help button home button Endocrine Society JCEM
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a related Letter to the Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Yoshida, S.
Right arrow Articles by Terakawa, N.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Yoshida, S.
Right arrow Articles by Terakawa, N.
The Journal of Clinical Endocrinology & Metabolism Vol. 87, No. 5 2376-2383
Copyright © 2002 by The Endocrine Society

Other Original Articles

Induction of Hepatocyte Growth Factor in Stromal Cells by Tumor-Derived Basic Fibroblast Growth Factor Enhances Growth and Invasion of Endometrial Cancer

Souichi Yoshida, Tasuku Harada, Tomio Iwabe, Fuminori Taniguchi, Akiko Fujii, Yasuko Sakamoto, Nobuhiro Yamauchi, Goshi Shiota and Naoki Terakawa

Department of Obstetrics and Gynecology (S.Y., T.H., T.I., F.T., A.F., Y.S., N.Y., N.T.) and 2nd Department of Internal Medicine (G.S.), Tottori University School of Medicine, Yonago 683-8504, Japan

Address all correspondence and requests for reprints to: Souichi Yoshida, M.D., Department of Obstetrics and Gynecology, Tottori University School of Medicine, Yonago 683-8504, Japan. E-mail: . souichi{at}grape.med.tottori-u.ac.jp

Abstract

Tumor progression is often regulated through interactions between carcinoma cells and host stromal cells. In this study of endometrial cancer, we investigated one mechanism potentially involved in hepatocyte growth factor (HGF)-mediated cancer-stromal interactions. Endometrial cancer cells (HEC-1 and ISHIKAWA) expressed the c-met receptor, but HGF did not. HGF, however, did stimulate the proliferation and invasion of these cells. The HGF gene was expressed in stromal cells, which had been separated from primary cultures of endometrial cancers, 6.4 times more than in isolated normal endometrial stromal cells. Immunohistochemical staining revealed immunoreactive HGF in cancer stromal cells, the staining intensity being more pronounced in cancer tissue than in normal endometrium. The conditioned medium from normal epithelial cells and cancer cell lines induced HGF production in normal stromal cells. We identified basic fibroblast growth factor as an HGF inducer derived from endometrial cancer cell lines. Basic fibroblast growth factor derived from tumor cells may induce HGF in endometrial stromal cells, whereas stromal cell-derived HGF leads to the invasive growth of carcinoma cells. These interactions, mediated by HGF and HGF inducers, may play a significant role in the progression of endometrial cancer.

SURGICAL AND HORMONAL therapies are effective in the early stage of endometrial cancer, a common neoplastic disease among women; however, no effective treatments exist for the advanced stage. Our lack of knowledge about the molecular mechanisms involved in endometrial cancer hampers progress toward an effective treatment. We do know that mesenchymal-epithelial cell interactions are essential for maintaining optimal cell function in normal adult tissue. We also know that the permissive counterparts of epithelial and stromal interactions may provide regulatory signals that maintain homeostasis. Previous studies have also suggested that interactions with host stromal cells influence the growth and invasive potential of epithelial malignant cells (1). Malignant transformation of epithelial cells leads to the alteration of reciprocal molecular exchange and disrupts normal homeostatic regulation. This could result in stromal cells that receive and transmit altered molecular signals. In fact, biological changes in stromal cells surrounding epithelial malignancy have been postulated to enhance the malignant phenotype of tumor cells (2, 3). Thus, host stromal-derived factors are likely to be key molecules that regulate tumor progression.

Hepatocyte growth factor (HGF), originally characterized as a potent mitogen for adult hepatocytes (4, 5), is a stromal-derived pleiotropic growth factor that elicits mitogenic, motogenic, and morphogenic activities on various types of cells, mainly as a paracrine factor (6, 7, 8). HGF also reportedly promotes angiogenesis in vitro and in vivo (9, 10). This effect of HGF appears both in normal tissue and malignant tumors by binding a high-affinity, membrane-spanning, c-met receptor, i.e. the c-met proto-oncogene product of heterodimeric tyrosine kinase (11). In normal uterine endometrium, stromal-derived HGF modulates regeneration during menstruation and promotes proliferation, migration, and lumen formation of endometrial epithelial cells (12). In carcinoma of the thyroid, gastrointestinal tract (13), ovary (14), pancreas (15), renal cells (16), breast (17), prostate (18), and endometrium (19), the c-met receptor is overexpressed. Moreover, in endometrial cancer, c-met expression levels correlated with surgical stage, histological grade, and shorter survival (19). Recent studies indicate that stromal-derived HGF promotes carcinoma cell proliferation and invasion, whereas carcinoma cells secrete several inducers of HGF production in stromal cells (20, 21). These data suggest that the HGF/c-met pathway plays a role as a mediator in cancer-stromal interaction. In endometrial cancer, however, no data are available to demonstrate an interaction mediated by stromal-derived HGF and tumor-derived HGF inducers.

In this study, we examined whether and how the HGF/c-met pathway is involved in endometrial cancer progression by testing the effects of HGF on endometrial carcinoma cell proliferation and invasion. Using quantitative real-time RT-PCR, the expression levels of HGF in stromal cells separated from primary cultures of endometrial cancer tissues were compared with those of normal endometrial stromal cells. HGF protein expressions were also evaluated by immunohistochemical staining. Finally, we identified an HGF inducer derived from endometrial cancer cells.

Materials and Methods

Tissue collection and processing

After obtaining informed consent, 10 neoplastic and 9 nonneoplastic endometrial tissues were obtained from patients undergoing hysterectomy for endometrial cancer and uterine leiomyoma, respectively. The specimens were dissected and rinsed in PBS. Part of the tissue was immediately frozen in liquid nitrogen for immunohistochemical study, and other parts were used for cell culture as described below. To confirm histopathology, routine hematoxylin and eosin-stained paraffin sections were prepared. The histological grade of the neoplastic specimens included in this study were: grade 1 (3 cases), grade 2 (4 cases), and grade 3 (3 cases).

Carcinoma and its stromal cells were collected from primary cultures of endometrial cancer tissues. Cancer tissue weighing 0.5–1.0 g wet were minced and disrupted with 0.5% collagenase in HBSS at 37 C for 60 min. The dispersed cells were filtered through a 70-µm nylon mesh to remove the undigested tissue pieces. The filtered fraction containing carcinoma and its stromal cells was placed on 10-cm culture dishes for 30 min at 37 C in 5% CO2 in air. After removing nonadhering cells, mixed cultures of carcinoma and stromal cells were maintained in Eagle’s minimum essential medium (EMEM) in the presence of 10% FBS. These cells were grown to confluence for 4–7 d. At confluence, the relative proportions of cancer colonies to surrounding stromal cells were approximately 30%. After reaching the confluence, medium was changed to DMEM/Ham’s F-12 (DMEM/F-12; 1:1, vol/vol) supplemented with 10% FBS, then cultured for 3–4 d. The cells were washed with PBS three times. Working under an inverted microscope, either the endometrial cancer cells or its stromal cells were scraped from cancer colonies or the spindle-shaped surrounding monolayer lesions, using 23-gauge needles, then cells were collected. Total RNA from each separated cell was extracted as described below. Immunocytochemical analysis of cancer and of stromal cells scraped from primary culture was performed. We used cytokeratin (DAKO Corp., Kyoto, Japan) as a marker of epithelial cells, vimentin (DAKO Corp.) as a marker of stromal cells, and factor VIII (DAKO Corp.) as a marker of endothelial cells. The results showed that the purification of cancer was 80–90%, and that of stromal cells was more than 95%. The separated carcinoma cell samples were contaminated with stromal cells.

Normal endometrial epithelial and stromal cells were isolated from nonneoplastic endometrial tissues. The stromal cells were isolated as previously described (22). Briefly, the normal endometrial tissues were minced and digested as described above. The dispersed cells were filtered through a 70-µm nylon mesh to remove the undigested tissue containing the glandular epithelium. The filtered fraction was separated further from epithelial cell clumps by differential sedimentation at unit gravity. The upper medium containing stromal cells was filtered through 40-µm nylon mesh. Final purification was achieved by allowing stromal cells, which attach rapidly to plates, to adhere selectively to culture dishes for 30 min at 37 C in 5% CO2 in air. Nonadherent epithelial cells were removed. Stromal cells were cultured in DMEM/F-12 supplemented with 10% FBS at 37 C in 5% CO2 in air. We used stromal cells in a monolayer culture after the first passage.

We isolated endometrial epithelial cells according to the method of Sugawara et al. (12), with some modifications. Lower medium (after separation of stromal cells by differential sedimentation) and fresh DMEM/F12 medium (back-washed through 40-µm nylon mesh) were collected in one 15-ml sterile polyethylene tube. After adding fresh medium up to 10 ml, the tube stood upright at 37 C for 30 min. The bottom 2 ml of cell suspension was recovered and layered over 10 ml medium and separated by differential sedimentation at unit gravity for 1 h. This process was repeated twice. The latter 2 ml containing endometrial epithelial cells was cultured in EMEM containing 10% FBS.

To confirm the purification of isolated normal endometrial epithelial and stromal cells, immunocytochemical analysis was performed, using cytokeratin, vimentin, and factor VIII. The results showed that the purity of epithelial and stromal cells was more than 90% and 98%, respectively.

Cell lines

HEC-1 cells, a model of moderately well-differentiated carcinoma, were obtained from the Health Science Research Resources Bank (Osaka, Japan). The ISHIKAWA cells (ISHIKAWA 3-H-12 No. 56), a model of well-differentiated endometrial adenocarcinoma, were a gift from Dr. M. Nishida (Tsukuba University, Ibaragi, Japan). These cell lines were maintained in EMEM in the presence of 10% FBS.

Cell proliferation assay

A mitogenic assay with recombinant human HGF (R & D Systems, Minneapolis, MN) was performed on the HEC-1 and ISHIKAWA cells. These cells were trypsinized and plated at a density of 2 x 103/well in a 96-well plate with EMEM serum-free medium supplemented with 2 mg/ml BSA (Sigma, St. Louis, MO) along with various concentrations of HGF (0–50 ng/ml). A monoclonal antibody against HGF (5 µg/ml; antihuman HGF, R & D Systems) was added to neutralize the specific effects of HGF. Monoclonal antibody mouse Ig G1k (COSMO BIO CO., LTD, Tokyo, Japan) was added solely or simultaneously with HGF. Each plate had one control column (six wells) containing medium free of HGF. After 48 h incubation, DNA synthesis and cell growth were measured. To examine DNA synthesis, 5-bromo-2'-deoxyuridine (Brdu) incorporation was assessed with a kit, Cell proliferation ELISA system, version 2 (Amersham Pharmacia Biotech, Little Chalfont, UK) according to the manufacturer’s instructions. Proliferation of cancer cell lines was determined spectrophotometrically by measuring the incorporation of terazolium dye [3-[4,5-dimethylthyazol-2-yl]-2,5-diphenyl tetrazolium bromide (MTT) assay]. The MTT assay used in this study was a previously described system (23). The ratio to the mean value of control (percent of control) was used for comparison. Results were presented as means ± SE of triplicate assays.

Matrigel invasion chamber assay

Biocoat Matrigel invasion chambers (Nippon Becton Dickinson and Co., Tokyo, Japan) were used. The HEC-1 and ISHIKAWA cells were inoculated in the inner cup coated with Matrigel at a density of 10 x 104 cells/ml in a vol of 0.5 ml serum-free EMEM containing 2 mg/ml BSA. In 24-well chamber plates, the EMEM containing 0–50 ng/ml HGF was added. To neutralize the specific effects of HGF, a monoclonal antibody against HGF (5 µg/ml) was used, and monoclonal antibody mouse Ig G1k was used as the control. The HEC-1 and ISHIKAWA cells were then cultured for 24 h and 72 h, respectively. The cell penetration was quantified by staining the porous membrane with a Diff-Quick stain kit (Green Cross, Kobe, Japan). Membrane pore size is 8 µm. The number of penetrating cells was counted per filter (x100). Each concentration of HGF was added to each of the three wells, and results were reported as means of the three membranes. The assay was repeated three times, and similar results were obtained. The cell numbers per 1 cm2 were used for comparison.

RT-PCR and Southern blotting

We used the guanidium thiocyanate method to extract total RNA from carcinoma cells that had been separated from primary cultures, as described above, and from endometrial cancer cell lines. RT of RNA from these cells into cDNA and PCR amplification for HGF and c-met was performed using the Gene Amp RNA PCR Core Kit (Perkin-Elmer Corp., Branchburg, NJ). For PCR analysis, the following specific primers for human HGF and human c-met were also used: glycerol-3-phosphate dehydrogenase (GAPDH) was examined as an internal control; HGF (sense): 5'-TCA CGA GCA TGA CAT GAC TCC-3' and (antisense): 5'-AGC TTA CTT GCA TCT GGT TCC-3', c-met (sense): 5'-GGT TGC TGA TTT TGG TCT TGC-3'and (antisense): 5'-TTC GGG TTG TAG GAG TCT TCT-3' (24), GAPDH (sense) 5'-ACC ACA GTC CAT GCC ATC AC-3' and (antisense) 5'-TCC ACC ACC CTG TTG CTG TA-3' (23). The distances between primers, including the primers, were 303 bp, 262 bp, and 450 bp, respectively. The PCR products were transferred to a nylon membrane and hybridized with internal biotin-labeled probes of HGF (5'-CAC ATG GAC AAG ATT-3') and c-met (5'-CGT CCT CTG GGA GCT-3'). The products on the membrane were detected by means of chemiluminescence. These procedures were previously described in detail (23).

Real-time RT-PCR

The levels of HGF mRNA were compared between isolated normal stromal cells and cancer stromal cells separated from primary cultures of endometrial cancer by means of real-time quantitative RT-PCR. PCR reactions were performed in the GeneAmp 5700 Sequence Detection System, which contains an Optical Detector and a GeneAmp PCR System 9600 (PE Applied Biosystems, Foster City, CA). The assay uses SYBERR Green I dye (SYBERR Green assay). The dye binds to double-stranded DNA during the extension phase of PCR, resulting in an increase in fluorescent emission, depending on the amount of PCR products. The GeneAmp 5700 system continuously measures the fluorescent spectra of all 96 wells of the thermal cycler and constructs amplification plots from the extension phase fluorescent emission data collected during PCR amplification. Threshold cycle values are calculated by determining the point at which the fluorescence exceeds a threshold limit. Primers were chosen with the assistance of the computer program, Primer Express (PE Applied Biosystems) and were located in 2 different exons as follows: HGF (sense): 5'-TTG GAT CAG GAC CAT GTG AGG-3' and (antisense): 5'-TCC ACG ACC AGG AAC AAT GA-3', GAPDH (sense) 5'-GAA GGT GAA GGT CGG AGT C-3' and (antisense) 5'-GAA GAT GGT GAT GGG ATT TC-3'. Amplification reactions (25 µl) contained 1 µl cDNA sample; 150 nM each primer; and 12.5 µl SYBER Green PCR Master Mix (the mix is x2 in concentration and contains SYBER Green I Dye, Ampli Taq Gold DNA polymerase, deoxy-nucleoside triphosphate, passive reference, optimized buffer components; PE Applied Biosystems). To generate standard curves of HGF and GAPDH, serial dilutions of a known amount of a cDNA sample were used. The threshold cycle values were plotted on these standard curves to obtain the amount of copies present in the initial cDNA sample. Each PCR amplification was performed in triplicate under the following conditions: 2 min at 50 C and 10 min at 95 C, followed by a total of 40 2-temperature cycles (15 sec at 95 C and 1 min at 60 C). Normalization to GAPDH was performed for each sample. The HGF/GAPDH ratio was calculated and reported as the means of triplicate assays. Dissociation curve analysis and gel electrophoresis were performed to confirm the absence of nonspecific bands and to determine correct size of the amplicons. Dissociation curves, which show the melting temperature of the products, were drawn by continuously collecting data obtained by slowly ramping the temperature of reaction solutions from 60 to 95 C. A lower melting temperature than that of the specific product indicates that nonspecific amplification is occurring. When the curve showed a lower melting temperature, the data were eliminated from the analysis.

Western blot analysis

Subconfluent HEC-1 and ISHIKAWA cells were lysed with ice-cold lysis buffer containing 50 mM Tris-HCl (pH 7.6), 150 mM NaCl, 0.1% SDS, 1 mM dithiothreitol, and 1 x Complete Protease Inhibitor Cocktail (Roche Molecular Biochemicals, Ingelheim, Germany). The lysate was centrifuged, and supernatant was prepared. Protein concentration of the supernatant was measured by Bradford assay. Eighty micrograms of the protein sample was resolved on a 4–20% gradient polyacrylamide gel by electrophoresis. These separated protein samples were then electroblotted onto a nitrocellulose membrane before blocking in 2.5% skimmed milk for 60 min. The membrane was then probed with anti-c-Met ß-chain antibodies (rabbit polyclonal; IBL, Gumma, Japan) at a dilution of 1:20 (5 µg/ml), followed by a peroxidase-conjugated secondary antibody (1:2000). Protein bands were visualized using an enhanced chemiluminescence system (Amersham Pharmacia Biotech). The protein band size was determined using a Full Range Rainbow (Amersham Pharmacia Biotech) molecular weight marker.

Immunohistochemical staining

In a cryostat, the frozen neoplastic and nonneoplastic tissues were cut into sections 8-µm thick, which were then thaw-mounted onto poly-L-lysine-coated slides. They were briefly air-dried and incubated in 0.3% hydrogen peroxide containing 70% methanol at room temperature for 15 min. Nonspecific activity was blocked by 1% normal goat serum at room temperature for 30 min. The slides were next incubated overnight at 4 C with either anti-HGF {alpha}-chain (rabbit polyclonal; IBL) or anti-c-Met (rabbit polyclonal; IBL) at a dilution of 1:20 (5 µg/ml). Treatment with secondary antibody fraction to rabbit Ig G [Histofine simple stain PO; Nichierei Co. (Japan), Tokyo, Japan] at room temperature for 30 min followed. The immunocomplex was visualized using diaminobenzidine chromogen in a peroxidase reaction kit [Nichierei Co. (Japan)]. Slides stained with the primary antibodies negative mixture were used as negative controls.

Preparation of conditioned media (CM)

To prepare CM from the subconfluent cancer cell lines and normal endometrial epithelial cells, these cells were washed three times and incubated in EMEM supplemented with 1% FBS for 48 h. The CM were collected and centrifuged to remove cell debris. The CM were stored at -20 C until needed for the experiments. At the time of the collection of CM, we confirmed that there were no significant differences of cell numbers among HEC-1, ISHIKAWA, and normal epithelial cells.

Induction of HGF gene in stromal cells

Isolated stromal cells were diluted with culture medium (DMEM/F-12 with 10% FBS) to a seeding density of 5 x 104/well and suspended in 24-well tissue culture plates. The cells were preincubated at 37 C for 48 h and allowed to reach subconfluence, and the culture medium exchanged to the EMEM medium was supplemented with 1% FBS for 24 h. TNF-{alpha}, IL-6, basic fibroblast growth factor (bFGF) (all from Genzyme Transgenics Corp., Cambridge, MA), or 17ß-E2 (Sigma) or CM from either a cancer cell line or normal epithelial cells with or without monoclonal antibody against bFGF (5 µg/ml, Sigma) were then added. After the cells were cultured for 6 h, total RNA was extracted. HGF gene expressions were compared by means of real-time RT-PCR, as described above.

Statistical analysis

To compare the values obtained with various treatments vs. the controls, results were analyzed using one-way ANOVA, followed by Fisher’s protected least-significant-difference test. The data are expressed as means ± SE. A level of P < 0.05 was accepted as indicating statistical significance.

Results

Expression of HGF and c-met in endometrial cancer cells

To determine whether the HGF/c-met pathway is involved in the progression of endometrial cancer, the gene expressions of HGF and c-met in the endometrial carcinoma cells of primary cultures and in HEC-1 and ISHIKAWA cell lines were examined by means of RT-PCR. HGF mRNA was not detected in the HEC-1 and ISHIKAWA cells. In the carcinoma cells that had been separated from primary cultures, HGF expressions were observed in only 3 of 10 cases. On the other hand, the c-met gene was revealed in 8 of 10 cases as well as in endometrial cancer cell lines, but differences were not detected among the histological grade of endometrial cancers (Fig. 1Go, A and B). As shown in Fig. 1CGo, high levels of 145-kDa c-met proteins, along with variable levels of its precursor (170 kDa) were detected in ISHIKAWA and HEC-1 cells. The c-met protein expression level in HEC-1 cells was higher than that in ISHIKAWA cells. Immunohistochemical study revealed positive staining for c-met in the carcinoma cells but none for stromal cells in the endometrial cancer tissue (Fig. 2Go). On the other hand, only faint and sporadic straining for c-met was observed in normal endometrial glands, consistent with the previous study (19) (data not shown).



View larger version (42K):
[in this window]
[in a new window]
  		Figure 1. A–C, Expressions of HGF and c-met receptor in endometrial cancer cells. Each gene expression in HEC-1, ISHIKAWA cell lines, and endometrial cancer cells separated from 10 cases of primary cultures were examined by RT-PCR and Southern blotting. The c-met protein expressions in endometrial cancer cell lines were determined by Western blot analysis. A, HGF gene expressions. Lane 1, HEC-1; lane 2, ISHIKAWA; lanes 3–12, expressions of HGF in cancer cells of primary cultures; lanes 3–5, 6–9, and 10–12, histological grades 1, 2, and 3 (G1, G2, and G3), respectively. B, c-met Receptor gene expressions. Lanes 1–10, Expressions of c-met gene in the same cases included in A. In 2 of 10 cases, c-met gene expression was not detected (data not shown). C, c-met Receptor protein expressions in ISHIKAWA and HEC-1 cell lines. Both the 145-kDa c-met protein and its precursor (170 kDa) were detected in ISHIKAWA and HEC-1 cells.

 


View larger version (146K):
[in this window]
[in a new window]
  		Figure 2. Immunohistochemical staining of c-met receptor in cryostat sections of endometrial cancer. A, The c-met receptor protein was detected in the cancer cells; B, negative control (x200).

 
Effects of HGF on endometrial cancer cell proliferation and invasion

Because endometrial carcinoma cells and cancer cell lines expressed the c-met receptor, we examined the effects of HGF on proliferation and invasion of HEC-1 and ISHIKAWA cells. After cells were treated with various concentrations of HGF, DNA synthesis and cell growth were measured by means of Brdu ELISA and MTT assay. The addition of 1–50 ng/ml HGF significantly increased proliferation of HEC-1 cells in a dose-dependent fashion. The increases in number of cells and DNA synthesis in the presence of 50 ng/ml HGF were 164% and 130% of control values, respectively. An antibody for HGF abrogated these stimulatory effects. In ISHIKAWA cells, we detected a subtle increase of cell proliferation (Fig. 3Go).



View larger version (31K):
[in this window]
[in a new window]
  		Figure 3. Effects of HGF on cell growth and DNA synthesis of HEC-1 and ISHIKAWA cells. HEC-1n and ISHIKAWA cells were seeded at a density of 2 x 103/well in a 96-well plate with EMEM serum-free medium, supplemented with 2 mg/ml BSA, along with various concentrations of HGF (0–50 ng/ml). After 48 h of incubation, effects on cell growth and DNA synthesis were evaluated by measuring MTT and Brdu incorporation. Results are presented as the percent of control value. Statistical significance of the differences vs. control value are indicated by: *, P < 0.01; **, P < 0.001.

 
The effect of HGF on the invasive capacity of the endometrial cancer cells was examined using the Matrigel invasion chamber assay. As shown in Fig. 4Go, addition of 1 and 10 ng/ml HGF significantly increased the number of penetrated HEC-1 and ISHIKAWA cells, 3.5 and 4.0 times more than those in the absence of HGF. These increased invasive activities induced by HGF were specifically negated by the addition of anti-HGF antibody. Adding anti-HGF antibody alone failed to affect cell proliferation and invasive ability, and the mouse Ig G did not influence the effects of HGF (data not shown).



View larger version (16K):
[in this window]
[in a new window]
  		Figure 4. Effects of HGF on the invasive capacity of HEC-1 and ISHIKAWA cells. The cells were seeded on 24-well Matrigel invasion chamber plates at a density of 5 x 104 cells/well and cultured in serum-free EMEM containing 2% BSA. HGF was added to the lower media; HEC-1 and ISHIKAWA cells were cultured for 24 and 72 h, respectively. The cell numbers per square centimeter that penetrated through Matrigel and porous membranes were used for comparison. Values are means ± SEM of three wells of each concentration of HGF. Statistical significance of the differences vs. control value are indicated by: *, P < 0.05, compared with control.

 
Increased HGF production in cancer stromal cells

Although HGF affects the progression of endometrial cancer, cancer cells did not express much HGF, suggesting that HGF mainly acts as a paracrine factor in endometrial cancer. We examined whether HGF would be produced in endometrial stromal cells and whether production would increase in cancer stromal cells as compared with normal endometrium. HGF gene expressions in cancer stromal cells that had been separated from primary cultures and in isolated normal stromal cells were compared by real-time RT-PCR. Amplification of the HGF gene was detected in all cases of normal and cancer stromal cells. The mean expression levels of HGF in cancer stromal cells were 6.4-fold more than in normal stromal cells (Fig. 5Go).



View larger version (11K):
[in this window]
[in a new window]
  		Figure 5. Comparison of HGF gene expression in cancer and normal stromal cells. HGF gene expressions in cancer stromal cells, separated from primary cultures, and in isolated normal stromal cells were compared by means of real-time RT-PCR. To neutralize to the GAPDH transcription levels, the HGF/GAPDH ratio was calculated and used for comparison. Data are means ± SEM of 10 samples of endometrial cancer stromal cells and eight samples of normal endometrial stromal cells. *, P < 0.001, compared with normal stromal cells.

 
Because the cancer stromal cells were less pure than normal endometrial stromal cells, a possible cause for the increased HGF transcription was that the resident carcinoma cells might affect the HGF mRNA expressions. To corroborate the result in Fig. 5Go, HGF protein expression levels in stromal cells of endometrial cancer and in normal endometrial tissues were compared by immunohistochemical staining. Although the staining was negative or weak in normal endometrial stromal cells, strong staining was observed in stromal cells associated with cancer cells in all cases that we tested (Fig. 6Go). In addition, positive staining for HGF was detected in carcinoma cells and in normal epithelial cells in some cases. In the cases presented in Fig. 6Go, normal epithelial cells were highly and uniformly positive, whereas the carcinoma cells were sporadically and weakly positive. The staining was seen as irrelevant to HGF gene expression in the carcinoma cells (Fig. 1AGo).



View larger version (138K):
[in this window]
[in a new window]
  		Figure 6. Immunohistochemical staining of HGF in cryostat sections of endometrial cancer (A) and normal endometrium (B). Negative control of endometrial cancer (C) and normal endometrium (D) (x 100).

 
Enhancement of HGF production by carcinoma-derived factor

The above results suggest that carcinoma cells, but not normal epithelial cells, might secrete factors that induce aberrant production of HGF in endometrial stromal cells. To confirm this, isolated normal endometrial stromal cells were cultured in the presence of CM from either normal endometrial epithelial cells or endometrial cancer cell lines. We used real-time RT-PCR to evaluate HGF mRNA expressions. The CM from HEC-1 and ISHIKAWA cells displayed a dose-dependent increase of HGF transcription. In addition, 3.4- and 2.7-fold increases were observed by adding 50% CM from HEC-1 and ISHIKAWA cells, respectively. However, CM from normal epithelial cell induced only a 1.4-fold increase of HGF gene expression (Fig. 7AGo).



View larger version (28K):
[in this window]
[in a new window]
  		Figure 7. Induction of HGF gene expression by CM from endometrial cancer cell lines and bFGF. Subconfluent normal endometrial stromal cells were cultured in EMEM medium supplemented with 1% FBS for 48 h. CM from normal endometrial epithelial cells or cancer cell lines (A) or EMEM containing bFGF (B) with or with out a bFGF antibody (5 µg/ml) were then added. After cells were cultured for 6 h, total RNA was extracted. HGF gene expressions were measured by means of real-time RT-PCR. The levels of HGF expressions were presented as the fold increases of the HGF/GAPDH ratio, relative to that of unstimulated normal stromal cells. Data are means ± SEM of triplicate measurements.

 
Induction of HGF production by bFGF in endometrial stromal cells

Finally, we investigated whether a particular molecule was a carcinoma-derived HGF inducer in endometrial stromal cells. We tested bFGF, IL-6, TNF-{alpha}, and E2, which are among the several factors previously reported as HGF inducers. The stimulatory effects of these factors on HGF gene expression in isolated endometrial stromal cells were examined. HGF mRNA expression was not enhanced by IL-6, TNF-{alpha}, or E2 (data not shown); bFGF (1 and 10 ng/ml) induced 1.5- and 3.2-fold increases of HGF transcription. These stimulatory effects of bFGF were completely abolished by adding bFGF antibody (Fig. 7BGo). On the other hand, the bFGF antibody partially negated the effects of CM from HEC-1 and ISHIKAWA cells but did not affect HGF-inducing ability of CM from normal endometrial cells (Fig. 7AGo).

Discussion

In the present study, we showed that HGF derived from endometrial stromal cells promotes the proliferation and invasion of endometrial cancer cells. We also demonstrated that HGF production is up-regulated in cancer stromal cells but not in normal endometrial stromal cells. Furthermore, we identified bFGF as a carcinoma cell-derived factor that induces HGF production in endometrial stromal cells. We determined that cancer cells induce HGF production in stromal cells by secreting stimulating factor, and stromal-derived HGF enhances the proliferation and invasion of endometrial cancer cells via a c-met receptor that is highly expressed in the cancer cells. Thus, we demonstrated that a mutual interaction between endometrial cancer and its stromal cells is mediated by cancer-derived bFGF and stromal-derived HGF.

Endometrial carcinoma cell lines did not express the HGF gene, as previously demonstrated in normal endometrial epithelial cells (12), and HGF mRNA was detected in all the stromal cells we examined. Consequently, HGF is likely produced mainly in endometrial stromal cells. In this study, however, HGF transcripts were detected in 3 of 10 carcinoma cells obtained from primary cultures. Because the purity of carcinoma cells separated from the primary cultures was 80–90%, the cells were contaminated by the stromal cells. The positive HGF expression in carcinoma cells may be attributable to stromal cell contamination. Another explanation would be that HGF expression was reported in lung and breast carcinomas (25, 26); therefore, some endometrial carcinoma cells would acquire the ability to express HGF during malignant transformation. On the other hand, although c-met receptor expressions were revealed in most carcinoma cells and endometrial cancer cell lines, we could not detect any differences among the histological grades of endometrial cancers. This is in contrast with an earlier study that showed the correlation of c-met expression levels and histological grade of endometrial cancers (19). The reason for the inconsistent findings may arise from a methodological problem. We used RT-PCR to determine the expression of c-met gene, but quantitative measurement of gene expression is limited with this method.

HGF stimulated HEC-1 cell growth and DNA synthesis, whereas it had no evident effect on proliferation of ISHIKAWA cells. Although HGF enhances the invasive ability of both endometrial carcinoma cell lines, the mitogenic activity of HGF on these cells seems to be cell type-specific variations of carcinoma cells. Another possible explanation is that the c-met receptor expression level was lower in ISHIKAWA cells than in HEC-1 cells, as shown in Fig. 1CGo, and the HGF enhancement of invasive ability was rather weak in ISHIKAWA cells, as indicated in Fig. 4Go. It seems likely that the cell reaction for HGF may correlate with the amount of c-met receptor.

In our immunohistochemical studies, we observed positive staining of HGF in some endometrial carcinoma cells and in normal epithelial cells, which is consistent with a previous report (19). In contrast, the results of RT-PCR analysis showed negative HGF expression in most endometrial carcinomas in this study, and in normal epithelial cells as demonstrated in the earlier report (12). This discrepancy between the results of immunohistochemical staining and RT-PCR may arise from the presence of HGF in the extracellular matrix. HGF is a heparin-binding growth factor (27). HGF binds to a heparin sulfate-modified CD 44, which concentrates HGF and protects from proteolytic degradation at the cell surface (28, 29). The immunoreactive HGF detected in normal epithelial cells and cancer cells may be proteins that bind to heparin sulfate-modified CD 44. The c-met receptor may be another cell surface binding site of HGF. Because the alteration of c-met expression levels in ovine endometrium during menstrual cycle was demonstrated (30), the difference of HGF staining in normal epithelial cells may depend on the levels of c-met expression, which may be controlled by the hormonal status of each case.

Our results in Fig. 7Go indicate that neoplastic endometrial epithelial cells secrete certain HGF inducers. We first identified that bFGF is a HGF-inducer in endometrial cancer. It is reported that bFGF is expressed in endometrial cancer cell lines, including HEC-1 and ISHIKAWA, and stimulates endometrial carcinoma cell proliferation (31); bFGF also enhanced the neovascularization and tumor growth in vivo (32). It was also observed that bFGF was overexpressed mainly in glandular epithelium of complex hyperplasia as well as carcinoma cells (33). These previous findings suggest that bFGF increased during malignant transformation and may promote the development of endometrial cancer in an autocrine manner. Our present data indicate that aberrant expression of bFGF in tumor cells may also act as a paracrine factor to stimulate HGF expression in the surrounding stromal cells. This, in turn, results in promoting progression to a more malignant phenotype.

Concerning the regulation of HGF production in endometrial stromal cells, the involvement of a PKC-dependent pathway was revealed in an earlier study (34). The expression of the FGF receptor-1, which is one of the high-affinity receptors for bFGF, and PLC activation by bFGF in endometrial stromal cells was also demonstrated (35). Because PLC produces diacylglycerol to resolve phosphatidylinositol 4,5-bisphosphate, diacylglycerol activates PKC. These data support our findings. Carcinoma cell-derived bFGF may bind to FGF receptor-1 expressed in endometrial stromal cells and may induce HGF production through a PKC-dependent pathway.

Although these results suggest that bFGF was one of the HGF inducers in endometrial stromal cells, the bFGF antibody did not show complete negation of HGF induction promoted by the CM of HEC-1 and ISHIKAWA cells. This suggests the existence of other HGF inducers in endometrial stromal cells. Several inducers, such as IL-1ß, IL-6, TNF-{alpha}, acidic FGF (FGF-1), bFGF (FGF-2), HST-1 (FGF-4), platelet-derived growth factor, and PGs, were reportedly observed in human fibroblasts (36, 37, 38, 39). In the murine ovary, estrogen stimulates HGF transcription (40). Therefore, we tested the effect of IL-6, TNF-{alpha}, bFGF, and estrogen on HGF production in endometrial stromal cells. Except for bFGF, we could not detect a stimulatory effect of these factors on HGF production. One author failed to demonstrate the stimulatory effects of several HGF inducers involved in HGF synthesis in endometrial stromal cells (34). These data indicate that the factors regulating HGF in endometrial stromal cells differ from those in other mesenchymal cell types.

Our current study and earlier studies suggest that malignant transformation of endometrial epithelial cells leads to overexpression of bFGF and disrupts homeostatic regulation. This could result in enhanced HGF production in stromal cells. Aberrant expression of HGF stimulates the invasive growth of endometrial cancer. The mutual interactions between carcinoma cells and host stromal cells, mediated by carcinoma-derived bFGF and stromal-derived HGF, may play a significant role in the progression of endometrial cancer.

Acknowledgments

We are grateful to Dr. M. Nishida for kindly providing the ISHIKAWA cell line and to Drs. Y. Ohata, M. Ito, and T. Tsudo for technical assistance.

Footnotes

Abbreviations: bFGF, Basic fibroblast growth factor; Brdu, 5-bromo-2'-deoxyuridine; CM, conditioned media; DMEM/F-12, DMEM/Ham’s F-12; EMEM, Eagle’s minimum essential medium; GAPDH, glycerol-3-phosphate dehydrogenase; HGF, hepatocyte growth factor; MTT, 3-[4,5-dimethylthyazol-2-yl]-2,5-diphenyl tetrazolium bromide.

Received April 16, 2001.

Accepted February 5, 2002.

References

  1. Camps JL, Chang S, Hus TC, Freeman MR, Hong SJ, Zhau HE, von Eschenbach AC, Chung LW 1990 Fibroblast-mediated acceleration of human epithelial tumor growth in vivo. Proc Natl Acad Sci USA 87:75–79[Abstract/Free Full Text]
  2. Hong WK, Sporn MB 1997 Recent advances in chemoprevention of cancer. Science 278:1073–1077[Abstract/Free Full Text]
  3. Kinzler KW, Vogelstein B 1998 Landscaping the cancer terrain. Science 280:1036–1037[Free Full Text]
  4. Nakamura T, Nawa K, Ichihara A 1984 Partial purification and characterization of hepatocyte growth factor from serum of hepatectomized rats. Biochem Biophys Res Commun 122:1450–1459[CrossRef][Medline]
  5. Nakamura T, Nishizawa T, Hagiya M, Seki T, Shimonishi M, Sugimura A, Tashiro K, Shimizu S 1989 Molecular cloning and expression of human hepatocyte growth factor. Nature 342:440–443[CrossRef][Medline]
  6. Nakamura T 1991 Structure and function of hepatocyte growth factor. Prog Growth Factor Res 3:67–86[CrossRef][Medline]
  7. Montesano R, Matsumoto K, Nakamura T, Orci L 1991 Identification of a fibroblast-derived epithelial morphogen as hepatocyte growth factor. Cell 67:901–908[CrossRef][Medline]
  8. Barros EJ, Santos OF, Matsumoto K, Makamura T, Nigam SK 1995 Differential tubulogenic and branching morphogenic activities of growth factors. Proc Natl Acad Sci USA 92:4412–4416[Abstract/Free Full Text]
  9. Bussolino F, DiRenzo MF, Ziche M, Bocchietto E, Olivero M, Naldini L, Gaudino G, Tamagnone L, Coffer A, Comoglio PM 1992 Hepatocyte growth factor is a potent angiogenic factor which stimulates endothelial cell motility and growth. J Cell Biol 119:629–641[Abstract/Free Full Text]
  10. Grant DS, Kleinman HK, Goldberg ID, Bhargava MM, Nickoloff BJ, Kinsella JL, Polverini P, Rosen EM Scatter factor induces blood vessel formation in vivo. Proc Natl Acad Sci USA 90:1937–1941
  11. Bottaro DP, Rubin JS, Faletto DL, Chan AM, Kmiecik TE, Vande Woude GF, Aaronson SA 1991 Identification of the hepatocyte growth factor receptor as the c-met proto-oncogene product. Science 251:802–804[Abstract/Free Full Text]
  12. Sugawara J, Fukaya T, Murakami T, Yoshida H, Yajima A 1997 Hepatocyte growth factor stimulated proliferation, migration, and lumen formation of human endometrial epithelial cells in vitro. Biol Reprod 57:936–942[Abstract]
  13. Prat M, Narsimhan RP, Crepaldi T, Nicotra MR, Natali PG, Comoglio PM 1991 The receptor encoded by the human c-MET oncogene is expressed in hepatocytes, epithelial cells and solid tumors. Int J Cancer 49:323–328[Medline]
  14. Corps AN, Sowter HM, Smith SK 1997 Hepatocyte growth factor stimulates motility, chemotaxis and mitogenesis in ovarian carcinoma cells expressing high levels of c-met. Int J Cancer 73:151–155[CrossRef][Medline]
  15. Furukawa T, Duguid WP, Kobari M, Matsuno S, Tsao MS 1995 Hepatocyte growth factor and Met receptor expression in human pancreatic carcinogenesis. Am J Pathol 147:889–895[Abstract]
  16. Natali PG, Prat M, Nicotra MR, Bigotti A, Olivero M, Comoglio PM, Di Renzo MF 1996 Overexpression of the met/HGF receptor in renal cell carcinomas. Int J Cancer 69:212–217[CrossRef][Medline]
  17. Beviglia L, Matsumoto K, Lin CS, Ziober BL, Kramer RH 1997 Expression of the c-Met/HGF receptor in human breast carcinoma: correlation with tumor progression. Int J Cancer 74:301–309[CrossRef][Medline]
  18. Humphrey PA, Zhu X, Zarnegar R, Swanson PE, Ratliff TL, Vollmer RT, Day ML 1995 Hepatocyte growth factor and its receptor (c-MET) in prostatic carcinoma. Am J Pathol 147:386–396[Abstract]
  19. Wagatsuma S, Konno R, Sato S, Yajima A 1998 Tumor angiogenesis, hepatocyte growth factor, and c-Met expression in endometrial carcinoma. Cancer 82:520–530[CrossRef][Medline]
  20. Nakamura T, Matsumoto K, Kiritoshi A, Tano Y, Nakamura T 1997 Induction of hepatocyte growth factor in fibroblasts by tumor-derived factors affects invasive growth of tumor cells: in vitro analysis of tumor-stromal interactions. Cancer Res 57:3305–3313[Abstract/Free Full Text]
  21. Matsumoto-Taniura N, Matsumoto K, Nakamura T 1999 Prostaglandin production in mouse mammary tumour cells confers invasive growth potential by inducing hepatocyte growth factor in stromal fibroblasts. Br J Cancer 81:194–202[CrossRef][Medline]
  22. Yoshioka H, Harada T, Iwabe T, Nagano Y, Taniguchi F, Tanikawa M, Terakawa N 1999 Menstrual cycle-specific inhibition of the proliferation of endometrial stromal cells by interleukin 6 and its soluble receptor. Am J Obstet Gynecol 180:1088–1094[CrossRef][Medline]
  23. Iwabe T, Harada T, Tsudo T, Tanikawa M, Onohara Y, Terakawa N 1998 Pathogenetic significance of increased levels of interleukin-8 in the peritoneal fluid of patients with endometriosis. Fertil Steril 69:924–930[CrossRef][Medline]
  24. Hori K, Shiota G, Kawasaki H 2000 Expression of hepatocyte growth factor and c-met receptor in gastric mucosa during gastric ulcer healing. Scand J Gastroenterol 35:23–31[CrossRef][Medline]
  25. Tsao M, Zhu H, Giaid A, Viallet J, Nakamura T, Park M 1993 Hepatocyte growth factor is an autocrine factor for human normal bronchial epithelial and lung carcinoma cells. Cell Growth Differ 4:571–579[Abstract]
  26. Wang Y, Selden C, Morgan N, Stamp G, Hodgson H 1994 Hepatocyte growth factor/scatter factor expression in human mammary epithelium. Am J Pathol 144:675–682[Abstract]
  27. Nakamura T, Teramoto H, Ichihara A 1986 Purification and characterization of a growth factor from rat platelets for mature parenchymal hepatocytes in primary cultures. Proc Natl Acad Sci USA 83:6489–6493[Abstract/Free Full Text]
  28. Bennett KL, Jackson DG, Simon JC, Tanczos E, Peach R, Modrell B, Stamenkovic I, Plowman G, Aruffo A 1995 CD44 isoforms containing exon V3 are responsible for the presentation of heparin-binding growth factor. J Cell Biol 128:687–698[Abstract/Free Full Text]
  29. van der Voort R, Taher TE, Wielenga VJ, Spaargaren M, Prevo R, Smit L, David G, Hartmann G, Gherardi E, Pals ST 1999 Heparin sulfate-modified CD44 promotes hepatocyte growth factor/scatter factor-induced signal transduction through the receptor tyrosine kinase c-Met. J Biol Chem 274:6499–6506[Abstract/Free Full Text]
  30. Chen C, Spencer TE, Bazer FW 2000 Expression of hepatocyte growth factor and its receptor c-met in the ovine uterus. Biol Reprod 62:1844–1850[Abstract/Free Full Text]
  31. Fujimoto J, Hori M, Ichigo S, Hirose R, Sakaguchi H, Tamaya T 1997 Plausible novel therapeutic strategy of uterine endometrial cancer with reduction of basic fibroblast growth factor secretion by progestin and O-(chloroacetylcarbamoyl) fumagillol (TNP-470; AGM-1470). Cancer Lett 113:187–194[CrossRef][Medline]
  32. Coltrini D, Gualandris A, Nelli EE, Parolini S, Molinari-Tosatti MP, Quarto N, Ziche M, Giavazzi R, Presta M 1995 Growth advantage and vascularization induced by basic fibroblast growth factor overexpression in endometrial HEC-1-B cells: an export-dependent mechanism of action. Cancer Res 55:4729–4738[Abstract/Free Full Text]
  33. Gold LI, Saxena B, Mittal KR, Marmor M, Goswami S, Nactigal L, Korc M, Demopoulos RI 1994 Increased expression of transforming growth factor ß isoforms and basic fibroblast growth factor in complex hyperplasia and adenocarcinoma of the endometrium: evidence for paracrine and autocrine action. Cancer Res 54:2347–2358[Abstract/Free Full Text]
  34. Nasu K, Sugano T, Matsui N, Narahara H, Kawano Y, Miyakawa I 1999 Expression of hepatocyte growth factor in cultured human endometrial stromal cells is induced through a protein kinase C-dependent pathway. Biol Reprod 60:1183–1187[Abstract/Free Full Text]
  35. Sangha RK, Li XF, Shams M, Ahmed A 1997 Fibroblast growth factor receptor-1 is a critical component for endometrial remodeling: localization and expression of basic fibroblast growth factor and FGF-R1 in human endometrium during the menstrual cycle and decreased FGF-R1 expression in menorrhagia. Lab Invest 77:389–402[Medline]
  36. Matsumoto K, Okazaki H, Nakamura T 1992 Up-regulation of hepatocyte growth factor gene expression by interleukin-1 in human skin fibroblasts. Biochem Biophys Res Commun 188:235–243[CrossRef][Medline]
  37. Matsumoto K, Okazaki H, Nakamura T 1995 Novel function of prostaglandins as inducers of gene expression of HGF and putative mediators of tissue regeneration. J Biochem 117:458–464[Abstract/Free Full Text]
  38. Matsumoto K, Tajima H, Hamanoue M, Kohno S, Kinoshita T, Nakamura T 1992 Identification and characterization of "injurin," an inducer of expression of the gene for hepatocyte growth factor. Proc Natl Acad Sci USA 89:3800–3804[Abstract/Free Full Text]
  39. Gohda E, Matsunaga T, Kataoka H, Takebe T, Yamamoto I 1994 Induction of hepatocyte growth factor in human skin fibroblasts by epidermal growth factor, platelet-derived growth factor and fibroblast growth factor. Cytokine 6:633–640[CrossRef][Medline]
  40. Liu Y, Lin L, Zarnegar R 1994 Modulation of hepatocyte growth factor gene expression by estrogen in mouse ovary. Mol Cell Endocrinol 104:173–181[CrossRef][Medline]



This article has been cited by other articles:


Home page
 Hum ReprodHome page
S. Battersby, K.J. Sales, A.R. Williams, R.A. Anderson, S. Gardner, and H.N. Jabbour
Seminal plasma and prostaglandin E2 up-regulate fibroblast growth factor 2 expression in endometrial adenocarcinoma cells via E-series prostanoid-2 receptor-mediated transactivation of the epidermal growth factor receptor and extracellular signal-regulated kinase pathway
Hum. Reprod., January 1, 2007; 22(1): 36 - 44.
[Abstract] [Full Text] [PDF]

Home page
 Reproductive SciencesHome page
M. B. Jones, A. P. Houwink, B. K. Freeman, T. M. Greenwood, J. M. Lafky, W. L. Lingle, A. Berchuck, G. L. Maxwell, K. C. Podratz, and N. J. Maihle
The Granulin-Epithelin Precursor Is a Steroid-Regulated Growth Factor in Endometrical Cancer
Reproductive Sciences, May 1, 2006; 13(4): 304 - 311.
[Abstract] [PDF]

Home page
 Hum ReprodHome page
S. Yoshida, T. Harada, T. Iwabe, F. Taniguchi, M. Mitsunari, N. Yamauchi, I. Deura, S. Horie, and N. Terakawa
A combination of interleukin-6 and its soluble receptor impairs sperm motility: implications in infertility associated with endometriosis
Hum. Reprod., August 1, 2004; 19(8): 1821 - 1825.
[Abstract] [Full Text] [PDF]

Home page
 J. Clin. Endocrinol. Metab.Home page
S. Yoshida, T. Harada, M. Mitsunari, T. Iwabe, Y. Sakamoto, S. Tsukihara, Y. Iba, S. Horie, and N. Terakawa
Hepatocyte Growth Factor/Met System Promotes Endometrial and Endometriotic Stromal Cell Invasion via Autocrine and Paracrine Pathways
J. Clin. Endocrinol. Metab., February 1, 2004; 89(2): 823 - 832.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a related Letter to the Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Yoshida, S.
Right arrow Articles by Terakawa, N.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Yoshida, S.
Right arrow Articles by Terakawa, N.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Endocrinology Endocrine Reviews J. Clin. End. & Metab.
Molecular Endocrinology Recent Prog. Horm. Res. All Endocrine Journals