Signaling pathways involving the Notch receptor system play a critical role in mediating cell-cell interactions which determine cell fate and organ development in a wide variety of organisms, including insects, nematode worms, echinoderms, and mammals.1, 2 Signaling is activated when Notch receptors, which are present as transmembrane proteins bind to ligands, also present as transmembrane proteins on adjacent cells. In humans, four Notch receptors (Notch-1, -2, -3, and -4) and five ligands (Jagged-1 and -2, and Delta-1, -3, and -4) have been identified.3 Notch signaling can also modulate apoptosis and cell proliferation.1, 4
Notch receptor and ligand proteins are constitutively expressed in various human tissues, including the liver. In human fetal liver, interactions between Jagged-1, expressed on the embryonic ductal plate, and Notch-3, expressed on the adjacent portal tract mesenchyme and hepatic arterial endothelium, are thought to play a role in the process of ductal plate remodeling and the subsequent development of the intrahepatic biliary tree.5 This suggestion is supported by the discovery that mutations of the Jagged-1 gene are present in the majority of patients with Alagille syndrome (AGS),6, 7 a disease characterized by defective development of interlobular bile ducts. In the normal human liver, expression of Notch-1, -2 and -3 receptors and Jagged-1 ligand has been observed to varying degrees on bile ducts, hepatic artery and/or portal vein endothelium, and hepatocytes.8–10 Altered expression of Notch receptor/ligand genes and proteins has also been described in a range of pediatric and adult liver diseases,5, 9–12 including expression on reactive bile ductules and small neovessels, both of which are thought to be important in the pathogenesis of the progressive periportal fibrosis that occurs in chronic cholestatic liver diseases associated with bile duct loss.
In experimental studies using mouse knockout models, Notch-1, Notch-2, and Jagged-1 have all been shown to be essential for normal embryonic development. Animals in which expression of these genes is disrupted die in utero, frequently from vascular defects.13–16 The development of suitable animal models for investigating defects in Notch signaling has thus been difficult. One notable exception is a mouse model of AGS in which animals doubly heterozygous for Jagged-1 and Notch-2 mutations developed a syndrome closely resembling AGS, including defective development of intrahepatic bile ducts.17 Jagged/Notch interactions have also been implicated in bile duct development in the normal mouse fetal liver, with upregulation of Notch-2 on hepatoblasts being associated with differentiation toward a biliary phenotype.18–20
Thus, while there is increasing evidence to suggest that the Notch signaling pathway is important in the development of intrahepatic bile ducts and blood vessels in health and disease, less is known about the role of the Notch signaling pathway in hepatocyte development and homeostasis, although in one study downregulation of Notch-2 expression on fetal mouse hepatoblasts appeared to be associated with hepatocellular differentiation.19
The paper by Croquelois and colleagues in this issue of HEPATOLOGY describes interesting novel observations, which provide a further insight into our understanding of the complex and important Notch signaling pathway.21 A previously developed mouse model was used in which inactivation of Notch-1 can be conditionally induced postnatally in mice with a defective but non-lethal Notch-1 mutation.22 In contrast to the Jagged-1/Notch-2 double heterozygous mouse model of AGS in which intrahepatic bile duct development was defective,17 bile ducts in the conditional Notch-1–deficient mice were well preserved and displayed a normal proliferative capacity in response to partial hepatectomy and bile duct ligation. Surprisingly, however, there was an 8-fold increase in hepatocellular proliferation in the Notch-1–deficient mice compared with normal controls, resulting in a 40% increase in liver mass relative to body weight. This was associated with downregulation of Notch-1 expression on hepatocytes and with histological features of nodular regenerative hyperplasia (NRH).
It is not clear to what extent the model developed in the study of Croquelois and colleagues can be applied to understanding the role of Notch signaling mechanisms in the pathogenesis of human liver disease. No liver diseases have been specifically linked to mutations in the Notch-1 receptor. The postnatal conditional inactivation of Notch-1 receptor expression used in the study of Croquelois et al. is probably more analogous to an acquired rather than a hereditary defect in Notch signaling, for which other epigenetic regulatory mechanisms should be considered.
In human liver disease, NRH is generally considered to be a response to vascular problems, particularly portal venous insufficiency.23 Notch signaling is known to be important in vascular morphogenesis,24, 25 and vascular anomalies been implicated in the hepatic and extra-hepatic manifestations of AGS.26, 27 Vascular abnormalities in the form of focal sinusoidal dilatation and large vascular spaces abutting directly upon hepatic parenchyma were frequent findings in the study of Croquelois et al. However, detailed studies, including sectioning of the whole liver in one case, failed to reveal any abnormalities of the portal veins or hepatic arteries. These observations suggest that Notch-1 activation may have a direct role in inhibiting hepatocyte proliferation. In contrast, other studies have suggested that Notch signaling may play a role in promoting rather than inhibiting hepatocellular regeneration and proliferation.28, 29 Hepatocyte proliferation following partial hepatectomy was reduced in the conditional Notch-1–deficient mice compared with control animals. These apparently conflicting findings suggest that other factors, including the overall liver mass, are important in regulating the regenerative response of the liver.
The notch signaling pathway involves four main stages.3, 30, 31 The first is the cell-cell surface interaction between Notch receptor and its corresponding ligand. This results in proteolytic cleavage of the Notch intracellular domain (NIC), which translocates to the nucleus. Within the nucleus NIC binds to transcription factors that control the activation and repression of Notch target genes. Finally, the signaling mechanism is switched off by a process involving ubiquitination and proteosome-dependent degradation of NIC. The factors controlling Notch signaling are complex and tightly regulated and may act at any of the four stages referred to above.3 Intracellular regulatory proteins include Numb, LNX, Itch, and SEL-10, which target cytoplasmic or nuclear NIC for ubiqutination and proteosomal degradation. Others such as Deltex and MINT are involved with transcriptional regulation. Extracellular modulation of Notch signaling is carried out by fringe proteins, including Lunatic Fringe (LFng), Manic Fringe (MFng) and Radical Fringe (RFng)—these are glycosyltransferases that regulate activation of Notch receptors by Notch ligands. The extracellular domains of the Notch receptor and ligand transmembrane proteins also have several epidermal growth factor-like repeats. The cytokine tumour necrosis factor (TNF) has been shown to induce the expression and nuclear translocation of Notch-1 in synovial fibroblasts.32 Cytokines and growth factors may therefore play a role in regulating Notch signaling. Conversely, signaling pathways involving Jagged-2 and Notch-1 have been implicated in the production of cytokines and growth factors, including interleukin-6 (IL-6), vascular endothelial growth factor (VEGF) and insulin-like growth factor-1 (IGF-1) by multiple myeloma-associated stromal cells.33 A number of these cytokines and growth factors (IL-6, TNF-alpha, IGF-1) are involved in regulating hepatocellular regeneration.34 Furthermore, transgenic mice expressing high levels of IL-6 develop nodular hepatocellular lesions closely resembling NRH.35 The results of these studies again suggest a potential role for Notch signaling in promoting rather than inhibiting hepatocellular proliferation, although the relevance of findings obtained from nonhepatic stromal cells is uncertain in this regard.
The study of Croquelois and colleagues may also have implications for understanding the pathogenesis of nodular hepatocellular lesions, including hepatocellular carcinoma (HCC), arising in livers without cirrhosis. One previous study has suggested that nodular regenerative hyperplasia (NRH) may be a precursor lesion in some cases of HCC developing in livers without cirrhosis.36 Notch signaling has the potential to influence cell proliferation and cell death by a number of pathways related to the cell cycle and apoptosis.37 Changes in Notch receptor expression have been observed in a number of cancers, including neoplasms of the breast, testis, skin, cervix and hematolymphoid system.38–43 For some tumors (e.g., T-cell acute lymphoblastic leukemia) Notch expression appears to have an oncogenic role, whereas for others (e.g., basal and squamous cell carcinomas of skin) it appears to have a tumor suppressor function. Little is known about the role of Notch signaling in hepatocellular neoplasms. One study showed that overexpression of Notch-1 was able to inhibit the growth of HCC cells in vitro and in vivo by downregulating a number of pathways involved in cell cycling and upregulating pathways involved in mediating apoptosis.44 The study by Croquelois et al. also suggests that expression of Notch-1 on hepatocytes has an antiproliferative function, although as indicated earlier other studies have produced apparently conflicting findings.28, 29 If Notch-1 is indeed a tumor-suppressor gene for HCC, this raises the possibility of developing novel therapeutic strategies targeted at promoting hepatocellular expression of Notch-1. However, it should be noted that the majority of HCCs arise in a background of liver cirrhosis where there are likely to be other abnormalities involving cell growth and differentiation.
In summary, the work presented by Croquelois and colleagues provides an interesting novel insight into the possible role of Notch signaling in regulating hepatocellular proliferation in health and disease. Their findings need to be interpreted in the light of other studies suggesting a conflicting role for Notch signaling in promoting rather than inhibiting hepatocellular proliferation. These apparently conflicting findings may well reflect the complexity of the mechanisms involved in Notch signaling and hepatocellular proliferation. Thus, further work is required to investigate the interactions between factors involved in Notch signaling and the many other factors that regulate hepatocyte proliferation and promote hepatocellular oncogenesis. Account should also be taken of other mechanisms involved in causing chronic liver inflammation and fibrosis, in many cases leading to cirrhosis, which remains the main risk factor for the development of HCC.