Nathalie Sphyris

Epithelial-mesenchymal transition and cancer stem cells: a dangerously dynamic duo in breast cancer progression

Aberrant activation of a latent embryonic program - known as the epithelial-mesenchymal transition (EMT) - can endow cancer cells with the migratory and invasive capabilities associated with metastatic competence. The induction of EMT entails the loss of epithelial characteristics and the de novo acquisition of a mesenchymal phenotype. In breast cancer, the EMT state has been associated with cancer stem cell properties including expression of the stem cell-associated CD44+/CD24-/low antigenic profile, self-renewal capabilities and resistance to conventional therapies. Intriguingly, EMT features are also associated with stem cells isolated from the normal mouse mammary gland and human breast reduction tissues as well as the highly aggressive metaplastic and claudin-low breast tumor subtypes. This has implications for the origin of these breast tumors as it remains unclear whether they derive from cells that have undergone EMT or whether they represent an expansion of a pre-existing stem cell population that expresses EMT-associated markers to begin with. In the present review, we consider the current evidence connecting EMT and stem cell attributes and discuss the ramifications of these newly recognized links for our understanding of the emergence of distinct breast cancer subtypes and breast cancer progression.

Epigenetic silencing of microRNA-203 is required for EMT and cancer stem cell properties

The epithelial-mesenchymal transition (EMT) imparts metastatic competence on otherwise non-metastatic cancer cells through decreased inter-cellular adhesions, increased migratory capacity, stem cell properties and anoikis and chemotherapy resistance. In this study, we profiled changes in microRNA expression during EMT in conjunction with changes in DNA methylation at microRNA promoters to discover essential mediators of EMT-imparted stemness properties. MicroRNA-203 (miR-203) expression is repressed following EMT induced by multiple different stimuli and in established claudin-low cell lines as well as the CD44hi/CD24lo stem cell-enriched fraction. Expression of miR-203 in mesenchymal cells compromises migratory and invasive capacity in vitro, and tumor initiation and metastasis in vivo. Unexpectedly, miR-203 expression affects the sphere-forming capacity of neighboring cells by indirectly enhancing expression of DKK1, a secreted inhibitor of Wnt signaling and stemness resulting in suppression of β-catenin protein levels. Our data suggest that restoring miR-203 expression levels may inhibit metastasis and combat deregulated Wnt signaling.

Improved retention of zymogen granules in cultured murine pancreatic acinar cells and induction of acinar-ductal transdifferentiation in vitro

Objectives: To establish a primary culture model for murine pancreatic acinar cells and to investigate the effects of different culture conditions on the phenotype and plasticity of these cells in extended culture. Methods: Acinar cells, cultured in Chee (CME) or Waymouth/Ham F-12 (WHME) media, exhibited 2 markedly dissimilar phenotypes. We employed 5′-bromo-2′-deoxyuridine (BrdU) incorporation, immunocytochemistry, and electron microscopy to investigate differences in cell cycle status and phenotype. Results: CME-cultured cells grew as discrete epithelial islands and retained zymogen granules and endoplasmic reticulum stacks, yet expressed cytokeratin 7, suggesting that they comprise an intermediate between the acinar and ductal cell types. Observed by time-lapse videomicroscopy, cells transferred to WHME formed a confluent monolayer, flattened and dedifferentiated to a duct-like phenotype typified by loss of secretory apparatus, variable β-catenin expression, an elongated teardrop shape, increased cell size, and nuclear pleomorphism. Transition between phenotypes did not involve apoptosis as assessed by morphologic criteria in Feulgen-stained cultures. The flattened cells exhibited increased BrdU incorporation and mitotic index, suggesting that dedifferentiation precedes the capacity for increased cell cycle entry, while the appearance of meganuclei is consistent with amplified DNA content. We also demonstrate greatly improved gene delivery to cultured acinar cells by adenovirus-mediated transduction compared with lipid-mediated transfection. Conclusions: This in vitro model confirms the plasticity of acinar cells and may serve to delineate the changes underlying acinar cell dedifferentiation and acquisition of a duct-like pluripotent phenotype.

GD3 synthase regulates epithelial–mesenchymal transition and metastasis in breast cancer

The epithelial–mesenchymal transition (EMT) bestows cancer cells with increased stem cell properties and metastatic potential. To date, multiple extracellular stimuli and transcription factors have been shown to regulate EMT. Many of them are not druggable and therefore it is necessary to identify targets, which can be inhibited using small molecules to prevent metastasis. Recently, we identified the ganglioside GD2 as a novel breast cancer stem cell marker. Moreover, we found that GD3 synthase (GD3S)—an enzyme involved in GD2 biosynthesis—is critical for GD2 production and could serve as a potential druggable target for inhibiting tumor initiation and metastasis. Indeed, there is a small molecule known as triptolide that has been shown to inhibit GD3S function. Accordingly, in this manuscript, we demonstrate that the inhibition of GD3S using small hairpin RNA or triptolide compromises the initiation and maintenance of EMT instigated by various signaling pathways, including Snail, Twist and transforming growth factor-β1 as well as the mesenchymal characteristics of claudin-low breast cancer cell lines (SUM159 and MDA-MB-231). Moreover, GD3S is necessary for wound healing, migration, invasion and stem cell properties in vitro. Most importantly, inhibition of GD3S in vivo prevents metastasis in experimental as well as in spontaneous syngeneic wild-type mouse models. We also demonstrate that the transcription factor FOXC2, a central downstream effector of several EMT pathways, directly regulates GD3S expression by binding to its promoter. In clinical specimens, the expression of GD3S correlates with poor prognosis in triple-negative human breast tumors. Moreover, GD3S expression correlates with activation of the c-Met signaling pathway leading to increased stem cell properties and metastatic competence. Collectively, these findings suggest that the GD3S-c-Met axis could serve as an effective target for the treatment of metastatic breast cancers.

Inhibition of FOXC2 restores epithelial phenotype and drug sensitivity in prostate cancer cells with stem-cell properties

Advanced prostate adenocarcinomas enriched in stem-cell features, as well as variant androgen receptor (AR)-negative neuroendocrine (NE)/small-cell prostate cancers are difficult to treat, and account for up to 30% of prostate cancer-related deaths every year. While existing therapies for prostate cancer such as androgen deprivation therapy (ADT), destroy the bulk of the AR-positive cells within the tumor, eradicating this population eventually leads to castration-resistance, owing to the continued survival of AR-/lo stem-like cells. In this study, we identified a critical nexus between p38MAPK signaling, and the transcription factor Forkhead Box Protein C2 (FOXC2) known to promote cancer stem-cells and metastasis. We demonstrate that prostate cancer cells that are insensitive to ADT, as well as high-grade/NE prostate tumors, are characterized by elevated FOXC2, and that targeting FOXC2 using a well-tolerated p38 inhibitor restores epithelial attributes and ADT-sensitivity, and reduces the shedding of circulating tumor cells in vivo with significant shrinkage in the tumor mass. This study thus specifies a tangible mechanism to target the AR-/lo population of prostate cancer cells with stem-cell properties.

p53 deficiency exacerbates pleiotropic mitotic defects, changes in nuclearity and polyploidy in transdifferentiating pancreatic acinar cells

In a primary culture model for pancreatic acinar–ductal transdifferentiation, cells exhibited increased proliferation, changes in nuclearity and polyploidy. We identify the ‘nucleus to centrosome’ ratio of the progenitor cell, the dissemination of centrosomes at spindle poles and cytokinesis failure as critical determinants of mitosis outcome and centrosome inheritance. Abortive cytokinesis of mononuclear cells contributes to the binuclear cell pool, whereas enclosure of entire mitotic formations, within a single nuclear envelope, perpetuates polyploidization. Binuclear cell nuclei combine their genomes on a single metaphase plate, doubling descendant ploidy. Moreover, ∼42% of binuclear and tetraploid cells assemble aberrant spindles with up to 8 centrosomes/poles. These phenotypes were exacerbated in p53-deficient cultures exhibiting increased S-phase entry, giant nuclei, multinucleation, multipolar mitoses and centrosome hyperamplification. The tendency of p53-proficient cells to spontaneously evade the tetraploidy checkpoint degenerates to uncontrolled polyploid progression in p53-deficient cultures, explaining why p53 abrogation alone rapidly descends to aneuploidy in this system. We detected constitutively nuclear mdm2, which may circumvent endogenous cell-cycle checkpoints, and pronounced accumulation of p21 and p27 in multinuclear cells and giant nuclei, consistent with roles in polyploidization. This in vitro model may recapitulate the processes underlying genomic instability in pancreatic tumours in vivo, and attests to the existence of a p53-dependent polyploidy checkpoint acting to limit the degree of polyploidization.

EMT-induced metabolite signature identifies poor clinical outcome

Metabolic reprogramming is a hallmark of cancer. Epithelial-mesenchymal transition (EMT) induces cancer stem cell (CSC) characteristics and promotes tumor invasiveness; however relatively little is known about the metabolic reprogramming in EMT. Here we show that breast epithelial cells undergo metabolic reprogramming following EMT. Relative to control, cell lines expressing EMT transcription factors show ≥1.5-fold accumulation of glutamine, glutamate, beta-alanine and glycylleucine as well as ≥1.5-fold reduction of phosphoenolpyruvate, urate, and deoxycarnitine. Moreover, these metabolic alterations were found to be predictive of overall survival (hazard ratio = 2.3 (95% confidence interval: 1.31–4.2), logrank p-value = 0.03) and define breast cancer molecular subtypes. EMT-associated metabolites are primarily composed of anapleurotic precursors, suggesting that cells undergoing EMT have a shift in energy production. In summary, we describe a unique panel of metabolites associated with EMT and demonstrate that these metabolites have the potential for predicting clinical and biological characteristics associated with patient survival.

Phosphorylation of serine 367 of FOXC2 by p38 regulates ZEB1 and breast cancer metastasis, without impacting primary tumor growth

Metastatic competence is contingent upon the aberrant activation of a latent embryonic program, known as the epithelial–mesenchymal transition (EMT), which bestows stem cell properties as well as migratory and invasive capabilities upon differentiated tumor cells. We recently identified the transcription factor FOXC2 as a downstream effector of multiple EMT programs, independent of the EMT-inducing stimulus, and as a key player linking EMT, stem cell traits and metastatic competence in breast cancer. As such, FOXC2 could serve as a potential therapeutic target to attenuate metastasis. However, as FOXC2 is a transcription factor, it is difficult to target by conventional means such as small-molecule inhibitors. Herein, we identify the serine/threonine-specific kinase p38 as a druggable upstream regulator of FOXC2 stability and function that elicits phosphorylation of FOXC2 at serine 367 (S367). Using an orthotopic syngeneic mouse tumor model, we make the striking observation that inhibition of p38-FOXC2 signaling selectively attenuates metastasis without impacting primary tumor growth. In this model, circulating tumor cell numbers are significantly reduced in mice treated with the p38 inhibitor SB203580, relative to vehicle-treated counterparts. Accordingly, genetic or pharmacological inhibition of p38 decreases FOXC2 protein levels, reverts the EMT phenotype and compromises stem cell attributes in vitro. We also identify the EMT-regulator ZEB1—known to directly repress E-cadherin/CDH1—as a downstream target of FOXC2, critically dependent on its activation by p38. Consistent with the notion that activation of the p38-FOXC2 signaling axis represents a critical juncture in the acquisition of metastatic competence, the phosphomimetic FOXC2(S367E) mutant is refractory to p38 inhibition both in vitro and in vivo, whereas the non-phosphorylatable FOXC2(S367A) mutant fails to elicit EMT and upregulate ZEB1. Collectively, our data demonstrate that FOXC2 regulates EMT, stem cell traits, ZEB1 expression and metastasis in a p38-dependent manner, and attest to the potential utility of p38 inhibitors as antimetastatic agents.

pIgR: Frenemy of Inflammation, EMT, and HCC Progression

Hepatocellular carcinoma (HCC) is the fifth most common neoplasm worldwide and a major cause of cancer-related death. The development of HCC has long been associated with inflammation-causing agents such as chronic viral infection with hepatitis B or C, alcoholic cirrhosis, or dietary exposure to fungal aflatoxins (1–3). Consequently, HCC progression unravels against a backdrop of persistent inflammation, extensive tissue remodeling, and excessive deposition of extracellular matrix components (2–4). Moreover, recently gained insights have linked inflammation to the aberrant activation of a latent embryonic program—termed the epithelial–mesenchymal transition (EMT)—that endows tumor cells with metastatic competence (5–7) and resistance to therapy (8). EMT is a complex process that enables the reprogramming of polarized epithelial cells toward a mesenchymal phenotype accompanied by shedding of epithelial characteristics, loss of apico-basal polarity, dissolution of intercellular contacts, and gain of intrinsic migratory and invasive capabilities (7). Both persistent inflammation and EMT have been independently implicated in wound healing and regeneration following tissue injury and in pathological conditions such as organ fibrosis and metastasis (7,9,10). Indeed, the induction of an inflammatory response plays dual and opposing roles in the context of tumor development. Initially, inflammation and immune surveillance serve to eliminate rogue premalignant or malignant cells, thus suppressing tumor formation. However, as tumors evolve, they not only evade immune surveillance but—somewhat paradoxically—provoke an inflammatory response, resulting in the recruitment of multiple immune cell types that secrete a diverse set of signaling molecules that promote cell proliferation and survival of resident cells and remodel the extracellular matrix to favor EMT. Accordingly, several inflammatory stimuli have been shown to activate and stabilize EMT-inducing transcription factors, thus providing a molecular basis for the links between inflammation and the EMT process (9,11). In this issue of the Journal, Ai et al. (12) identify the polymeric immunoglobulin receptor (pIgR)—a key inflammatory mediator—as a prognostic biomarker for HCC and a molecular player in hepatitis B infection, chronic liver inflammation, the induction of EMT, HCC recurrence, and metastatic progression. Whereas pIgR aberrant expression has long been associated with HCC (13), its relevance to malignancy has remained unclear. Thus, to date, the only known function of pIgR is in mediating transcytosis of polymeric immunoglobulins from the basolateral to the apical surface of epithelia, ultimately facilitating the secretion of IgA and IgM, which comprise the first-line of defense against infection (14). The study by Ai et al. (12) reveals previously unrecognized roles for pIgR by demonstrating that pIgR overexpression is capable of eliciting EMT in MDCK cells as well as in immortalized or transformed hepatic cell lines. Conversely, pIgR–small hairpin RNA knockdown restored epithelial characteristics in tumor necrosis factor-α–treated HT29 cells and MDCK cells ectopically expressing pIgR. In vivo, pIgR-overexpressing cells yielded a higher number of experimental lung metastases compared with control counterparts, confirming that pIgR overexpression can promote colonization. Consistent with EMT, Ai et al. (12) detected decreased levels of epithelial markers (E-cadherin, cytokeratins) and increased levels of the mesenchymal marker, vimentin, and phospho-Smad2/3 in pIgR-overexpressing HCC specimens. At the mechanistic level, Ai et al. (12) implicate pIgR in the EMT initiated by cross talk of transforming growth factor-β (TGF-β) with inflammatory mediators (tumor necrosis factor-α, interferon-γ, interleukin-4). Thus, they demonstrated that pIgR overexpression enhances Smad2/3 nuclear translocation following TGF-β/cytokine treatment and identified pIgR as a novel partner of the Smad complex that activates Smad signaling by recruiting Smad2. It is well established that TGF-β functions as a tumor suppressor early in tumorigenesis, whereas in later stages of carcinogenesis, it exacerbates tumor progression by promoting immune evasion and angiogenesis (15,16). Furthermore, loss of key TGF-β signaling mediators (eg, Smad4) enables tumor cells to become refractory to cytostasis and primed for EMT (16). For example, Battaglia et al. (17) showed that the hepatitis C virus core protein decreases Smad3 activation in hepatocytes, switching the TGF-β response from cytostasis to EMT. The study by Ai et al. (12) is the first demonstration of a host immunoglobulin receptor that synergizes with TGF-β/Smad signaling and the inflammatory milieu to engage EMT, thus bestowing metastatic competence upon disseminating HCC cells (12). The study by Ai et al. (12) ascribes novel functions to pIgR but also raises intriguing questions that warrant further investigation: 1) Given that pIgR is expressed on the surface of several glandular epithelia, including those of the liver, intestine, and breast (14), does pIgR activation play similar roles in other carcinomas known to be exacerbated by inflammation? 2) Is the expression of pIgR necessary for TGF-β-induced EMT? 3) How does pIgR modulate the transcriptional output of Smad signaling? 4) Given that Smad2/3 and pIgR interact at the early endosome, does the entire pIgR-Smad complex translocate to the nucleus and participate in target gene regulation? Or, does Smad4 displace pIgR from the pIgR-Smad complex before nuclear translocation? 5) Given that pIgR potentiates Smad-mediated EMT through a physical interaction with Smad2, does pIgR activation affect Smad-independent noncanonical TGF-β signaling cascades (18)? 6) Which kinases and/or phosphatases regulate pIgR activation, in response to inflammation, thus influencing Smad recruitment and downstream signaling? 7) What additional proteins—if any—comprise the pIgR-Smad complex (eg, SARA, Smad4) (16,19)? 8) Is pIgR-Smad complex formation necessary for the EMT programs induced by other inflammatory cytokines? Furthermore, although Ai et al. (12) demonstrate the enhanced colonization ability of pIgR-overexpressing cells when introduced via injection into the tail vein, additional experiments are necessary to determine its effects on spontaneous metastasis originating from orthotopic or autochthonous tumor models. The generation of mouse models specifically and temporally expressing pIgR in the liver or, indeed, knock-in mice in which pIgR activation is disrupted by mutation of serines S682 and/or S734 (to alanine) will be useful to investigate the pathogenic consequences of pIgR-mediated EMT in distinct liver cell types in situ. Moreover, examining the relevance of pIgR in established models of HCC and liver fibrosis (eg, bile duct ligation, carbon tetrachloride) will enable lineage tracing and serve to determine which pIgR-expressing cell types may contribute to liver disease, be it fibrosis or malignancy. Importantly, given that HCCs are highly resistant to chemotherapy (3,20) and EMT has been linked to increased liver cell survival post-cytotoxic insult (20), such tractable models will address whether and how pIgR inhibition may modulate resistance to therapy. Given the recently established connection between EMT and stem cell properties (21,22), the findings of Ai et al. (12) raise the intriguing possibility that pIgR may also affect stemness. This is especially important in view of the emerging literature on the roles of EMT and stem cells in HCC progression, heterogeneity, and resistance to therapy (20,23–25). A key finding of this study (12) is that different and separable portions of pIgR are important in mediating its transcytosis functions and the induction of EMT through the recruitment of R-Smads. Thus, mutations in tyrosine and serine residues (Y677F and Y743F, S673A, and S735A) that are important for transcytosis do not seem to affect pIgR-induced EMT, but mutations in two serine residues (S682A and S734A) unrelated to transcytosis attenuate the ability of pIgR to elicit EMT. These results may guide the development of therapies specifically directed toward the inhibition of pIgR-elicited EMT without affecting its roles in transcytosis. Thus, treatment strategies directed at inhibiting pIgR may help curb recurrent HCC or metastasis following tumor resection and may improve patient responsiveness to immune-based therapies as they may alleviate the immunosuppressive effects of TGF-β. Whether inhibiting pIgR can indeed be achieved without eliminating its functions in transcytosis and antigen presentation and whether such agents will interfere with the physiological roles of TGF-β in normal tissue homeostasis or the immune system remains to be seen.

Mathematical modelling of phenotypic plasticity and conversion to a stem-cell state under hypoxia

Hypoxia, or oxygen deficiency, is known to be associated with breast tumour progression, resistance to conventional therapies and poor clinical prognosis. The epithelial-mesenchymal transition (EMT) is a process that confers invasive and migratory capabilities as well as stem cell properties to carcinoma cells thus promoting metastatic progression. In this work, we examined the impact of hypoxia on EMT-associated cancer stem cell (CSC) properties, by culturing transformed human mammary epithelial cells under normoxic and hypoxic conditions, and applying in silico mathematical modelling to simulate the impact of hypoxia on the acquisition of CSC attributes and the transitions between differentiated and stem-like states. Our results indicate that both the heterogeneity and the plasticity of the transformed cell population are enhanced by exposure to hypoxia, resulting in a shift towards a more stem-like population with increased EMT features. Our findings are further reinforced by gene expression analyses demonstrating the upregulation of EMT-related genes, as well as genes associated with therapy resistance, in hypoxic cells compared to normoxic counterparts. In conclusion, we demonstrate that mathematical modelling can be used to simulate the role of hypoxia as a key contributor to the plasticity and heterogeneity of transformed human mammary epithelial cells.