Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Twist2 contributes to breast cancer progression by promoting an epithelial–mesenchymal transition and cancer stem-like cell self-renewal

Abstract

The epithelial to mesenchymal transition (EMT) is a highly conserved cellular programme that has an important role in normal embryogenesis and in cancer invasion and metastasis. We report here that Twist2, a tissue-specific basic helix-loop-helix transcription factor, is overexpressed in human breast cancers and lymph node metastases. In mammary epithelial cells and breast cancer cells, ectopic overexpression of Twist2 results in morphological transformation, downregulation of epithelial markers and upregulation of mesenchymal markers. Moreover, Twist2 enhances the cell migration and colony-forming abilities of mammary epithelial cells and breast cancer cells in vitro and promotes tumour growth in vivo. Ectopic expression of Twist2 in mammary epithelial cells and breast cancer cells increases the size and number of their CD44high/CD24low stem-like cell sub-populations, promotes the expression of stem cell markers and enhances the self-renewal capabilities of stem-like cells. In addition, exogenous expression of Twist2 leads to constitutive activation of STAT3 (signal transducer and activator of transcription 3) and downregulation of E-cadherin. Thus, the overexpression of Twist2 may contribute to breast cancer progression by activating the EMT programme and enhancing the self-renewal of cancer stem-like cells.

Introduction

The epithelial to mesenchymal transition (EMT) is a well-coordinated developmental programme that has a very important role in the development of the mesoderm from the epithelium during embryogenesis (Acloque et al., 2009; Kalluri and Weinberg, 2009). During the EMT process, epithelial cells undergo dramatic phenotypic changes, lose expression of E-cadherin and other components of epithelial cell junctions, adopt a mesenchymal cell phenotype and acquire motility and invasive properties that allow them to migrate through the extracellular matrix. EMT is triggered by several extracellular signals, including components of the extracellular matrix and growth factors, and is mediated by the activation of EMT transcription factors such as Twist1, Snai1, Slug, ZEB1 and ZEB2 (Maeda et al., 2005; Thiery and Sleeman, 2006; Ouyang et al., 2010). Accumulating evidence suggests that aberrant activation of the EMT developmental programme contributes to tumour invasion, metastatic dissemination and acquisition of therapeutic resistance (Yang et al., 2004; Yang and Weinberg, 2008; Ruan et al., 2009; Thiery et al., 2009; Singh and Settleman, 2010; Hanahan and Weinberg, 2011).

The Twist proteins Twist1 and Twist2 (also named Dermo-1) are highly conserved basic helix-loop-helix (bHLH) transcription factors that are structurally characterised by a conserved domain, which contains a stretch of basic amino acids adjacent to two amphipathic α-helices that are separated by an interhelical loop. Twist1 is a regulator of embryonic development that participates in EMT during differentiation of the mesoderm in Drosophila and in neural crest migration in vertebrates (Thisse et al., 1987; Chen and Behringer, 1995). Twist1 is overexpressed in various human solid tumours and is involved in tumour invasion and metastasis. Ectopic overexpression of Twist1 remarkably enhances the invasive and metastatic capacities of Madin–Darby canine kidney epithelial cells, human mammary epithelial cells and breast cancer cells by inducing EMT (Yang et al., 2004; Ansieau et al., 2008; Ansieau et al., 2010). Twist2, which was cloned as an E12-binding partner using a yeast two-hybrid screen, shows more than 90% identity with Twist1 in the bHLH and carboxy-terminal domains and regulates a series of developmental processes in embryogenesis (Li et al., 1995). During mouse embryogenesis, the expression pattern of Twist2 extensively overlaps, but is distinct from, that of mouse Twist1. Twist2 is predominantly expressed in a subset of mesodermal- and ectodermal-derived tissues and may function as a potent negative regulator of gene expression during differentiation of a subset of mesenchymal cell lineages. Twist2 represses the transcription of the myogenic bHLH protein MyoD. In addition, Twist2 inhibits osteoblast maturation and maintains cells in a preosteoblast phenotype (Lee et al., 2000; Gong and Li, 2002; Franco et al., 2011). Similar to Twist1, Twist2 overrides oncogene-induced premature senescence by promoting EMT in human epithelial cells (Ansieau et al., 2008). Twist2 is also involved in p12CDK2−Ap1-induced EMT of hamster cheek pouch carcinoma-I cells (Tsuji et al., 2008). Twist2 mRNA is overexpressed in a large variety of human primary tumours; however, the transcriptional expression pattern of Twist2 differs from that of Twist1 in some of these cancers (Ansieau et al., 2008). These differences suggest that Twist1 and Twist2 may have overlapping but distinct roles in different sets of tumours.

Recently, Mani et al. (2008) demonstrated that Twist1 transforms non-tumourigenic, immortalised human mammary epithelial cells into mesenchymal-like cells and dedifferentiates HER2/neu-infected human mammary epithelial cells into CD44high/CD24low cancer stem-like cells via EMT. Cancer stem cells (CSCs), which are also known as tumour initiating cells or tumourigenic cells, are cell populations in tumour tissues with stem cell-like properties (Al-Hajj et al., 2003; Brabletz et al., 2005; Bao et al., 2006; Mani et al., 2008; Gupta et al., 2009a). Twist1 induces a breast CSC phenotype by directly reducing the level of CD24 (Vesuna et al., 2009). Yin et al. (2010) reported that Twist1 may be an important regulator of stemness in epithelial ovarian cancer cells. The EMT-derived cells, which are induced by Twist1 in human mammary epithelial cells, exhibited the potential for multilineage differentiation similar to mesenchymal stem cells (Battula et al., 2010). Twist2, similar to Twist1, is a potential mediator of mesenchymal stem cell self-renewal and lineage commitment. In addition, Twist2 may function to regulate critical transcription factors and osteo-/chondrogenic-inductive factors in the early stages of cell fate determination in human mesenchymal stem cells (Isenmann et al., 2009). However, the functional significance of Twist2 in human breast cancer and in regulating CSC stemness as well as the underlying cellular and molecular mechanisms are poorly studied.

In this study, we demonstrate that Twist2 is a potent inducer of EMT. In human mammary epithelial cells and breast cancer cells, overexpression of Twist2 upregulates characteristic mesenchymal markers, downregulates epithelial markers and promotes cell migration. Moreover, Twist2 increases the abundance of CD44high/CD24low sub-populations and promotes self-renewal of stem-like cells. Furthermore, Twist2 increases the colony-forming activity of mammary epithelial cells and breast cancer cells in vitro and promotes tumourigenesis in vivo. Taken together, these data suggest that Twist2 contributes to breast cancer progression by promoting EMT and cancer stem-like cell self-renewal.

Results

Twist2 is overexpressed in breast cancer

Although the Twist2 gene is overexpressed (more than twofold) in only 3.6% of breast cancer tissue samples compared with healthy breast tissue, the protein level of Twist2 in human breast cancer remains unknown (Ansieau et al., 2008). We therefore measured the protein levels of Twist2 in human breast cancer samples, consisting of 33 specimens of primary breast cancer tissue, 12 specimens of matched non-cancer breast tissue and 8 specimens of matched breast metastatic tumours from the lymph nodes, using immunohistochemistry (Figure 1). Of the 33 primary breast cancer tissue specimens, 31 (93.9%) were positive for Twist2. Among these 31 Twist2-positive samples, 28 (90.3%) showed positive for Twist2 in both the cytoplasm and the nucleus, whereas Twist2 was primarily located in the nucleus in the other 3 cases. Interestingly, all eight breast metastatic tumour specimens were Twist2 positive. Twist2 was distributed in both the cytoplasm and the nucleus in six of these samples, whereas Twist2 was mostly localised in the nucleus in the other two samples. In contrast, most non-cancer breast tissue samples (8/12) were negative for Twist2 staining, whereas the other 4 non-cancer breast tissues samples were positive for the staining of Twist2, which was mainly located in the nucleus. These data indicate that Twist2 may be involved in tumourigenesis and metastasis of human breast cancer.

Figure 1
figure1

Twist2 is overexpressed in human breast cancer, whereas E-cadherin is downregulated. Pictures from one representative case are presented. (a) Most non-cancer breast tissues were negative for Twist2 staining. (b) A total of 31 (93.9%) specimens of primary breast cancer tissue were positive for Twist2 staining. Twist2 was distributed in both the cytoplasm and the nucleus in most of the primary breast cancer tissue specimens (28/31). (c) All eight specimens of breast metastatic tumours were Twist2 positive. Twist2 was distributed in both the cytoplasm and the nucleus in most cases (six out of eight). (d) Non-cancer breast tissues were strongly positive for E-cadherin staining. Compared with the matched non-cancer breast tissues, the levels of E-cadherin in six of eight specimens of primary breast cancer tissues (e) and breast metastatic tumours from lymph node (f) were decreased.

The expression of E-cadherin correlates to the invasive potential of several types of human cancer including breast cancer. To determine whether a correlation existed between the expression patterns of Twist2 and E-cadherin in human breast cancer, we also analysed E-cadherin expression using immunochemical staining in the eight surgical specimens of human breast cancer tumour tissues with lymph node metastases mentioned above. Compared with the matched non-cancer breast tissues, we found that six of eight cases of breast tumour tissues and their matched six lymph node metastatic specimens displayed decreased E-cadherin. Among these six cases, five cases showed strong positive staining for Twist2 in the cytoplasm and nucleus (Figure 1). These results suggest that a reverse correlation may exist between the expression patterns of Twist2 and E-cadherin in human breast cancer progression.

Twist2 induces EMT in mammary epithelial cells and breast cancer cells

On the basis of our observations that E-cadherin may be involved in the functions of Twist2 in human breast cancer progression, we next investigated whether Twist2 was capable of inducing the EMT in human mammary epithelial cells. We used MCF-10A cells to generate stable cell lines that constitutively expressed either Flag-tagged Twist2 or empty vectors by retroviral infection. MCF-10A cells are immortalised, non-transformed human mammary epithelial cells that lack endogenous Twist2 expression. Stable ectopic expression of Twist2 in MCF-10A cells was confirmed using as anti-Flag antibody and western blotting analysis (Figure 2a). Twist2 was located in both the nuclei and the cytoplasm, but mainly in the nuclei of MCF-10A/Twist2 cells as shown by immunofluorescence staining (Figure 2b). Compared with vector-infected cells, Twist2-expressing cells underwent a striking morphological change, which entailed a transformation from a cobblestone-like epithelial morphology to an elongated fibroblast-like morphology when the cells were plated at either low or high densities (Figure 2c). Western blot analysis of MCF-10A cells that were grown in dense cultures showed a relatively moderate downregulation of the epithelial markers E-cadherin and β-catenin in Twist2-expressing cells, whereas the mesenchymal markers fibronectin, N-cadherin and vimentin were notably upregulated (Figure 2d). In addition, real-time reverse transcriptase–PCR analysis demonstrated a gene expression pattern consistent with EMT, including E-cadherin repression and the concomitant induction of Vimentin, N-cadherin, Fibronectin, transforming growth factor-β1 (TGF-β1), Twist1, Snai1, Slug, ZEB1 and ZEB2 (Figures 2e and f; P<0.05). To further examine whether Twist2 overexpression induced an EMT phenotype in human breast cancer cells, we used the low-tumourigenic human breast cancer cell line MCF-7 to establish Twist2-expressing cells (Figure 2g). The MCF-7/Twist2 cells showed a mesenchymal-like morphology (Figure 2h), a decreased level of E-cadherin and β-catenin, and an increase in fibronectin and N-cadherin expression in dense culture (Figure 2i). Twist2 was distributed in both the nucleus and the cytoplasm in MCF-7/Twist2 cells (Figure 2j).

Figure 2
figure2

Ectopic overexpression of Twist2 in human mammary epithelial cells and breast cancer cells results in epithelial–mesenchymal transition (EMT). (a) MCF-10A cells constitutively expressing either Flag-tagged Twist2 or empty vector were generated by retroviral infection with a retroviral pBABE-puro and confirmed by western blot analysis using specific antibodies against Twist2 and the Flag tag. (b) Analysis using immunofluorescence reveals that Twist2 is located in both the nuclei and the cytoplasm but mainly in the nuclei of MCF-10A/Twist2 cells. (c) Twist2 induced mesenchymal morphology in MCF-10A cells. Cells were plated at low (top row) or high (bottom row) densities. Cells were monitored for morphological changes using phase-contrast microscopy. ( × 100.) (d) Western blot analysis was used to evaluate EMT markers in cell lysates. Ectopic overexpression of Twist2 in MCF-10A cells increases the expression of the mesenchymal markers fibronectin, N-cadherin and vimentin but decreases the expression of the epithelial markers E-cadherin and β-catenin. (e) Relative levels of mRNAs encoding E-cadherin, vimentin, N-cadherin and fibronectin in MCF-10A cells expressing vector or Twist2 were determined by real-time reverse transcriptase–PCR. GAPDH mRNA was used to normalise variability in template loading. Error bars represent the mean±s.d. of triplicate experiments. *P<0.05. (f) Twist2 increases the transcription of other known EMT inducers in MCF-10A cells. Real-time reverse transcriptase–PCR analysis showed the relative mRNA expression levels of known EMT inducers in MCF-10A cells expressing vector or Twist2. GAPDH mRNA was used to normalise variability in template loading. Error bars represent the mean±s.d. of triplicate experiments. *P<0.05. (g) MCF-7 cells that constitutively express either Flag-tagged Twist2 (Twist2) or empty vector (vector) were generated by retroviral infection and confirmed by immunoblotting analysis. (h) Twist2 induces mesenchymal morphology in MCF-7 cells. ( × 100.) (i) Twist2 induces changes in the expression of EMT markers in MCF-7 cells. The expression of epithelial and mesenchymal markers was analysed by western blotting. (j) Analysis using immunofluorescence reveals the nuclear and cytoplasm localisation of Twist2 in MCF-7/Twist2 cells.

Analysis using immunofluorescence staining further revealed that vimentin (Figure 3a) and fibronectin (Figure 3b) were strongly induced in Twist2-expressing MCF-10A cells. In contrast, levels of E-cadherin, which showed membrane localisation in vector cells, were dramatically decreased in MCF-10A/Twist2 cells (Figure 3c). β-Catenin was modestly downregulated and translocated into the cytoplasm in MCF-10A/Twist2 cells (Figure 3d). Ectopic expression of Twist2 in MCF-7 cells also decreased the levels of E-cadherin (Figure 3e) and β-catenin (Figure 3f).

Figure 3
figure3

Overexpression of Twist2 induces EMT marker alterations in MCF-10A (ad) and MCF-7 (e, f) cells in two-dimensional culture. Twist2 induces the expression of vimentin (a) and fibronectin (b), downregulates the expression of E-cadherin (c) and modestly alters the localisation of β-catenin (d) in MCF-10A cells. Ectopic expression of Twist2 in MCF-7 cells decreases the levels of E-cadherin (e) and β-catenin (f).

To extend our analysis to a model with more physiological relevance, we used an MCF-10A three-dimensional cell culture model to test some markers of EMT using immunofluorescence. The expression of E-cadherin, which is a central component of cell–cell adhesion junctions, was dramatically downregulated, whereas vimentin was sharply upregulated in the acinar structures formed by Twist2-overexpressing MCF-10A cells compared with vector-infected cells (Figure 4a). We also investigated whether Twist2 alters epithelial cell polarity. Unlike monolayer cultures, mammary epithelial cells grown in three dimensions recapitulate various features of in vivo breast epithelium, such as apicobasal polarisation (Debnath et al., 2003). To analyse changes in polarity, we used immunofluorescence to visualise the orientation of the Golgi apparatus in acini using the cis-Golgi protein GM130. In addition, we examined the distribution of α6 integrin, which is a basal marker. In vector-cell-derived acini, the apical portion of the Golgi was always oriented towards the lumen. However, Twist2-expressing cell-derived acini showed a modest disruption in Golgi orientation; some of the Golgi apparatus was oriented towards the basal surface, which was rarely seen in control acini (Figure 4b). Consistent with these results, α6 integrin, which was located on the basal surface of acini formed by vector cells, was depolarised in acini derived from Twist2-expressing cells (Figure 4c). In addition, Twist2-expressing cell-derived acini elicited excess proliferation and nearly filled luminal space with an increased proliferation marker Ki-67 (Figure 4d) and a decreased apoptotic marker activated caspase 3 (Figure 4e). Collectively, these observations demonstrate that the ectopic expression of Twist2 promotes an EMT phenotype in human mammary epithelial cells and breast cancer cells, indicating that Twist2 is a potent inducer of EMT.

Figure 4
figure4

Overexpression of Twist2 induces EMT marker alterations in MCF-10A cells in three-dimensional culture. In acinar structures formed by vector- and Twist2-expressing MCF-10A cells, Twist2 downregulates the expression of E-cadherin, upregulated the expression of vimentin (a), alters the orientation of polar proteins GM130 (b) and α6 integrin (c), increases the level of proliferation marker Ki-67 (d) and decreases the level of apoptotic marker activated caspase 3 (e).

Twist2 enhances cell migration and colony formation in vitro and promotes tumour growth in vivo

The EMT is characterised by increased cell motility and is also a critical step in tumour progression. Given the mesenchymal phenotype of MCF-10A/Twist2 cells, we hypothesised that ectopic Twist2 expression enhances cell motility as a consequence of the EMT process. To test this hypothesis, we examined whether Twist2 expression affects the motility of MCF-10A cells in a wound healing assay. Confluent cell monolayers of vector- and Twist2-expressing cells were wounded with a pipette tip. After 16 h, Twist2-expressing cells migrated away from the wound edge and showed wound closure, whereas the wound of vector-transfected cells did not heal (Figure 5a). These results suggest that increased Twist2 expression promotes cell motility. To investigate whether Twist2 enhances the colony-forming ability of MCF-10A cells in vitro, we seeded 200 cells in triplicate wells of six-well plates for a colony-forming assay. After 10 days of culture, Twist2-expressing cells formed colonies that were more numerous and larger than those of the vector-transfected cells (Figure 5b).

Figure 5
figure5

Twist2 enhances cell migration and the colony formation of MCF-10A and MCF-7 cells in vitro and promotes tumour growth in vivo. (a) Twist2 expression in MCF-10A cells promotes cell migration in a wound-healing assay. ( × 100.) (b) Twist2-expressing MCF-10A cells form more and larger colonies compared with the empty vector-expressing cells in six-well plates. (c) Twist2 enhances MCF-7 cell migration in a wound-healing assay. ( × 100.) (d) Twist2-expressing MCF-7 cells form more and larger colonies compared with vector cells in six-well plates. (e) Twist2 promotes tumour growth in vivo. Eight female BALB/c nude mice, which were aged 4–6 weeks, were randomly divided into two groups. MCF-7 vector- or Twist-expressing cells (5 × 106) were subcutaneously injected into BALB/c nude mice.

MCF-10A cells are immortalised, non-transformed human mammary epithelial cells, and do not show capabilities of transforming and tumour growth even with ectopic overexpression of oncogene K-Ras, H-RasV12 or LBX1 (Ansieau et al., 2008; Yu et al., 2009). Therefore, we used MCF-7 cells ectopically expressing Twist2 to determine the effects of Twist2 expression on cell migration and colony-forming activity in vitro and on tumourigenicity in nude mice. Wound healing assays showed that Twist2-expressing cells showed greater migration potential than vector cells (Figure 5c). Twist2-expressing MCF-7 cells formed more and larger colonies than the vector cells in six-well plates after 20 days of incubation (Figure 5d). These data suggest that Twist2 enhances the transformative activity of human breast cancer cells in vitro. To further examine whether Twist2 expression impacted tumour growth in vivo, we subcutaneously injected the same number of Twist2-expressing or control MCF-7 cells into nude mice. As shown in Figure 5e, Twist2 overexpression markedly enhanced the subcutaneous tumour growth over the control group after 4 weeks, and the volumes of the tumours in the nude mice injected with Twist2-expressing cells were larger than those of the control group. Taken together, our results indicate that Twist2 has prooncogenic activity and functions to promote tumour growth both in vitro and in vivo.

Overexpression of Twist2 in MCF-10A and MCF-7 cells increases their CD44high/CD24low progenitor sub-population and promotes stem-like cell self-renewal

As noted earlier, EMT generates cells with stem cell-like properties (Mani et al., 2008). However, the functional significance of Twist2 in breast CSCs remains unexplored. We determined whether stable expression of Twist2 induced such stem cell-like phenotypes in human mammary epithelial cells and breast cancer cells. Representative stem cell markers were analysed by western blot. As shown in Figure 6a, compared with the vector control, Twist2 expression caused an increase in the protein levels of the stem cell markers CD44, Bmi-1 and Sox2. To test the self-renewal ability of MCF-10A/Twist2 cells, we turned to a well-characterised tumoursphere culture model to examine the capacity for mammosphere formation in non-adherent serum-free medium. Mammosphere formation assays of Twist2-expressing MCF-10A cells revealed an increase in the size and number of mammospheres compared with vector-transduced cells (P<0.05; Figures 6b and c). To further explore the role of Twist2 in stem cell phenotypes in human breast cancer, we performed flow cytometric analysis to identify CD44high/CD24low populations, which is a phenotype associated with normal breast stem cells and potential breast cancer progenitors. Flow cytometric analysis demonstrated that the proportion of CD44high/CD24low cells was 7.50% in the vector control but reached 35.65% in MCF-10A/Twist2 cells (P<0.01; Figure 6d). We further investigated the effect of Twist2 overexpression on the stemness of cancer stem-like cells in human breast cancer cells. As shown in Figure 6e, Twist2 overexpression upregulated the expression of stem cell markers CD44, Bmi-1 and Sox2. The number and size of mammospheres was increased in MCF-7/Twist2 cells (P<0.01) (Figures 6f and g). Moreover, Twist2 augmented the proportion of CD44high/CD24low cancer stem-like cells from 2.43% in MCF-7/vector cells to 15.04% in MCF-7/Twist2 cells (P<0.01; Figure 6h). These data demonstrate that the overexpression of Twist2 can endow human mammary epithelial cells and breast cancer cells with a stem cell-like phenotype.

Figure 6
figure6

Ectopic overexpression of Twist2 in mammary epithelial cells (ad) and breast cancer cells (eh) increases their CD44high/CD24low progenitor sub-population and promotes stem-like cell self-renewal. Twist2 upregulated the expression of the stem cell markers CD44, Bmi-1 and Sox2 in MCF-10A (a) and MCF-7 (e) cells. Twist2 increases the size and number of mammospheres in MCF-10A (b, c) and MCF-7 (f, g) cells *P<0.05, **P<0.01. Twist2 increases the size of the CD44high/CD24low progenitor sub-population in MCF-10A (d) and MCF-7 (h) cells. Flow cytometry was used to analyse CD44high/CD24low sub-populations in the cells expressing either empty vector or Twist2. The percentages of the mean CD44high/CD24low population are indicated.

Twist2 represses E-cadherin promoter activity

Members of the bHLH family of transcription factors dimerise to form heterodimers, which in turn bind the E-box consensus sequence (CANNTG) in GC-rich promoter regions to regulate a diverse set of genes (Li et al., 1995; Gong and Li, 2002). To determine whether the decreased expression of E-cadherin observed in Twist2-overexpressing cells was the consequence of transcriptional downregulation, MCF-10A/vector and MCF-10A/Twist2 cells were transiently transfected with a reporter plasmid containing the luciferase gene under the control of an 887 bp fragment of the human E-cadherin promoter (from −680 to +207; Fernando et al., 2010). As shown in Figure 7a, reporter activity was significantly repressed in MCF-10A/Twist2 cells compared with control cells (P<0.01). These results show that Twist2 directly or indirectly regulates E-cadherin transcription in these cells.

Figure 7
figure7

Ectopic Twist2 expression significantly downregulates E-cadherin promoter activity and constitutively activates STAT3. Luciferase reporter activity is significantly repressed in MCF-10A/Twist2 cells compared with MCF-10A/vector cells (**P<0.01) (a). STAT3 is constitutively activated and primarily localised to the nuclei in MCF-10A/Twist2 (b, c) and MCF-7/Twist2 (d, e) cells, whereas no obvious pSTAT3 (Y705) is observed in empty vector-transfected cells.

STAT3 (signal transducer and activator of transcription 3) is constitutively active in Twist2-expressing cells

Constitutive activation of STAT3 and its downstream targets are involved in multiple steps of tumourigenesis from initiation to progression. STAT3 is constitutively active in more than 50% of primary breast tumours and tumour-derived cell lines. Levels of STAT3, which is phosphorylated at tyrosine residue 705 (Y705), are frequently elevated in breast carcinomas (Lo et al., 2007). A previous report has shown that STAT3 activates Twist1 expression by binding to its promoter (Cheng et al., 2008). In addition, ectopic Twist1 expression in MCF-7 cells induces constitutive STAT3 activation (Sullivan et al., 2009). To determine the relationship exists between Twist2 and STAT3, we examined the expression and localisation of STAT3 that was phosphorylated at Y705 in MCF-10A/Twist2 cells using western blot analysis. As shown in Figure 7b, phosphorylated STAT3 (Y705) was upregulated in Twist2-expressing cells compared with that in the controls. However, we were unable to find another activated form of STAT3, pSTAT3 (S727), in either the vector- or the Twist2-transfected cells (data not shown). We also used immunofluorescence to confirm the expression and localisation of phosphorylated STAT3 (Y705) in MCF-10A/Twist2 cells. Analysis using immunofluorescence revealed that STAT3 was constitutively activated and mainly located in the nuclei of MCF-10A/Twist2 cells, whereas no obvious pSTAT3 was observed in the vector-transfected cells (Figure 7c). We further examined the expression and localisation of pSTAT3 (Y705) in human breast cancer cells by immunoblotting and immunofluorescence. Ectopic expression of Twist2 also constitutively activated STAT3 (Figure 7d) and the activated pSTAT3 (Y705) was mainly distributed in the nuclei of MCF-7/Twist2 cells (Figure 7e). These results indicate that STAT3 activity is associated with Twist2 function in mammary epithelial cells and breast cancer cells.

Discussion

Breast carcinogenesis is a multistep and multipath disease process during which normal mammary epithelial cells progressively develop into hyperplasia, carcinoma in situ, invasive cancer and metastasis. Twist2 has been shown to be involved in some human primary tumours; however, little is known about the functional significance of Twist2 in human breast cancer. In this study, our results demonstrate that Twist2 is overexpressed in human breast cancers and lymph node metastases and promotes an EMT phenotype in human mammary epithelial cells and breast cancer cells, indicating that Twist2 may function as a potent driver of EMT during breast cancer development. Of note, there is a huge discrepancy between the percentage of Twist2-positive human primary breast tumours assessed by immunohistochemistry, which detected the protein level of the Twist2 gene in this report, and the percentage of human breast samples harbouring an overexpression of Twist2 mRNA levels (more than twofold), which was measured by real-time quantitative PCR analysis (Ansieau et al., 2008). We also observed that the Twist2 mRNA level and the Twist2 protein expression in two cases of human breast cancer tissue samples and their matched non-cancer and lymph node metastatic specimens are not always correlated (data not shown). Further work will be necessary to characterise the discrepancy between the mRNA level and the protein expression of Twist2 gene in breast cancer progression.

Increasing evidence suggests that cancer cells often reactivate latent developmental programmes to efficiently regulate the multistep process of tumourigenesis. EMT is a highly coordinated developmental process during embryogenesis and a pathological feature in tumourigenesis (Acloque et al., 2009; Kalluri and Weinberg, 2009; Hanahan and Weinberg, 2011). Developmental EMT regulators including Twist1, Twist2, Snai1, Slug and Six1, as well as developmental signalling pathways, including TGF-β and Wnt/β-catenin, are aberrantly activated in cancers and correlate with poor clinical outcomes. Recently, EMT inducers or regulators, such as Twist1, Snai1, TGF-β, ZEB1, YB-1, LBX1 and Six1, have been shown to induce well-differentiated cells and cancer cells to form populations with stem cell-like characteristics via promoting EMT, indicating that there is a cross-talk between the EMT programme and the pathways involved in regulating stemness in stem cells (Mani et al., 2008; Morel et al., 2008; Evdokimova et al., 2009; Iliopoulos et al., 2009; McCoy et al., 2009; Polyak and Weinberg, 2009; Wellner et al., 2009; Yu et al., 2009; Ouyang et al., 2010; Singh and Settleman, 2010). In this study, we demonstrated that Twist2 has a critical role in EMT and is involved in reprogramming differentiated cells and cancer cells into stem-like cells. The ectopic expression of Twist2 in mammary epithelial cells and breast cancer cells increases the CD44high/CD24low stem-like cell sub-population and the self-renewal capability of these stem-like cells. This finding was further confirmed by the increased expression of stem cell markers such as CD44, Bmi-1 and Sox2 in Twist2-expressing cells. Our study supports a link between the EMT programme and the gain of epithelial stem cell-like properties.

Of all the EMT markers regulated by Twist2, E-cadherin is considered a hallmark of epithelial cells. E-cadherin is a central adhesion molecule of cell–cell adhesion junctions that is essential for the formation and maintenance of the epithelial cell phenotype. Loss of E-cadherin, which is consistently observed at sites of EMT in embryonic development and tumourigenesis, has wide-ranging transcriptional and functional consequences for human breast epithelial cells (Peinado et al., 2007; Onder et al., 2008). In a previous study, short hairpin RNA-mediated knockdown of E-cadherin triggered EMT and resulted in the acquisition of a mesenchymal phenotype and increased CSC activity in HMLER breast cancer cells (Gupta et al., 2009b). Moreover, the E-cadherin promoter is frequently repressed directly or indirectly by transcription factors such as Twist1, Snai1, Slug, ZEB1, ZEB2, FOXC2, KLF8 and E47, thereby disrupting the polarity of epithelial cells and maintaining a mesenchymal phenotype (Yang and Weinberg, 2008; Thiery et al., 2009; Ouyang et al., 2010). Our results show that Twist2 directly or indirectly binds to the promoter of E-cadherin and downregulates the expression of E-cadherin. This result indicates that Twist2 may promote EMT and generate mesenchymal cells with stemness by down-egulating the expression of E-cadherin.

The polycomb group protein Bmi-1 may be involved in Twist2-induced repression of E-cadherin expression. In this study, we demonstrated that the expression of stem cell marker Bmi-1 is increased in Twist2-expressing human mammary epithelial cells and breast cancer cells. Bmi-1, a member of the polycomb-repressive complex 1, is commonly deregulated in various tumours, such as breast cancer, and has a critical role in maintaining self-renewal in normal and malignant human mammary stem cells (Dimri et al., 2002; Liu et al., 2005, 2006). Bmi-1 inhibits phosphatase and tensin homologue, induces EMT in human nasopharyngeal epithelial cells, and is involved in the regulation of self-renewal and differentiation of stem cells (Song et al., 2009b). A recent report showed that Twist1 directly regulated Bmi-1. Bmi-1-containing polycomb-repressive complex directly represses E-cadherin expression. Bmi-1 and Twist1 are mutually essential to promote EMT and tumour-initiating capability of human head and neck squamous cell carcinoma cells (Yang et al., 2010). Together with our findings, Bmi-1 may be involved in Twist2-induced repression of E-cadherin expression. Further investigation is required to understand the role of Bmi-1 in Twist2-induced downregulation of E-cadherin.

Constitutive activation of STAT3 has been observed in a number of human epithelial malignancies, including tumours of the prostate, breast, lung, head and neck, brain and pancreas (Bromberg, 2002; Chan et al., 2004). On phosphorylation of a tyrosine residue, activated STAT3 translocates to the nucleus and regulates the transcription of numerous target genes (Levy and Darnell, 2002). STAT3 activation has an important role in TGF-β-mediated EMT and mediates epidermal growth factor-/epidermal growth factor receptor-induced EMT by increasing the expression of Twist1 (Yang et al., 2006; Lo et al., 2007). Moreover, as a prosurvival pathway, STAT3 signalling is critical for breast cancer stem-like cell maintenance (Zhou et al., 2007). In this study, we found that STAT3 was constitutively active and mainly located in the nuclei of Twist2-expressing mammary epithelial cells and breast cancer cells. Therefore, our current data suggest that STAT3 may be involved in Twist2-induced EMT and stem-like cell self-renewal in human mammary epithelial cells and breast cancer cells.

In summary, Twist2 contributes to breast cancer progression by promoting the activation of EMT and by enhancing the self-renewal of cancer stem-like cells. The critical role of Twist2, along with Twist1, Snai1, TGF-β, YB-1 and LBX1, in promoting EMT and stem cell self-renewal indicates that cancer cells often reactivate latent developmental programmes to efficiently regulate the multistep process of tumourigenesis. These EMT inducers or regulators are often kept silent in adults to maintain epithelial homeostasis and integrity. Therefore, knowledge of the molecular details of the physiology and pathology of EMT and its function in regulating the stemness of stem cells may facilitate the development of new biomarkers for cancer detection as well as more specific and potent anti-invasive drugs.

Materials and methods

Cell culture and generation of cell lines

Human mammary epithelial cell line MCF-10A, breast cancer cell line MCF-7, packaging cell line GP2-293 and retroviral vector pBABE-puro were generous gifts from Professor Kunxin Luo (Department of Molecular and Cell Biology, University of California, Berkeley, CA, USA). MCF-10A cells were maintained as described (Debnath et al., 2003). MCF-7 and GP2-293 cells were cultured in Dulbecco's modied Eagle's media supplemented with 10% fetal bovine serum. The human Twist2-Flag complementary DNA was cloned into a pBABE-puro expression vector. Cell lines expressing either Twist2-Flag (MCF-10A/Twist2 and MCF-7/Twist2) or empty vector (MCF-10A/vector and MCF-7/vector) were generated by retroviral infection. Briefly, the GP2-293 packaging cells were grown to 60% confluency and co-transfected with 5 μg of pBABE-puro or pBABE-puro-Twist2-Flag and 5 μg of pCMV-VSVG using the Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA). The viral supernatants were collected at 48 h after transfection and spun at 3000 g for 3 min to remove cells and debris. The supernatant was transferred to a new tube and filtered through a 0.45 μm filter. For retroviral infection of cells, MCF-10A or MCF-7 cells were plated at 105 cells per well on a six-well plate. After 24 h, the medium was removed and 2 ml of the culture supernatant, which contained retroviral vectors and polybrene (Sigma-Aldrich, St Louis, MO, USA) at a final concentration of 8 μg/ml, was added to the wells. After 24 h of infection, the culture medium with viral supernatant was removed and replaced with fresh medium containing 2 μg/ml of puromycin for selection. MCF-10A and MCF-7 cells that constitutively expressed either Flag-tagged Twist2 or empty vector were confirmed by western blot analysis with specific antibodies against Twist2 and the Flag tag.

Western blot analysis

Western blotting was performed as described previously (Song et al., 2009a). Cell lysates were subjected to SDS–polyacrylamide gel and immunoblot analysis with antibodies to Twist2 (Abnova, Taipei, Taiwan), Flag (Sigma-Aldrich), E-cadherin, β-catenin, N-cadherin, fibronectin, GM130, CD44, CD24 (BD Biosciences, Bedford, MA, USA), vimentin, Sox2 (R&D Systems, Minneapolis, MN, USA), STAT3, pSTAT3 (Tyr705), pSTAT3 (Ser727), active (cleaved) caspase 3 (Cell Signaling Technology, Danvers, MA, USA), α6 integrin (Millipore, Billerica, MA, USA), Ki-67 (Zymed, San Francisco, CA, USA), Bmi-1 and β-actin (Millipore).

Immunofluorescence

Cells were seeded on coverslips in six-well plates. After 24 h, the cells were washed with phosphate-buffered saline (PBS) twice, permeabilised in 0.1% Triton X-100 and fixed with 4% paraformaldehyde in PBS for 20 min. The paraformaldehyde was quenched with three washes in 100 mM glycine in PBS. Coverslips were permeabilisated and then blocked for 1 h in blocking buffer (PBS containing 10% newborn calf serum, 1% bovine serum albumin and 0.02% Triton X-100). After being blocked, coverslips were incubated with antibodies in staining buffer (blocking buffer lacking bovine serum albumin) overnight at 4 °C. Coverslips were washed three times with staining buffer and then incubated with Alexa 488-conjugated and/or Alexa 546-conjugated secondary antibodies (Invitrogen) for 1 h at room temperature. Coverslips were then washed twice with 10% fetal bovine serum/0.02% Triton X-100 in PBS and stained with Hoechst 33 258. Finally, coverslips were washed three times with PBS and one time with water before being mounted on slides. Immunofluorescence was observed with a Leica confocal microscope (Leica, Solms, Germany).

Real-time reverse transcriptase–PCR

Total cellular RNA was prepared using TRIzol reagent (Invitrogen) and the expression levels of E-cadherin, Vimentin, N-cadherin, Fibronectin, TGF-β1, Twist1, Snai1, Slug, ZEB1 and ZEB2 mRNAs were determined by real-time reverse transcriptase–PCR using SYBR Green. Data shown are normalised to GAPDH expression and represent the average of three repeated experiments. Primer sequences were as follows. E-cadherin, forward: 5′-IndexTermTGCCCAGAAA ATGAAAAAGG-3′, reverse: 5′-IndexTermGTGTATGTGGCAATGCGTTC-3′; Vimentin, forward: 5′-IndexTermGAGAACTTTGCCGTTGAAGC-3′, reverse: 5′-IndexTermGCTTCCTGTAGGTGGCAATC-3′; N-cadherin, forward: 5′-IndexTermACAGTGGCCACCTACAAAGG-3′, reverse: 5′-IndexTermCCGAGATGGGGTTGATAATG-3′; Fibronectin, forward: 5′-IndexTermCAGTGGGAGACCTCGAGAAG-3′, reverse: 5′-IndexTermTCCCTCGGAACATCAGAAAC-3′; TGF-β1, forward: 5′-IndexTermCAGAAATACAGCAACAATTCCTGG-3′, reverse: 5′-IndexTermTTGCAGTGTGTTATCCCTGCTGTC-3′; Twist1, forward: 5′-IndexTermGGAGT CCGCAGTCTTACGAG-3′, reverse: 5′-IndexTermTCTGGAGGACCT GGTAGAGG-3′; Snai1, forward: 5′-IndexTermCCTCCCTGTCAGATGAGGAC-3′, reverse: 5′-IndexTermCCAGGCTGAGGTATTCCTTG-3′; Slug, forward: 5′-IndexTermGGGGAGAAGCCTTTTTCTTG-3′, reverse: 5′-IndexTermTCCTCATGTTTGTGCAGGAG-3′; ZEB1, forward: 5′-IndexTermAACCCAACTTGAACGTCACA-3′, reverse: 5′-IndexTermATTACACCCAGACTGCGTCA-3′; ZEB2, forward: 5′-IndexTermTTCCTGG GCTACGACCATAC-3′, reverse: 5′-IndexTermTGTGCTCCATCAAGCA ATTC-3′; GAPDH, forward: 5′-IndexTermACCCAGAAGACTGTGGATGG-3′, reverse: 5′-IndexTermTCTAGACGGCAGGTCAGGTC-3′.

Generation of E-cadherin reporter construct and luciferase reporter assay

Genomic DNA was isolated from MCF-10A cells and the E-cadherin promoter was amplified using the following primers: forward, 5′-(BglII) IndexTermCAAAAAATTAGGCTGCTAGCTC-3′ and reverse, 5′-(HindIII) IndexTermAATGCGTCCCTCGCAAGTCA-3′. MCF-10A/vector or MCF-10A/Twist2 cells were seeded in six-well plates at 2.5 × 104 cells per well. After 24 h, cells were transfected with 2 μg E-cadherin promoter luciferase plasmid in triplicate. After 48 h of transfection, cell lysates were prepared to measure luciferase activity.

Wound healing assay

Cells were seeded in 6-cm dishes at a density of 5 × 105. At 24 h later, a wound was incised in the central area of the confluent culture, followed by careful washing to remove detached cells and the addition of fresh medium. Phase contrast images of the wounded area were recorded using an inverted microscope at indicated time.

Flow cytometry

Identification of CD44high/CD24low cells was performed using monoclonal anti-CD44-fluorescein isothiocyanate (clone G44-26) and anti-CD24-PE (clone ML5) antibodies (BD Biosciences). Cells were labelled and CD44/CD24 markers were analysed using a FACSCalibur flow cytometer (BD Biosciences).

Mammosphere culture

Mammosphere culture was performed as described in Dontu et al. (2003) with a slight modification. Briefly, to induce sphere formation, cell cultures were dissociated to single cells by 0.05% trypsin–EDTA solution and plated into 24-well ultra-low attachment plates (Corning Life Sciences, Lowell, MA, USA) at a density of 4000 viable cells/ml. Cells were grown in a serum-free Dulbecco's modied Eagle's medium/F12 medium supplemented with B27 (Invitrogen), 20 ng/ml epidermal growth factor, 20 ng/ml basic fibroblast growth factor, 4 μg/ml heparin (Sigma-Aldrich) and 1% methyl cellulose. The mammospheres were cultured for 7–10 days.

Three-dimensional culture assay

Cells were grown in three-dimensional culture on a layer of growth factor-reduced Matrigel (BD Biosciences) in eight-well chamber slides. Indirect immunofluorescence staining and analysis were performed as outlined previously (Debnath et al., 2003).

Colony formation assay

Cells were seeded in six-well plates at a density of 200 cells/well and maintained in complete medium for 10–20 days. After most of the colonies had expanded to more than 50 cells, the cells were washed with PBS, fixed in methanol for 15 min and stained with crystal violet for 15 min. The plates were then photographed, and the colonies were counted. At least three independent experiments were carried out for each assay.

Tumour growth assay

To measure tumourigenicity in vivo in nude mice, an MCF-7/vector or MCF-7/Twist2 cell suspension (5 × 106 cells/ml) was mixed with an equal volume of Matrigel (BD Biosciences) and subcutaneously injected (200 μl total) into 4- to 6-week-old female BALB/c nude mice. These eight mice were treated weekly with 20 μl of a 10−2M ethanolic solution of E2 applied to the skin of the neck. Mice were killed at 4 weeks after injection and examined for the growth of subcutaneous tumours. All animals were used in accordance with institutional guidelines, and the current experiments were approved by the Use Committee for Animal Care and performed in the Cancer Research Centre, Xiamen University, China.

Immunohistochemical staining of breast cancer specimens

Tissue sections from 33 specimens of primary breast cancer, 12 specimens of matched non-cancer breast tissue and 8 specimens of matched breast metastatic tumours from the lymph node were fixed in 10% formaldehyde and embedded in paraffin. Sections were then cut and stained using immunohistochemistry as previously described (Bao et al., 2004). Briefly, paraffin-embedded tissue sections were processed for antigen retrieval by heating the sections in 10 mM sodium citrate (pH 6.0) at 95 °C for 20 min. Sections were immunostained with the monoclonal antibodies anti-Twist2 (1:200) or anti-E-cadherin (1:200). The immunostaining was performed with an ABC staining system (Santa Cruz Biotechnology, Santa Cruz, CA, USA) using an avidin-biotinylated-peroxidase detection method. Tissue procurement was approved by the institutional review board of Xiamen University.

Statistical analysis

The results of the experimental studies were expressed as the mean±s.d. Statistical differences were analysed by Student's t-test. P<0.05 was regarded as significant.

References

  1. Acloque H, Adams MS, Fishwick K, Bronner-Fraser M, Nieto MA . (2009). Epithelial-mesenchymal transitions: the importance of changing cell state in development and disease. J Clin Invest 119: 1438–1449.

    CAS  Article  Google Scholar 

  2. Al-Hajj M, Wicha MS, Benito-Hernandez A, Morrison SJ, Clarke MF . (2003). Prospective identification of tumorigenic breast cancer cells. Proc Natl Acad Sci USA 100: 3983–3988.

    CAS  Article  Google Scholar 

  3. Ansieau S, Bastid J, Doreau A, Morel AP, Bouchet BP, Thomas C et al. (2008). Induction of EMT by twist proteins as a collateral effect of tumor-promoting inactivation of premature senescence. Cancer Cell 14: 79–89.

    CAS  Article  Google Scholar 

  4. Ansieau S, Morel AP, Hinkal G, Bastid J, Puisieux A . (2010). TWISTing an embryonic transcription factor into an oncoprotein. Oncogene 29: 3173–3184.

    CAS  Article  Google Scholar 

  5. Bao S, Ouyang G, Bai X, Huang Z, Ma C, Liu M et al. (2004). Periostin potently promotes metastatic growth of colon cancer by augmenting cell survival via the Akt/PKB pathway. Cancer Cell 5: 329–339.

    CAS  Article  Google Scholar 

  6. Bao S, Wu Q, McLendon RE, Hao Y, Shi Q, Hjelmeland AB et al. (2006). Glioma stem cells promote radioresistance by preferential activation of the DNA damage response. Nature 444: 756–760.

    CAS  Google Scholar 

  7. Battula VL, Evans KW, Hollier BG, Shi Y, Marini FC, Ayyanan A et al. (2010). Epithelial-mesenchymal transition-derived cells exhibit multilineage differentiation potential similar to mesenchymal stem cells. Stem Cells 28: 1435–1445.

    CAS  Article  Google Scholar 

  8. Brabletz T, Jung A, Spaderna S, Hlubek F, Kirchner T . (2005). Opinion: migrating cancer stem cells—an integrated concept of malignant tumour progression. Nat Rev Cancer 5: 744–749.

    CAS  Article  Google Scholar 

  9. Bromberg J . (2002). Stat proteins and oncogenesis. J Clin Invest 109: 1139–1142.

    CAS  Article  Google Scholar 

  10. Chan KS, Sano S, Kiguchi K, Anders J, Komazawa N, Takeda J et al. (2004). Disruption of Stat3 reveals a critical role in both the initiation and the promotion stages of epithelial carcinogenesis. J Clin Invest 114: 720–728.

    CAS  Article  Google Scholar 

  11. Chen ZF, Behringer RR . (1995). Twist is required in head mesenchyme for cranial neural tube morphogenesis. Genes Dev 9: 686–699.

    CAS  Article  Google Scholar 

  12. Cheng GZ, Zhang WZ, Sun M, Wang Q, Coppola D, Mansour M et al. (2008). Twist is transcriptionally induced by activation of STAT3 and mediates STAT3 oncogenic function. J Biol Chem 283: 14665–14673.

    CAS  Article  Google Scholar 

  13. Debnath J, Muthuswamy SK, Brugge JS . (2003). Morphogenesis and oncogenesis of MCF-10A mammary epithelial acini grown in three-dimensional basement membrane cultures. Methods 30: 256–268.

    CAS  Article  Google Scholar 

  14. Dimri GP, Martinez JL, Jacobs JJ, Keblusek P, Itahana K, Van Lohuizen M et al. (2002). The Bmi-1 oncogene induces telomerase activity and immortalizes human mammary epithelial cells. Cancer Res 62: 4736–4745.

    CAS  PubMed  Google Scholar 

  15. Dontu G, Abdallah WM, Foley JM, Jackson KW, Clarke MF, Kawamura MJ et al. (2003). In vitro propagation and transcriptional profiling of human mammary stem/progenitor cells. Genes Dev 17: 1253–1270.

    CAS  Article  Google Scholar 

  16. Evdokimova V, Tognon C, Ng T, Ruzanov P, Melnyk N, Fink D et al. (2009). Translational activation of snail1 and other developmentally regulated transcription factors by YB-1 promotes an epithelial-mesenchymal transition. Cancer Cell 15: 402–415.

    CAS  Article  Google Scholar 

  17. Fernando RI, Litzinger M, Trono P, Hamilton DH, Schlom J, Palena C . (2010). The T-box transcription factor Brachyury promotes epithelial-mesenchymal transition in human tumor cells. J Clin Invest 120: 533–544.

    CAS  Article  Google Scholar 

  18. Franco HL, Casasnovas J, Rodríguez-Medina JR, Cadilla CL . (2011). Redundant or separate entities?—Roles of Twist1 and Twist2 as molecular switches during gene transcription. Nucleic Acids Res 39: 1177–1186.

    CAS  Article  Google Scholar 

  19. Gong XQ, Li L . (2002). Dermo-1, a multifunctional basic helix-loop-helix protein, represses MyoD transactivation via the HLH domain, MEF2 interaction, and chromatin deacetylation. J Biol Chem 277: 12310–12317.

    CAS  Article  Google Scholar 

  20. Gupta PB, Chaffer CL, Weinberg RA . (2009a). Cancer stem cells: mirage or reality? Nat Med 15: 1010–1012.

    CAS  Article  Google Scholar 

  21. Gupta PB, Onder TT, Jiang G, Tao K, Kuperwasser C, Weinberg RA et al. (2009b). Identification of selective inhibitors of cancer stem cells by high-throughput screening. Cell 138: 645–659.

    CAS  Article  Google Scholar 

  22. Hanahan D, Weinberg RA . (2011). Hallmarks of cancer: the next generation. Cell 144: 646–674.

    CAS  Article  Google Scholar 

  23. Iliopoulos D, Polytarchou C, Hatziapostolou M, Kottakis F, Maroulakou IG, Struhl K et al. (2009). MicroRNAs differentially regulated by Akt isoforms control EMT and stem cell renewal in cancer cells. Sci Signal 2: ra62.

    Article  Google Scholar 

  24. Isenmann S, Arthur A, Zannettino AC, Turner JL, Shi S, Glackin CA et al. (2009). TWIST family of basic helix-loop-helix transcription factors mediate human mesenchymal stem cell growth and commitment. Stem Cells 27: 2457–2468.

    CAS  Article  Google Scholar 

  25. Kalluri R, Weinberg RA . (2009). The basics of epithelial-mesenchymal transition. J Clin Invest 119: 1420–1428.

    CAS  Article  Google Scholar 

  26. Lee MS, Lowe G, Flanagan S, Kuchler K, Glackin CA . (2000). Human Dermo-1 has attributes similar to twist in early bone development. Bone 27: 591–602.

    CAS  Article  Google Scholar 

  27. Levy DE, Darnell Jr JE . (2002). Stats: transcriptional control and biological impact. Nat Rev Mol Cell Biol 3: 651–662.

    CAS  Article  Google Scholar 

  28. Li L, Cserjesi P, Olson EN . (1995). Dermo-1: a novel twist-related bHLH protein expressed in the developing dermis. Dev Biol 172: 280–292.

    CAS  Article  Google Scholar 

  29. Liu S, Dontu G, Mantle ID, Patel S, Ahn NS, Jackson KW et al. (2006). Hedgehog signaling and Bmi-1 regulate self-renewal of normal and malignant human mammary stem cells. Cancer Res 66: 6063–6071.

    CAS  Article  Google Scholar 

  30. Liu S, Dontu G, Wicha MS . (2005). Mammary stem cells, self-renewal pathways, and carcinogenesis. Breast Cancer Res 7: 86–95.

    CAS  Article  Google Scholar 

  31. Lo HW, Hsu SC, Xia W, Cao X, Shih JY, Wei Y et al. (2007). Epidermal growth factor receptor cooperates with signal transducer and activator of transcription 3 to induce epithelial-mesenchymal transition in cancer cells via up-regulation of TWIST gene expression. Cancer Res 67: 9066–9076.

    CAS  Article  Google Scholar 

  32. Maeda M, Johnson KR, Wheelock MJ . (2005). Cadherin switching: essential for behavioral but not morphological changes during an epithelium-to-mesenchyme transition. J Cell Sci 118: 873–887.

    CAS  Article  Google Scholar 

  33. Mani SA, Guo W, Liao MJ, Eaton EN, Ayyanan A, Zhou AY et al. (2008). The epithelial-mesenchymal transition generates cells with properties of stem cells. Cell 133: 704–715.

    CAS  Article  Google Scholar 

  34. McCoy EL, Iwanaga R, Jedlicka P, Abbey NS, Chodosh LA, Heichman KA et al. (2009). Six1 expands the mouse mammary epithelial stem/progenitor cell pool and induces mammary tumors that undergo epithelial-mesenchymal transition. J Clin Invest 119: 2663–2677.

    CAS  Article  Google Scholar 

  35. Morel AP, Lievre M, Thomas C, Hinkal G, Ansieau S, Puisieux A . (2008). Generation of breast cancer stem cells through epithelial-mesenchymal transition. PLoS One 3: e2888.

    Article  Google Scholar 

  36. Onder TT, Gupta PB, Mani SA, Yang J, Lander ES, Weinberg RA . (2008). Loss of E-cadherin promotes metastasis via multiple downstream transcriptional pathways. Cancer Res 68: 3645–3654.

    CAS  Article  Google Scholar 

  37. Ouyang G, Wang Z, Fang X, Liu J, Yang CJ . (2010). Molecular signaling of the epithelial to mesenchymal transition in generating and maintaining cancer stem cells. Cell Mol Life Sci 67: 2605–2618.

    CAS  Article  Google Scholar 

  38. Peinado H, Olmeda D, Cano A . (2007). Snail, Zeb and bHLH factors in tumour progression: an alliance against the epithelial phenotype? Nat Rev Cancer 7: 415–428.

    CAS  Article  Google Scholar 

  39. Polyak K, Weinberg RA . (2009). Transitions between epithelial and mesenchymal states: acquisition of malignant and stem cell traits. Nat Rev Cancer 9: 265–273.

    CAS  Article  Google Scholar 

  40. Ruan K, Bao S, Ouyang G . (2009). The multifaceted role of periostin in tumorigenesis. Cell Mol Life Sci 66: 2219–2230.

    CAS  Article  Google Scholar 

  41. Singh A, Settleman J . (2010). EMT, cancer stem cells and drug resistance: an emerging axis of evil in the war on cancer. Oncogene 29: 4741–4751.

    CAS  Article  Google Scholar 

  42. Song G, Ouyang G, Mao Y, Ming Y, Bao S, Hu T . (2009a). Osteopontin promotes gastric cancer metastasis by augmenting cell survival and invasion through Akt-mediated HIF-1alpha up-regulation and MMP9 activation. J Cell Mol Med 13: 1706–1718.

    Article  Google Scholar 

  43. Song LB, Li J, Liao WT, Feng Y, Yu CP, Hu LJ et al. (2009b). The polycomb group protein Bmi-1 represses the tumor suppressor PTEN and induces epithelial-mesenchymal transition in human nasopharyngeal epithelial cells. J Clin Invest 119: 3626–3636.

    CAS  Article  Google Scholar 

  44. Sullivan NJ, Sasser AK, Axel AE, Vesuna F, Raman V, Ramirez N et al. (2009). Interleukin-6 induces an epithelial-mesenchymal transition phenotype in human breast cancer cells. Oncogene 28: 2940–2947.

    CAS  Article  Google Scholar 

  45. Thiery JP, Acloque H, Huang RY, Nieto MA . (2009). Epithelial-mesenchymal transitions in development and disease. Cell 139: 871–890.

    CAS  Article  Google Scholar 

  46. Thiery JP, Sleeman JP . (2006). Complex networks orchestrate epithelial-mesenchymal transitions. Nat Rev Mol Cell Biol 7: 131–142.

    CAS  Article  Google Scholar 

  47. Thisse B, el Messal M, Perrin-Schmitt F . (1987). The twist gene: isolation of a Drosophila zygotic gene necessary for the establishment of dorsoventral pattern. Nucleic Acids Res 15: 3439–3453.

    CAS  Article  Google Scholar 

  48. Tsuji T, Ibaragi S, Shima K, Hu MG, Katsurano M, Sasaki A et al. (2008). Epithelial-mesenchymal transition induced by growth suppressor p12CDK2-AP1 promotes tumor cell local invasion but suppresses distant colony growth. Cancer Res 68: 10377–10386.

    CAS  Article  Google Scholar 

  49. Vesuna F, Lisok A, Kimble B, Raman V . (2009). Twist modulates breast cancer stem cells by transcriptional regulation of CD24 expression. Neoplasia 11: 1318–1328.

    CAS  Article  Google Scholar 

  50. Wellner U, Schubert J, Burk UC, Schmalhofer O, Zhu F, Sonntag A et al. (2009). The EMT-activator ZEB1 promotes tumorigenicity by repressing stemness-inhibiting microRNAs. Nat Cell Biol 11: 1487–1495.

    CAS  Article  Google Scholar 

  51. Yang J, Mani SA, Donaher JL, Ramaswamy S, Itzykson RA, Come C et al. (2004). Twist, a master regulator of morphogenesis, plays an essential role in tumor metastasis. Cell 117: 927–939.

    CAS  Article  Google Scholar 

  52. Yang J, Weinberg RA . (2008). Epithelial-mesenchymal transition: at the crossroads of development and tumor metastasis. Dev Cell 14: 818–829.

    CAS  Article  Google Scholar 

  53. Yang MH, Hsu DS, Wang HW, Wang HJ, Lan HY, Yang WH et al. (2010). Bmi1 is essential in Twist1-induced epithelial-mesenchymal transition. Nat Cell Biol 12: 982–992.

    Article  Google Scholar 

  54. Yang Y, Pan X, Lei W, Wang J, Shi J, Li F et al. (2006). Regulation of transforming growth factor-beta 1-induced apoptosis and epithelial-to-mesenchymal transition by protein kinase A and signal transducers and activators of transcription 3. Cancer Res 66: 8617–8624.

    CAS  Article  Google Scholar 

  55. Yin G, Chen R, Alvero AB, Fu HH, Holmberg J, Glackin C et al. (2010). TWISTing stemness, inflammation and proliferation of epithelial ovarian cancer cells through MIR199A2/214. Oncogene 29: 3545–3553.

    CAS  Article  Google Scholar 

  56. Yu M, Smolen GA, Zhang J, Wittner B, Schott BJ, Brachtel E et al. (2009). A developmentally regulated inducer of EMT, LBX1, contributes to breast cancer progression. Genes Dev 23: 1737–1742.

    CAS  Article  Google Scholar 

  57. Zhou J, Wulfkuhle J, Zhang H, Gu P, Yang Y, Deng J et al. (2007). Activation of the PTEN/mTOR/STAT3 pathway in breast cancer stem-like cells is required for viability and maintenance. Proc Natl Acad Sci USA 104: 16158–16163.

    CAS  Article  Google Scholar 

Download references

Acknowledgements

We thank Professor Kunxin Luo (Department of Molecular and Cell Biology, University of California, Berkeley), Professor Shideng Bao (Stem Cell Biology and Regenerative Medicine, Lerner Research Institute, Cleveland Clinic) and reviewers for critical comments. This work was supported by grants from the National Nature Science Foundation of China (no. 30871242, 31071302), National Basic Research Program of China (no. 2009CB941601, 2010CB732402), the Fundamental Research Funds for the Central Universities (no. 2010121095), the Science Planning Program of Fujian Province (2009J1010), the Outstanding Young Science Foundation of Fujian Province (no. 2010J06013) and the 985 Project grant from Xiamen University.

Author information

Affiliations

Authors

Corresponding author

Correspondence to G Ouyang.

Ethics declarations

Competing interests

The authors declare no conflict of interest.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Fang, X., Cai, Y., Liu, J. et al. Twist2 contributes to breast cancer progression by promoting an epithelial–mesenchymal transition and cancer stem-like cell self-renewal. Oncogene 30, 4707–4720 (2011). https://doi.org/10.1038/onc.2011.181

Download citation

Keywords

  • Twist2
  • epithelial–mesenchymal transition
  • cancer stem cell
  • breast cancer
  • metastasis

Further reading

Search

Quick links