EphA2 overexpression has been reported in many cancers and is believed to play an important role in tumor metastasis and angiogenesis. We show that the activated epidermal growth factor receptor (EGFR) and the cancer-specific constitutively active EGFR type III deletion mutant (EGFRvIII) induce the expression of EphA2 in mammalian cell lines, including the human cancer cell lines A431 and HN5. The regulation is partially dependent on downstream activation of mitogen-activated protein kinase/extracellular signal–regulated kinase kinase and is a direct effect on the EphA2 promoter. Furthermore, EGFR and EphA2 both localize to the plasma membrane and EphA2 coimmunoprecipitates with activated EGFR and EGFRvIII. Ligand activation of EphA2 and EphA2 knockdown by small interfering RNA inhibit EGF-induced cell motility of EGFR-overexpressing human cancer cells, indicating a functional role of EphA2 in EGFR-expressing cancer cells. (Mol Cancer Res 2007;5(3):283–93)

The epidermal growth factor (EGF) and its receptor (EGFR) constitute a well-characterized receptor-ligand system that plays an important role in the regulation of cell growth, proliferation, survival, and motility (1-4). EGFR is involved in oncogenic transformation, which can be caused by receptor overexpression, autocrine ligand loops, gene amplification, or activating mutations such as the EGFR type III deletion mutant (EGFRvIII; refs. 5-8).

In a global search for EGFR and EGFRvIII regulated genes, we recently found the mRNA level of the receptor tyrosine kinase EphA2 to be up-regulated by ligand-activated EGFR and the constitutively active variant EGFRvIII (9).

The cDNA sequence of EphA2 (previously known as epithelial cell kinase) was first described by Lindberg and Hunter in 1990 (10). EphA2 belongs to the Eph receptor family, which is the largest subfamily of receptor tyrosine kinases (11). Based on similarities in the extracellular domain sequences, the Eph receptors are divided into two classes, EphA and EphB. The Eph receptors bind cell membrane–anchored ligands, known as ephrins, which are divided into two classes, ephrin-A and ephrin-B. The ephrin-A class are associated to the membrane by a glycosylphosphatidylinositol anchor whereas the ephrin-B class are transmembrane proteins (12). Generally, EphA receptors bind to ephrin-A ligands and EphB receptors to ephrin-B ligands, although a few exceptions exist (13).

Interaction between an Eph receptor and its ligand results in heterodimers that, by clustering, form heterotetramers (12), leading to transphosphorylation of tyrosine residues of the juxtamembrane domain and to activation of the tyrosine kinase (14, 15).

The Eph family of receptors and their ligands are expressed in adult human tissues and in the developing nervous system (16). These receptor-ligand systems are involved in cell-cell repulsion and attraction and play a role in tissue patterning, neuronal targeting, and angiogenesis during embryonic development (17-20). In addition, EphA2 overexpression has been reported in many cancers including those of the breast, lung, esophagus, stomach, cervix, ovary, colon, prostate, kidney, and skin (21-29). Overexpression of EphA2 has been correlated with increased metastatic potential and decreased survival (22-24, 27, 29). Furthermore, increasing evidence suggests that EphA2 is involved in tumor angiogenesis (26, 30, 31).

Little is known about the regulation of EphA2 expression. However, members of the p53 family of transcription factors have been shown to increase EphA2 transcription (32). High levels of EphA2 have also been observed in cells with overactive RAS signaling (33-35).

Understanding the regulation and function of EphA2 in cancer cells may lead to therapeutics for several types of aggressive cancers. Therefore, we are interested in the mechanisms that control the regulation and the function of EphA2 in cancer cells. The aim of this study was to investigate the regulation of EphA2 by ligand-activated EGFR and by the constitutively active EGFRvIII. Furthermore, we wanted to investigate the significance of EphA2 up-regulation in EGFR-expressing cells by analyzing the effect of ligand stimulation on EphA2 localization and EGF-induced cancer cell migration.

We show that activated EGFR and EGFRvIII rapidly stimulate EphA2 expression in several mammalian cell lines including the human cancer cell lines A431 and HN5, which both naturally overexpress EGFR. The stimulation of EphA2 expression is dependent on the tyrosine kinase activity of EGFR/EGFRvIII and partially on the activation of mitogen-activated protein kinase/extracellular signal–regulated kinase kinase (MEK). Our results further show that EphA2 and EGFR localize to the plasma membrane and that ligand stimulation of EphA2 or EphA2 knockdown by small interfering RNA (siRNA) affects the EGF-induced cell motility by inhibiting cell migration and modulating the migration pattern to a less uniform and more scattered type. These results indicate that EphA2 could play a significant role in regulating EGF-induced migration of human cancer cells.

Activation of EGFR Increases EphA2 mRNA and Protein Levels

Recently, we have investigated changes in the transcriptome induced by ligand activation of EGFR or the constitutively active mutant EGFRvIII in mouse fibroblast cell lines expressing either wild-type EGFR (NR6wtEGFR) or EGFRvIII (NR6M) using Affymetrix oligonucleotide arrays (9). One of the genes, whose expression was found to be induced by ligand-activated EGFR and by EGFRvIII, encodes the tyrosine kinase receptor EphA2. To confirm and extend these results in human cancer cells, time profiles of the EGFR-induced expression of EphA2 were investigated in the human head and neck carcinoma cell line HN5 and in the human epidermoid carcinoma cell line A431, both overexpressing EGFR. Serum-starved cells were stimulated with EGF for up to 24 h and processed for quantitative real-time reverse transcription-PCR (Fig. 1A). In both cell lines, a maximum increase in EphA2 mRNA levels was observed after 2 h of EGF stimulation. A 5-fold increase in EphA2 mRNA level was detected in the A431 cell line whereas a 9-fold increase was seen in HN5 cells (Fig. 1A).

FIGURE 1.

A. Serum-starved A431 and HN5 cells were stimulated with 10 nmol/L EGF for the indicated periods. EphA2 mRNA levels were detected in Trizol-extracted RNA samples by quantitative real-time reverse transcription-PCR analyses. Points, average of three to four independent experiments; bars, SE. B. Western blot analysis showing levels of EphA2 in serum-starved NR6, NR6wtEGFR, A431, and HN5 cells stimulated with 10 nmol/L EGF for indicated periods and in NR6M cells inhibited with 5 μmol/L AG1478 for 48 h and then released from inhibition at indicated periods. Tubulin levels confirm equal protein loading. Representative of four independent experiments. C. Relative increases in EphA2 protein levels quantified by measuring band intensities detected by Western blot analysis, in which a representative experiment of four is shown in B. Bars, SE.

FIGURE 1.

A. Serum-starved A431 and HN5 cells were stimulated with 10 nmol/L EGF for the indicated periods. EphA2 mRNA levels were detected in Trizol-extracted RNA samples by quantitative real-time reverse transcription-PCR analyses. Points, average of three to four independent experiments; bars, SE. B. Western blot analysis showing levels of EphA2 in serum-starved NR6, NR6wtEGFR, A431, and HN5 cells stimulated with 10 nmol/L EGF for indicated periods and in NR6M cells inhibited with 5 μmol/L AG1478 for 48 h and then released from inhibition at indicated periods. Tubulin levels confirm equal protein loading. Representative of four independent experiments. C. Relative increases in EphA2 protein levels quantified by measuring band intensities detected by Western blot analysis, in which a representative experiment of four is shown in B. Bars, SE.

Close modal

To verify that EphA2 protein levels are induced by EGFR activation, serum-starved A431, HN5, NR6, and NR6wtEGFR cells were stimulated with EGF for up to 48 h and EphA2 protein levels were analyzed (Fig. 1B). An increase in EphA2 protein levels was detected in the EGFR-expressing cell lines after 6 h of EGF stimulation and the level remained high during the 48-h period. A similar increase in EphA2 protein levels was seen when releasing the cell line expressing constitutive activated EGFRvIII (NR6M) from AG1478 inhibition, thus allowing receptor autophosphorylation (Fig. 1B).

Quantification of the induced EphA2 protein levels in the EGFR-expressing cell lines showed that EGF stimulation increased the protein levels of EphA2 three to five times with a doubling time ranging from 3.0 to 4.3 h (Fig. 1C). Similarly, releasing the NR6M cell line from AG1478 inhibition increased the protein levels of EphA2 with a doubling time of 3.9 h (Fig. 1C). EGF stimulation of the parental NR6 cell line, which lacks endogenous EGFR expression, had no effect on the level of EphA2 (Fig. 1B and C). These results show that EGF stimulation is an important inducer of EphA2 mRNA and protein levels in mammalian cell lines expressing EGFR including two human cancer cell lines. Likewise, activation of EGFRvIII similarly induces EphA2 expression.

The EGFR-Stimulated Increase in EphA2 Protein Levels Is Dependent on MEK Activity

A panel of small molecular weight inhibitors were used to identify which signaling pathways downstream of ligand-activated EGFR might contribute to the EGFR-induced expression of EphA2. The EGFR-specific inhibitors AG1478 and gefitinib clearly decreased the EGFR- and EGFRvIII-induced expression of EphA2 in the NR6wtEGFR, NR6M, and HN5 cell lines (Fig. 2A). However, AG1478 had no effect on the EphA2 protein level in A431 cells. Detection of EGFR tyrosine phosphorylation (Y1173) showed that AG1478 treatment had no effect on EGFR phosphorylation in A431 cells (not shown). The MEK-specific inhibitors U0126 and PD98059 had partial inhibitory effects on the EGF-induced increase in EphA2 level in the A431, NR6wtEGFR, and NR6M cell lines and a strong inhibitory effect in the HN5 cell line. Inhibitors of phosphatidylinositol 3-kinase (wortmannin and LY294002), phospholipase Cγ (U73122), p38 (SB203580), Janus-activated kinase-2 (AG490), and Src (PP2 and SU6656) had no effect on the EGFR/EGFRvIII–induced expression of EphA2 (not shown). These results indicate that the EGFR- and EGFRvIII-induced expression of EphA2 is dependent on MEK activity; however, the degree of dependence seems to be cell type specific.

FIGURE 2.

A. Western blot analysis showing EphA2 protein levels in serum-starved NR6wtEGFR, NR6M, A431, and HN5 cells treated with 5 μmol/L AG1478, 2 μmol/L gefitinib, 10 μmol/L U0126, or 50 μmol/L PD98059 and/or with 10 nmol/L EGF for 48 h. Tubulin levels confirm equal protein loading. B. Percent increase in luciferase activity of HN5 cells and A431 cells transiently transfected with −4030-EphA2-Luc and stimulated with 5 nmol/L EGF for 6 h in the presence or absence of 5 μmol/L AG1478 or 1 μmol/L gefitinib. Representative data of two independent experiments done in duplicates and normalized to SV40 promoter activity (internal control) for each condition. Bars, SE. *, P < 0.05, compared with unstimulated cells.

FIGURE 2.

A. Western blot analysis showing EphA2 protein levels in serum-starved NR6wtEGFR, NR6M, A431, and HN5 cells treated with 5 μmol/L AG1478, 2 μmol/L gefitinib, 10 μmol/L U0126, or 50 μmol/L PD98059 and/or with 10 nmol/L EGF for 48 h. Tubulin levels confirm equal protein loading. B. Percent increase in luciferase activity of HN5 cells and A431 cells transiently transfected with −4030-EphA2-Luc and stimulated with 5 nmol/L EGF for 6 h in the presence or absence of 5 μmol/L AG1478 or 1 μmol/L gefitinib. Representative data of two independent experiments done in duplicates and normalized to SV40 promoter activity (internal control) for each condition. Bars, SE. *, P < 0.05, compared with unstimulated cells.

Close modal

EGFR Activation Stimulates EphA2 Promoter Activity

To determine if the EGFR-induced expression of EphA2 was through activation of the EphA2 promoter, a luciferase reporter assay was done. A human EphA2 promoter fragment spanning nucleotides −4030 to +21 relative to the putative EphA2 transcriptional start site, previously cloned into a luciferase reported vector (−4030-EphA2-Luc), was used (36). Initially, HN5 cells were transiently transfected with −4030-EphA2-Luc and stimulated with various concentrations of EGF and in different stimulation periods to determine the optimal conditions for stimulation of EphA2 promoter activity. A concentration of 5 nmol/L EGF and a stimulation period of 6 h were found to give the highest increase in EphA2 promoter activity (not shown) and were used in subsequent experiments. To test if the increase in EphA2 promoter activity was dependent on EGFR kinase activity, HN5 and A431 cells transiently transfected with 4030-EphA2-Luc were stimulated with EGF alone or in combination with AG1478 or gefitinib. EGF stimulation significantly increased the EphA2 promoter activity in both cell lines. However, the response was more pronounced in HN5 cells as compared with A431 cells, most likely due to the high basal activity level in A431 cells as shown by Western blotting (Fig. 2A). The EphA2 promoter activity significantly decreased in response to AG1478 and gefitinib treatment (Fig. 2B). Thus, these results show that EGFR activation induces expression of EphA2 in human cancer cell lines by a stimulatory effect on the EphA2 promoter.

EphA2 Localizes with EGFR at the Cell Membrane

Overexpression of EphA2 has been shown to affect the subcellular localization of EphA2 (37). Therefore, we wished to investigate the effects of ligand stimulation of EGFR or EphA2 on the localization of both receptors. Cells were serum starved and treated with either EGF or soluble EphA2-ligand (ephrin-A1/Fc) for 48 h followed by fixation and indirect staining of EphA2 and EGFR. The cells were investigated by confocal microscopy and images of A431 and HN5 cells revealed that EGFR and EphA2 primarily localized to the cell membrane in untreated cells (Fig. 3A). EGF stimulation increased EphA2 levels in both cell lines, and the receptor seemed to be distributed at the plasma membrane or intracellularly. EGF stimulation induced an almost complete loss of EGFR at the cell membrane of A431 cells, and residual EGFR was found intracellularly (Fig. 3A). A less pronounced response to EGF stimulation on EGFR localization was observed in the HN5 cell line. Ephrin-A1 stimulation resulted in removal of EphA2 from the plasma membrane without affecting EGFR localization. Occasionally, some EphA2 staining was seen in the nucleus (Fig. 3A).

FIGURE 3.

A. Representative confocal images of A431 and HN5 cells stained for EphA2 (green), EGFR (red), and nuclei (blue) incubated for 48 h with 10 nmol/L EGF or 1 μg/mL ephrin-A1/Fc or left unstimulated. Bars, 20 μm. B. Scatter diagrams of pixel intensities of EphA2 and EGFR staining from the images of A431 cells shown in A. Red lines on axis, threshold for background pixels; blue lines, mean intensity of EphA2 or EGFR staining above the background threshold. C and D. Columns, mean staining intensities of EphA2 and EGFR in A431 cells (C) and HN5 cells (D) incubated for 48 h with 10 nmol/L EGF or 1 μg/mL ephrin-A1/Fc or left unstimulated. More than 80 cells in 10 randomly selected areas of each sample were used for measuring the staining intensities in each setting. Bars, 95% confidence interval. *, P < 0.05; **, P < 0.001; ***, P < 0.00001, compared with unstimulated cells.

FIGURE 3.

A. Representative confocal images of A431 and HN5 cells stained for EphA2 (green), EGFR (red), and nuclei (blue) incubated for 48 h with 10 nmol/L EGF or 1 μg/mL ephrin-A1/Fc or left unstimulated. Bars, 20 μm. B. Scatter diagrams of pixel intensities of EphA2 and EGFR staining from the images of A431 cells shown in A. Red lines on axis, threshold for background pixels; blue lines, mean intensity of EphA2 or EGFR staining above the background threshold. C and D. Columns, mean staining intensities of EphA2 and EGFR in A431 cells (C) and HN5 cells (D) incubated for 48 h with 10 nmol/L EGF or 1 μg/mL ephrin-A1/Fc or left unstimulated. More than 80 cells in 10 randomly selected areas of each sample were used for measuring the staining intensities in each setting. Bars, 95% confidence interval. *, P < 0.05; **, P < 0.001; ***, P < 0.00001, compared with unstimulated cells.

Close modal

The pixel intensities of EphA2 and EGFR staining in images of unstimulated, EGF-stimulated, and ephrin-A1–stimulated cells were extracted and processed with confocal microscope associated software. Scatter diagrams with pixel intensities of EGFR staining and EphA2 staining from the images of unstimulated and EGF-stimulated A431 cells seen in Fig. 3A are illustrated in Fig. 3B. By examining the pixel distribution in the scatter diagrams and the mean staining intensities, it is seen that EGF stimulation resulted in increased EphA2 staining and decreased EGFR staining (Fig. 3B). In agreement with this, by calculating the mean staining intensities of EphA2 and EGFR staining from images with low magnification and settings optimal for quantification, the levels of EphA2 staining were found to increase 2.2- and 1.6-fold on EGF stimulation in the A431 and HN5 cell lines, respectively (Fig. 3C and D). The levels of EGFR staining decreased 1.9- and 1.3-fold in the A431 and HN5 cell lines, respectively (Fig. 3C and D). The changes in EGFR and EphA2 staining intensities were statistically significant (P < 0.05-0.00001). Ephrin-A1 stimulation slightly decreased the levels of EphA2 and EGFR in both cell lines, but this was not statistically significant (Fig. 3C and D). Quantification of staining intensities from images acquired with high magnification and settings optimal for visual presentation gave similar results (not shown). These results show that EphA2 and EGFR both localize to the plasma membrane of cancer cells and that EGF stimulation increases the level of EphA2 staining and induces EGFR down-regulation.

Activated EGFR and EGFRvIII Associate with EphA2

Based on the observed EphA2 and EGFR colocalization at the cell membrane, we speculated whether an interaction between EGFR and EphA2 could be detected by coimmunoprecipitation and whether the phosphorylation status of EGFR could influence this association. Coimmunoprecipitation experiments were carried out using NR6, NR6M, NR6wtEGFR, A431, and HN5 cells (Fig. 4A). Serum-starved cells were left untreated or stimulated with EGF for 24 h and EGFR was immunoprecipitated from the cellular lysates. Neither EGFR nor EphA2 could be detected in the EGFR-negative cell line NR6 after immunoprecipitation with EGFR antibody. Stimulation of NR6wtEGFR and A431 cells with EGF resulted in a clear increase in coprecipitated EphA2, whereas EphA2 coprecipitated with EGFRvIII irrespective of EGF stimulation in the NR6M cell line expressing constitutively active EGFRvIII (Fig. 4A). These results indicate that EGFR/EGFRvIII activation is required for EphA2 coimmunoprecipitation with EGFR/EGFRvIII. The level of EGFR phosphorylation in the immunoprecipitated samples was detected by Western blot analysis with antibody against phosphorylated tyrosine (PY-20; not shown). The results showed high phosphorylation level in HN5 cells independent of EGF stimulation, possibly explaining why EphA2 coimmunoprecipitated with EGFR independent of EGF stimulation in this cell line (Fig. 4A).

FIGURE 4.

A. Immunoprecipitation with monoclonal anti-EGFR antibody and cell lysates from serum-starved cells (NR6, NR6wtEGFR, NR6M, A431, and HN5) left untreated or stimulated with 10 nmol/L EGF for 24 h. The immunoprecipitates were subjected to Western blot analysis with antibodies directed against EphA2, EGFR, or EGFRvIII. Results from one of two similar experiments. B. Western blot analysis showing levels of EphA2 and EGFR/EGFRvIII in immunoprecipitated samples using antiphosphotyrosine antibody (PY-20) of serum-starved NR6, NR6wtEGFR, NR6M, A431, and HN5 cells stimulated with 1 μg/mL ephrin-A1/Fc or 10 nmol/L EGF for 15 min.

FIGURE 4.

A. Immunoprecipitation with monoclonal anti-EGFR antibody and cell lysates from serum-starved cells (NR6, NR6wtEGFR, NR6M, A431, and HN5) left untreated or stimulated with 10 nmol/L EGF for 24 h. The immunoprecipitates were subjected to Western blot analysis with antibodies directed against EphA2, EGFR, or EGFRvIII. Results from one of two similar experiments. B. Western blot analysis showing levels of EphA2 and EGFR/EGFRvIII in immunoprecipitated samples using antiphosphotyrosine antibody (PY-20) of serum-starved NR6, NR6wtEGFR, NR6M, A431, and HN5 cells stimulated with 1 μg/mL ephrin-A1/Fc or 10 nmol/L EGF for 15 min.

Close modal

To investigate if the association between EGFR and EphA2 leads to receptor transphosphorylation, cells were stimulated with either EGF or ephrin-A1/Fc. The effect of EGF and ephrin-A1/Fc on either EphA2 or EGFR tyrosine phosphorylation was analyzed by immunoprecipitation of tyrosine-phosphorylated proteins followed by detection of EGFR and EphA2 by Western blot analysis (Fig. 4B).

The results showed that 15 min of EGF stimulation induced phosphorylation of EGFR and that 15 min of ephrin-A1/Fc stimulation induced EphA2 phosphorylation, but either ligand failed to induce tyrosine phosphorylation of the other receptor (Fig. 4B). Preliminary experiments showed that EphA2 and EGFR did not coimmunoprecipitate after EGF stimulation for 15 min. Furthermore, EGF stimulation for 24 h induced EphA2 phosphorylation whereas ephrin-A1/Fc had no effect on EGFR phosphorylation (not shown). Thus, these data suggest that EGFR and EphA2 associate after prolonged EGF stimulation and that the outcome of this association could involve transphosphorylation of EphA2.

EphA2 Activation Modulates EGF-Induced Motility

EGFR activation is known to stimulate cell proliferation and motility (1, 2). Therefore, we inquired if activation of EphA2 could influence EGF-induced proliferation and motility of EGFR-expressing cells. Ligand activation of EphA2 did not have an effect on EGF-induced proliferation (not shown) but inhibited EGF-induced motility (Fig. 5A-D). The motility of HN5 cells was investigated by a wound healing assay (Fig. 5A and B) and a three-dimensional spheroid motility assay (Fig. 5C and D). In both assays, ephrin-A1/Fc stimulation of HN5 cells inhibited EGF-induced motility. Quantification showed that ephrin-A1/Fc decreased EGF-induced wound closure by 41% after 24 h (P < 0.0001) and 30% after 48 h (P < 0.001; Fig. 5B). Intriguingly, we found that ephrin-A1/Fc stimulation of HN5 spheroids not just reduced the EGF-induced migration of HN5 cells out from the spheroids but also altered the motility to a less uniform and more scattered type (Fig. 5C). This effect was not seen in the wound healing assay and suggests that the effect of EphA2 ligand stimulation on cell motility depends on the three-dimensional organization of the cells. To further investigate the role of EphA2 in EGF-induced motility, the effect of EphA2 knockdown by siRNA was investigated by a wound healing assay (Fig. 5E and F). Quantification of the migration distance (wound closure) showed that the EGF-induced wound closure was significantly decreased in EphA2 knockdown cells compared with the negative control cells transfected with sc-siRNA (P < 0.001). Importantly, these results show that knockdown of EphA2 results in reduced EGF-induced migration and indicate that EphA2 receptors, in the absence of ephrin-A1/Fc stimulation, stimulate EGF-induced cell motility.

FIGURE 5.

A. Wound healing assay using confluent serum-starved HN5 cells gently wounded through the central axis. Cells were stimulated with 2 μg/mL ephrin-A1/Fc and/or 1 nmol/L EGF for 48 h. Images were taken at 0, 24, and 48 h. B. Quantification of cell motility by measuring the distance between the invading front of cells in six random selected microscopic fields for each condition and time point. The degree of motility is expressed as percent of wound closure as compared with the zero time point. C. Spheroid motility assay using spheroids of HN5 cells allowed to adhere and migrate in the presence of 2 μg/mL ephrin-A1/Fc and/or 1 nmol/L EGF. Images were taken at 24 and 48 h with 4× and 10× objectives. D. Quantification of cell motility by measuring the distance between the edge of the spheroid and farthest migrated cell at three random selected areas for three spheroids for each condition and time point. The degree of motility is expressed as migrated distance compared with unstimulated cells. E. Wound healing assay using confluent HN5 cells transfected with 25 nmol/L siRNA targeting EphA2 (EphA2-siRNA) or 25 nmol/L negative control siRNA (sc-siRNA) and serum starved overnight. Cells were wounded and left untreated or stimulated with 1 nmol/L EGF for 48 h. Images were taken at 0, 24, and 48 h and the migration distances were quantified as in B. Representative of two experiments. F. Western blot analysis showing EphA2 protein levels in HN5 cell lysates prepared from the cells used in the motility assay shown in E. B, D, and E. Bars, SE. *, P < 0.05; **, P < 0.001; ***, P < 0.0001, compared with the zero time point.

FIGURE 5.

A. Wound healing assay using confluent serum-starved HN5 cells gently wounded through the central axis. Cells were stimulated with 2 μg/mL ephrin-A1/Fc and/or 1 nmol/L EGF for 48 h. Images were taken at 0, 24, and 48 h. B. Quantification of cell motility by measuring the distance between the invading front of cells in six random selected microscopic fields for each condition and time point. The degree of motility is expressed as percent of wound closure as compared with the zero time point. C. Spheroid motility assay using spheroids of HN5 cells allowed to adhere and migrate in the presence of 2 μg/mL ephrin-A1/Fc and/or 1 nmol/L EGF. Images were taken at 24 and 48 h with 4× and 10× objectives. D. Quantification of cell motility by measuring the distance between the edge of the spheroid and farthest migrated cell at three random selected areas for three spheroids for each condition and time point. The degree of motility is expressed as migrated distance compared with unstimulated cells. E. Wound healing assay using confluent HN5 cells transfected with 25 nmol/L siRNA targeting EphA2 (EphA2-siRNA) or 25 nmol/L negative control siRNA (sc-siRNA) and serum starved overnight. Cells were wounded and left untreated or stimulated with 1 nmol/L EGF for 48 h. Images were taken at 0, 24, and 48 h and the migration distances were quantified as in B. Representative of two experiments. F. Western blot analysis showing EphA2 protein levels in HN5 cell lysates prepared from the cells used in the motility assay shown in E. B, D, and E. Bars, SE. *, P < 0.05; **, P < 0.001; ***, P < 0.0001, compared with the zero time point.

Close modal

In the present study, we report that activated EGFR and EGFRvIII induce expression of EphA2 and that this regulation seems to involve downstream activation of the MEK signaling pathway and stimulation of EphA2 promoter activity. Furthermore, we found that EphA2 localizes with EGFR at the plasma membrane and that the two receptors can be coimmunoprecipitated. Our results further show that ligand stimulation of EphA2 or EphA2 knockdown by siRNA inhibits EGF-induced motility of human cancer cells overexpressing EGFR. These observations indicate that EphA2 plays a role in EGFR-mediated cancer cell motility.

Activated EGFR Induces Expression of EphA2

EGF stimulation was found to increase EphA2 mRNA and protein levels in a number of EGFR-expressing cell lines including two human cancer cells lines (A431 and HN5). Similarly, the cancer-specific constitutively active variant EGFRvIII was found to increase EphA2 mRNA and protein levels in a mouse fibroblast cell line stably transfected with EGFRvIII (NR6M). The results were supported by confocal microscopy, which showed that EGF stimulation significantly increased the staining intensity of EphA2 in both A431 and HN5 cells.

The increased EphA2 expression was dependent on EGFR tyrosine kinase activity and partially on MEK activity. These findings are in agreement with recent reports showing increased EphA2 levels in RAS-transformed cells (34, 35, 37). The partial effect of the MEK inhibitors in two of the three EGFR expressive cell lines could be due to incomplete inhibition in these cell lines. Alternatively, it could indicate involvement of other signaling pathways; however, further studies using a panel of molecular inhibitors targeting signaling pathways downstream of ligand-activated EGFR showed no effect on the level of EphA2.

A luciferase reporter assay was done to investigate if the increase in EphA2 expression was through a direct effect on the EphA2 promoter. EGF stimulation of the transfected cells resulted in a significant increase in luciferase activity, indicating that the increase in EphA2 levels mediated by EGFR is through activation of the EphA2 promoter.

EphA2 is frequently overexpressed in human cancers, and our results suggest that activated EGFR and EGFRvIII could contribute to this through activation of the MEK signaling pathway and induction of EphA2 promoter activation.

EphA2 Localizes with EGFR at the Cell Membrane

E-Cadherin has been reported to affect the subcellular localization of EphA2 by directing EphA2 into cell-cell contact (38). In this article, the subcellular localization of EphA2 was investigated in EGFR-expressing human cancer cells by confocal microscopy. In A431 cells, EphA2 was mainly located at cell-cell contacts, whereas in the HN5 cell line, EphA2 was more evenly distributed throughout the cells. The fact that A431 cells express high levels of E-cadherin whereas HN5 cells are devoid of E-cadherin and that E-cadherin has been shown to direct EphA2 into cell-cell contacts (38-40) may explain the difference between the observed distributions of EphA2.

Some EphA2 staining was occasionally seen concentrated in the nucleus. It is established that the intracellular fragment of another receptor tyrosine kinase, ErbB4, can be transported to the nucleus. However, whether full-length receptor tyrosine kinases can be transported to the nucleus is controversial (41). The possible function of EphA2 in the nucleus needs to be investigated further. EphA2 showed an overlapping distribution with EGFR at the cell membrane in both HN5 and A431 cells. The colocalization was less pronounced in EGF-stimulated cells due to EGF-induced EGFR down-regulation. The colocalization of EGFR and EphA2 led us to investigate a possible association between EGFR and EphA2. Immunoprecipitation of EGFR pulled down EphA2 in EGF-stimulated cells and cells expressing EGFRvIII, suggesting that EphA2 associates with activated EGFR. The observation that EGFRvIII associates with EphA2 hints that the association occurs intracellularly because EGFRvIII lacks extracellular domain I and two thirds of domain II (42). However, it remains to be elucidated whether EphA2 interacts directly or indirectly with EGFR. We investigated if association between EGFR and EphA2 could result in receptor transphosphorylation of ligand stimulation. After 15 min of ligand stimulation, neither EphA2 nor EGFR showed detectable effects on phosphorylation of the other receptor. However, further experiments showed that EphA2 transphosphorylation could be detected after a longer period (24 h) of EGF stimulation. Similarly, receptor association was detected after 24 h and not after 15 min of EGF stimulation; therefore, it is possible that association of EphA2 and EGFR results in transphosphorylation. In addition, we investigated cross-talk at the level of ERK because ligand stimulation of EphA2 has been shown to affect ERK phosphorylation (35, 43). Although ligand stimulation of EphA2 inhibited basal ERK phosphorylation, no effect on EGF-induced ERK phosphorylation was detected (not shown).

EphA2 Functions by Modulating EGF-Induced Cell Motility

The ephrin-EphA2 ligand system plays an important role in repulsion and attraction of neurons and endothelial cells during development (20, 44). In cancer, EphA2 overexpression has been correlated with metastasis (45, 46), indicating that EphA2 might be involved in cancer cell motility. EGFR is known to induce cell migration (3), and we therefore speculated if EphA2 could be involved in regulating EGF-induced cancer cell motility.

The results show that ligand stimulation of EphA2 significantly inhibits EGF-induced cell migration in a wound healing assay and in a three-dimensional motility assay (Fig. 5A-D). In correlation, Guo et al. (47) recently reported that EphA2 has tumor suppressive functions in mammalian skin and suggested that EphA2 overexpression in tumors is the result of a compensatory feedback defensive mechanism against malignant transformation. In addition, Pedersen et al. have shown that activated EGFR induces transcription of a panel of genes (IFN-γ responsive genes) whose products have known inhibitory effects on EGFR activity and function, indicative of a negative feedback mechanism abrogating or limiting the tumor-promoting effects of overactivated EGFR (9, 48). Furthermore, Macrae et al. (35) recently showed that RAF activation stimulates EphA2 expression and that ligand-stimulated EphA2 attenuates activation of the mitogen-activated protein kinase pathway thereby forming a negative feedback loop. Moreover, they reported that expressions of EphA2 and its ligands are mutually exclusive in breast cancer cell lines and suggested that the interaction between EphA2 and its ligands occurs between cells of different types. Based on these results, Macrae et al. (35) proposed that ephrin ligands may suppress neighboring EphA2-expressing tumor cells depending on RAS signaling and that escape from this inhibition could be a necessary step in development of some cancers. In conjunction, these observations support the proposed tumor suppressive function of the ephrin-EphA2 ligand system and is depicted in Fig. 6A. However, EphA2 is overexpressed in a broad range of cancers, and results have shown that EphA2 overexpression is able to induce cancer cell transformation (37). The results from this study show that EphA2 knockdown by siRNA inhibits EGF-induced cancer cell migration in a wound healing assay. These results suggest that EphA2 has a ligand-independent stimulatory effect on EGF-induced cancer cell motility and that EphA2 down-regulation results in termination of these effects (Fig. 6B). In support, the results from the confocal microscopy analysis showed that EphA2 ligand stimulation resulted in removal of EphA2 from the plasma membrane. Ligand-induced down-regulation of EphA2 is reported in other studies (49, 50). Furthermore, unlike most other receptor tyrosine kinases, results have shown that EphA2 is constitutively active and does not require ligand-based activation and phosphorylation for enzyme activity (15, 51). Rather, ligand activation seems to control EphA2 turnover in the plasma membrane.

FIGURE 6.

A schematic presentation of two proposed mechanisms based on the presented results. A. EGF stimulation of EGFR induces EphA2 gene transcription. EphA2 and EGFR colocalize and interact on the plasma membrane of the cancer cells, where ligand stimulation (ephrin-A1) of EphA2 leads to inhibition of EGF-induced cancer cell motility. B. EGF stimulation of EGFR induces EphA2 gene transcription. EphA2 and EGFR colocalize and interact on the plasma membrane of the cancer cells, where EphA2 has ligand-independent effects on EGF-induced cancer cell motility. Ligand (ephrin-A1) stimulation of EphA2 leads to receptor internalization, down-regulation, and termination of ligand-independent effects on cell motility.

FIGURE 6.

A schematic presentation of two proposed mechanisms based on the presented results. A. EGF stimulation of EGFR induces EphA2 gene transcription. EphA2 and EGFR colocalize and interact on the plasma membrane of the cancer cells, where ligand stimulation (ephrin-A1) of EphA2 leads to inhibition of EGF-induced cancer cell motility. B. EGF stimulation of EGFR induces EphA2 gene transcription. EphA2 and EGFR colocalize and interact on the plasma membrane of the cancer cells, where EphA2 has ligand-independent effects on EGF-induced cancer cell motility. Ligand (ephrin-A1) stimulation of EphA2 leads to receptor internalization, down-regulation, and termination of ligand-independent effects on cell motility.

Close modal

In summary, our results show that EGFR is an important inducer of EphA2 expression and that this mechanism may contribute to the high frequency of EphA2 expression found in many human cancers of different origin (26). Furthermore, our results suggest that EphA2 may play an important role in EGFR-mediated cell motility. Therefore, these results support the further development of novel anticancer therapeutics targeting EphA2.

Materials

Anti-EphA2 and anti–glyceraldehyde-3-phosphate dehydrogenase antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, CA); polyclonal antibodies to EGFR were from Fitzgerald Industries (Concord, MA); monoclonal antibodies to EGFR (Ab-1) were from Calbiochem (San Diego, CA); monoclonal anti-EGFRvIII antibodies were from Novocastra (Newcastle, United Kingdom); antibodies against phosphorylated and total p44/p42 ERK were from Cell Signaling Technology (Danvers, MA); antiphosphotyrosine (PY-20) monoclonal antibodies were from Zymed Laboratories (San Francisco, CA); and antibody to tubulin was from NeoMarkers (Fremont, CA).

Recombinant human EGF, AG1478, U73122, PP2, and SU6656 were purchased from Calbiochem. LY294002 was purchased from Biosource (Camarillo, CA). U0126 and PD98059 were purchased from Promega (Mannheim, Germany). SB203580 and wortmannin were purchased from Upstate Cell Signaling Solution (Lake Placid, NY). Ephrin-A1/Fc was purchased from R&D Systems Europe Ltd. (Abingdon, United Kingdom). Gefitinib was kindly provided by AstraZeneca A/S (Albertslund, Denmark).

Cell Culture

The cell lines NR6, NR6wtEGFR, and NR6M have previously been described (9). The human epidermoid carcinoma cell line A431 was obtained from the American Type Culture Collection and the head and neck carcinoma cell line HN5 was kindly provided by Dr. Jiri Bartek (Department of Cell Cycle and Cancer, Danish Cancer Society, Copenhagen, Denmark). All cells were maintained in DMEM (Invitrogen, Taastrup, Denmark) supplemented with 10% FCS, 50 units/mL penicillin, and 50 μg/mL streptomycin.

Immunoblot Analyses

For the immunoblot analyses, cells were seeded in DMEM with 10% FCS, allowed to grow for 24 h until 80% confluent, washed once in PBS, and serum starved in DMEM supplemented with 0.5% FCS for 48 h. Serum-starved cells were treated with different inhibitors using the following concentrations: 5 μmol/L AG1478, 1 μmol/L gefitinib, 5 μmol/L LY294002, 50 μmol/L PD98059, 10 μmol/L PP2, 10 μmol/L SB203580, 15 μmol/L SU6656, 20 μmol/L U0126, 100 nmol/L U73122, and 0.1 μmol/L wortmannin. Immunoblot analyses were done as previously described (9). Briefly, 5 μg of whole-cell lysate were resolved by SDS-PAGE and electroblotted onto nitrocellulose membranes. After transfer and blocking in 5% nonfat milk, incubation with primary antibody was done overnight at 4°C and secondary antibody staining was done for 1 h at room temperature. The enhanced chemiluminescence detection method was used for all Western blot experiments. Band intensities were determined using the public domain Java image-processing program ImageJ and analyzed using GraphPad Prism with nonlinear regression.

Quantitative Real-time Reverse Transcription-PCR

Total RNA was extracted from the cells using Trizol reagent (Invitrogen, Taastrup, Denmark). The first-strand cDNA synthesis and the real-time reverse transcription-PCR reaction were carried out using the SuperScript III Platinum Two-Step qRT-PCR Kit with SYBR Green (Invitrogen) with the following EphA2 primer sets: EPHA2F, 5′-CACCTCTGCCAGCGACGTG-3′; EPHA2R, 5′-CCGCCATGAAGTGCTCCGTA-3′ (TAG Copenhagen, Denmark). The housekeeping primers (RPL13A-R and RPL13A-F) were from DNA Technology (Aarhus, Denmark). The PCR cycling conditions used were as follows: 95°C for 10 min, 95°C for 20 s, 56.7°C for 20 s, and 72°C for 20 s for 40 cycles. The EphA2 PCR products were normalized to RPL13A PCR products, the level of EphA2 mRNA was determined, and the relative amounts of EphA2 mRNA were calculated from the standard curve obtained using known amounts of cDNA. The experiment was repeated twice and each sample was assayed in duplicates.

Immunoprecipitation

Cell lysates were precleared and equal amounts of protein (1,000 μg) were immunoprecipitated with 5 μg of mouse anti-EGFR (Ab-1; Calbiochem) or 5 μg of mouse antiphosphotyrosine (PY-20; Zymed Laboratories). The immunocomplexes were precipitated with protein G agarose beads (Upstate Cell Signaling Solution), collected by centrifugation, and washed thrice with ice-cold PBS. Antibody control was prepared by incubating 1 mL of precleared radioimmunoprecipitation assay buffer with 5 μg of mouse anti-EGFR antibody followed by incubation with antibody and then protein G agarose beads. Immunoprecipitated/coimmunoprecipitated proteins were analyzed by Western blot analyses.

Luciferase Reporter Assay

The plasmid −4,030-EphA2-Luc contains a human EphA2 promoter fragment spanning nucleotides −4,030 to +21 relative to the putative EphA2 transcriptional start site in front of a luciferase reporter gene and was kindly provided by David M. Cohen (Oregon Health and Science University and the Portland Veterans Affairs Medical Center, Portland, OR; ref. 36).

A431 and HN5 cells, seeded in 24-well plates (100,000 per well) in DMEM supplemented with 10% FCS and cultured overnight, were transiently transfected with the −4,030-EphA2-Luc plasmid or with a control plasmid containing the SV40 promoter using LipofectAMINE 2000 reagent (Invitrogen). Twenty-four hours after the transfection, cells were switched to low-serum medium (0.5%). After 2 h of starvation, cells were stimulated with 5 nmol/L EGF alone or in combinations with 5 μmol/L AG1478 or 1 μmol/L gefitinib for 6 h. Cells were then lysed by adding 100 μL/well of Passive Lysis Buffer (Promega) and luciferase activity was measured using the Luciferase Assay System (Promega) as previously described (52). Differences in luciferase activity were tested using two-sided t test for comparing means.

Immunofluorescence Confocal Microscopy

Serum-starved A431 and HN5 cells plated on CC2-coated chamber slides (Lab-Tek; Nalgene Nunc International, Rochester, NY) were incubated with 10 nmol/L EGF, 1 μg/mL ephrin-A1/Fc (secreted forms of the EphA2 ligand ephrin-A1 attached to Fc portion of human immunoglobulin G1 heavy chain), or left unstimulated for 48 h. Cells were fixated in 2% paraformaldehyde for 20 min. Nonspecific binding was blocked and cells were permeabilized with 5% goat serum (Dako, Glostrup, Denmark) and 0.2% saponin in PBS (blocking buffer) for 20 min followed by incubation with anti-EphA2 rabbit antibody (Santa Cruz Biotechnology) in blocking buffer for 1 h, wash with PBS, and incubation with antirabbit antibody conjugated to Alexa Fluor 488 (Molecular Probes, Eugene, OR). After washing with PBS and incubation with blocking buffer for 20 min, cells were incubated with anti-EGFR mouse antibody (Ab-1; Calbiochem) followed by wash with PBS and incubation with antibody conjugated to Alexa Fluor 568 and To-Pro-3 (Molecular Probes). Cells were washed with PBS and mounted with Fluoromount G (Southern Biotechnology Associates, Birmingham, AL). Images were acquired with a LSM 510 Meta confocal microscope (Carl Zeiss, Jena, Germany) with 20× or 40× objectives and processed using the LSM software v. 3.2 (Carl Zeiss) and Adobe Photoshop 7.0.

Images of randomly selected areas from each sample were acquired with one of two settings of scanning parameters (low magnification: 20× objective with 3-μm confocal slide, or high magnification: 40× objective with 1-μm confocal slide). Using the LSM software v. 3.2, the mean intensity of EphA2 and EGFR staining was calculated for both settings. Background pixels were discarded in the mean intensity calculation by setting a threshold corresponding to intensity in a background area with no cells. Difference between images of unstimulated and stimulated cells was tested using two-sided t test for comparing means. Five to 10 images with 41 to 386 cells from each setting were used.

Wound Healing Assay

HN5 cells were seeded in triplicate in fibronectin-coated six-well plates or T25 cell culture flasks and allowed to grow until 70% confluence. Cells seeded in T25 flasks were transfected using 2 μL of LipofectAMINE 2000 reagent (Invitrogen), 4,000 μL of OptiMEM I Reduced Serum Medium (Invitrogen), and 25 nmol/L siRNA targeting EphA2 (Hs_EphA2_8 and Hs_EphA2_6, Qiagen, Ballerup, Denmark), 25 nmol/L stealth RNAi negative control (Invitrogen), or 25 nmol/L BlOCK-iT Fluorescent Oligo siRNA (Invitrogen). Transfection complex was removed from the cells after 4 h and the cells were grown overnight in low-serum media. Fluorescence was detected to analyze transfection efficiency and EphA2 protein levels were detected by Western blot analysis to verify EphA2 knockdown. The cell layer was gently “wounded” through the central axis of the plate using a pipette tip. Loose cells were washed off and the remaining cells treated with 1 nmol/L EGF and/or 2 μg/mL ephrin-A1/Fc in low-serum media. Migration of cells into the wound was observed at three preselected time points (0, 24, and 48 h) in six random selected microscopic fields for each condition and time point. Images were acquired with a Nikon Eclipse TS 100 phase-contrast microscope with 10× objectives and processed using Adobe Photoshop 7.0. Differences in cell migration distances were tested using two-sided t test for comparing means.

Spheroid Motility Assay

Spheroids of HN5 cells were generated using the hanging drop method as described by Del Duca et al. (53). Briefly, 50,000 cells were placed in droplets of 20 μL on the lids of 100-mm dishes, which were then inverted over dishes containing DMEM. The droplets were then incubated overnight. The cellular aggregates were harvested using a Pasteur pipette, transferred to 100-mm dishes coated with 0.75% agarose, filled with DMEM, and incubated overnight for spheroid formation. A minimum of five spheroids for each condition were then transferred to fibronectin-coated six-well plates and cells allowed to adhere and migrate for 48 h in the presence of 1 nmol/L EGF and/or 2 μg/mL ephrin-A1/Fc. Photographs were taken at 24 and 48 h with a Nikon Eclipse TS 100 phase-contrast microscope with 4× and 10× objectives and processed using Adobe Photoshop 7.0. Three to nine images from each condition and time point were used to quantify the migration distances, which were tested using two-sided t test for comparing means.

Grant support: Danish Cancer Society, the Danish Medical Research Council, the University of Copenhagen, the Novo Nordic Foundation, the John and Birthe Meyer Foundation, and the Danish Cancer Research Foundation.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1
Pedersen MW, Tkach V, Pedersen N, Berezin V, Poulsen HS. Expression of a naturally occurring constitutively active variant of the epidermal growth factor receptor in mouse fibroblasts increases motility.
Int J Cancer
2004
;
108
:
643
–53.
2
Ullrich A, Coussens L, Hayflick JS, et al. Human epidermal growth factor receptor cDNA sequence and aberrant expression of the amplified gene in A431 epidermoid carcinoma cells.
Nature
1984
;
309
:
418
–25.
3
Wells A, Gupta K, Chang P, Swindle S, Glading A, Shiraha H. Epidermal growth factor receptor-mediated motility in fibroblasts.
Microsc Res Tech
1998
;
43
:
395
–411.
4
Di Fiore PP, Pierce JH, Fleming TP, et al. Overexpression of the human EGF receptor confers an EGF-dependent transformed phenotype to NIH 3T3 cells.
Cell
1987
;
51
:
1063
–70.
5
Chaffanet M, Chauvin C, Laine M, et al. EGF receptor amplification and expression in human brain tumours.
Eur J Cancer
1992
;
28
:
11
–7.
6
Kuan CT, Wikstrand CJ, Bigner DD. EGF mutant receptor vIII as a molecular target in cancer therapy.
Endocr Relat Cancer
2001
;
8
:
83
–96.
7
Tang P, Steck PA, Yung WK. The autocrine loop of TGF-α/EGFR and brain tumors.
J Neurooncol
1997
;
35
:
303
–14.
8
Moscatello DK, Montgomery RB, Sundareshan P, McDanel H, Wong MY, Wong AJ. Transformational and altered signal transduction by a naturally occurring mutant EGF receptor.
Oncogene
1996
;
13
:
85
–96.
9
Pedersen MW, Pedersen N, Damstrup L, et al. Analysis of the epidermal growth factor receptor specific transcriptome: effect of receptor expression level and an activating mutation.
J Cell Biochem
2005
;
96
:
412
–27.
10
Lindberg RA, Hunter T. cDNA cloning and characterization of eck, an epithelial cell receptor protein-tyrosine kinase in the eph/elk family of protein kinases.
Mol Cell Biol
1990
;
10
:
6316
–24.
11
Aasheim HC, Patzke S, Hjorthaug HS, Finne EF. Characterization of a novel Eph receptor tyrosine kinase, EphA10, expressed in testis.
Biochim Biophys Acta
2005
;
1723
:
1
–7.
12
Himanen JP, Rajashankar KR, Lackmann M, et al. Crystal structure of an Eph receptor-ephrin complex.
Nature
2001
;
414
:
933
–8.
13
Himanen JP, Chumley MJ, Lackmann M, et al. Repelling class discrimination: ephrin-A5 binds to and activates EphB2 receptor signaling.
Nat Neurosci
2004
;
7
:
501
–9.
14
Nowakowski J, Cronin CN, McRee DE, et al. Structures of the cancer-related Aurora-A, FAK, and EphA2 protein kinases from nanovolume crystallography.
Structure (Camb)
2002
;
10
:
1659
–67.
15
Wybenga-Groot LE, Baskin B, Ong SH, Tong J, Pawson T, Sicheri F. Structural basis for autoinhibition of the Ephb2 receptor tyrosine kinase by the unphosphorylated juxtamembrane region.
Cell
2001
;
106
:
745
–57.
16
Hafner C, Schmitz G, Meyer S, et al. Differential gene expression of Eph receptors and ephrins in benign human tissues and cancers 2.
Clin Chem
2004
;
50
:
490
–9.
17
Kullander K, Klein R. Mechanisms and functions of Eph and ephrin signalling.
Nat Rev Mol Cell Biol
2002
;
3
:
475
–86.
18
Parri M, Buricchi F, Taddei ML, et al. EphrinA1 repulsive response is regulated by an EphA2 tyrosine phosphatase.
J Biol Chem
2005
;
280
:
34008
–18.
19
Holmberg J, Clarke DL, Frisen J. Regulation of repulsion versus adhesion by different splice forms of an Eph receptor.
Nature
2000
;
408
:
203
–6.
20
Brantley-Sieders DM, Caughron J, Hicks D, Pozzi A, Ruiz JC, Chen J. EphA2 receptor tyrosine kinase regulates endothelial cell migration and vascular assembly through phosphoinositide 3-kinase-mediated Rac1 GTPase activation.
J Cell Sci
2004
;
117
:
2037
–49.
21
Fox BP, Kandpal RP. Invasiveness of breast carcinoma cells and transcript profile: Eph receptors and ephrin ligands as molecular markers of potential diagnostic and prognostic application.
Biochem Biophys Res Commun
2004
;
318
:
882
–92.
22
Herrem CJ, Tatsumi T, Olson KS, et al. Expression of EphA2 is prognostic of disease-free interval and overall survival in surgically treated patients with renal cell carcinoma.
Clin Cancer Res
2005
;
11
:
226
–31.
23
Kinch MS, Moore MB, Harpole DH, Jr. Predictive value of the EphA2 receptor tyrosine kinase in lung cancer recurrence and survival.
Clin Cancer Res
2003
;
9
:
613
–8.
24
Miyazaki T, Kato H, Fukuchi M, Nakajima M, Kuwano H. EphA2 overexpression correlates with poor prognosis in esophageal squamous cell carcinoma.
Int J Cancer
2003
;
103
:
657
–63.
25
Nakamura R, Kataoka H, Sato N, et al. EPHA2/EFNA1 expression in human gastric cancer.
Cancer Sci
2005
;
96
:
42
–7.
26
Ogawa K, Pasqualini R, Lindberg RA, Kain R, Freeman AL, Pasquale EB. The ephrin-A1 ligand and its receptor, EphA2, are expressed during tumor neovascularization.
Oncogene
2000
;
19
:
6043
–52.
27
Thaker PH, Deavers M, Celestino J, et al. EphA2 expression is associated with aggressive features in ovarian carcinoma.
Clin Cancer Res
2004
;
10
:
5145
–50.
28
Walker-Daniels J, Coffman K, Azimi M, et al. Overexpression of the EphA2 tyrosine kinase in prostate cancer.
Prostate
1999
;
41
:
275
–80.
29
Wu D, Suo Z, Kristensen GB, et al. Prognostic value of EphA2 and EphrinA-1 in squamous cell cervical carcinoma.
Gynecol Oncol
2004
;
94
:
312
–9.
30
Cheng N, Brantley DM, Liu H, et al. Blockade of EphA receptor tyrosine kinase activation inhibits vascular endothelial cell growth factor-induced angiogenesis.
Mol Cancer Res
2002
;
1
:
2
–11.
31
Cheng N, Brantley D, Fang WB, et al. Inhibition of VEGF-dependent multistage carcinogenesis by soluble EphA receptors.
Neoplasia
2003
;
5
:
445
–56.
32
Dohn M, Jiang J, Chen X. Receptor tyrosine kinase EphA2 is regulated by p53-family proteins and induces apoptosis.
Oncogene
2001
;
20
:
6503
–15.
33
Andres AC, Reid HH, Zurcher G, Blaschke RJ, Albrecht D, Ziemiecki A. Expression of two novel eph-related receptor protein tyrosine kinases in mammary gland development and carcinogenesis.
Oncogene
1994
;
9
:
1461
–7.
34
Zantek ND, Walker-Daniels J, Stewart J, et al. MCF-10A-NeoST: a new cell system for studying cell-ECM and cell-cell interactions in breast cancer.
Clin Cancer Res
2001
;
7
:
3640
–8.
35
Macrae M, Neve RM, Rodriguez-Viciana P, et al. A conditional feedback loop regulates Ras activity through EphA2.
Cancer Cell
2005
;
8
:
111
–8.
36
Xu H, Tian W, Lindsley JN, et al. EphA2: expression in the renal medulla and regulation by hypertonicity and urea stress in vitro and in vivo.
Am J Physiol Renal Physiol
2005
;
288
:
F855
–66.
37
Zelinski DP, Zantek ND, Stewart JC, Irizarry AR, Kinch MS. EphA2 overexpression causes tumorigenesis of mammary epithelial cells.
Cancer Res
2001
;
61
:
2301
–6.
38
Zantek ND, Azimi M, Fedor-Chaiken M, Wang B, Brackenbury R, Kinch MS. E-cadherin regulates the function of the EphA2 receptor tyrosine kinase.
Cell Growth Differ
1999
;
10
:
629
–38.
39
Sovova V, Kucerova D, Vojtechova M, Sloncova E, Tuhackova Z. Transactivation of E-cadherin is not involved in the activity of EGF receptor in colorectal carcinoma cells.
Int J Oncol
2004
;
25
:
1459
–64.
40
Ara T, Deyama Y, Yoshimura Y, et al. Membrane type 1-matrix metalloproteinase expression is regulated by E-cadherin through the suppression of mitogen-activated protein kinase cascade.
Cancer Lett
2000
;
157
:
115
–21.
41
Linggi B, Carpenter G. ErbB receptors: new insights on mechanisms and biology.
Trends Cell Biol
2006
;
16
:
649
–56.
42
Moscatello DK, Holgado-Madruga M, Godwin AK, et al. Frequent expression of a mutant epidermal growth factor receptor in multiple human tumors.
Cancer Res
1995
;
55
:
5536
–9.
43
Pratt RL, Kinch MS. Activation of the EphA2 tyrosine kinase stimulates the MAP/ERK kinase signaling cascade.
Oncogene
2002
;
21
:
7690
–9.
44
Flanagan JG, Vanderhaeghen P. The ephrins and Eph receptors in neural development 3.
Annu Rev Neurosci
1998
;
21
:
309
–45.
45
Brantley DM, Cheng N, Thompson EJ, et al. Soluble Eph A receptors inhibit tumor angiogenesis and progression in vivo.
Oncogene
2002
;
21
:
7011
–26.
46
Fang WB, Brantley-Sieders DM, Parker MA, Reith AD, Chen J. A kinase-dependent role for EphA2 receptor in promoting tumor growth and metastasis.
Oncogene
2005
;
24
:
7859
–68.
47
Guo H, Miao H, Gerber L, et al. Disruption of EphA2 receptor tyrosine kinase leads to increased susceptibility to carcinogenesis in mouse skin.
Cancer Res
2006
;
66
:
7050
–8.
48
Andersen P. Role of the tumor supressor IRF-1 activated by the epidermal growth factor receptor in human cancer cells. Master Thesis. Denmark: Department of Radiation Biology, Copenhagen University Hospital; 2006. p. 1–91.
49
Walker-Daniels J, Riese DJ, Kinch MS. c-Cbl-dependent EphA2 protein degradation is induced by ligand binding.
Mol Cancer Res
2002
;
1
:
79
–87.
50
Wang Y, Ota S, Kataoka H, et al. Negative regulation of EphA2 receptor by Cbl.
Biochem Biophys Res Commun
2002
;
296
:
214
–20.
51
Binns KL, Taylor PP, Sicheri F, Pawson T, Holland SJ. Phosphorylation of tyrosine residues in the kinase domain and juxtamembrane region regulates the biological and catalytic activities of Eph receptors.
Mol Cell Biol
2000
;
20
:
4791
–805.
52
Pedersen N, Pedersen MW, Lan MS, Breslin MB, Poulsen HS. The insulinoma-associated 1: a novel promoter for targeted cancer gene therapy for small-cell lung cancer.
Cancer Gene Ther
2006
;
13
:
375
–84.
53
Del Duca D, Werbowetski T, Del Maestro RF. Spheroid preparation from hanging drops: characterization of a model of brain tumor invasion.
J Neurooncol
2004
;
67
:
295
–303.