Primary epithelial cells from the endometrium, well-differentiated endometrial carcinoma RL952 and immortalized malignant endometrioid ovarian cancer cell TOV-112D (EOCC), clear cell ovarian cancer cell TOV-21G, serous ovarian cancer cell OV-90 cell lines (ATCC, Rockville, MD, USA) were used. Ovarian cancer and primary endometrial cells were cultured in medium 199 and medium 105 mixtures (Invitrogen Life Technologies Inc., New York, NY). RL-952 was maintained in Dulbecco's modified Eagle's medium F-12 (GIBCO: Invitrogen, NY, USA). These media were supplemented with 10% FBS. Hanks Balanced Salt Solution containing trypsin 0.25 mM EDTA was obtained from Sigma (St. Louis, MO, USA). The concentrations of human IL-1α, IL-1β and IL-1RA in cell culture supernatants were measured by using ELISA kit (R&D Systems Inc., Minneapolis, MN). Monoclonal mouse anti-human IL-1α and IL-1RA and antibody Alexa Fluor 594-labelled goat anti-mouse were respectively purchased from R&D Systems Inc. (Minneapolis, MN, USA) and Molecular Probes (Invitrogen, Carlsbad, CA, USA). 4, 6-diaminido-2-phenyl-indole (DAPI) was obtained from Sigma Aldrich (St Louis, MO, USA). Reverse Transcriptase Supercript II and SYBR Green Master Mix were purchased respectively from Invitrogen (Carlsbad, CA, USA) and Applied Biosystems (Foster City, CA, USA).

Endometrial biopsies were obtained from 5 healthy fertile patients undergoing gynecological surgery for tubal ligation. The study was approved by the CHUS Ethics Human Research Committee on Clinical Research. All participants gave written consent. Tissues were washed in HBSS minced into small pieces and dissociated with collagenase as previously described 41 . Endometrium was finely minced and incubated in sterile Hank's balanced salt solution (HBSS) (GIBCO Invitrogen Corp., Burlington, ON, Canada) containing 20 mM Hepes, 100 U/ml penicillin, 100 μg/ml streptomycin and 1 mg/ml collagenase at 37°C in a shaking water bath during 60 minutes. Fragments of epithelial glands from collegenase digestion were isolated by filtration through a 45-μm nylon mesh.

Endometrial and ovarian cancer cells were seeded at a density of 2 × 10⁶ cells per 1 ml in 12-well plates containing medium with 10% FBS and cultured overnight. Medium was exchanged and cells were cultured for a further 48 hr. The culture supernatants were collected and microfuged at 1,500 rpm for five min to remove particles and the supernatants frozen at -20°C until use in ELISA. The concentration of IL-1α, IL-1β and IL-1RA in the supernatants per 2 × 10⁶ cells was measured using an ELISA kit (R&D Systems, Minneapolis, MN) according to the manufacturer's instructions.

To evaluate intracellular, membrane-bound IL-1α and intracellular IL-1RA, immunostaining was performed according to Akoum et al. 42 . Briefly, cell lines were grown on glass coverslips overnight and fixed with formaldehyde in PBS. The cells were permeabilized by treatment with 0.1% Triton X-100 (PBS/TX) in PBS for 15 min at room temperature and incubated with a monoclonal mouse anti-human IL-1α or IL-1RA antibody in 1% BSA/PBS for 2 hours. After washing with PBS, the cells were incubated with secondary antibody goat Anti-Mouse Alexa Fluor 594 for 1 hour. Nuclei were identified by 4', 6'-diamidino-2-phenylindole staining for 15 min at room temperature. Following mounting, cells were observed under the Leica microscope. Experiments have been done five times. Immunostained cells were scanned with iCys imaging cytometer (Compucyte, Cambridge, MA). Immuno-staining was detected using Argon ion (488 nm) excitation laser with green (530 nm/30 nm) detection PMT. DNA staining was detected using violet diode (405 nm) excitation laser with bleu (463 nm/39 nm) detection PMT. Image for cellular morphology was acquired using scattering of the Argon ion laser. Scanning was performed at 0,5 μm × 0,25 μm pixel size resolution. Cellular event selection was performed using a virtual channel obtained by adding green and blue fluorescence signals to insure detection and quantification of cytoplasmic signal. Immuno-staining intensity and cellular area were measured and used to compare IL-1α and IL-1RA proteins expression between EOCC, EC and the others subtypes of ovarian cancers. An experimented scorer selected the scoring thresholds for immuno-staining intensity. All cell selections were confirmed by visualizing a gallery of at least 250 representative cells.

IL-1α, IL-1β and IL-1RA mRNA extraction was achieved using trizol. To evaluate the level of gene expression, real-time PCR with SYBR Green dye was applied. Experiments have been done five times. The Rotor-Gene (Corbett Research, Sydney, Australia) equipment for reaction monitoring was used. β actine gene was used as internal control. The forward sequence GAATGACgCCCTCAATCAAAGT and reverse sequence TCATCTTGGGCAGTCACATACA were used for human IL-Rα. For human IL-1RA, the forward and reverse sequences were AATCCAGCAAGATGCAAGCC and ACGCCTTCGTCAGGCATATT, respectively. Forward and reverse sequences for human IL-1β were AAACAGATGAAGTGCTCCTTCCAGG and TGGAGAACACCACTTGTTGCTCCA respectively. For β actine, the forward and reverse sequences were CATGTACGTTGCTATCCAGGC and CTCCTTAATGTCACGCACGAT, respectively. The PCR reaction was performed in 20 μl final volume using 36-well plates. The reaction mixture contained 10 μl SyberGreenSuperMix, 100 nM of each primer (forward and reverse) and 1 μl cDNA. All samples were run in duplicate. The thermal protocol was as follows: 1 min 90°C, followed by 60 cycles (20 s at 95°C - denaturation, 20 s at 60°C - annealing and 20 s at 72°C - elongation - when the signal was acquired). Each sample was normalized on the basis of its GAPDH content according to the formula , EC representing endometrial cells; OCC, ovarian cancer cells and C_T the threshold cycle.

IL-1α and IL-1RA staining scores follow an ordinal scale. Data followed a parametric distribution and were shown as means ± SD. We used one-way analysis of variance (ANOVA) and the Bonferroni's test post hoc for multiple comparisons or the unpaired t-test for comparison of two groups. Statistical analyses were performed using excel and GraphPad Software, Prism 4.0 (GraphPad Software, San Diego, CA, USA). Differences were considered as statistically significant whenever a P value < 0.05 occurred.

The intensity of IL-1 ligands system proteins staining was scored using quantitative imaging cytometry. Immunofluorescence analysis clearly showed that IL-1α protein (Figure 1 and Figure 2A) and IL-1RA protein (Figure 3 and Figure 4A) were expressed in all types of studied cells. Whereas incubation of cells without primary antibodies (negative controls), did not result in any noticeable staining. As shown in Table 1, statistical analysis comparing endometrial cells and EAOC subtypes showed that IL-1α staining was more intense in clear cell line (TOV-21G) (Figure 1 and Figure 2A; P < 0.05), whereas IL-1RA staining was higher in serous cell line (OV-90) and very low in endometrioid ovarian cell line (TOV-112D) (Figure 3 and Figure 4A; P < 0.05).

To further analyze IL-1α and IL-1RA at level of transcription, gene expression was achieved by real-time quantitative PCR kinetics using SybrGreen I chemistry. The baseline adjustment method of the Rotor Gene software was used to determine the threshold cycle in each reaction. A melting curve was constructed for each primer pair to verify the presence of one gene-specific peak and the absence of primer dimmer. A representative Real-Time-PCR of IL-1α and IL-1RA mRNA in EAOC subtypes compared to endometrial cell line RL-952 and primary endometrial cells are shown in Table 1. IL-1α mRNA expression was higher in TOV-21G cells (Figure 2B; P < 0.05), whereas no statically changes of IL-1α mRNA expression was observed between endometrial cells and the other epithelial ovarian cancer cell lines (Figure 2B). Analysis of mRNA levels showed a marked decrease in the expression of IL-1RA in EOCC (Figure 4B; P < 0.001); and an increase in the expression of IL-1β in TOV-112D and OV-90 cells (Table 1, Figure 5).

Concentrations of the cytokines released by endometrial cells and ovarian cancer cells are shown in Table 2.

The results of this study demonstrated the presence of IL-1α, IL-1β and IL-1RA in all cell lines. The levels of IL-1α secretion were higher in endometrial cells than ovarian cancer cells. The levels of IL-1β were significantly higher in the supernatant of EOCC than both of endometrial cells (P < 0.05). The levels of IL-1β in the supernatant of all ovarian cancer cell lines studied were significantly higher than endometrial cells (P < 0.05). Moreover, we found a high concentration of IL-1β in OV-90 cell line (P < 0.01). The levels of IL-1RA are significantly lower in EOCC compared to both endometrial cells (P < 0.05). However, IL-1RA concentrations were also lowers in EOCC compared to other ovarian cancer cell lines, with high expression in OV-90 cell line (P < 0.01).