10.1245/s10434-006-9320-5
Annals of Surgical Oncology 14:1565-1574 (2007)
© 2007 Society of Surgical Oncology
Change of E-Cadherin by Hepatocyte Growth Factor and Effects on the Prognosis of Hypopharyngeal Carcinoma
Chul-Ho Kim, MD, PhD1,2,
JangHee Kim, MD3,
Hison Kahng, MD1,2 and
Eun Chang Choi, MD, PhD4
1 Department of Otolaryngology, School of Medicine, Ajou University, Suwon, Korea
2 Center for Cell Death Regulating Biodrug, School of Medicine, Ajou University, Suwon, Korea
3 Department of Pathology, School of Medicine, Ajou University, Suwon, Korea
4 Department of Otorhinolaryngology, College of Medicine, Yonsei University, 134 Shinchon-dong, Seodaemun-gu, Seoul, 120-752, Korea
Correspondence: Address correspondence and reprint requests to: Eun Chang Choi, MD, PhD; E-mail: ostium{at}ajou.ac.kr
 |
ABSTRACT
|
|---|
Background: Hepatocyte growth factor (HGF) is known to induce scattering in various epithelial cells, and E-cadherin plays important roles in the maintenance of cell-cell adhesion. However, the mechanisms surrounding these actions are not fully understood. Therefore, we examined how HGF affects the expression and distribution of E-cadherin. In addition, we observed the relationship between prognosis and modulation of E-cadherin by HGF in hypopharyngeal carcinoma.
Methods: Tumor tissues from 66 patients with hypopharyngeal squamous cell carcinoma were evaluated for the expression of HGF, its receptor (c-Met), and E-cadherin. Reverse transcriptase–polymerase chain reaction (RT-PCR) and Western blot test were performed on hypopharyngeal cancer tissues. The association and changes of E-cadherin with HGF treatment in a hypopharyngeal cancer cell line were investigated by RT-PCR, Western blot analysis, inhibition assay, immunofluorescence staining, and invasion assay.
Results: E-cadherin expression was found in 87.9% of squamous cell carcinomas; these could be further classified as membranous type (46.9%) or nonmembranous type (53.1%). The expression of HGF in tumors with nonmembranous type E-cadherin expression was far higher than in tumors with membranous expression. Nonmembranous type E-cadherin expression correlated significantly with lymph node metastasis, distant metastasis, and recurrence (P < .05). HGF decreased the expression of E-cadherin and induced the translocation of E-cadherin to the cytoplasm. HGF and E-cadherin neutralizing antibody stimulated dispersion, and HGF significantly enhanced the invasion of hypopharyngeal cancer cells in a dose-dependent manner (P < .05).
Conclusions: These results suggest that HGF can modulate the expression and intracellular localization of E-cadherin in hypopharyngeal cancer cells. In addition, these results indicate that changes in E-cadherin by HGF can affect the prognosis of hypopharyngeal carcinoma.
Key Words: HGF • E-cadherin • Hypopharyngeal cancer
|
INTRODUCTION
|
|---|
Hypopharyngeal carcinoma is characterized by aggressive invasion and metastasis. Therefore, most cases of hypopharyngeal carcinoma are detected in the advanced stage (including invasion to adjacent vital structures, a strong tendency for submucosal spreading, and enlarged ipsilateral or bilateral cervical lymph node deposits). These features combine to make locoregional control of these cancers difficult. The 5-year overall survival rate for all patients with hypopharyngeal carcinoma is only 30%; this is essentially unchanged from two decades ago, despite various treatments and intensive research. This is attributed to the rising incidence of distant metastases and intercurrent diseases, as well as second primary malignancies. More effective therapies for hypopharyngeal carcinoma are thus needed. Although various studies have tried to find the mechanism of invasion and metastasis of cancer, the precise molecular mechanisms of invasion and metastasis of head and neck cancers are not fully understood.
Hepatocyte growth factor (HGF), a peptide growth factor well known as a potent stimulator of hepatocyte growth, can promote proliferation, motility, morphogenesis, and angiogenesis in many types of cells, including various tumor cells.1–4 HGF is primarily produced by mesenchymal cells (such as fibroblasts);5 its biological signal is transmitted from mesenchymal cells to epithelial cells through the HGF receptor,6 a c-met proto-oncogene product.7 Previously, we reported that immunohistochemical staining of hypopharyngeal cancer tissue showed high expression of HGF and c-Met. We also demonstrated that the level of HGF greatly increased in relation to lymph node metastasis and clinical stage of the tumor, whereas c-Met expression rose only according to the clinical stage of the tumor. Moreover, the expression of both mRNA and protein of HGF and c-Met were higher in hypopharyngeal cancer tissue than in normal tissue.8
E-cadherin plays important roles in the maintenance of cell-cell adhesions in epithelial cells.9 Downregulation of E-cadherin in transformed cell lines has been associated with dedifferentiation and acquisition of the ability to invade, suggesting a possible role of this protein as a tumor suppressor.10 Reduced E-cadherin expression and HGF were detected in gastric cancer cells,11 and HGF induces the transfer of E-cadherin from cell-cell contact sites to the cytoplasm in prostate cancer cells.12
In the present study, we first studied the correlation between HGF, c-Met, and E-cadherin in human hypopharyngeal carcinoma tissues from curatively resected specimens, finding a strong correlation between the expression of HGF and the expression and/or localization of E-cadherin. We also showed that HGF can induce the translocation of an E-cadherin in human hypopharyngeal cancer cell lines, resulting in invasion into a collagen membrane.
|
MATERIALS AND METHODS
|
|---|
Patients
Sixty-six patients admitted for hypopharyngeal squamous cell carcinoma (SCC) at Ajou University Hospital and Yonsei University Medical Center in Korea were enrolled onto this study between 1994 and 2000. Surgically excised specimens of hypopharyngeal SCC (excluding carcinoma-in-situ and verrucous carcinoma) were used. None of the patients had undergone radiotherapy or chemotherapy before surgical excision. The average age at clinical onset was 62 years, with a range of 41–78 years. There were 7 patients at stage I, 2 patients at stage II, 19 patients at stage III, and 38 patients at stage IV (based on the 2002 American Joint Committee on Cancer TNM classification system).
Cells and Cell Culture
FaDu cells, a hypopharyngeal cancer cell line, were obtained from American Type Culture Collection (ATCC, Washington, DC) and were cultured in Eagle minimum essential medium containing 25 mM of NaHCO3 (EMEM), 10% fetal bovine serum, glutamine, 1% essential amino acids, vitamins, and streptomycin under 5% CO2 atmosphere at 37°C.
Expression of HGF, c-Met, and E-Cadherin in Hypopharyngeal Cancer Tissues
Formalin-fixed, paraffin-embedded tissue was cut into 5-µm sections. The paraffin was removed, and the sample was rehydrated by serial passage through xylene and a graded series of ethanol. Endogenous peroxidase activity was blocked by incubation for 45 minutes in 6% hydrogen peroxide in methanol. The primary antibodies were a human HGF polyclonal antibody used at 50 µg/mL (R&D Systems, Minneapolis, MN), human-met c-28 polyclonal antibody, and monoclonal mouse anti–E-cadherin (HECD-1; Zymed Laboratory, San Francisco, CA) used at .625 µg/mL (Santa Cruz Biotechnology, Santa Cruz, CA). Primary antibody incubations were carried out for 2 hours at 25°C. The sections were thoroughly washed in phosphate-buffered saline, then incubated with the appropriate biotinylated second antibody followed by avidin-peroxidase. Goat and rabbit Vectastain-elite immunoperoxidase kits (Vector Laboratories, Burlington, VT) were purchased for this purpose and were used according to the manufacturer’s instructions. The chromogenic reaction was carried out with 3–3'-diaminobenzidine in a peroxidase substrate solution for 4 minutes. Hematoxylin was used for nuclear staining. For each experiment, negative controls (which included the omission of either the primary or secondary antibody) were included to examine nonspecific staining. The slides were reviewed by at least two pathologists and scored semiquantitatively by stain intensity and stain area (double scoring system) as follows: -, no staining (score 0); ± , definite but weak staining (score 1); +, moderate staining (score 2); ++, strong staining (score 3); stain < 35% (score 1); 35%–75% (score 2); and stained > 75% of cancer cells (score 3).
Reverse Transcriptase–Polymerase Chain Reaction
Total RNA from fresh hypopharyngeal cancer tissues and FaDu cells was extracted with TriZol (Invitrogen, Groningen, Netherlands). cDNA was prepared with the Omniscript Reverse Transcriptase kit (Qiagen, Germany) according to the manufacturer’s instructions. To determine the effects of HGF, cells were treated with HGF (0, 10 ng/mL, 30 ng/mL). The sequences of PCR primers were as follows: HGF-F: 5'-ACA TCG TCA CTT CTG GC-3'; HGF-R: 5'-ATC CAT CCT ATG TTT GTT CG-3'; c-Met-F: 5'-AGT AGC CTG ATT GTG CAT TT-3'; c-Met-R: 5'-TCT TTC ATG ATG CCC TC-3'; E-cadherin–F: 5'-ACA TCG TCA CTT CTG GC-3'; E-cadherin–R: 5'ATC CAT CCT ATG TTT GTT CG-3'. PCR products were separated by electrophoresis in 1.5% agarose gels and were detected under ultraviolet light (Bio-Rad, Hercules, CA).
Western Blot Test
Fresh hypopharyngeal cancer tissues were prepared, and exponentially growing cells in 10-cm2 dishes were rinsed several times with phosphate-buffered saline and fed with EMEM supplemented with 10% fetal bovine serum and human recombinant HGF (0, 10 ng/mL, 30 ng/mL) (R&D Systems). The cells were cultured for 3 days in a humidified CO2 incubator. Control cells were similarly washed and cultured in medium without HGF. Cells were washed with phosphate-buffered saline and extracted with EBC buffer (120 mmol/L of NaCl, .5% NP-40, 40 mmol/L of Tris, pH 8.0, 1 mmol/L of ethylenediaminetetraacetic acid) with protease inhibitors (100 µg/mL of phenylmethylsulfonyl fluoride and 1 µg/mL of leupeptin). The extracts were centrifuged at 10,000 x g for 10 minutes; the supernatants were used for Western blot analyses. Protein content was measured with a protein assay kit (Bio-Rad). Twenty micrograms of protein was resolved by sodium dodecyl sulfate–polyacrylamide gel electrophoresis transferred to nitrocellulose filters (Amersham, Arlington Heights, IL), and incubated overnight at 4°C with anti-human c-Met antibodies, anti-human HGF antibodies, or anti-human E-cadherin antibodies. After washing, the filters were incubated with peroxidase-conjugated donkey anti-rabbit antibody (Amersham) or donkey anti-mouse antibody (Amersham) and were visualized with an enhanced chemiluminescence detection system (Amersham).
Effect of HGF on Colony Dispersion
FaDu cells were subcultured and maintained in growth medium until colonies > 16 cells were established. Cultures were then deprived of growth factors and serum for 48 hours before treatment with HGF (0, 10 ng/mL, 30 ng/mL) at the times and conducted in cells pretreated for 2 hours with mitomycin C (8 µg/mL). To confirm the effect of HGF, cells were treated with HGF (0, 10 ng/mL, 30 ng/mL) and antihuman E-cadherin mouse monoclonal antibody (3 µg/mL). Colony dispersion was documented by photography at 6, 12, 18, and 24 hours.
Immunofluorescence Staining
Cells grown on glass slides were fixed with 3.7% formaldehyde containing 10 mM of CaCl2 in Tris-buffered saline (TBS) at room temperature for 15 minutes. The fixed cells were treated with .2% Triton X-100 in TBS for 15 minutes and washed with TBS three times. After soaking with TBS containing 10% fetal calf serum (FCS) (10% FCS/TBS) for 1 hour, the samples were incubated overnight with primary antibody in 10% FCS/TBS. The slides were then washed with TBS three times, followed by incubation with secondary antibody in 10% FCS/TBS for 1 hour. After incubation, the cells were washed with TBS three times, embedded in TBS containing 50% glycerol, and analyzed with a Zeiss confocal laser scanning microscope.
Invasion Assay
Transwell chambers (Costar) were used to verify the degree of invasiveness depending on the administration of HGF. First, type I collagen (6 µg per filter) melted in EMEM 100 µL was poured into the upper part of a polyethylene filter (8 µm pore size); coating was carried out in a laminar flow hood for one night. A total of 500 µL of .5% fetal bovine serum medium was put into the lower part of each well, and the wells were adjusted to HGF densities of 0, 10, and 30 ng/µL. After preprocessing with mitomycin C (8 µg/mL) for 30 minutes, 105 cells (in 100 µL of growth medium) were attached to the top of the filter of the upper well. After this chamber was cultivated in 5% CO2 at 37°C for 48 hours, the filter of the upper well was removed and the cells passed through the pore. The attached cells on the lower part were dyed with hematoxylin and counted with a light microscope.
Statistical Analysis
Student’s t-test and one-way analysis of variance (for the invasion assay) were used for statistical analyses of the data. Patient survival rates were calculated by the Kaplan-Meier method, and statistically significant differences in survival were identified by the log rank test. Multivariate analysis was performed by the Cox proportional hazard model. All statistical analyses were conducted by SPSS 10.0 statistical software (SPSS, Chicago, IL). A P value of less than .05 was considered statistically significant.
|
RESULTS
|
|---|
Expression of HGF, c-Met, and E-Cadherin and in Human Hypopharyngeal Cancer Tissue
HGF overexpression was observed in 74.5% of hypopharyngeal cancer tissues. In most of the specimens, HGF expression was primarily seen in the stromal cells around the cancer cells. In most cases, the cytoplasm of the stromal cells was diffusely and darkly stained; in some cases, the cell membrane was also stained (Fig. 1A
). There was a statistically significant correlation between HGF expression and lymph node metastasis and clinical stage (data not shown). Compared with HGF, c-Met expression was not observed in stromal cells, but was primarily in carcinoma cells and was occasionally (and weakly) in normal cells. In most of the specimens, c-Met was expressed strongly and diffusely in the cytoplasm of carcinoma cells (Fig. 1B). There was a statistically significant correlation between the expression of c-Met and lymph node metastasis (data not shown).
Received for publication September 30, 2006.
Accepted for publication November 22, 2006.
TOP
ABSTRACT
INTRODUCTION
) or nonmembranous type (53.1%) (Fig. 1D
- and ß-catenin does not change with HGF stimulation. Instead, only E-cadherin expression was decreased, and this was mediated by endocytosis to the cytoplasm, as suggested by Western blot and immunofluorescence studies.12,25 The effects of HGF on E-cadherin–mediated cell-cell adhesion were also observed in melanoma cells and prostate cancer cells.12,25 Li et al.25 reported that prolonged HGF stimulation causes downregulation of E-cadherin, which is MAPK and PI3K dependent. In melanoma cells, activation of Snail expression plays an important role in downregulation of E-cadherin.26 The role of Snail/Slug in HGF-induced downregulation of E-cadherin and desmoglein is now under further investigation. Grotegut et al.27 reported that Snail upregulation by HGF is mediated via the MAPK/Egr-1 signaling pathway and that both Snail and Egr-1 play a critical role in HGF-induced cell scattering, migration, and invasion. 
