Abstract
Aberrant expression of erythropoietin-producing hepatocellular carcinoma cell (EPH) receptors has been reported in a variety of human cancer types. In addition to modulating cell proliferation and migration, EPH receptors are also involved in tumor progression. The transcriptional activation and silencing of EPH receptors are also associated with tumorigenesis. However, the mechanisms underlying the involvement of EPH receptors in tumorigenesis have not been completely deciphered. We have investigated and described the role of EPHB6, a kinase-deficient receptor, in modulating the abundance of cadherin 17 and activation of other intracellular signaling proteins. We previously showed that EPHB6 alters the tumor phenotype of breast carcinoma cells. However, the mechanisms underlying these phenotypic changes had not previously been investigated. Herein we demonstrated the downstream effects of EPHB6 expression on the abundance of cadherin 17, mitogen-activated protein kinase (MEK2), extracellular signal-regulated kinase (ERK), phospho-ERK, β-catenin, phospho- glycogen synthase kinase 3 beta (GSK3β) (ser21/9), cell morphology and actin cytoskeleton. These comparisons were made between EPHB6-deficient MDA-MB-231 cells transfected with an empty pcDNA3 vector and cells stably transfected with an expression construct of EPHB6. The results indicate elevated levels of MEK2 and phospho-ERK. While there was no change in the amount of ERK, the abundance of cadherin 17, β-catenin and phospho-GSK3β was significantly reduced in EPHB6-transfected cells. These studies clearly demonstrate an inverse relationship between the levels of phospho-ERK and the abundance of cadherin 17, β-catenin and phospho-GSK3β in EPHB6-expressing MDA-MB-231 cells. From these data we conclude that EPHB6-mediated alterations arise due to changes in abundance and localization of cadherin 17 and activation of WNT signaling pathway. Transcriptional silencing of EPHB6 in native MDA-MB-231 cells and consequent effects on cadherin 17 and WNT pathway may, thus, be responsible for the invasive behavior of these cells.
Erythropoietin-producing hepatocellular carcinoma cell receptor B6 (EPHB6) is a member of the largest family of receptor tyrosine kinases (1, 2). EPHB6 and EPHA10 are the only kinase-deficient members of this family. These kinase-deficient receptors are likely to transduce signals by associating with other kinase-sufficient receptors (3).The modulation of cell attachment and cell migration by EPH receptors has been correlated to invasiveness of tumor cells (4-6). The decrease in the expression of EPHB6 receptor has been linked to aggressiveness and invasiveness in melanoma (7), neuroblastoma (8) and non-small cell lung carcinoma (9). The invasiveness of MDA-MB-231 cells has been attributed, in part, to loss of EPHB6 (10), and ectopic expression of EPHB6 has been shown to suppress in vitro invasiveness of MDA-MB-231 cells (11). However, the mechanism by which EPHB6 receptor suppresses the invasive phenotype of MDA-MB-231 cells remains unclear.
In addition to transducing signals from the cell surface to the nucleus, EPH receptors also contribute to cell adhesion. Similarly, a superfamily of cell surface molecules, which contribute to cell adhesion and signal transduction, is comprised of cadherins. Cadherins are transmembrane receptors that mediate calcium-dependent homophilic or heterophilic adhesion between cells. Members of the cadherin superfamily include classical cadherins, desmosomal cadherins, protocadherins and products of tumor-suppressor genes such as c-rearranged during transfection (RET) gene and FAT (12). Classical cadherins enable adhesion between cells, maintain cell polarity and preserve tissue integrity (13). Cadherins are linked with the actin cytoskeleton via α, β, and γ isoforms of cytoplasmic catenins (14-16), and facilitate a variety of molecular changes during cell development and morphogenesis (17, 18). Aberrant expression of cadherins has been associated with a variety of tumors (19, 20), and oncogenic properties of some cadherins have been confirmed (21, 22). Altered expression of cadherin 17, in particular, has been implicated in several types of cancers (21, 23), and it has also been considered a useful marker for gastrointestinal carcinomas (24). The association of cadherin 17 expression with caudal-type homeobox protein 2 (CDX2) transcription factor in ovarian cancer (23), and epidermal growth factor receptor and nuclear factor kappa-light-chain-enhancer of activated B cells (NF-ϰB) signaling proteins in gastric cancer (21, 25) suggests important roles of cadherin 17 in signal transduction. Although EPH receptors and cadherins have been investigated extensively, detailed investigations of interactions between these two families of cell surface molecules have not been performed. Some recent investigations, however, demonstrate modulation of signal transduction by cadherin and EPH receptor (26, 27).
The altered regulation of various EPH receptors has been observed in a variety of cancer types (28). However, the downstream intracellular effector molecules have not been characterized completely. Some EPH receptor signaling events include EPHB4 dependent modulation of estrogen receptor-alpha in breast cancer (29), activation of mammalian target of rapamycin complex 1 and extracellular-regulated kinase pathways by EPHA2 (30), interaction of EPHB1 receptor with tumor-suppressor phosphatase and tensin homolog (31), and EPHB3-mediated suppression of metastasis via protein phosphatase 2(PP2A)/RAC/AKT signaling pathway in non-small cell lung carcinoma (32). Morphological changes mediated by EPH have been characterized by co-localization of EPHA2 and integrin α3 at cell edges and protrusions in a glioblastoma cell line (33). The modulation of cell proliferation, migration and apoptosis has been attributed to specific activation of EPH receptors by their cognate ephrin ligands (34-39). Given the diversity of intracellular proteins affected by EPH receptors, it is not surprising that EPHB4 has been designated as both a tumor promoter and a tumor suppressor (40).
In light of the background described above, we investigated the association of cadherin 17 expression with kinase-deficient EPHB6 receptor in the breast cancer cell line MDA-MB-231, and demonstrated the effects of EPHB6 on cytoskeleton components and the levels of intracellular signaling proteins such as mitogen-activated protein kinase (MEK2), extracellular signal-regulated kinase (ERK), β-catenin and phosphorylated glycogen synthase kinase 3 beta (GSK3β). We discussed these molecular changes in the context of morphological and cytoskeletal rearrangements in native and EPHB6-expressing MDA-MB-231 cells. Our studies suggest that EPHB6 down-regulates the levels of cadherin 17, and possibly also alters its cellular localization.
Materials and Methods
Cell culture. MCF10A, MCF7, BT20, MDA-MB-435, and MDA-MB-468 cells were grown as described previously (10, 11). Native MDA-MB-231 cells transfected with pcDNA3 vector and EPHB6-transfected MDA-MB-231 cells (11) were cultured at 37°C/7% CO2 in Dulbecco’s modified eagle’s medium (DMEM) supplemented with 2 mM L-glutamine, 1 mM sodium pyruvate, 100 units/ml PenStrep, and 10% fetal bovine serum (all Life Technologies, Grand Island, NY, USA).
Total RNA isolation and reverse transcription polymerase chain reaction (RT PCR). Cells were grown until 85-95% confluence in 25-mm cell culture flasks. RNA was isolated from various cell lines using TRI reagent (Molecular Research Center, Cincinnati, OH, USA) according to the manufacturer’s protocol. The amount of RNA was determined by NanoDrop (Thermo Scientific, Waltham, MA, USA), and its purity checked by determining the ratio of absorbance at 260 nm and 280 nm. The RNA preparation was treated with RNase-free DNase and 1.5 μg RNA was reverse transcribed with oligo dT using the Superscript III First-Strand RNA synthesis kit (Life Technologies) according to the manufacturer’s protocol. The abundance of cadherin 17 (CDH17) transcript was quantified by PCR using 5’ACAATCGACCCACGTTTCTC3’ and 5’ATATTGTGCACCGGGATCAT3’ oligonucleotides (Integrated DNA Technologies, Inc., Coralville, IA, USA) as forward and reverse primers, respectively. The reaction mixture (20 μl) containing 10 ng cDNA, 6.0 pmol each of forward and reverse primers, 4.0 nmol dATP, dCTP, dGTP and dTTP, 0.4 units Taq DNA polymerase and 1X reaction buffer was amplified for 35 cycles of denaturation at 95°C for 1 min, annealing at 65°C for 1 min and extension at 72°C for 1 min. The amplified product was separated by electrophoresis on a 1% agarose gel and visualized by staining with ethidium bromide. PCR amplifications were also carried out with primers corresponding to housekeeping gene β-actin transcript to normalize the amounts of cDNA templates used from various cell lines.
Western blotting. MDA-MB-231 cells, stably transfected with an EPHB6 construct (MDA-MB-231-EPHB6) or an empty pcDNA3 vector (MDA-MB-231-pCDNA), were grown in DMEM to 85-90% confluence. The culture medium was aspirated and the dish washed with 3 ml of ice-cold phosphate buffered saline (PBS). Cells were lysed by incubating with 250 μl of RIPA buffer (10 mM Tris, pH 7.4, 150 mM NaCl, 5 mM EDTA, 8% glycerol, 1% Triton X-100, 0.1% sodium dodecyl sulfate (SDS), 0.5% sodium deoxycholate, 10 mM Na3VO4, 1 mM sodium fluoride) and complete protease inhibitor mixture (Roche Diagnostics, Indianapolis, IN, USA) for 15 min with gentle shaking. The lysates were cleared by centrifugation at 16,000 g for 10 min at 4°C, and the concentration of protein in the supernatant was determined using BCA protein estimation kit (Thermo Scientific, Waltham, MA, USA). Forty micrograms of total protein per well were separated by SDS-polyacrylamide gel electrophoresis (PAGE) and the separated proteins were electroblotted onto methanol-presoaked polyvinylidene fluoride membranes. The membrane was blocked for 60 min with 5% bovine serum albumin (BSA) containing 0.1% Tween 20 in Tris-buffered saline at room temperature and then incubated overnight at 4°C with monoclonal antibodies to phospho-ERK1/2 (Thr202/Tyr204, 1:1,000; Cell Signaling Technology, Danvers, MA, USA), ERK1/2 (1:2,000; Cell Signaling Technology), β-catenin (1:1,000; Cell Signaling Technology), phospho-GSK3 (ser 21/9, 1:1,000; Cell Signaling Technology) or cadherin 17 (1: 1000; R&D, Minneapolis, MN, USA). Antigen-antibody complexes were detected by enhanced chemiluminescence (Amersham/GE Healthcare, Piscataway, NJ, USA) after incubating with the corresponding horseradish peroxidaseHRP-conjugated secondary antibodies. The amounts of protein in individual lanes were normalized by probing the membrane with antibody to glyceraldehyde 3-phosphate dehydrogenase (GAPDH) (1:10000 dilution; Millipore, Billerica, MA, USA). The intensity of specific bands corresponding to cadherin 17, MEK2, phosphor-ERK and GAPDH were quantified using ImageJ software (National Institutes of Health, Bethesda, MD, USA). Intensity ratios of cadherin 17/GAPDH, MEK2/GAPDH, phospho-ERK/GAPDH, phospho-GSK3β and β-catenin in MDA-MB-231-pcDNA3 and MDA-MB-231-EPHB6 cells were plotted using GraphPad software (GraphPad Software Inc., La Jolla, CA, USA).
Immunofluroscence. Cells were grown in a glass-bottom dish in DMEM supplemented with 2 mM L-glutamine, 1 mM sodium pyruvate (Life Technologies), 100 U/ml of PenStrep, and 10% fetal bovine serum (all Life Technologies) at 37°C/7% CO2 for 72 h. Culture medium was changed every 24 h. The dishes were washed twice with ice cold PBS (pH 7.4) and treated with Histochoice MB tissue fixative (Amresco, Solon, OH, USA) for 10 min. Fixative solution was aspirated, cells were permeabilized with 0.3 Triton X-100 made in PBS for 10 min, and incubated with antibody to cadherin 17 (1:100 dilution, R&D) overnight at 4°C. Cells were washed five times with ice-cold PBS (pH 7.4), incubated with alexa flour 546 donkey antibody to mouse IgG (1: 500 dilution) for 2 h at room temperature, and treated with 4’,6-diamidino-2-phenylindole (DAPI)-containing mounting solution. The fluorescence-labeled cells were analyzed on a Nikon A1 3 color confocal microscope (Nikon Inc., Melville, NY, USA) using ×20 objective.
Phalloidin labeling. To study the cytoskeletal re-arrangement, cells were labeled with Phalloidin-fluorescein isothiocyanate (FITC) (Enzo Life Sciences, Farmingdale, NY, USA) and treated with DAPI-containing mounting solution. The fluorescence was analyzed on Nikon A1 3-color confocal microscope using a ×100 objective.
Results
Semi-quantitative PCR of cadherin 17 transcript in breast carcinoma cells. To address the involvement of cell adhesion molecules in invasiveness of breast carcinoma cells, we investigated the expression pattern of cadherin 17 transcripts in MCF10A, MCF7, BT-20, MDA-MB-231, MDA-MB-435 and MDA-MB-468 cell lines. We had previously observed a differential expression of a variety of cadherin transcripts in these cell lines (data not shown). However, the expression of CDH17 transcript was detectable only in the MDA-MB-231 cell line (Figure 1A). In light of our earlier observation that EPHB6 suppresses invasiveness of MDA-MB-231 cells (10, 11) and our current results demonstrating the expression of cadherin 17 selectively in MDA-MB-231, we investigated the relationship between EPHB6 and cadherin 17 by characterizing the abundance of CDH17 transcript in MDA-MB-231 cells stably transfected with either an empty pcDNA3 vector or EPHB6-containing pcDNA3. The levels of CDH17 transcript were found to be higher in MDA-MB-231-pcDNA3 cells than MDA-MB231-EPHB6 cells (Figure 1B and C).
Abundance of cadherin 17 protein in breast carcinoma cells. As described above, the levels of CDH17 transcript were lower in EPHB6-expressing MDA-MB-231 cells, compared to cells transfected with empty pcDNA3 vector. The abundance of transcript in these cell lines was, therefore, compared to the levels of corresponding protein by western blotting. The amount of cadherin 17 protein in EPHB6-expressing cells was lower than the corresponding empty vector-transfected cells (Figure 1D and E). It warrants mentioning that all quantitations were performed in three biological replicates with three technical replicates of each experiment.
Cellular localization of cadherin 17 and its EPHB6-dependent down-regulation. Semi-quantitative PCR revealed a decrease in the abundance of CDH17 transcript in MDA-MB-231-EPHB6 cells compared to empty vector-transfected cells. This result prompted us to examine the in situ expression of cadherin 17 protein in MDA-MB231-EPHB6 and MDA-MB-231-pcDNA3 cell lines. Figure 2 shows localization and abundance of cadherin 17 protein in these cell lines. MCF10A cells, which have no detectable levels of cadherin 17 protein, were used as a negative control for these experiments. Figure 2A and C clearly show that MDA-MB-231-pcDNA cells express significantly higher amounts of the protein as compared to EPHB6-expressing cells (Figure 2D and F). Although the distinction between membranous and subcellular localization of cadherin 17 was not obvious from the images shown, optical sectioning of fluorescent images (data not shown) in other experiments tentatively suggested the localization of cadherin 17 on the cell surface in empty pcDNA3 vector-transfected MDA-MB-231 cells and around the nucleus in EPHB6-transfected cells.
EPHB6 activates ERK/MAPK pathway in MDA-MB-231 cell line. In order to delineate the signal transduction pathways mediated by EPHB6, expression of ERK/MAPK was investigated in vector-transfected and EPHB6-transfected MDA-MB-231 cell lines as described in Materials and Methods. As shown in Figure 3, the amount of MEK2 protein was significantly higher in MDA-MB-231-EPHB6 cells than in MDA-MB-231-pcDNA3 cells (Figure 3A and B). We subsequently quantified the levels of total ERK and phospho-ERK in these two cell lines. As shown in Figure 3C, the levels of total ERK were found to be comparable in MDA-MB-231-pcDNA3 and MDA-MB-231-EPHB6 cells. Although we did not observe any 44-kDa isoform of phospho-ERK under these conditions, the abundance of the 42-kDa phosphorylated form of ERK (Thr202/Tyr204) was elevated in MDA-MB-231-EPHB6 cells compared to MDA-MB-pCDNA3 cells (Figure 3C and D).
Cadherin 17 expression in breast carcinoma cell lines. A: Total RNA was isolated from various cell lines and cadherin 17 (CHD17) transcript was amplified by PCR with CHD17-specific primers as described in the Materials and Methods section. The amplified product from MCF10A (lane 2), MCF7 (lane 3), BT20 (lane 4), MDA-MB-231 (lane 5), MDA-MB-435 (lane 6) and MDA-MB-468 (lane 7) cells was separated on an agarose gel. Lane 1 represents a no-template control and lane 8 shows a molecular weight marker. The arrow indicates the presence of 480-basepair (bp) amplicon corresponding to CHD17 transcript. B: Total RNA was isolated and cadherin 17 transcript was amplified as described above. The amplified products from pcDNA3 vector-transfected cell line MDA-MB-231-pcDNA3 (lane 1) and EPHB6-transfected cell line MDA-MB-231-EPHB6 (lane 2) were separated on an agarose gel. The amounts of cDNA used for amplification were normalized by amplifying actin transcript in each sample. The arrows indicate the amplified products of expected size. C: The amounts of amplified products corresponding to MDA-MB-231-pcDNA3 and MDA-MB-231-EPHB6 were determined by densitometry using imageJ program and the ratios of CHD17 and actintranscripts in cell lines were compared as a histogram. The data are presented as the mean and standard deviation of three experiments and triplicate amplifications in each experiment. D: Total protein was isolated from MDA-MB-231-pcDNA3 and MDA-MB-231-EPHB6 cell lines, separated by electrophoresis, probed with antibody against cadherin 17 or glyceraldehyde 3-phosphate dehydrogenase (GAPDH) and the antigen-antibody complexes detected by chemiluminescence. E: The amounts of protein corresponding to cadherin 17 and GAPDH in MB-231-pcDNA3 and MDA-MB-231-EPHB6 were determined by densitometry using the imageJ program and the data presented as the mean and standard deviation of three experiments.
In situ expression of cadherin 17 in pcDNA3 vector-transfected MDA-MB-231-pcDNA3 and EPHB6-transfected MDA-MB231-EPHB6 cell lines. MDA-MB-231-pcDNA3 (panels A, B and C) and MDA-MB-231-EPHB6 (panels D, E and F) cell lines were grown on glass slides, fixed and incubated with fluorescence-tagged cadherin 17 antibody and 4’, 6’ diamino-2-phenylindole (DAPI). The stained cells were imaged on a Nikon A1 3-color confocal microscope using a ×100 objective. Panels A and D show cadherin 17 expression, panels B and E indicate DAPI-stained nuclei, and panels C and F are merged images. All images were taken at the same magnification as indicated by the line scale at the bottom of each panel.
Activated ERK alters WNT signaling in MDA-MB-231 cells. To investigate the downstream effects of signals transduced by EPHB6 receptor, we examined GSK3β as an intracellular target protein for phosphorylation. The levels of phoshorylated (serine 21/9) GSK3β were significantly higher in MDA-MB231-pcDNA3 cells compared to EPHB6-transfected MDA-MB-231 cells (Figure 4A and B). A concomitant decrease in the amount of β-catenin was also observed in MDA-MB-231-EPHB6 cells (Figure 4C and D). EPHB6 induces re-organization of actin cytoskeleton. EPHB6-mediated suppression of invasiveness of MDA-MB-231 cells prompted us to examine actin organization in EPHB6-transfected MDA-MB-231 cells. MDA-MB-231-EPHB6 and MDA-MB-231-pcDNA3 cell lines were grown as described in the Materials and Methods, and stained with FITC-conjugated phalloidin. The comparison of the actin cytoskeleton indicated morphological as well as size alterations in EPHB6-transfected cells. The pronounced contours of the cytoskeleton in MDA-MB-231-EPHB6 cells are clearly evident in Figure 5.
Discussion
EPH receptors are involved in regulating cell development and differentiation (41, 42), and aberrant expression of these molecules has been also associated with a variety of human cancers (43, 44). We have previously reported transcriptional silencing of kinase-deficient EPHB6 receptor in invasive breast carcinoma cell lines (45), and demonstrated EPHB6-mediated phenotypic changes in these cells (11). In light of our unpublished observations on differential expression of various cadherin transcripts in breast carcinoma cell lines and in order to characterize the functional relationship of EPHB6 with cadherin family of cell adhesion molecules, we investigated the abundance of cadherin 17 transcript in a variety of breast cell lines. Although these cell lines were distinguishable by the expression of a combination of various cadherin transcripts, cadherin 17 was uniquely present in MDA-MB-231 cells. Combined with our earlier results (10, 11), MDA-MB-231 cells can thus be recognized by the presence of cadherin 17 and the absence of EPHB6. These two observations set the basis for investigating any possible cross-talk between EPHB6 and cadherin 17 and to characterize EPHB6-modulated changes in the abundance of some intracellular signaling proteins. Herein we showed that EPHB6 modulates the expression of cadherin 17, and the phenotypic changes in EPHB6-expressing MDA-MB-231 cells are mediated by MAPK and WNT signaling pathways. To our knowledge, this is the first report showing expression of liver-intestine cadherin in a breast carcinoma cell line and correlating its expression with EPHB6 levels. Based on the intracellular proteins linked to EPHB6 signaling in MDA-MB-231 cells (3) and a report suggesting expression of EPHB6 in breast tumors (46), it is reasonable to explain these results as context-dependent consequences of EPHB6 in normal and tumor cells.
Mitogen-activated protein kinase (MEK2) and extracellular signal-regulated kinase (ERK) expression in EPHB6-transfected MDA-MB-231-EPHB6 and pcDNA3 vector-transfected MDA-MB231-pcDNA3 cell lines. A: Total protein was isolated from MDA-MB-231-pcDNA3 and MDA-MB-231-EPHB6 cell lines, separated by electrophoresis, probed with antibody against MEK2 or glyceraldehyde 3-phosphate dehydrogenase (GAPDH) and the antigen-antibody complex detected by chemiluminescence. The arrow indicates the protein corresponding to 45 kDa MEK2 in MDA-MB-231-pcDNA3 (lane 1) and MDA-MB231-EPHB6 (lane 2) cells. The bottom portion shows the band corresponding to GAPDH. B: The histogram shows the ratios of MEK2 and GAPDH proteins in MDA-MB-231-pcDNA3 and MDA-MB-231-EPHB6 cell lines. The results are the mean of three experiments, and error bars indicate the standard deviation. C: Protein blots were probed with antibodies indicated as above. The arrows indicate the bands corresponding to phospho-(p)-ERK (Thr202/Tyr204), GAPDH and total (t)-ERK in MDA-MB-231-pcDNA3 (lane 1) and MDA-MB-231-EPHB6 (lane 2) cells. D: The intensity of bands in panel C was determined by densitometry and ratios of p-ERK and GAPDH proteins in MDA-MB-231-pcDNA3 and MDA-MB-231-EPHB6 cells are shown in the histogram. Experiments were repeated three times and each experiment contained three lanes for each sample. Error bars indicate standard deviation.
The phosphorylation-dependent signal transduction by the kinase-deficient EPHB6 receptor can be interpreted by the ability of other receptors of this family to dimerize with and cross-phosphorylate EPHB6. In fact, it has been shown that EPHB6 can be phosphorylated upon stimulation of cells with EPHB2 (47), and its phosphorylation has also been observed in a complex with EPHB1. Our recent results demonstrate the interaction of EPHB6 with kinase-sufficient receptors EPHB2 and EPHA2 (48). Both receptors have been implicated in a variety of cancer types (26, 34, 49-52). The above observations provide a clear rationale for the altered levels of phosphorylated intracellular proteins in response to altered expression of a kinase-deficient receptor. These changes in intracellular proteins are likely to have arisen by the binding of kinases or regulator proteins to phosphorylated amino acid residues on EPHB6 receptor. Although detailed mechanisms of EPHB6 effects have not been worked out, it has been shown to activate MAPK pathway in lung adenocarcinoma (53).
The levels of phosphorylated glycogen synthase kinase 3 beta (p-GSK3β) and β-catenin in pcDNA3 vector transfected MDA-MB-pcDNA3 and EPHB6-transfected MDA-MB-231-EPHB6 cell lines. A: Total protein was isolated from MDA-MB-231-pcDNA3 and MDA-MB-231-EPHB6 cells, separated by electrophoresis, probed with an antibody against GSK3β or glyceraldehyde 3-phosphate dehydrogenase (GAPDH) as indicated, and the antigen-antibody complexes were detected by chemiluminescence. The arrows indicate the bands corresponding to pGSK3β and GAPDH in MDA-MB-231-pcDNA3 (lane 1) and MDA-MB-231-EPHB6 (lane 2) cells. B: The intensity of bands in panel A was determined by densitometry and ratios of pGSK3β and GAPDH proteins in MDA-MB-231-pcDNA3 and MDA-MB-231-EPHB6 cells are shown in the histogram. Experiments were repeated three times and each experiment contained three lanes for each sample. Error bars indicate standard deviation. C: Protein blots were probed with antibody against β-catenin or GAPDH as indicated, and the antigen-antibody complexes were detected by chemiluminescence. The arrows indicate the bands corresponding to β-catenin and GAPDH in MDA-MB-231-pcDNA3 (lane 1) and MDA-MB-231-EPHB6 (lane 2) cells. D: The intensity of bands in panel C was determined by densitometry and the ratios of β-catenin and GAPDH proteins in MDA-MB-231-pcDNA3 and MDA-MB-231-EPHB6 cells are shown in the histogram. Experiments were repeated three times and each experiment contained three lanes for each sample. Error bars indicate standard deviation.
To explain the EPHB6-mediated invasiveness phenotype of MDA-MB-231 cells, we reasoned that EPHB6 receptor may affect adhesion between two cells. Thus, the expression profiling of cell adhesion molecules appeared to be a logical choice. Based on the unique expression of cadherin 17 in MDA-MB-231 cells, we investigated the effect of EPHB6 on the abundance of cadherin 17. As described in the results section, the amount of cadherin 17 was significantly lower in cells stably transfected with EPHB6. The decrease in cadherin 17 levels in EPHB6-transfected cells strengthens the argument that cadherin 17 is oncogenic and EPHB6 likely suppresses invasiveness of MDA-MB-231 cells. These observations are supported by the reports demonstrating cadherin 17-mediated transformation of premalignant liver progenitor cells to liver carcinomas in mice (54), and its altered expression in colorectal (22, 55), pancreatic (56) and gastric (57-59) cancer. Thus breast carcinoma can be added to the growing list of human cancer types that show aberrant expression of cadherin 17. It warrants mention that aberrant expression of cadherin 17, a cell surface molecule, is associated with tumor cell metastasis (19, 55, 57, 60, 61), and thus is indicative of its involvement in cell adhesion. The altered cellular localization of cadherin 17 in EPHB6-transfected cells suggests that cadherin 17 may play a dual role in adhesion and cell signaling. Based on these observations, we postulate that cadherin 17 modulates the activation of EPHB6, and a reciprocal association likely exists between the amounts of these two proteins. Such a linkage is supported by the observation that EPH and cadherin interaction facilitates the recruitment of a disintegrin and metalloproteinase domain-containing protein 10 (ADAM10) protease at the plasma membrane for dampening signal transduction (27). Our observations, therefore, suggest a novel dimension to the regulation of EPH receptors and cadherins.
Staining of pcDNA3 vector-transfected MDA-MB-231-pcDNA3 and EPHB6-transfected MDA-MB-EPHB6 cells with fluorescien isothiocyanateFITC-phalloidin. MDA-MB-231-pcDNA3 (panels A, B and C) and MDA-MB-231-EPHB6 (panels D, E and F) cells were grown on glass slides, fixed and incubated with fluorescence-tagged phalloidin and 4’, 6’ diamino-2-phenylindole (DAPI). The stained cells were imaged on a Nikon A1 3-color confocal microscope using a ×100 objective. Panels A and D show phalloidin signal, panels B and E indicate DAPI-stained nuclei, and panels C and F are merged images. All images were taken at the same magnification as indicated by the line scale at the bottom of each panel.
The increased amount of MEK2 and the 42 kDa isoform of phospho-ERK (p42) in MDA-MB-231-EPHB6 cells clearly demonstrates EPHB6-mediated modulation of MAPK pathway in breast carcinoma cells. These results are supported by the observations made in lung adenocarcinoma cells (53). The signals received by EPHB6 appear to be relayed by activation of MAPK pathway. We have therefore explored the downstream targets of MAPK pathway mediated by EPHB6 receptor. Based on the reports that down-regulation of cadherin 17 inactivates WNT signaling and inhibits tumor growth (54), we reasoned GSK3β and β-catenin as likely downstream targets of EPHB6 receptor.
After establishing an inverse relationship between EPHB6 and cadherin expression, we investigated the abundance of phospho-GSK3β and β-catenin in MDA-MB-231 cells transfected with EPHB6. The levels of both phospho-GSK3β and β-catenin were lower in EPHB6-expressing cells, thus suggesting a collective decrease in the amounts of cadherin 17, phospho-GSK3β and β-catenin. However, the cause and effect linkage between these proteins remains unclear. GSK3β is known to be constitutively active in non-proliferating cells (62, 63), and its activity is modulated by phosphorylation of tyrosine residue 216 and serine residue 9 (64, 65). GSK3β is also known to suppress as well as promote tumor growth, and these activities are related to its phosphorylation status. While the kinase-inactive enzyme promotes mammary tumorigenesis (66), a GSK3 inhibitor caused cell death through cyclin D1 depletion in breast cancer cells (67). The effects of GSK inhibitor have been attributed to an increase in the inactivating phosphorylation of GSK3α at Ser 21 and GSK3β at Ser 9, and a decrease in the activating phosphorylation of GSK3β at Tyr216 (67). The overexpression of GSK3β has been reported in human ovarian, colon and pancreatic carcinoma (68, 69), and GSK3β has also been linked to cell survival and proliferation in ovarian cancer cells (70).
Our previous observations indicating suppression of invasiveness of EPHB6-transfected MDA-MB-231 cells can be explained by implicating cytoskeleton rearrangements, modulation of intracellular proteins and altered transcription of genes. Cytoskeletal arrangements play a central role in migration and invasiveness of tumor cells (71, 72). In line with the involvement of the cytoskeleton in these processes, we observed distinct and extensive cell protrusions in EPHB6-transfected MDA-MB-231 cells. Such protrusions were particularly visible in cells sharing close proximity, and thus indicated EPHB6 mediated interactions between cells. These observations have further clarified the likely role of EPHB6 receptor in modulating the invasiveness of MDA-MB-231 cells. It is noteworthy that effects of EPHB3 on tumor suppression and cell adhesion in colon cancer have also been attributed to cytoskeleton rearrangement (73).
We believe that the tumor-suppressor activities of EPHB6 are mediated by dampening tumor-promoting activities of GSK3β and β-catenin. However, our data are insufficient to suggest a mechanism for altered phosphorylation and decreased abundance of GSK3β and β-catenin in EPHB6-transfected MDA-MB-231 cells. Our data are in agreement with biological activities of intracellular proteins that appear to be modulated by differential stimulation of EPHB6 receptor. The phenotypic changes in MDA-MB-231 breast carcinoma cell lines presented here and EPHB6 expression observed in breast tumors (46) suggest that EPHB6-mediated alterations are context dependent. Such context dependence of specific receptors and regulatory molecules can, in large part, be attributed to redundancy of receptors, as well as intracellular signaling proteins.
In conclusion, we presented the first report describing the expression of cadherin 17, a liver-intestine cadherin, in a breast carcinoma cell line. We have also demonstrated modulation of its expression by EPHB6 receptor. Furthermore, our results have allowed us to suggest a biological association of EPHB6 receptor with intracellular signaling proteins that mediate MAPK and WNT signaling pathways in normal and carcinoma cells. These results provide adequate basis for predicting an association of cadherin 17 with a subset of invasive breast tumors. Such an association of cadherin 17 with breast tumors, if confirmed, can potentially be applied to diagnostic and therapeutic purposes.
- Received June 26, 2014.
- Revision received July 17, 2014.
- Accepted July 18, 2014.
- Copyright © 2014 The Author(s). Published by the International Institute of Anticancer Research.
