Predictive Markers and Cancer Prevention
EphA2 overexpression correlates with poor prognosis in esophageal squamous cell carcinoma
Version of Record online: 9 DEC 2002
Copyright © 2002 Wiley-Liss, Inc.
International Journal of Cancer
Volume 103, Issue 5, pages 657–663, 20 February 2003
How to Cite
Miyazaki, T., Kato, H., Fukuchi, M., Nakajima, M. and Kuwano, H. (2003), EphA2 overexpression correlates with poor prognosis in esophageal squamous cell carcinoma. Int. J. Cancer, 103: 657–663. doi: 10.1002/ijc.10860
- Issue online: 18 DEC 2002
- Version of Record online: 9 DEC 2002
- Manuscript Accepted: 14 OCT 2002
- Manuscript Revised: 11 OCT 2002
- Manuscript Received: 25 JUN 2002
- esophageal squamous cell carcinoma;
- lymph mode metastasis;
EphA2 is a member of the Eph family of receptor tyrosine kinases, which interact with cell-bound ligands known as ephrins. EphA2 expression was investigated by immunohistochemistry with an anti-EphA2 monoclonal antibody in 80 patients with esophageal squamous cell carcinoma (ESCC) who had undergone surgery. EphA2 overexpression was positive in 40 of the 80 patients (50%). A significant correlation was observed between EphA2 expression and regional lymph node metastasis (p=0.023), number of lymph node metastases (p=0.011) and poor degree of tumor differentiation (p=0.004). The survival rates of EphA2-positive patients were poorer than those of EphA2-negative patients (p=0.014). The 5-year survival rate of patients without EphA2 overexpression was 68%, whereas that of patients with EphA2 overexpression was 29%. EphA2 expression was also investigated in 7 ESCC cell lines (TE-1, -2, -8, -13, -15, TT and TTn) and 1 immortalized human esophageal keratinocyte cell line (CHEK-1). Western blotting revealed different levels of EphA2 expression in the 8 cell lines. EphA2 was expressed at a high level in the ESCC cell lines compared to CHEK-1. EphA2 phosphorylation was demonstrated in all cell lines. Northern blot analysis showed that EphA2 mRNA expression in TE-1 was greater than that in the other ESCC cell lines. The observation of small gaps on Western blot analysis of the ESCC cell lines suggests that there may be a mechanism for EphA2 regulation at the point of translation. In conclusion, EphA2 overexpression appears to be related to poor degree of tumor differentiation and lymph node metastasis in ESCC. Consequently, patients with EphA2 overexpression have a poorer prognosis than those without. EphA2 is a potential target to prevent ESCC cells spreading into the lymphatic drainage. © 2002 Wiley-Liss, Inc.
Esophageal cancer ranks among the 10 most frequent cancers in the world, occurring mostly in developing countries with marked regional variations in incidence.1, 2 Mortality rates are very similar to the incidence rates3 due to relatively late diagnosis and inefficiency of treatment. The survival rate at 5 years was reported to be < 10%,4 although recent advancements in surgical techniques and adjuvant therapy have improved the 5-year survival rate to about 40%.5
Molecules that are highly expressed by human esophageal squamous cell carcinomas (ESCC) may serve as therapeutically relevant targets or important tumor markers. Recently, several groups have identified overexpression of HER2 or epidermal growth factor receptor in some tumors and have used this knowledge to develop successful approaches for therapeutic targeting of cancer cells.6
Protein tyrosine phosphorylation generates the powerful signals necessary for the growth, migration and invasion of normal and malignant cells.7 A number of tyrosine kinases have been reported to have a relationship with cancer progression and patient prognosis.8 EphA2 (Mr 130,000) is a member of the Eph family of receptor tyrosine kinases, which interact with cell-bound ligands known as ephrins.9 The Eph receptor becomes phosphorylated at specific tyrosine residues in its cytoplasmic domain following ligand binding.10, 11, 12 The phosphorylated motifs serve as sites of interaction with certain cytoplasmic signaling proteins, and as a result, it is thought that ephrin-Eph signals are transmitted. It has been proposed that Eph-ephrin interactions act to limit cell migration across embryonic boundaries.13 Two recent studies have supported this hypothesis by showing that EphB-ephrin B interactions mediate cell sorting to the correct segments and prevent cell intermingling across segmental boundaries.14, 15 The function of Eph kinase has been studied in detail in normal cells and the protein has been suggested to regulate proliferation, differentiation and migration. Following on from these observations, it was suspected that Eph kinase might affect malignant ability in cancer. Recent studies have demonstrated EphA2 expression in human melanoma,16 colon cancer,17 prostate cancer18 and mammary cancer.19 A cell line study has also indicated that EphA2 is a powerful oncoprotein in breast cancer and that its overexpression causes malignant transformation.19 A 10–100-fold overexpression of EphA2 has also been reported in metastatic prostate cancer cells compared to non-invasive prostatic epithelial cells.18 However, the relationship between EphA2 expression and clinical features (tumor invasiveness, metastasis and prognosis) in patients with cancer is not known. Furthermore, few studies have investigated EphA2 expression in ESCC.
In our study, an immunohistochemical analysis of EphA2 protein expression was performed to determine the relationship between EphA2 overexpression and clinicopathological factors in ESCC, and to elucidate mechanism of EphA2 overexpression and signal transduction in ESCC, we also investigate Western blot and Northern blot analysis in human cell lines.
MATERIAL AND METHODS
Patients and tissue samples
The tissue specimens examined in our study were removed from 80 patients with thoracic esophageal cancer who had undergone surgery at the Gunma University Hospital between 1983 and 2000. Written informed consent to participate in the study was obtained from each patient before surgery, according to the ethical guidelines of our university. All patients underwent potentially curative surgery without preoperative therapy. The tissue specimens examined in our study were diagnosed pathologically as R0 after resection. The patients included 69 men and 11 women, who were 40–78 years of age (mean age: 61.6 years). The tumor stage was classified according to the 5th edition of the TNM classification of the International Union against Cancer (UICC). The mean postoperative follow-up period for the 80 patients was 32.8 months (range: 8.6–192.2 months). Specimens were fixed in 10% formaldehyde solution and embedded in paraffin.
Seven human esophageal cancer cell lines and 1 immortalized human esophageal cell line were used: TE-1, TE-2, TE-8, TE-13, TE-15, TT, TTn and CHEK-1. The TE cell lines were kindly provided by Dr. T. Nishihira, Institute of Development, Aging and Cancer, Tohoku University School of Medicine, Sendai, Japan. TT and TTn cells (JCRB0262 and 0261) were kindly provided by Dr. K. Takahashi. CHEK-1 cells were kindly provided by Dr. H. Matsubara. This latter cell line was established by transduction of human papillomavirus type 16 E6/E7 into primary cultures of human esophageal keratinocytes.20 All cancer cell lines were derived from ESCC with varying degrees of differentiation.21 TE cell lines and CHEK-1 were cultured in RPMI-1640 medium (Sigma Chemical Co., St. Louis, MO) supplemented with 10% fetal bovine serum and antibiotics (100 units/ml penicillin and 100 μg/ml streptomycin). TT and TTn were cultured in 1:1 Dulbecco's modified Eagle medium and Ham's F-12 medium (Sigma Chemical Co.) containing 10% fetal bovine serum and antibiotics as described above. All cell lines were cultured to 60–80% confluence.
Antibodies were purchased from the following manufacturers: monoclonal antibody (MAb) specific for EphA2 (clone D7) and MAb specific for FAK (clone 4.47) (Upstate Biotechnology, Inc., Lake Placid, NY); polyclonal antibody specific for ephrin-A1 (V-18) and MAb specific for phosphotyrosine (PY-20) (Santa Cruz Biotechnology, Inc., Santa Cruz, CA); MAb specific for Kip1/p27 (57) (Transduction Laboratories, Lexington, KY); MAb specific for E-cadherin (HECD-1) (Takara Shuzo, Kyoto, Japan). MAb specific for β-actin was purchased from Sigma Chemical Co.
Immunohistochemistry for EphA2 protein
Immunohistochemical staining was performed by the standard avidin-biotin-peroxidase complex (ABC) method. Briefly, each 4 μm tissue section was deparaffinized, rehydrated and incubated with fresh 0.3% H2O2 in methanol for 30 min at room temperature. After rehydration through a graded ethanol series, the sections were autoclaved in citrate buffer at 120°C for 5 min and then cooled to 30°C. After incubation with normal horse serum, the tissue sections were applied for 30 min and removed by blotting. The sections were then incubated with anti-EphA2 MAb at a dilution of 1:250 in phosphate-buffered saline (PBS) containing 1% bovine serum albumin at 4°C overnight, washed in PBS and incubated with secondary antibody for 30 min at room temperature. Immunohistochemistry was performed using the ABC system (Vectastain Lab. Inc., Burlingame, CA). The chromogen was 3,3′-diaminobenzidine tetrahydrochloride, applied as a 0.02% solution containing 0.0055% H2O2 in 50 mM ammonium acetate-citric acid buffer (pH 6.0). The sections were lightly counterstained with hematoxylin. Negative controls were prepared by substituting normal mouse serum for primary antibody, and no detectable staining was evident.
Evaluation of staining for EphA2
When > 40% of carcinoma cells in a given specimen were stained more intensely than the normal epithelium the sample was classified as EphA2-positive (+).
Cell extraction and Western blotting
Lysates from exponentially growing cell lines were prepared in buffer (20 mM Tris-HCl, pH 7.6, 1 mM EDTA, 140 mM NaCl, 1% Nonidet P-40, 1% aprotinin, 1 mM phenylmethylsulfonyl fluoride and 1 mM sodium vanadate). The protein concentration was determined with a BCA Protein Assay Kit (Pierce, Rockford, IL). Protein (30 μg) from each cell line was resuspended in sodium dodecyl sulfate (SDS) sample buffer (100 mM Tris-HCl, pH 8.8, 0.01% bromophenol blue, 36% glycerol and 4% SDS) containing 1 mM dithiothreitol, boiled for 5 min and subjected to a 5–10% gradient Ready-Gel (Bio-Rad, Tokyo, Japan). Proteins were electrotransferred to a Hybond enhanced chemiluminescence nitrocellulose membrane (Amersham Pharmacia Biotech, Buckinghamshire, UK). Proteins were immunoblotted using anti-EphA2 (clone D7; Upstate Biotechnology). The bands were detected using the enhanced chemiluminescence detection system (Amersham Pharmacia Biotech). For re-blotting, membranes were stripped according to the manufacturer's protocol. Proteins were re-blotted using anti-ephrinA2 (clone V-18; Santa Cruz) or anti-β-actin (Sigma Chemical Co.). Anti-β-actin (Sigma Chemical Co.) antibody served as the control. For stimulation by Eph ligand, cells were incubated for 10 min with 1 μg/ml ephrinA1-Fc Chimera (R & D Systems Inc., Minneapolis, MN) and then subjected to immunoblotting and immunoprecipitation. For inhibition of proteasomal degradation, cells were incubated for 4 hr with 50 μM MG132 (Peptide Institute, Osaka, Japan).
For immunoprecipitation, 850 μg of protein from each cell line was incubated overnight at 4°C with 5 μg anti-EphA2 antibody or 5 μg anti-FAK antibody and then with 30 μl protein G-Sepharose (Amersham Pharmacia Biotech) for 30 min at 4°C. Immunoprecipitates were washed 3 times in lysis buffer, resuspended in SDS sample buffer and resolved by SDS-PAGE.
RNA extraction and Northern blotting
Total RNA was extracted from the cells with TRIZOL (Gibco BRL, Rockville, MD). Twenty micrograms of RNA per lane was electrophoresed in 1.2% agarose gels containing 2.2 mol/l formaldehyde and blotted onto a Biodyne B membrane (Pall, Tokyo, Japan). The cDNA probe was labeled by using a random primed DNA labeling kit (Roche Molecular Biochemicals, Mannheim, Germany) and α-32P dCTP (Amersham Pharmacia Biotech). The human EphA2 probe was digested from pNeoMSV-EphA2 (generously provided by Dr. T. Hunter, SALK Institute, La Jolla, CA) with BglII and BamHI to yield a 2.9 kbp cDNA fragment. Membranes were prehybridized at 42°C for more than 2 hr. Hybridization was performed overnight at 42°C. The membranes were washed in 2 × SSC, 0.1% SDS for 15 min and 0.2 × SSC, 0.1% SDS for 15 min at 42°C. The washed membrane was exposed to X-ray film using an intensifying screen. 28S and 18S rRNA were stained with methylene blue after transfer to membranes to ensure that the RNAs were not degraded and were present in each sample in approximately equal quantities.
Statistical analysis was performed using the χ2 test, Fisher's exact test and Mann-Whitney U test. Survival curves were calculated by the Kaplan-Meier method and analysis was carried out by the log-rank test.
Relationship between EphA2 expression and clinicopathological features
EphA2 expression in ESCC was investigated by immunohistochemical analysis of formalin fixed, paraffin-embedded specimens using an EphA2-specific MAb. In normal esophageal tissue, immunostaining of EphA2 was detected in the cytoplasm of the basal cells, parabasal cells and leukocytes (Fig. 1a,b). Immunostaining of EphA2 was seen in the cell membrane and cytoplasm of cancer cells, particularly in cells located in the peripheral layers of the cancer cell nest (Fig. 1a,c,d). Since heterogeneous expression of EphA2 was noted in the tumor, the sample was classified as EphA2-positive (+) when > 40% of carcinoma cells were stained more intensely than the normal epithelium. EphA2 expression was positive in 40 of the 80 patients (50%). The relationship between the clinicopathological characteristics of patients with ESCC and EphA2 expression in their tumors is summarized in Table I. A significant correlation was observed between EphA2 overexpression and regional lymph node metastasis (p=0.023) and number of lymph node metastases (p=0.011). EphA2 overexpression is correlated with a poor degree of tumor differentiation (p=0.004). However, there was no significant association with age, gender, tumor location, depth of tumor invasion, distant metastasis or disease stage.
|Age (mean ± SD, years)1||62.0 ± 7.8||61.2 ± 9.2||—||0.686|
|TNM clinical classification|
|Number of lymph node metastases (mean ± SD)||3.0 ± 4.1||1.7 ± 4.1||—||0.011|
The survival rates of patients with EphA2-positive cancer were significantly lower than those of patients with EphA2-negative cancer (p=0.014; Fig. 2). The 5-year survival rate of patients without EphA2 overexpression was 68%, whereas that of patients with EphA2 overexpression was 29%. Although multivariate statistical analysis was carried out, EphA2 was not a prognostic factor by itself in contrast to depth of tumor invasion, lymph node metastasis or disease stage (data not shown).
EphA2 expression and tyrosine phosphorylation at the protein level in cell lines
The expression of EphA2 at the protein level was investigated in 8 cell lines derived from ESCC and immortalized esophageal keratinocytes. Western blotting revealed different levels of expression of EphA2 (Fig. 3). EphA2 was expressed at a high level in TE1 cells, at a lower level in TE2, TE8, TE13, TE15 and TTn cells, and at a very low level in TT and CHEK-1 cells. We investigated the level of E-cadherin protein because E-cadherin regulates the function of EphA2 and loss of E-cadherin function alters the localization of EphA2.22 E-cadherin was also expressed at different levels in each cell line, with CHEK-1 expressing very low levels of E-cadherin. Western blot analysis revealed no relationship between EphA2 and E-cadherin. EphrinA1 was expressed at various levels in each cell line, indicating that each has a different ephrinA1 ligand stimulation level.
The Eph receptor becomes phosphorylated at specific tyrosine residues in its cytoplasmic domain following ligand binding.10, 11, 12 The phosphorylated motifs serve as sites of interaction with certain cytoplasmic signaling proteins, and as a result, it is thought that ephrin-Eph signals are transmitted. To investigate whether EphA2 is tyrosine-phosphorylated in each cell line, immunoprecipitation was performed with an anti-EphA2 antibody and blotting with an anti-phosphotyrosin-specific antibody (Fig. 3, lowest panel). Various levels of EphA2 phosphorylation were demonstrated in all the cell lines. Phosphorylated EphA2 exhibited the highest level in TT cells, in which Western blotting showed the lowest level of EphA2. This indicates that almost all ESCC cells have different endogenous levels of phosphorylated EphA2, and that although EphA2 level may be relatively low in TT cells, they may become highly activated.
Tyrosine phosphorylation of EphA2 and focal adhesion kinase (FAK) by EphrinA1-Fc treatment in cell lines
Furthermore, we used ephrinA1-Fc fusion protein as an EphA2 ligand to confirm the downstream target of EphA2. Treatment of PC3 prostate tumor cells with ephrinA1-Fc fusion protein has been reported to induce tyrosine dephosphorylation of focal adhesion kinase (FAK).23 On the other hand, Carter et al.24 reported that similar treatment of the NIH3T3 immortalized mouse fibroblast cell line increased the tyrosine phosphorylation of FAK and EphA2. They considered that FAK constitutes an important downstream effector of the ephrinA1-Eph receptor tyrosine kinase system in fibroblasts. The CHEK-1, TE1 and TE8 cell lines exhibited heavy tyrosine phosphorylation of EphA2, whereas TT cells exhibited little such change upon ephrinA1-Fc stimulation (Fig. 4). Also, tyrosine-phosphorylation of FAK was clearly increased in the CHEK-1, TE1 and TE8 cell lines upon ephrinA1-Fc stimulation. However, the tyrosine phosphorylation of FAK exhibited no change in TT cells. These results indicate that the ephrinA1-EphA2 receptor tyrosine kinase system cannot appropriately transmit downstream in TT cells.
Expression of EphA2 at the mRNA level in cell lines
In view of the marked variations seen in the level of expression of EphA2 protein between the 8 cell lines, Northern blotting was performed to examine the underlying effects of EphA2 on tumor cell regulation. This analysis indicated that the level of expression of EphA2 mRNA was relatively steady and did not always correlate with the amount of EphA2 protein in the cells (Fig. 5). EphA2 was expressed at a high level in the TE1 cell line. However, expression of mRNA in TT was similar to that in all other cell lines.
Effect of EphA2 expression by proteasome inhibitor treatment
To investigate the stability of EphA2, EphA2 expression was compared in cells incubated with 50 μM MG132 and in control cells. No difference in EphA2 expression was observed in cells incubated with MG132. The proteasome inhibitor, MG132, abolished the decrease in Kip1/p27 protein (Fig. 6). This result indicates that EphA2 levels do not appear to be regulated by the ubiquitin-proteasome pathway.
The results of immunohistochemistry in our study suggest that expression of EphA2 protein is correlated with cell differentiation, regional lymph node metastasis and number of lymph node metastases. Furthermore, the prognosis of patients overexpressing EphA2 was significantly less favorable than that of the other patients. It has been reported that EphA2 is a powerful oncoprotein in breast cancer and that EphA2 overexpression causes malignant transformation.19 A 10–100-fold overexpression of EphA2 has also been reported in metastatic prostate cancer cells compared to non-invasive prostatic epithelial cells.18 The results of our study support these observations. Activation of EphA2 has been reported to inhibit the Ras/MAPK pathway.25 However, this assay was performed using non-malignant cell lines and a study using a human prostate cancer cell line exhibited that EphA2 activation did not inhibit proliferation. Miao et al.25 suggested that these cells may have developed mechanisms to escape the growth-inhibitory effects of EphA2 activation. Further work is clearly required to investigate the relationship between EphA2 overexpression and tumor metastasis.
The expression levels of EphA2 were investigated in 7 ESCC cell lines and 1 immortalized esophageal keratinocyte cell line. All ESCC cell lines except TT expressed higher levels of EphA2 than the non-transformed cell line. Phosphorylation of EphA2 in the immortalized cell line was weaker than that in the cancer cell lines despite the presence of abundant immunoprecipitated EphA2 protein. In the ligand stimulation assay, some cell lines were able to transmit downstream of EphA2, but we could not identify any change in the tyrosine phosphorylation of FAK in TT cells upon ephronA1-Fc treatment. Because the level of phosphorylated EphA2 was highest in TT cells, EphA2 may be active continuously but unable to transmit downstream. These results indicate that the level of EphA2 protein reflects tumor malignancy in ESCC and that EphA2 phosphorylates itself and transmits downstream of the ephrin-EphA2 system in a majority ESCCs, whereas the ephrin-EphA2 system cannot function appropriately in some ESCCs. Northern blot analysis showed that more EphA2 mRNA was expressed in TE1 cells than in the other ESCC cell lines. This result is similar to that of Western blot analysis. However, because there were little gaps among the ESCC cell lines by Western blot analysis, there may also be a mechanism of EphA2 regulation at the point of translation. Judging from these data and the assay using cells incubated with the proteasome inhibitor MG132, EphA2 levels do not appear to be regulated at the transcription level, nor by the ubiquitin-proteasome pathway.
It has been reported that E-cadherin regulates the function of EphA2 and that loss of E-cadherin function altered the localization of EphA2.22 Orsulic et al.26 showed that the Eph receptor and ephrins were differentially regulated by E-cadherin. In our study, there appeared to be no relationship between EphA2 and E-cadherin on Western blot analysis. EphA2 and E-cadherin could not be coimmunoprecipitated (data not shown) and no relationship could be demonstrated between EphA2 and E-cadherin in ESCC.
In our study, patients with EphA2 overexpression had a poor prognosis for overall survival. Although multivariate statistical analysis was performed, EphA2 was not a prognostic factor by itself. Thus, it was presumed that our result was influenced by lymph node metastasis. Metastasis, the main cause of death in most cancer patients, remains one of the most important but least understood aspects of cancer. In particular, lymph node metastasis is associated with a poor prognosis in esophageal cancer.27, 28 EphA2 may be a good target to prevent ESCC cells spreading into the lymphatic drainage. The intensity of EphA2 expression in preoperative biopsy specimens may be an indicator of advanced disease with a high probability of tumor spread.
In conclusion, EphA2 overexpression was related to poor degree of tumor differentiation and lymph node metastasis. Consequently, patients with EphA2 overexpression had a poor prognosis. EphA2 may be a good target to prevent ESCC cells spreading into the lymphatic drainage.
We are grateful to Dr. T. Nishihira (Institute of Development, Aging and Cancer, Tohoku University School of Medicine, Sendai, Japan), Dr. K. Takahashi (Department of Oral Surgery, School of Medicine Chiba University, Japan) and Dr. H. Matsubara (Department of Academic Surgery, Chiba University Graduate School of Medicine, Japan) for providing cell lines and Dr. T. Hunter (SALK Institute, La Jolla, CA) for providing pNeoMSV-EphA2.
- 4Survival of cancer patients in Europe-the Eurocare Study. Lyon: IARC, 1995., , , , , .