Samples of 45 endometrial adenocarcinomas and 13 normal endometrium tissues were obtained from surgical pathology specimens at the Department of Gynecology, Qilu Hospital of Shandong University. The samples were immediately frozen in liquid nitrogen and stored at −70°C until analyzed. The specimens were processed for histopathological, immunohistochemical examination and western blot analysis. Information included body mass index (BMI), stage and grade. BMI was calculated by dividing the weight in kilograms by the height in meters squared. We defined obesity as a BMI ≥25 (24). Pathological grading was determined using histopathological analysis and the staging process following the FIGO system. None of these patients received preoperative chemotherapy and/or hormonal therapy or pelvic irradiation. The present study was approved by the Research Ethics Board of Qilu Hospital.

All specimens were routinely processed (10% formalin-fixed for 24–48 h), paraffin-embedded and thin-sectioned (4 μm). Antigen retrieval was achieved by heating the slides using a microwave at 95°C for 15 min in citric acid buffer (2 mmol/l citric acid, 9 mmol/l trisodium citrate dehydrate, pH 6.0). The dilutions of antibodies used in the present study were as follows: 1:50 for PPARγ (ab19481), 1:50 for ERα (ab37438) and 1:100 for ERβ (ab3576) (all from Abcam, Cambridge, MA, USA). A Histostain^®-Plus kit (SP-9000; ZSGB-BIO, Beijing, China) was used to detect the immunostaining of PPARγ, ERα and ERβ. 3,3′-Diaminobenzidine tetrahydrochloride (DAB) was used to visualize the reaction, followed by counterstaining with hematoxylin. For each tissue section, at least three fields were photographed under light microscopy. Two hundred cancer cells in each field were selected, and the labeling index (LI) was determined as the percentage of positive cells/200 cells. Cases with an LI >10% were considered positive EC in the present study.

The endometrial carcinoma cell lines ECC-1 and KLE were purchased from the American Type Culture Collection (ATCC; Manassas, VA, USA). ECC-1 cells were cultured in RPMI-1640 medium with 10% fetal bovine serum (FBS) (Gibco, Carlsbad, CA, USA). KLE cells were cultured in a mixture of Dulbecco’s modified Eagle’s medium (DMEM) and Ham’s F-12 1:1 supplemented with 10% FBS in a 5% CO₂ environment at 37°C.

The PPARγ expression vector pGST-PPARγ plasmid was a gift of Dr Bert Vogelstin (Johns Hopkins University, Baltimore, MD, USA). Inhibition of PPARγ function was carried out using small interfering RNA synthesized by GenePharma Company (Shanghai, China). The siRNA sequences were as follows: nonsense negative control (5′-CTG CTG ACT TTA CAG AAG AAA CA-3′) and PPARγ siRNA (5′-AAG CCC ATT GAA GAC ATT CAA GA-3′). The cells were grown to 80% confluency in 6-well plates for plasmid DNA transfection. In addition, 4 μg of plasmid DNA was incubated with 8 μl of Lipofectamine 2000 transfection reagent (Invitrogen, Carlsbad, CA, USA) in 0.5 ml of Opti-MEM™ I reduced serum-free medium (Gibco) for 30 min at room temperature, followed by an additional 1.5 ml of serum-free medium. A total of 2 ml of this liposomal complex was then added to cells. To transfect siRNA, the cells were grown to 40% confluency in 6-well plates. In each transfection reaction, 100 pmol of RNA was incubated with 5 μl of Lipofectamine 2000. Cells were then incubated in a 5% CO₂ environment at 37°C and then switched to 2 ml of complete medium with 10% FBS after 5 h. Twenty-four hours after transfection, the cells were plated for proliferation, migration and invasion assays. The cells were harvested for RNA and protein analyses at 48 h after transfection.

Total RNA was extracted using TRIzol reagent (Invitrogen) according to the manufacturer’s instructions as previously described (25). RNA concentrations were quantified spectrophotometrically (Thermo Fisher Scientific, Waltham, MA, USA). Then, 1 μg of RNA was reverse transcribed in a total volume of 20 μl using a PrimeScript RT reagent kit according to the manufacturer’s protocol. qRT-PCR was performed using SYBR Premix Ex Taq (both from Takara, Tokyo, Japan) according to the manufacturer’s protocol. The LightCycler system (Roche Diagnostics GmbH, Basle, Switzerland) was used to quantify the mRNA expression levels. The expression of each target gene was normalized to the expression of β-actin. Primers were synthesized by Takara and are listed in Table I.

Tissues were pulverized using a mortar and pestle, with lysis buffer (50 mM Tris-HCl, 150 mM NaCl, 1% Triton X-100, 0.1% sodium deoxycholate, 0.1% SDS and protease inhibitor mixture). The cells were washed twice with ice-cold PBS and then lysed using lysis buffer. After incubation on ice for 20 min, the lysates were cleared by centrifugation for 15 min at 12,000 rpm at 4°C. Protein concentrations were quantified using a BCA protein assay kit (Beyotime Biotechnology, Shanghai, China). Western blotting was performed as previously described (26). Briefly, 30 μg of total protein was separated by SDS-PAGE for 2 h at 80 V and then transferred onto PVDF membranes (Millipore, Billerica, MA, USA) for 1.5 h at 200 mA. The membranes were blocked for 2 h at room temperature in 5% non-fat dry milk. The primary antibody was incubated overnight at 4°C, and the secondary antibody was incubated for 1 h at room temperature. Protein expression was detected using ECL (Millipore), and the mean intensity of the bands was quantified using ImageJ (version 1.45). β-actin was also evaluated as an internal control. The dilutions of the primary antibodies used In the present study were as follows: 1:500 for PPARγ (#2453; Cell Signaling Technology, Danvers, MA, USA), 1:100 for ERα (ab37438), 1:1,000 for ERβ (ab3576) (both from Abcam, Cambridge, MA, USA) and 1:3,000 for β-actin (AP0060; Bioworld, Minneapolis, MN, USA). The dilution of the secondary goat anti-rabbit IgG-HRP antibody (BS10350; Bioworld) was 1:20,000.

For the Transwell migration assays, 1×10⁵ cells were plated in the top chamber with a non-coated membrane (24-well insert; 8-μm pore size; Corning Costar, Tewksbury, MA, USA). For the invasion assays, 2×10⁵ cells were plated in the top chamber with a Matrigel-coated membrane (BD Biosciences, San Jose, CA, USA). In both assays, the cells were plated in medium without serum, and medium supplemented with 10% serum was used as a chemoattractant in the lower chamber. Following incubation for 24 h, the cells that did not migrate or invade through the pores were removed by a cotton swab. Cells on the underside of the membrane were fixed with methanol, stained by 0.1% crystal violet, and photographed at ×200 magnification. The cells numbers were counted in five randomly selected fields.

Cell proliferation was determined using the Cell Counting Kit-8 (Beyotime) according to the manufacturer’s instructions. Briefly, 24 h after transfection, the cells were plated for the proliferation assays, with 5×10³ cells/well seeded in a 96-well plate and grown at 37°C for 24 h. After 10 μl of WST-8 dye was added to each well, the cells were incubated at 37°C for 2 h, and the absorbance was finally determined at 450 nm using a microplate reader (Thermo Fisher Scientific).

All results, including transfection, were repeated using independent experiments in triplicate. The χ² test was used to analyze the distribution of cases considered positive for the biological parameters. The correlation between the expressions was analyzed by the Pearson correlation. Statistical analysis between small groups of subjects was performed using the non-parametric Mann-Whiney U test. Statistical significance was assumed at P<0.05. All calculations were performed using SPSS 17.0 software.

PPARγ, ERα and ERβ immunoreactivity were identified in the cell nuclei. PPARγ immunoreactivity was significantly lower in the EC tissues than that in the normal endometrium (P<0.05). The statistical analysis indicated a significant correlation between PPARγ expression and the clinicopathological variables (P<0.05). The expression of ERα was gradually reduced in the moderately and poorly differentiated endometrial carcinoma (P<0.05). The expression of ERβ was only decreased in the poorly differentiated endometrial carcinoma, and no significant associations were detected between ERβ and the clinicopathological variables (Fig. 1, Table II). Furthermore, we found a positive linear variation for PPARγ and ERα immune expression (P<0.05) and no correlations between the expression of ERα and ERβ, or ERβ and PPARγ (Fig. 2). The decreased expression of PPARγ and ERα in the endometrial carcinomas suggest that aberrant PPARγ and ERα expression may be an early molecular event in cancer development.

The PPARγ immunoreactivity results demonstrated a strong association between PPARγ and ERα in EC (P<0.05), suggesting a possible interaction of these two nuclear receptors in human EC cells. Thus, we used two EC cell lines in the following experiments in vitro. The expression of ERs in each cell line was evaluated using western blotting. Only ERβ was expressed in the KLE cells, whereas ERα and ERβ were expressed in the ECC-1 cells, and PPARγ was expressed in both cell lines (Fig. 3A).

When KLE cells were transfected with the expression vector pGST-PPARγ for 48 h, the expression of PPARγ protein increased 49.04%, and the expression of ERβ decreased 28.15% significantly (P<0.01), when compared with the negative control cells (Fig. 3B). In the transfected ECC-1 cells under the same conditions, we found increased expression of PPARγ protein of 36.96% and decreased expression of ERα of 23.41% (P<0.01); however, we did not observe the noticeable depression of ERβ protein in this cell line (Fig. 3C). According to previous studies, PPARγ inhibits ER transcriptional activity through its interaction with ER response elements (ERE) (27,28). To provide further insight into the effects of PPARγ on the activity of ERs, we investigated mRNA levels using qRT-PCR. As expected, the downstream ER activity was confirmed as a decrease in ER gene expression after transfection with the PPARγ expression vector (Fig. 3D). As shown by qRT-PCR and western blot analysis, stimulation of PPARγ did not significantly inhibit transactivation of ERβ in the ECC-1 cells, indicating that in the ECC-1 cell line, the PPARγ effect occurred predominantly through ERα.

To obtain a better understanding of whether suppression of PPARγ expression is associated with ER expression, we transfected PPARγ siRNA and then analyzed both the protein and mRNA levels of ERs. As shown in Fig. 4A and B, after PPARγ siRNA transfection, the PPARγ protein levels in the KLE and ECC-1 cells were decreased by 53.64 and 48.71%, respectively. We also noted increased ERα expression in the ECC-1 cell line, yet not ERβ, in both cell lines. The mRNA levels were confirmed by qRT-PCR analysis (Fig. 4C). These data demonstrate a negative crosstalk between the PPARγ and ER signaling pathways (P<0.05). ECC-1 expressed both the ERα and ERβ receptors, and KLE only expressed the ERβ receptor. Although we did not evaluate whether or not PPARγ regulated ER expression via ERα only, ERα may be more closely related with the mechanism than ERβ.

Having analyzed the interactions of protein and gene expression between PPARγ and ERs, we next evaluated the biological effects of upregulating or downregulating the PPARγ expression in EC cells. In the present study, we investigated cell migration and invasion using Transwell migration and invasion assays. After 24 h of transfection with the PPARγ expression vector, the migratory or invasive cell numbers were significantly decreased compared with the controls in both cell lines (P<0.01) (Fig. 5). However, after 24 h of transfection with PPARγ siRNA in KLE cells, there were no differences in migration or invasive cell numbers compared with the controls (Fig. 5A). On the contrary, downregulation of PPARγ expression enhanced the migratory and invasive abilities in the ECC-1 cells (P<0.01) (Fig 5B).

The stimulation of PPARγ has demonstrated growth inhibitory effects on different tumor cell types, including colon (5), lung (6) and breast cancer (7). Therefore, PPARγ has been considered as a molecular target for cancer chemoprevention (29). Thus, we would expect that PPARγ activation results in decreased cell proliferation in endometrial carcinoma cells. Here, to investigate whether the aberrant expression of PPARγ influences EC cell viability, we performed a CCK-8 assay in each cell line. As expected, the number of KLE and ECC-1 cells was significantly decreased after transfection with the PPARγ expression vector (P<0.05) (Fig. 6). Moreover, our data revealed that the proliferation of the two cell lines was significantly promoted after transfection with PPARγ siRNA (P<0.05) (Fig. 6).