Claudin-1 as a promoter of EMT in hepatocellular carcinoma

Article metrics


Claudins are integral structural and functional components of apical cell adhesions (tight junctions). Loss of such adhesions has been associated with malignant transformation, a process most often accompanied by a concomitant loss of claudin expression. A growing body of evidence reveals the highly contextual upregulation of claudin expression in certain cancer types, and moreover their relevance in promoting cancer cell invasion and metastatic progression. In this issue of Oncogene, Suh et al. reported on claudin-1 expression in hepatocellular carcinoma (HCC), including its role as a promoter of the epithelial−to−mesenchymal transition via the c-Abl/Raf/Ras/ERK signaling pathway. Considering the limited therapeutic options in HCC, evaluation of its role as a target merits further investigation.


Hepatocellular carcinoma (HCC) is the fifth most common malignancy and the third most common cause of cancer-related deaths globally. The treatment of HCC remains a challenge, with low survival rates of 20 and 5% for the 1- and 3-year survival, respectively. As in other cancer types, metastases are responsible for more than 90% of HCC-associated mortality.1 Suh et al.2 report that claudin-1 overexpression in HCC facilitates acquisition of invasive molecular changes consistent with epithelial−to−mesenchymal transition (EMT); in doing so they propose that targeting claudin-1 here could represent a new therapeutic opportunity.

Claudins are obligatory constituents of tight junctions, which also contain occludins and junctional adhesion molecules. The relevance of tight junctions exceeds that of mere ‘mechanical sealing points’ between adjacent cells. Tight junctions are gatekeepers of tissue integrity and cellular polarity by maintaining the differential composition of the apical and basolateral cellular domains. More recently, it has become apparent that tight junction molecules are involved in cellular signaling affecting proliferation, motility and invasion.3 This occurs through their ability to recruit signaling proteins, generating contact between the extracellular milieu, intracellular signaling pathways and the cytoskeleton. The claudin family consists of 27 tetra-span, trans-membrane proteins (20–27 kDa), each composed of four membrane spanning domains, two extracellular loops and a short carboxyl intracellular tail.4

Claudins interact with each other through homo- and heterophilic interactions, regulating the charge and size of molecules that pass through the paracellular spaces. It has been shown that stimulation by growth factors (epidermal growth factor and transforming growth factor-β), as well as β-catenin/Tcf signaling, regulate the expression of various claudins. Also, through the E-box elements in their promoter region, claudins fall under transcriptional control by a family of zinc-finger repressors, Slug/Snail. Claudin-1 can be phosphorylated by MAPK at Thr203, and by protein phosphatase PP2A, which enhances its barrier function via facilitating integration into tight junctions, while the WNK4 kinase phosphorylates Claudin-3 and Claudin-4 decreasing tight junction function5 (Figure 1).

Figure 1

Claudins are expressed at the site of tight junctions localized on the apical cell membrane. During cancer development/ progression, tight junctions are often lost, or claudins translocate to other cellular compartments. This facilitates malignant transformation and EMT. Boxes indicate claudin interactions and regulators.

Disruption of tight junctions, and thereby the expression and localization of claudins, has a causal rather than consequential role in malignant cellular transformation, and thereby in cancer formation and progression. Early research predominantly identified claudins as tumor suppressors in human malignancies. To that effect, claudin-1 is reduced in basal-like breast cancers known to carry particularly poor prognosis and tendency to display chemo-resistance, as well as in colon cancer. Claudin-7 is downregulated in invasive breast cancer and in head and neck squamous carcinomas. So far, the general consensus has been that claudin expression would decrease during tumorigenesis because tight junctions are lost. This is now challenged and it has become apparent that the qualitative and quantitative deregulation of claudins in cancer is highly contextual, tissue- and disease-stage specific. Claudin-1 overexpression has thus far been observed in colon-, nasopharyngeal, oral squamous cell- and ovarian cancer. More recently, a novel claudin-1-high group of mammary carcinomas has been identified, attributing a dual, tumor suppressor and oncogenic role to claudin-1 in breast cancer.6 Claudin-2 is overexpressed in colorectal-, fibrolamellar HCC, gastric- and intestinal primary tumors, and interestingly, its abundance is highly specific for breast cancer derived liver metastasis compared with bone and lung secondary deposits.7 Claudin-3 and -4 are frequently elevated in pancreatic ductal adenocarcinoma, prostate, uterine, ovarian and breast cancer. Claudin-7 abundance has been noted in prostate, ovarian and renal cancer.3

The authors of the current paper, have previously attempted to elucidate the oncogenic signaling mechanism of claudins in liver cancer. To that effect they have established that claudin-1 overexpressing HCC cells (SNU-354, -423 and -449) display induced matrix metalloproteinase- 2 activity and increased cell invasion and migration compared to normal liver Chang cells and other HCC cell lines (SNU-398, -475) that express relatively low levels of claudin-1. Their previous work implicated both c-Abl and PKCδ in this claudin-mediated process.8 Now, they elegantly consolidate their own findings with the previously established oncogenic contribution of the Ras/Raf/MEK/ERK in HCC and yield an in depth mechanistic profile of claudin-1 driven pro-metastatic signaling in liver cells.

The authors exploit several cellular models to demonstrate their findings, namely normal liver cells with stable overexpression of exogenous claudin-1, and HCC cells with endogenously high claudin-1 (SNU-354, -423 and -449). They note a decrease in expression of membrane β-catenin and E-cadherin with a concomitant increase of the mesenchymal cell markers N-cadherin, Vimentin, Slug and Zeb1. In an opposing experiment, small interfering RNA targeting of CLD1 in claudin-1-expressing normal liver cells restored the expression levels of E-cadherin, N-cadherin and Vimentin to the levels of parental normal liver cells. Furthermore, they noted increased levels of the active form of Ras in the presence of claudin-1, as well as accumulation of the phosphorylated ERK1/2. They also reiterate their original observation of increased c-Abl protein levels in the context of claudin-1 expression. To place claudin-1 in the c-Abl-Raf-Ras-ERK axis, they systematically inhibited each level of the signaling cascade using small interfering RNA and relevant molecular inhibitors, to ultimately deduce that claudin-1 sits upstream of c-Abl/Raf/Ras/ERK, as a required trigger for EMT in liver cells. Interestingly, in liver cells, claudin-1 is not affected by modulation of any of Snail, Slug or Zeb1; Snail is not required and it does not mediate claudin-1 proinvasive effects in HCC. Of note is the absence of morphological transformation in claudin-1 over-expressing liver cells, despite the obvious activation of the EMT program, suggesting that a contributing factor may be required for complete phenotypical, malignant cellular reprograming.

These in vitro data were corroborated by analysis of the human HCC tissue and establishing overexpression of claudin-1 in HCC (n=98) compared with normal liver tissue. Moreover, there appears to be a direct correlation of high claudin-1 and Slug levels (P<0.05) in these specimens. Intriguingly, localization of claudin-1 in HCC is observed at the cell membrane and to an even greater extent in the cytoplasm. This supports the hypothesis of cellular translocation of claudin-1 from the site of tight junctions. Similar observations were noted previously in primary colon carcinomas that were associated with the loss of APC tumor suppressor function and activation of nuclear β-catenin, and in respective metastatic lesions. Both showed enhanced and non junctional claudin-1 staining, localized largely in the cytoplasm and even the nucleus.9 Therefore, cytoplasmic localization of claudin-1 could be an event which facilitates its oncogenic role.

Key signal transduction pathways have already been implicated in the pathogenesis of HCC, including epidermal growth factor receptor, vascular endothelial growth factor, platelet-derived growth factor, insulin-like growth factor, Wnt/β-catenin and phosphatidylinositol-3-kinase. Importantly, activation of the Ras pathway was observed in 100% of HCC specimens analyzed when compared with non-neoplastic surrounding tissue and normal livers. Furthermore, ERK activation is a known predictor of poor survival in HCC. For this reason, an effective blockade targeting Raf/MEK/ERK pathway using small molecules is being considered for the treatment of HCC.10

In support of the hypothesis that claudins are emerging targets for cancer treatment, the Clostridium perfringens enterotoxin is a known claudin-targeting toxin that has demonstrated antiproliferative effects when used in xenograft models of primary pancreatic cancer and breast cancer derived brain metastasis.11 In this issue of Oncogene, the work of Suh et al. identifies claudin-1 as a relevant activator of the c-Abl/Raf/Ras/ERK cascade. It highlights the possibility of attenuating invasive behavior of HCC cells, as well as inhibiting the c-Abl/Raf/Ras/ERK oncogenic signaling axis by targeting claudin molecules. This underscores the need to further validate claudins as both potential biomarkers for the prediction of invasive cancer behavior and as druggable targets for cancer treatment.


  1. 1

    Roxburgh P, Evans TR . Systemic therapy of hepatocellular carcinoma: are we making progress? Adv Ther 2008; 25: 1089–1104.

  2. 2

    Suh Y, Yoon CH, Kim RK, Lim EJ, Oh YS, Hwang SG et al. Claudin-1 induces epithelial-mesenchymal transition through activation of the c-Abl-ERK signaling pathway in human liver cells. Oncogene 2013; 32: 1743–1868.

  3. 3

    Turksen K, Troy TC . Junctions gone bad: claudins and loss of the barrier in cancer. Biochim Biophys Acta 2011; 1816: 73–79.

  4. 4

    Singh AB, Sharma A, Dhawan P . Claudin family of proteins and cancer: an overview. J oncol 2010; 2010: 541957.

  5. 5

    Morin PJ . Claudin proteins in human cancer: promising new targets for diagnosis and therapy. Cancer Res 2005; 65: 9603–9606.

  6. 6

    Myal Y, Leygue E, Blanchard AA . Claudin 1 in breast tumorigenesis: revelation of a possible novel ‘claudin high’ subset of breast cancers. J Biomed Biotechnol 2010; 2010: 956897.

  7. 7

    Tabaries S, Dupuy F, Dong Z, Monast A, Annis MG, Spicer J et al. Claudin-2 promotes breast cancer liver metastasis by facilitating tumor cell interactions with hepatocytes. Mol Cell Biol 2012; 32: 2979–2991.

  8. 8

    Yoon CH, Kim MJ, Park MJ, Park IC, Hwang SG, An S et al. Claudin-1 acts through c-Abl-protein kinase Cdelta (PKCdelta) signaling and has a causal role in the acquisition of invasive capacity in human liver cells. J Biol Chem 2010; 285: 226–233.

  9. 9

    Dhawan P, Singh AB, Deane NG, No Y, Shiou SR, Schmidt C et al. Claudin-1 regulates cellular transformation and metastatic behavior in colon cancer. J Clin Invest 2005; 115: 1765–1776.

  10. 10

    Cervello M, McCubrey JA, Cusimano A, Lampiasi N, Azzolina A, Montalto G . Targeted therapy for hepatocellular carcinoma: novel agents on the horizon. Oncotarget 2012; 3: 236–260.

  11. 11

    Suzuki H, Kondoh M, Takahashi A, Yagi K . Proof of concept for claudin-targeted drug development. Ann NY Acad Sci 2012; 1258: 65–70.

Download references


JS and GG are supported by grants from Cancer Research UK, and Action against Cancer.

Author information

Correspondence to G Giamas.

Ethics declarations

Competing interests

The authors declare no conflict of interest.

Rights and permissions

Reprints and Permissions

About this article


  • Claudin-1
  • hepatocellular cancer
  • invasion
  • Ras
  • ERK
  • c-Abl

Further reading