Article Text

Download PDFPDF

Gene of the month: E-cadherin (CDH1)
  1. Tamara M H Gall,
  2. Adam E Frampton
  1. HPB Surgical Unit, Department of Surgery & Cancer, Imperial College, Hammersmith Hospital campus, London, UK
  1. Correspondence to Adam E Frampton, HPB Surgical Unit, Department of Surgery & Cancer, Imperial College, Hammersmith Hospital campus, Du Cane Road, London W12 0HS, UK; a.frampton{at}

Statistics from

Epithelial-cadherin (E-cadherin; encoded by CDH1) is a member of the classical cadherins (the others being neural cadherin (N-cadherin) and vascular endothelial cadherin (VE-cadherin)). These single-pass transmembrane glycoproteins are expressed by a variety of tissues and are involved in Ca2+-dependent cell–cell adhesion. Initially described as liver cell adhesion molecule in chickens1 and uvomorulin in mice,2 the name E-cadherin was first used by Takeichi and colleagues in the early 1980s.3 ,4 Since then its role in normal epithelial cell architecture and tissue formation, as well as a tumour suppressor gene in cancer development and progression, has been extensively studied. Cell–cell adhesions are vital to maintain the integrity of cells and cohesion of tissues, and the control of these junctions therefore plays an important part in tumourigenesis. E-cadherin mediates cell–cell contact at the basolateral membrane in adherens junctions and its expression is the hallmark of epithelial cell layers.5 This short review focuses on the structure and function of E-cadherin at the cell junction, including the cadherin–catenin complex and its involvement in epithelial-to-mesenchymal transition (EMT). Finally, the role of E-cadherin in cancer and the therapeutic implications are discussed.

Structure and function

The CDH1 gene is located on chromosome 16q22.1, spanning a region of approximately 100 kb.6 CDH1 comprises 16 exons and 15 introns and is highly conserved between species.6 The resulting E-cadherin protein is a 120 kDa glycoprotein consisting of an extracellular domain of five tandem repeated domains, a cytoplasmic domain and a single transmembrane domain.7 ,8 The extracellular domain has binding sites for Ca2+ ions and extends from the cell surface to bind to cadherins on adjacent cells by lateral dimerisation.9 This allows a cadherin–cadherin interface and thus cell–cell adhesion. The cytoplasmic domain consists of the juxtamembrane domain (JMD) and the catenin-binding domain (CBD), each with around 30–35 residues. The JMD allows clustering of cadherins and contributes to the adhesive strength via p120-catenin (p120-ctn).10 The CBD interacts with β-catenin and γ-catenin (homologous to plakoglobin). α-catenin (homologous to vinculin) then links the bound β-catenin to the actin cytoskeleton (figure 1). This promotes protein clustering at the adherens junction and stabilises cell adhesion.11 The C-terminal of the cytoplasmic domain, consisting of around 150 residues that include the CBD, has been found to be essential in calcium-dependent cell aggregation.12 In normal adult epithelial tissue, E-cadherin mediates cell adhesion as highlighted above. In addition, the cadherin–catenin complex can activate certain signalling cascades and has an active role in EMT.

Figure 1

The E-cadherin interface mediating cell–cell adhesion. The extracellular domain binds to the extracellular domain of E-cadherin on adjacent cells by calcium activated dimerisation. The intracellular domain binds to p120ctn and to β-catenin. The complex formed with β-catenin allows α-catenin to link this complex to the cellular actin cytoskeleton.

E-cadherin activated signalling cascades

Wnt and cadherin pathways are linked by the activities of β-catenin, thereby co-ordinating gene expression and cell adhesion.13 β-catenin is one of the key regulators of the Wnt signalling pathway. This pathway involves the binding of Wnt proteins to cell-surface receptors of the Frizzled family. In response to Wnt signalling, β-catenin translocates from the cytoplasm to the nucleus. Here, it activates T-cell factor/lymphoid enhancer factor complex, resulting in the transcription of target genes. This includes the important oncogene c-MYC,14 which promotes cell migration, cell proliferation, cell growth and can also induce apoptosis in the absence of a survival signal (figure 2). The Wnt pathway has a role in embryonic development and controls processes such as cell migration, proliferation and cellular adhesion. Levels of cytoplasmic β-catenin are in turn controlled by the adherens junction and E-cadherin as detailed above. Disruption of this junction will result in altered levels of β-catenin and will thus influence the Wnt pathway. The stability of the adherens junction and the E-cadherin/β-catenin complex is modulated by its serine/threonine phosphorylation status. Serine/threonine phosphorylation of β-catenin or E-cadherin results in increased stabilisation of the cadherin–catenin complex, while tyrosine phosphorylation of β-catenin disrupts the complex, resulting in a loss of cell–cell adhesion; release of cytoplasmic β-catenin; increase in Wnt signalling; and increased tumour invasiveness.15 ,16

Figure 2

The role of β-catenin in the Wnt signalling pathway. A Wnt protein binds to a cell-surface receptor of the Frizzled family. This mediates the translocation of β-catenin from the cytoplasm into the nucleus. Here it activates the TCF (T-cell factor)/LEF (lymphoid enhancer factor) complex resulting in the transcription of target genes, such as c-MYC with its resultant actions of cell migration, cell proliferation, cell growth or conversely apoptosis. TK (tyrosine kinase) phosphorylation of the E-cadherin/β-catenin complex disrupts this junction releasing free β-catenin into the cytoplasm for involvment in the Wnt signalling cascade.

Epithelial-to-mesenchymal transition

EMT is an important embryonic process that is also seen in an uncontrolled fashion in malignant tumours.17 Polarised epithelial cells undergo multiple biochemical changes resulting in a mesenchymal cell phenotype, with an associated loss of epithelial cell proteins, the main one being E-cadherin.18 The non-polarised mesenchymal cells are highly motile and invasive.19 By making this reversible change, epithelial cells can move to a distant site and redifferentiate to form a new structure. However, in order to do this they must dissolve cell adhesions and follow a chemoattractive path through the extracellular matrix. Cancer cells that have undergone EMT gain stem cell characteristics, including self renewal and the ability to initiate new tumours, as well as increasing their resistance to chemotherapy.20 ,21 Uncontrolled EMT therefore allows tumour invasion and metastatic spread. E-cadherin is not expressed in mesenchymal cells, thus resulting in a loss of cell–cell adhesion and an increase in Wnt signalling. Moreover, it may be that a loss of E-cadherin can actually induce EMT. Its downregulation during EMT can be due to promoter methylation22 or upregulation of transcriptional repressors, including members of the SNAIL23 ,24 and ZEB gene families of zinc-finger transcription factors.25 MicroRNAs (miRNAs) are important gene regulators at the post-transcriptional level and members of the miR-200 family play a critical role in the regulation of EMT, as they target ZEB1 and ZEB2, thereby releasing the direct repression of E-cadherin.26 The re-expression of E-cadherin can reverse the transformed mesenchymal cells to an epithelial phenotype (mesenchymal-to-epithelial transition, MET),27 which facilitates tumour colonisation of secondary locations. Indeed, several reports have shown increased E-cadherin levels in metastatic deposits compared with their primary tumours.28 ,29 Indeed, E-cadherin binding activates ERK MAP kinase and Akt/PKB signalling pathways, thus providing a survival advantage for metastatic carcinoma cells in a challenging ectopic environment.30

E-cadherin in human cancers

The functional role of E-cadherin suggests that genetic and epigenetic alterations of its encoding gene, CDH1, may have a significant impact on tumour invasion and metastatic spread, with a loss or reduced expression of E-cadherin resulting in a more invasive tumour. Indeed, a number of studies have identified a reduced expression of E-cadherin, due to loss of heterozygosity (LOH) at 16q22.1, inactivating mutations, CpG-island hypermethylation of CDH1 gene promoter or silencing of CHD1 gene expression by specific transcription factors, in many epithelial tumours, including breast cancer (BC), pancreatic ductal adenocarcinoma (PDAC), gastric cancer (GC) and colorectal cancer (CRC), as well as hepatocellular carcinoma;31 sqamous cell carcinomas of the skin,32 head and neck;33 oesophageal carcinoma;34 and melanoma.35

E-cadherin in BC

There is a suggestion that a reduced expression of E-cadherin could lead to the development of BC as the loss of 16q is an early event36 and LOH at 16q is seen in up to 55% of sporadic BCs.37 A reduced expression of E-cadherin is seen in the ductal-type and even more commonly (>90%) in lobular BC.38 This loss of function may be due to homozygous deletions of CDH1, mutations of CDH1, hypermethylation of CDH1 or transcriptional inactivation.39 ,40 Invasive lobular BCs often show inactivating mutations combined with LOH of the wild-type CDH1 allele. While a more heterogeneous loss is observed in the invasive ductal type.41 Hence, loss of E-cadherin is recognised as a hallmark diagnostic feature of lobular neoplasia and invasive lobular carcinomas.42 Patients whose tumours have reduced E-cadherin expression have a worse disease-free survival (DFS)43 and overall survival (OS).44 Loss is also associated with an increased tumour size, higher histological grade and the development of distant metastases.44 Furthermore, it is an independent prognostic marker in triple negative BC.45 Interestingly, forced expression of E-cadherin in BC cells results in growth inhibition in vitro and in vivo in murine tumour models,46 perhaps suggesting a therapeutic role.

E-cadherin in pancreatic cancer

A total or partial loss of its expression has been seen in 43% of patients with PDAC.47 Its expression may be used as a prognostic indicator as it is reduced in poorly compared with moderately-differentiated adenocarcinomas48 and reduced expression is an independent predictor of OS in PDAC.48 ,49 Furthermore, a role for E-cadherin expression in chemotherapy resistance has been established in vitro, as chemoresistant cancer cells have an increased expression of ZEB1 (which represses E-cadherin and regulates EMT).50 Interestingly, by silencing ZEB1 and increasing E-cadherin expression, chemosensitivity was restored.50

E-cadherin in GC

Most studies of GC have concentrated on the familial form and its associated genetic mutations. CDH1 mutations are the most common seen in diffuse GC, detectable in about 50% of cases.51 Germline mutations of CDH1 have been reported in families susceptible to this disease52–54 and these CDH1 alterations, including LOH, point mutations and hypermethylation, have since also been found in 29% of non-familial GCs. Those with structural alterations of the CDH1 gene have poorer survival rates.55 Reduced E-cadherin expression in GC also correlates with a greater depth of invasion, increased lymph-node metastasis and more advanced stage.56 These data suggest that in some families with a predisposition to GC, E-cadherin plays a crucial role as a tumour suppressor, and loss worsens outcomes.

E-cadherin in CRC

Similar results have also been seen in CRC where a loss of E-cadherin is associated with poorer DFS.57 Interestingly, reduced expression at the invasive tumour margin is associated with low levels of miR-200c.26 Loss of E-cadherin here corresponds to high tumour budding, perineural invasion and a worse prognosis.58 Furthermore, patients that respond to chemotherapy are more likely to have retained E-cadherin expression than those who are non-responders.57 ,59

An oncogenic role for E-cadherin

While the role of E-cadherin as a tumour suppressor is well established, more recently there are some reports supporting its role as an oncogene through alternative pathways. These include inflammatory breast carcinoma, ovarian cancer and glioblastoma. For example, ovarian cancer does not normally undergo EMT, but actually shows a more epithelial phenotype in early tumour progression.60 Interestingly, E-cadherin is highly expressed in ovarian cancer, but rare in normal ovarian tissues.61 Depending on cellular context, this oncogenic effect may be related to ligand-independent EGFR activation or phosphoinositide 3-kinase (PI3K)/AKT activation.62 Thus, E-cadherin may allow some tumours to form strong cell–cell adhesions actually promoting their growth and survival.

Take home points

  • E-cadherin is an important glycoprotein necessary for effective cell–cell adhesion.

  • E-cadherin determines epithelial cell differentiation.

  • Loss of E-cadherin expression is accepted as the hallmark of the epithelial-to-mesenchymal transition, which is an essential step in the metastatic progression of human cancers.

  • E-cadherin also regulates β-catenin signalling in the canonical Wnt pathway.

  • E-cadherin is traditionally known as a tumour supressor gene and reduced expression is seen in the vast majority of epithelial cancers, promoting tumour invasiveness and leading to worse patient prognosis.


View Abstract


  • Contributors Both authors contributed to the design and writing equally.

  • Competing interests None.

  • Provenance and peer review Commissioned; internally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.