Role of emmprin in endometrial cancer

Patients with normal endometrium (n = 20), endometrial hyperplasia (n = 10) (simple, n = 1; complex, n = 2; atypical complex, n = 7), adenocarcinoma and carcinosarcoma (n = 134) were treated at Okayama University Hospital between January 2000 and November 2009. All of the patients underwent a total abdominal hysterectomy, bilateral salpingo-oophorectomy and partial omentectomy with or without pelvic and/or para-aortic lymphadenectomy. Pelvic lymph node dissection included the right and left common iliac, external iliac, suprainguinal, internal iliac, obturator, sacral and parametrial nodal chains. Para-aortic lymph node dissection included the nodes located from the bifurcation of the aorta to the level of the renal vein (n = 36). Adjuvant chemotherapy was used depending the FIGO stage, grade, patient preference and physician discretion. Our standard chemotherapy consisted of paclitaxel (175 mg/m² infused over 3 h) and carboplatin (dosed for an area under the curve of 5) for 3–6 cycles (n = 52). Patients with neoadjuvant chemotherapy were excluded from this study. Tumor specimens were obtained at the time of surgery, immediately fixed in 10% neutral-buffered formalin and embedded in paraffin. The histological cell types were diagnosed according to the WHO classification, and all 134 cancer specimens were classified as endometrioid adenocarcinomas and carcinosarcoma. The histological grades were assigned according to the International Federation of Gynecology and Obstetrics (FIGO) staging classification as follows: grade 1, n = 51; grade 2, n = 55; grade 3, n = 22; and carcinosarcoma, n = 6) The surgical stages were reviewed based on the FIGO staging system as follows: stage I, n = 74; stage II, n = 11; stage III, n = 39; stage IV, n = 10. The median age at the time of surgery was 57.7 years (range, 28–85). The disease-free survival (DFS) and overall survival (OS) rates were defined as the interval between the initial operation and clinically or radiologically proven recurrence or death, respectively. This study was approved by the Institutional Review Board of Okayama University Hospital.

Formalin-fixed paraffin-embedded sections (4-μm thickness) were deparaffinized with xylene and rehydrated in ethanol. Endogenous peroxidase activity was quenched by treatment with methanol containing 0.3% hydrogen peroxidase for 15 min. The sections were then incubated with an anti-emmprin antibody (Santa Cruz Biotechnology, Santa Cruz, CA) at room temperature, followed by staining with a streptavidin-biotin -peroxidase kit (Nichirei, Tokyo, Japan). The sections were counterstained with hematoxylin. The levels of emmprin staining in epithelial cells were classified into three groups by scoring the percentages of positive cells: 2, strong, >50% of cells stained; 1, moderate, 10–50% of cells stained; 0, weak, <10% of cells stained. Microscopic analyses were independently conducted by two independent examiners with no prior knowledge of the clinical data. Final decisions in controversial cases were made using a conference microscope.

The HEC-50B (Japanese Collection of Research Bioresources (JCRB) number: JCRB1145) and KLE (ATCC number: CRL-1622) cell lines evaluated were derived from human endometrial cancers. The HEC-50B cell line was maintained in Dulbecco’s modified eagle’s medium (DMEM) (Life Technologies, Grand Island, NY) supplemented with 10% fetal bovine serum (FBS). The KLE cell line was maintained in DMEM/F12 medium (Life Technologies, Grand Island, NY) supplemented with 10% FBS. HEC-50B and KLE cells were trypsinized and plated in culture dishes. At ~50% confluency, the cells were transfected with an annealed emmprin siRNA (sc-35298) or scramble siRNA as an empty vector for gene silencing (final concentration, 100 nmol/L) using a siRNA transfection reagent (sc-29528; Santa Cruz Biotechnology). The cells were cultured with each siRNA for 24, 48 and 72 h.

Cell lysates were collected and estimated using a Protein Assay system (Bio-Rad, Hercules, CA) according to the manufacturer’s protocols. Proteins from each cell line were subjected to SDS-PAGE, and then transferred onto nitrocellulose membranes. The polyclonal and monoclonal antibodies used for immunoblotting were as follows: anti-emmprin, anti-NF-κB p65 and anti-NF-κB phospho-p65 (p-p65) (Santa Cruz Biotechnology, Santa Cruz, CA); and anti-β-actin (Sigma Chemical Co., St. Louis, MO). The working dilution of all the primary antibodies was 1:1000. The membranes were then incubated with appropriate secondary antibodies. The resulting antigen-antibody complexes were detected with an enhanced chemiluminescence kit (Amersham Biosciences, Piscataway, NJ).

To evaluate the effects of emmprin on cell proliferation, MTS and cell viability assays were performed. For the MTS assay, cells were seeded into 96-well plates and transfected when the cell density reached 5 × 10⁴ cells/well. After transfection of the cells with the emmprin siRNA for 24, 48 and 72 h, MTS (Promega, Madison, WI) was added for 2 h. The absorbances were measured at 490 nm using an ELISA plate-reader (Bio-Rad Systems, Hercules, CA). The cell viability was analyzed using SYTO 10 green fluorescent nucleic acid stain and dead red (ethidium homodimer-2) nucleic acid stain (Live/Dead^® reduced biohazard viability/cytotoxicity kit; Invitrogen, Eugene, OR). Briefly, cells were transfected with the emmprin siRNA for 72 h and then incubated with SYTO 10 green fluorescent nucleic acid stain and dead red nucleic acid stain for 15 min. The cell fluorescence was observed using a fluorescence microscope (Olympus, Tokyo, Japan).

To investigate the effects of the emmprin siRNA on cell motility and invasion, we used monolayer wounding (scratch) and chemotaxicells (polycarbonate filter, pore size 8 μm; Kurabo, Osaka, Japan) coated with type IV collagen. For the monolayer wounding assay, cells were allowed to form a monolayer on a culture dish, and a wound was made by scratching the monolayer with a pipette tip. After removal of the scratched cells and transfection with the emmprin siRNA, the cells were cultured for 24 h. For the chemotaxicells coated with type IV collagen, cells (1 × 10⁵) in 100 μL of DMEM/F12 medium or DMEM supplemented with 0.1% bovine serum albumin were placed in the upper compartment, transfected with the emmprin siRNA and then incubated for 72 h. After the incubation, the cells on the upper surface of the filter were wiped off with a cotton swab. The cells on the lower surface were stained with hematoxylin and counted in 10 randomly selected fields.

To investigate the differences in the Matrigel invasion abilities among cells expressing emmprin, we used BD BioCoat Matrigel Invasion Chambers (BD Bioscience, Bedford, MA). HEC-50B and KLE cells transfected with the emmprin siRNA were added in situ with 10 μg/mL of DiI (Invitrogen) in DMEM/F12 or DMEM supplemented with 10% FBS for 1 h. Cells (5 × 10⁴) of each genotype were added to the inserts, and 0.75 mL of medium was added to the bottom of each well. After 72 h of incubation, the membranes were removed from the insert and mounted on slides, and the numbers of invading cells were counted under a microscope. The Matrigel assays were performed in triplicate.

Total RNA was extracted from the cell lines using an acid guanidinium –phenol -chloroform method (ISOGEN; Nippon Gene, Tokyo, Japan) according to the manufacturer's instructions. Real-time RT-PCR was performed using a LightCycler rapid thermal cycler instrument (Roche Diagnostics, Mannheim, Germany) under the conditions recommended by the manufacturer. The real-time RT-PCR used primers for emmprin, EGF and TGF-β as previously described [30, 31]. The PCR products were checked by melting point analyses and their electrophoretic mobilities. Standard curves for calculation of the numbers of transcripts were created using plasmids containing the respective amplified fragments as inserts, and were adjusted to use glyceraldehyde-3-phosphate dehydrogenase (GAPDH) as a reference gene. In addition, the PCR products were analyzed by 1.5% agarose gel electrophoresis. As an internal control, GAPDH mRNA was also measured by quantitative RT-PCR. The quantitative RT-PCR used primers for MMP-2, MMP-9, VEGF, E-cadherin, Vimentin, Snail and GAPDH as described previously [32].

pNF-κB-responsive and pAP-1-responsive elements were used for NF-κB and AP-1 signaling reporter assays, respectively. pNF-κB-Luc and pAP-1-Luc were purchased from Clontech (Palo Alto, CA). Transient transfections were performed using Lipofectamine™ 2000 reagent (Life Technologies). For the luciferase reporter assays, cells were transfected with 0.5 μg of NF-κB-responsive plasmid, AP-1-responsive plasmid, estrogen-responsive plasmid or progesterone-responsive plasmid in combination with 0.05 μg of pTK-RLUC (Promega) as an internal control. Their proteins were extracted using a Dual-Luciferase reporter assay system (Promega). The firefly and Renilla luciferase activities were measured concurrently for 12 sec using a luminometer (LUMAT LB9507; Berthold, Wildbad, Germany). The assays were carried out for quadruplicate transfection experiments, and at least three independent values were analyzed to confirm reproducibility.

For evaluation of cell growth in monolayers, cells were plated at a density of 3 × 10⁴ cells/well in 6-well plates containing DMEM or DMEM/F12 supplemented with 10% FBS. The cell numbers were counted in triplicate after 1, 3, 5 and 7 days using a hemocytometer to assess cell proliferation.

Statistical analyses were performed using the Mann–Whitney U-test for comparisons with controls and one-factor ANOVA followed by Fisher's protected least significance difference test for all pairwise comparisons. The survival rates were calculated by the Kaplan–Meier method, and the differences between the survival curves were examined by using the log-rank test. The analyses were performed with the software package StatView version 5.0 (Abacus Concepts, Berkeley, CA). Differences were considered significant at p < 0.05.

Emmprin-dependent activation may have a role in conventional endometrial tumorigenesis, and therefore the emmprin activation machinery might be important. The expression levels of emmprin were examined in human endometrial tissues by immunostaining. Figure 1 show representative immunostaining patterns of emmprin. Weak epithelial staining was observed in 25 cases (15.2%), moderate staining in 64 cases (38.9%) and strong staining in 76 cases (45.9%). The mean scores of the epithelial staining for emmprin were 0.85 for normal human endometrium, 0.9 for hyperplasia and 1.42 for cancer samples. Interestingly, endometrial cancer had the strongest emmprin expression compared with normal human endometrium and endometrial hyperplasia (p < 0.05, Mann–Whitney U-test) (Figure 1C).

Table 1 shows the distribution of cases scored as positive for each of the biological parameters examined, according to the clinicopathological characteristics in the overall population. As expected, the expression of emmprin had significant associations with clinicopathological parameters such as FIGO stage (p = 0.009), histology (p = 0.017), depth of myometrial invasion (p = 0.001), cervical involvement (p = 0.001), lymph node metastasis (p < 0.001), lymph vascular space (LVS) involvement (p < 0.001) and peritoneal cytology (p = 0.031), whereas age and ovarian metastasis did not have significant associations (p < 0.05, Mann–Whitney U-test). Emmprin was most significantly associated with the DFS and OS rates among the prognostic factors using the log-rank test for endometrial cancer. Figure 2 shows the DFS and OS curves of the 134 patients with endometrial cancer, according to the emmprin expression status. The DFS and OS rates of patients with high emmprin expression (score 2) were significantly higher than those of patients with low emmprin expression (scores 0–1) (DFS: p < 0.001; OS: p < 0.001). Furthermore, we examined independent prognostic factor for DFS and OS of the clinicopathologic factors including stage, histology, lymph node metastasis, deep myometrial invasion, ovarian metastasis and emmprin by multiple analysis. The ovarian metastasis was strongest independent prognostic factor for DFS and OS by multiple analysis (P = 0.0245 and P = 0.0222). However, emmprin expression was not an independent prognostic factor for DFS and OS.

We first examined the protein levels of emmprin in human endometrial cancer cell lines. As shown in Figure 3A, emmprin protein was highly expressed in KLE, HEC-50B and Ishikawa cells, weakly expressed in HEC-251 cells and absent in HEC-1A cells. Among the endometrial cancer cell lines, we chose KLE and HEC-50B endometrial cancer cells for further study. Use of the emmprin siRNA to trigger differentiation in endometrial cancer cells led to a time-dependent reduction in the emmprin mRNA that peaked around 72 h (Figure 3B). The expression of emmprin protein was significantly decreased in KLE and HEC-251 cells after transfection with the emmprin siRNA for 72 h (Figure 3C). The cell proliferation of HEC-50B and KLE cells transfected with the emmprin siRNA for 72 h was significantly inhibited as evaluated by MTS assays (Figure 3D). The cell viabilities of HEC-50B and KLE cells were evaluated after transfection with the emmprin siRNA. The percentages of viable cells were decreased to 72.1% (HEC-50B) and 53.6% (KLE) of the wild-type cell viabilities after 72 h of transfection with the emmprin siRNA (Figure 3E).

The purpose of these experiments was to study the role of endometrial cancers in the motility, invasiveness and Matrigel invasion of cells transfected with the emmprin siRNA. The cells transfected with the emmprin siRNA were much slower than cells transfected with the wild-type or empty vector in the motility and invasiveness assay (Figures 4A and 4B). These findings suggest that emmprin is associated with cell adhesion and spreading. The percentages of cells reaching the bottom of the filter were decreased to 61.5% and 48.9% in the Matrigel invasion assay after transfection of the emmprin siRNA for 72 h into HEC-50B and KLE cells, respectively (Figure 4C).

MMP activities have been implicated in EMT by activating growth factors and their receptors and cleaving cell-cell and cell-ECM adhesion molecules. We examined the transcriptional repressors of EMT families and each growth factor. The expression of TGF-β and EGF were decreased after transfection of the emmprin siRNA into HEC-50B and KLE cells (Figure 5A). Loss of cell-cell adhesion is a prerequisite for EMT and involves functional loss of E-cadherin. Transfection of the emmprin siRNA caused significant increases in the expression of E-cadherin in HEC-50B and KLE cells. Moreover, the expression of E-cadherin was also increased, accompanied by decreased expression of Vimentin after transfection of the emmprin siRNA into HEC-50B and KLE cells. Furthermore, the expression of Snail was significantly decreased after transfection of the emmprin siRNA into KLE cells (Figure 5B).

Emmprin can promote tumor cell invasion via activation of NF-κB and JNK pathways. The activation of AP-1 transcriptional activity may be induced by signaling mediated by JNK and extracellular signal-regulated kinase (ERK) pathways. In the present study, we examined the effects of the emmprin siRNA on the transcriptional activities of NF-κB and AP-1 in HEC-50B and KLE cells. The NF-κB transcriptional activities were decreased after transfection of the emmprin siRNA into HEC-50B and KLE cells. However, the AP-1 transcriptional activities were not decreased after transfection of the emmprin siRNA into HEC-50B and KLE cells (Figure 5C). NF-κB p65 was not affected by transfection of the emmprin siRNA into HEC-50B and KLE cells. However, NF-κB p-p65 was significantly decreased after transfection of the emmprin siRNA into KLE and HEC-251 cells (Figure 5D). NF-κB p50, JNK and ERK were not decreased after transfection of the emmprin siRNA into KLE and HEC-251 cells (data not shown).

VEGF, MMP-2 and MMP-9 are thought to be critically involved in the processes of tumor cell migration, invasion and metastasis. We examined VEGF, MMP-2 and MMP-9 expression after transfection of the emmprin siRNA into HEC-50B and KLE cells. Transfection of the emmprin siRNA caused significant decreases in the expression of MMP-2 and VEGF in HEC-50B and KLE cells. The expression of MMP-9 was significantly decreased by transfection of the emmprin siRNA into KLE cells (Figure 5E).

The effects of emmprin expression on cell proliferation were analyzed after transfection of the emmprin siRNA into the HEC-50B and KLE endometrial cancer cell lines. We found significant inhibitory effects on cell proliferation after transfection of the emmprin siRNA into HEC-50B and KLE cells compared with transfection of the wild-type or empty vector (p < 0.05) (Figure 5F).