The proliferation, apoptosis, invasion of endothelial-like epithelial ovarian cancer cells induced by hypoxia

Human epithelial ovarian carcinoma cell lines SKOV-3 and ES-2 were purchased from American Type Culture Collection (ATCC, Manassas, VA), and were maintained in McCoy's 5a. Primary human umbilical vein endothelial cells (HUVEC) were isolated from umbilical vein and cultured as described previously [14]

Thirty microliters of Matrigel (B&D, Bedford, MA) were dropped onto each glass coverslip in a 12-well culture plate and polymerized for 1 h at room temperature, followed by 30 min's incubation at 37°C in a humidified 5% CO₂ incubator, as described previously [15]. Tumor cells (1 × 10⁴) were seeded onto the three-dimensional gel. The medium supplied with 15% FBS was changed every 36 h. Hypoxic condition was created by flushing 5% CO₂ and 95% N₂ through a modified chamber (Mitsubishi, Japan), until O₂ concentration was reduced to 1%, measured with a Mini oxygen meter. The culture system was sealed and incubated at 37°C [16]. The cells were treated with 50 nM Sirolimus (Sigma, St. Louis, MO) in DMSO to inhibit the role of HIF-1α under hypoxia when necessary.

For the proliferation assay, 1 × 10⁴ SKOV-3, ES-2 and HUVEC cells, were seeded into a flat bottom 96-well plate and incubated at 37°C for 3 and 7 d under normoxia or hypoxia (1% O₂) respectively, prior to the addition of 20 μL of MTT solution (5 mg/ml in PBS). After incubated for additional 4 h at 37°C, absorbance at 490 nm was measured with a multi-function reader (Tecan GENios, Zurich, Switzerland) to determine cell viability.

Cell cycle and apoptosis assay were performed on cells with or without hypoxia treatment (for 3 or 7 d) to determine whether hypoxia regulates the growth phase and apoptosis of epithelial ovarian cells. Cells were trypsinized and centrifuged at 300 × g (1000 rpm) for 5 min, then resuspended (1 × 10⁶ cells/ml) and fixed with 70% ice-cold ethanol for 30 min, followed by centrifuged, washed and resuspended in 500 μl PBS contained 10 μl of DNase free RNase (final concentration is 1‰). After 30 min incubation, pyridine iodide (PI, 0.05 mg/ml) was added to the solution to incubate for an additional 15 min in the dark and filtered by a nylon mesh to remove cell clusters. The fluorescence of PI was measured using FACS Calibur Flow Cytometer (Becton-Dickinson, San Jose, CA). Cell subpopulations in G0/G1, S and G2/M phases and apoptosis were calculated by gating analysis based on differences in DNA content. At least 20000 cells were analyzed per sample. Cell proliferation characters were indexed by the ratio in S-phase.

Invasion assays were performed in a 24-well transwell chamber (Costar, Bodenheim, Germany) as previously described [17]. Briefly, the 8 μm pore inserts were coated with 15 μg of Matrigel. Cells were seeded to coated filters (5 × 10⁴ cells/filter) in 200 μL of serum-free medium in triplicate. Another 500 μL of serum-free media was added in the lower parts of the chambers. After 7d's incubation under hypoxia, the upper Matrigel coated surface was wiped off using a cotton swab. Cells migrated through the filters were fixed, stained with Giemsa (Sigma, St. Louis, MO), photographed, and counted.

Fifteen microliters of Matrigel were mounted on ethylene vinyl acetate (EVA) membrane (Leica, Wetzlar, Germany) with frame instead of coverslip in 9-cm dishes and treated to establish three-dimensional culture as described above. The density of tumor cells seeded onto gel was adjusted to 1 × 10⁵. After 7 d, samples on EVA membrane were washed with PBS-DEPC and air-dried, channels formed by endothelial-like cells (ELs) were selected by microscopy and microdissected with laser capture microdissection (LCM) system (Leica). About 1,500-2,000 ELs were laser-captured from each EVA membrane. The cells were immersed in digestion buffer for quantitative real-time reverse transcription polymerase chain reaction (RT-PCR) and telomerase activity assay.

Total RNA was extracted from 2 × 10⁴ cells (including HUVEC, SKOV-3, SKOV-3 EL, ES-2, ES-2 EL, or the SKOV-3 or ES-2 cells treated by 50 nM Sirolimus) using TRIzol reagent (Invitrogen, Carlsbad, CA). Aliquots of RNA were reverse transcribed to cDNA using a Superscribe First-Strand synthesis system (Invitrogen). Real-time PCR analysis was performed to quantify mRNA expression of HIF-1α, VEGF, Flk-1, Cyclin D1, p53, and V-src by a Rotor-Gene3000 PCR system (Corbett, Australia) using SYBR-Green PCR Master mix (Qiagen, Hilden, Germany). The PCR reaction consisted of 12.5 μl of SYBR-Green PCR Master mix, 1.0 μl of forward and reverse primers (0.4 μM final concentration), and 2.0 μl of 1:10-diluted template cDNA in a total volume of 25 μl. Amplification was initiated at 50°C for 2 min, 95°C for 70 sec, followed by 40 cycles of 95°C for 20 sec, 58°C for 20 sec, and 72°C for 30 sec. To verify only a single product produced, a dissociation protocol was added after thermocycling. The assay included a no-template control, a standard curve of four serial dilution points (in steps by 10-fold) of a cDNA mixture. All data were controlled by Rotor-Gene software (version 6.0) for quantity of RNA input, an endogenous reference gene (β-actin) was performed as control in the same reverse transcription reaction. Data were presented as the means ± S.E from three separate experiments. The primers used in this experiment were shown in Table 1. All primers were designed using Primer3 web software (Whitehead Institute, Cambridge, MA) and were synthesized by Sangon Biological Engineering Technology and Service Co., Ltd. (Shanghai, P.R. China).

The telomerase activity of all the cells (including HUVEC, SKOV-3, SKOV-3 EL, ES-2, ES-2 EL, or the SKOV-3 or ES-2 cells treated by 50 nM Sirolimus) was tested by telomerase repeat sequence amplification-enzyme linked immunosorbent assay (TRAP-ELISA) using the kit from Huamei Biotechnology Co., Ltd. (Shanghai, China) according to the manufacturer's instruction.

ANOVA analysis or paired-samples t-test were performed to identify differences, using SPSS11.5 statistical software (Lead, US). Statistical significance was assumed at P < 0.05, P-values are presented as two-tailed.

To investigate the morphology of the endothelial-like cells from ovarian cancer induced by hypoxia, the SKOV-3 and ES-2 cells were cultured in the 3-dimensional Matrigel system on EVA membrane under 1% O₂ for 7 d before harvested by LCM. The morphology of the endothelial-like cells induced by hypoxia were pictured by microscope and shown in Figure 1. As it shown, after incubated under hypoxia, the ovarian cancer cells extended and reshaped, developed ELs and connected with each other (A and B), eventually forming network structures and channels (C and D). The original and microdissected by LCM of the single cell were shown in Fig. 1A and 1B, Fig. 1C and 1D indicated the original and microdissected grouped cells.

In order to elucidate the biological behaviors changes in SKOV-3, ES-2 and HUVEC cells by hypoxia, the proliferation, cell cycle, apoptosis and invasion were detected by MTT, FCM and transwell chamber after induced by hypoxia for 3 or 7 d. As shown in Fig. 2A, the proliferation of SKOV-3 was inhibited significantly on 3rd d while there was no difference after 7d's incubation. As for the proliferation of ES-2 cells, there has no significant difference after incubation under hypoxia. The proliferation of HUVEC cells were inhibited by incubation under hypoxia for 3 d and further inhibited after 7 d's incubation.

The percent of cells in S-phase and apoptosis after incubation for 3 or 7 d under hypoxia were shown in Fig. 2B and 2C. As they shown, in the case of SKOV-3 and ES-2 cells, the percent in S-phase were decreased and those of apoptosis were increased after 3 d's incubation, however, there had no difference in S-phase and apoptosis after 7 d's incubation of the two cell lines. On the other hand, the percent of S-phase of HUVEC cells was decreased and that of apoptosis was increased after both 3 and 7 d's incubation.

The numbers of cell migrated through basement membrane of the transwell chamber were shown in Fig. 3D (after 3 d's incubation) and 3E (after 7 d's incubation). Compared to normoxia control, the numbers decreased significantly in SKOV-3 after 3 and 7 d's incubation under hypoxia while it decreased significantly in ES-2 only after 3 d's incubation. The numbers of HUVEC cells were decreased significantly after both 3 and 7 d's incubation.

In order to study the malignant of the ovarian cancer cells, the activities of telomerase of SKOV-3, ES-2 and HUVEC cells incubated under hypoxia, normoxia or hypoxia with Sirolimus were detected by TRAP-ELISA. As shown in Table 2, the activities of telomerase were positive in all the SKOV-3 endothelial-like cells, SKOV-3 under normoxia or with Sirolimus. The activities of telomerase were negative in ES-2 endothelial-like cells and ES-2 with Sirolimus but positive in ES-2 under normoxia. As we expected, the activity of telomerase was negative in HUVEC cells.

In order to elucidate the underlying mechanisms for the biological behaviors changes of the ELs by hypoxia, the mRNA expression of HIF-1α, CyclinD1, VEGF, Flk-1, p53 and V-src in SKOV-3, ES-2 and HUVEC cells incubated under hypoxia, normoxia or hypoxia with Sirolimus were detected by Real-time PCR. The genes expression mentioned above in SKOV-3 and SKOV-3 relative cells were shown in Fig. 3A and Fig. 3B indicated the genes expression in ES-2 and ES-2 relative cells.

As shown in Fig. 3, HIF-1α mRNA expression in both of the two tumors' ELs was significantly higher than that in the cells under normoxia and with Sirolimus, and than that in HUVEC cells.

VEGF mRNA expression in both of the two tumors' ELs was significantly higher than that in the cells under normoxia and with Sirolimus, but was greatly lower than that in HUVEC cells.

Flk-1 mRNA expression in both of the two tumors' ELs was significantly higher than that in the cells under normoxia, but was greatly lower than that in HUVEC cells. On the other hand, Flk-1mRNA expression in ES-2 endothelial-like cells was significantly higher than that in cells treated with Sirolimus, however, there was no difference in Flk-1 mRNA expression between SKOV-3 endothelial-like cells and SKOV-3 cells treated with Sirolimus.

Cyclin D1 mRNA expression in both of the two tumors' ELs was greatly lower than that in the cells under normoxia, while there was no difference in Cyclin D1 mRNA expression in the cells treated with Sirolimus and HUVEC cells.

p53 mRNA expression in both of the two tumors' ELs was significantly higher than that in the cells under normoxia and in HUVEC cells, however, there was no significant changes after treated with Sirolimus.

V-src mRNA didn't express in all kinds of cells under hypoxia or normoxia.