E XP E RI M ENTA L CE L L R E S EA RC H

3 20 (2014) 2 33 –246

Available online at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/yexcr

Research Article

Role of the EphB2 receptor in autophagy, apoptosis and invasion in human breast cancer cells Sahiti Chukkapallia,1, Mohamed Amessoua,1, Ashok K. Dillya,1, Hafedh Dekhilb, Jing Zhaoc, Qiang Liuc, Alex Bejnaa, Ron D. Thomasa, Sudeshna Bandyopadhyaya, Tarek A. Bismard, Daniel Neilla, Laurent Azoulaye, Gerald Batistc,e, Mustapha Kandouza,f,n a

Department of Pathology, Wayne State University School of Medicine, Detroit, MI, USA Obesity Research Center, College of Medicine, King Saud University, Kingdom of Saudi Arabia c Montréal Centre for Experimental Therapeutics in Cancer, Segal cancer Centre, Lady Davis Institute for Medical Research, Jewish General Hospital, Montréal, Québec, Canada d Departments of Pathology & Laboratory Medicine, Oncology, Biochemistry & Molecular Biology, University of Calgary, Calgary, Canada e Department of Oncology, McGill University, Montréal, Québec, Canada f Karmanos Cancer Institute, Wayne State University, Detroit, MI, USA b

article information

abstract

Article Chronology:

The Eph and Ephrin proteins, which constitute the largest family of receptor tyrosine kinases, are

Received 24 May 2013

involved in normal tissue development and cancer progression. Here, we examined the

Received in revised form

expression and role of the B-type Eph receptor EphB2 in breast cancers. By immunohistochem-

28 October 2013

istry using a progression tissue microarray of human clinical samples, we found EphB2 to be

Accepted 29 October 2013

expressed in benign tissues, but strongly increased in cancers particularly in invasive and

Available online 6 November 2013

metastatic carcinomas. Subsequently, we found evidence that EphB2, whose expression varies in

Keywords: EphB2 Ephrin Breast cancer Autophagy Apoptosis Invasion

established cell breast lines, possesses multiple functions. First, the use of a DOX-inducible system to restore EphB2 function to low expressers resulted in decreased tumor growth in vitro and in vivo, while its siRNA-mediated silencing in high expressers increased growth. This function involves the onset of apoptotic death paralleled by caspases 3 and 9 activation. Second, EphB2 was also found to induce autophagy, as assessed by immunofluorescence and/or immunoblotting examination of the LC3, ATG5 and ATG12 markers. Third, EphB2 also has a pro-invasive function in breast cancer cells that involves the regulation of MMP2 and MMP9 metalloproteases and can be blocked by treatment with respective neutralizing antibodies. Furthermore, EphB2-induced invasion is kinase-dependent and is impeded in cells expressing a kinase-dead mutant EphB2. In summary, we identified a mechanism involving a triple role for EphB2 in breast cancer progression, whereby it regulates apoptosis, autophagy, and invasion. & 2013 Elsevier Inc. All rights reserved.

Abbreviations: EphB2, B-type Eph receptor 2; MMP, matrix metalloproteinase; RTK, receptor tyrosine kinases; TMA, tissue microarray; DOX, doxycycline n

Corresponding author at: Department of Pathology, Wayne State University School of Medicine, 5101 Cass Ave, Chemistry Bldg, Rm 425, Detroit, MI 48202, USA. E-mail address: [email protected] (M. Kandouz). 1 These authors contributed equally. 0014-4827/$ - see front matter & 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.yexcr.2013.10.022

234

EX P ER I ME NTAL C E LL RE S E ARCH

Introduction The Eph receptor tyrosine kinases (RTK) and their membranebound ligands, the Ephrins mediate direct cell–cell communication and are thus involved in important biological processes including neuro-navigation during the development of the central nervous system, angiogenesis and formation of spatial boundaries and tissue morphogenesis during embryogenesis (for reviews see [1,10,19–22,31,35,43,50,58,60,70,74]). Ephs and Ephrins have been classified into two subfamilies, A-type (EphA) and B-type (EphB) receptors, and A-type (EphrinA) and B-type (EphrinB) ligands, based on their sequence similarity and receptor–ligand binding affinities. EphA receptors bind almost exclusively to EphrinAs, which are tethered to the cell surface by a glycosylphosphatidylinositol (GPI)-anchor. EphB receptors bind almost exclusively to EphrinBs which are bound to the membranes by a single transmembrane domain [23]. However, there is some crosstalk between A and B types. For example, the EphB2 receptor can bind EphrinB ligands as well as EphrinA5 [29]. B-type Eph receptors and Ephrin ligands have a transmembrane domain and an intracellular domain, which undergoes tyrosine phosphorylation and binds signaling proteins with a PDZ domain. An important feature of the Eph/Ephrin family is that, following receptor-ligand interactions, it engages in a bi-directional signaling event initiated both from the receptor (forward signaling) and the ligand (reverse signaling). This family organization and task distribution makes it complicated to comprehend the functions of individual receptors or ligands or their various interactive pairs. In fact, the same Eph or Ephrin could demonstrate multiple and sometimes opposing functions, especially in the context of their role in cancer [39]. A growing amount of data suggests that Ephs and Ephrins could have either positive or negative roles in tumorigenesis [14]. For example, loss of expression of EphB2 was observed in colon cancers [47], where it has been suggested as a potential tumor suppressor [4]. Similarly, it seems to function as a prostate tumor suppressor and somatic inactivating mutations were found in
10% of sporadic prostate tumors [36]. Also, a K1019X mutation was associated with increased risk for prostate cancer in men with a positive family history [42]. These data clearly suggest a role for EphB2 in cancer. There is also less direct data. For example, deletions at chr. 1p36 region, where EphB2 is localized, were very frequently found in breast cancers [11,62], but the specific role of EphB2 in these cancers is not known. In the present work we assessed EphB2 expression in different models of breast cell lines and human clinical tissues and we identified an important role for this receptor in regulating three different homeostatic processes: autophagy, apoptosis and invasion. This finding provides important elements in the understanding of the role of EphB2 in breast cancer.

Materials and methods Immunoblotting Immunoblotting was performed using a standard protocol and a monoclonal antibody against EphB2 (Acris Antibodies Inc.). Other antibodies used in imunoblotting were obtained from Cell

3 20 (2014 ) 233 – 246

Signaling Inc. (phospho-ERK, ERK, ATG5, LC3, and Active Casp9), Epitomics (Active Casp3), or Millipore (Actin). A polyclonal antibody against EphB2 was a generous gift from Dr. Tony Pawson (Samuel Lunenfeld Research Institute, Toronto).

Tissue microarray and immunohistochemistry We analyzed the expression levels of the EphB2 receptor in a panel of human benign and cancer breast specimens using a commercially available TMA (Breast Cancer Progression Tissue Array BR2082, Biomax.us), including 206 cases: 32 metastatic, 68 invasive ductal, 22 each of lobular and intraductal carcinomas, four each of squamous cell and lobular carcinoma in situ, eight fibroadenoma, 16 each of hyperplasia and inflammation, 10 adjacent normal tissue and six normal tissue. IHC staining and scoring of this array was performed, using a monoclonal antiEphB2 from Acris Antibodies Inc. at a 1/100 dilution, and analyzed by the study pathologist (S.B). The antibodies were titrated to breast samples exclusively. Protein expression was measured semi-quantitatively (0: absent, 1þ: weak, 2þ: moderate and 3þ: strong expression). The results were illustrated as the frequency (%) of specimens with different staining levels (Fig. 1b). In addition to determining the staining intensity, the % of tumor cells within each and every sample, that showed strong positive staining was recorded and used to generate a quantitative score (Table 2) by the study statistician (LA), according to a modified histological score (H-Score) method [2,49]. Here, the intensity scores reflected the percentage of cells with the maximal intensity, and thus ranged between 0 and 300 (a score of 3 in 100% of cells). Since data were not normally distributed, instead of means we reported medians, along with the standard deviation. As a consequence, we reported the interquartile range, which is the value at the first quartile (Q1) to the third quartile (Q3), Q2 being by definition the median. Therefore, the Q1 to Q3 range provides an idea of what the values are for 75% of the data. Each cancer progression category was compared to the “Non-Cancer” category used as a reference, using Mann-Whitney U tests.

Cell culture and cell transfection All various cell lines were grown as recommended by ATCC. The NHMEC-E6/E7 cells were derived by immortalization from normal human mammary epithelial cells (NHMECs), originally obtained from Clonetics, and were a kind gift by Dr. Al-Moustafa (McGill University).

Colony formation assay Cells plated onto 6 cm-plates (7  105 cells/well) were allowed to attach overnight in FBS-supplemented RPMI. T47DhEphB2-p2 and T47DDsR-p1 cells were obtained by transfection of T47D cells with a pDsRed-based vector expressing the human EphB2 cDNA and the empty vector respectively (generous gift of Dr. Spyro Mousses, Translational Genomics Research Institute, Cancer Drug Development Laboratory, Gaithersburg, MD, USA). After 2–3 weeks in culture under selection with G418 at 600 mg/mL, Crystal Violet (Sigma) was added to cell cultures and rinsed away with PBS. Colonies were counted in representative squares. Experiments were performed in triplicates, and data were presented as mean7SD.

E XP E RI ME N TAL CE L L R ES E ARC H

32 0 (2 014 ) 2 33 –24 6

235

Fig. 1 – Immunohistochemical staining of EphB2 in human breast clinical specimens. The TMA was immunostained with a monoclonal antibody against EphB2 (Acris Antibodies Inc.). (a) Representative microphotographs include samples with low (left panels) and high (right panels) levels, of normal samples (N), Intraductal carcinomas (ID), invasive ductal carcinomas (InvD), metastases (Met) and Invasive lobular carcinomas (InvL). Both low (  40) and high (  200, inset) magnifications are shown in all photographs. (b) A stacked column chart summarizing the IHC results and indicating the frequency of specimens with EphB2 staining intensity scores of 0 to 3þ. Note that the “normal” or non-cancer category includes normal adjacent to tumor, hyperplasia, mastitis, Fibroadenoma, and normal tissue.

TetON inducible expression system The generation of MDA435 cells that inducibly express the EphB2 protein was performed using the Tet-On Advanced Inducible Gene Expression System and the Retro-X Universal Packaging System (Clontech). The human EphB2 ORF was amplified from a constitutive expression construct (Origene) and cloned in the BamHI and EcoRI sites in the retroviral pRetroX-Tight-Pur vector (Clontech), producing the pRetro–EphB2 construct. GP2-293 packaging cells were transfected with either the pTet-On-Advanced vector or the pRetro–EphB2 vector and the viral envelope-producing pVSV-G construct. To collect viral particles, the cell culture medium was collected 2–4 days after transfection, centrifuged and filtered.

A first round of infection was performed with the regulator plasmid pTet-On-Advanced. G418-resistant clones were selected and screened by transient transfections with the pTRE-Tight-Luc for clones with low background and high DOX-dependent induction, using a luciferase assay (Promega). A clone of MDA435 cells was chosen and used for a second infection with the pRetro–EphB2 construct. Puromycin-resistant cells were selected and assayed for DOX-dependent induction of EphB2. Because there has been some concern regarding the identity of the MDA435 cells, based on expression profiling [15,61,64], and which appears to be attributed to a phenomenon called “lineage infidelity” found in other breast cancer cell lines as well [9,32,33,51,66], we have reproduced our main experiments in another breast cancer cell line, SKBR3 cells.

236

EX P ER I ME NTAL C E LL RE S E ARCH

MTT growth assay Cells were counted and plated with or without DOX treatment. They were washed with PBS and the MTS/PMS solution, diluted in RPMI1640 phenol red free medium (100 μl of medium/10 μl of MTS/PMS solution) was added to the cells, incubated for 2 h at 371 and the absorbance was read in microplate reader at 490 nm. The medium was changed every alternate day. The kit used is the Cell titer 96 aqueous non-radioactive cell proliferation assay (promega).

Real time RT-PCR Total RNA was isolated from breast cell lines using the RNAquous4 PCR Kit (Ambion) or the Nucleospin RNA II kit (Macherey Nagel) according to the manufacturers' instructions. The cDNA was prepared using the High Capacity cDNA reverse trascription kit (Applied Biosystems). Expression levels were examined using the RT SYBR green ROX qPCR mastermix (Qiagen or SAB biosciences) as directed by the manufacturer. The list of primers used is provided in Table 1. The primers for Birc5 were obtained from realtimeprimers.com.

Flow cytometry analysis MDA435tetEphB2 cells were treated for 5 days with or without DOX (1 μg/mL) without changing the medium. Total cells (Adherent and floating) were collected and washed with PBS. The cells were counted and resuspended in the Annexin V Binding Buffer (ABB). The quantification of apoptosis was performed by Fluorescence Activated Cell Sorting (FACS) using the Dead Cell Apoptosis Kit with Annexin V Alexa Fluor 488 and Propidium Iodide (Kit #V13241, Life Technologies), according to the manufacturer's recommended protocol. Flow cytometry work was done at the Microscopy, Imaging and Cytometry Resources (MICR) Core at Wayne State University.

Autophagy microscopic monitoring Two approaches were used to monitor autophagy by immunofluorescence. First, MDA435tetEphB2 cells were grown on glass slides and induced or not with DOX (5 days, 1 μg/mL), to express EphB2. Immunofluorescence staining was performed using an Table 1 – Sequences of the primers used in real time RTPCR. Genes

Primer sequences

MMP2

S: TTGACGGTAAGGACGGACTC AS: ACTTGCAGTACTCCCCATCG S: CTCTGGAGGTTCGACGTG AS: GTCCACCTGGTTCAACTCAC S: TCCGCATCAGGAAGGCTAGA AS: AGGACCAGGCCTCCAAGCT S: GCGGAAGAGGTGGATGTAC AS: GCCTTGAAAGTCCCAGATGG S: ATCAAGTTCCAGGAGTTCAGC AS: TGATCTTCATGGTGCGTGTG S: GACGTCCAGAACTAGAAGCTG AS: CAATCCCTGCAAATAAGGCC

MMP9 BCL2 EPHB2 EFNB1 EFNB2

3 20 (2014 ) 233 – 246

anti-LC3B antibody (LC3B (D11) XPs Rabbit mAb #3868, Cell Signaling Inc.). In a second set of experiments, cells were first transfected with a plasmid encoding a yellow fluorescence protein (YFP)-tagged LC3 (a kind gift from Dr. Tamotsu Yoshimori, National Institute of Genetics, Japan), induced with DOX (5 days, 1 μg/mL) and grown on glass slides. Different microscopy fields were photographed and the percentage of autophagic cells, defined as cells with strong punctate fluorescence signal, was determined from efficiently transfected fluorescently labeled cells. Imaging was performed with a 40  objective using an EVOSs FL Cell Imaging System (AMG). Image overlay was done with Adobe Photoshop (Adobe Systems Inc.).

Gene silencing EphB2 silencing in MDA-MB-231 cells was done using a pGIPZshRNA sequence (RHS4430-98514044 from OpenBiosystems) or an siRNA sequence (Qiagen), with the respective nonsilencing sequences used as controls. Briefly, MDA-MB-231 cells were plated 16 h before transfection to produce monolayers that were 60% confluent and these were transfected with 30 nM of either control siRNA (#1027280, Qiagen) or EphB2 siRNA (#SI02224789, Qiagen) using the Lipofectamine 2000 reagent according to the manufacturer's instructions (Invitrogen). Transfection efficiency was monitored by measuring the level of EphB2 mRNA using real time qRT-PCR or immunoblotting analysis.

Modified Boyden chamber cell invasion assay The invasion assay was performed on MDA435tetEphB2, SKBR3tetEphB2 or MDA-MB-231 cells after 3 days of EphB2 induction or silencing respectively. Cell invasion assays were done in modified Boyden chambers fitted with Matrigel-coated 8-mm transwell filters (250 μg/mL Matrigel, BD Biosciences Cat #356231) as described [69]. Cells were seeded in the top of transwells of the 24-well plate at a density of (2  105 cells)/(0.5 mL of 1% BSA in RPMI medium). Fetal bovine serum at 10% was used as a chemoattractant in the bottom chambers. For the MMP blocking experiments, 4 μg of neutralizing Anti-MMP-9 or Anti-MMP-2 antibodies or control IgGs were added to the upper chamber in serum-free medium. In other experiments, MDA435 cells were transfected with a control empty vector, plasmids encoding a wild type (WT) EphB2 or a kinase dead (KD) mutant EphB2 (Generous gift from Dr. Tony Pawson, Samuel Lunenfeld Research Institute, Toronto, Canada) for 72 h. After 18 h, the cells on the upper side of the membrane were removed and cells penetrating the filter were stained with a Cell Stain Solution (Chemicon) and imaged by bright field microscopy. Relative cell attachment was determined by counting the number of cells in at least three microscopy fields.

In vivo tumorigenesis assay 4  106 MDA435tetEphB2 cells were injected subcutaneously into athymic nude mice. Tumors were allowed to develop for 14 days, where tumors were visible, and EphB2 expression was induced in vivo by 2 mg/mL Doxycycline (Clontech) in 5% sucrose in drinking water. Tumor growth was measured and the tumor volume (cubic centimeter) was calculated as length x width2/2.

E XP E RI ME N TAL CE L L R ES E ARC H

Statistical analysis

32 0 (2 014 ) 2 33 –24 6

237

Expression of EphB2 elicits growth-inhibitory effects in breast cancer cells

Data shown as means7SD were analyzed using the Student's t test (*Po0.05, **Po0.01, ***Po0.001). P values less than 0.05 were considered to be statistically significant.

Results Dynamic expression of EphB2 in breast cancer specimens To determine the expression levels of EphB2 in breast tissues, we assessed it by immunohistochemistry (IHC) using a breast tumor progression tissue microarray (TMA). Representative staining is illustrated in Fig. 1a. Protein expression was measured semiquantitatively using a 4-tiered system (0¼ negative, 1þ¼ weak, 2þ ¼moderate and 3þ¼ strong expression). We first established an overall staining distribution of EphB2 in various categories of breast cancer specimens. We found that in reference to the noncancer tissues, the malignant specimens were significantly more positive for EphB2 expression. This is evidenced by the higher % of cases with medium (2þ) or high (3þ) staining in all groups of cancer specimens (Fig. 1b). Next, and because this score distribution does not take into account the extent and heterogeneity of staining within the specimens, we used a quantitative intensity scoring method to reflect the % of cells with the maximal intensity (3þ) within each specimen. The calculated staining scores of 80 for intraductal carcinomas, 35 for invasive ductal carcinomas to 90 for metastatic specimens (Table 2), in comparison with the normal specimens' category, considered as a reference group, show that the cancer specimens have stronger EphB2 expression levels than the normal tissues. Only in invasive lobular specimens does this scoring method not show a similar change (Tables 2, P ¼0.0072). This could be explained by the fact that this category does not include any sample with 3þ staining level (Fig. 1b), and suggest also that there might be differences between ductal and lobular tumors with regard to EphB2 expression. Taken together, these IHC results show that overall EphB2 staining is increased in cancer tissues and suggest different possible roles at different cancer progression stages, i.e. early in breast tumorigenesis and later at invasion and metastases.

We first surveyed several established human mammary tumor cell lines to identify the most suitable cellular models for our studies. We examined the expression pattern of the EphB2 receptor in a panel of 12 mammary cancer and benign cell lines by immunoblotting. Both two immortalized non tumorigenic cell lines, MCF10A and NHMEC-E6/E7, and a group of four cancer cell lines showed high EphB2 expression, with the levels in cancer cells appearing to be stronger (Fig. 2a). Extracts from all cell lines were collected at approximately 80% confluence. However, since EphB2 is a cell–cell communication protein, we examined whether its low expression in many cell lines is sensitive to cell density; in fact, it is known that some proteins are destabilized by losing cell–cell contacts when they are grown at low confluence. Therefore, we wanted to rule out that the inability of some breast cancer cell lines to express EphB2 was not due to the level of confluence at which the protein extracts were collected. To this end, cells were grown at densities ranging from sparse (25% confluence), or medium (50%) to dense (100%). While EphB2 expression was found indeed to be cell density-dependent in the EphB2expressing benign MCF-10A and cancer (Hs578T, MDA-MB-231) cells, it did not significantly change in the EphB2-devoid (T47D, ZR-75-1 and MCF-7) cells (Fig. 2b), where EphB2 silencing seemed more persistent, suggesting that the low EphB2 levels in these cells are not related to cell density and cell–cell interactions, but rather to other modes of gene regulation.

Table 2 – Comparison of median scores of different cancer progression categories for EphB2 expression. Samples (number)

R median (Q1–Q3), P-value

Non-cancer (N¼ 56)a Intraductal (N ¼22) Invasive ductal (N¼ 68) Invasive lobular (N¼ 22) Metastasis (N¼ 32)b

0 (0–0), reference 80 (20–170), o0.0001 35 (0–120), o0.0001 0 (0–10), 0.0072 90 (30–190), o0.0001

Scores were calculated by multiplying the intensity score (3) by the % cells with the maximal intensity, the maximal value being 300 (3  100%). a Includes normal adjacent to tumor, hyperplasia, mastitis, Fibroadenoma, and normal tissue. b All specimens are ductal carcinoma metastases.

Fig. 2 – Expression of EphB2 in human breast cell lines. (a) Immunoblots were performed on total protein extracts from different cancer (Hs578T, BT-20, BT-549, T47D, MDA-435, MCF7, ZR-75-1, MDA-MB-468, MDA-MB-231 and SKBR3) and benign (MCF-10A and NHMEC-E6/E7) cells grown at high confluence. (b) The effect of cell density on EphB2 expression in human breast cell lines was analyzed by immunoblotting. Cells were seeded to reach different confluences (25%, 50% and 100%) the next day for extraction. Total protein lysates were analyzed by immunoblotting using anti-EphB2 (Rabbit polyclonal). An anti-β-Actin antibody was used as a loading control.

238

EX P ER I ME NTAL C E LL RE S E ARCH

In conclusion, among the different cell lines, we elected to use T47D, SKBR3 and MDA435 cells, which showed very low levels of EphB2 and MDA-MB-231 as EphB2 expressing cells, for subsequent in vitro and in vivo experiments. EphB2 has been reported as a potential tumor suppressor gene in the colon [4] and the prostate [36]. However, we have no evidence about the role of this gene in breast cancer. To examine the effect of EphB2 overexpression on breast cancer cell growth, we performed a colony formation assay by transfecting T47D cells with an EphB2-encoding or the control DsR vector (Supplemental Fig. S1a). EphB2 expression resulted in decreased T47D colony formation (Supplemental Fig. S1b). The rounded up morphology of the EphB2-expressing cells (Supplemental Fig. S1c) is reminiscent of a cell death phenotype, as we previously observed in 293T cells [40]. These results are indicative of a growth inhibitory effect of EphB2 in breast cancer cells. Indeed, after antibiotic resistance selection, very small loosely attached colonies were obtained in the EphB2transfected cells and no viable clone of cells stably expressing EphB2 could be maintained for more than few generations. Therefore, we developed a doxycycline (DOX)-inducible system using MDA435 and SKBR3 cells, here referred to as MDA435tetEphB2 and SKBR3tetEphB2 cells respectively, to express EphB2 on demand. The induction of EphB2 resulted in an increase in phosphorylation of ERK1/2 (Fig. 3a), a downstream EphB2 signaling target [40], indicating the functionality of the gene. In order to examine the effect of restoring EphB2 expression using the DOX-inducible system, we performed an MTT assay over a period of 5 days, and found that EphB2 induces growth inhibition (Fig. 3b). To rule out the possibility that any observed effects might be an artifact due to protein overexpression, we performed an experiment where different DOX doses were used to fine-tune EphB2 expression (Fig. 3c). We found that even at a dose as low as 30 ng/mL which resulted in very low induction of EphB2, the growth-inhibitory effects were significant (Fig. 3d). Therefore, the inducible system faithfully models EphB2negative and -positive cells, thus validating the observed growthinhibitory effects. We next asked whether EphB2 could inhibit breast tumor growth in vivo, using the DOX-inducible system. MDA435tet-EphB2 cells were inoculated subcutaneously to nude mice and tumor growth was measured over a period of 80 days with or without DOX treatment. Tumors were left to build up until day 14 post-inoculation and then the mice groups were randomized and a group was given DOX in the drinking water. Induction of EphB2 expression resulted in a drastic decrease in tumor growth in comparison with non-induced tumors (Fig. 3e). This result provides strong evidence that sustained EphB2 expression inhibits tumor growth in vivo in mammary cells. In conclusion, we have shown that EphB2 possesses growth inhibitory functions in breast cancer cells.

EphB2 expression induces apoptotic cell death in breast cancer cells We next sought to determine the mechanisms involved in the growth-inhibitory effects of EphB2. MDA435TetEphB2 cells were treated for 5 days with or without DOX (1 μg/mL) and total cells were collected and analyzed by flow cytometry. The quantification of apoptosis by Annexin V and propidium iodide staining showed a significant 4-fold increase in the fraction of dead cells upon induction of EphB2 expression (Fig. 4a). The involvement of apoptosis is further illustrated by an increased level of active caspases 3 and 9, assessed

3 20 (2014 ) 233 – 246

by immunoblotting (Fig. 4b) and the decreased expression of antiapoptotic genes Bcl2 and Birc5 (Survivin), quantified by real time RTPCR (Fig. 4c). These results indicate that the growth-inhibitory effects of EphB2 involve an apoptotic cell death.

EphB2 expression induces autophagy in breast cancer cells Autophagy is increasingly recognized as a critical factor in the modulation of tumor suppression functions of many proteins and in various cancers, including in breast cancer [13,76]. Although autophagy is viewed primarily as a pro-survival mechanism, its sustainment might result in secondary apoptotic cell death. We have previously demonstrated that EphB2 induces autophagy in 293T cells [40]. Therefore, we tested whether the growth inhibitory effect of EphB2 expression in mammary cells is associated with its proautophagic function. A commonly and reliably used marker of autophagy is LC3 that is converted from the LC3-I form to a lipidated and cleaved LC3-II form, prone to accumulate and localize at autophagosomal vesicles. We first used immunofluorescent staining with an anti-LC3 antibody in MDA435tetEphB2 cells grown on glass slides and induced or not with DOX (5 days, 1 μg/mL), to express EphB2. Immunofluorescence staining showed an increase in the LC3 signal upon EphB2 expression (Fig. 4d). To obtain quantitative data, in a second set of experiments, we transfected the cells with a plasmid encoding a yellow fluorescence protein (YFP)-tagged LC3, prior to induction with DOX (5 days, 1 μg/mL) and growth on glass slides. Different microscopy fields were photographed and the percentage of autophagic cells, defined as cells with strong punctate fluorescence signal, was determined out of the efficiently transfected fluorescently labeled cells. Using this method, we found that EphB2 expression significantly increased the fraction of cells with LC3-YFP-labeled vesicles (Fig. 4e). As a further indication that autophagy occurs, EphB2 expression was indeed found to induce accumulation of ATG5, ATG12 and the LC3II form of the LC3 protein, which are important markers of autophagy (Fig. 4f). These data show that in addition to inducing apoptosis, EphB2 induces autophagy in breast cancer cells. Using another cell line, SKBR3 cells, we have confirmed the pro-apoptotic and pro-autophagic functions of EphB2 (Supplemental Fig. S2a). In summary, our data suggest that although EphB2 induces autophagy, its sustained expression also inhibits the growth of breast cancer cells by inducing apoptosis.

EphB2 has pro-invasive functions in breast cancer cells In view of the results of our IHC analysis showing higher EphB2 expression in invasive and metastatic specimens, and since it has become clear that Ephs and Ephrins can demonstrate multiple and sometimes conflicting functions in tumor progression [39], we assessed the role of EphB2 in invasion. Using the DOXinducible system and the Boyden chamber-like invasion assay, we found that EphB2 overexpression significantly induced the invasive potential of the MDA435tetEphB2 (Fig. 5a) and SKBR3tetEphB2 cells (Supplemental Fig. S2b). To further determine the mechanism of action, we analyzed the effect of EphB2 on the expression of invasion-regulatory genes. We observed that the expression of the pro-invasive genes matrix metalloproteases MMP2 and MMP9 was strongly increased (Fig. 5b). Moreover, the use of MMP2 or MMP9 neutralizing antibodies was able to counteract EphB2-induced invasiveness of MDA435tetEphB2 cells,

E XP E RI ME N TAL CE L L R ES E ARC H

32 0 (2 014 ) 2 33 –24 6

239

Fig. 3 – Dose and time course analysis of growth inhibitory effects of EphB2 expression. (a) The inducible system enables the DOXinduced expression of EphB2 in MDA435tetEphB2 cells, at different times (shown induction at 3 or 5 days). The receptor activation is demonstrated by the resulting ERK activation examined by immunoblotting of ERK1/2 and phospho-ERK1/2. The GAPDH level is shown as a protein loading control. (b) In parallel, growth inhibition was assessed by MTT assay (1–5 days, 1 μg/mL). (c) The system allows the fine tuning of EphB2 expression levels, as shown by a dose response to DOX treatment in MDA435tetEphB2 cells. Cells were treated with an increasing concentration of DOX and EphB2 levels were analyzed by immunoblotting. (d) In parallel, growth inhibition was assessed by MTT assay of MDA435tetEphB2 cells treated with increasing doses of DOX. (e) Tumor growth of MDA435tetEphB2 cells inducibly expressing EphB2 injected subcutaneously into nude mice. DOX administration was started at day 14 after cell injection. Tumor growth was monitored in untreated mice ( DOX) or mice treated with DOX (þDOX). (***Po0.001).

while the control IgG antibodies were ineffective, indicating a role for MMP2 and MMP9 in this invasive process (Fig. 5c). We asked whether the kinase activity is essential for the proinvasive function of EphB2. Using a kinase-dead mutant (KD), along with the wild type (WT) EphB2 and the control vector (3.1), transfected into wild type MDA435 cells, we found that EphB2mediated cell invasion is dramatically impaired (Fig. 6a and b) in the KD mutant which has no ability to undergo auto-phosphorylation, as verified by immunoprecipitation and immunoblotting of EphB2 (Fig. 6c). Therefore, we conclude that the role of EphB2 in MDA435 cell invasion is kinase-dependent.

EphB2 knockdown inhibits its functions in breast cancer cells In a complementary set of experiments, we sought to further confirm the various functions of EphB2 in mammary cells, using a gene silencing approach. We used siRNAs to silence EphB2 expression in MDA-MB-231 cells, which express high levels of this receptor (Fig. 2a). A colony formation assay showed that lower EphB2 levels resulted in increased growth of MDA-MB-231 cells, in comparison with the non-silenced control counterparts (Fig. 7a). Another MTT growth assay performed over a period of 3

240

EX P ER I ME NTAL C E LL RE S E ARCH

days showed significant growth induction as early as 1 day after siRNA silencing (Fig. 7b). The siRNA sequences were verified to efficiently knockdown the expression of EphB2 with a silencing rate of over 90% (Fig. 7c and d). To assess the effect of EphB2 knockdown

3 20 (2014 ) 233 – 246

on the activation of the autophagic and apoptotic pathways, we used real time RT-PCR and immunoblotting to analyze the expression of a panel of genes involved in these two processes. We found that expression of the anti-apoptotic genes Bcl2 and Birc5 (Survivin) was

E XP E RI ME N TAL CE L L R ES E ARC H

increased (Fig. 7c), while the expression of pro-autophagic genes ATG5 and ATG12 was decreased after EphB2 silencing (Fig. 7d). The level of the LC3II form was only slightly decreased by EphB2 knockdown (Fig. 7d). This is expected in this reverse approach where autophagy has not been triggered, unlike what we observed when autophagy has been actively induced by EphB2 expression (Fig. 4d). These results support our conclusions that EphB2 concomitantly regulates pro-autophagic and pro-apoptotic functions in mammary cells. Furthermore, when EphB2 expression was decreased by siRNA knock down in MDA-MB-231 cells, the latter's invasiveness was reduced by over 40% (Fig. 8a). This effect translated into a decrease in levels of the pro-invasive genes MMP2 and MMP9 (Fig. 8b). Taken together, these data demonstrate that, while EphB2 controls autophagy and apoptosis, it also possesses pro-invasive functions in mammary cancer cells.

Discussion In this work, we characterized the expression and functions of EphB2 in breast cancer progression. First, we demonstrated very clearly that EphB2 expression suppresses the growth of human breast cancer cells both in vitro and in vivo. Moreover, at the center of this function we identified a regulatory mechanism that involves both autophagy and apoptosis. Importantly, we also found that EphB2 promotes cell invasiveness. These findings raise important dilemmas. At first examination, the functional data related to tumor growth suppression are somewhat unorthodox in view of the IHC data. While increased EphB2 in invasive or metastatic specimens makes sense with regard to the demonstrated pro-invasive function of EphB2, the growth-inhibitory function appears counter-intuitive in view of the increased EphB2 levels in the non-invasive in situ specimens. It is thus possible that, although we have uncovered different EphB2 functions using in vitro and in vivo models, the actual coordination of these functions and the dominance of one function over the others might depend on other clinical parameters. In other words, increases in EphB2 levels might result in distinct functions depending on the cellular context, rather than being of an overall fixed and linear pattern throughout tumor progression. For example, it might be that in the DCIS tumors, which express EphB2, the pro-autophagic function is dominant, thus favoring a pro-survival and prooncogenic mechanism. As an illustrative antecedent model, it was previously shown that autophagy is required for the survival of abnormal precursor cells that pre-exist in DCIS [16]. BNIP3, a gene

32 0 (2 014 ) 2 33 –24 6

241

involved in autophagy, cell survival and death is overexpressed in DCIS and invasive breast carcinomas. While its expression is associated with an increased risk of recurrence and shorter disease-free survival in DCIS, it is also associated with good survival outcome in invasive carcinomas, which led to the suggestion that BNIP3 undergoes a functional switch from a cell death to survival function during the transition from pre-invasive to invasive breast cancer [67]. An equally challenging question is the observation that EphB2 shows a pro-apoptotic function, which is a feature of tumor suppressor proteins, while it appears to be overexpressed in cancer tissues. As we have shown previously, EphB2-induced cell death can be significantly reduced by inhibiting autophagy [40], suggesting that the pro-death function of EphB2 is mediated largely through the autophagic process. It is therefore plausible that in the context of mammary cells, autophagy serves as a gateway for apoptotic cell death (Hypothetical model in Fig. 9). This would be consistent with the proposed dual functions of autophagy as both a pro-survival and pro-death mechanism [3,12,13,41,44,73] and with the described interplay between autophagy and apoptosis [18,63,65]. Understanding cross-talk between the signaling pathways and regulatory mechanisms involved in both autophagy and apoptosis has a critical impact both in the basic study of cancer and on its translational aspect. Autophagy and apoptosis share many of their regulators and effectors such as ATG5 and BCL2 [48], which we found here to be downstream targets of EphB2. These processes have been shown to either cooperate or oppose each other, depending on the cellular context and thus should not be considered as mutually exclusive. Our data suggest that EphB2 regulates the expression of autophagy-related genes such as Atg5, Atg12, and apoptotic genes such as Bcl2, Birc5 and caspases 3 and 9. It is now our challenge to identify how the pathways involving these genes are integrated to result in the final growth regulation outcome. If a ‘functional switch’ occurs here which blocks apoptotic signaling, EphB2-mediated autophagy is likely to result in increased cell survival and tumor growth (Fig. 9). Since, according to the results described in this work, EphB2 also has proinvasive functions, the concomitance of apoptosis blockade, onset of autophagy and invasion might result in increased conditions for the metastatic process to occur. This constellation of events might explain the higher percentage of EphB2-positive metastases observed by IHC and also as suggested by an earlier immunohistochemistry-based report of a negative association between increased levels of EphB2 and overall patient survival [72]. Conflicting data have been previously reported regarding the role of EphB receptors in metastasis in other cancers. For instance, on one hand, a trend for decreased EphB2 expression was observed in

Fig. 4 – EphB2 expression induces apoptosis and autophagy in MDA435tetEphB2 cells. (a) Flow cytometry analysis using Annexin V/propidium iodide (PI) staining shows a significant increase in the levels of cell death upon DOX-induced EphB2 expression. The calculated “Cell Death” values shown in the histogram include the fraction of cells undergoing either early or late apoptosis/ necrosis. (b) EphB2 expression induces the activation of apoptotic signaling in MDA435tetEphB2 cells as shown by caspases-3 and -9 activation and (c) lower expression of the anti-apoptotic genes Bcl2 and Birc5, assessed by immunoblotting and real time PCR respectively. Expression levels were assessed after 5 days of DOX induction (1 μg/mL). (d) EphB2 induces the increase in LC3 punctate staining shown by immunofluorescence using an anti-LC3 antibody. DAPI indicates nuclear staining. (e) Transfection of MDA435tetEphB2 cells with a plasmid encoding a YFP-tagged LC3 shows that EphB2 expression increases the fraction of autophagic cells, identified as cells with strong punctate fluorescence (stars). DAPI indicates nuclear staining. The histogram illustrates the results of three different counts. Note that the counting reflects the fraction of autophagic cells within the efficiently transfected fluorescently labeled cells rather than the total cell population. However, transfection efficiency is the same between the two conditions, since cells were first transfected and later seeded and DOX-induced to express EphB2. (f) EphB2 expression induces the activation of autophagic signaling in MDA435tetEphB2 cells as shown by the expression of ATG5 and ATG12, as well as the levels of the LC3-II form, assessed by immunoblotting. (***Po0.001).

242

EX P ER I ME NTAL C E LL RE S E ARCH

Fig. 5 – EphB2 expression induces cell invasion in MDA435tetEphB2 cells. (a) DOX-induced EphB2 expression increases cell invasiveness in a modified Boyden chamber assay. Quantification was performed by counting the invading cells in at least three microscopic fields. (b) This effect is reflected by the expression of the pro-invasive genes MMP2 and MMP9, as measured by real time RT-PCR. (c) The invasive effect was inhibited as a result of treatment with either an MMP2 or MMP9-neutralizing antibody. An IgG treatment was used as a negative control. (**Po0.01, ***Po0.001, and ns: not significant).

3 20 (2014 ) 233 – 246

Fig. 6 – EphB2-mediated invasion is kinase-dependent. (a) A kinase-dead mutant EphB2 (KD-EphB2 or KD), devoid of autophosphorylation, is unable to induce cell invasion, in transfected MDA435 cells, as assessed in a Boyden chamber-like assay. (b) Quantification was performed by counting the invading cells in at least three microscopic fields and compared with the wild type EphB2 (WT-EphB2 or WT) and control vector (pcDNA3.1 or 3.1) transfectants. (c) Immunoprecipitation, performed using an antiphosphotyrosine antibody, was followed by immunoblotting with an anti-EphB2 antibody, showing the inability of KD-EphB2 to undergo autophosphorylation. An immunoblotting was performed on EphB2 using total protein extracts to control for equal EphB2 expression in both the WT-EphB2 and KD-EphB2transfected cells. Protein extracts from DOX-induced MDA435tetEphB2 cells were used as a positive control. (***Po0.001).

metastatic lesions of colorectal cancers (CRC), but no significant association was observed between the EphB2 expression and advanced tumor grade in this study [37]. Also, CRC patients with lower EphB2 expression were found to have more advanced tumor

E XP E RI ME N TAL CE L L R ES E ARC H

32 0 (2 014 ) 2 33 –24 6

243

Fig. 7 – EphB2 expression knockdown increases cell growth and expression of anti-apoptotic genes, decreases expression of proautophagic genes. (a) siRNA-mediated EphB2 silencing increases the growth of MDA-MB-231 cells, as examined by colony formation. (b) This effect was also examined by MTT assay. (c) This function was associated with induction of the pro-survival genes Bcl2 and Birc5, and (d) decrease of the autophagy markers ATG5 and ATG12 as assessed by real time RT-PCR and immunoblotting respectively. (***Po0.001 and ns: not significant).

stage, poor differentiation, poor overall survival and disease-free survival [26]. Another study suggested that low expression of EphB2 is correlated with CRC metastasis to the liver [59]. Overexpression of EphB2 inhibited colon cancer migration [26]. On the other hand, EphB receptors, and especially EphB4, have been shown to promote the migration of melanoma cells [75]. EphB4 also promotes tumor growth by stimulating angiogenesis [56], a process that is essential for the metastatic process. It has also been reported that migrating glioblastoma (GBM) cells express high levels of EphB2 in vitro and in vivo [52] and EphB2 overexpression in glioma cells results in increased cell invasion [53]. More recently, using a model of GBM neurosphere formation, it was shown that EphB2 expression stimulates GBM cell migration and invasion [71]. In other respects, it is important to note that the functions of EphB2 are likely to be modulated by the co-expression and mutual activation by and of its Ephrin ligands. Since EphB2 has different Ephrin ligands, we chose to use cellular models with no or low levels of ligand expression (Supplemental Fig. S3), to be able to control the study parameters. Needless to say that the role of the ligands is very important, particularly in the Eph/Ephrin family, where the membrane-bound ligands behave as receptors that elicit reverse signaling. Since receptor tyrosine kinase (RTK) can be autophosphorylated upon ectopic expression, independently from the ligands [28], in this study only EphB2 forward signaling (receptor-originated) is initiated by virtue of its overexpression and auto-activation, thus

mimicking ligand activation without having to take into account the reverse signaling. Nevertheless, it would be necessary to take into account the levels of expression and activation of the different EphB2 ligands, to fully comprehend its role in autophagy, apoptosis and invasion.

Conclusion Overall our data lead to a testable model for future studies: in a normal context (Fig. 9), EphB2-mediated autophagy induction triggers pro-apoptotic signals resulting in tumor growth inhibition. However, in a cancer initiated context, a blockade of apoptosis occurs that favors the pro-survival function of autophagy, thus allowing emergence of the pro-invasive function of EphB2. The translational impact of this study is significant. Little yet important data are available regarding the role of many Ephs/ Ephrins in breast cancer. These were mainly reported for the Atype EphA2 and the B-type EphB4 and EphB6 [5–8,17,27, 30,38,45,46,55,57,68,77]. As a result of research done in the previous decade, much work is going into the direction of targeting Eph receptors in multiple malignancies including breast cancer [24,54]. However, given the contradictory data on the biological impact of Eph receptors on the growth of various tumors, both pro-tumor progression effects in certain conditions

244

EX P ER I ME NTAL C E LL RE S E ARCH

3 20 (2014 ) 233 – 246

Fig. 9 – Summary of the EphB2 functions and suggested mechanisms in breast cancer cells. EphB2 regulates three different processes: apoptosis via caspases (CASPs) activation, autophagy via ATG genes (ATGs) and LC3, and invasion via metalloproteases MMP2 and MMP9 (MMPs). Our working hypothesis is that the three processes are linked and their differential modulation could drive either tumor growth inhibition or tumor progression and invasion. For instance, in normal or non-invasive cells, the maintenance of the proautophagic function of EphB2 could trigger apoptotic cell death. However, in conditions where the apoptotic machinery is blocked, the pro-survival role of autophagy is more prominent, thus favoring the invasive phenotype. Fig. 8 – EphB2 expression knockdown inhibits cell invasion and expression of pro-invasive genes. (a) shRNA-mediated EphB2 silencing (EphB2sh44) decreases MDA-MB-231 cells' invasiveness in comparison with the non-silencing (NSsh) control counterpart, as assessed using the Boyden chamberlike assay. Quantification was performed by counting the invading cells in at least three microscopic fields. (b) Following EphB2 silencing, the expression of the pro-invasive genes MMP2 and MMP9 was significantly decreased in EphB2silenced (siRNA) versus control counterparts (Sc), as measured by real time RT-PCR. (*Po0.05, **Po0.01, and ***Po0.001).

and anti-tumor effects in others, this requires more detailed investigation before an application of Eph targeting can be reliably developed. As illustrated by our study, tumor-suppressive and pro-invasive functions can even be controlled by the same Eph or Ephrin protein. In colon, EphB2 was shown to control both cell proliferation and cell positioning and migration independently [25,34]. In Glioblastomas, EphB2 controls both proliferation and migration [71]. It is therefore critical to understand the biology of these molecules if they are to be targeted therapeutically. Our work demonstrates that EphB2 has a dual role in breast cancer. On one hand it drives tumor growth inhibition. On the other hand, it has pro-autophagic and pro-invasive functions. This information is critical in understanding the conditions in which EphB2 functions can be activated and in designing therapeutic strategies aimed at this RTK, at the right time and cellular context.

Conflict of interest The authors declare no conflict of interest.

Acknowledgments This work was supported by a Breast Cancer Concept Award from the US Department of Defense and by a grant from the Canadian Institute for Health Research, to M.K. and G.B., and by a start-up fund from Wayne State University School of Medicine to M.K. The Microscopy, Imaging and Cytometry Resources Core is supported, in part, by NIH Center Grant P30CA022453 to The Karmanos Cancer Institute, Wayne State University and the Perinatology Research Branch of the National Institutes of Child Health and Development, Wayne State University. We thank Dr. Fazlul Zarkar and Dr. Ramzi Mohammad for access to the fluorescent microscopy instrument.

Appendix A.

Supporting information

Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.yexcr.2013.10.022.

reference [1] D. Arvanitis, A. Davy, Eph/Ephrin signaling: networks, Genes Dev. 22 (2008) 416–429. [2] S. Bacus, J.L. Flowers, M.F. Press, J.W. Bacus, K.S. McCarty Jr., The evaluation of estrogen receptor in primary breast carcinoma by computer-assisted image analysis, Am. J. Clin. Pathol. 90 (1988) 233–239. [3] E.H. Baehrecke, Autophagy: dual roles in life and death?, Nat. Rev. Mol. Cell Biol. 6 (2005) 505–510.

E XP E RI ME N TAL CE L L R ES E ARC H

[4] E. Batlle, J. Bacani, H. Begthel, S. Jonkheer, A. Gregorieff, B.M. van de, N. Malats, E. Sancho, E. Boon, T. Pawson, S. Gallinger, S. Pals, H. Clevers, EphB receptor activity suppresses colorectal cancer progression, Nature 435 (2005) 1126–1130. [5] G. Berclaz, B. Flutsch, H.J. Altermatt, V. Rohrbach, V. Djonov, A. Ziemiecki, E. Dreher, A.C. Andres, Loss of EphB4 receptor tyrosine kinase protein expression during carcinogenesis of the human breast, Oncol. Rep. 9 (2002) 985–989. [6] L. Bhushan, R.P. Kandpal, EphB6 receptor modulates micro RNA profile of breast carcinoma cells, PLoS One 6 (2011) e22484. [7] M. Brantley-Sieders, A. Jiang, K. Sarma, A. Badu-Nkansah, D.L. Walter, Y. Shyr, J. Chen, Eph/Ephrin profiling in human breast cancer reveals significant associations between expression level and clinical outcome, PLoS One 6 (2011) e24426. [8] D.M. Brantley-Sieders, G. Zhuang, D. Hicks, W.B. Fang, Y. Hwang, J.M. Cates, K. Coffman, D. Jackson, E. Bruckheimer, R.S. Muraoka-Cook, J. Chen, The receptor tyrosine kinase EphA2 promotes mammary adenocarcinoma tumorigenesis and metastatic progression in mice by amplifying ErbB2 signaling, J. Clin. Invest. 118 (2008) 64–78. [9] A.F. Chambers, MDA-MB-435 and M14 cell lines: identical but not M14 melanoma?, Cancer Res. 69 (2009) 5292–5293. [10] Y. Chen, A.K. Fu, N.Y. Ip, Eph receptors at synapses: implications in neurodegenerative diseases, Cell Signal. 24 (2012) 606–611. [11] N. Chunder, S. Mandal, D. Basu, A. Roy, S. Roychoudhury, C.K. Panda, Deletion mapping of chromosome 1 in early onset and late onset breast tumors—a comparative study in eastern India, Pathol. Res. Pract. 199 (2003) 313–321. [12] K.N. Dalby, I. Tekedereli, G. Lopez-Berestein, B. Ozpolat, Targeting the prodeath and prosurvival functions of autophagy as novel therapeutic strategies in cancer, Autophagy 6 (2010) 322–329. [13] J. Debnath, The multifaceted roles of autophagy in tumorsimplications for breast cancer, J. Mammary Gland Biol. Neoplasia 16 (2011) 173–187. [14] V.C. Dodelet, E.B. Pasquale, Eph receptors and Ephrin ligands: embryogenesis to tumorigenesis, Oncogene 19 (2000) 5614–5619. [15] G. Ellison, T. Klinowska, R.F. Westwood, E. Docter, T. French, J.C. Fox, Further evidence to support the melanocytic origin of MDAMB-435, Mol. Pathol. 55 (2002) 294–299. [16] V. Espina, B.D. Mariani, R.I. Gallagher, K. Tran, S. Banks, J. Wiedemann, H. Huryk, C. Mueller, L. Adamo, J. Deng, E.F. Petricoin, L. Pastore, S. Zaman, G. Menezes, J. Mize, J. Johal, K. Edmiston, L.A. Liotta, Malignant precursor cells pre-exist in human breast DCIS and require autophagy for survival, PLoS One 5 (2010) e10240. [17] W.B. Fang, D.M. Brantley-Sieders, M.A. Parker, A.D. Reith, J. Chen, A kinase-dependent role for EphA2 receptor in promoting tumor growth and metastasis, Oncogene 24 (2005) 7859–7868. [18] G.M. Fimia, M. Piacentini, Regulation of autophagy in mammals and its interplay with apoptosis, Cell Mol. Life Sci. 67 (2010) 1581–1588. [19] J.G. Flanagan, P. Vanderhaeghen, The Ephrins and Eph receptors in neural development, Annu. Rev. Neurosci. 21 (1998) 309–345. [20] A.M. Flenniken, N.W. Gale, G.D. Yancopoulos, D.G. Wilkinson, Distinct and overlapping expression patterns of ligands for Ephrelated receptor tyrosine kinases during mouse embryogenesis, Dev. Biol. 179 (1996) 382–401. [21] G.C. Friedman, D.D. O'Leary, Eph receptor tyrosine kinases and their ligands in neural development, Curr. Opin. Neurobiol. 6 (1996) 127–133. [22] J. Frisen, J. Holmberg, M. Barbacid, Ephrins and their Eph receptors: multitalented directors of embryonic development, EMBO J. 18 (1999) 5159–5165. [23] N.W. Gale, S.J. Holland, D.M. Valenzuela, A. Flenniken, L. Pan, T.E. Ryan, M. Henkemeyer, K. Strebhardt, H. Hirai, D.G. Wilkinson, T. Pawson, S. Davis, G.D. Yancopoulos, Eph receptors and ligands comprise two major specificity subclasses and are reciprocally compartmentalized during embryogenesis, Neuron 17 (1996) 9–19. [24] K. Garber, Of Ephs and Ephrins: companies target guidance molecules in cancer, J. Natl. Cancer Inst. 102 (2010) 1692–1694.

32 0 (2 014 ) 2 33 –24 6

245

[25] M. Genander, M.M. Halford, N.J. Xu, M. Eriksson, Z. Yu, Z. Qiu, A. Martling, G. Greicius, S. Thakar, T. Catchpole, M.J. Chumley, S. Zdunek, C. Wang, T. Holm, S.P. Goff, S. Pettersson, R.G. Pestell, M. Henkemeyer, J. Frisen, Dissociation of EphB2 signaling pathways mediating progenitor cell proliferation and tumor suppression, Cell 139 (2009) 679–692. [26] D.L. Guo, J. Zhang, S.T. Yuen, W.Y. Tsui, A.S. Chan, C. Ho, J. Ji, S.Y. Leung, X. Chen, Reduced expression of EphB2 that parallels invasion and metastasis in colorectal tumours, Carcinogenesis 27 (2006) 454–464. [27] G.C. Harburg, L. Hinck, Navigating breast cancer: axon guidance molecules as breast cancer tumor suppressors and oncogenes, J. Mammary Gland Biol. Neoplasia 16 (2011) 257–270. [28] L. He, K. Hristova, Physical–chemical principles underlying RTK activation, and their implications for human disease, Biochim. Biophys. Acta 1818 (2012) 995–1005. [29] J.P. Himanen, M.J. Chumley, M. Lackmann, C. Li, W.A. Barton, P.D. Jeffrey, C. Vearing, D. Geleick, D.A. Feldheim, A.W. Boyd, M. Henkemeyer, D.B. Nikolov, Repelling class discrimination: Ephrin-A5 binds to and activates EphB2 receptor signaling, Nat. Neurosci. 7 (2004) 501–509. [30] N. Hiramoto-Yamaki, S. Takeuchi, S. Ueda, K. Harada, S. Fujimoto, M. Negishi, H. Katoh, Ephexin4 and EphA2 mediate cell migration through a RhoG-dependent mechanism, J. Cell Biol. 190 (2010) 461–477. [31] N. Holder, R. Klein, Eph receptors and Ephrins: effectors of morphogenesis, Development 126 (1999) 2033–2044. [32] A. Hollestelle, J.H. Nagel, M. Smid, S. Lam, F. Elstrodt, M. Wasielewski, S.S. Ng, P.J. French, J.K. Peeters, M.J. Rozendaal, M. Riaz, D.G. Koopman, T.L. Ten Hagen, B.H. de Leeuw, E.C. Zwarthoff, A. Teunisse, P.J. van der Spek, J.G. Klijn, W.N. Dinjens, S.P. Ethier, H. Clevers, A.G. Jochemsen, M.A. den Bakker, J.A. Foekens, J.W. Martens, M. Schutte, Distinct gene mutation profiles among luminal-type and basal-type breast cancer cell lines, Breast Cancer Res. Treat. 121 (2010) 53–64. [33] A. Hollestelle, M. Schutte, Comment Re: MDA-MB-435 and M14 cell lines: identical but not M14 Melanoma?, Cancer Res. 69 (2009) 7893. [34] J. Holmberg, M. Genander, M.M. Halford, C. Anneren, M. Sondell, M.J. Chumley, R.E. Silvany, M. Henkemeyer, J. Frisen, EphB receptors coordinate migration and proliferation in the intestinal stem cell niche, Cell 125 (2006) 1151–1163. [35] M. Hruska, M.B. Dalva, Ephrin regulation of synapse formation, function and plasticity, Mol. Cell Neurosci. 50 (2012) 35–44. [36] P. Huusko, D. Ponciano-Jackson, M. Wolf, J.A. Kiefer, D.O. Azorsa, S. Tuzmen, D. Weaver, C. Robbins, T. Moses, M. Allinen, S. Hautaniemi, Y. Chen, A. Elkahloun, M. Basik, G.S. Bova, L. Bubendorf, A. Lugli, G. Sauter, J. Schleutker, H. Ozcelik, S. Elowe, T. Pawson, J.M. Trent, J.D. Carpten, O.P. Kallioniemi, S. Mousses, Nonsense-mediated decay microarray analysis identifies mutations of EPHB2 in human prostate cancer, Nat. Genet. 36 (2004) 979–983. [37] A.M. Jubb, F. Zhong, S. Bheddah, H.I. Grabsch, G.D. Frantz, W. Mueller, V. Kavi, P. Quirke, P. Polakis, H. Koeppen, EphB2 is a prognostic factor in colorectal cancer, Clin. Cancer Res. 11 (2005) 5181–5187. [38] P. Kaenel, M. Mosimann, A.C. Andres, The multifaceted roles of Eph– Ephrin signaling in breast cancer, Cell Adhes. Migr. 6 (2012) 138–147. [39] M. Kandouz, The Eph/Ephrin family in cancer metastasis: communication at the service of invasion, Cancer Metastasis Rev. 31 (2012) 353–373. [40] M. Kandouz, K. Haidara, J. Zhao, M.L. Brisson, G. Batist, The EphB2 tumor suppressor induces autophagic cell death via concomitant activation of the ERK1/2 and PI3K pathways, Cell Cycle 9 (2010) 398–407. [41] A.C. Kimmelman, The dynamic nature of autophagy in cancer, Genes Dev. 25 (2011) 1999–2010. [42] R.A. Kittles, A.B. Baffoe-Bonnie, T.Y. Moses, C.M. Robbins, C. Ahaghotu, P. Huusko, C. Pettaway, S. Vijayakumar, J. Bennett, G. Hoke, T. Mason, S. Weinrich, J.M. Trent, F.S. Collins, S. Mousses,

246

[43] [44] [45]

[46]

[47]

[48]

[49]

[50] [51]

[52]

[53]

[54]

[55]

[56]

[57]

[58] [59]

[60]

EX P ER I ME NTAL C E LL RE S E ARCH

J. Bailey-Wilson, P. Furbert-Harris, G. Dunston, I.J. Powell, J.D. Carpten, A common nonsense mutation in EphB2 is associated with prostate cancer risk in African American men with a positive family history, J. Med. Genet. 43 (2006) 507–511. R. Klein, Bidirectional modulation of synaptic functions by Eph/ Ephrin signaling, Nat. Neurosci. 12 (2009) 15–20. G. Kroemer, B. Levine, Autophagic cell death: the story of a misnomer, Nat. Rev. Mol. Cell Biol. 9 (2008) 1004–1010. S.R. Kumar, J. Singh, G. Xia, V. Krasnoperov, L. Hassanieh, E.J. Ley, J. Scehnet, N.G. Kumar, D. Hawes, M.F. Press, F.A. Weaver, P.S. Gill, Receptor tyrosine kinase EphB4 is a survival factor in breast cancer, Am. J. Pathol. 169 (2006) 279–293. M. Lu, K.D. Miller, Y. Gokmen-Polar, M.H. Jeng, M.S. Kinch, EphA2 overexpression decreases estrogen dependence and tamoxifen sensitivity, Cancer Res. 63 (2003) 3425–3429. A. Lugli, H. Spichtin, R. Maurer, M. Mirlacher, J. Kiefer, P. Huusko, D. Azorsa, L. Terracciano, G. Sauter, O.P. Kallioniemi, S. Mousses, L. Tornillo, EphB2 expression across 138 human tumor types in a tissue microarray: high levels of expression in gastrointestinal cancers, Clin. Cancer Res. 11 (2005) 6450–6458. S. Luo, D.C. Rubinsztein, Atg5 and Bcl-2 provide novel insights into the interplay between apoptosis and autophagy, Cell Death Differ. 14 (2007) 1247–1250. R.A. McClelland, U. Berger, L.S. Miller, T.J. Powles, E.V. Jensen, R.C. Coombes, Immunocytochemical assay for estrogen receptor: relationship to outcome of therapy in patients with advanced breast cancer, Cancer Res. 46 (1986) 4241s–4243s. A. Merlos-Suarez, E. Batlle, Eph–Ephrin signalling in adult tissues and cancer, Curr. Opin. Cell Biol. 20 (2008) 194–200. V. Montel, M. Suzuki, C. Galloy, E.S. Mose, D. Tarin, Expression of melanocyte-related genes in human breast cancer and its implications, Differentiation 78 (2009) 283–291. M. Nakada, J.A. Niska, H. Miyamori, W.S. McDonough, J. Wu, H. Sato, M.E. Berens, The phosphorylation of EphB2 receptor regulates migration and invasion of human glioma cells, Cancer Res. 64 (2004) 3179–3185. M. Nakada, J.A. Niska, N.L. Tran, W.S. McDonough, M.E. Berens, EphB2/R-Ras signaling regulates glioma cell adhesion, growth, and invasion, Am. J. Pathol. 167 (2005) 565–576. E. Nievergall, T. Saunders, M. Lackmann, Targeting of EPH receptor tyrosine kinases for anticancer therapy, Crit. Rev.Oncog. 17 (2012) 211–232. N.K. Noren, G. Foos, C.A. Hauser, E.B. Pasquale, The EphB4 receptor suppresses breast cancer cell tumorigenicity through an Abl-Crk pathway, Nat. Cell Biol. 8 (2006) 815–825. N.K. Noren, M. Lu, A.L. Freeman, M. Koolpe, E.B. Pasquale, Interplay between EphB4 on tumor cells and vascular Ephrin-B2 regulates tumor growth, Proc. Natl. Acad. Sci. USA 101 (2004) 5583–5588. N.K. Noren, M. Lu, A.L. Freeman, M. Koolpe, E.B. Pasquale, Interplay between EphB4 on tumor cells and vascular Ephrin-B2 regulates tumor growth, Proc. Natl. Acad. Sci. USA 101 (2004) 5583–5588. D.D. O'Leary, D.G. Wilkinson, Eph receptors and Ephrins in neural development, Curr. Opin. Neurobiol. 9 (1999) 65–73. T. Oshima, M. Akaike, K. Yoshihara, M. Shiozawa, N. Yamamoto, T. Sato, N. Akihito, Y. Nagano, S. Fujii, C. Kunisaki, N. Wada, Y. Rino, K. Tanaka, M. Masuda, T. Imada, Overexpression of EphA4 gene and reduced expression of EphB2 gene correlates with liver metastasis in colorectal cancer, Int. J. Oncol. 33 (2008) 573–577. E.B. Pasquale, Eph–Ephrin bidirectional signaling in physiology and disease, Cell 133 (2008) 38–52.

3 20 (2014 ) 233 – 246

[61] J.M. Rae, C.J. Creighton, J.M. Meck, B.R. Haddad, M.D. Johnson, MDA-MB-435 cells are derived from M14 melanoma cells—a loss for breast cancer, but a boon for melanoma research, Breast Cancer Res.Treat. 104 (2007) 13–19. [62] G. Ragnarsson, G. Eiriksdottir, J.T. Johannsdottir, J.G. Jonasson, V. Egilsson, S. Ingvarsson, Loss of heterozygosity at chromosome 1p in different solid human tumours: association with survival, Br. J. Cancer 79 (1999) 1468–1474. [63] H. Rikiishi, Novel insights into the interplay between apoptosis and autophagy, Int. J. Cell Biol. 2012 (2012) 317645. [64] D.T. Ross, U. Scherf, M.B. Eisen, C.M. Perou, C. Rees, P. Spellman, V. Iyer, S.S. Jeffrey, M. Van de Rijn, M. Waltham, A. Pergamenschikov, J.C. Lee, D. Lashkari, D. Shalon, T.G. Myers, J.N. Weinstein, D. Botstein, P.O. Brown, Systematic variation in gene expression patterns in human cancer cell lines, Nat. Genet. 24 (2000) 227–235. [65] F. Scarlatti, R. Granata, A.J. Meijer, P. Codogno, Does autophagy have a license to kill mammalian cells?, Cell Death Differ. 16 (2009) 12–20. [66] S. Sellappan, R. Grijalva, X. Zhou, W. Yang, M.B. Eli, G.B. Mills, D. Yu, Lineage infidelity of MDA-MB-435 cells: expression of melanocyte proteins in a breast cancer cell line, Cancer Res. 64 (2004) 3479–3485. [67] E.Y. Tan, L. Campo, C. Han, H. Turley, F. Pezzella, K.C. Gatter, A.L. Harris, S.B. Fox, BNIP3 as a progression marker in primary human breast cancer; opposing functions in in situ versus invasive cancer, Clin. Cancer Res. 13 (2007) 467–474. [68] L. Truitt, T. Freywald, J. DeCoteau, N. Sharfe, A. Freywald, The EphB6 receptor cooperates with c-Cbl to regulate the behavior of breast cancer cells, Cancer Res. 70 (2010) 1141–1153. [69] C.V. Ustach, M.E. Taube, N.J. Hurst Jr., S. Bhagat, R.D. Bonfil, M.L. Cher, L. Schuger, H.R. Kim, A potential oncogenic activity of platelet-derived growth factor d in prostate cancer progression, Cancer Res. 64 (2004) 1722–1729. [70] H.U. Wang, Z.F. Chen, D.J. Anderson, Molecular distinction and angiogenic interaction between embryonic arteries and veins revealed by Ephrin-B2 and its receptor Eph-B4, Cell 93 (1998) 741–753. [71] S.D. Wang, P. Rath, B. Lal, J.P. Richard, Y. Li, C.R. Goodwin, J. Laterra, S. Xia, EphB2 receptor controls proliferation/migration dichotomy of glioblastoma by interacting with focal adhesion kinase, Oncogene 31 (2012) 5132–5143. [72] Q. Wu, Z. Suo, B. Risberg, M.G. Karlsson, K. Villman, J.M. Nesland, Expression of Ephb2 and Ephb4 in breast carcinoma, Pathol. Oncol. Res. 10 (2004) 26–33. [73] W.K. Wu, S.B. Coffelt, C.H. Cho, X.J. Wang, C.W. Lee, F.K. Chan, J. Yu, J.J. Sung, The autophagic paradox in cancer therapy, Oncogene 31 (2012) 939–953. [74] G.D. Yancopoulos, M. Klagsbrun, J. Folkman, Vasculogenesis, angiogenesis, and growth factors: Ephrins enter the fray at the border, Cell 93 (1998) 661–664. [75] N.Y. Yang, E.B. Pasquale, L.B. Owen, I.M. Ethell, The EphB4 receptor-tyrosine kinase promotes the migration of melanoma cells through Rho-mediated actin cytoskeleton reorganization, J. Biol. Chem. 281 (2006) 32574–32586. [76] Z.J. Yang, C.E. Chee, S. Huang, F.A. Sinicrope, The role of autophagy in cancer: therapeutic implications, Mol. Cancer Ther. 10 (2011) 1533–1541. [77] D.P. Zelinski, N.D. Zantek, J.C. Stewart, A.R. Irizarry, M.S. Kinch, EphA2 overexpression causes tumorigenesis of mammary epithelial cells, Cancer Res. 61 (2001) 2301–2306.