Potential conflict of interest: Nothing to report.
Pre-“EMT”ing key processes in liver carcinogenesis: Growing evidence for how malignant hepatocytes invade and conquer†
Article first published online: 25 JUN 2010
Copyright © 2010 American Association for the Study of Liver Diseases
Volume 52, Issue 1, pages 384–388, July 2010
How to Cite
Teoh, N. C. (2010), Pre-“EMT”ing key processes in liver carcinogenesis: Growing evidence for how malignant hepatocytes invade and conquer. Hepatology, 52: 384–388. doi: 10.1002/hep.23777
- Issue published online: 25 JUN 2010
- Article first published online: 25 JUN 2010
Chen L, Chan THM, Yuan Y-F, Hu L, Huang J, Ma S, et al. CHD1L promotes hepatocellular carcinoma progression and metastasis in mice and is associated with these processes in human patients. J Clin Invest 2010;120:1178-1191. (Reprinted with permission.)
Chromodomain helicase/ATPase DNA binding protein 1–like gene (CHD1L) is a recently identified oncogene localized at 1q21, a frequently amplified region in hepatocellular carcinoma (HCC). To explore its oncogenic mechanisms, we set out to identify CHD1L-regulated genes using a chromatin immunoprecipitation–based (ChIP-based) cloning strategy in a human HCC cell line. We then further characterized 1 identified gene, ARHGEF9, which encodes a specific guanine nucleotide exchange factor (GEF) for the Rho small GTPase Cdc42. Overexpression of ARHGEF9 was detected in approximately half the human HCC samples analyzed and positively correlated with CHD1L overexpression. In vitro and in vivo functional studies in mice showed that CHD1L contributed to tumor cell migration, invasion, and metastasis by increasing cell motility and inducing filopodia formation and epithelial-mesenchymal transition (EMT) via ARHGEF9-mediated Cdc42 activation. Silencing ARHGEF9 expression by RNAi effectively abolished the invasive and metastatic abilities of CHD1L in mice. Furthermore, investigation of clinical HCC specimens showed that CHD1L and ARHGEF9 were markedly overexpressed in metastatic HCC tissue compared with healthy tissue. Increased expression of CHD1L was often observed at the invasive front of HCC tumors and correlated with venous infiltration, microsatellite tumor nodule formation, and poor disease-free survival. These findings suggest that CHD1L-ARHGEF9-Cdc42-EMT might be a novel pathway involved in HCC progression and metastasis.
The development of hepatocellular carcinoma (HCC) in a chronically diseased liver, often referred to as hepatocarcinogenesis, is a multistep process characterized by the progressive accumulation and interplay of genetic alterations causing aberrant growth and malignant transformation of liver parenchymal cells, followed by vascular invasion and metastasis. Molecular analyses from tumor samples, either in the form of unbiased approaches using expression profiling strategies, or more targeted approaches to elicit specific alterations in signaling pathways, have been intensely pursued in an attempt to identify key processes contributory to HCC, and to stratify individuals at high risk of recurrence.
Cirrhosis, the setting for the majority of HCCs, is characterized by nodules of proliferating and apoptotic hepatocytes trapped within dense fibrosis. The risk of HCC with hepatitis B (HBV) and hepatitis C virus infection is much higher when chronic hepatitis is present, demonstrated by outcome studies after antiviral therapy and in HBV transgenic mice.1, 2 The contribution of inflammation to hepatocarcinogenesis was recently demonstrated in human HCC tissue,3 as well as in mice exposed to diethylnitrosamine.4 Although it remains unclear whether replacement by stem cell–like precursor cells leads to HCC, or whether damaged hepatocytes themselves revert to de-differentiated, autopropagative tumor cells, what is beyond question is that >90% of HCCs exhibit chromosomal abnormalities.5-7 In human HCC, chromosomal abnormalities include 1, 4q, 8, 9p, 11, 13, 16q, and 17p.4, 5-7 Although viral DNA sequences in HBV-associated HCCs integrate at multiple sites to cause chromosomal rearrangements, predominantly deletions,5, 8 some may induce amplification of genes. Amplification of 1q21 is a frequently reported genetic aberration in HCC, and it has been associated with the metastatic potential of tumor cells.7, 9
In the April 2010 issue of the Journal of Clinical Investigation, Chen et al. report the significance of chromodomain helicase DNA binding protein 1-like (CHD1L), an oncogene isolated from the 1q21 amplicon, in hepatocarcinogenesis in a culmination of elegant translational studies defining the role of CHD1L in tumor cell migration, invasion, and metastasis by enhancing cell motility, inducing filopodia formation and epithelial-mesenchymal transition (EMT).10 Chromodomain helicase/adenosine triphosphatase DNA-binding protein 1–like gene, CHD1L, belongs to the sucrose nonfermenting 2 (SNF2)-like family of proteins which participate in various nuclear activities such as transcriptional activation, repression, DNA repair, and homologous recombination.11 The authors previously described spontaneous liver tumor development in a mouse CHD1L transgenic model.12 They now use a chromatin immunoprecipitation-based cloning strategy to first identify the genes potentially regulated by CHD1L, and found ARHGEF9 (Rho guanine nucleotide exchange factor 9) to be of interest for its ability to activate the Rho family of small guanosine triphosphatases (GTPases).10, 13 Rho GTPases function as molecular switches, cycling between inactive and active guanosine diphosphate–bound states, to regulate actin cytoskeleton scaffolds within cells and to modulate the interaction of cadherin-dependent adherens junctions between cells.13, 14 During tumor progression, invasion, and metastasis, changes in the activity of Rho GTPase and reorganization of the actin cytoskeleton occur with the loss of normal cell–cell polarity and contact.13 Among the 20 genes encoding the different members of the Rho family, cell division cycle 42 (Cdc42) has been well studied for its role in actin cytoskeleton modification. Rearrangement of the actin cytoskeleton promotes cancer progression via the acquisition of migratory and invasive properties of dissociated cells, thereby enabling them to actively pass through the basement membrane, into blood vessels, and traverse to distant sites.14 Because ARHGEF is a guanine nucleotide exchange factor (GEF) specific for Cdc42, Chen and coworkers proceeded to show that CHD1L activates Cdc42 via enhanced ARHGEF expression in HCC cell lines.10 Notably, in primary HCC specimens derived from patients of Chinese ethnicity, they confirmed that up-regulation of ARHGEF significantly correlated with CDH1L overexpression in >50% tumors (n = 35). Although it was not stipulated, it is presumed these HCCs were HBV-related. Using in vitro and in vivo assays, they then showed CHD1L overexpression promoted HCC cell motility, tumor invasion, metastasis, and intriguingly, induced filopodia formation and EMT.10 The epithelial markers E-cadherin, α-catenin and β-catenin were down-regulated, whereas the mesenchymal markers N-cadherin, vimentin, and α-smooth muscle actin were enhanced in CHD1L-transfected cells, and in tumors induced by CHD1L-expressing cells in nude mice.10 Moreover, silencing of CHD1L inhibited the EMT phenotype, invasiveness, and tumorigenic ability of HCC cells in such animals. These studies were further strengthened by confirmation of the functional effects of the CHD1L-ARHGEF9-Cdc42 pathway in 16 of the 35 primary human HCCs, as well as the presence of EMT by E-cadherin and vimentin protein expression in a smaller subset of tumors.10
The conversion of an epithelial cell to a mesenchymal cell is critical to metazoan embryogenesis and is a defining structural feature of organ development.15 EMT is a multistep process beginning with well-polarized and adhesive epithelial cells leading to the formation of diassembled, motile mesenchymal cells. Of note, mesenchymal migration is mechanistically different from epithelial movement. Epithelial cells move en bloc, whereas mesenchymal migration is more dynamic (and/or haphazard). The fundamental difference of such movements in embryogenesis is that they are subtle and controlled, whereas in tumorigenesis, migration is aggressive and unchecked (Fig. 1).16 Turning an epithelial cell into a mesenchymal cell requires alteration in morphology, cellular architecture, adhesion potential, and migratory ability. Commonly used molecular markers for EMT include increased expression of N-cadherin, vimentin, nuclear localization of β-catenin, increased production of transcription factors such as Snail1 (Snail), Snail2 (Slug), Twist, and E47, which inhibit E-cadherin.15-20 Phenotypic markers for EMT include increased migration capacity, three-dimensional invasion, and resistance to anoikis/apoptosis.15, 16 In order to investigate the significance of EMT in HCC, and its impact on metastatic potential and prognosis, Chen et al. probed a different set of 50 primary HCCs and their paired metastatic tumors.10 They found a significantly greater proportion of metastatic HCCs (68%) expressed CHD1L compared to primary HCCs only, but also CHD1L overexpression was associated with venous invasion, microsatellite tumor formation, and poorer survival. Interestingly, the leading edge of HCCs and cells invading into surrounding tissue and blood vessels stained strongly positive for CHD1L (Fig. 1).
This discovery bolsters mounting evidence that implicates EMT in liver carcinogenesis. Twist and Snail have been found to be expressed in up to 70% of human HCC and correlate with adherens junction dissociation and poor prognosis, whereas ectopic expression of these genes in HCC cell lines can increase cell motility and invasiveness.21-23 The hepatitis B X (HBx) gene may also induce EMT in cultured hepatocytes by signal transducer and activator of transcription 5b activation.24 The wingless-type (Wnt)/β-catenin and Hedgehog signaling pathways, which are essential for stem cell function and early development, also play a major role in EMT and cancer progression.25 Studies have identified a possible role for the Hedgehog pathway in HCC with expression of Sonic in up to 60% of human HCC samples, with concomitant down-regulation of Gli-related targets.26, 27 Wnt signaling intermediates have also been shown to be up-regulated in HCCs, with more than 10 studies demonstrating β-catenin mutations.28 Others have also reported that preneoplastic lesions with β-catenin activation have a higher risk of malignant transformation than their counterparts, with β-catenin mutation rates ranging from 0%-45% in some tumors.27-30
In contrast, the concept of EMT in liver fibrosis remains controversial and has been the subject of intense study.31 EMT provides an attractive potential mechanism by which a large number of cells of fibrogenic capacity can be mobilized, following injury, to signal and invoke wound healing. Profibrogenic factors present in cirrhotic liver, such as transforming growth factor-β, can induce primary hepatocytes and cholangiocytes to undergo EMT.31, 32 Also, various markers of EMT have been detected in liver from patients with cirrhosis and in animal models of hepatic fibrosis.31 Hence, this begs the question: could hepatocytes under siege in a milieu of chronic inflammation, oxidative stress, and cell death be driven to EMT to survive and perpetuate? The in vivo evidence for hepatocyte EMT comes from the study by Zeisberg et al.32 The expression of fibroblast-specific protein 1 (FSP-1) in cells with epithelial markers has been widely used to define EMT in vivo. Using a murine CCl4-induced fibrosis model whereby hepatocytes were irreversibly tagged with β-galactosidase (Albumin-Cre; Rosa26-stop-β-gal), they observed that a significant population of hepatocyte-derived cells expressed FSP-1. However, more recent work by Taura et al., which focused on collagen synthesis as a primary marker of EMT in hepatocytes, dispute the above findings.33 They subjected triple transgenic mice (Rosa26-stop-β-gal;Albumin-Cre;ColI-green fluorescent protein) to CCl4 treatment; such mice permanently, and heritably, expressed β-galactosidase but also had type I collagen-expressing cells labeled with green fluorescent protein. Although hepatocytes derived from these animals assumed fibroblast morphology upon transforming growth factor-β1 stimulation, they did not produce type I collagen, nor did they express either α-smooth muscle actin or FSP1.
In sum, although the work by Chen et al. is of major clinical relevance, the true impact of their observations await further validation in other HCC patient cohorts that are geographically and etiologically distinct. Although they have meticulously and mechanistically demonstrated the importance of the CHD1L pathway in advanced and metastatic HCCs, what remains unresolved are (1) the role of preneoplastic cells and EMT in the “at risk” cirrhotic liver, nor (2) have they addressed the pathogenic contributions of CHD1L and involvement of EMT as key steps early in hepatocarcinogenesis. Regardless, this exciting discovery may serve well in predicting clinical outcome, stratifying the prognostic variables and different biological behavioral profiles associated with advanced HCC.