This article is mentioned in:

Introduction

Vascular endothelial growth factor (VEGF) has a fundamental role in the physiological angiogenesis and the vascularization of the follicular luteinizing granulosa layer during corpus luteum (CL) formation (1–4). Inhibition of VEGF in vivo during the luteal phase prevents luteal angiogenesis and subsequent progesterone secretion (5–8), while an excess VEGF generation during the vascularization of multiple follicles is also thought to cause ovarian hyperstimulation syndrome (OHSS) (9–11). Therefore, the molecular regulation of luteal VEGF expression becomes more and more important.

A previous study by our group reported that hypoxia-inducible factor (HIF)-1α contributes to the transcriptional regulation of VEGF in luteal cells (LCs) (12). HIF-1, a helix-loop-helix transcriptional factor, which consists of HIF-1α and HIF-1β, has been cloned and characterized as a transcriptional activator of numerous oxygen-sensitive genes, including erythropoietin (13,14), heme oxygenases (15,16), transferrin (16), and several glycolytic enzymes (13–17). It has been indicated that HIF-1α is inducible by a decrease in tissue or cellular O2 (15,16). HIF-1β is not inducible, but it can bind to HIF-1α to form a dimer to activate the transcription of numerous genes containing cis hypoxia-response element (HRE) in their promoter or enhancer regions (15,16). Previously, a chromatin immunoprecipitation (ChIP) assay indicated that estrogen can simultaneously induce the recruitment of HIF-1α as well as -β to the upstream HRE and ERα to the proximal GC-rich region of the VEGF promoter, which mediates transcriptional activation of the murine VEGF gene (18–20).

It has been demonstrated that PHDs are the major enzymes to promote the degradation of HIF-1α (21–23). PHD catalyzes site-specific proline hydroxylation of HIF-1α, and PHD is recognized and targeted for degradation by the ubiquitin-proteasome pathway. Three isoforms of PHD, including PHD1, -2, and -3, have been identified (21–24). Previous studies by our group have indicated that PHD2 participates in functional regulation, including sensing high-salt intake and maintaining normal blood pressure, through regulating HIF-α levels (25) and also has a role in angiogenesis (26).

Given the role of PHD2 in the regulation of HIF-1α levels, it was hypothesized in the present study that the PHD2 signaling pathway has a role in ovarian angiogenesis and that overexpression of PHD2 attenuates the expression of VEGF induced by hypoxia in LCs. The present study examined the effect of hypoxia on the expression of HIF-1α and determined the role of PHD2, the primary isoform of PHDs, in hypoxia-induced activation of HIF-1α by transfection of PHD2-expressing vectors into LCs. The effect on VEGF mRNA expression was also examined. The results indicated that PHD2 is a mediator of cellular HIF-1α and its target gene VEGF in LCs, which may be an important regulatory mechanism of VEGF-dependent angiogenesis during mammalian CL development.

Materials and methods

Animals

25-day-old Female Sprague-Dawley (SD) rats (Fuzhou Animal Center, Fuzhou, China) were used in the present study. All animals were maintained under a 12-h light/dark cycle with food and water available ad libitum. The study was conducted in accordance with guidelines of the Institutional Animal Care and Use Committee and was approved by the Ethics Committee on Animal Experimentation of Fujian Normal University. All efforts were made to minimize animal discomfort and to reduce the number of animals used.

Isolation and culture of rat LCs

Rat LCs were isolated and cultured according to previously described procedures (27–29). Briefly, the rats received a subcutaneous injection of equine chorionic gonadotropin [eCG; Sigma-Aldrich, St. Louis, MO, USA; 50 international units (IU)] and an ovulatory dose (25 IU) of human chorionic gonadotropin (hCG; Sigma-Aldrich) 64 h later. Ovaries was obtained at day 5 after hCG injection and minced with a razor blade. Tissue was digested in medium 199 (Gibco-BRL, Invitrogen Life Technologies, Carlsbad, CA, USA) containing 1% fetal calf serum (FCS; Gibco-BRL) and 2,000 IU collagenase (Gibco-BRL) plus 3,000 IU of DNase (Sigma-Aldrich) per gram of tissue for 1 h at 37°C under 95% air with 5% CO2. The contents of the flask were filtered through nylon mesh (BD Biosciences, Franklin Lakes, NJ, USA) and centrifuged (100 xg, 5 min); the supernatant fraction was discarded and the pellet was washed three times with fresh medium. The final cell concentration was 106 cells per ml and cells were incubated in six-well culture plates. Cell numbers were determined with a hemocytometer and cell viability was >90% as assessed by Trypan blue (Sigma-Aldrich) staining exclusion.

Transfection of plasmids expressing rat PHD2 into the cells

Plasmids encoding rat full-length PHD2 cDNA were a generous gift from Dr Ningjun Li (Department of Pharmacology & Toxicology, School of Medicine, Virginia Commonwealth University, Richmond, VA, USA). The expression and function of rat PHD2 protein by the plasmids has been validated by previous studies (25,30–32). Plasmid transfections were performed using lipids (DOTAP/DOPE; Avanti Polar Lipids, Inc., Alabaster, AL, USA) according to the manufacturer's instructions. In brief, 5 µg DNA was mixed with a lipid solution at a DNA/lipid ratio of 1:10 (w/w) in serum-free culture medium (5 ml for a 10-cm dish). Cells were incubated with this transfection medium for 5 h and switched to normal medium for another 16 h. The cells were then ready for the subsequent experiments.

Cell hypoxia

Cell culture under hypoxia with 3% O2 was performed as described previously (33,34). Briefly, LCs were plated in six-well plates 24 h prior to the experiments to reach sub-confluence. To decrease O2 pressure in the culture medium, the plates and dishes were transferred to a sealed, humidified modular chamber, the atmospheric air was evacuated by a vacuum pump, and the chamber was re-filled with a non-standard gas mixture containing 3% O2 and 5% CO2 in an N2 base. After repeating this evacuation and inflow five times, the cells were cultured for 6 h. After hypoxia, the culture medium was rapidly replaced by TRIzol solution (Invitrogen Life Technologies), and total RNA was extracted for determination of HIF-1α and VEGF mRNA. For HIF-1α protein determination, the cultured cells were scraped immediately and placed in ice-cold homog-enization buffer [25 mM Tris-HCl, 300 mM sucrose, 2 mM EDTA, protease inhibitor cocktail (Roche Diagnostics, Basel, Switzerland), pH 7.4], frozen in liquid nitrogen and then stored at −80°C for western blot analysis.

RNA extraction and quantitative reverse transcription polymerase chain reaction (RT-PCR) analysis of PHD1, 2 and 3, HIF-1α and VEGF

Total RNA was extracted using TRIzol solution and then reverse-transcribed uding a cDNA Synthesis kit (Bio-Rad Laboratories, Inc., Hercules, CA, USA). The RT products were amplified using SYBR green (Bio-Rad Laboratories, Inc.) and the following primers: PHD1 forward, 5′-GCT GCT GCG TTG GTTAC-3′ and reverse; 5′-GCC TCC TGG TTC TCTTG-3′ (GenBank accession no.: NM001004083); PHD2 forward, 5′-CTG GGA CGC CAA GGTGA-3′ and reverse, 5′-CAA TGT CAG CAA ACTGG-3′ (GenBank accession no.: NM178334); PHD3 forward, 5′-GTT CAG CCC TCC TATGC-3′ and reverse, 5′-ACC ACC GTC AGT CTTTA-3′ (GenBank accession no.: NM019371); HIF-1α forward, 5′-CTG GCA CGG GGA TGA TAC AGC-3′ and reverse, 5′-TCT CAT CCA TTG ACT GCC CCAG-3′ (GenBank accession no.: AF057308) and VEGF forward, 5′-ACG AAG CGC AAG AAA TCCC-3′ and reverse, 5′-TTA ACT CAA GCT GCC TCGCC-3′ (GenBank accession no.: M32167) with an iCycler iQ Real Time PCR Detection System (Bio-Rad Laboratories, Inc.). The levels of 18 rRNA [forward, 5′-CGC CGC TAG AGG TGA AATTC-3′ and reverse, 5′-TCT TGG CAA ATG CTT TCGC-3′ (GenBank accession no.: M11188)] were used as an endogenous control. The 25 µl PCR reaction mix included 12.5 µl SYBR Green PCR Master Mix, 2.5 µl 10X primers, 1 µl cDNA template and 9 µl RNase-free water. The PCR conditions were as follows: 50°C for 2 min, 95°C for 15 min, followed by 40 cycles at 94°C for 15 sec, 60°C for 30 sec and 72°C for 30 sec. Relative gene expression was calculated in accordance with the ΔΔCt method, with relative mRNA levels calculated as 2−ΔΔCt.

Preparation of nuclear extracts and cytosolic protein, and western blot analysis of protein levels of HIF-1α and PHD

Nuclear protein was prepared as previously described (12,25,32). Cytosolic protein and nuclear protein were collected separately (Beyotime Institute of Biotechnology, Haimen, China). The cytosolic protein was used for western blot analysis of PHD2, while the nuclear fraction was used for western blot analyses of HIF-1α. Briefly, protein concentration was determined using a Bio-Rad assay (Bio-Rad Laboratories, Inc.) with bovine serum albumin standards. The protein samples (30 µg) were separated by 8% SDS-PAGE, and were electrophoretically transferred to a polyvinylidene fluoride membrane (Pall Corporation, Pensacola, FL, USA). The membrane was washed with phosphate-buffered saline with 0.2% Tween-20 (PBST; Sigma-Aldrich), blocked with 5% nonfat dried milk in PBST and probed with the following primary antibodies (Novus Biologicals, Littleton, CO, USA): Mouse monoclonal anti-HIF-1α (1:300; cat. no. NB100-105), rabbit polyclonal anti-PHD2 (1:300; cat. no. NB100-137), and rabbit polyclonal anti-β-actin (cat. no. NB600-503H) overnight at 4°C. After washing, the membranes were then incubated with horseradish peroxidase-conjugated goat anti-mouse (1:5,000; cat. no. NB7570) or goat anti-rabbit (1:5,000, cat. no. NBP2-30348H) immunoglobulin G secondary antibodies (Novus Biologicals) for 1 h at room temperature. β-actin was used as a loading control (1:5,000, cat. no. NB600-501H). To detect the immunoblotting signal 2 ml enhanced chemiluminescence detection solution (Thermo Scientific, Rockford, IL, USA) was used, and the membrane was exposed to Kodak OMAT film (Eastman Kodak, Rochester, NY, USA). The blots were quantified using ImageJ software (National Institutes of Health, Bethesda, MD, USA)

Determination of HIF prolyl hydroxylase activity

HIF-1α peptide-specific conversion of 2-oxoglutarate into succinate provides a hydroxyl group for HIF-1α to be prolyl hydroxylated. This reaction has been widely used for the determination of PHD activity by measuring HIF-1α-dependent conversion rate of 2-oxoglutarate into succinate (25,35,36).

Statistics

Values are expressed as the mean ± standard error. The significance of differences in mean values within and between multiple groups was evaluated using analysis of variance followed by a Duncan's multiple range test. Statistical analysis was conducted using Sigmastat 3.02 (Sigma-Aldrich). P<0.05 was considered to indicate a statistically significant difference.

Results

PHD2 is the most abundantly expressed PHD in LCs, and is not affected by hypoxia

RT-qPCR indicated that all of three PHDs were present, among which PHD2 was most abundantly expressed in LCs (Fig. 1), while no significant changes of these mRNA levels were found after hypoxia treatment (Fig. 1).

Figure 1 Effects of hypoxia on the mRNA expression of PHD2 in luteal cells. Relative mRNA levels of PHDs were assessed by reverse transcription polymerase chain reaction analysis. Values are expressed as the mean ± standard error (n=6). Different letters denote statistically significant differences between values (P<0.05). PHD, hypoxia-inducible factor prolyl-hydroxylase.

Efficient PHD2 transfection in LCs is not affected by hypoxia

PHD2 mRNA levels increased significantly after PHD2 transfection, even following hypoxia treatment (Fig. 2). However hypoxia was found to have no marked effect on PHD2 mRNA expression (Fig. 2), indicating the high transfection efficiency and the high expression levels of PHD2 plasmid.

Figure 2 Effects of hypoxia and PHD2 plasmid on PHD2 mRNA levels in luteal cells. Relative mRNA levels of PHD2 were assessed by reverse transcription polymerase chain reaction analysis. Values are expressed as the mean ± standard error (n=6). Different letters denote statistically significant differences between values (P<0.05). PHD, hypoxia-inducible factor prolyl-hydroxylase.

Furthermore, PHD2 protein levels were detected by western blot analysis (Fig. 3). The results indicated that, similar to mRNA levels, PHD2 protein levels were enhanced following transfection, but that hypoxia had no effects on PHD2 protein levels (Fig. 3).

Figure 3 Effects of hypoxia and PHD2 plasmid on PHD2 protein levels in luteal cells. (A) Representative western blot depicting the protein levels of PHD2. (B) Quantified intensities of PHD2 blots normalized to ACTB. Values are expressed as the mean ± standard error (n=6). Different letters denote statistically significant differences between values (P<0.05). PHD, hypoxia-inducible factor prolyl-hydroxylase; ACTB, β-actin.

Effects of hypoxia and PHD2 transfection on PHD2 protein levels and PHD2 activity in LCs

To further confirm stable transfection of PHD2 into LCs, PHD2 biological activity was tested using the 2-oxoglutarate conversion assay (Fig. 4). In the PHD2-transfected groups, 2-oxoglutarate conversion was enhanced, however, not to a statistically significant extent. Of note, a significant decrease of PHD activity was found in LCs under hypoxic conditions, even following transfection with PHD2 plasmids (Fig. 4).

Figure 4 Effects of hypoxia and PHD2 plasmid on PHD activity in luteal cells. PHD activity in each group by [14C]-2-oxoglutarate conversion rate. Values are expressed as the mean ± standard error (n=6). Different letters denote statistically significant differences between values (P<0.05). PHD, hypoxia-inducible factor prolyl-hydroxylase.

Hypoxia increases and PHD2 transfection decreases VEGF mRNA levels in LCs, while HIF-1α mRNA is unaffected

In the present study, hypoxia significantly increased VEGF mRNA expression in LCs (Fig. 5). To better understand the role of PHD2 in LCs, the mRNA levels of VEGF and HIF-1α in LCs transfected with PHD2 plasmid were also examined (Figs. 5 and 6). Of note, a marked decrease in VEGF levels was observed in PHD2-transfected LCs (Fig. 5), while HIF-1α mRNA levels were not significantly altered in LCs following hypoxia treatment (Fig. 6).

Figure 5 Effects of hypoxia and PHD2 plasmid on VEGF mRNA levels in luteal cells. Relative mRNA levels of VEGF were assessed by reverse transcription polymerase chain reaction analysis. Values are expressed as the mean ± standard error (n=6). Different letters denote statistically significant differences between values (P<0.05). PHD, hypoxia-inducible factor prolyl-hydroxylase; VEGF, vascular endothelial growth factor.

Figure 6 Effects of hypoxia and PHD2 plasmid on HIF-1alpha mRNA levels in luteal cells. The relative mRNA levels of HIF-1alpha were assessed by reverse transcription polymerase chain reaction analysis. Values are expressed as the mean ± standard error (n=6). No statistically significant differences between values were present. HIF, hypoxia-inducible factor; PHD, HIF prolyl-hydroxylase.

Effects of hypoxia and PHD2 transfection on HIF-1α protein levels in LCs

It has been demonstrated that PHDs are major enzymes to promote the degradation of HIF-1α (21–24); therefore, the present study examined the HIF-1α protein levels in each group (Fig. 7). Hypoxia inhibited HIF-1α protein expression under normoxic and hypoxic conditions (Fig. 7), and a significant decrease of HIF-1α protein was found in PHD2-transfected LCs, which was consistent with the results of previous studies by our group (25,32), indicating that PHD2 may regulate VEGF expression via the HIF-1α pathway in LCs during CL development.

Figure 7 Effects of human chorionic gonadotropin and PHD2 plasmid on HIF-1alpha protein levels in luteal cells. (A) Representative western blot depicting the protein levels of HIF-1alpha. (B) Quantified intensities of HIF-1alpha normalized to ACTB. Values are expressed as the mean ± standard error (n=6). Different letters denote statistically significant differences between values (P<0.05). HIF, hypoxia-inducible factor; PHD, HIF prolyl-hydroxylase; ACTB, β-actin.

Discussion

The results of the present study clearly demonstrated that hypoxia induced VEGF through inhibiting HIF-1α signaling in LCs, which was attenuated by overexpression of PHD2, suggesting that PHD2-mediated VEGF expression via the HIF-1α pathway may be an important mechanism of VEGF-dependent angiogenesis during mammalian CL development.

The CL is a temporary endocrine structure in mammals, which has an important role in the female reproductive cycle and is formed temporarily from a ruptured and ovulated follicle with rapid angiogenesis (5,6,11,37,38). VEGF is thought to have a paramount role in the regulation of normal and abnormal angiogenesis in the ovary (2–4,6,9,11,39,40), particularly in the newly formed CL. Numerous studies have shown that reproductive hormones such as hCG also take part in the primary regulation of VEGF expression in the ovary. For example, VEGF mRNA expression in human luteinized granulosa cells has been shown to be dose- and time-dependently enhanced by hCG in vitro (9,10). Chronic or acute exposure to hCG directly stimulates VEGF production and secretion by monkey (1) and human luteinized granulosa cells (5,9,10,40). Furthermore, the administration of a gonadotropin-releasing hormone antagonist decreased VEGF mRNA expression in the CL of monkeys (41). In addition, luteal vascularization and the development of ovarian hyperstimulation syndrome (OHSS) are absolutely dependent on LH/hCG stimulation (9,10). Furthermore, in a fully formed, highly vascular CL hCG also up-regulates VEGF expression (5). However, a previous study by our group has already provided direct evidence that VEGF is transcriptionally activated by a HIF-1-mediated mechanism in LCs under hypoxia (12), which is caused by ovulation of the ruptured follicle with bleeding and an immature vascula-ture (2,38). Therefore, the present study examined the induced effect of hypoxia and PHD2 overexpression on VEGF mRNA expression in LCs. Of note, VEGF expression was induced by hypoxia in LCs and hypoxia-stimulated HIF-1α protein expression was highly correlated with VEGF expression, indicating hypoxia stimulated VEGF expression via the HIF-1α signaling pathway.

Numerous studies have indicated that HIF-1α regulates the expression of numerous genes whose protein products have critical roles in developmental and physiological processes, including angiogenesis, erythropoiesis, glycolysis, iron transport and cell proliferation/survival (5,38,42–45). Because HIF-1α activates the transcription of VEGF, which is required for angiogenesis, it is possible that hypoxia may mediate angiogenesis via the HIF-1α/VEGF pathway. In addition to a detailed exploration of the downstream mechanism of HIF-1α (12,46–48), recent studies have clarified the upstream process modulated by PHDs (49), which regulates HIF-1α degradation by the ubiquitin-proteasome pathway. In particular, PHD2 has been drawing considerable attention because PHD2 is considered to be the key oxygen sensor of all identified PHD enzymes (25,26), as knockdown of PHD2 resulted in elevated HIF protein levels (49), and several recent studies have also highlighted the importance of PHD2 in tumourigenesis (26). The present study reported that hypoxia induced the activation of HIF-1α and overexpression of PHD2 blocked this activation of HIF-1α and its target gene VEGF following hypoxia treatment. These results indicated that PHD2 is involved in HIF-1α-mediated gene activation in LCs under hypoxia, which may present a novel mechanism of VEGF-dependent angiogenesis during mammalian CL development.

In conclusion, the present study clearly demonstrated that hypoxia induces HIF-1α and VEGF expression in LCs. To the best of our knowledge, the present study was also the first to provide direct evidence indicating that PHDs are expressed in rat LCs and that hypoxia-induced VEGF expression can be blocked by overexpression of PHD2 in LCs. This PHD2-mediated VEGF expression may be one of the important mechanisms of VEGF-dependent angiogenesis during CL formation in the mammalian ovary. Furthermore, this PHD2 antagonism presents an opportunity for the development of novel treatments for fertility control and for certain types of ovarian dysfunction (26,44,45), particularly conditions characterized by pathological angiogenesis and excessive vascular permeability, including polycystic ovarian syndrome (PCOS), OHSS and ovarian neoplasia.

Acknowledgments

This study was supported by the National Natural Science Foundation of China (nos. 31101032 and 31271255), the Program for New Century Excellent Talents in University of the Ministry of Education of China (no. NCET-120614), the Doctoral Foundation of the Ministry of Education of China (no. 20113503120002) and Fujian Provincial Science and Technology Projects of the Department of Education (no. JB14041). The authors particularly thank Dr Ningjun Li from Department of Pharmacology & Toxicology, School of Medicine, Virginia Commonwealth University in the USA for kindly providing the PHD and control plasmids.

References