The relevance of the TGF- Paradox to EMT-MET programs

  • Published on

  • View

  • Download


  • T. Se We


    grotione co

    grams stimulated by TGF-b, they readily display a variety of acquired phenotypes that provide a selective

    F-b) is-relate

    of human pathologies, including brosis, autoimmune diseases,

    cally exerts tumor suppressing activities in normal cells, as wellas in early-stage carcinomas through its ability to induce cell cyclearrest and apoptotic reactions. Interestingly, as carcinomas con-tinue to evolve and ultimately acquire metastatic phenotypes,the tumor suppressing functions of TGF-b are circumvented such

    otypes [5,6naling ulti

    converges in the nucleus to regulate the expression and activity

    taining EMT reactions. Amongst the EMT transcription factors tar-geted by TGF-b are members of the Snail (SNAI1 and SNAI2/Slug),ZEB (ZEB1 and ZEB2/SIP1), basic helix-loop-helix (Twist1 andTwist2), Six family of homeobox (Six1), and Forkhead (FOXC2), aswell as members of the High Mobility Group proteins (HMG2a),which modify DNA structure to enhance transcription factor bind-ing [3]. Recently, EMT reactions have been subcategorized intothree distinct programs, including (i) Type 1 EMT, which transpiresduring embryonic development of the endocardial cushion, neural

    Corresponding author. Tel.: +1 216 368 5763; fax: +1 216 368 1166.E-mail address: (W.P. Schiemann).

    Cancer Letters xxx (2013) xxxxxx

    Contents lists available at


    .e l1 These authors equally contributed to this work.and cancer [1]. With respect to human malignancies, TGF-b typi- of a variety of master EMT transcription factors operant in main-topoietic cell lineages [3,4]. Aberrant TGF-b signaling has also beenlinked to the initiation, development, and progression of a variety

    quired migratory, invasive, and metastatic phening its induction of EMT programs, TGF-b sig0304-3835/$ - see front matter 2013 Published by Elsevier Ireland Ltd.

    Please cite this article in press as: C.D. Morrison et al., The relevance of the TGF-b Paradox to EMT-MET programs, Cancer Lett. (2013), http://dx.d10.1016/j.canlet.2013.02.048]. Dur-matelycytokines, including the activins and inhibins, bone morphogeneticproteins (BMPs), Nodal, and growth and differentiation factors[1,2]. TGF-b regulates a vast array of physiological processes rang-ing from embryonic development and tissue morphogenesis to celldifferentiation and survival to cellular and organ homeostasis; italso inhibits the proliferation of epithelial, endothelial, and hema-

    programs [25], which reect an evolutionarily conserved cascadein which polarized epithelial cells adopt mesenchymal cell charac-teristics that include (i) reduced apicobasolateral polarity and celladhesion; (ii) enhanced chemoresistance and evasion from hostimmunosurveillance; (iii) expanded stem-like and tumor-initiatingactivities; (iv) elevated resistance to apoptotic stimuli; and (v) ac-1. Introduction

    Transforming growth factor-b (TGber of a large family of structurallyadvantage to growing carcinomas, including (i) enhanced cell migration and invasion; (ii) heightenedresistance to cytotoxic agents, targeted chemotherapeutic, and radiation treatments; and (iv) boostedexpansion of cancer-initiating and stem-like cell populations that underlie tumor metastasis and diseaserecurrence. At present, the molecular, cellular, and microenvironmental mechanisms that enable post-EMT and metastatic carcinoma cells to hijack the oncogenic activities of TGF-b remain incompletelyunderstood. Additionally, the molecular mechanisms that counter EMT programs and limit theaggressiveness of late-stage carcinomas, events that transpire via mesenchymal-epithelial transition(MET) reactions, also need to be further elucidated. Here we review recent advances that provide newinsights into how TGF-b promotes EMT programs in late-stage carcinoma cells, as well as how theseevents are balanced by MET programs during the development and metastatic progression of humancarcinomas.

    2013 Published by Elsevier Ireland Ltd.

    the prototypical mem-d and multifunctional

    that TGF-b is utilized as an oncogenic factor that promotes carci-noma growth, invasion, and metastasis. The paradoxical switchin TGF-b function during tumorigenesis has been linked to the abil-ity of TGF-b to induce epithelial-mesenchymal transition (EMT)Keywords:Epithelial plasticity

    whose manifestations are intimately linked to the initiation of epithelial-mesenchymal transition (EMT)programs in developing and progressing carcinomas. Indeed, as carcinoma cells emerge from EMT pro-Mini-review

    The relevance of the TGF-b Paradox to EM

    Chevaun D. Morrison 1, Jenny G. Parvani 1, William PCase Comprehensive Cancer Center, Division of General Medical Sciences-Oncology, CasOH 44106, United States

    a r t i c l e i n f o

    Article history:Available online xxxx

    a b s t r a c t

    The role of transformingreecting its ability to funcpromoter in their late-stag


    journal homepage: www-MET programs

    chiemann stern Reserve University, Wolstein Research Building, 2103 Cornell Road Cleveland,

    wth factor-b (TGF-b) during tumorigenesis is complex and paradoxical,as a tumor suppressor in normal and early-stage cancers, and as a tumor

    unterparts. The switch in TGF-b function is known as the TGF-b Paradox,

    SciVerse ScienceDirect


    sevier .com/locate /

  • ability of Smad2/3/4 complexes to interact physically with a vari-ety of transcriptional activators or repressors that are expressed

    pressors or tumor promoters, and as such, miRNA expression istypically dysregulated in numerous human diseases, including

    er Lein a cell- and context-specic manner. Canonical Smad2/3 signal-ing is also regulated by their being targeted by several adaptermolecules, including SARA, Hgs, and Dab2 [5,11], and by the inhib-itory Smad, Smad7, which inhibits Smad2/3 phosphorylation byTbR-I and induces its internalization and degradation [5,11].

    Besides its ability to activate Smad2/3, TGF-b also alters cellbehavior by stimulating an ever expanding list of noncanonicalTGF-b effectors, including several nonreceptor protein tyrosine ki-nases (e.g., Src and FAK), mediators of cell survival (e.g., NF-jB andPI3K/Akt), mediators of immune responses (e.g., TAK1/TAB andTRAF6), small GTP-binding proteins (e.g., Ras, Rho, and Rac1), andMAP kinases (e.g., ERK1/2, p38 MAPK, and JNK), and; Fig. 1)[2,11,18]. Moreover, crosstalk between various noncanonicalTGF-b effectors has been shown to drive the oncogenic activitiesof this multifunctional cytokine. Indeed, we recently dened acrest, and closure and fusion of the palate; (ii) Type 2 EMT, whichtranspires during tissue remodeling, wound healing, and brosis;and (iii) Type 3 EMT, which transpires during tumor metastasis[7]. In addition, EMT programs are countered and reversed bymesenchymal-epithelial transitions, which also play essential rolesduring embryogenesis and tissue morphogenesis, as well as duringcarcinoma progression and metastatic outgrowth [8]. TGF-b is amaster regulator of all EMT subtypes and readers desiring amore thorough summary of the mechanisms whereby TGF-bdrives EMT programs are directed to several comprehensive re-views [5,9,10]. Here we discuss recent ndings related to the par-adoxical role of TGF-b in regulating oncogenic Type 3 EMT-METprograms, as well as its function in creating EMT-permissivemicroenvironments during carcinoma development and metastaticprogression.

    2. TGF-b signaling

    Mammals express three genetically distinct TGF-b ligands (TGF-bs 13), whose mature and biologically active forms are 97%identical and exhibit redundant activities in vitro [2,11]. TGF-b sig-naling commences upon binding to its high-afnity receptors,namely the TGF-b type I (TbR-I), type II (TbR-II), and type III(TbR-III or betaglycan) receptors [2,11]. TbR-I and TbR-II both con-tain serine/threonine protein kinases in their cytoplasmic domainsthat produce intracellular signals in response to TGF-b. In contrast,the cytoplasmic domain of TbR-III lacks intrinsic protein kinaseactivity; however, this TGF-b receptor is highly expressed in cellsand modulates the binding and presentation of TGF-b to its signal-ing receptors, as well as functions as a tumor suppressor in a vari-ety of tissues, including the breast, ovary, prostate, lung, pancreas,and kidney [12]. The binding of TGF-b to TbR-II results in its trans-phosphorylation and activation of TbR-I, which phosphorylates andactivates the latent transcription factors, Smad2 and Smad3(Fig. 1). Once activated, Smad2/3 form heteromeric complexes withthe common Smad, Smad4, which accumulate en masse in the nu-cleus to govern gene transcription, an event referred to as canon-ical TGF-b signaling. Recent evidence also indicates that TGF-breceptors can activate the BMP-regulated Smads, Smad1/5/8(Fig. 1), leading to the acquisition of migratory and invasive pheno-types in carcinomas [13], and to the induction of proliferative andmigratory phenotypes in endothelial cells (Fig. 1) [1417]. The pre-cise mechanisms and functional consequences of this unconven-tional coupling remain to be elucidated. Importantly, thediversity of canonical TGF-b signaling is mediated in part via the

    2 C.D. Morrison et al. / CancTbRI:xIAP:TAB1:TAK1:IKKb signaling complex that activates NF-jB and promotes Cox-2 expression that is essential in elicitingEMT programs by TGF-b [1921]. In addition, others demonstrated

    Please cite this article in press as: C.D. Morrison et al., The relevance of the TG10.1016/j.canlet.2013.02.048cancer [26,27]. Recently, aberrant miRNA expression, includingthose responsive to TGF-b [2629], has been linked to the initiationof the TGF-b Paradox and its manifestation of oncogenic TGF-bsignaling [30]. Indeed, the coupling of oncogenic TGF-b to miRNAswas rst described by Goodall and colleagues who found TGF-b toinhibit the expression of miR-200 family members, which functionin suppressing the expression of the EMT transcription factors,ZEB1 and ZEB2 (SIP1) [31,32]. During EMT and metastasis stimu-lated by TGF-b, the loss of miR-200 family members leads to thestabilization of ZEB1 and ZEB2, resulting in their binding to andrepression of the Cdh1 (E-cadherin) promoter [32]. More recently,aberrant miR-200 family member expression has been observedto induce a powerful feed-forward signaling loop between TGF-band miR-200 family members to drive oncogenic EMT and carci-noma metastasis [33]. Finally, two recent studies demonstratedthat highly metastatic 4T1 breast cancer cells are more epithe-lial-like as compared to their isogenic and nonmetastatic 4T07counterparts [34,35]. Amongst the many unique differences be-tween these two isogeneic cell types is the re-expression of miR-200 in metastatic 4T1 cells, leading to the synthesis and secretionof metastasis-promoting proteins necessary for metastatic out-growth [35]. Collectively, these ndings establish miR-200 familymembers as tumor suppressors capable of preventing oncogenicTGF-b signaling and its induction of EMT in normal epithelial cellsa physical interaction between TRAF6 and TbRI that activates TAK1,leading to the initiation of JNK and p38 MAPK signaling [22,23].TRAF6 also ubiquitinates TbRI, which promotes its cleavage byTACE and the accumulation of the intracellular domain of TbR-I(ICD) to accumulate in the nucleus where it functions in conjunc-tion with other regulators of transcription to alter gene expression[24]. It should be noted that both canonical and noncanonical TGF-b effectors participate in driving EMT programs by TGF-b. Indeed,Smads 3 and 4 function as positive regulators of EMT programs,while Smad2 functions as a negative regulator of EMT [10,25].Along these lines, noncanonical TGF-b effectors are essential in ini-tiating EMT reactions and maintaining phenotypes in cells thathave emerged from EMT programs [2,11,18]. Although the precisemechanisms that coordinate the dynamics between the canonicaland noncanonical TGF-b signaling systems remain to be fully elu-cidated, recent evidence has shown that imbalances between thesetwo branches of the TGF-b signaling system, particularly thosedominated by noncanonical TGF-b effectors, underscores theTGF-b Paradox and the acquisition of oncogenic activity byTGF-b in developing and progressing carcinomas [5,11]. In the suc-ceeding sections, we highlight recent ndings that depict the func-tion of TGF-b and its effector molecules as potent promoters ofcarcinoma EMT-MET programs and their role in mediating meta-static progression (Fig. 2).

    3. TGF-b and post-transcriptional regulation of EMT

    3.1. TGF-b, microRNAs, and EMT

    MicroRNAs (miRNAs) are small, noncoding RNAs that post-transcriptionally control gene expression by inhibiting mRNAtranslation, or by promoting mRNA degradation [26,27]. Similarto the vast array of biological activities regulated by TGF-b, miRNAsalso oversee a wide variety of physiological processes, such as cellproliferation, differentiation, and apoptosis. Additionally, accumu-lating evidence demonstrates that miRNAs can serve as tumor sup-

    tters xxx (2013) xxxxxx[36].Recently, the homeobox transcription factor Six1 has been

    shown to transform mammary epithelial cells and promote their

    F-b Paradox to EMT-MET programs, Cancer Lett. (2013),

  • er LeC.D. Morrison et al. / Cancmetastasis by facilitating the conversion of TGF-b from a tumorsuppressor to a tumor promoter [3739]. The enhancement inTGF-b signaling mediated by Six1 transpires in part via its induc-tion of the miR-106b-25 cluster, which represses Smad7 expres-sion. The net-effect of this event results in elevated TbR-Iexpression that not only promotes EMT programs in response toheightened autocrine TGF-b signaling, but also expands the popu-lation of tumor-initiating cells that signicantly shortens the timeto disease relapse [40]. Along these lines, miR-29a is highly ex-pressed in mammary epithelial cells undergoing EMT reactions.Interestingly, miR-29a downregulates the expression of triste-traproline, which normally functions to degrade mRNAs whose3-UTRs are rich in AU-repeats. When combined with oncogenicRas signaling, miR-29a-mediated repression of tristetraprolineexpression promotes EMT and metastasis of mammary epithelialcells in mouse models, and associates with the appearance of inva-sive ductal carcinomas in human breast cancers [41].

    TGF-b and its activation of Smad4 also couple to the expressionof miR-155 in mammary epithelial cells, which downregulatestheir expression of RhoA and promotes tight junction disassembly[42]. Along these lines, TGF-b also inhibits RhoA activity by induc-ing the expression of miR-24, which represses the expression of theRho guanine nucleotide exchange factor (GEF), Net1 [43]. Similarly,TGF-b-mediated induction of Twist results in upregulation of miR-10b, which suppresses HOXD10 expression. The net-effect of these

    Fig. 1. Schematic of TGF-b and BMP signaling pathways coupled to EMT and MET progranormal epithelial cells. Shown are the receptors for TGF-b and BMP ligands, as well as the(BMP) reactions in normal and malignant cells. ALK-1 is primarily expressed on endothprecise signaling components required to mediate these physiological effects remain to

    Please cite this article in press as: C.D. Morrison et al., The relevance of the TG10.1016/j.canlet.2013.02.048tters xxx (2013) xxxxxx 3events results in the dramatic expression of the prometastaticgene, RhoC and elicits breast cancer invasion and metastasis [44].Somewhat surprisingly, miR-10b expression has also been associ-ated with diminished breast cancer invasion via its ability to targetthe Rac1 GEF, Tiam1 (T-cell lymphoma invasion and metastasis 1)[45]. Interestingly, Tiam1 expression is also governed by miRNAsmiR-21 and miR-31 to enhance colon cancer cell EMT, migration,and invasion [46]. miRNAs 21 and 31 are both strongly inducedby TGF-b [46], and a high expression of miR-21 in breast cancersassociates with a poor clinical outcomes and predicts for increasedTGF-b1 expression in cancers of the breast [47]. Lastly, Smad3 hasrecently been shown to associate with the p68/Drosha/DGCR8miRNA processing complex during the maturation of miR-21 tran-scripts, a reaction that transpires independent of Smad4 expressionand activity [48,49]. Collectively, these ndings highlight the coop-erative manner in which canonical and noncanonical TGF-b effec-tors regulate miRNA expression in developing and progressingcarcinomas, as well as demonstrate the importance of aberrantmiRNA expression in facilitating the TGF-b Paradox.

    3.2. TGF-b, alternative mRNA splicing, and EMT

    Alternative mRNA splicing is now gaining acceptance as animportant and integral mechanism operant in promoting EMTand MET programs [5052], which greatly increases the variability,

    ms, respectively. TGF-b and BMP signaling systems typically oppose one another inir downstream effectors operant in mediating the initiation of EMT (TGF-b) and METelial cells and functions to regulate proliferation, migration and angiogenesis. Thebe fully elucidated. See text for additional details.

    F-b Paradox to EMT-MET programs, Cancer Lett. (2013),

  • er Le4 C.D. Morrison et al. / Cancexibility, and complexity of the human genome [53]. TGF-b sig-naling plays a major role in regulating mRNA splicing, particularlyin developing carcinomas by suppressing their expression of epi-thelial splicing regulatory proteins (ESRPs) that bind RNA and gov-ern the alternative splicing of epithelial gene transcripts [54,55].The ability of TGF-b to downregulated ESRPs transpires via adEF1- and ZEB2/SIP1-dependent manner [54]. Importantly, down-regulated expression of ESRP results in the appearance of an EMTsplicing signature capable of differentiating luminal versus basalbreast cancer subtypes [50]. Moreover, the downregulation of ESRPby TGF-b also elicits differential isoform expression of CD44, p120catenin, and hMENA [52,54], events that further enhance thedevelopment of EMT programs induced by TGF-b. Accordingly, en-forced expression of ESRP attenuates the coupling of TGF-b to EMTprograms in epithelial cells [52]. Lastly, amongst the best studiedexamples of alternative splicing governed by TGF-b is that of thebroblast growth factor receptor (FGFR) gene located on humanchromosome 10q26, which is expressed as two alternative splicevariants, namely FGFR2-IIIb and FGFR2-IIIc [8]. FGFR2-IIIb is pri-marily expressed in epithelial cells, while their mesenchymalcounterparts primarily express FGFR2-IIIC, particularly in post-EMT cells stimulated by TGF-b [52,56]. Along these lines, stableexpression of FGFR2-IIIb variants in mesenchymal cells promotestheir acquisition of MET phenotypes, while stable expression ofFGFR1-IIIC variants in epithelial cells elicits EMT reactions reminis-cent of those stimulated by TGF-b [8,56]. In addition to continuingto identify novel splice variants operant in mediating EMT andmetastasis stimulated by TGF-b, future studies also need to

    Fig. 2. Cell and environmental factors coupled to the induction of EMT and MET progracomplex cascade of expression and repression that is regulated by a variety of factorsmicroenvironment. Targets of TGF-b and BMP signaling include microRNAs, inltratingtransitioning carcinoma cells further ne tune the balance between EMT:MET reactions tspecic expression of proteins in epithelial versus mesenchymal cell states. See text for

    Please cite this article in press as: C.D. Morrison et al., The relevance of the TG10.1016/j.canlet.2013.02.048tters xxx (2013) xxxxxxdevelop high-throughput methods to detect and assay the poten-tial of essential splice variants to serve as novel predictive bio-markers and chemotherapeutic targets in carcinoma cells.

    4. The inuence of the tumor microenvironment on EMTprograms induced by TGF-b

    Studies over the last two decades have continually afrmed theoverall importance of the tumor microenvironment (TME) in regu-lating carcinoma EMT and metastasis. Critical components of thetumor microenvironment include cancer-associated broblasts,endothelial cells, and a variety of inltrating immune cells suchas T and B cells, dendritic cells (DCs), and tumor-associated macro-phages [57]. In addition, hypoxia and alterations in the ECM com-position and its biomechanical properties also dictate the responseof carcinoma cells to TGF-b and its ability to drive EMT reactionsand metastatic progression [58,59]. In the succeeding sections,we highlight the central role of the TME in facilitating EMT andoncogenic signaling by TGF-b.

    4.1. TGF-b, cancer-associated broblasts, and EMT

    Recent ndings have established that normal and malignantcells respond differently to TGF-b in a manner that reects altera-tions in the composition of the surrounding ECM [3]. Indeed, TGF-bpotently regulates the activation status of adjacent cancer-associ-ated broblasts (CAFs) and myobroblasts, which are the predom-inant cell types housed within the TME. When activated by TGF-b,

    ms in carcinoma cells. Activation of TGF-b and BMP signaling systems couples to aboth within the carcinoma cells themselves, and within their accompanying hostimmune cells, and numerous microenvironmental factors and cytokines. Finally,hrough employment of alternative gene splicing events, which gives rise to isoformadditional details.

    F-b Paradox to EMT-MET programs, Cancer Lett. (2013),

  • er Lethese cells produce and secrete a variety of growth factors, cyto-kines, chemokines, and ECM proteins capable of promoting tumorinitiation and progression in adjacent epithelial cells [2,58]. Indeed,conditional deletion of TbR-II expression to inactivate TGF-b sig-naling in stromal broblasts elicited carcinoma formation in theprostate and stomach, as well as generated signicant expansionof the stromal compartment [60]. Similar initiation of carcinomaformation was observed following conditional deletion of TbR-IIin mammary broblasts, which enhanced the growth and invasionof breast cancer cells via elevated broblast secretion of cytokinesand growth factors, including TGF-a, MSP (macrophage stimulat-ing protein), and HGF [61]. Likewise, genetic inactivation of TbR-II in as little as 50% of human prostate broblasts was sufcientto transform normal human prostate epithelial cells in mouse tis-sue recombination models of prostate cancer development andprogression [62]. Importantly, these mouse-based ndings havealso been observed in the stroma of human colon cancers, whosetumor-associated broblasts exhibit downregulated expression ofTbR-II that correlated with increased lymph node metastasis anddecreased overall patient survival [63,64]. Collectively, these nd-ings highlight the tumor suppressing activities of TGF-b that tran-spire through its inhibition of broblast activities, such that loss ofTGF-b signaling in tumor-associated broblasts upregulates theirproduction of environmental factors capable of driving tumor for-mation and metastatic progression.

    4.2. TGF-b and tissue compliance during EMT programs

    TGF-b also regulates the differentiation status of broblasts,particularly their acquisition of myobroblast phenotypes and al-tered expression patterns of ECM proteins and growth factors ascompared to normal broblasts [65]. Along these lines, alterationsin the biomechanics of the ECM within developing tumors hasbeen linked to carcinoma development and metastasis. Indeed, tu-mor-associated myobroblasts are responsible for manifestingdesmoplastic reactions through their secretion of large amountsof collagen into the tumor microenvironment. Collagen becomescrosslinked to elastin through the activities of lysyl oxidases (LOXs)and tumor-associated myobroblasts further contribute toenhancing tumor rigidity by their production of a-smooth muscleactin [6568]. Recently, increased microenvironmental rigidity hasbeen shown to promote EMT programs stimulated by TGF-b. Forinstance, matrix rigidity converts TGF-b from a proapoptotic mol-ecule to an inducer of EMT reactions in NMuMG and MDCK cells,doing so via enhanced coupling of TGF-b to the PI3K/Akt pathway[69]. We too observed matrix rigidity to not only enhance the abil-ity of TGF-b to induce EMT programs in normal and malignantmammary epithelial cells, but also to underlie the manifestationsof the TGF-b Paradox, such that the tumor suppressing functionsof TGF-b are favored in compliant microenvironments, while its tu-mor promoting functions are favored in rigid microenvironments[68]. This dichotomous change in TGF-b behavior reects theupregulated expression of LOX and its ability to enhance the cou-pling of TGF-b to p38 MAPK [68]. Along these lines, elevated LOXexpression also elicits the activation of PI3K and increases the stiff-ness and tumorigenicity of mammary tumors that arose in MMVT-Neu transgenic mice [70]. It is interesting to note that biomechan-ically rigid tumors are typically more hypoxic than their compliantcounterparts, and as such, rigid and hypoxic tumor microenviron-ments drive LOX expression in a hypoxia-inducible factor-1a (HIF-1a)-dependent manner [71]. In addition, hypoxia-dependent LOXexpression is also essential in promoting breast cancer invasionand metastasis [7173], and in engendering the formation of the

    C.D. Morrison et al. / Cancpremetastatic niche operant in receiving disseminated cells[72,73]. Finally, LOX-like2 (LOXL2) was observed to interact phys-ically with the EMT transcription factor Snail, leading to the

    Please cite this article in press as: C.D. Morrison et al., The relevance of the TG10.1016/j.canlet.2013.02.048activation of EMT programs and acquisition of aggressive pheno-types by human squamous cell carcinomas [74,75]. Moreover, ren-dering basal-like human breast cancer cells decient in LOXL2expression not only prevented their pulmonary metastasis, butalso induced these cells to undergo MET programs [76]. Collec-tively, these ndings highlight the emerging importance of howbiochemical changes within tumor microenvironments functionin driving oncogenic TGF-b and its stimulation of EMT and metas-tasis in developing carcinomas.

    4.3. TGF-b and hypoxia during EMT programs

    Hypoxia is an integral part of tumorigenesis, which culminatesin the expression of HIFs and the urokinase receptor (uPAR) thatpromote EMT programs [5,77,78]. As alluded to above, hypoxia in-duces breast cancer cells to exhibit the classical hallmarks of EMT,including E-cadherin downregulation and/or delocalization fromthe plasma membrane, as well as the nuclear accumulation of Snailand b-catenin [79]. In fact, reoxygenation enables post-EMT cells toundergo MET programs and reestablish junctional protein com-plexes [77]. Moreover, hypoxia-mediated EMT reactions are alsoassociated with a gain in stem-like characteristics [80], whichmay in part be driven by the Notch signaling system [81]. Likewise,EMT induced by hypoxia is further characterized by the formationof NF-jB-HIF-1a complexes, which enhance the expression of c-Jun [82]. Collectively, these studies demonstrate a critical role forhypoxia in regulating the balance between EMT and MET states,particularly during the evolution of metastatic phenotypes in car-cinomas. Given the striking parallels between hypoxia and onco-genic TGF-b signaling, future studies need to address the specicroles played by TGF-b and BMP during hypoxia-driven EMT pro-grams and vice versa.

    4.4. TGF-b and integrins during EMT programs

    The differential expression and activity of integrins clearlymodulates how normal and malignant cells sense and respond toTGF-b in compliant and rigid microenvironments. Integrins act astransmembrane scaffolds that link the ECM to the actin cytoskele-tal system; they also function as mediators of mechanotransduc-tion and regulate cell proliferation, migration, and invasion, aswell as cell survival [67,83]. For instance, expression of a b1 inte-grin mutant that mimics rigidity-induced integrin clustering (i.e.,V737N-b1 integrin) readily enhanced the invasion of premalignantmammary epithelial cells by promoting the formation of focaladhesion complexes [70]. Even more remarkably, culturingV737N-b1 integrin-expressing mammary epithelial cells in compli-ant 3D-organotypic cultures elevated their activation of the PI3Kpathway to levels typically observed in their rigid 3D-organotypiccounterparts [70]. Thus, b1 integrin clustering plays a prominentrole in eliciting oncogenic signaling events in rigid carcinomamicroenvironments. Along these lines, we identied integrins asprominent mediators of oncogenic TGF-b signaling and its couplingto EMT programs. Indeed, upregulated expression of b3 integrin inresponse to TGF-b is essential for its initiation of EMT reactions[8486]. Mechanistically, b3 integrin forms a complex with TbR-II that is bridged by FAK, leading to Src-mediated phosphorylationof TbR-II at Tyr284 that recruits ShcA and Grb2 necessary in cou-pling TGF-b to the amplied activation of p38 MAPK [8486].Inhibiting these events alleviates oncogenic TGF-b signaling,including its coupling to NF-jB, Cox-2, and PGE2 signaling [1921,87], as well as prevents the acquisition of EMT and metastaticphenotypes driven by TGF-b [5]. The formation of TbR-II:b3 inte-

    tters xxx (2013) xxxxxx 5grin complexes also results in the activation of FAK, which medi-ates TGF-b stimulation of breast cancer cell invasion both in vitroand in vivo [88], and in the upregulated expression of p130Cas,

    F-b Paradox to EMT-MET programs, Cancer Lett. (2013),

  • mediating metastatic outgrowth [34,91]. Finally, we recently iden-tied integrin switching as a novel modulator of the oncogenic

    ple-negative breast cancer subtypes. However, a newly described

    N-cadherin and vimentin, and accompanying increases in epithe-

    er Leactivities of TGF-b. Indeed, pharmacologic or genetic inactivationof b1 integrin elicits a dramatic compensatory upregulation of b3integrin expression that not only restores the oncogenic functionsof TGF-b, but also enhances its ability to drive metastasis at earlierstages of mammary tumor development (J.G. Parvani and W.P.Schiemann, unpublished observation). Thus, metastatic carcinomacells and their acquisition of EMT programs provides the meansnecessary to evade single agent integrin-based therapies throughintegrin-switching mechanisms [96,97]. As such, future studiesneed to identify the collection of integrins operant in mediatingthe oncogenic activities of TGF-b, as well as determine how EMTprograms and chemotherapies alter the composition of integrinsin TGF-b-responsive carcinoma cells.

    5. TGF-b and cancer stem cells (CSCs) during EMT programs

    Developing carcinomas are highly heterogeneous and containnumerous distinct cell populations, some of which can reinitiatetumor growth and development with varying degrees of efciency.Amongst the most efcient tumor-initiating cells are cancer stemcells (CSCs) that comprise a small fraction of the primary tumorand possess self-renewing capabilities [98,99]. Initial evidencelinking EMT programs to the generation of CSCs was provided byWeinberg and colleagues [100] who demonstrated that engineer-ing immortalized mammary epithelial cells to stably express Snailor Twist, or stimulating them with TGF-b readily produced a post-EMT population of cells that displayed the markers (e.g., CD44high/CD24low) and features (e.g., mammosphere and tumor-initiatingbehaviors) of stem-like cells. Subsequent studies showed that thissame post-EMT population of cells can also assume the character-istics of mesenchymal stem cells (MSCs), including their ability to(i) differentiate into bone, adipocytes, or chrondrocytes; (ii) mi-grate to breast cancer cells; and (iii) home to wounded tissues[101]. Similar ndings by others have clearly established EMT pro-grams as major drivers of the selection and expansion of CSCs[100,102104], resulting in the appearance of chemoresistant phe-notypes and disease recurrence [99,105108]. The importance ofwhich elicits diminished expression of Smad3 and enhances thepulmonary metastasis of breast cancers [89]. Thus, aberrant cou-pling of TGF-b to avb3 integrin expression is sufcient to manifestthe TGF-b Paradox and elicit the oncogenic activities of TGF-b inresponsive cells.

    Besides b3 integrin, TGF-b signaling is also impacted by itsphysical interaction with b1 integrin [84], which also regulatesTGF-b stimulation of EMT programs and p38 MAPK [90]. Recently,we linked TGF-b stimulation of EMT to the initiation of prolifera-tive programs and metastatic outgrowth of disseminated breastcancer cells. In doing so, we observed compliant microenviron-ments to prevent TGF-b from activating Smad3/4 [34,91,92], whichblocks the expression of Pyk2 necessary to mediate escape frommetastatic dormancy [91]. Interestingly, these events are also gov-erned by the reciprocal expression patterns of E-cadherin and b1integrin, such that dormant cells are typically E-cadherin high, b1integrin low, while outgrowth procient cells are typically b1 inte-grin high, E-cadherin low [34,64,91,93,94]. Importantly, the acqui-sition of EMT phenotypes represents the molecular switch thatallows cells to toggle betweenmetastatic dormancy and outgrowthdue to diminished E-cadherin expression, which interacts in a het-erotypic manner with the extracellular domain of b1 integrin [95],leading to a loss of b1 integrin and Pyk2 expression operant in

    6 C.D. Morrison et al. / CancTGF-b in mediating these events is highlighted by the nding that(i) TGF-b gene signatures associate with metastatic progressionand poor clinical outcomes, and (ii) pharmacological inactivation

    Please cite this article in press as: C.D. Morrison et al., The relevance of the TG10.1016/j.canlet.2013.02.048lial markers, such as E-cadherin and CK-19 [8,119]. The inductionof MET programs has been postulated as playing an integral partof the metastatic cascade because metastatic lesions typically ap-pear more differentiated as compared to their primary tumor oforigin [120]. Thus, the current paradigm states that EMT drives car-cinoma cells to disseminate from the primary tumor, while METdrives disseminated carcinoma cells to reinitiate proliferative pro-grams necessary for their formation of secondary tumors [120].The latter response is important because TGF-b stimulation ofEMT typically inhibits cell cycle progression, and as such, it appearsthat the ability of carcinoma cells to disseminate and ultimatelydevelop into macroscopic lesions may in fact represent two mutu-ally exclusive processes. Indeed, the ability of disseminated tumorcells to reestablish epithelial phenotypes viaMET programs may infact represent the rate-limiting step of metastasis, whose failure totranspire could potentially lock metastatic loci into a state of per-petual dormancy [120]. Here we highlight recent ndings of themolecular mechanisms that underlie MET reactions and their rele-vance to TGF-b and metastasis.

    6.1. Bone morphogenetic proteins (BMPs) and MET programs

    BMPs are multifunctional cytokines of the TGF-b superfamilythat were originally identied based on their ability to induce bonerare breast cancer subtype termed claudin-low was observedto be enriched in post-EMT and stem-like properties [99,109],and to exhibit signicantly worse prognoses as compared to otherbreast cancer subtypes [110]. Consistent with its designation as amaster regulator of EMT programs, TGF-b signaling drives theappearance of claudin-low post-EMT cell populations [111].

    Conversely, recent evidence also suggests that EMT suppressesCSC formation, such that CSCs may reside in the epithelial-like ba-sal cells in prostate cancer [112]. Along these lines, other studieshave also shown that epithelial-like tumor cells are in fact capableof forming distant metastases [113116]. Accordingly, epithelial-like prostate cancer cells were enriched in subpopulations of met-astatic tumor initiating cells, whereas their mesenchymal-likecounterparts were observed to display reduced tumor initiatingcapabilities [115,116]. Clearly, future studies are warranted to fullydene the relationship between epithelial-like and mesenchymal-like cell types in promoting the expansion of CSCs and tumor-ini-tiating cell populations during carcinoma development, metastasis,and recurrence.

    6. Mesenchymal-epithelial transition (MET) programs

    As alluded to above, MET is the process whereby mesenchymal-like cells transdifferentiate to establish polarized epithelial-likecells, complete with the formation of strong cellcell junctionsand adherens complexes. These junctional complexes are com-prised of various transmembrane molecules and accessory pro-teins, which form intricate networks with the actin cytoskeleton.Additional descriptions related to the regulation of these junctionalcomplexes and their role in EMT programs are directed to severalrecent reviews [11,117,118]. MET programs are also characterizedby decreases in the expression of mesenchymal markers, such asof TGF-b signaling in metastatic breast cancer cells induced aMET program indicative of diminished tumorigenicity [102]. Withrespect to breast cancers, mesenchymal phenotypes derived fromEMT programs are predominantly associated with basal-like or tri-

    tters xxx (2013) xxxxxxdifferentiation and formation [121]. Similar to TGF-b, BMP signal-ing transpires through its binding to specic type I and type IIreceptors [121]. However, while TGF-b binds exclusively to

    F-b Paradox to EMT-MET programs, Cancer Lett. (2013),

  • er LeTbR-II, BMP ligands are more promiscuous and bind not only to theBMP type II receptor (BMPR2), but also to the activin type II recep-tors, ActRII and ActRIIB [121]. Once activated, BMP receptors acti-vate Smads 1, 5, or 8, which translocate into the nucleus bound toSmad4 to bring about changes in gene expression [121]. In additionto regulating bone formation, BMPs have also been implicated inmediating MET programs. For instance, administration of BMP7to human breast cancer cells results in their upregulated expres-sion of E-cadherin, as well as inhibits TGF-b signaling, an event thatculminates in decreased vimentin expression [122]. Along theselines, therapeutic administration of BMP7 diminished breast can-cer metastasis to bone [122] and reduced the pool of CSCs associ-ated with EMT reactions [123]. Furthermore, functional disruptionof the BMPR2 dramatically enhanced pulmonary metastases in aMMTV-PyMT mouse model [124]. Interestingly, intravital imaginganalyses have identied a reduction in TGF-b signaling in pulmon-ary metastases [92,125], suggesting that BMPs may antagonize theactivities of TGF-b at micrometastatic sites to facilitate metastaticoutgrowth of secondary lesions [120]. Future studies need to inves-tigate the spatiotemporal contexts that govern BMP signaling atdistinct stages of metastasis, as well as how these events differ be-tween primary tumors and their metastatic lesions.

    6.2. Transcriptional regulators of MET

    While EMT programs are clearly required for the generation ofstem-like CSCs and MSCs, the production of inducible pluripotentstem cells (iPSCs) requires the initiation of MET programs [126]. In-deed, the reprogramming of broblasts requires the simultaneousexpression of four embryonic stem cell transcription factors,namely Oct4, Sox2, Klf4, and c-Myc. These transcription factorscollectively induce a MET reaction that inhibits TGF-b by downreg-ulating TbR-II expression, as well as that of Snail [126]. Conversely,augmenting TGF-b signaling signicantly impeded the generationof iPSCs [126], indicating that EMT and MET programs opposeone another during the generation of iPSCs, and consequently, thatEMT programs driven by TGF-b limit cellular reprogramming andthe pluripotency of fully transitioned post-EMT cells. Future stud-ies need to further our understanding of the dynamics and inter-play that exist between EMT and MET programs in regulatingcell reprogramming, and to determine the therapeutic potentialof targeting these differences as a means to alleviate CSCs and dis-ease recurrence.

    6.3. The miR-200 family and MET programs

    The miR-200 family of miRNAs (i.e., miR-200a, miR-200b, miR-200c, miR-141 and miR-429) has been implicated in promotingMET via their ability to repress the expression of ZEB1 and ZEB2,leading to upregulated E-cadherin expression [32]. In fact, enforcedexpression of the miR-200 family enhances the epithelial charac-teristics of breast cancer cells [127]. Besides their ability to allevi-ate the repression of essential epithelial gatekeepers, miR-200family members also target cytoskeletal components, such asWAVE3 and ECM proteins, such as bronectin, both of which areprominently associated with mesenchymal phenotypes [127]. Re-cently, the tumor suppressor p53 was shown to mediate theexpression of miR-200c, as the loss of p53 expression in mammaryepithelial cells resulted in the appearance of mesenchymal cellpopulations that possessed stem-like properties [128]. As men-tioned previously, TGF-b is a potent suppressor of the miR-200family, doing so by promoting the hypermethylation of the pro-moters for these miRNAs [127]. It should be noted that while

    C.D. Morrison et al. / CancmiR-200 family members are clearly linked to maintaining epithe-lial cell integrity, aberrantly upregulated or downregulated expres-sion of these miRNAs has also been associated with the acquisition

    Please cite this article in press as: C.D. Morrison et al., The relevance of the TG10.1016/j.canlet.2013.02.048of metastatic phenotypes [35,127]. For instance, MET programs canbe initiated by expression of miR-200 family members, or by inac-tivation of EMT-inducing miRNAs [27]. Importantly, the nuances ofexpressing MET-inducing miRNAs versus inhibiting EMT-inducingmiRNAs reects the summation of global changes in gene expres-sion that transpire in response to these miRNAs. Along these lines,the interplay between miR-200s and other plasticity-associatedmiRNAs remain incompletely understood [27]. Future studies needto address these relationships, as well as to better dene their rolein driving metastatic progression.

    6.4. Grainyhead-like-2 (GRHL2) and MET programs

    GRHL2 is a developmentally regulated gene that is associatedwith neural tube closure [129]. Recently, decreased expression ofGRHL2 has been observed in basal-like and claudin-low breast can-cer subtypes, whose re-expression of GRHL2 induced cadherinswitching consistent with the induction of MET programs (i.e., N-cadherin in GRHL2low to E-cadherin in GRHL2high) [129]. Indeed,GRHL2 promotes MET programs through its ability to inhibitTGF-b-mediated activation of Smad2/3, which leads to the loss ofZEB1 expression and accompanying upregulation of miR-200expression [129]. Finally, GRHL2 reduces the expression of CD44,suggesting GRHL2 not only alleviates EMT phenotypes, but dimin-ishes the proportion of carcinoma cells that possess CSC and stem-like characteristics [129]. Given the parallels between MET pro-grams induced by BMPs and GRHL2, future studies need to deter-mine the extent to which GRHL2 mediates MET reactionsstimulated by BMPs, and conversely, how these events are inacti-vated by TGF-b in developing and progressing carcinoma cells.

    6.5. Endoplasmic reticulum protein 29 (ERp29) and MET programs

    ERp29 is a resident endoplasmic reticulum protein that func-tions as a chaperone during the unfolding and secretion of proteinsthrough the vesicular transport system [130,131]. With respect todeveloping carcinomas, endoplasmic reticulum homeostasis is of-ten disrupted as a result of DNA damage or oxidative stress; how-ever, the mechanistic role of ERp29 during carcinogenesis remainsincompletely understood [130]. Interestingly, overexpression ofERp29 in mesenchymal-like and metastatic MDA-MB-231 cells in-duced a MET program that manifested in the appearance of corticalactin staining, elevated expression of epithelial markers (e.g., E-cadherin and CK-19), and decreased expression of mesenchymalmarkers (e.g., vimentin and bronectin), events associated withdiminished expression of Slug, Snail, ZEB2, and Twist [131]. Con-versely, disruption of ERp29 expression in epithelial-like and non-metastatic MCF-7 cells induced a loss of junctional integrity thatwas reminiscent of those observed in EMT programs [131]. Collec-tively, these ndings demonstrate a novel role for ERp29 and endo-plasmic reticulum homeostasis in driving MET programs, as well asin inhibiting TGF-b signaling in mammary carcinomas.

    6.6. c-Abl and MET programs

    c-Abl is a nonreceptor protein tyrosine kinase (PTK) that gov-erns a variety of physiological process, including cell proliferation,motility, and survival, as well as cell and tissue homeostasis [132134]. c-Abl is perhaps best known for its causative role in promot-ing hematologic cancer development, typically arising in chronicmyelogenous leukemia (CML) cells that harbor the Philadelphiatranslocation on chromosome 22 wherein c-Abl is translocatedand fused to the break-point cluster region, resulting in the pro-

    tters xxx (2013) xxxxxx 7duction of a constitutively-active c-Abl kinase fusion protein[132134]. Importantly, rational drug design led to the generationof the small molecule c-Abl antagonist, Imatinib, which ushered in

    F-b Paradox to EMT-MET programs, Cancer Lett. (2013),

  • trials designed to assess the therapeutic effects of Imatinib on ad-

    we recently established c-Abl as an essential gatekeeper charged

    The TGF-b superfamily regulates various aspects of EMT and

    [26] K. Ruan, X. Fang, G. Ouyang, MicroRNAs: novel regulators in the hallmarks ofhuman cancer, Cancer Lett. 285 (2009) 116126.

    er LeMET programs in normal and malignant epithelial cells. With re-spect to the ability of TGF-b to induce EMT reactions, numerousstudies have documented the integral role played by both canoni-cal and noncanonical TGF-b effectors to elicit EMT and metastaticprogression of developing carcinomas [5,11]. Collectively, thesesignaling events coalesce to induce EMT programs by modulatingthe expression of miRNAs, by creating a permissive tumor micro-environment, and by inducing the acquisition of stem-like andchemoresistant phenotypes. Although our understanding of EMTprograms seems to grow on a daily basis, our knowledge relatedto MET programs and their regulation by BMP continues to lagand remain highly elusive. Given the proposed role of MET pro-grams in functioning as the rate-limiting step of metastasis[120], future studies need to elucidate the molecular mechanismsthat enable TGF-b and BMP signals to diverge and oppose one an-other in normal epithelial cells, as well as identify the signaling de-fects that enable these pathways to converge in metastaticcarcinoma cells. In doing so, science and medicine will undoubt-edly glean valuable therapeutic insights capable of targeting andalleviating metastatic disease in carcinoma patients.

    Conict of interest

    None declared.


    Members of the Schiemann Laboratory are thanked for criticalreading of the manuscript. W.P.S. was supported in part by grantswith maintaining epithelial phenotypes in mammary epithelialcells. Indeed, heterologous expression of a constitutively-active c-Abl mutant (CST-Abl) in late-stage breast cancer cells preventedtheir acquisition of EMT and metastatic phenotypes in responseto TGF-b [132]. In doing so, we observed CST-Abl expression inlate-stage breast cancer cells to promote the formation of strongcellcell junctions and reduce cell spreading in 2D cultures, as wellas to elicit the normalization and hollowing of acinar structures in3D cultures [132]. Conversely, pharmacologic or genetic inactiva-tion of c-Abl in normal mammary epithelial cells initiated EMTprograms, complete with the downregulated expression of epithe-lial markers (e.g., E-cadherin and MMPs) and upregulated expres-sion of their mesenchymal counterparts (e.g., vimentin, N-cadherin, and Twist1) [132]. Thus, these morphological and pheno-typic changes brought about by c-Abl activation clearly reect theinitiation of MET programs and their ability to render late-stagebreast cancers benign when engrafted into mice [132]. Futurestudies need to develop the means to harness and exploit the tu-mor suppressing and MET-inducing functions of c-Abl as a novelmeans to alleviate carcinoma metastasis and disease recurrence.

    7. Conclusion and future directionsvanced breast cancers were halted due to a lack of efcacy [135137], ndings we recapitulated in preclinical mouse models ofmammary tumor development and metastasis [132]. Importantly,the era of targeted drug therapies and signicantly improved theclinical course of CML patients [132134]. Interestingly, the suc-cess of Imatinib in treating liquid tumors has not been replicatedin their solid tumor counterparts. For instance, three recent clinical

    8 C.D. Morrison et al. / Cancfrom the National Institutes of Health (CA129359), the Department

    Please cite this article in press as: C.D. Morrison et al., The relevance of the TG10.1016/j.canlet.2013.02.048[27] P.A. Gregory, C.P. Bracken, A.G. Bert, G.J. Goodall, MicroRNAs as regulators ofepithelial-mesenchymal transition, Cell Cycle. 7 (2008) 31123118.

    [28] M.D. Bullock, A.E. Sayan, G.K. Packham, A.H. Mirnezami, MicroRNAs: criticalregulators of epithelial to mesenchymal (EMT) and mesenchymal toof Defense (BC084561), and pilot funding from the Case Compre-hensive Cancer Center (P30 CA043703).


    [1] J. Massague, TGFb in cancer, Cell 134 (2008) 215230.[2] M.A. Taylor, Y.H. Lee, W.P. Schiemann, Role of TGF-b and the tumor

    microenvironment during mammary tumorigenesis, Gene Expr. 15 (2011)117132.

    [3] M.K. Wendt, M. Tian, W.P. Schiemann, Deconstructing the mechanisms andconsequences of TGF-b-induced EMT during cancer progression, Cell TissueRes. 347 (2012) 85101.

    [4] M. Tian, J.R. Neil, W.P. Schiemann, Transforming growth factor-b and thehallmarks of cancer, Cell Signal. 23 (2011) 951962.

    [5] M.A. Taylor, J.G. Parvani, W.P. Schiemann, The pathophysiology of epithelial-mesenchymal transition induced by transforming growth factor-b in normaland malignant mammary epithelial cells, J. Mammary Gland Biol. Neoplasia15 (2010) 169190.

    [6] J.P. Thiery, H. Acloque, R.Y. Huang, M.A. Nieto, Epithelial-mesenchymaltransitions in development and disease, Cell 139 (2009) 871890.

    [7] R. Kalluri, R.A. Weinberg, The basics of epithelial-mesenchymal transition, J.Clin. Invest. 119 (2009) 14201428.

    [8] D. Yao, C. Dai, S. Peng, Mechanism of the mesenchymal-epithelial transitionand its relationship with metastatic tumor formation, Mol. Cancer Res. 9(2011) 16081620.

    [9] M.K. Wendt, T.M. Allington, W.P. Schiemann, Mechanisms of the epithelial-mesenchymal transition by TGF-b, Future Oncol. 5 (2009) 11451168.

    [10] C.H. Heldin, M. Vanlandewijck, A. Moustakas, Regulation of EMT by TGFb incancer, FEBS Lett. 586 (2012) 19591970.

    [11] J.G. Parvani, M.A. Taylor, W.P. Schiemann, Noncanonical TGF-b signalingduring mammary tumorigenesis, J. Mammary Gland Biol. Neoplasia 16 (2011)127146.

    [12] C.E. Gatza, S.Y. Oh, G.C. Blobe, Roles for the type III TGF-b receptor in humancancer, Cell Signal. 22 (2010) 11631174.

    [13] S. Bharathy, W. Xie, J.M. Yingling, M. Reiss, Cancer-associated transforminggrowth factor b type II receptor gene mutant causes activation of bonemorphogenic protein-Smads and invasive phenotype, Cancer Res. 68 (2008)16561666.

    [14] I.M. Liu, S.H. Schilling, K.A. Knouse, L. Choy, R. Derynck, X.F. Wang, TGFb-stimulated Smad1/5 phosphorylation requires the ALK5 L45 loop andmediates the pro-migratory TGFb switch, EMBO J. 28 (2009) 8898.

    [15] M.J. Goumans, G. Valdimarsdottir, S. Itoh, F. Lebrin, J. Larsson, C. Mummery, S.Karlsson, P. ten Dijke, Activin receptor-like kinase (ALK)1 is an antagonisticmediator of lateral TGF-b/ALK5 signaling, Mol. Cell. 12 (2003) 817828.

    [16] M.J. Goumans, G. Valdimarsdottir, S. Itoh, A. Rosendahl, P. Sideras, P. tenDijke, Balancing the activation state of the endothelium via two distinct TGF-b type I receptors, EMBO J. 21 (2002) 17431753.

    [17] B.N. Ray, N.Y. Lee, T. How, G.C. Blobe, ALK5 phosphorylation of the endoglincytoplasmic domain regulates Smad1/5/8 signaling and endothelial cellmigration, Carcinogenesis 31 (2010) 435441.

    [18] A. Moustakas, C.H. Heldin, Non-Smad TGF-b signals, J. Cell Sci. 118 (2005)35733584.

    [19] J.R. Neil, K.M. Johnson, R.A. Nemenoff, W.P. Schiemann, Cox-2 inactivatesSmad signaling and enhances EMT stimulated by TGF-b through a PGE2-dependent mechanisms, Carcinogenesis 29 (2008) 22272235.

    [20] J.R. Neil, W.P. Schiemann, Altered TAB1:IjB kinase interaction promotestransforming growth factor b-mediated nuclear factor-jB activation duringbreast cancer progression, Cancer Res. 68 (2008) 14621470.

    [21] J.R. Neil, M. Tian, W.P. Schiemann, X-linked inhibitor of apoptosis protein andits E3 ligase activity promote transforming growth factor-b-mediated nuclearfactor-jB activation during breast cancer progression, J. Biol. Chem. 284(2009) 2120921217.

    [22] A. Sorrentino, N. Thakur, S. Grimsby, A. Marcusson, V. von Bulow, N. Schuster,S. Zhang, C.H. Heldin, M. Landstrom, The type I TGF-b receptor engages TRAF6to activate TAK1 in a receptor kinase-independent manner, Nat. Cell Biol. 10(2008) 11991207.

    [23] M. Yamashita, K. Fatyol, C. Jin, X. Wang, Z. Liu, Y.E. Zhang, TRAF6 mediatesSmad-independent activation of JNK and p38 by TGF-b, Mol. Cell. 31 (2008)918924.

    [24] Y. Mu, R. Sundar, N. Thakur, M. Ekman, S.K. Gudey, M. Yakymovych, A.Hermansson, H. Dimitriou, M.T. Bengoechea-Alonso, J. Ericsson, C.H. Heldin,M. Landstrom, TRAF6 ubiquitinates TGF-b type I receptor to promote itscleavage and nuclear translocation in cancer, Nat. Commun. 2 (2011) 330.

    [25] A. Moustakas, C.H. Heldin, Induction of epithelial-mesenchymal transition bytransforming growth factor b, Semin. Cancer Biol. 22 (2012) 446454.

    tters xxx (2013) xxxxxxepithelial transition (MET) in cancer progression, Biol. Cell. 104 (2012) 312.

    F-b Paradox to EMT-MET programs, Cancer Lett. (2013),

  • er Le[29] J. Zavadil, M. Narasimhan, M. Blumenberg, R.J. Schneider, Transforminggrowth factor-b and microRNA:mRNA regulatory networks in epithelialplasticity, Cells Tissues Organs. 185 (2007) 157161.

    [30] M. Tian, W.P. Schiemann, The TGF-b paradox in human cancer: an update,Future Oncol. 5 (2009) 259271.

    [31] M. Korpal, Y. Kang, The emerging role of miR-200 family of microRNAs inepithelial-mesenchymal transition and cancer metastasis, RNA Biol. 5 (2008)115119.

    [32] P.A. Gregory, A.G. Bert, E.L. Paterson, S.C. Barry, A. Tsykin, G. Farshid, M.A.Vadas, Y. Khew-Goodall, G.J. Goodall, The miR-200 family and miR-205regulate epithelial to mesenchymal transition by targeting ZEB1 and SIP1,Nat. Cell Biol. 10 (2008) 593601.

    [33] P.A. Gregory, C.P. Bracken, E. Smith, A.G. Bert, J.A. Wright, S. Roslan, M. Morris,L. Wyatt, G. Farshid, Y.Y. Lim, G.J. Lindeman, M.F. Shannon, P.A. Drew, Y.Khew-Goodall, G.J. Goodall, An autocrine TGF-b/ZEB/miR-200 signalingnetwork regulates establishment and maintenance of epithelial-mesenchymal transition, Mol. Biol. Cell 22 (2011) 16861698.

    [34] M.K. Wendt, M.A. Taylor, B.J. Schiemann, W.P. Schiemann, Downregulation ofepithelial cadherin is required to initiate metastatic outgrowth of breastcancer, Mol. Biol. Cell 22 (2011) 24232435.

    [35] M. Korpal, B.J. Ell, F.M. Buffa, T. Ibrahim, M.A. Blanco, T. Celia-Terrassa, L.Mercatali, Z. Khan, H. Goodarzi, Y. Hua, Y. Wei, G. Hu, B.A. Garcia, J. Ragoussis,D. Amadori, A.L. Harris, Y. Kang, Direct targeting of Sec23a by miR-200sinuences cancer cell secretome and promotes metastatic colonization, Nat.Med. 17 (2011) 11011108.

    [36] M.J. Schliekelman, D.L. Gibbons, V.M. Faca, C.J. Creighton, Z.H. Rizvi, Q. Zhang,C.H. Wong, H. Wang, C. Ungewiss, Y.H. Ahn, D.H. Shin, J.M. Kurie, S.M. Hanash,Targets of the tumor suppressor miR-200 in regulation of the epithelial-mesenchymal transition in cancer, Cancer Res. 71 (2011) 76707682.

    [37] S.M. Farabaugh, D.S. Micalizzi, P. Jedlicka, R. Zhao, H.L. Ford, Eya2 is requiredto mediate the pro-metastatic functions of Six1 via the induction of TGF-bsignaling, epithelial-mesenchymal transition, and cancer stem cell properties,Oncogene 31 (2012) 552562.

    [38] D.S. Micalizzi, K.L. Christensen, P. Jedlicka, R.D. Coletta, A.E. Baron, J.C. Harrell,K.B. Horwitz, D. Billheimer, K.A. Heichman, A.L. Welm, W.P. Schiemann, H.L.Ford, The Six1 homeoprotein induces human mammary carcinoma cells toundergo epithelial-mesenchymal transition and metastasis in mice throughincreasing TGF-b signaling, J. Clin. Invest. 119 (2009) 26782690.

    [39] D.S. Micalizzi, C.A. Wang, S.M. Farabaugh, W.P. Schiemann, H. Ford,Homeoprotein Six1 increases TGF-b type I receptor and converts TGF-bsignaling from suppressive to supportive for tumor growth, Cancer Res. 70(2010) 1037110380.

    [40] A.L. Smith, R. Iwanaga, D.J. Drasin, D.S. Micalizzi, R.L. Vartuli, A.C. Tan, H.L.Ford, The miR-106b-25 cluster targets Smad7, activates TGF-b signaling, andinduces EMT and tumor initiating cell characteristics downstream of Six1 inhuman breast cancer, Oncogene (2012),

    [41] C.A. Gebeshuber, K. Zatloukal, J. Martinez, MiR-29a suppresses tristetraprolin,which is a regulator of epithelial polarity and metastasis, EMBO Rep. 10(2009) 400405.

    [42] W. Kong, H. Yang, L. He, J.J. Zhao, D. Coppola, W.S. Dalton, J.Q. Cheng,MicroRNA-155 is regulated by the transforming growth factor b/Smadpathway and contributes to epithelial cell plasticity by targeting RhoA, Mol.Cell Biol. 28 (2008) 67736784.

    [43] E. Papadimitriou, E. Vasilaki, C. Vorvis, D. Iliopoulos, A. Moustakas, D.Kardassis, C. Stournaras, Differential regulation of the two RhoA-specic GEFisoforms Net1/Net1A by TGF-b and miR-24: role in epithelial-to-mesenchymal transition, Oncogene 31 (2011) 28622875.

    [44] L. Ma, J. Teruya-Feldstein, R.A. Weinberg, Tumour invasion and metastasisinitiated by microRNA-10b in breast cancer, Nature 449 (2007) 682688.

    [45] C.H. Moriarty, B. Pursell, A.M. Mercurio, MiR-10b targets Tiam1: implicationsfor Rac activation and carcinoma migration, J. Biol. Chem. 285 (2010) 2054120546.

    [46] C.L. Cottonham, S. Kaneko, L. Xu, MiR-21 and miR-31 converge on TIAM1 toregulate migration and invasion of colon carcinoma cells, J. Biol. Chem. 285(2010) 3529335302.

    [47] B. Qian, D. Katsaros, L. Lu, M. Preti, A. Durando, R. Arisio, L. Mu, H. Yu, HighmiR-21 expression in breast cancer associated with poor disease-free survivalin early stage disease and high TGF-b1, Breast Cancer Res. Treat. 117 (2009)131140.

    [48] B.N. Davis, A.C. Hilyard, G. Lagna, A. Hata, SMAD proteins control DROSHA-mediated microRNA maturation, Nature 454 (2008) 5661.

    [49] B.N. Davis, A.C. Hilyard, P.H. Nguyen, G. Lagna, A. Hata, Smad proteins bind aconserved RNA sequence to promote microRNA maturation by Drosha, Mol.Cell 39 (2010) 373384.

    [50] I.M. Shapiro, A.W. Cheng, N.C. Flytzanis, M. Balsamo, J.S. Condeelis, M.H.Oktay, C.B. Burge, F.B. Gertler, An EMT-driven alternative splicing programoccurs in human breast cancer and modulates cellular phenotype, PLoSGenet. 7 (2011) e1002218.

    [51] M. Dutertre, M. Lacroix-Triki, K. Driouch, P. de la Grange, L. Gratadou, S. Beck,S. Millevoi, J. Tazi, R. Lidereau, S. Vagner, D. Auboeuf, Exon-based clustering ofmurine breast tumor transcriptomes reveals alternative exons whoseexpression is associated with metastasis, Cancer Res. 70 (2010) 896905.

    [52] C.C. Warzecha, T.K. Sato, B. Nabet, J.B. Hogenesch, R.P. Carstens, ESRP1 and

    C.D. Morrison et al. / CancESRP2 are epithelial cell-type-specic regulators of FGFR2 splicing, Mol. Cell33 (2009) 591601.

    Please cite this article in press as: C.D. Morrison et al., The relevance of the TG10.1016/j.canlet.2013.02.048[53] E.T. Wang, R. Sandberg, S. Luo, I. Khrebtukova, L. Zhang, C. Mayr, S.F.Kingsmore, G.P. Schroth, C.B. Burge, Alternative isoform regulation in humantissue transcriptomes, Nature 456 (2008) 470476.

    [54] K. Horiguchi, K. Sakamoto, D. Koinuma, K. Semba, A. Inoue, S. Inoue, H. Fujii,A. Yamaguchi, K. Miyazawa, K. Miyazono, M. Saitoh, TGF-b drives epithelial-mesenchymal transition through dEF1-mediated downregulation of ESRP,Oncogene 31 (2012) 31903201.

    [55] C.C. Warzecha, P. Jiang, K. Amirikian, K.A. Dittmar, H. Lu, S. Shen, W. Guo, Y.Xing, R.P. Carstens, An ESRP-regulated splicing programme is abrogatedduring the epithelial-mesenchymal transition, EMBO J. 29 (2010) 32863300.

    [56] T. Shirakihara, K. Horiguchi, K. Miyazawa, S. Ehata, T. Shibata, I. Morita, K.Miyazono, M. Saitoh, TGF-b regulates isoform switching of FGF receptors andepithelial-mesenchymal transition, EMBO J. 30 (2011) 783795.

    [57] J. Fuxe, M.C. Karlsson, TGF-b-induced epithelial-mesenchymal transition: alink between cancer and inammation, Semin. Cancer Biol. 22 (2012) 455461.

    [58] H.P. Naber, P. ten Dijke, E. Pardali, Role of TGF-b in the tumor stroma, Curr.Cancer Drug Targets 8 (2008) 466472.

    [59] Y. Jing, Z. Han, S. Zhang, Y. Liu, L. Wei, Epithelial-mesenchymal transition intumor microenvironment, Cell Biosci. 1 (2011) 29.

    [60] N.A. Bhowmick, A. Chytil, D. Plieth, A.E. Gorska, N. Dumont, S. Shappell, M.K.Washington, E.G. Neilson, H.L. Moses, TGF-b signaling in broblastsmodulates the oncogenic potential of adjacent epithelia, Science 303 (2004)848851.

    [61] N. Cheng, N.A. Bhowmick, A. Chytil, A.E. Gorksa, K.A. Brown, R. Muraoka, C.L.Arteaga, E.G. Neilson, S.W. Hayward, H.L. Moses, Loss of TGF-b type II receptorin broblasts promotes mammary carcinoma growth and invasion throughupregulation of TGF-a-, MSP- and HGF-mediated signaling networks,Oncogene 24 (2005) 50535068.

    [62] O.E. Franco, M. Jiang, D.W. Strand, J. Peacock, S. Fernandez, R.S. Jackson 2nd,M.P. Revelo, N.A. Bhowmick, S.W. Hayward, Altered TGF-b signaling in asubpopulation of human stromal cells promotes prostatic carcinogenesis,Cancer Res. 71 (2011) 12721281.

    [63] D. Bacman, S. Merkel, R. Croner, T. Papadopoulos, W. Brueckl, A. Dimmler,TGF-b receptor 2 downregulation in tumour-associated stroma worsensprognosis and high-grade tumours show more tumour-associatedmacrophages and lower TGF-b1 expression in colon carcinoma: aretrospective study, BMC Cancer 7 (2007) 156.

    [64] D. Barkan, H. Kleinman, J.L. Simmons, H. Asmussen, A.K. Kamaraju, M.J.Hoenorhoff, Z.Y. Liu, S.V. Costes, E.H. Cho, S. Lockett, C. Khanna, A.F.Chambers, J.E. Green, Inhibition of metastatic outgrowth from singledormant tumor cells by targeting the cytoskeleton, Cancer Res. 68 (2008)62416250.

    [65] B. Elenbaas, R.A. Weinberg, Heterotypic signaling between epithelial tumorcells and broblasts in carcinoma formation, Exp. Cell Res. 264 (2001) 169184.

    [66] M.J. Paszek, N. Zahir, K.R. Johnson, J.N. Lakins, G.I. Rozenberg, A. Gefen, C.A.Reinhart-King, S.S. Margulies, M. Dembo, D. Boettiger, D.A. Hammer, V.M.Weaver, Tensional homeostasis and the malignant phenotype, Cancer Cell. 8(2005) 241254.

    [67] J.T. Erler, V.M. Weaver, Three-dimensional context regulation of metastasis,Clin. Exp. Metastasis 26 (2009) 3549.

    [68] M.A. Taylor, J. Amin, D.A. Kirschmann, W.P. Schiemann, Lysyl oxidasecontributes to mechanotransduction-mediated regulation of transforminggrowth factor-b signaling in breast cancer cells, Neoplasia 13 (2011) 406418.

    [69] J.L. Leight, M.A. Wozniak, S. Chen, M.L. Lynch, C.S. Chen, Matrix rigidityregulates a switch between TGF-b1-induced apoptosis and epithelial-mesenchymal transition, Mol. Biol. Cell 23 (2012) 781791.

    [70] K.R. Levental, H. Yu, L. Kass, J.N. Lakins, M. Egeblad, J.T. Erler, S.F. Fong, K.Csiszar, A. Giaccia, W. Weninger, M. Yamauchi, D.L. Gasser, V.M. Weaver,Matrix crosslinking forces tumor progression by enhancing integrin signaling,Cell 139 (2009) 891906.

    [71] L.M. Postovit, D.E. Abbott, S.L. Payne, W.W. Wheaton, N.V. Margaryan, R.Sullivan, M.K. Jansen, K. Csiszar, M.J. Hendrix, D.A. Kirschmann, Hypoxia/reoxygenation: a dynamic regulator of lysyl oxidase-facilitated breast cancermigration, J. Cell Biochem. 103 (2008) 13691378.

    [72] J.T. Erler, K.L. Bennewith, T.R. Cox, G. Lang, D. Bird, A. Koong, Q.T. Le, A.J.Giaccia, Hypoxia-induced lysyl oxidase is a critical mediator of bone marrowcell recruitment to form the premetastatic niche, Cancer Cell 15 (2009) 3544.

    [73] J.T. Erler, K.L. Bennewith, M. Nicolau, N. Dornhofer, C. Kong, Q.T. Le, J.T. Chi,S.S. Jeffrey, A.J. Giaccia, Lysyl oxidase is essential for hypoxia-inducedmetastasis, Nature 440 (2006) 12221226.

    [74] H. Peinado, M. Del Carmen Iglesias-de la Cruz, D. Olmeda, K. Csiszar, K.S. Fong,S. Vega, M.A. Nieto, A. Cano, F. Portillo, A molecular role for lysyl oxidase-like2 enzyme in snail regulation and tumor progression, EMBO J. 24 (2005) 34463458.

    [75] H. Peinado, G. Moreno-Bueno, D. Hardisson, E. Perez-Gomez, V. Santos, M.Mendiola, J.I. de Diego, M. Nistal, M. Quintanilla, F. Portillo, A. Cano, Lysyloxidase-like 2 as a new poor prognosis marker of squamous cell carcinomas,Cancer Res. 68 (2008) 45414550.

    [76] G. Moreno-Bueno, F. Salvador, A. Martin, A. Floristan, E.P. Cuevas, V. Santos, A.

    tters xxx (2013) xxxxxx 9Montes, S. Morales, M.A. Castilla, A. Rojo-Sebastian, A. Martinez, D. Hardisson,K. Csiszar, F. Portillo, H. Peinado, J. Palacios, A. Cano, Lysyl oxidase-like 2

    F-b Paradox to EMT-MET programs, Cancer Lett. (2013),

  • er Le(LOXL2), a new regulator of cell polarity required for metastaticdissemination of basal-like breast carcinomas, EMBO Mol. Med. 3 (2011)528544.

    [77] M. Jo, R.D. Lester, V. Montel, B. Eastman, S. Takimoto, S.L. Gonias, Reversibilityof epithelial-mesenchymal transition (EMT) induced in breast cancer cells byactivation of urokinase receptor-dependent cell signaling, J. Biol. Chem. 284(2009) 2282522833.

    [78] R.D. Lester, M. Jo, V. Montel, S. Takimoto, S.L. Gonias, UPAR induces epithelial-mesenchymal transition in hypoxic breast cancer cells, J. Cell Biol. 178 (2007)425436.

    [79] S. Cannito, E. Novo, A. Compagnone, L. Valfre di Bonzo, C. Busletta, E. Zamara,C. Paternostro, D. Povero, A. Bandino, F. Bozzo, C. Cravanzola, V. Bravoco, S.Colombatto, M. Parola, Redox mechanisms switch on hypoxia-dependentepithelial-mesenchymal transition in cancer cells, Carcinogenesis 29 (2008)22672278.

    [80] E. Louie, S. Nik, J.S. Chen, M. Schmidt, B. Song, C. Pacson, X.F. Chen, S. Park, J.Ju, E.I. Chen, Identication of a stem-like cell population by exposingmetastatic breast cancer cell lines to repetitive cycles of hypoxia andreoxygenation, Breast Cancer Res. 12 (2010) R94.

    [81] F. Xing, H. Okuda, M. Watabe, A. Kobayashi, S.K. Pai, W. Liu, P.R. Pandey, K.Fukuda, S. Hirota, T. Sugai, G. Wakabayshi, K. Koeda, M. Kashiwaba, K. Suzuki,T. Chiba, M. Endo, Y.Y. Mo, K. Watabe, Hypoxia-induced Jagged2 promotesbreast cancer metastasis and self-renewal of cancer stem-like cells, Oncogene30 (2011) 40754086.

    [82] P. Bendinelli, E. Matteucci, P. Maroni, M.A. Desiderio, NF-jB activation,dependent on acetylation/deacetylation, contributes to HIF-1 activity andmigration of bone metastatic breast carcinoma cells, Mol. Cancer Res. 7(2009) 13281341.

    [83] J.S. Desgrosellier, D.A. Cheresh, Integrins in cancer: biological implicationsand therapeutic opportunities, Nature Rev. Cancer 10 (2010) 922.

    [84] A.J. Galliher, W.P. Schiemann, b3 Integrin and Src facilitate transforminggrowth factor-b mediated induction of epithelial-mesenchymal transition inmammary epithelial cells, Breast Cancer Res. 8 (2006) R42.

    [85] A.J. Galliher, W.P. Schiemann, Src phosphorylates Tyr284 in TGF-b type IIreceptor and regulates TGF-b stimulation of p38 MAPK during breast cancercell proliferation and invasion, Cancer Res. 67 (2007) 37523758.

    [86] A.J. Galliher-Beckley, W.P. Schiemann, Grb2 binding to Tyr284 in TbR-II isessential for mammary tumor growth and metastasis stimulated by TGF-b,Carcinogenesis 29 (2008) 244251.

    [87] M. Tian, W.P. Schiemann, PGE2 receptor EP2 mediates the antagonistic effectof COX-2 on TGF-b signaling during mammary tumorigenesis, FASEB J. 24(2010) 11051116.

    [88] M.K. Wendt, W.P. Schiemann, Therapeutic targeting of the focal adhesioncomplex prevents oncogenic TGF-b signaling and metastasis, Breast CancerRes. 11 (2009) R68.

    [89] M.K. Wendt, J.A. Smith, W.P. Schiemann, P130Cas is required for mammarytumor growth and transforming growth factor-b-mediated metastasisthrough regulation of Smad2/3 activity, J. Biol. Chem. 284 (2009) 3414534156.

    [90] N.A. Bhowmick, R. Zent, M. Ghiassi, M. McDonnell, H.L. Moses, Integrin b1signaling is necessary for transforming growth factor-b activation ofp38MAPK and epithelial plasticity, J. Biol. Chem. 276 (2001) 4670746713.

    [91] M.K. Wendt, B.J. Schiemann, J.G. Parvani, Y.H. Lee, Y. Kang, W.P. Schiemann,TGF-b stimulates Pyk2 expression as part of an epithelial-mesenchymaltransition program required for metastatic outgrowth of breast cancer,Oncogene (2012).

    [92] M. Korpal, J. Yan, X. Lu, S. Xu, D.A. Lerit, Y. Kang, Imaging transforming growthfactor-beta signaling dynamics and therapeutic response in breast cancerbone metastasis, Nature Med. 15 (2009) 960966.

    [93] T. Shibue, R.A. Weinberg, Integrin b1-focal adhesion kinase signaling directsthe proliferation of metastatic cancer cells disseminated in the lungs, Proc.Natl. Acad. Sci. USA 106 (2009) 1029010295.

    [94] D. Barkan, L.H. El Touny, A.M. Michalowski, J.A. Smith, I. Chu, A.S. Davis, J.D.Webster, S. Hoover, R.M. Simpson, J. Gauldie, J.E. Green, Metastatic growthfrom dormant cells induced by a Col-I-enriched brotic environment, CancerRes. 70 (2010) 57065716.

    [95] J.D. Whittard, S.E. Craig, A.P. Mould, A. Koch, O. Pertz, J.r. Engel, M.J.Humphries, E-cadherin is a ligand for integrin a2b1, Matrix Biol. 21 (2002)525532.

    [96] G.J. Mizejewski, Role of integrins in cancer: survey of expression patterns,Proc. Soc. Exp. Biol. Med. 222 (1999) 124138.

    [97] V.M. Weaver, S. Lelievre, J.N. Lakins, M.A. Chrenek, J.C. Jones, F. Giancotti, Z.Werb, M.J. Bissell, b4 Integrin-dependent formation of polarized three-dimensional architecture confers resistance to apoptosis in normal andmalignant mammary epithelium, Cancer Cell 2 (2002) 205216.

    [98] T. Watabe, K. Miyazono, Roles of TGF-b family signaling in stem cell renewaland differentiation, Cell Res. 19 (2009) 103115.

    [99] C.J. Creighton, J.C. Chang, J.M. Rosen, Epithelial-mesenchymal transition(EMT) in tumor-initiating cells and its clinical implications in breast cancer, J.Mammary Gland Biol. Neoplasia 15 (2010) 253260.

    [100] S.A. Mani, W. Guo, M.J. Liao, E.N. Eaton, A. Ayyanan, A.Y. Zhou, M. Brooks, F.Reinhard, C.C. Zhang, M. Shipitsin, L.L. Campbell, K. Polyak, C. Brisken, J. Yang,R.A. Weinberg, The epithelial-mesenchymal transition generates cells with

    10 C.D. Morrison et al. / Cancproperties of stem cells, Cell 133 (2008) 704715.[101] V.L. Battula, K.W. Evans, B.G. Hollier, Y. Shi, F.C. Marini, A. Ayyanan, R.Y.

    Wang, C. Brisken, R. Guerra, M. Andreeff, S.A. Mani, Epithelial-mesenchymal

    Please cite this article in press as: C.D. Morrison et al., The relevance of the TG10.1016/j.canlet.2013.02.048transition-derived cells exhibit multilineage differentiation potential similarto mesenchymal stem cells, Stem Cells 28 (2010) 14351445.

    [102] M. Shipitsin, L.L. Campbell, P. Argani, S. Weremowicz, N. Bloushtain-Qimron, J. Yao, T. Nikolskaya, T. Serebryiskaya, R. Beroukhim, M. Hu, M.K.Halushka, S. Sukumar, L.M. Parker, K.S. Anderson, L.N. Harris, J.E. Garber, A.L.Richardson, S.J. Schnitt, Y. Nikolsky, R.S. Gelman, K. Polyak,Molecular denition of breast tumor heterogeneity, Cancer Cell 11 (2007)259273.

    [103] I. Ben-Porath, M.W. Thomson, V.J. Carey, R. Ge, G.W. Bell, A. Regev, R.A.Weinberg, An embryonic stem cell-like gene expression signature in poorlydifferentiated aggressive human tumors, Nat. Genet. 40 (2008) 499507.

    [104] A.P. Morel, M. Lievre, C. Thomas, G. Hinkal, S. Ansieau, A. Puisieux, Generationof breast cancer stem cells through epithelial-mesenchymal transition, PLoSONE 3 (2008) e2888.

    [105] C.J. Creighton, X. Li, M. Landis, J.M. Dixon, V.M. Neumeister, A. Sjolund, D.L.Rimm, H. Wong, A. Rodriguez, J.I. Herschkowitz, C. Fan, X. Zhang, X. He, A.Pavlick, M.C. Gutierrez, L. Renshaw, A.A. Larionov, D. Faratian, S.G.Hilsenbeck, C.M. Perou, M.T. Lewis, J.M. Rosen, J.C. Chang, Residual breastcancers after conventional therapy display mesenchymal as well as tumor-initiating features, Proc. Natl. Acad. Sci. USA 106 (2009) 1382013825.

    [106] S.G. Penheiter, R.D. Singh, C.E. Repellin, M.C. Wilkes, M. Edens, P.H. Howe, R.E.Pagano, E.B. Leof, Type II transforming growth factor-b receptor recycling isdependent upon the clathrin adaptor protein Dab2, Mol. Biol. Cell 21 (2010)40094019.

    [107] E.A. Turley, M. Veiseh, D.C. Radisky, M.J. Bissell, Mechanisms of disease:epithelial-mesenchymal transitiondoes cellular plasticity fuel neoplasticprogression?, Nat Clin. Pract. Oncol. 5 (2008) 280290.

    [108] M.K. Asiedu, J.N. Ingle, M.D. Behrens, D.C. Radisky, K.L. Knutson, TGFb/TNFa-mediated epithelial-mesenchymal transition generates breast cancer stemcells with a claudin-low phenotype, Cancer Res. 71 (2011) 47074719.

    [109] B.T. Hennessy, A.M. Gonzalez-Angulo, K. Stemke-Hale, M.Z. Gilcrease, S.Krishnamurthy, J.S. Lee, J. Fridlyand, A. Sahin, R. Agarwal, C. Joy, W. Liu, D.Stivers, K. Baggerly, M. Carey, A. Lluch, C. Monteagudo, X. He, V. Weigman, C.Fan, J. Palazzo, G.N. Hortobagyi, L.K. Nolden, N.J. Wang, V. Valero, J.W. Gray,C.M. Perou, G.B. Mills, Characterization of a naturally occurring breast cancersubset enriched in epithelial-to-mesenchymal transition and stem cellcharacteristics, Cancer Res. 69 (2009) 41164124.

    [110] A. Prat, J.S. Parker, O. Karginova, C. Fan, C. Livasy, J.I. Herschkowitz, X. He, C.M.Perou, Phenotypic and molecular characterization of the claudin-lowintrinsic subtype of breast cancer, Breast Cancer Res. 12 (2010) R68.

    [111] J.H. Taube, J.I. Herschkowitz, K. Komurov, A.Y. Zhou, S. Gupta, J. Yang, K.Hartwell, T.T. Onder, P.B. Gupta, K.W. Evans, B.G. Hollier, P.T. Ram, E.S.Lander, J.M. Rosen, R.A. Weinberg, S.A. Mani, Core epithelial-to-mesenchymaltransition interactome gene-expression signature is associated with claudin-low and metaplastic breast cancer subtypes, Proc. Natl. Acad. Sci. USA 107(2010) 1544915454.

    [112] R.A. Taylor, R. Toivanen, M. Frydenberg, J. Pedersen, L. Harewood, B.Australian Prostate Cancer, A.T. Collins, N.J. Maitland, G.P. Risbridger,Human epithelial basal cells are cells of origin of prostate cancer,independent of CD133 status. Stem Cells, vol. 30, 2012, pp. 108796.

    [113] S. Floor, W.C. van Staveren, D. Larsimont, J.E. Dumont, C. Maenhaut, Cancercells in epithelial-to-mesenchymal transition and tumor-propagating-cancerstem cells: distinct, overlapping or same populations, Oncogene 30 (2011)46094621.

    [114] T. Tsuji, S. Ibaragi, K. Shima, M.G. Hu, M. Katsurano, A. Sasaki, G.F. Hu,Epithelial-mesenchymal transition induced by growth suppressor p12CDK2-AP1 promotes tumor cell local invasion but suppresses distant colonygrowth, Cancer Res. 68 (2008) 1037710386.

    [115] T. Celia-Terrassa, O. Meca-Cortes, F. Mateo, A.M. de Paz, N. Rubio, A. Arnal-Estape, B.J. Ell, R. Bermudo, A. Diaz, M. Guerra-Rebollo, J.J. Lozano, C. Estaras,C. Ulloa, D. Alvarez-Simon, J. Mila, R. Vilella, R. Paciucci, M. Martinez-Balbas,A.G. de Herreros, R.R. Gomis, Y. Kang, J. Blanco, P.L. Fernandez, T.M. Thomson,Epithelial-mesenchymal transition can suppress major attributes of humanepithelial tumor-initiating cells, J. Clin. Invest. 122 (2012) 18491868.

    [116] Z. Yu, T.G. Pestell, M.P. Lisanti, R.G. Pestell, Cancer stem cells, Int. J. Biochem.Cell Biol. 44 (2012) 21442151.

    [117] M. Itoh, M.J. Bissell, The organization of tight junctions in epithelia:implications for mammary gland biology and breast tumorigenesis, J.Mammary Gland. Biol. Neoplasia 8 (2003) 449462.

    [118] G. Agiostratidou, J. Hulit, G.R. Phillips, R.B. Hazan, Differential cadherinexpression: potential markers for epithelial to mesenchymal transformationduring tumor progression, J. Mammary Gland. Biol. Neoplasia 12 (2007) 127133.

    [119] C.L. Chaffer, E.W. Thompson, E.D. Williams, Mesenchymal to epithelialtransition in development and disease, Cells Tissues Organs. 185 (2007) 719.

    [120] T. Brabletz, To differentiate or not routes towards metastasis, Nature Rev.,Cancer 12 (2012) 425436.

    [121] The TGF-b Family, Cold Spring Harbor Laboratory Press, Cold Spring Harbor,2008.

    [122] J.T. Buijs, N.V. Henriquez, P.G. van Overveld, G. van der Horst, I. Que, R.Schwaninger, C. Rentsch, P. Ten Dijke, A.M. Cleton-Jansen, K. Driouch, R.Lidereau, R. Bachelier, S. Vukicevic, P. Clezardin, S.E. Papapoulos, M.G.

    tters xxx (2013) xxxxxxCecchini, C.W. Lowik, G. van der Pluijm, Bone morphogenetic protein 7 inthe development and treatment of bone metastases from breast cancer,Cancer Res. 67 (2007) 87428751.

    F-b Paradox to EMT-MET programs, Cancer Lett. (2013),

  • [123] J.T. Buijs, G. van der Horst, C. van den Hoogen, H. Cheung, B. de Rooij, J. Kroon,M. Petersen, P.G. van Overveld, R.C. Pelger, G. van der Pluijm, The BMP2/7heterodimer inhibits the human breast cancer stem cell subpopulation andbone metastases formation, Oncogene 31 (2012) 21642174.

    [124] P. Owens, M.W. Pickup, S.V. Novitskiy, A. Chytil, A.E. Gorska, M.E. Aakre, J.West, H.L. Moses, Disruption of bone morphogenetic protein receptor 2(BMPR2) in mammary tumors promotes metastases through cellautonomous and paracrine mediators, Proc. Natl. Acad. Sci. USA 109 (2012)28142819.

    [125] S. Giampieri, C. Manning, S. Hooper, L. Jones, C.S. Hill, E. Sahai, Localized andreversible TGF-b signalling switches breast cancer cells from cohesive tosingle cell motility, Nat. Cell Biol. 11 (2009) 12871296.

    [126] R. Li, J. Liang, S. Ni, T. Zhou, X. Qing, H. Li, W. He, J. Chen, F. Li, Q. Zhuang, B.Qin, J. Xu, W. Li, J. Yang, Y. Gan, D. Qin, S. Feng, H. Song, D. Yang, B. Zhang, L.Zeng, L. Lai, M.A. Esteban, D. Pei, A mesenchymal-to-epithelial transitioninitiates and is required for the nuclear reprogramming of mouse broblasts,Cell Stem. Cell 7 (2010) 5163.

    [127] E.N. Howe, D.R. Cochrane, J.K. Richer, The miR-200 and miR-221/222microRNA families: opposing effects on epithelial identity, J. MammaryGland. Biol. Neoplasia 17 (2012) 6577.

    [128] C.J. Chang, C.H. Chao, W. Xia, J.Y. Yang, Y. Xiong, C.W. Li, W.H. Yu, S.K.Rehman, J.L. Hsu, H.H. Lee, M. Liu, C.T. Chen, D. Yu, M.C. Hung, P53 regulatesepithelial-mesenchymal transition and stem cell properties throughmodulating miRNAs, Nat. Cell Biol. 13 (2011) 317323.

    [129] B. Cieply, P.t. Riley, P.M. Pifer, J. Widmeyer, J.B. Addison, A.V. Ivanov, J. Denvir,S.M. Frisch, Suppression of the epithelial-mesenchymal transition bygrainyhead-like-2, Cancer Res. 72 (2012) 24402453.

    [130] D. Zhang, D.R. Richardson, Endoplasmic reticulum protein 29 (ERp29): Anemerging role in cancer, Int. J. Biochem. Cell Biol. 43 (2011) 3336.

    [131] I.F. Bambang, Y.K. Lee, D.R. Richardson, D. Zhang, Endoplasmic reticulumprotein 29 regulates epithelial cell integrity during the mesenchymal-epithelial transition in breast cancer cells, Oncogene (2012).

    [132] T.M. Allington, A.J. Galliher-Beckley, W.P. Schiemann, Activated Abl kinaseinhibits oncogenic transforming growth factor-b signaling and tumorigenesisin mammary tumors, FASEB J. 23 (2009) 42314243.

    [133] T. Hunter, Treatment for chronic myelogenous leukemia: the long road toimatinib, J. Clin. Invest. 117 (2007) 20362043.

    [134] D.W. Sherbenou, B.J. Druker, Applying the discovery of the Philadelphiachromosome, J. Clin. Invest. 117 (2007) 20672074.

    [135] H.K. Chew, W.E. Barlow, K. Albain, D. Lew, A. Gown, D.F. Hayes, J. Gralow, G.N.Hortobagyi, R. Livingston, A phase II study of imatinib mesylate andcapecitabine in metastatic breast cancer: Southwest Oncology Group Study0338, Clin. Breast Cancer 8 (2008) 511515.

    [136] M. Cristofanilli, P. Morandi, S. Krishnamurthy, J.M. Reuben, B.N. Lee, D.Francis, D.J. Booser, M.C. Green, B.K. Arun, L. Pusztai, A. Lopez, R. Islam, V.Valero, G.N. Hortobagyi, Imatinib mesylate (Gleevec) in advanced breastcancer-expressing C-Kit or PDGFR-b: clinical activity and biologicalcorrelations, Ann. Oncol. 19 (2008) 17131719.

    [137] S. Modi, A.D. Seidman, M. Dickler, M. Moasser, G. DAndrea, M.E. Moynahan, J.Menell, K.S. Panageas, L.K. Tan, L. Norton, C.A. Hudis, A phase II trial ofimatinib mesylate monotherapy in patients with metastatic breast cancer,Breast Cancer Res. Treat. 90 (2005) 157163.

    C.D. Morrison et al. / Cancer Letters xxx (2013) xxxxxx 11Please cite this article in press as: C.D. Morrison et al., The relevance of the TG10.1016/j.canlet.2013.02.048F-b Paradox to EMT-MET programs, Cancer Lett. (2013),

    The relevance of the TGF- Paradox to EMT-MET pr1 Introduction2 TGF- signaling3 TGF- and post-transcriptional regulation of E3.1 TGF-, microRNAs, and EMT3.2 TGF-, alternative mRNA splicing, and EMT

    4 The influence of the tumor microenvironment on4.1 TGF-, cancer-associated fibroblasts, and EM4.2 TGF- and tissue compliance during EMT progr4.3 TGF- and hypoxia during EMT programs4.4 TGF- and integrins during EMT programs

    5 TGF- and cancer stem cells (CSCs) during EMT 6 Mesenchymal-epithelial transition (MET) programs6.1 Bone morphogenetic proteins (BMPs) and MET programs6.2 Transcriptional regulators of MET6.3 The miR-200 family and MET programs6.4 Grainyhead-like-2 (GRHL2) and MET programs6.5 Endoplasmic reticulum protein 29 (ERp29) and MET programs6.6 c-Abl and MET programs

    7 Conclusion and future directionsConflict of interestAcknowledgementsReferences


View more >