SEARCH

SEARCH BY CITATION

Keywords:

  • ductal outgrowths;
  • luminal epithelial cells;
  • repopulation assay;
  • stem cells;
  • wnt-signaling

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Conflict of interest
  9. References
  10. Supporting Information

Postnatally, the mammary gland undergoes continuous morphogenesis and thereby is especially prone to malignant transformation. Thus, the maintenance of the epithelium depends on a tight control of stem cell recruitment. We have previously shown that epithelial overexpression of the EphB4 receptor results in defective mammary epithelial development and conferred a metastasizing tumor phenotype on experimental mouse mammary tumors accompanied by a preponderance of progenitor cells. To analyze the effect of EphB4 overexpression on mammary epithelial cell fate, we have used Fluorescence Activated Cell Sorting (FACS) analyses to quantify epithelial sub-populations and repopulation assays of cleared fat pads to investigate their regenerative potential. These experiments revealed that deregulated EphB4 expression leads to an augmentation of bi-potent progenitor cells and to a shift of the differentiation pathway towards the luminal lineage. The analyses of the ductal outgrowths indicated that EphB4 overexpression leads to enforced branching activity, impedes ductal differentiation and stimulates angiogenesis. To elucidate the mechanisms forwarding EphB4 signals, we have compared the expression profile of defined cell populations between EphB4 transgene and wild type mammary glands concentrating on the wnt signaling pathway and on genes implicated in cell migration. With respect to wnt signaling, the progenitor cell population was the most affected, whereas the stem cell-enriched population showed the most pronounced deregulation of migration-associated genes. Thus, the luminal epithelial EphB4 signaling contributes, most likely via wnt signaling, to the regulation of migration and cell fate of early progenitors and is involved in the determination of branching points along the ductal tree.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Conflict of interest
  9. References
  10. Supporting Information

The mammary parenchyme is a tissue that grows into the surrounding mesenchyme in strictly separated periods, beginning with embryonic sprouting from the initial mammary buds to subsequently form the rudimentary mammary anlagen. After a quiescent stage until the onset of puberty, the mammary anlagen invade the entire fat pad to form the virgin adult mammary ductal network and now enter a stage of tissue homeostasis with minor cycles of proliferation, differentiation and involution during the estrous cycles. And finally, with the hormonal stimulation at pregnancy the ductal network undergoes secondary and tertiary branching and side-sprouting followed by terminal differentiation to form the secretory lobulo-alveolar units for milk production and secretion (Sternlicht 2006). The nature of these processes including cycles of proliferation, differentiation and tissue remodeling with epithelial/mesenchymal replacement and also the endocrine hormonal stimulation as master regulator makes the mammary epithelium an interesting model to investigate benign tissue invasion. Especially in this aspect, breast cancer research relies on a better understanding of mammary gland development through postnatal progression during puberty, estrous cycles and reproductive cycles. Although a major progress in the understanding of the pivotal processes is noticeable, especially the molecular mechanisms by which branching points are determined, epithelial structures surpass the basement membrane and invade the surrounding tissue to attain their functionality need further investigation.

The composition of the surrounding extracellular matrix (ECM) was repeatedly shown to influence branching morphogenesis (Stahl et al. 1997; Wiseman & Werb 2002; Fata et al. 2004; Lo et al. 2012) and recently Nguyen-Ngoc and Ewald were able to directly correlate a defined composition of ECM components with ductal elongation, branching and migration in vitro (Nguyen-Ngoc & Ewald 2013). Of the postnatal morphogenic events, the ductal elongation and primary branching through the terminal end buds (TEB) are the best characterized. The formation of bifurcations through interaction with the stroma and subsequent deposition of stromal matrix acts as initiator of primary branching. Simultaneously, ECM in direction of growth is being degraded by secreted matrix metalloproteinases (MMPs), which in turn give rise to bioactive fragments stimulating epithelial migration, which is further supported by the sheer cell mass of TEB body cells (Koshikawa et al. 2000). At the trailing end of the TEBs macrophages contribute to collagen I deposition causing the club shape of TEBs and eliciting ductal formation (Ingman et al. 2006). On the other hand, for side-sprout formation epithelial cells first have to cross the myoepithelial cell layer before they can pass the (degraded) basement membrane and invade the surrounding fibrous stroma (Wiseman & Werb 2002). These cells, however, lack the high proliferative activity observed in the body cells of the TEBs to promote movement. The integrity of the basement membrane appears crucial for not only correct differentiation and polarity of the cell but also functional responsiveness to hormonal stimulation. Integrins thereby function as control points to maintain signaling cues essential for proper epithelial polarity and allocation. Thus, interaction with the stroma when reforming the basement membrane around the growing side-sprouts, accompanied by an epithelial-to-mesenchymal transition (EMT)-like process to allow collective epithelial migration might be coordinated by direct cell–cell communication among epithelial cells in local niches (Fata et al. 2004). As side-sprouting is mainly induced during the estrous cycle and prominently during pregnancy, hormonal stimulation and paracrine mechanisms interpreting these signals are certainly important mediators (Mallepell et al. 2006; Brisken & Duss 2007). Among others, amphiregulin and epidermal growth factor (EGF) signaling are key players in paracrine communication, probably also by influencing the interaction with the stroma via integrins (Luetteke et al. 1999). Interestingly, it has recently been shown that EGF signaling is involved in the transmission of estrogen stimulation on Esrα-negative breast cancer stem cells (Harrison et al. 2013).

EphB4/ephrin-B2 signaling might represent a possible mediator between endocrine stimulation and locally restricted signal interpretation in spatially defined niches. EphB4 is a member of the Eph receptor tyrosine kinase family, which is divided into the EphA and EphB subgroups according to their binding affinity to either ephrin-A or ephrin-B ligands. Ephrin-A ligands are bound to the cell membrane via a GPI-tail, whereas ephrin-B ligands are true transmembrane proteins. Thus, receptor ligand interaction requires direct cell–cell contact and results in bi-directional signaling: forward signaling elicited by the receptor and reverse signaling emanating from the ligand (Bush & Soriano 2012). Eph-ephrin signaling is involved in pattern formation, tissue compartmentalization and guided cell migration, thereby playing an essential role not only during embryonic development but also in adult tissue homeostasis. In particular, EphB4 and its cognate ligand ephrin-B2 are intimately involved in angiogenesis, bone homeostasis and intestinal stem cell regulation (Adams 2002; Batlle et al. 2002; Cheng et al. 2002; Zhao et al. 2006; Pasquale 2010). In the mammary gland, EphB4 is expressed in an estrogen-dependent manner preferentially in the basal cell compartment of the parenchyma, induced during puberty and downregulated during the luteal phase of the estrous cycle and pregnancy. In contrast, ephrin-B2 is constitutively expressed predominantly in the luminal epithelium (Nikolova et al. 1998). The phenotypic impact of deregulated EphB4 or ephrin-B2 expression in the mammary epithelium of transgenic mice clearly revealed an important role of both proteins in mammary gland morphogenesis (Munarini et al. 2002; Haldimann et al. 2009). Mouse Mammary Tumor Virus – Long Terminal Repeat (MMTV-LTR) driven expression of EphB4 resulted in overexpression of EphB4 in the basal and luminal epithelial cells starting at puberty and culminating during pregnancy and lactation. This unscheduled overexpression of EphB4 led to a delay in mammary development during puberty and pregnancy, as well as to a disturbed ductal and alveolar architecture. Moreover, epithelial overexpression of EphB4 also interfered with mammary vascularization and imposed an aggressive and metastasizing phenotype on NeuT induced mammary tumors (Munarini et al. 2002; Kaenel et al. 2011). Thus, the EphB4-ephrin-B2 crosstalk may be involved in the specification of hormonal signaling cues required for the establishment and maintenance of a functional organ structure. Due to its kinase activity upon ephrin-B2 binding, EphB4 can interact with several other signaling pathways. EphB receptors control integrin-mediated matrix interaction (Huynh-Do et al. 1999, 2002; Miao et al. 2005) and in epithelial tissues EphB/ephrin-B signaling was shown to regulate cell–cell adhesion via ADAM10 and E-cadherin cleavage (Solanas et al. 2011). Thus, EphB4 signaling has the potential to influence cell fate decisions of mammary epithelial cells at several stages of development.

The reproductive and non-reproductive cycles of the mammary gland originate from stem cells and their progenitors, which are located in their niches dispersed throughout the epithelial tree (Kobayashi et al. 2012). Although the understanding of the molecules involved in the homeostasis of the stem cell niche is beginning to emerge (Brisken & Duss 2007), little is still known about the mechanisms governing the migration of the progenitors out of the niche and controlling cell fate decisions. In particular, the origin of cells maintaining the luminal cell population is still controversially discussed (Smalley et al. 2012). Classically, the repopulation assay of cleared mammary fat pads, a procedure introduced in 1959 by DeOme et al. (1959), represents the gold standard to investigate the developmental potential of isolated mammary cell population including stem cell enriched populations (Shackleton et al. 2006; Stingl et al. 2006). Together with in vitro experiments this assay helped identify an epithelial cell hierarchy in the mammary gland, although numerous questions still remain to be tackled. The differentiation hierarchy initiates from a cell population that is capable of repeatedly reconstituting an entire mammary gland in vivo, the mammary stem cells. The stem cells progress into bi-potent progenitors, capable of differentiating into either lineage of the mammary epithelium the ductal luminal cells, the secretory alveolar cells and the contractile myoepithelial cells. These bi-potent cells, however, exhibit a limited life-span and have lost the capacity to repopulate cleared mammary fat pads. After this stage, the (irreversible?) cell fate decision takes place and the cells differentiate either into luminal or basal precursors. Finally, the ductal epithelium differentiates from the luminal precursors and the myoepithelium develops from the basal precursor cells (Shackleton et al. 2006; Stingl et al. 2006). However, whether luminal precursors also give rise to alveolar cells or whether the alveolar cells derive from the basal cell precursors is still a matter of debate. Similarly, the differentiation hierarchy of the estrogen receptor-positive sensory cells remains to be defined (Visvader & Smith 2011; Stingl 2009). The stem cells as well as their descendants are thought to be localized at the tip of the growing end buds during puberty and thereafter they are dispersed along the ductal epithelium embedded in the basal myoepithelial cell layer (Brisken & Duss 2007).

The repopulation assay, however, cannot only be used to verify the stem cell characteristics of isolated mammary cells but also allows the investigation of the differentiation pathways and pattern formation of genetically modified cells (Cohn et al. 2010; LaMarca et al. 2010; Vafaizadeh et al. 2010; Kaenel et al. 2012; Gastaldi et al. 2013). In order to characterize the developmental processes regulated by EphB4, we have adopted this assay and investigated the differentiation potential of transgenic mammary epithelial cells overexpressing the EphB4 receptor. In this report we demonstrate that the deregulated EphB4 expression does not affect the self-renewal of mammary stem cells, but is shifting their differentiation pathway towards the luminal cell fate. Moreover, the outgrowths were characterized by extensive secondary and tertiary side-branching as well as by prominent vascularization. RNA profiling revealed that the luminal EphB4 expression exerts its effects most probably by interfering with wnt signaling in the progenitor cell populations thereby altering their matrix interaction and migration. Thus, our results indicate that EphB4 signaling contributes to the regulation of migration and cell fate decision of early progenitors and is involved in the determination of branching activity.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Conflict of interest
  9. References
  10. Supporting Information

Transgenic lines

The establishment of the EphB4 transgenic mouse line has been described previously (Munarini et al. 2002). The cDNA of the murine EphB4 depleted of its polyadenylation was placed under the control of the MMTV-LTR promoter and transgenic mice were established by pronuclear injection. Transgenic mice were crossed over six generations with inbred C57/Bl6 mice (Charles River, Wiga, Sulzfeld, Germany) in order to reduce complexity of the genetic background.

All animal studies were conducted according to the institutional ethics guidelines.

Cell isolation, FACS analysis and cell sorting

The inguinal #4 mammary glands were removed from adult female virgin mice. After mechanical dissection, the tissue was transferred to enzymatic digestion in complete EpiCult-B medium supplemented with collagenase/hyaluronidase and 5% (v/v) fetal calf serum (all reagents StemCell Technologies, Grenoble, France). Generation of a single cell suspension for antibody staining was performed according to the protocol of the manufacturer.

The antibodies used for flow cytometry analysis and cell sorting were purchased from Biolegend (San Diego, CA, USA) and included biotinylated TER119 (clone TER-119), biotinylated CD31 (clone MEC13.3), biotinylated CD45 (clone 30-F11), CD49f-fluorescein isothiocyanate (FITC) (clone GoH3) and CD24-PE (clone M1/69). Isotype controls were performed with the corresponding mouse IgG1 κ Isotype Control antibodies from Biolegend. Compensation was performed using the BD CompBeads Compensation Particles Set (BD Biosciences, San Jose, CA, USA).

For staining, cells were incubated at 5–10 × 105 cells/mL in phosphate-buffered saline (PBS), 2.5% (v/v) fetal calf serum, 10 mmol/L ethylenediaminetetraacetic acid (EDTA), 0.2% (w/v) sodium azide containing the primary antibodies for 25 min at 4°C and subsequently 15 min at 4°C with Streptavidin-PE/Cy7 (Biolegend. Analysis was carried out with FACSDiva software on a BD LSRII Special Order System (BD Biosciences) and cells were sorted on a BD FACS ARIA (BD Biosciences). Samples were primarily gated on forward- and side-scatter. Contaminating endothelial and hematopoietic cells were excluded by gating for CD31, CD45 and TER119 negative cells. Endothelial and hematopoietic lineage negative cells were analyzed according to their CD49f and CD24 expression as previously shown (Kaenel et al. 2012).

Mammary fat pad cell transplantations and repopulation analysis

Mammary glands of 18–21 days old C57/Bl6 mice (Charles River, Wiga, Sulzfeld, Germany) were cleared of endogenous epithelium according to the procedure introduced by DeOme et al. (1959). We used recipient mice with an intact immune system in order to allow for any cellular interaction participating in epithelial morphogenesis. Test cells were prepared as described above and suspended at required concentrations in Dulbecco's modified eagle medium (DMEM) (Sigma Aldrich, Buchs, Switzerland) with 2% (v/v) FCS. 10 000 cells in 5–10 μL were injected into each cleared fat pad. After 8 or 12 weeks of growth, the mice were killed and mammary glands were excised and fixed in Carnoy′s Solution for 4 h. Subsequently, mammary glands were rehydrated and stained with Carmine Alum for outgrowth analysis. After recording, mammary glands were embedded in paraffin and sectioned for microscopy.

Immunohistochemistry and laser scanning microscopy

Hematoxylin and Eosin (Merck, Darmstadt, Germany) stainings were performed as previously described (Nikolova et al. 1998). Antibodies used for immunohistochemical analysis included EphB4 (Acris, Herford, Germany, polyclonal), CD271 (Millipore, Zug, Switzerland, clone EP1039Y), CD31 (Haldimann et al. 2009), CD61 (Milliporeclone EPR2417Y), MUC-1 (Acris, Herford, Germany, polyclonal), CK19 (Santa Cruz, Heidelberg, Germany, polyclonal), EpCAM (OriGene, Rockvielle, MD, USA, clone 144), smooth muscle actin (SMA) (Sigma Aldrich, clone 1A4), Par-3 (Proteintech Group, Manchester, UK, polyclonal), ZO-1 (Millipore, clone R40.76), Ki67 (Novus Biologicals, Littleton, CO, USA, clone Sp6), CD133 (Millipore, clone 17A6.1) and Fzd-2 (Novus Biologicals, polyclonal). Tissue preparation and conventional immunohistochemistry were done as described (Nikolova et al. 1998) and epitope retrieval was done either by microwaving in citrate buffer pH 5.5 (CD31, CD61, CD133, MUC-1) or in Tris-EDTA (TE) buffer pH 8.0 (CD271, CK19, EpCAM, EphB4, Fzd2, Ki67, Par-3, SMA, ZO-1).

For laser scanning microscopy the secondary antibodies included anti-rabbit IgG Alexa Fluor 488 (Life Technologies, Zug, Switzerland), anti-mouse IgG Cy3, anti-rabbit IgG Cy3, anti-rabbit IgG Cy5 and anti-goat IgG FITC (all from Millipore) antibodies. After incubation with primary antibodies overnight the sections were washed three times with TBS and labelled secondary antibodies were applied in mixtures according to the demanded combinations and at the concentrations indicated by the manufacturer. Sections were incubated for 2–4 h at RT in the dark, rinsed, stained with 4´6´-diamidino-2-phenylindole dihydrochloride (DAPI) (Sigma Aldrich), embedded in Mowiol (Sigma Aldrich) and analyzed using a LSM Zeiss Meta 510 (Carl Zeiss, Jena, Germany).

Gene expression using RT2 PCR arrays

RNA was isolated from sorted cell populations using the RNeasy Micro Kit from Qiagen (Qiagen, Hombrechtikon, Switzerland) and cDNA was synthesized using the Superscript III Reverse Transcriptase kit from Life Technologies according to the manufacturer's instructions.

The Mouse Wnt Signaling Pathway polymerase chain reaction (PCR) array (PAMM-043, SABiosciences, Qiagen, Hombrechtikon, Switzerland) and a custom PCR array (Table S1) were performed according to the manufacturer`s protocol with the required corresponding components (SABiosciences, Qiagen, Hombrechtikon, Switzerland).

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Conflict of interest
  9. References
  10. Supporting Information

In our previous work we have shown that unscheduled activation of EphB4 signaling results in defects/delays during mammary epithelial development and vascularization (Munarini et al. 2002; Andres & Djonov 2010). Moreover, EphB4 overexpression conferred a metastasizing tumor phenotype, which was accompanied by a preponderance of progenitor cells suggesting a disturbed developmental hierarchy in the epithelial cell population (Kaenel et al. 2011). In order to analyze the effect of EphB4 overexpression on mammary epithelial cell fate, we have used FACS analyses to quantify defined epithelial cell populations and repopulation assays of cleared fat pads to investigate their regenerative potential (Fig. S1).

EphB4 overexpression leads to an increase in luminal and progenitor cells

Mammary epithelial cell sub-populations can be identified by the constellation of cell surface marker expression. In particular, the level of CD24 and CD49f expression is indicative for the luminal (CD24+/CD49flow) and the basal (CD24+/CD49f+) epithelial cell fraction, as well as a progenitor cell-enriched (CD24+/CD49f++) and the stem cell-enriched (CD24low/CD49f++) cell populations (Stingl 2009). We have used this protocol to separate these epithelial sub-populations from adult virgin wild type (WT) and transgenic EphB4 mice by FACS analysis. Endothelial, hematopoietic and stromal cells have been sorted out using CD31, CD45 and TER119 before plotting for the markers CD24 and CD49f (Fig. S2). Quantification of the basal, luminal, progenitor and stem cell population in the two mouse strains revealed that EphB4 overexpression has hardly any effect on the proportion of basal and stem cells (Fig. 1A). In contrast, we observed a large increase in luminal cells in the EphB4 mice (= 0.0016) and also a significant increase in the bi-potent progenitor population (= 0.04). We therefore quantified proliferating cells by staining for Ki67 in adult virgin EphB4 and WT mammary glands to supply a possible explanation for the increase in these two populations. The amount of proliferating epithelial cells increased from 1.33 ± 1.88% in WT to 12.66 ± 0.9% in the EphB4 mammary glands (= 0.003). These cells were predominantly located basally between the luminal epithelial and myoepithelial cell layers (data not shown) and thus, could indeed represent bi-potent cells differentiating towards the luminal pathway.

image

Figure 1. Quantification and repopulation capability of EphB4 transgenic and wild type (WT) mammary epithelial cells. (A) Epithelial cell sub-populations were quantified by fluorescence-activated cell sorting (FACS) analysis plotting lineage negative epithelial cells for their CD24 and CD49f expression. Quantified populations included basal cells (CD24+/CD49f+), luminal cells (CD24+/CD49flow), a bi-potent progenitor-enriched population (CD24+/CD49f++) and a stem cell-enriched population (CD24low/CD49f++). Grey bars represent the WT control and red bars the EphB4 transgenic cells. Asterisks mark significant differences. (B) Isolated epithelial bulk populations of WT and EphB4 transgenic mammary glands were injected into cleared fat pads of 3 week old WT recipient mice and analyzed by whole mount staining after 8 and 12 weeks. Outgrowths were classified either as ductal outgrowths, formation of cell aggregates or no growth. (C) EphB4 expression in outgrowths. Sections from cell aggregates at 8 weeks or ductal outgrowths at 12 weeks were reacted with anti-EphB4 antibody. Scale bars represent 40 μm.

Download figure to PowerPoint

In order to analyze the growth potential, the collective epithelial cell populations of WT and EphB4 transgenic mice have been used to repopulate cleared mammary fat pads of recipient mice. We found a similar repopulation frequency comparing WT and EphB4 expressing cells, confirming the equal frequency of stem cells in both strains found by the FACS quantification. Both the WT (12.5%) and the EphB4 cells (8.8%) have a low success rate of ductal outgrowth after 8 weeks of regeneration (Fig. 1B). Hence, we extended the regeneration time after cell transplantation from 8 to 12 weeks. At this time point, cells of both strains showed a remarkable increase of outgrowths, with 62.5% for WT and 60% for EphB4-derived cells. Strikingly, the majority of analyzed fat pads merely exhibited large cell aggregates after 8 weeks of growth, which contained a mixed population of cells. This suggests that the aggregates represent surviving injected cells, which are the point of origin of the ductal outgrowths seen after 12 weeks. Yet, at 8 weeks the injected EphB4 transgenic cells expressed similar levels of EphB4 as the WT control. With ongoing differentiation, EphB4 expression was reduced in WT outgrowths and was mainly restricted to the myoepithelial/basal cell layer. In contrast, in the transgenic outgrowths EphB4 expression was considerably induced in the luminal cells most probably representing the MMTV promoter activation (Fig. 1C).

Transplanted EphB4 cells stimulate angiogenesis in cleared fat pads

At early stages of repopulation it appears that transplanted cells form aggregates before they grow out to build ductal structures. At this stage only a minority of cells expressed epithelial markers such as CK18, CK14 or EpCAM (data not shown). In some of these aggregates, however, the formation of ductal structures was already recognizable by a ring-like arrangement of SMA positive cells (Fig. 2A,B). Yet, the aggregates are surrounded by a thick layer of stromal cells, many of which show expression of CD271 (Figs 2A,B). Furthermore, the aggregates exhibit CD61 expression in distinct regions. CD61 is a marker for luminal precursor cells (Shackleton et al. 2006) and thus its presence in the aggregates indicate a certain degree of epithelial differentiation (Fig. 2C,D). This confirms our hypothesis that the cell aggregates represent a heterogeneous cell population from which the ductal outgrowths start during subsequent development.

image

Figure 2. Formation of heterogeneous cell aggregates prior to ductal outgrowth after 8 weeks of regeneration. Sections from cell aggregates in fat pads were reacted with anti-smooth muscle actin (SMA) and anti-CD271 antibodies and analyzed by confocal microscopy (A, B), and anti-CD61 (C, D) or anti-CD31 (E, F) antibodies analyzed by conventional microscopy. Rudimentary duct formation in cell aggregates derived from wild type (WT) (A) or EphB4 (B) cells visualized by SMA-expression surrounded by CD271-positive stroma. Dashed lines mark CD271hi stroma and arrows indicate ductal myoepithelial-like cells positive for SMA. Nuclei were counterstained with DAPI (4´6´-diamidino-2-phenylindole dihydrochloride). Scale bars represent 30 μm. Epithelial progenitors in heterogeneous wild type (WT) (C) and EphB4 (D) cell aggregates marked by CD61 expression. Arrows indicate regions where ductal structures are perceptible and arrow heads indicate distinct, smaller cell clusters with high expression of CD61. Scale bars represent 40 μm. Blood vessels and single CD31-positive cells in and around WT (E) and EphB4 (F) cell aggregates. Arrows indicate vessels surrounding and penetrating cell aggregates and arrow heads indicate CD31-positive cells in the heterogeneous cell mass. Scale bars represent 40 μm.

Download figure to PowerPoint

In order to survive, grow and differentiate, the injected cells are dependent on an adequate blood supply. After 8 weeks of regeneration, the cell aggregates of WT cells are surrounded by a sparse capillary network in direct contact with the heterogeneous cell mass as detected by immunohistochemical staining for CD31. In addition, single cells within the aggregates express CD31 (Fig. 2E). The circumferential capillary network is more pronounced around the aggregates of EphB4 derived cells (Fig. 2F) yet the frequency of internal CD31 expressing cells does not differ significantly at this stage (WT 17.03 ± 4.91%; EphB4 13.24 ± 3.57%, = 0.17).

Later in the regeneration process after 12 weeks, the ductal structures of the outgrowths become apparent. In the WT outgrowths, the epithelial cells formed a lumen and were surrounded by SMA positive myoepithelial cells. CD271 positive stromal cells became rare in the vicinity of the ducts (Fig. 3A). Ductal/alveolar differentiation into myoepithelial and luminal epithelial cells was also apparent in the outgrowth of EphB4 expressing cells, although lumen formation seemed to be delayed. Here, however, the parenchymal structures were still surrounded by a prominent layer of CD271 positive fibroblasts and SMA+/CD271+ double-positive myoepithelial cells were frequently observed (Fig. 3B). CD61 positive cells were still present in the outgrowths of either strain, although they seemed more prominent in the ducts of EphB4 transgenic cells and more abundant in the alveolar structures of WT outgrowths (Fig. 3C,D).

image

Figure 3. Ductal outgrowths and vascularization after 12 weeks of regeneration. Sections from ductal outgrowths were reacted with anti-smooth muscle actin (SMA) and anti-CD271 antibodies and analyzed by confocal microscopy (A, B), and anti-CD61 (C, D) or anti-CD31 (E, F) antibodies analyzed by conventional microscopy. Alveolar-like structures derived from wild type (WT) (A) or EphB4 (B) cells show a distinct myoepithelal layer marked by SMA-expression. EphB4 transgenic outgrowth were surrounded by CD271-positive stroma. Arrows indicate CD271-positive stromal cells, arrow heads indicate CD271-positive myoepithelial cells. Nuclei were counterstained with DAPI (4´6´-diamidino-2-phenylindole dihydrochloride). Scale bars represent 30 μm. CD61 expression in WT (C) and EphB4 (D) outgrowths. Scale bars represent 40 μm. Blood vessels and single CD31-positive cells in and around WT (E) and EphB4 (F) ducts. Arrows indicate vessels surrounding ducts and arrow head indicates CD31-positive cells within the duct. Scale bars represent 40 μm.

Download figure to PowerPoint

With regard to vascularization, the WT-derived outgrowths were surrounded by a sparse capillary network localized within or adjacent to the stromal cell layer surrounding the ducts, similar to the situation found in native virgin mammary glands. Intraepithelial CD31 positive cells were hardly found at this stage and showed merely weak expression of CD31 (Fig. 3E). In contrast, vascularization of EphB4 derived outgrowths was highly increased and capillaries were in intimate contact with or even found within the parenchyma (Fig. 3F). Moreover, intraepithelial cells strongly expressing CD31 were frequently present in the ducts (WT 3.03 ± 0.99%; EphB4 12.16 ± 4.57%, < 0.005). These results demonstrate that EphB4 expressing epithelial cells stimulate vascularization already at an early stage of regeneration, which is maintained during subsequent differentiation.

EphB4 overexpression leads to increased branching in repopulated mammary glands

Wholemount staining was used to look at the gross morphology of the outgrowths. In the WT we observed a ductal network as expected in a virgin state, with ducts dividing into two descending branches as it is typical for primary and secondary branching (Fig. 4A,C). Only rarely we were able to detect side-sprouting and alveoli formation in the WT outgrowths (Fig. 4C). In contrast, EphB4 derived outgrowths, although the parenchymal network was also established by primary and secondary branching, exhibited an extended ductal system with excessive side sprouting and the formation of numerous alveolar structures (Fig. 4B,D). Compared to the WT, hardly any ductal walls were discernible in EphB4 derived outgrowths as they were largely enclosed by side sprouts. Quantification of ductal branches and side sprouts revealed a significant (< 0.005) increase in branching and side-sprouting in the transgenic outgrowths (Fig. 4E).

image

Figure 4. Morphology of mammary outgrowths after 12 weeks of regeneration. Fat pads were removed after 12 weeks of regeneration and wholemounts were stained with carmine alum for morphological analysis. Fat pads of both wild type (WT) (A, C) and EphB4 transgenic outgrowths (B, D) exhibited complete repopulation with secondary branching (A, B) or ongoing fat pad invasion by terminal endbuds (TEBs) and primary branching (C, D). Filled arrows indicate primary or secondary branching points, arrow heads indicate TEBs with bifurcations and stars indicate side sprouts. Scale bars represent 350 μm. Branching and side-sprouting was quantified in wholemount stainings in random areas of four repopulated mammary glands of either strain (E). WT = 7, EphB4 = 14.

Download figure to PowerPoint

In order to characterize the outgrowth structures in more detail, we looked at the epithelial organization by analyzing EpCAM and SMA expression. Although EphB4 derived outgrowths exhibited a multilayered epithelium, SMA and EpCAM expression reveal an intact differentiation into an inner epithelial and outer myoepithelial cell layer, comparable to the WT outgrowths (Fig. 5A,B).

image

Figure 5. Compartmentalization of myoepithelial and epithelial cells in mammary epithelial outgrowths. Sections from wild type (WT) (A) and EphB4 transgenic (B) ductal outgrowths were reacted with anti-smooth muscle actin (SMA) and anti-EpCAM antibodies and analyzed by confocal microscopy. Nuclei were counterstained with DAPI (4´6´-diamidino-2-phenylindole dihydrochloride). SMA expression marks the myoepithelial cells (arrows) surrounding the EpCAM-positive epithelium (arrow head). Scale bars represent 30 μm.

Download figure to PowerPoint

Polarization represents a further indication of proper ductal organization of the luminal epithelial cells. Therefore, we have analyzed the expression of Par-3 and ZO-1, two proteins intimately involved with cell polarity (Ebnet et al. 2003; Suzuki & Ohno 2006). Par-3, as a crucial regulator molecule for the establishment of apical-basal polarity and mediator of adherence junctions, is mainly located on the apical epithelial border, but can also be found in myoepithelial cells. ZO-1, a tight junction protein, co-localizes with Par-3 on the apical border, setting up the epithelial barrier. In the WT we find this typical expression pattern both in smaller structures (Fig. 6A) as well as in the larger ductal system (Fig. 6C). EphB4 transgenic outgrowths also show a normal localization of both Par-3 and ZO-1 in smaller alveolar structures (Fig. 6B). In larger ducts with multilayered epithelium, however, there is no consistent apical border along the epithelial compartment and apparently cytoplasmic co-localization of Par-3 and ZO-1 was found in a subset of luminal epithelial cells (Fig. 6D). The localization and pattern of Par-3 in the myoepithelial cells appears normal, however, it is more prominent compared to the WT (Fig. 6D). These results indicate that luminal epithelial cells in EphB4 transgenic outgrowths were not able to undergo complete polarization. The failure in proper polarization is further illustrated by the expression of Muc-1 and CD133. CD133, an epithelial cell surface marker being involved in primary branching (Anderson et al. 2011) was weakly expressed on the apical membrane of WT outgrowths predominantly at sites of bifurcation (Fig. 6E). Interestingly, in the EphB4 derived outgrowths we found a CD133 expressing cell population in ducts lacking a normal established lumen, as well as single positive cells in regions that have a stromal morphology (Fig. 6F). Similarly, we could detect the Muc-1 protein, a protective glycoprotein localized on the apical membrane of differentiated epithelial cells (Spicer et al. 1995), only at the apical pole of the epithelial cells facing the lumen in the WT outgrowths (Fig. 6G), whereas the EphB4 transgenic outgrowths showed Muc-1 protein on the basal-lateral side of the ductal epithelial cells in addition to the apical localization (Fig. 6H). These defects in proper epithelial differentiation correlated with a preponderance of CK19 positive cells (71.79 ± 9.19% of total epithelial cells) and an increased rate of proliferation, especially in alveolar structures as indicated by staining for Ki67 (WT 7.61 ± 1.6%; EphB4 47.68 ± 19.66%, < 0.005). In WT-derived outgrowths, only a subset of the epithelial cells was positive for CK19 (43.37 ± 4.78%) and only a few of these CK19 positive cells were proliferating (Fig. 7A–D). These results indicate that in EphB4 transgenic outgrowths the luminal epithelial compartment is in a less differentiated state than in WT ductal outgrowths.

image

Figure 6. Polarization and differentiation in mammary outgrowths of wild type (WT) and EphB4 transgenic cells. Sections from ductal outgrowths were reacted with anti-Par-3 and anti-ZO-1 antibodies and analyzed by confocal microscopy (A–D), and anti-CD133 (E, F) or anti-MUC-1 (G, H) antibodies analyzed by conventional microscopy (E–H). Polarization was analyzed by staining for Par-3 and ZO-1 in WT (A, C) and EphB4 (B, D) alveolar-like structures (A, B) and ducts (B, D). Nuclei were counterstained with DAPI (4´6´-diamidino-2-phenylindole dihydrochloride). Arrows indicate co-localization of Par-3 and ZO-1 on the apical membranes, arrow heads indicate basal Par-3 localization. Scale bars represent 30 μm. CD133 expression in WT (E) and EphB4 transgenic (F) outgrowths. Arrows indicate the continuous luminal localization of CD133, arrow heads indicate single CD133-positive cells. Scale bars represent 40 μm. MUC-1 expression in WT (G) and EphB4 transgenic (H) outgrowths. Arrows indicate apical expression of MUC-1, arrow head indicates intraepithelial MUC-1 expression. Scale bars represent 40 μm.

Download figure to PowerPoint

image

Figure 7. CK19 and proliferation in mammary outgrowths from wild typ (WT) and EphB4 transgenic cells. Sections of WT (A, C) and EphB4 (B, D) ductal outgrowths were reacted against anti-CK19 and anti-Ki67 and analyzed by confocal microscopy in ductal (A, B) and alveolar (C, D) structures. Nuclei were counterstained with DAPI (4´6´-diamidino-2-phenylindole dihydrochloride). Arrows indicate proliferating CK19-positive cells. Scale bars represent 30 μm.

Download figure to PowerPoint

In summary, overexpression of EphB4 in mammary epithelial cells affects the differentiation hierarchy by favoring the luminal pathway and interferes with proper ductal differentiation, branching and vascularization. In order to characterize the signaling mechanisms responsible for the EphB4 induced phenotype we performed pathway-oriented gene expression arrays on basal, luminal, progenitor and stem cell-enriched cell populations isolated from virgin WT and EphB4 transgenic mammary glands. The cells were sorted according to the CD24 and CD49f marker expression mentioned above (Fig. 1) and gene expression was analyzed on a qPCR basis. To these ends we used custom arrays containing genes involved in cell movement, cell adhesion and EMT, as well as a commercially available array representative for the wnt signaling pathway.

EphB4 derived outgrowths show a direct effect on major wnt signaling mediators

As wnt signaling is intimately involved with mammary epithelial differentiation (Hatsell et al. 2003; Howe & Brown 2004; Gauger et al. 2012) and our EphB4 transgenic outgrowths were characterized by alterations concerning proliferation and differentiation, we used an array covering crucial genes involved at various levels in wnt signaling. These included the wnt molecules as well as receptors, co-receptors, regulators/inhibitors and transcription factors. A first observation is that of the four EphB4 cell populations the most discrepancies from the WT were found in the progenitor population (Fig. 8). Here, we found a strong upregulation in secreted frizzled-related protein 2 (sfrp2 5.8), transducing-like enhancer of split 2 (tle2 9.8) and wnt 5a (8.2) and minor upregulations in wnt 11 (3.4), dishevelled associated activator of morphogenesis 1 (daam1 2.29), F-box and WD-40 domain protein 4 (fbxw4 2.4), sentrin specific peptidase 2 (senp2 2.8) and porcupine homolog (porcn 4.7). Furthermore, we found significant downregulations in frat 1 (−4.7), wnt 5b (−3.3) as well as wnt 10a (−3.9) and close to significance cyclin D2 (ccnd2 −3.8) and transcription factor 7-like 1 (tcf7 l1 −4.9; Table 1). In contrast, in the basal cell population, only significant downregulation was detected concerning frizzled 2 (fzd2 −4.8), secreted frizzled related protein 1 (sfrp1 −4.4) and casein kinase 1 alpha 1 (csnk1a1 −2.4; Table 1). In the luminal cells frat 1 was −12.55-fold downregulated, whereas one member of the wnt family was down- and one upregulated (wnt2b (−2.5) and wnt11 (4.4), respectively; Table 1). In the stem cell-enriched cell population we hardly found any variations, F-box and WD-40 domain protein 4 (fbxw4) being the only significant change in gene expression (−2.3; Table 1).

Table 1. Changes in expression levels of genes involved in the wnt signaling pathway
Basal cellsLuminal cellsProgenitorsStem cells
GeneFold diff./P-valueGeneFold diff./P-valueGeneFold diff./P-valueGeneFold diff./P-value
  1. Fold diff. indicates the expression level in EphB4 transgenic epithelial cell populations in relation to their wild type (WT) counterparts.

Csnk1a1−2.4/0.04Frat1−12.5/0.003Frat1−4.7/0.03Fbxw4−2.34/0.04
Fzd2−4.8/0.002Wnt2b−2.5/0.05Wnt5b−3.3/0.007  
Sfrp1−4.4/0.05Fzd2−5.0/0.09Wnt10a−3.9/0.03  
    Ccnd2−3.8/0.06  
Fbxw43.1/0.07Wnt114.4/0.002Tcf7 l1−4.9/0.06  
Senp23.9/0.09      
    Daam12.3/0.007  
    Fbxw42.4/0.004  
    Senp22.8/0.002  
    Sfrp25.8/0.02  
    Tle29.8/0.01  
    Wnt5a8.2/0.008  
    Wnt113.4/0.0003  
    Porcn4.7/0.06  
image

Figure 8. Wnt signaling gene expression array with isolated epithelial cell populations. Volcano plots of wnt signaling pathway-oriented gene expression in EphB4 derived basal (A), luminal (B), progenitor (C) and stem cells (D) in relation to the corresponding wild type (WT) populations. Each analysis was done in triplets and values with < 0.05 (blue line) were regarded as significant, whereas up- or downregulation of threefold or more was taken into account (red lines). The identity of significantly deregulated genes is listed in Table 1.

Download figure to PowerPoint

The stem cell and bi-potent progenitor cells are disturbed in pathways regulating adhesion, migration and epithelial-stromal interaction

Due to the defects in polarity, epithelial junction and barrier formation in the EphB4 outgrowths we also had a look at numerous genes crucial for cell adhesion, ECM interaction, EMT, transforming growth factor (TGF)β- and bone morphogenetic protein (BMP) signaling. We only had few significant alternations in the basal and the luminal cells (Fig. 9). The basal cells revealed a significant and strong downregulation in insulin-like growth factor binding protein 3 (igfbp3 −8.5), vascular cell adhesion molecule 1 (vcam1 −8.4) and TGFβ-induced protein (tgfbi −3.1) as well as upregulation of growth differentiation factor 5 (gdf5 4.2; Table 2). In the luminal cells interleukin 1 receptor antagonist (il1rn −9) was clearly downregulated, whereas adamts8 (−2.7), integrin β2 (itgb2 −2.6), integrin αM (itgam −2.6) and transforming growth factor β2 (tgfb2 −3.1) exhibited more moderate changes. On the other hand, bone morphogenetic protein 1 was slightly upregulated (bmp1 2.2; Table 2). Interestingly, many of the genes analyzed were upregulated in the progenitor cells of the EphB4 mice. The highest upregulation concerned estrogen receptor α (esr1 13.1), followed by inhibin α (inha 6.6) and integrin αE (itgae 6. 7). Less distinct upregulations were found in bone morphogenetic protein 1 (bmp1 4.3), integrin α3 (itga3 2.1), integrin α5 (itga5 2.8), integrin β4 (itgb4 2.5) and transforming growth factor β1 (tgfb1 3. 7). In addition, downregulation of interleukin 1 receptor anatagonist (il1rn −4.1), adamts8 (−7.6), insulin-like growth factor 1 (igf1 −7.5) and integrin αX (itgax −2.3) was observed (Table 2). The most affected cell population, however, was the stem cell enriched fraction and remarkably, only significant upregulations were detected. A very strong upregulation was found for fibronectin 1 (fn1 58.2), tenascin C (tnc 35.8), bone morphogenetic protein 6 (bmp6 19.6) and transforming growth factor β induced (tgfbi 11). Several members of the integrin family were also upregulated (itga3, itga5, itgb2, itgb3, itgb4), as well as three members of the matrix metalloproteinase family (mmp2, mmp9, mmp14). The other genes that were upregulated included activin A receptor type 1 (acvr1 2.2), connective tissue growth factor (ctgf 3.1), regulator of G-protein signaling 2 (rgs2 6.1), MAD homolog 3 (smad3 2.1), transforming growth factor β receptor I and II (tgfbr1 2.5, tgfbr2 2.5) and tissue inhibitor of metalloproteinase 2 (timp2 3.6; Table 2).

Table 2. Changes in expression levels of genes involved in adhesion, migration and epithelial-to-mesenchymal transition (EMT)
Basal cellsLuminal cellsProgenitorsStem cells
GeneFold diff./P-valueGeneFold diff./P-valueGeneFold diff./P -valueGeneFold diff./P-value
  1. Fold diff. indicates the expression level in EphB4 transgenic epithelial cell populations in relation to their wild type (WT) counterparts.

Igfbp3−8.5/0.02Adamts8−2.7/0.01Adamts8−7.6/0.06Acvr12.2/0.03
Vcam1−8.4/0.005Il1rn−9.0/0.01Il1rn−4.1/0.02Bmp619.6/0.003
Tgfbi−3.1/0.06Itgb2−2.6/0.04Igf1−7.5/0.06Ctgf3.1/0.003
  ItgaM−2.6/0.05Itgax−2.3/0.05Fn158.2/0.0003
Gdf54.2/0.04Tgfb2−3.1/0.05  Itga32.8/0.01
    Bmp14.3/0.003Itga58.1/0.02
  Bmp12.17/0.004Esr113.1/0.02Itgb23.6/0.02
    Inha6.6/0.007Itgb33.2/0.02
    Itga32.1/0.05Itgb43.5/0.01
    Itga52.8/0.02Mmp23.3/0.03
    ItgaE6.7/0.02Mmp92.4/0.03
    Itgb42.5/0.003Mmp143.7/0.00002
    Tgfb13.7/0.02Rgs26.6/0.001
    Tgfbr22.0/0.005Smad32.1/0.002
      Tgfbi11.0/0.001
      Tgfbr12.5/0.02
      Tgfbr22.5/0.003
      Timp23.6/0.007
      Tnc35.8/0.007
image

Figure 9. Custom gene expression array specific for adhesion, migration, and extracellular matrix (ECM) interaction with isolated epithelial cell populations. Volcano plots of gene expression in EphB4 derived basal (A), luminal (B), progenitor (C) and stem cells (D) in relation to the corresponding wild type (WT) populations. Each analysis was done in triplets and values with < 0.05 (blue line) were regarded as significant, whereas up- or downregulation of threefold or more was taken into account (red lines). The identity of significantly deregulated genes is listed in Table 2.

Download figure to PowerPoint

As a confirmation of the results obtained from the PCR arrays we analyzed frizzled 2 expression at the protein level since this gene was downregulated in the two large cell populations, the basal and the luminal cell fractions of isolated EphB4 transgenic epithelial cells. Furthermore, as a major receptor for wnt growth factors it plays a central role in wnt signaling. Thus, we immunohistochemically analyzed Fzd2 expression in sections of WT and EphB4 transgenic outgrowths. Indeed, also on protein level Fzd2 was expressed at a lower rate in the EphB4 transgenic outgrowths (Fig. 10). Whereas Fzd2 could be detected at a high level in both larger ductal structures as well as in smaller alveolar structures in the WT (Fig. 10A,C), almost no expression was found in larger ducts of EphB4 transgenic outgrowths (Fig. 10B). Although Fzd2 expression could readily be detected at sites of branching and in alveoli it appears that Fzd2 there is also irregularly distributed and does not show the uniform patterning found in the WT outgrowths (Fig. 10D).

image

Figure 10. Fzd2 expression in mammary outgrowths. Sections from ductal outgrowths derived from wild type (WT) (A, C) and EphB4 transgenic (B, D) cells were reacted with anti-Fzd2 antibodies and analyzed by conventional microscopy. Fzd2 was expressed in alveolar-like structures (A, B) and ducts (B, D). Scale bars represent 40 μm.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Conflict of interest
  9. References
  10. Supporting Information

The understanding of the regulatory mechanisms orchestrating the continuous tissue dynamics in the parenchymal compartment appears as a crucial milestone for the understanding of the homeostasis of the mammary epithelium and thereby also for an improved clinical management of breast cancer. In this work we have concentrated on the effect of overexpression of EphB4 in the mammary epithelium as this receptor has been associated with endocrine estrogen signaling, paracrine communication as well as intercellular adhesion (Nikolova et al. 1998; Noren & Pasquale 2007; Heroult et al. 2010; Martin et al. 2010; Schmitt et al. 2013).

Altered EphB4 expression affects the mammary epithelial differentiation pathway

The FACS-based quantification of defined mammary epithelial sub-populations revealed that overexpression of EphB4 leads to a substantial increase of cells in the luminal epithelial cell compartment. Whereas WT mice exhibited an almost even ratio between luminal and basal cells, the transgenic epithelium contained a normal quantity of basal cells, while the luminal fraction was augmented. The cellular composition of the mammary parenchyma, especially the ratio of myoepithelial and epithelial cells, reflects the functional state of the parenchyma. During the development from a virgin to a pregnant mammary gland, the epithelial cell mass dramatically increases, whereas the myoepithelial cell number remains almost constant and one observes a shift in the ratio between these two cell types (Dulbecco et al. 1982). In our analyses, however, we have compared virgin animals and thus pregnancy-induced augmentation of luminal cells does not account for this result. In this context, it would be interesting to determine the portion of Esrα positive cells in the luminal fraction, since Schmitt et al. (2013) identified EphB4 as a positive regulator of Esrα expression in various breast cancer cell lines. Moreover, we have observed that the expression of Esrα is significantly increased in the bi-potent progenitor cell fraction. Interestingly, the bi-potent progenitor cell population was also significantly increased in the EphB4 transgenic epithelium, whereas the stem cell-enriched fraction remained unaffected. The similar content of mammary stem cells was also reflected by the repopulation potential, where ductal outgrowths derived from EphB4 transgenic and WT cells were observed at a similar frequency. Since we have previously shown, that the repopulation-competent cells can be assigned to the stem cell fraction only (Kaenel et al. 2012), we have renounced to repeat the repopulation assays with the different separated EphB4 transgenic cell fractions. FACS-based quantification and repopulation frequency revealed that EphB4 overexpression does not affect the stem cells themselves but has an instant effect on their closest descendants and favors the differentiation pathway towards the luminal cell fate. A similar increase in bi-potent and luminal cells has also been observed in transgenic mammary epithelium overexpressing ephrin-B2 (Kaenel et al. 2012). These cells, however, were capable of growing and differentiating normally, while proliferation and differentiation were severely affected by EphB4 overexpression. This supports our previous notion that ephrin-B2 reverse signaling favors differentiation, whereas over-stimulation of EphB4 forward signaling impedes it.

Ductal outgrowth is preceded by the formation of cellular aggregates

In our initial experiments, we performed the repopulation assays with a latency time of 8 weeks (Kaenel et al. 2012). After this period, however, the repopulation frequency in both strains was low and thus, we prolonged the regeneration time to 12 weeks. After 8 weeks both the WT control as well as the EphB4 transgenic cells mainly gave rise to cell aggregates, whereas complete ductal outgrowths were frequently observed after 12 weeks. This suggests that injected, differentiation-competent cells first form aggregates before growing out into a ductal network. The presence of CD61 positive cells within these aggregates indicates that indeed epithelial cells are present. As the aggregates represent a pre-stage of epithelial outgrowth it is conceivable that at this stage an expansion of the stem cells takes place, which give rise to the progenitor cells. Thus, the primary steps of stem cell regulation and differentiation can be expected in these cell aggregates. Additionally, this developmental point most likely features the recruitment of stromal cells including fibroblasts, endothelial and hematopoietic cells supporting epithelial survival, outgrowth and organization. This is exemplified by our finding, that SMA-, CD271 and CD31-positive cells were found within and adjacent to the aggregates. CD271 is described as a myoepithelial marker in human breast tissue and was used to assign a basal-like phenotype to breast tumor cells. Through its interaction with other receptors it can mediate cell survival or apoptosis (Popnikolov et al. 2005). In adults, CD271 is best described in the nervous system but is also found in various tissues at mesenchymal/epithelial boundaries. We could find high expression of CD271 mainly in the stroma surrounding the ducts growing out or surrounding cell aggregates. At this stage, CD271 might function as a regulator of cell survival in regions where the injected cells start to form a functional parenchyme by mediating mesenchymal–epithelial interactions. Interaction of the lymphatic and cardiovascular system and stromal cells with the mammary epithelium has already previously been shown to have a major impact on mammary morphogenesis (Shekhar et al. 2000; Ingman et al. 2006; Polyak & Kalluri 2010). Remodeling of the epithelial parenchyme occurs together with remodeling of lymphatic and blood vessels during pregnancy (Andres & Djonov 2010; Betterman et al. 2012). We interpret the transition of the cell aggregates to epithelial structures as a comparable process, during which blood vessels are crucial for the supply of not only nutrition but also hematopoietic cells (T-cells, macrophages), which help to remodel the tissue. Interestingly, EphB4 expression was observed to a similar extent in the transgenic and WT-derived aggregates. Thus, it is not astonishing, that at this stage the distribution of the markers analyzed does not differ significantly between the two experimental groups.

EphB4 overexpression interferes with differentiation and pattern formation

After 12 weeks of latency, ductal outgrowth could be readily observed in both experimental groups. In the case of EphB4 transgenic cells, expression of the receptor was strikingly stronger in ductal epithelial cells than in the preceding cell aggregates. As the EphB4 transgene expression lies under the control of the MMTV-LTR promoter this finding suggests that glucocorticoid receptor expression/response is only initiated after these early cell aggregates have begun to differentiate into ductal structures. This is a strong indication that the phenotype we observe in mammary glands repopulated by EphB4 cells is unlikely due to deregulated EphB4 expression in mammary stem cells but rather due to the effect of luminal cells on their niche and the stem and progenitor cells therein.

Concomitant with the onset of EphB4 transgene expression, alterations in the architecture of the outgrowths became evident. Most strikingly, the outgrowths of EphB4 transgenic cells exhibited an abnormally high number of side-branches, the formation of numerous periductal alveolar structures and the presence of a multi-layered ductal epithelium. The localization of SMA and EpCAM expressing cells, however, revealed that the compartmentalization of epithelial and myoepithelial cells was normal. Thus, the tissue compartmentalization does not seem to be the origin of enhanced branching. Interestingly, whereas ductal structures were multilayered and luminal epithelial cells did not reach end-differentiation, the alveolar structures were almost normal. This could reflect the hypothesis that alveolar cells develop from the basal cell population while ductal epithelium arises from the luminal precursors as it was proposed by Visvader & Smith (2011).

Yet, the multilayered epithelium led us to analyze the polarization of these cells. The co-localization of Par-3 and ZO-1 is an imperative reference for correct polarization (Ohno 2001). In smaller structures of EphB4 transgenic outgrowths with single-layered epithelia and throughout the WT cell-derived outgrowths we found perfect co-localization on the apical border of the luminal epithelium. In contrast, the multi-layered EphB4 outgrowths exhibited double-positive cells within the ducts and an apical localization was not discernible. It seems that the cells fail to arrange with a common polarity and consequently, also proper lumen formation failed. This is further highlighted by the expression pattern of Muc-1 and CD133. It is conceivable that the EphB4 transgenic outgrowths have defects in the remodeling program from their initial cell aggregates to a properly organized functional glandular tissue. This is supported by the persistent proliferation in apparently fully developed ducts and the abundant presence of cytokeratin 19 positive cells, which under normal condition should become scarce at this developmental stage.

In addition to the epithelium, we also found effects of EphB4 overexpression on non-epithelial cells. Most prominent was the significantly more pronounced periductal vascular network in the EphB4 transgenic outgrowths compared to WT cell-derived structures and even intraductal CD31 positive cells could be observed. This correlates with our findings in the EphB4 transgenic mammary glands where also a surplus of irregular capillary around the alveoli was observed (Andres & Djonov 2010) indicating that the epithelial-endothelial cross-talk was maintained in the outgrowths. Similarly, it has been shown that tumor cell derived EphB4 expression stimulates angiogenesis in breast cancer, thereby promoting tumor growth (Noren et al. 2004) and it is thought that non-endothelial EphB4 expression is involved in endothelial cell guidance and activation of filopodial extrusions of sensory endothelial cells (Zhang & Hughes 2006; Sawamiphak et al. 2010). Vice versa, upregulation of granulosa cell derived ephrin-B expression is thought to trigger angiogenesis in the corpus luteum (Egawa et al. 2003; Xu et al. 2006). In addition to the endothelium, also stromal cells seem to be affected by epithelial EphB4 expression, exemplified by CD271 staining. In ductal outgrowths after 12 weeks of regeneration, we found CD271 mainly expressed by single stromal cells surrounding the epithelium and in a minor subset of basal cells of the epithelium in the WT. On the other hand, in EphB4 outgrowths, especially the alveolar-like structures were in some parts completely surrounded by CD271-expressing stromal cells and also more basal cells expressed CD271. In breast cancer, CD271 is indicative for an aggressive tumor phenotype with basal cell-like activity as well as tumor-initiating and invasive capability (Kim et al. 2012). CD271 seems important for fat pad invasion in the mammary repopulation at early stages in the WT but its expression decreased when the ductal system was established after 12 weeks. In contrast, CD271 expression remained at a high level in and around the EphB4 outgrowths specifically concentrated at the numerous side-branches. It would be interesting to determine in what manner CD271 might regulate epithelial-mesenchymal interaction and whether it directly influences the protrusions of the epithelium into the surrounding fat pad. There might be a correlation with carcinogenesis in the sense, that tumor cells or in this case injected stem and/or progenitor cells might manipulate the adipocytes which in turn promote adipose tissue invasion (Nieman et al. 2013).

Members of wnt signaling and factors in adhesion, EMT and ECM-interaction are differentially expressed in EphB4 overexpressing epithelial cells

In order to gain information about the mechanisms by which EphB4 impedes differentiation we have compared the expression profile of the four defined cell fractions between EphB4 transgenic and WT mice. Since wnt signaling is involved in many aspects of mammary gland morphogenesis including branching and the control of cell fate (Jarde & Dale 2012) and based on the described cross-talk between wnt and EphB4 (Batlle et al. 2002), we have concentrated on the expression of members of the wnt signaling pathway. These experiments revealed significant discrepancies predominantly in the progenitor fraction. These differences included members of the wnt growth factor family involved in both canonical (wnt2b, 10a) and non-canonical wnt signaling (wnt5a, 5b, 11; Siar et al. 2012), indicating that both pathways are affected by aberrant EphB4 expression. Some of the deregulated wnt members have no assigned function in mammary gland development and it would be interesting to analyze whether they simply amplify/adopt functions of known wnt molecules or whether they initiate a completely new cell response. In addition, key regulators/mediators of wnt such as the sFRPs, Fzd2, Frat1 and Senp2 are deregulated in more than one cell population. Representatively for others, we could confirm the alterations found in gene expression on the protein levels of Fzd2 in mammary gland sections. These results suggest that EphB4 exerts its effects at least in part through altering wnt signaling, which may result in the observed stimulation of branching activity and differentiation failure. Interestingly, a similarly massive side-sprouting after reconstitution of a cleared mammary fat pad has also been found for Sfrp1−/− mammary epithelial cells (Gauger et al. 2012). As sFRPs act as universal inhibitors of wnt signaling we assume that indeed wnt signaling in general is a key pathway regulating branching morphogenesis. In addition to wnt, effects on branching morphogenesis was also described for other signaling pathways, such as BMP, integrin/mitogen-activated protein kinase (MAPK) and fibroblast growth factor (FGF) signaling (Bradbury et al. 1995; Parsa et al. 2008; Gauger et al. 2012; Forsman et al. 2013; Mori et al. 2013; Pond et al. 2013), some of which are known to directly interact with Eph/ephrin signaling. Thus, it would be interesting to investigate if and in which hierarchical order EphB4 expression also activates these signaling cascades.

Members of the Eph family have been shown to induce EMT via wnt signaling in gastric cancer cells (Huang et al. 2013). As we had major defects in the outgrowth morphology caused by or leading to disturbed cellular organization and polarity, we also included genes involved in EMT, migration/adhesion and ECM interaction in our gene expression profiling. Similar to wnt signaling, only minor alterations were found in the basal and luminal cell fractions. Interestingly, downregulation of il1rn and adamts8 was found in the luminal cell population as well as in the progenitor population. IL1RN is expressed in a specific form by epithelial cells where it is involved in the regulation of an autocrine IL1-mediated differentiation pathway (Haskill et al. 1991; Arend et al. 1998). As we found that deregulated EphB4 has a stronger impact on luminal cells than basal cells and even stronger on progenitors it is conceivable that the deregulated expression of EphB4 represses the expression of il1rn in luminal precursors thereby impeding their differentiation pathway.

A disintegrin and metalloproteinases (ADAMs) are important mediators for growth factor release induced by stromal-epithelial interaction. In the mammary gland, ADAM17-dependent processing of amphiregulin in the EGF signaling pathway has been shown (Sternlicht et al. 2005) and Solanas et al. (2011) could correlate EphB3 expression with ADAM10-mediated E-cadherin release in the intestine. Less information is available about the role of the ADAMTS family in epithelial cell signaling; however, adamts8 was found to be downregulated in response to TGFβ in alveolar lung cells (Keating et al. 2006). Functionally, ADAMTS8 is best known as a potent inhibitor of angiogenesis (Vazquez et al. 1999) and therefore, its downregulation might be a molecular effector of the increased angiogenesis we found in the EphB4 transgenic outgrowths. Thus, the luminal cell fraction might be a decisive and key regulator for angiogenic stimulation. This would be interesting to follow up, as we found only minor other alterations of gene expression in the luminal fraction.

On the other hand, we had a dramatic increase in the expression of numerous genes in the progenitor- and stem cell fractions including several integrins, BMP- and TGFβ signaling components, MMPs and their inhibitor, as well as the ECM components fibronectin 1 and tenascin C. All of these genes are involved in the regulation of cell migration and thus might contribute to the enhanced side-sprouting by preparing the surrounding stroma and fatty tissue for epithelial invasion. This also correlates nicely with our previous finding in a double transgenic tumor model where we could show that EphB4 overexpression confers an invasive and metastasizing phenotype on NeuT-induced mammary tumors, which was accompanied by induction of tenascin expression and a preponderance of progenitor cells (Munarini et al. 2002; Scherberich et al. 2005; Kaenel et al. 2011). We speculate that after repopulation the epithelial cells impeded in their differentiation acquire a more invasive phenotype similar to the observation that the introduction of MMTV-wnt1 cancer stem cells into the intact mammary gland leads to the activation of an invasive remodeling program (Parashurama et al. 2012). Moreover, since the stem cell fraction was the most affected, deregulated EphB4 expression may also facilitate the migration of stem cells out of their niches as it was observed in the colon (Batlle et al. 2002).

In summary, we have shown that overexpression of EphB4 in mammary epithelial cells leads to a shift in the differentiation pathway and promotes an invasive remodeling program for mammary outgrowths featuring enhanced side-sprouting. Two possible explanations arise for this phenotype. Firstly, the onset of EphB4 transgene expression in differentiating epithelial cells interferes with their own differentiation program by imposing a motile phenotype. Alternatively, MMTV-controlled EphB4 activation in differentiating epithelial cells influences the stem/progenitor cells, which then differentiate abnormally. Since the gene expression profiling revealed the most extensive changes in the progenitor and stem cell fraction, this latter mechanism seems to be the more probable. In addition, the observed alterations in the peri-ductal vasculature and stroma may equally contribute to the observed phenotype. Although we cannot discriminate between the impacts of the different cell–cell communication options, it seems that the wnt signaling pathway plays a central role in the interpretation of the EphB4 induced signaling. Taken together, these results demonstrate that EphB4-ephrin-B2 signaling is indispensable for the mammary stem cell homeostasis and correct cell fate decisions. Moreover, comparable to their roles in the intestinal system (Batlle et al. 2002), EphB4 and ephrin-B2 orchestrate (progenitor-) cell migration and tissue pattern formation and thus play a central role in the establishment of a functional glandular architecture.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Conflict of interest
  9. References
  10. Supporting Information

This study was supported by the Swiss National Science Foundation (3100A0-127168; 3100A_143853/1), the Swiss Cancer League (KLS-2825-08-2011) and the Schweizerische Stiftung für Klinisch-Experimentelle Tumorforschung.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Conflict of interest
  9. References
  10. Supporting Information
  • Adams, R. H. 2002. Vascular patterning by Eph receptor tyrosine kinases and ephrins. Semin. Cell Dev. Biol. 13, 5560.
  • Anderson, L. H., Boulanger, C. A., Smith, G. H., Carmeliet, P. & Watson, C. J. 2011. Stem cell marker prominin-1 regulates branching morphogenesis, but not regenerative capacity, in the mammary gland. Dev. Dyn. 240, 674681.
  • Andres, A. C. & Djonov, V. 2010. The mammary gland vasculature revisited. J. Mammary Gland Biol. Neoplasia 15, 319328.
  • Arend, W. P., Malyak, M., Guthridge, C. J. & Gabay, C. 1998. Interleukin-1 receptor antagonist: role in biology. Annu. Rev. Immunol. 16, 2755.
  • Batlle, E., Henderson, J. T., Beghtel, H., van den Born, M.M., Sancho, E., Huls, G., Meeldijk, J., Robertson, J., van de Wetering, M., Pawson, T. & Clevers, H.2002. Beta-catenin and TCF mediate cell positioning in the intestinal epithelium by controlling the expression of EphB/ephrinB. Cell 111, 251263.
  • Betterman, K. L., Paquet-Fifield, S., Asselin-Labat, M. L., Visvader, J.E., Butler, L.M., Stacker, S.A., Achen, M.G. & Harvey, N.L. 2012. Remodeling of the lymphatic vasculature during mouse mammary gland morphogenesis is mediated via epithelial-derived lymphangiogenic stimuli. Am. J. Pathol. 181, 22252238.
  • Bradbury, J. M., Edwards, P. A., Niemeyer, C. C. & Dale, T. C. 1995. Wnt-4 expression induces a pregnancy-like growth pattern in reconstituted mammary glands in virgin mice. Dev. Biol. 170, 553563.
  • Brisken, C. & Duss, S. 2007. Stem cells and the stem cell niche in the breast: an integrated hormonal and developmental perspective. Stem Cell Rev. 3, 147156.
  • Bush, J. O. & Soriano, P. 2012. Eph/ephrin signaling: genetic, phosphoproteomic, and transcriptomic approaches. Semin. Cell Dev. Biol. 23, 2634.
  • Cheng, N., Brantley, D. M. & Chen, J. 2002. The ephrins and Eph receptors in angiogenesis. Cytokine Growth Factor Rev. 13, 7585.
  • Cohn, E., Ossowski, L., Bertran, S., Marzan, C. & Farias, E. F. 2010. RARalpha1 control of mammary gland ductal morphogenesis and wnt1-tumorigenesis. Breast Cancer Res. 12, R79.
  • DeOme, K. B., Faulkin, L. J. Jr, Bern, H. A. & Blair, P. B. 1959. Development of mammary tumors from hyperplastic alveolar nodules transplanted into gland-free mammary fat pads of female C3H mice. Cancer Res. 19, 515520.
  • Dulbecco, R., Henahan, M. & Armstrong, B. 1982. Cell types and morphogenesis in the mammary gland. Proc. Natl Acad. Sci. USA 79, 73467350.
  • Ebnet, K., Aurrand-Lions, M., Kuhn, A., Kiefer, F., Butz, S., Zander, K., Meyer zu Brickwedde, M.K., Suzuki, A., Imhof, B.A. & Vestweber, D. 2003. The junctional adhesion molecule (JAM) family members JAM-2 and JAM-3 associate with the cell polarity protein PAR-3: a possible role for JAMs in endothelial cell polarity. J. Cell Sci. 116, 38793891.
  • Egawa, M., Yoshioka, S., Higuchi, T., Sato, Y., Tatsumi, K., Fujiwara, H. & Fujii, S. 2003. Ephrin B1 is expressed on human luteinizing granulosa cells in corpora lutea of the early luteal phase: the possible involvement of the B class Eph-ephrin system during corpus luteum formation. J. Clin. Endocrinol. Metab. 88, 43844392.
  • Fata, J. E., Werb, Z. & Bissell, M. J. 2004. Regulation of mammary gland branching morphogenesis by the extracellular matrix and its remodeling enzymes. Breast Cancer Res. 6, 111.
  • Forsman, C. L., Ng, B. C., Heinze, R. K., Kuo, C., Sergi, C., Gopalakrishnan, R., Yee, D., Graf, D., Schwertfeger, K.L. & Petryk, A. 2013. BMP-binding protein twisted gastrulation is required in mammary gland epithelium for normal ductal elongation and myoepithelial compartmentalization. Dev. Biol. 373, 95106.
  • Gastaldi, S., Sassi, F., Accornero, P., Torti, D., Galimi, F., Migliardi, G., Molyneux, G., Perera, T., Comoglio, P.M., Boccaccio, C., Smalley, M.J., Bertotti, A. & Trusolino, L. 2013. Met signaling regulates growth, repopulating potential and basal cell-fate commitment of mammary luminal progenitors: implications for basal-like breast cancer. Oncogene 32, 14281440.
  • Gauger, K. J., Shimono, A., Crisi, G. M. & Schneider, S. S. 2012. Loss of SFRP1 promotes ductal branching in the murine mammary gland. BMC Dev. Biol. 12, 25.
  • Haldimann, M., Custer, D., Munarini, N., Stirnimann, C., Zurcher, G., Rohrbach, V., Djonov, V., Ziemiecki, A. & Andres, A.C. 2009. Deregulated ephrin-B2 expression in the mammary gland interferes with the development of both the glandular epithelium and vasculature and promotes metastasis formation. Int. J. Oncol. 35, 525536.
  • Harrison, H., Simoes, B. M., Rogerson, L., Howell, S. J., Landberg, G. & Clarke, R. B. 2013. Oestrogen increases the activity of oestrogen receptor negative breast cancer stem cells through paracrine EGFR and Notch signalling. Breast Cancer Res. 15, R21.
  • Haskill, S., Martin, G., Van Le, L., Morris, J., Peace, A., Bigler, C.F., Jaffe, G.J., Hammerberg, C., Sporn, S.A. & Fong, S. 1991. cDNA cloning of an intracellular form of the human interleukin 1 receptor antagonist associated with epithelium. Proc. Natl Acad. Sci. USA 88, 36813685.
  • Hatsell, S., Rowlands, T., Hiremath, M. & Cowin, P. 2003. Beta-catenin and Tcfs in mammary development and cancer. J. Mammary Gland Biol. Neoplasia 8, 145158.
  • Heroult, M., Schaffner, F., Pfaff, D., Prahst, C., Kirmse, R., Kutschera, S., Riedel, M., Ludwig, T., Vajkoczy, P., Graeser, R. & Augustin, H.G. 2010. EphB4 promotes site-specific metastatic tumor cell dissemination by interacting with endothelial cell-expressed ephrinB2. Mol. Cancer Res. 8, 12971309.
  • Howe, L. R. & Brown, A. M. 2004. Wnt signaling and breast cancer. Cancer Biol. Ther. 3, 3641.
  • Huang, J., Xiao, D., Li, G., Ma, J., Chen, P., Yuan, W., Hou, F., Ge, J., Zhong, M., Tang, Y., Xia, X. & Chen, Z. 2013. EphA2 promotes epithelial-mesenchymal transition through the Wnt/beta-catenin pathway in gastric cancer cells. Oncogene doi: 10.1038/onc.2013.238. [Epub ahead of print].
  • Huynh-Do, U., Stein, E., Lane, A. A., Liu, H., Cerretti, D. P. & Daniel, T. O. 1999. Surface densities of ephrin-B1 determine EphB1-coupled activation of cell attachment through alphavbeta3 and alpha5beta1 integrins. EMBO J. 18, 21652173.
  • Huynh-Do, U., Vindis, C., Liu, H., Cerretti, D.P., McGrew, J.T., Enriquez, M., Chen, J. & Daniel, T.O. 2002. Ephrin-B1 transduces signals to activate integrin-mediated migration, attachment and angiogenesis. J. Cell Sci. 115, 30733081.
  • Ingman, W. V., Wyckoff, J., Gouon-Evans, V., Condeelis, J. & Pollard, J. W. 2006. Macrophages promote collagen fibrillogenesis around terminal end buds of the developing mammary gland. Dev. Dyn. 235, 32223229.
  • Jarde, T. & Dale, T. 2012. Wnt signalling in murine postnatal mammary gland development. Acta Physiol. 204, 118127.
  • Kaenel, P., Antonijevic, M., Richter, S., Kuchler, S., Sutter, N., Wotzkow, C., Strange, R. & Andres, A.C. 2012. Deregulated ephrin-B2 signaling in mammary epithelial cells alters the stem cell compartment and interferes with the epithelial differentiation pathway. Int. J. Oncol. 40, 357369.
  • Kaenel, P., Schwab, C., Mulchi, K., Wotzkow, C. & Andres, A. C. 2011. Preponderance of cells with stem cell characteristics in metastasising mouse mammary tumours induced by deregulated EphB4 and ephrin-B2 expression. Int. J. Oncol. 38, 151160.
  • Keating, D. T., Sadlier, D. M., Patricelli, A., Smith, S.M., Walls, D., Egan, J.J. & Doran, P.P. 2006. Microarray identifies ADAM family members as key responders to TGF-beta1 in alveolar epithelial cells. Respir. Res. 7, 114.
  • Kim, J., Villadsen, R., Sorlie, T., Fogh, L., Gronlund, S.Z., Fridriksdottir, A.J., Kuhn, I., Rank, F., Wielenga, V.T., Solvang, H., Edwards, P.A., Borresen-Dale, A.L., Ronnov-Jessen, L., Bissell, M.J. & Petersen, O.W. 2012. Tumor initiating but differentiated luminal-like breast cancer cells are highly invasive in the absence of basal-like activity. Proc. Natl Acad. Sci. USA 109, 61246129.
  • Kobayashi, S., Sugiura, H., Ando, Y., Shiraki, N., Yanagi, T., Yamashita, H. & Toyama, T. 2012. Reproductive history and breast cancer risk. Breast Cancer 19, 302308.
  • Koshikawa, N., Giannelli, G., Cirulli, V., Miyazaki, K. & Quaranta, V. 2000. Role of cell surface metalloprotease MT1-MMP in epithelial cell migration over laminin-5. J. Cell Biol. 148, 615624.
  • LaMarca, H. L., Visbal, A. P., Creighton, C. J., Liu, H., Zhang, Y., Behbod, F. & Rosen, J.M., 2010. CCAAT/enhancer binding protein beta regulates stem cell activity and specifies luminal cell fate in the mammary gland. Stem Cells 28, 535544.
  • Lo, A. T., Mori, H., Mott, J. & Bissell, M. J. 2012. Constructing three-dimensional models to study mammary gland branching morphogenesis and functional differentiation. J. Mammary Gland Biol. 17, 103110.
  • Luetteke, N. C., Qiu, T. H., Fenton, S. E., Troyer, K.L., Riedel, R.F., Chang, A. & Lee, D.C. 1999. Targeted inactivation of the EGF and amphiregulin genes reveals distinct roles for EGF receptor ligands in mouse mammary gland development. Development 126, 27392750.
  • Mallepell, S., Krust, A., Chambon, P. & Brisken, C. 2006. Paracrine signaling through the epithelial estrogen receptor alpha is required for proliferation and morphogenesis in the mammary gland. Proc. Natl. Acad. Sci. USA 103, 21962201.
  • Martin, T. J., Allan, E. H., Ho, P. W., Gooi, J.H., Quinn, J.M., Gillespie, M.T., Krasnoperov, V. & Sims, N.A. 2010. Communication between ephrinB2 and EphB4 within the osteoblast lineage. Adv. Exp. Med. Biol. 658, 5160.
  • Miao, H., Strebhardt, K., Pasquale, E. B., Shen, T. L., Guan, J. L. & Wang, B. 2005. Inhibition of integrin-mediated cell adhesion but not directional cell migration requires catalytic activity of EphB3 receptor tyrosine kinase. Role of Rho family small GTPases. J. Biol. Chem. 280, 923932.
  • Mori, H., Lo, A. T., Inman, J. L., Alcaraz, J., Ghajar, C.M., Mott, J.D., Nelson, C.M., Chen, C.S., Zhang, H., Bascom, J.L., Seiki, M. & Bissell, M.J. 2013. Transmembrane/cytoplasmic, rather than catalytic, domains of Mmp14 signal to MAPK activation and mammary branching morphogenesis via binding to integrin beta1. Development 140, 343352.
  • Munarini, N., Jager, R., Abderhalden, S., Zuercher, G., Rohrbach, V., Loercher, S., Pfanner-Meyer, B., Andres, A.C. & Ziemiecki, A. 2002. Altered mammary epithelial development, pattern formation and involution in transgenic mice expressing the EphB4 receptor tyrosine kinase. J. Cell Sci. 115, 2537.
  • Nguyen-Ngoc, K. V. & Ewald, A. J. 2013. Mammary ductal elongation and myoepithelial migration are regulated by the composition of the extracellular matrix. J. Microsc. 251, 21223.
  • Nieman, K. M., Romero, I. L., Van Houten, B. & Lengyel, E. 2013. Adipose tissue and adipocytes support tumorigenesis and metastasis. Biochim. Biophys. Acta 1831, 153341.
  • Nikolova, Z., Djonov, V., Zuercher, G., Andres, A. C. & Ziemiecki, A. 1998. Cell-type specific and estrogen dependent expression of the receptor tyrosine kinase EphB4 and its ligand ephrin-B2 during mammary gland morphogenesis. J. Cell Sci. 111(Pt 18), 27412751.
  • Noren, N. K., Lu, M., Freeman, A. L., Koolpe, M. & Pasquale, E. B. 2004. Interplay between EphB4 on tumor cells and vascular ephrin-B2 regulates tumor growth. Proc. Natl Acad. Sci. USA 101, 55835588.
  • Noren, N. K. & Pasquale, E. B. 2007. Paradoxes of the EphB4 receptor in cancer. Cancer Res. 67, 39943997.
  • Ohno, S. 2001. Intercellular junctions and cellular polarity: the PAR-aPKC complex, a conserved core cassette playing fundamental roles in cell polarity. Curr. Opin. Cell Biol. 13, 641648.
  • Parashurama, N., Lobo, N. A., Ito, K., Mosley, A.R., Habte, F.G., Zabala, M., Smith, B.R., Lam, J., Weissman, I.L., Clarke, M.F. & Gambhir, S.S 2012. Remodeling of endogenous mammary epithelium by breast cancer stem cells. Stem Cells 30, 21142127.
  • Parsa, S., Ramasamy, S. K., De Langhe, S., Gupte, V.V., Haigh, J.J., Medina, D. & Bellusci, S. 2008. Terminal end bud maintenance in mammary gland is dependent upon FGFR2b signaling. Dev. Biol. 317, 121131.
  • Pasquale, E. B. 2010. Eph receptors and ephrins in cancer: bidirectional signalling and beyond. Nat. Rev. Cancer 10, 165180.
  • Polyak, K. & Kalluri, R. 2010. The role of the microenvironment in mammary gland development and cancer. Cold Spring Harb. Perspect. Biol. 2, a003244.
  • Pond, A. C., Bin, X., Batts, T., Roarty, K., Hilsenbeck, S. & Rosen, J. M. 2013. Fibroblast growth factor receptor signaling is essential for normal mammary gland development and stem cell function. Stem Cells 31, 178189.
  • Popnikolov, N. K., Cavone, S. M., Schultz, P. M. & Garcia, F. U. 2005. Diagnostic utility of p75 neurotrophin receptor (p75NTR) as a marker of breast myoepithelial cells. Mod. Pathol. 18, 15351541.
  • Sawamiphak, S., Seidel, S., Essmann, C. L., Wilkinson, G.A., Pitulescu, M.E., Acker, T. & Acker-Palmer, A. 2010. Ephrin-B2 regulates VEGFR2 function in developmental and tumour angiogenesis. Nature 465, 487491.
  • Scherberich, A., Tucker, R. P., Degen, M., Brown-Luedi, M., Andres, A. C. & Chiquet-Ehrismann, R. 2005. Tenascin-W is found in malignant mammary tumors, promotes alpha8 integrin-dependent motility and requires p38MAPK activity for BMP-2 and TNF-alpha induced expression in vitro. Oncogene 24, 15251532.
  • Schmitt, F., Nguyen, P. H., Gupta, N. & Mayer, D. 2013. Eph receptor B4 is a regulator of estrogen receptor alpha in breast cancer cells. J. Recept. Signal Transduct. Res. 33, 2448.
  • Shackleton, M., Vaillant, F., Simpson, K. J., Stingl, J., Smyth, G.K., Asselin-Labat, M.L., Wu, L., Lindeman, G.J. & Visvader, J.E. 2006. Generation of a functional mammary gland from a single stem cell. Nature 439, 8488.
  • Shekhar, M. P., Werdell, J. & Tait, L. 2000. Interaction with endothelial cells is a prerequisite for branching ductal-alveolar morphogenesis and hyperplasia of preneoplastic human breast epithelial cells: regulation by estrogen. Cancer Res. 60, 439449.
  • Siar, C. H., Nagatsuka, H., Han, P. P., Buery, R.R., Tsujigiwa, H., Nakano, K., Ng, K.H. & Kawakami, T. 2012. Differential expression of canonical and non-canonical Wnt ligands in ameloblastoma. J. Oral Pathol. Med. 41, 332339.
  • Smalley, M. J., Kendrick, H., Sheridan, J. M., Regan, J.L., Prater, M.D., Lindeman, G.J., Watson, C.J., Visvader, J.E. & Stingl, J. 2012. Isolation of mouse mammary epithelial subpopulations: a comparison of leading methods. J. Mammary Gland Biol. Neoplasia 17, 9197.
  • Solanas, G., Cortina, C., Sevillano, M. & Batlle, E. 2011. Cleavage of E-cadherin by ADAM10 mediates epithelial cell sorting downstream of EphB signalling. Nat. Cell Biol. 13, 11001107.
  • Spicer, A. P., Rowse, G. J., Lidner, T. K. & Gendler, S. J. 1995. Delayed mammary tumor progression in Muc-1 null mice. J. Biol. Chem. 270, 3009330101.
  • Stahl, S., Weitzman, S. & Jones, J. C. 1997. The role of laminin-5 and its receptors in mammary epithelial cell branching morphogenesis. J. Cell Sci. 110(Pt 1), 5563.
  • Sternlicht, M. D. 2006. Key stages in mammary gland development: the cues that regulate ductal branching morphogenesis. Breast Cancer Res. 8, 201.
  • Sternlicht, M. D., Sunnarborg, S. W., Kouros-Mehr, H., Yu, Y., Lee, D. C. & Werb, Z. 2005. Mammary ductal morphogenesis requires paracrine activation of stromal EGFR via ADAM17-dependent shedding of epithelial amphiregulin. Development 132, 39233933.
  • Stingl, J. 2009. Detection and analysis of mammary gland stem cells. J. Pathol. 217, 229241.
  • Stingl, J., Eirew, P., Ricketson, I., Shackleton, M., Vaillant, F., Choi, D., Li, H.I. & Eaves, C.J. 2006. Purification and unique properties of mammary epithelial stem cells. Nature 439, 993997.
  • Suzuki, A. & Ohno, S. 2006. The PAR-aPKC system: lessons in polarity. J. Cell Sci. 119, 979987.
  • Vafaizadeh, V., Klemmt, P., Brendel, C., Weber, K., Doebele, C., Britt, K., Grez, M., Fehse, B., Desrivieres, S. & Groner, B. 2010. Mammary epithelial reconstitution with gene-modified stem cells assigns roles to Stat5 in luminal alveolar cell fate decisions, differentiation, involution, and mammary tumor formation. Stem Cells 28, 928938.
  • Vazquez, F., Hastings, G., Ortega, M. A., Lane, T.F., Oikemus, S., Lombardo, M. & Iruela-Arispe, M.L. 1999. METH-1, a human ortholog of ADAMTS-1, and METH-2 are members of a new family of proteins with angio-inhibitory activity. J. Biol. Chem. 274, 2334923357.
  • Visvader, J. E. & Smith, G. H. 2011. Murine mammary epithelial stem cells: discovery, function, and current status. Cold Spring Harb. Perspect. Biol. 3.
  • Wiseman, B. S. & Werb, Z. 2002. Stromal effects on mammary gland development and breast cancer. Science 296, 10461049.
  • Xu, Y., Zagoura, D., Keck, C. & Pietrowski, D. 2006. Expression of Eph receptor tyrosine kinases and their ligands in human Granulosa lutein cells and human umbilical vein endothelial cells. Exp Clin Endocrinol Diabetes 114, 590595.
  • Zhang, J. & Hughes, S. 2006. Role of the ephrin and Eph receptor tyrosine kinase families in angiogenesis and development of the cardiovascular system. J. Pathol. 208, 453461.
  • Zhao, C., Irie, N., Takada, Y., Shimoda, K., Miyamoto, T., Nishiwaki, T., Suda, T. & Matsuo, K2006. Bidirectional ephrinB2-EphB4 signaling controls bone homeostasis. Cell Metab. 4, 111121.

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Conflict of interest
  9. References
  10. Supporting Information
FilenameFormatSizeDescription
dgd12126-sup-0001-FigS1-S2-TablesS1-S2.pdfapplication/PDF509K

Fig. S1. Schematic overview of the experimental strategy.

Fig. S2. Quantification of epithelial cell populations in mammary glands of wild type and EphB4 transgenic mice by flow cytometry.

Table S1. List of genes investigated in WNT Signaling Pathway PCR Array (SABiosciences).

Table S2. List of genes investigated in a customized PCR Array including genes involved in ECM interaction, adhesion/migration and EMT (SABiosciences).

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.