutations in Notch3 are associated with inherited vascular diseases, including degenerative vascular disorder, cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (22).

The Notch signaling pathway is significantly associated with liver disease (23). In liver cirrhosis, the expression of Notch and Toll-like receptor signaling pathways are disordered (24). Based on the analysis of Notch and its target genes in HCC, the expression of Notch3 and Notch4 is aberrantly increased when compared with para-cancer tissue (25). Notch3 and 4 are rarely observed in normal liver tissue or para-carcinoma chronic hepatitis. In cirrhotic liver tissues, Notch3 mRNA and Hes6 mRNA expression levels are lower compared with those observed in HCC tissues. By contrast, high expression of Notch3 and low expression of Notch4 have been observed in human HCC HepG2 cell lines (26). Compared with HCC tissues of neoplastic cells, non-cancerous tissues adjacent to tumors have a high expression of Notch1 in the cytoplasm and Notch4 in nucleus, and a low expression of Notch2; however there is no significant difference between Notch4 and Notch3 expression. Thus, Notch receptor expression is abnormal in HCC (27). Rodent models with Notch loss or gain of function have demonstrated that Notch is involved in several stages of IHBD morphogenesis (28). Murine knockout (KO) studies for each of the mammalian Notch receptors and ligands have been conducted, and the resulting effects on the liver phenotype are presented in Table I (23–33).

Liver cirrhosis is commonly observed during the early stages of HCC and ICC (58). Previously, alterations in the Notch pathway have been identified in various solid tumors, such as breast, ovarian, pancreatic and liver cancer, melanoma and glioblastoma (59). The Notch signaling pathway exhibits a carcinogenic role in HCC. However, various members of the Notch signaling pathway may function as suppressors or enhancers of HCC. For example, the suppressive effect of Runt-related transcription factor 3 in HCC may be a result of Notch signaling inhibition (60). The Notch signaling pathway is involved in stem cell self-renewal and differentiation, and Notch signaling has been reported to promote the self-renewal of cancer stem-like cell niches in primary and metastatic tumors. Notch activation triggers epithelial-mesenchymal transformation (61). Several studies have revealed that a loss of the epithelial phenotype via epithelial-mesenchymal-transition promotes the acquisition of a stem-like phenotype and drug resistance (62–64). In addition, in patients with HCC an epithelial gene signature has been associated with sensitivity to the epidermal growth factor receptor inhibitors gefitinib and cetuximab (65). Deregulated expression of Notch, including the overexpression and activation of Notch proteins, ligands and targets, has been identified in numerous solid tumors, including cervical, endometrial, renal, lung and gastric carcinomas (66). These findings suggest that additional signaling pathways may affect whether Notch functions as a tumor suppressor or oncogene in a particular tissue (67).

Gain or loss of function mutations in the Notch pathway have been identified in several types of cancers, including neural stem cell tumors, lung carcinomas and prostate cancer (68). Although Notch does not directly lead to unregulated cell proliferation or genetic alterations that are associated with tumor progression, it alters the developmental state of cells and consequently maintains cells in a proliferative or undifferentiated state (69). In chronic HBV infection (CHB), repression of Notch receptors was demonstrated to lead to immune dysfunction (70). The contribution that a decreased expression of Notch receptors makes to ongoing fibrosis, cirrhosis and HCC is unclear; however, repression of Notch receptors in CHB has been suggested to repress immune regulation, resulting in the inhibition of differentiation and proliferation of effector cells leading to additional pathogenesis of CHB (71). Notably, the pro-mitogenic function of Notch was demonstrated in a model of partial hepatectomy (72). The pro-oncogenic function of Notch was also investigated by genome wide analysis of HCC samples, which revealed that the Notch coactivator MAML2 is a target of genetic alterations (73). Additionally, Notch signaling is crucial for the differentiation of hepatocytes into biliary lineage cells during the early stages of ICC developments. Gain and loss of function studies have demonstrated that ICC develops via the Notch-mediated differentiation of hepatocytes into biliary lineage cells, and that the malignancy and progression of the tumor are dependent on the intensity of Notch signaling in hepatocytes (74). Notably, hepatitis-infected hepatocytes may be converted into biliary lineage cells via Notch signal activation and thus become the point of origin for ICC (75). Therefore, suppression of Notch signaling may present a novel strategy for the treatment of ICC, since it may inhibit the conversion of hepatocytes into biliary lineage cells during the early stages of ICC development (76). Mice with liver-specific constitutive activation of Notch1 intracellular domain (N1ICD) may develop HCC when they reach adult age (77). Genomic profiling technology has revealed that a Notch-specific gene expression signature reported in mice overexpressing NICD was also present in a cluster of patients with HCC (78). Histological analyses of the mouse liver tissue exhibited similar features compared with that of the HCC patients, which included the presence of proliferating K-19 positive cells. Constitutive Notch2 overexpression causes HPCs to spontaneously develop into dedifferentiated HCC cells (79). In addition, Notch-induced malignant hepatocyte transformation is associated with downregulation of hepatocyte-associated genes and Sox9 expression (80). Fate-mapping studies have demonstrated that clear-cell adenocarcinoma (CCA) cells derived from hepatocytes may be converted to a biliary K-19 positive phenotype (81). Accordingly, in CCA development, N1ICD association with protein kinase B signaling in hepatocytes stimulated their malignant dedifferentiation (82). The stimulation of N1ICD expression by inflammatory mediators has also been reported in human CCA, which additionally supports the role of Notch in liver cancer (83) (Table II) (9,12,24,31,84–87).

Persistent activation of Notch signaling may lead to oncogenesis depending on modifier factors, including the inflammatory environment or the presence of other carcinogenic conditions that may cause HCC or CCA (88). A Notch signature has been reported in a subset of HCC patients and an overexpression of Notch receptors has been reported in human CCAs, which is fundamentally required for the development of targeted therapies (89). Silencing of the Notch pathway may potentially inhibit Notch-driven tumor progression and interfere with tumor aggressiveness, since Notch activation has been associated with a more malignant phenotype (84). However, the identification of a reliable tissue-specific biomarker of Notch inhibition is critical for the application of Notch-targeted therapy. Notably, the hepatic Notch target gene Sox9 is associated with a poor prognosis in liver cancers (90). Therefore, the role of Sox9 as a potential biomarker of Notch involvement and indication for Notch-targeted treatment requires additional study.

Roma et al (91) reported that pharmacodynamically active drugs, including emtricitabine/rilpivirine/tenofovir or ‘GSI’, inhibits Notch signaling, which prevents metastasis and recurrence of tumors and increases the disease-free survival time of patients. However, gastrointestinal toxicity is a primary side-effect of GSI use and these drugs are more effective against tumors that exhibit upregulated Notch signaling. Therefore, other potential modulating factors remain to be identified (92).

An increased understanding of the mechanisms involved in liver cancer proliferation and differentiation may aid the development of therapeutic strategies for liver cancer. The Notch pathway is emerging as a critical signaling pathway, which regulates cell proliferation, differentiation and necrosis associated with normal development, as well as stem cell renewal and differentiation. In conclusion, loss or disruption of Notch signaling may be a key contributing factor in bile ductular disorders, sinusoidal capillarization and the neovascularization of portal regions in the liver. To develop effective therapeutic approaches for the treatment of liver cancer, additional studies that investigate the Notch pathway are urgently required.