TGF-beta-Induced Epithelial-to-Mesenchymal Transition Promotes Breast Cancer Progression

In the U.S. alone, approximately 250,000 women are diagnosed with breast cancer every year.1 Forty thousand women will die of the disease, which means that only lung carcinomas cause more cancer-related deaths. Critically, the response to therapy and clinical course are dependent on the subtype of tumor diagnosed.2 For example, luminal A and luminal B type breast cancers are characterized by a low chance of metastasis and relatively good clinical outcomes. However, basal-like breast cancers (BLBC) are highly invasive, progress aggressively to distal tumors, and are associated with poor prognoses.

The invasive nature of each tumor subtype is dependent on epithelial cells increasing their capacity for migration through a process known as the epithelial-to-mesenchymal transition (EMT).3 During EMT, epithelial cancer cells shed their epithelial characteristics and acquire more migratory mesenchymal cell-like properties. The reverse process, mesenchymal-to-epithelial transition (MET), facilitates the subsequent integration of cells at secondary locations. Recent studies have focused on the role of TGF-beta, an established inducer of EMT, in breast cancer progression.4

Breast cancer metastasis is dependent on EMT.
View Larger Image
Breast cancer metastasis is dependent on EMT. Epithelial cancer cells in a breast milk duct undergo epithelial-to-mesenchymal transition (EMT) to acquire a more migratory mesenchymal phenotype. EMT induced by TGF-beta suppresses epithelial genes and promotes the expression of mesenchymal proteins. Highly motile mesenchymal-like breast cancer cells invade pulmonary epithelia and proliferate as secondary tumors. This process requires EMT (1), mesenchymal cell intravasation (2), migration through the vasculature (3), adherence (4), extravasation (5), invasion of a secondary tissue (6), mesenchymal-to-epithelial transition (7), and distal proliferation (8).

Many groups have shown that TGF-beta can induce EMT, but the precise signaling cascades involved are not completely understood.5, 6, 7 Classic TGF-beta signaling requires binding of TGF-beta to type II TGF-beta receptors, transphosphorylation of type I receptors, and subsequent phosphorylation of Smad2 and Smad3. Phosphorylated Smad2/3 forms a trimer with Smad4 that then translocates to the nucleus and interacts with transcription factors, co-activators, and co-repressors to suppress epithelial genes and promote the expression of mesenchymal proteins. In addition, non-Smad signal­ing through activation of ERK MAP Kinases, Rho GTPases, and PI 3-Kinase/Akt has also been implicated in TGF-beta-induced EMT.8, 9, 10

A recent paper investigated the hypothesis that Annexin A1 (AnxA1), an actin regulatory protein, is functionally involved in breast cancer progression.11 de Graauw et al. observed consistently greater AnxA1 expression in BLBC-like, compared to luminal-like, breast cancer cell lines. Using AnxA1 small interfering RNA (siRNA), the authors could drive BLBC-like cells from a mesenchymal to an epithelial morphology, an effect that was reversed by ectopic AnxA1 expression. AnxA1 siRNA also reduced TGF-beta-induced Smad2 phosphorylation and nuclear translocation of Smad4, indicating that AnxA1 was able to regulate TGF-beta signaling. Further in vitro studies showed that expression of AnxA1 in the luminal-like MCF-7 human breast cancer cell line increased cell scattering and Smad3/4 transcriptional reporter activity, effects that could be blocked by the TGF-beta receptor inhibitor SB-431542.

To extend their studies to an in vivo model, highly invasive 4T1 mouse breast cancer cells were injected into mouse mammary fat pads. In this model, knockdown of AnxA1 using short hairpin RNA (shRNA) had no effect on primary tumor growth, but significantly reduced the number of surface metastases in the lungs. To investigate the clinical relevance of these findings, de Graauw et al. examined tissue microarrays from breast cancer patients.11 Analysis of these samples revealed that AnxA1 expression correlated with pathological tumor grade, and was significantly higher in BLBC compared to other tumor subtypes.

Other groups have investigated Smad-dependent TGF-beta signaling in the context of breast cancer progression. Recently, Araki et al. stud­ied the effect of TGF-beta on the p53 tumor suppressor protein.12 The authors showed that TGF-beta increased the expression of the E3 ubi­quitin ligase human double minute 2 (HDM2) in a Smad3/4-de­pendent manner. HDM2 conjugates ubiquitin to p53, tagging it for degradation by the proteasome, and eliminating its capacity for tumor suppression. Using a mouse mammary epithelial cell line, Araki et al. reported similar changes in the expression of murine double minute 2 (MDM2) and p53 during TGF-beta-induced EMT. Following examination of human clinical samples, the authors discovered a significant correlation between Smad3 activation and HDM2 levels in ductal and lobular breast carcinomas. Importantly, increased HDM2 expression and Smad3 activation were not detected in surrounding normal epithelial cells, supporting the specificity of these findings to breast cancer pathology.

Collectively, these recent studies reveal a paradoxical role for TGF-beta in regard to breast cancer progression. Although TGF-beta is known to suppress epithelial cell proliferation and therefore primary tumorigenesis, it is now believed to promote metastasis via the induction of EMT.13 Elucidation of the signaling pathways involved may present a novel pharmacological target for the prevention of breast cancer progression via the inhibition of EMT.


  1. American Cancer Society:
  2. Sørlie, T. et al. (2001) Proc. Natl. Acad. Sci. USA 98:10869.
  3. Kalluri, R. & R.A. Weinberg (2009) J. Clin. Invest. 119:1420.
  4. Xu, J. et al. (2009) Cell Research 19:156.
  5. Sato, Y. et al. (2010) Am. J. Path. 771:141.Cites the use of R&D Systems Products
  6. Takahashi, E. et al. (2010) J. Biol. Chem. 285:4060.Cites the use of R&D Systems Products
  7. Vincent, T. et al. (2010) Nat. Cell. Biol. 11:943.Cites the use of R&D Systems Products
  8. Santibáñez, J.F. et al. (2010) FEBS Lett. 584:2305.Cites the use of R&D Systems Products
  9. Lee, J. et al. (2010) J. Biol. Chem. [Epub ahead of print]Cites the use of R&D Systems Products
  10. Deng, B. et al. (2010) Mol. Cell. Biochem. 340:21.
  11. de Graauw, M. et al. (2010) Proc. Natl. Acad. Sci. USA 107:6340.Cites the use of R&D Systems Products
  12. Araki, S. et al. (2010) J. Clin. Invest. 120:290.Cites the use of R&D Systems Products
  13. Singh, G. et al. (2010) J. Biol. Chem. 285:27241.

Cites the use of R&D Systems Products This symbol denotes references that cite the use of R&D Systems products.