Benhamouche S, Curto M, Saotome I, Gladden AB, Liu CH, Giovannini M, et al. Nf2/Merlin controls progenitor homeostasis and tumorigenesis in the liver. Genes Dev 2010;24:1718-1730. (Reprinted with permission.)
The molecular signals that control the maintenance and activation of liver stem/progenitor cells are poorly understood, and the role of liver progenitor cells in hepatic tumorigenesis is unclear. We report here that liver-specific deletion of the neurofibromatosis type 2 (NF2) tumor suppressor gene in the developing or adult mouse specifically yields a dramatic, progressive expansion of progenitor cells throughout the liver without affecting differentiated hepatocytes. All surviving mice eventually developed both cholangiocellular and hepatocellular carcinoma, suggesting that Nf2−/− progenitors can be a cell of origin for these tumors. Despite the suggested link between NF2 and the Hpo/Wts/Yki signaling pathway in Drosophila, and recent studies linking the corresponding Mst/Lats/Yap pathway to mammalian liver tumorigenesis, our molecular studies suggest that Merlin is not a major regulator of YAP in liver progenitors, and that the overproliferation of Nf2−/− liver progenitors is instead driven by aberrant epidermal growth factor receptor (EGFR) activity. Indeed, pharmacologic inhibition of EGFR blocks the proliferation of Nf2−/− liver progenitors in vitro and in vivo, consistent with recent studies indicating that the NF2-encoded protein Merlin can control the abundance and signaling of membrane receptors such as EGFR. Together, our findings uncover a critical role for NF2/Merlin in controlling homeostasis of the liver stem cell niche.
Scientific and medical literature on liver regeneration often mentions the Greek god Prometheus, who was chained to a rock in the Caucasus; there, on a daily basis, his liver was devoured by an eagle, and then it grew back every night. Thus, the liver is the only internal human organ with the unique characteristic of natural regeneration: after an injury, as little as 25% of the remaining liver is sufficient for the complete recovery of the liver mass. This ability of the liver is predominantly due to either hepatocytes entering the cell cycle or hepatic oval cells (OCs), which can differentiate into hepatocytes or cholangiocytes. As shown in Fig. 1, together with bone marrow cells, hepatocytes and OCs are sources of liver progenitor or stem cells. However, the exact origin of OCs is a matter of debate; some authors have suggested that OCs arise from unidentified intrahepatic stem cells1 or from the hematopoietic system.2 Lately, studies using label retention have supported the idea that OCs arise from intraductal and periductal locations within the most proximal branches of the biliary tree.3
The course of hepatocarcinogenesis can last longer than 30 years after first diagnosed with hepatitis B or hepatitis C virus. This multistep disease starts with concomitant cirrhosis and chronic hepatitis to finally develop tumorigenesis, which subsequently induces liver regeneration via OC activation.4 During the 1990s, evidence of a linkage between OCs and hepatocellular carcinoma (HCC) increased with experimental proof. Dunsford and Sell5 and Hixson et al.6 attempted to analyze the phenotypic relationships between OCs, bile duct cells, and adult and fetal cells. They found that OCs, preneoplastic foci, early tumor nodules, and primary HCC express both OC and hepatocyte antigens. This suggests a cause-effect relationship between OCs and HCC. Their results corroborated the idea that OCs appear and proliferate in the liver as previously reported by Farber7 and Hewitt8 in the late 1950s, and they were confirmed by Dumble et al.9 nearly a half-century later.
Rodent models of liver tumorigenesis have been based on chemical induction, which yields HCC almost exclusively and cholangiocarcinoma (CC) only rarely. Unfortunately, animal models of CC have been limited primarily to the Syrian hamster model, murine models of gallbladder adenocarcinoma, and the administration of furan to rats.10 Thus, liver-specific neurofibromatosis type 2 (Nf2−/−)–deleted mice11 not only represent an excellent model of liver tumorigenesis for both HCC and CC but also offer an excellent tool for studying the involvement of OCs in liver malignancies. These mice develop a great variety of histopathological types of HCC (including trabecular, solid, pseudoductular, and acinar HCC) and early CC that resemble human tumorigenesis.11
NF2 is an inherited disorder characterized by the development of Schwann cell tumors of the vestibulocochlear nerve. Several tumors of the nervous system, including schwannomas, meningiomas, and ependymomas, have been associated with mutations in the NF2 locus.12 The NF2 gene codes for a 595–amino acid protein called Merlin; Merlin is highly related to the ezrin, radixin, and moesin proteins, which are actively involved in the regulation of the cytoskeleton and signal transduction pathways.13 Merlin caught the attention of cancer researchers because it was found to be a negative regulator of the Hippo/Warts/Yorkie tumor suppressor pathway in Drosophila. However, the function of Merlin in the regulation of the analogous macrophage stimulating 1 (Mst)/large tumor suppressor (Lats)/yes-associated protein (Yap) pathway in mammals is not clear yet.14
In the actual study, McClatchey's group11 used different experimental approaches to investigate whether NF2/Merlin regulates Mst/Lats/Yap. They observed that the absence of NF2/Merlin does not change the phosphorylation, localization, or expression of Yap1-related genes after the endogenous or exogenous administration of Merlin or short hairpin RNA knockdown in liver-specific NF2-deleted OCs.15 These results strongly support the notion that NF2/Merlin is not a major regulator of Yap1; thus, an NF2 deficiency in the liver is not sufficient to inhibit Yap1 activation in the liver. Furthermore, these data open a new line of research on the inactivation of the Mst/Lats/Yap pathway in the liver that is parallel to but distinct from previous reports.16
Indeed, strong evidence suggests that NF2/Merlin functions as a tumor suppressor by blocking epidermal growth factor receptor (EGFR)–dependent signaling.17 Curto et al.18 previously showed that overproliferation of Nf2−/− cells in vitro and in vivo is EGFR-dependent. Consistently, Benhamouche et al.11 found that the pharmacological inhibition of EGFR by erlotinib in liver-specific NF2-deficient mice caused reductions in the lesion size, liver/body weight ratio, cell proliferation, and EGFR targets. The outcomes of these experiments were consistent with a great number of studies indicating a role for EGFR signaling in OC proliferation and liver tumorigenesis in mice and humans.19
The histological and anatomical observations of liver-specific NF2-deficient mice agree with the involvement of NF2/Merlin in the proliferation of facultative OCs. Liver-specific NF2-deleted mice exhibit extensive hyperplasia of facultative OCs. These OCs originate from the portal tracts, progressively infiltrate the surroundings, and thus compromise the normal architecture of the liver. As a result of hepatomegaly-derived ascites, the mice die at approximately 30 weeks of age. However, the mice that outlive this time barrier represent an animal model important for studying not only the development of OCs but also the development of HCC and CC in the same liver.
Next, Benhamouche et al.11 addressed the crucial problem of defining the cells that initiate HCC growth. Several studies have documented that HCC develops from OCs.20 Therefore, to confirm this hypothesis, these authors performed partial hepatectomy in two different experimental models with conditional NF2-knockout mice. They either deleted NF2 from the normal adult liver after an infection with a Cre recombinase-expressing adenovirus or stimulated Cre recombinase under the control of the interferon-responsive Mx1 promoter mice with polyinosinic:polycytidylic acid in order to achieve interferon-dependent deletion. Both models resembled the histological features of liver-specific NF2-deleted mice with OC hyperplasia and the subsequent development of HCC and CC. Thus, partial hepatectomy triggers the overproliferation of Nf2−/− cells, and this is consistent with the role of NF2/Merlin in the down-regulation of epidermal growth factor.
In summary, the main take-home messages of Benhamouche and colleagues' work11 are as follows. First, NF2/Merlin plays an important role in the initial establishment of the liver progenitor niche both in intercellular communication and in growth factor signaling. Second, NF2/Merlin in the liver appears to be independent of the Mst/Lats/Yap pathway, although more in-depth studies are needed because this relationship remains unclear (Fig. 1). Third, the liver-specific NF2 deficiency provides a unique animal model for studying the homeostasis of liver progenitor cells. Indeed, NF2/Merlin appears to directly control liver progenitor proliferation and neoplastic transformation. Lastly, this work provides evidence that the deletion of a single gene is sufficient to activate the proliferation and development of both embryonic and adult liver progenitor cells and thus reproduce the two major forms of liver cancer: HCC and CC. This raises the interesting possibility of analyzing and associating human mutations in the NF2 gene with liver tumorigenesis with the goal of gene-based treatment options.