Progestogens stimulate prostacyclin production by human endothelial cells

C. Hermenegildo1,2,5, P.J. Oviedo3, M.C. García-Martínez3, M.A. García-Pérez1, J.J. Tarín4 and A. Cano3

1Research Unit, Hospital Clínico Universitario of Valencia and Departments of 2 Physiology, 3 Pediatrics, Obstetrics and Gynecology and 4 Functional Biology and Physical Anthropology, University of Valencia, 46010 Spain

5 To whom correspondence should be addressed at: Department of Physiology, Faculty of Medicine and Dentistry, Av. Blasco Ibañez, 17, E 46010 Valencia, Spain. Email: carlos.hermenegildo{at}uv.es


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
BACKGROUND: The effects of progestogens on endothelial physiology are poorly studied. Prostacyclin is a potent vasodilator synthesized by two isoforms of cyclooxygenase (COX) in endothelium. We examined the effects of two clinically used progestogens, progesterone and medroxyprogesterone acetate (MPA), on prostacyclin production by cultured human umbilical vein endothelial cells (HUVEC) and the possible role of progesterone receptors and both COX enzymes. METHODS: Cells were exposed to 1–100 nmol/l of either progesterone or MPA and prostacyclin production was measured in culture medium. RESULTS: Both progestogens significantly increased prostacyclin release in a time- and dose-dependent manner, being higher than control after 24 h. Progesterone and MPA, both at 10 nmol/l, increased mRNA expression and protein content of both COX. All these effects were mediated through progesterone receptor activation, since they were abolished by treatment of cells with the progesterone receptor antagonist RU-486. Selective inhibitors of COX-1 and -2 (SC-560 and NS-398 respectively) reduced basal prostacyclin release, and eliminated increased production in response to progestogens. In combination with estradiol, progestogens had an additive effect without eliminating estradiol-induced prostacyclin production. CONCLUSIONS: Our results support the hypothesis that progesterone and MPA increased HUVEC prostacyclin production in a progesterone receptor-dependent manner, by enhancing COX-1 and COX-2 expression and activities.

Key words: cyclooxygenase/endothelial function/hormones/prostaglandins/vasoactive agents


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Vascular endothelium plays a leading role in vascular physiology. The endothelium is crucial to the modulation of vessel tone and to the control of platelet adhesion and aggregation, two key factors in the initiation and development of atherosclerosis (Ross, 1999Go). These actions are mainly mediated through the release of such vasorelaxing factors as prostacyclin and nitric oxide, or of vasoconstrictors, such as endothelin-1, angiotensin II or thromboxane A2.

Prostacyclin is a potent endogenous anticoagulant for platelets and a strong vasodilator. Prostacyclin is a prostaglandin produced from free arachidonic acid through the catalytic activity of two different cyclooxygenases (COX), termed COX-1 and COX-2. COX represent the main control mechanism for prostacyclin production. COX-1 is considered to be expressed in a constitutive manner, whereas COX-2 is inducible by mitogens, cytokine growth factors and endotoxins and is over-expressed in inflammatory processes (Smith et al., 2000Go; Parente and Perretti, 2003Go).

Clinical and experimental data support the consideration of endothelium as a target for sexual hormones (Mendelsohn and Karas, 1999Go). Estrogen receptors have been found in endothelium and estradiol actions have been exhaustively studied (Mendelsohn and Karas, 1999Go; White, 2002Go). For instance, estradiol enhances endothelial prostacyclin production through both COX-1 (Jun et al., 1998Go) and COX-2 (Akarasereenont et al., 2000Go) and reduces production of vasoconstrictors, reducing blood pressure (Dubey et al., 2002Go).

Much less is known, however, regarding progestogen actions on the vascular wall. Several types of endothelial cells express progesterone receptors (PR) (Vazquez et al., 1999Go). Progestogens have been shown to regulate many physiological processes that impact on the atherosclerotic progression. For instance, progestogens can inhibit vasorelaxation (White et al., 1995Go) and decrease endothelial cell proliferation (Vazquez et al., 1999Go). Also, PR can regulate the vascular injury response (Karas et al., 2001Go).

Progestogens are a second component of hormone substitution in post-menopausal women, and are recommended in women with uterus to reduce the risk of endometrial cancer. The effects of exogenous administration of progestogens in combination with estradiol on endothelial function are also unclear, given that in some studies, progestogens counteract estradiol's beneficial vascular effects, while in other trials progestogens do not (see reviews Cano, 1999Go; Dubey et al., 2002Go; Ganz, 2002Go).

Cardiovascular effects of hormone therapy are presently under discussion (Herrington and Klein, 2003Go; Kuller, 2003Go). A better knowledge of the endothelial effects of two clinically used progestogens, progesterone (the natural progestogen) and medroxyprogesterone acetate (MPA), could contribute to clarify the physiological and clinical effects. Moreover, all progestogens do not have similar effects on vascular tone (Sarrel, 1999Go).

The objectives of the work were the following: (i) to study the effects of two progestogens, progesterone and MPA, on prostacyclin production by endothelial cells; (ii) to uncover whether these actions are mediated by PR; (iii) to illuminate whether the observed effects are due to regulation of one or both COX isozymes; and (iv) to test whether progestogens interact with estradiol effects.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Cell culture and experimental design
Primary HUVEC were isolated by collagenase treatment of human umbilical veins as described in Jaffe et al. (1973)Go. Briefly, HUVEC were grown in 25 cm2 flasks (Orange Scientific, Belgium) in human endothelial cell-specific medium EBM-2 (Clonetics, BioWhittaker, USA), supplemented with EGM-2 (Clonetics), in an incubator at 37 °C with 5% CO2.

Cells were identified as endothelial by their characteristic cobblestone morphology and the presence of von Willebrand factor by immunocytochemistry using a specific antibody (F-3520; Sigma, Spain).

Cells from passages 4–6 were seeded onto 6-well plates with fibronectin-treated coverslips for immunocytochemistry, onto 24-well plates for prostacyclin measurements, onto 96-well plates for measurement of cell viability, and onto 25 cm2 flasks for western blot and mRNA isolation. When cells were at 75% of confluence, culture medium was exchanged for a phenol red-free medium 199 (Gibco-BRL, Life Technologies, UK) supplemented with 20% charcoal/dextran-treated fetal bovine serum (Gibco-BRL) and maintained for 24 h. The culture medium was then eliminated and immediately replaced with phenol red-free medium 199. The desired concentrations of progesterone, MPA and estradiol were obtained by successive dilutions of a stock solution (1 mmol/l in ethanol) with phenol red-free medium. The desired concentrations of RU-486, SC-560 and NS-398 were obtained by successive dilutions of a stock solution [10 mmol/l in dimethylsulphoxide (DMSO)] with phenol red-free medium. Therefore, cells were exposed to <0.1% ethanol and/or <0.1% DMSO. Control cells were exposed to the same concentrations of ethanol and/or DMSO as treated cells.

Immunoblotting
HUVEC were treated in 25 cm2 flasks for 24 h with the desired products. Flasks were then washed twice with pre-warmed medium 199. A volume of 150 µl of lysis buffer (0.1% Triton X-100, 0.5% sodium deoxicholate acid, 0.1% sodium dodecyl sulphate (SDS), in 100 ml of phosphate saline buffer containing protease inhibitors: 1 µg/ml leupeptin, 0.5 µg/ml pepstatin and 1 µg/ml bestatin) was added and incubation was maintained at 4 °C for 30 min. Cells were then collected using a cell scraper, boiled for 5 min and sonicated for 10 s. Protein content was measured (Lowry et al., 1951Go) and samples were frozen at –20 °C until assay.

Equal amounts of protein (ranging 40–125 µg) were then separated by 10% of SDS–polyacrylamide gel electrophoresis, and the protein was transferred to PVDF sheets (PVDF Transfer Membrane Westran, Schleicher & Schuell, Germany). Immunostaining was achieved using specific antibodies anti-PR (sc-538; Santa Cruz Biotechnology, USA), anti-COX-1 (sc-1752; Santa Cruz Biotechnology) or anti-COX-2 (cat no. 160107; Cayman Chemical, USA). Development was performed with alkaline phosphatase-linked anti-goat antibody (for COX-1) or anti-rabbit antibody (for PR and COX-2) (both from Sigma), followed with nitroblue tetrazolium (NBT)/5-bromo-4-chloro-3-indolyl phosphate, p-toluidine salt (BCIP) colour development reaction. Blots were digitized using a Gelprinter Plus (TDI, Spain), and the densities of spots were analysed with the program 1-D Manager. Equivalent protein loading and transfer efficiency were verified by staining for {beta}-actin (from Sigma).

Assay of prostacyclin
After incubation with the desired products, medium was collected and stored at –20 °C until prostacyclin assay. Culture wells were then washed with PBS and adherent cells were collected in 0.5 mol/l NaOH solution for protein determination by the modified Lowry method using bovine serum albumin as standard (Lowry et al., 1951Go).

The amount of prostacyclin produced, calculated as the concentration of stable hydrolysis product, 6-keto-prostaglandin F1{alpha}, was assessed in duplicate by a commercial enzyme immunoassay kit (Cayman Chemical). The production of prostacyclin was expressed as ng prostacyclin/mg protein.

With the aim of discerning which of two COX isoenzymes was implicated in prostacyclin production, selective COX-1 (0.1 µmol/l SC-560; Cayman Chemical) or COX-2 (1 µmol/l NS-398; Cayman Chemical) inhibitors were added to some wells.

RNA isolation and real-time RT–PCR assay
Total cellular RNA was extracted by using the TRIzol® reagent (Invitrogen, USA) following the manufacturer's instructions. RT was carried out using SuperScriptTM First-Strand Synthesis System for RT–PCR (Invitrogen) by using a personal Mastercycler Eppendorf Thermocycler (Eppendorf, Germany). One microgram of total RNA was reverse-transcribed to cDNA following the manufacturer's instructions. For each RT, a blank was prepared using all the reagents except the RNA sample (for which an equivalent volume of diethylpyrocarbonate-treated water was substituted) and was used as non-template control in real-time PCR experiments.

Primers for quantitative RT–PCR were designed using the Primers Express Software (Applied Biosystems, USA) and synthesized by Custom Primers (Life Technologies, Spain). The sequence of the glyceraldehyde-3-phosphate dehydrogenase (GAPDH) sense primer was 5'-CTGCTCCTCCTGTTCGACAGT-3' and that of the antisense primer was 5'-CCGTTGACTCCGACCTTCAC-3' (NCBI#: NM_002046) giving rise to an expected PCR product of 100 bp. The COX-1 primers were designed to amplify a 168 bp PCR product and these were the following: 5'-TACTCACAGTGCGCTCCAAC-3' for the sense primer and 5'-GCAACTGCTTCTTCCCTTTG-3' for the antisense primer (NCBI#: AF440204). Related to COX-2, primers used were the following: 5'-ATCATAAGGAGGGCCAGCT-3' for the sense primer and 5'-AAGGCGCAGTTTACGCTGTC-3' for the antisense primer, and a 101 bp product was expected (NCBI#: D28235).

RT–PCR was performed using an ABI Prism 7700 Sequence Detection System (Applied Biosystems) with a heated lid (105 °C), an initial denaturation step at 95 °C for 10 min, followed by 40 cycles of 95 °C for 15 s and 60 °C for 1 min. To amplify cDNA, the RT samples were diluted 1/100 and 1/200. In each reaction a total of 1 µl from each RT tube was mixed with 12.5 µl of SYBR Green PCR master mix (Applied Biosystems) containing nucleotides, Taq DNA polymerase, MgCl2 and reaction buffer with SYBR green; 1.5 µl of each 2.5 µmol/l specific primers and double-distilled water were added to a final volume of 25 µl. Each sample was amplified in duplicate for COX-1, COX-2 and GAPDH. In parallel, 5-fold serial dilutions of well-known cDNA concentrations were run as calibration curves. Data were analysed with the ABI Prism 1.7 analysis software (Applied Biosystems). Duplicates showing >5% variation were discarded. To validate a RT–PCR, standard curves with r>0.95 and slope values between –3.1 and –3.4 were required. The amounts of COX-1 and COX-2 were normalized to the corresponding values of the housekeeping gene GAPDH to estimate and compare the relative COX-1 and COX-2 expression among samples. Experiments were performed four times.

In some samples, PCR products were purified by using MiniElute PCR Purification Kit (Qiagen, USA) and were sequenced to prove that the amplified bands corresponded to previously published COX-1, COX-2 and GAPDH corresponding sequences. Agarose gel electrophoreses were also performed to demonstrate that RT–PCR yielded a unique band.

Cell viability measurement
Cell respiration, an indicator of cell viability, was assessed by mitochondrial-dependent reduction of 3-(4.5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) to formazan. Experiments were performed in parallel and with the same protocol described for prostacyclin production, but were performed in 96-well plates. At the end of each experiment, medium was discarded and cells were incubated with 0.1 mg/ml MTT dissolved in phenol red-free medium 199 for 3 h. Medium was then removed by aspiration and formazan contained in cells was solubilized with 100 µl DMSO. The extent of reduction of MTT to formazan was quantified through the measurement of optical density at 540 nm by using a microplate reader (Bio-Rad, USA). Results were expressed as relative percentage of formazan produced by cells maintained in phenol red-free medium 199 without treatments.

Statistical analysis
Values are reported as means ± SEM. Repeated-measures analysis of variance test was applied for comparisons of means, and then Student's t-test was performed. P<0.05 was considered significant.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
The first step of the present work was to ensure the presence of PR on HUVEC. Western blot analysis confirmed the presence of PR with the use of an antibody that recognizes both forms of PR, PRA and PRB. Cells clearly exhibited a 94 kDa band, corresponding to PRA, and a less evident 114 kDa band, corresponding to PRB (Figure 1). There were no significant differences between both PR protein expressions after the different treatments.



View larger version (20K):
[in this window]
[in a new window]
 
Figure 1. Progesterone receptor expression in cultured endothelial cells. Human umbilical vein endothelial cells were exposed to different combinations of progesterone (10 nmol/l) or medroxyprogesterone acetate (10 nmol/l) with 10 µmol/l RU-486 (mifepristone, a progesterone receptor antagonist) for 24 h. Equal amounts of protein (range: 80–125 µg) were subjected to 10% acrylamide gels electrophoresis and immunoblotted with a specific antibody anti-PR, as described in Materials and methods. A typical immunoblotting image with the bands corresponding to PRA (~94 kDa) and PRB (~114 kDa) is presented.

 
To investigate the effects of progesterone and MPA on prostacyclin production, HUVEC were first exposed to either 10 nmol/l of progesterone or 10 nmol/l MPA during different times of incubation, up to 48 h (Figure 2). There was a sustained, spontaneous production of prostacyclin in control, non-stimulated endothelial cells (exposed only to vehicle), for 48 h. Control prostacyclin production at 16, 24 and 48 h was significantly higher (P<0.05) than those of shorter incubation times. Exposure to 10 nmol/l progesterone or 10 nmol/l MPA increased prostacyclin production, compared to control values, after 24 h (P<0.05), and remained augmented up to 48 h. Therefore, the remaining experiments were performed at 24 h of incubation with different compounds. In overall experiments, prostacyclin production in control endothelial cells was 1.53±0.12 ng/mg protein.



View larger version (18K):
[in this window]
[in a new window]
 
Figure 2. Time-course of progesterone and medroxyprogesterone acetate effects on 6-keto-prostaglandin F1{alpha} production by cultured endothelial cells. Human umbilical vein endothelial cells were exposed to 10 nmol/l of either progesterone or medroxyprogesterone acetate (MPA) for the indicated time periods (8–48 h), culture medium was then collected and 6-keto-prostaglandin F1{alpha} concentration was measured as described in Materials and methods. Data are expressed as ng of 6-keto-prostaglandin F1{alpha}/mg of protein, and are mean ± SEM of six to nine duplicated determinations corresponding to two different experiments performed in cells from different cultures. At each time-point, control values were higher (P<0.05) than previous. *P<0.05 versus control values for both progesterone and medroxyprogesterone acetate (MPA) treatments at the same time-point.

 
HUVEC exposure to three different, near-physiological concentrations of progesterone (1–100 nmol/l) resulted in a dose-dependent increase of prostacyclin production (P<0.005 versus control values), being higher with 100 nmol/l than with 1 nmol/l progesterone (P<0.05) (Figure 3). The same concentrations of MPA also increased prostacyclin production (P<0.005 versus control values), but there were no differences between the effects induced by the three tested concentrations. These effects are mediated through PR activation, since treatment of cells with 10 µmol/l RU-486 completely abolished both progesterone (P<0.005)- and MPA (P<0.02)-increased prostacyclin production (Figure 4).



View larger version (27K):
[in this window]
[in a new window]
 
Figure 3. Dose-dependent stimulation of prostacyclin production by endothelial cells after exposure to progesterone or medroxyprogesterone acetate (MPA). Human umbilical vein endothelial cells were exposed to different concentrations (1–100 nmol/l) of either progesterone or MPA for 24 h, culture medium was then collected and 6-keto-prostaglandin F1{alpha} concentration was measured as described in Materials and methods. Data are expressed as percentage of control values, and are mean ± SEM of eight to 21 duplicated determinations corresponding to four different experiments performed in cells from different cultures. Average control values for all experiments were 1.67±0.24 ng/mg protein (range: 0.45–3.77 ng/mg protein). *P<0.005 versus control values and {dagger}P<0.05 versus 1 nmol/l progesterone values.

 


View larger version (24K):
[in this window]
[in a new window]
 
Figure 4. Progesterone and medroxyprogesterone acetate (MPA) stimulate prostacyclin production by endothelial cells through progesterone receptor activation. Human umbilical vein endothelial cells were exposed to different combinations of progesterone (10 nmol/l) or MPA (10 nmol/l) with 10 µmol/l RU-486 (mifepristone, a progesterone receptor antagonist) for 24 h. Culture medium was then collected and 6-keto-prostaglandin F1{alpha} concentration was measured as described in Materials and methods. Data are expressed as percentage of control values, and are mean±SEM of 12–18 duplicated determinations corresponding to five different experiments performed in cells from different cultures. Average control values for all experiments were 1.85 ± 0.17 ng/mg protein (range: 0.60–2.37 ng/mg protein). *P<0.005 versus control, {dagger}P<0.005 versus progesterone and {ddagger}P<0.02 versus MPA values.

 
To examine the role of COX-1 or COX-2 in the observed effects, experiments were conducted to study COX mRNA expression, protein content as well as COX-1 and COX-2 activities. Both progesterone as well as MPA treatments resulted in an increased HUVEC expression of COX-1 and COX-2 mRNA (P<0.05 versus control values) (Figure 5). Protein concentration also reflected the changes in mRNA expression, since not only progesterone, but also MPA increased COX-1 and COX-2 protein concentration (P<0.05 versus control values) (Figure 6). Treatment of cells with RU-486 alone modified neither mRNA nor protein content of either COX, but completely abolished the effects afforded by progesterone and MPA on mRNA expression and protein content of both COX-1 and COX-2 (P>0.05 versus control values) (Figures 5 and 6).



View larger version (20K):
[in this window]
[in a new window]
 
Figure 5. Progesterone and medroxyprogesterone acetate (MPA) stimulate COX-1 and COX-2 relative mRNA expression through progesterone receptor activation. Human umbilical vein endothelial cells were exposed to different combinations of progesterone (10 nmol/l) or MPA (10 nmol/l) with 10 µmol/l RU-486 (mifepristone, a progesterone receptor antagonist) for 24 h. Total RNA was extracted and relative expression of COX-1 and -2 was quantified by RT–PCR, as described in Materials and methods. Data are expressed as percentage of control values, and are mean ± SEM of five to eight duplicated determinations corresponding to four different experiments performed in cells from different cultures. Average control values were 1.09±0.13 (range: 0.55–1.44) and 1.70±0.13 (range: 1.30–1.86) for COX-1 and COX-2 respectively. *P<0.05 versus control values.

 


View larger version (34K):
[in this window]
[in a new window]
 
Figure 6. Progesterone and medroxyprogesterone acetate (MPA) increase both COX-1 and COX-2 protein content in human endothelial cells through progesterone receptor activation. Human umbilical vein endothelial cells were exposed to different combinations of progesterone (10 nmol/l) or MPA (10 nmol/l) with 10 µmol/l RU-486 (mifepristone, a progesterone receptor antagonist) for 24 h. Equal amounts of protein (range: 40–80 µg) were subjected to 10% acrylamide gels electrophoresis and immunoblotted with specific antibodies anti-COX-1 or anti-COX-2, as described in Materials and methods. Typical immunoblots and relative levels assessed by densitometry of bands of 70 kDa (COX-1) and of 72 kDa (COX-2) are presented. Data are expressed as percentage of control values, and are mean±SEM of four to six western blots corresponding to three different experiments performed in cells from different cultures. *P<0.05 versus respective control values.

 
Although progesterone- and MPA-increased production of prostacyclin seems to be mediated through increased mRNA and protein expression of both COX-1 and COX-2, experiments were performed to study the role of both enzyme activities on prostacyclin production by using specific COX-1 and COX-2 inhibitors (Figure 7). The use of 0.1 µmol/l SC-560 alone, an inhibitor of COX-1, decreased prostacyclin production to 44% of control values (P<0.005). When used in combination with progesterone or MPA, SC-560 completely abolished the increased production of prostacyclin induced by both progestogens (P<0.001), decreasing prostacyclin content to the same level of SC-560 alone. NS-398 (1 µmol/l, a COX-2 inhibitor) induced similar results. Inhibition of COX-2 activity diminished prostacyclin production to 41% of control values (P<0.005). When used in combination with either progesterone or MPA, NS-398 significantly (P<0.001) decreased prostacyclin production to the same levels as when used alone.



View larger version (24K):
[in this window]
[in a new window]
 
Figure 7. Role of COX-1 or COX-2 inhibition on progesterone- and medroxyprogesterone acetate (MPA)-stimulated prostacyclin production by human endothelial cells. Human umbilical vein endothelial cells were exposed to different combinations of progesterone (10 nmol/l) or MPA (10 nmol/l) with 0.1 µmol/l SC-560 (a COX-1 antagonist) or with 1 µmol/l NS-398 (a COX-2 antagonist) for 24 h. Culture medium was then collected and 6-keto-prostaglandin F1{alpha} concentration was measured as described in Materials and methods. Data are expressed as percentage of control values, and are mean ± SEM of nine to 16 duplicated determinations corresponding to five different experiments performed in cells from different cultures. Average control values for all experiments were 1.48 ± 0.12 ng/mg protein (range: 0.32–2.37 ng/mg protein). *P<0.005 versus control values, and {dagger}P<0.001 versus either progesterone or MPA values.

 
Results from Figure 7 revealed that pharmacological blockade of COX-1 and COX-2 inhibited prostacyclin production to virtually the same extent and left a residual prostacyclin production in control HUVEC. We used the non-selective COX inhibitor indomethacin to block both COX-1 and COX-2 activities (data not shown). In cells exposed to 10 µmol/l indomethacin, prostacyclin production was 29±4% of control values (P<0.001). Exposure of HUVEC to indomethacin and either progesterone or MPA resulted in a prostacyclin production of 33±5 and 31 ± 6% of control values respectively (P<0.001 versus control and indomethacin for both progestogens). Therefore, there was also a residual production of prostacyclin when both COX were inhibited, probably reflecting an intracellular pool of prostacyclin released to culture medium after COX inhibition.

The physiological interaction between estradiol and progestogens on prostacyclin production was tested by exposing cells to a physiological concentration of estradiol and either progesterone or MPA (Figure 8). Treatment of cells with 1 nmol/l estradiol alone increased prostacyclin production by 25% (P<0.005 versus control values). When used in combination with progestogens, the amount of prostacyclin released was the addition of the effects of estradiol (25%) and progestogens (56 and 55% for progesterone and MPA alone respectively). In both cases, prostacyclin production induced by co-exposure to estradiol and progestogens was significantly higher than by exposure to estradiol alone (P<0.05).



View larger version (26K):
[in this window]
[in a new window]
 
Figure 8. Progesterone and medroxyprogesterone acetate (MPA) do not counteract estradiol-induced prostacyclin production by endothelial cells. Human umbilical vein endothelial cells were exposed to different combinations of progesterone (10 nmol/l) or MPA (10 nmol/l) with 1 nmol/l estradiol for 24 h. Culture medium was collected and 6-keto-prostaglandin F1{alpha} concentration was measured as described in Materials and methods. Data are expressed as percentage of control values, and are mean ± SEM of seven to 14 duplicated determinations corresponding to four different experiments performed in cells from different cultures. Average control values for all experiments were 1.61 ± 0.48 ng/mg protein (range: 0.58–4.70 ng/mg protein). *P<0.005 versus control and {dagger}P<0.05 versus estradiol alone.

 
The possible toxic effect on HUVEC of some of the compounds used was discarded by experiments performed in parallel and the measurement of MTT production. Cell viability after all treatments (progesterone, MPA, RU-486, SC-560, NS-398, estradiol and their combinations) was the same as control cells maintained without treatments (data not shown).


    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
In this study we demonstrated that both progesterone and MPA are able to increase HUVEC prostacyclin production within 24 h of incubation, through PR-mediated mechanisms, probably involving both enhanced COX-1 and COX-2 expressions and activities.

Our results confirm previous reports regarding the presence of both forms of PR on cultured endothelial cells (Vazquez et al., 1999Go). Nevertheless, PRB expression is much less detectable than PRA expression, which is in contrast with data from previous studies. For instance, some authors have obtained similar levels of protein expression for PRA and PRB in other types of human endothelial cells (Vazquez et al., 1999Go). Also, mRNA expression studies demonstrated a high expression of the PRB isoform in HUVEC (Tatsumi et al., 2002Go). The relative roles of PRA and PRB tend to support the view that PRB is the active PR, whereas PRA is either inactive or acts as an inhibitor of PRB activity. However, both isoforms are often co-expressed in normal physiology and exert both cooperative actions and distinct activities (Graham and Clarke, 2002Go).

Progesterone increased prostacyclin production in a dose-dependent manner. Physiological concentrations of progesterone were used in the present study, since in the follicular phase, plasma progesterone levels are 0.15–0.70 ng/ml (corresponding to 0.5–2.2 nmol/l) and in the luteal phase are 2.00–25.0 ng/ml (corresponding to 6.4–79.5 nmol/l). Nevertheless, the use of the same concentrations of MPA resulted in an increased, although not dose-dependent, prostacyclin production. The explanation for that difference remains elusive, although the wide range of values could be the main reason, since there is a tendency to increase prostacyclin production with increased MPA concentrations.

Increased prostacyclin production induced by progestogens supports an active role for these hormones on endothelial function, thus discarding previous reports in which it has been denied (Lewis et al., 1986Go; Mueck et al., 2002Go). Furthermore, the production of prostacyclin by HUVEC after exposure to progesterone or MPA is comparable to that afforded by physiological concentrations of other steroid hormones such as estradiol (Mikkola et al., 1995Go, 1996bGo; Akarasereenont et al., 2000Go).

Time-course analysis reveals a considerable latency required for the increased prostacyclin production, as progesterone and MPA effects are evident only after 24 h of incubation, thus suggesting a genomic effect. The use of the antiprogestogen RU-486 at a dose within the adequate range to ensure PR antagonism (Xu et al., 2002Go) permits the ascription of the observed effects to PR. RU-486 alone had no effect on basal prostacyclin synthesis by HUVEC, as has already been demonstrated in pulmonary artery endothelial cells (Jun et al., 1999Go). When used in combination with progesterone or MPA, RU-486 completely abolished their effects on prostacyclin production. Taken together, data from Figures 2 and 4 indicate that progesterone and MPA enhanced prostacyclin release which is, at least in part, a genomic action mediated through PR activation.

Nevertheless, other mechanisms cannot be completely refuted. For instance, a rapid (within 5 min) stimulation of prostacyclin production by physiological concentrations of progesterone (1–100 nmol/l) in rat aorta strips has recently been reported, suggesting a non-genomic mechanism. The possible contribution of PR, however, has not been studied (Selles et al., 2002Go).

It is generally believed that COX-1 is ubiquitously expressed (termed as constitutive isoform), whereas COX-2 is induced by mitogens, growth factors, bacterial endotoxin, and cytokines (inducible COX) (Parente and Perretti, 2003Go). Nevertheless, in endothelium [and other tissues, such as uterus (Kim et al., 1999Go)], things may not be so simple, and both COX shared characteristics of constitutive and inducible enzymes.

Data from Figures 5 and 6 demonstrated a basal expression of COX-1 and COX-2, suggesting a constitutive component of COX-dependent prostacyclin production. A careful consideration should be done, however, since other factors, including cytokines, growth factors, etc., may be regulating this level and have not been identified in the present work.

Our results demonstrated a PR-mediated induction of both enzymes after exposure to progesterone or MPA, measured as protein content and as mRNA expression (Figures 5 and 6). These results suggest that COX mRNA in endothelium after exposure to progesterone is highly correlated with protein expression levels, indicating a regulation at the mRNA level (Habermehl et al., 2000Go). Moreover, the use of COX-1 and COX-2 inhibitors (Figure 7) revealed not only that basal production of prostacyclin by HUVEC was equally mediated through both enzymes, but also that both inhibitors similarly abolished progestogen-increased prostacyclin production. Taken together, results from Figures 5Go7 support a role for COX-1 and COX-2 in both basal- and progestogen-induced prostacyclin production.

Therefore, it seems that COX-2 is, at least in part, a constitutive enzyme in HUVEC, and COX-1 an inducible one. Consistent with this role of COX-2, clinical studies with celecoxib, a selective inhibitor of COX-2, have shown that this enzyme exerts control of most systemic prostacylin production in healthy humans (McAdam et al., 1999Go).

In addition, COX-1 induction in cultured endothelial cells has previously been reported. As stated in the Introduction, estradiol augments the protein content of both COX-1 (Jun et al., 1998Go) and COX-2 (Akarasereenont et al., 2000Go). Agreeing with our data, 10 day treatment with progesterone increases COX-1 expression in ovine renal artery endothelium (Rupnow et al., 2002Go). Also, pregnancy (a state of both high estrogen and progesterone) increases both COX-1 protein and mRNA expression in endothelial cells from sheep (Janowiak et al., 1998Go). The dependence of COX-1 on progesterone and PR activity has been shown in baboon endometrium, where COX-1 expression was completely inhibited by the use of an antiprogestogen (Kim et al., 1999Go).

Our data could have important clinical and therapeutic implications, taking into account that progestogens are currently and frequently used as a second component of hormonal substitution in post-menopausal women. On the one hand, progestogens are thought to have a neutral effect or even to counteract the estradiol effects on several cardiovascular and haemostatic parameters, such as blood pressure, vascular tone, lipid profile, fibrinogen plasma concentration or the fibrinolytic system (Winkler, 1999Go; Kawano et al., 2001Go; Dubey et al., 2002Go; Ganz, 2002Go). Related to prostacyclin, progestogens counteract estrogen-induced prostacyclin release in endothelial cultured cells exposed to serum or plasma from post-menopausal women in some studies (Mikkola et al., 2000Go), but not others (Mikkola et al., 1996aGo). Results of the present study support an active role for progestogens per se on endothelium-dependent vasodilator production. Moreover, progestogens do not counteract estradiol-induced prostacyclin production and in fact both hormones have an additional effect (Figure 8). Reported differences in the progestogen used or in the design of the experiments seem crucial to an adequate understanding of those results.

On the other hand, and related to the foregoing, our study reveals an equal action for progesterone and MPA. Despite the differences between them on lipid metabolism (Writing Group for the PEPI Trial, 1995Go), blood pressure (Dubey et al., 2002Go), vascular tone (Minshall et al., 1998Go), experimental atherosclerosis (Adams et al., 1997Go), or HUVEC expression of vascular cell adhesion molecule-1 (VCAM-1) (Otsuki et al., 2001Go), with a more prejudicial profile for MPA, both progestogens behaved similarly in every parameter assayed in our study.

Moreover, results of the present work cannot explain the differences observed in the Women's Health Initiative (WHI) clinical trial, which showed that there were no cardioprotective effects of estradiol + MPA, as judged by comparing the number of adverse cardiovascular events in the estradiol + MPA group with that in the placebo group (Rossouw et al., 2002Go). As stated before, there were no differences between MPA and MPA + estradiol-induced prostacyclin production (Figure 8).

Nevertheless, the possible benefits of increased prostacyclin production by progestogens should be carefully considered. It seems clear that the primary effects of prostacyclin should be to induce vasodilatation, but enhanced expression and activity of both COX could also increase the production of other prostanoids with proaggregatory and vasoconstrictor activity, such as thromboxane A2 or prostaglandin F2{alpha} (Ospina et al., 2003Go). In that case, vasodilator and vasoconstrictor effects should be unbalanced, and possible pathological mechanisms promoted.

The reported COX stimulation by progestogens could have other consequences. For instance, it could have an important role on embryo implantation, since estradiol promotes uterine vascular permeability whereas progesterone stimulates uterine angiogenesis during pregnancy (Ma et al., 2001Go), probably through COX-2-derived prostaglandins (Matsumoto et al., 2002Go).

In conclusion, our results demonstrate that progesterone and MPA stimulate prostacyclin production by HUVEC through PR-mediated mechanisms, probably involving both COX-1 and COX-2 enzymes, without counteracting the effects of estradiol.


    Acknowledgements
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
The authors are indebted to Mrs Rosa Aliaga and Mrs Elvira Calap for their excellent technical assistance. Supported by grants 01/0917 and 03/0831 from Fondo de Investigación Sanitaria, Madrid, Spain, SAF2003-08387 from the Ministerio de Ciencia y Tecnología, Madrid, Spain, and grant GV01-69 from Oficina de Ciencia y Tecnología, Generalitat Valenciana, Valencia, Spain. Pilar J.Oviedo is recipient of a fellowship from the Fundación José y Ana Royo, Valencia, Spain.


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Adams MR, Register TC, Golden DL, Wagner JD and Williams JK (1997) Medroxyprogesterone acetate antagonizes inhibitory effects of conjugated equine estrogens on coronary artery atherosclerosis. Arterioscler Thromb Vasc Biol 17, 217–221.[Abstract/Free Full Text]

Akarasereenont P, Techatraisak K, Thaworn A and Chotewuttakorn S (2000) The induction of cyclooxygenase-2 by 17beta-estradiol in endothelial cells is mediated through protein kinase C. Inflamm Res 49, 460–465.[CrossRef][ISI][Medline]

Cano A (1999) Progestins and coronary heart disease. Hum Reprod Update 5, 189–190.[Free Full Text]

Dubey RK, Oparil S, Imthurn B and Jackson EK (2002) Sex hormones and hypertension. Cardiovasc Res 53, 688–708.[CrossRef][ISI][Medline]

Ganz P (2002) Vasomotor and vascular effects of hormone replacement therapy. Am J Cardiol 90, F11–F16.[CrossRef][ISI][Medline]

Graham JD and Clarke C (2002) Progesterone receptors—animal models and cell signaling in breast cancer: expression and transcriptional activity of progesterone receptor A and progesterone receptor B in mammalian cells. Breast Cancer Res 4, 187–190.[CrossRef][ISI][Medline]

Habermehl DA, Janowiak MA, Vagnoni KE, Bird IM and Magness RR (2000) Endothelial vasodilator production by uterine and systemic arteries. IV. Cyclooxygenase isoform expression during the ovarian cycle and pregnancy in sheep. Biol Reprod 62, 781–788.[Abstract/Free Full Text]

Herrington DM and Klein KP (2003) Randomized clinical trials of hormone replacement therapy for treatment or prevention of cardiovascular disease: a review of the findings. Atherosclerosis 166, 203–212.[CrossRef][ISI][Medline]

Jaffe EA, Nachman RL, Becker CG and Minick CR (1973) Culture of human endothelial cells derived from umbilical veins. Identification by morphologic and immunologic criteria. J Clin Invest 52, 2745–2756.[ISI][Medline]

Janowiak MA, Magness RR, Habermehl DA and Bird IM (1998) Pregnancy increases ovine uterine artery endothelial cyclooxygenase-1 expression. Endocrinology 139, 765–771.[Abstract/Free Full Text]

Jun SS, Chen Z, Pace MC and Shaul PW (1998) Estrogen upregulates cyclooxygenase-1 gene expression in ovine fetal pulmonary artery endothelium. J Clin Invest 102, 176–183.[Abstract/Free Full Text]

Jun SS, Chen Z, Pace MC and Shaul PW (1999) Glucocorticoids downregulate cyclooxygenase-1 gene expression and prostacyclin synthesis in fetal pulmonary artery endothelium. Circ Res 84, 193–200.[Abstract/Free Full Text]

Karas RH, van Eickels M, Lydon JP, Roddy S, Kwoun M, Aronovitz M, Baur WE, Conneely O, O'Malley BW and Mendelsohn ME (2001) A complex role for the progesterone receptor in the response to vascular injury. J Clin Invest 108, 611–618.[Abstract/Free Full Text]

Kawano H, Motoyama T, Hirai N, Yoshimura T, Kugiyama K, Ogawa H, Okamura H and Yasue H (2001) Effect of medroxyprogesterone acetate plus estradiol on endothelium-dependent vasodilation in postmenopausal women. Am J Cardiol 87, 238–240.[CrossRef][ISI][Medline]

Kim JJ, Wang J, Bambra C, Das SK, Dey SK and Fazleabas AT (1999) Expression of cyclooxygenase-1 and -2 in the baboon endometrium during the menstrual cycle and pregnancy. Endocrinology 140, 2672–2678.[Abstract/Free Full Text]

Kuller LH (2003) Hormone replacement therapy and risk of cardiovascular disease: implications of the results of the Women's Health Initiative. Arterioscler Thromb Vasc Biol 23, 11–16.[Abstract/Free Full Text]

Lewis GD, Campbell WB and Johnson AR (1986) Inhibition of prostaglandin synthesis by glucocorticoids in human endothelial cells. Endocrinology 119, 62–69.[Abstract]

Lowry OH, Rosebrough NJ, Farr AL and Randall RJ (1951) Protein measurement with the Folin phenol reagent. J Biol Chem 196, 265–275.[ISI]

Ma W, Tan J, Matsumoto H, Robert B, Abrahamson DR, Das SK and Dey SK (2001) Adult tissue angiogenesis: evidence for negative regulation by estrogen in the uterus. Mol Endocrinol 15, 1983–1992.[Abstract/Free Full Text]

Matsumoto H, Ma W, Daikoku T, Zhao X, Paria BC, Das SK, Trzaskos JM and Dey SK (2002) Cyclooxygenase-2 differentially directs uterine angiogenesis during implantation in mice. J Biol Chem 277, 29260–29267.[Abstract/Free Full Text]

McAdam BF, Catella-Lawson F, Mardini IA, Kapoor S, Lawson JA and FitzGerald GA (1999) Systemic biosynthesis of prostacyclin by cyclooxygenase (COX)-2: the human pharmacology of a selective inhibitor of COX-2. Proc Natl Acad Sci USA 96, 272–277.[Abstract/Free Full Text]

Mendelsohn ME and Karas RH (1999) The protective effects of estrogen on the cardiovascular system. New Engl J Med 340, 1801–1811.[Free Full Text]

Mikkola T, Turunen P, Avela K, Orpana A, Viinikka L and Ylikorkala O (1995) 17 beta-estradiol stimulates prostacyclin, but not endothelin-1, production in human vascular endothelial cells. J Clin Endocrinol Metab 80, 1832–1836.[Abstract]

Mikkola T, Ranta V, Orpana A, Viinikka L and Ylikorkala O (1996a) Hormone replacement therapy modifies the capacity of plasma and serum to regulate prostacyclin and endothelin-1 production in human vascular endothelial cells. Fertil Steril 66, 389–393.[ISI][Medline]

Mikkola T, Ranta V, Orpana A, Ylikorkala O and Viinikka L (1996b) Effect of physiological concentrations of estradiol on PGI2 and NO in endothelial cells. Maturitas 25, 141–147.[CrossRef][ISI][Medline]

Mikkola T, Viinikka L and Ylikorkala O (2000) Administration of transdermal estrogen without progestin increases the capacity of plasma and serum to stimulate prostacyclin production in human vascular endothelial cells. Fertil Steril 73, 72–74.[CrossRef][ISI][Medline]

Minshall RD, Stanczyk FZ, Miyagawa K, Uchida B, Axthelm M, Novy M and Hermsmeyer K (1998) Ovarian steroid protection against coronary artery hyperreactivity in rhesus monkeys. J Clin Endocrinol Metab 83, 649–659.[Abstract/Free Full Text]

Mueck AO, Seeger H and Wallwiener D (2002) Medroxyprogesterone acetate versus norethisterone: effect on estradiol-induced changes of markers for endothelial function and atherosclerotic plaque characteristics in human female coronary endothelial cell cultures. Menopause 9, 273–281.[CrossRef][ISI][Medline]

Ospina JA, Duckles SP and Krause DN (2003) 17beta-Estradiol decreases vascular tone in cerebral arteries by shifting COX-dependent vasoconstriction to vasodilation. Am J Physiol Heart Circ Physiol 285, H241–H250.[Abstract/Free Full Text]

Otsuki M, Saito H, Xu X, Sumitani S, Kouhara H, Kishimoto T and Kasayama S (2001) Progesterone, but not medroxyprogesterone, inhibits vascular cell adhesion molecule-1 expression in human vascular endothelial cells. Arterioscler Thromb Vasc Biol 21, 243–248.[Abstract/Free Full Text]

Parente L and Perretti M (2003) Advances in the pathophysiology of constitutive and inducible cyclooxygenases: two enzymes in the spotlight. Biochem Pharmacol 65, 153–159.[CrossRef][ISI][Medline]

Ross R (1999) Atherosclerosis: an inflammatory disease. New Engl J Med 340, 115–126.[Free Full Text]

Rossouw JE, Anderson GL, Prentice RL, LaCroix AZ, Kooperberg C, Stefanick ML, Jackson RD, Beresford SA, Howard BV, Johnson KC et al. (2002) Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results from the Women's Health Initiative randomized controlled trial. J Am Med Assoc 288, 321–333.[Abstract/Free Full Text]

Rupnow HL, Phernetton TM, Modrick ML, Wiltbank MC, Bird IM and Magness RR (2002) Endothelial vasodilator production by uterine and systemic arteries. VIII. Estrogen and progesterone effects on cPLA2, COX-1, and PGIS protein expression. Biol Reprod 66, 468–474.[Abstract/Free Full Text]

Sarrel PM (1999) The differential effects of oestrogens and progestins on vascular tone. Hum Reprod Update 5, 205–209.[Abstract/Free Full Text]

Selles J, Polini N, Alvarez C and Massheimer V (2002) Nongenomic action of progesterone in rat aorta: role of nitric oxide and prostaglandins. Cell Signal 14, 431–436.[CrossRef][ISI][Medline]

Smith WL, DeWitt DL and Garavito RM (2000) Cyclooxygenases: structural, cellular, and molecular biology. Annu Rev Biochem 69, 145–182.[CrossRef][ISI][Medline]

Tatsumi H, Kitawaki J, Tanaka K, Hosoda T and Honjo H (2002) Lack of stimulatory effect of dienogest on the expression of intercellular adhesion molecule-1 and vascular cell adhesion molecule-1 by endothelial cell as compared with other synthetic progestins. Maturitas 42, 287–294.[CrossRef][ISI][Medline]

Vazquez F, Rodriguez-Manzaneque JC, Lydon JP, Edwards DP, O'Malley BW and Iruela-Arispe ML (1999) Progesterone regulates proliferation of endothelial cells. J Biol Chem 274, 2185–2192.[Abstract/Free Full Text]

White RE (2002) Estrogen and vascular function. Vasc Pharmacol 38, 73–80.

White MM, Zamudio S, Stevens T, Tyler R, Lindenfeld J, Leslie K and Moore LG (1995) Estrogen, progesterone, and vascular reactivity: potential cellular mechanisms. Endocr Rev 16, 739–751.[CrossRef][ISI][Medline]

Winkler UH (1999) Effects of progestins on cardiovascular diseases: the haemostatic system. Hum Reprod Update 5, 200–204.[Abstract/Free Full Text]

Writing Group for the PEPI Trial (1995) Effects of estrogen or estrogen/progestin regimens on heart disease risk factors in postmenopausal women. The Postmenopausal Estrogen/Progestin Interventions (PEPI) Trial. J Am Med Assoc 273, 199–208.[Abstract]

Xu X, Otsuki M, Sumitani S, Saito H, Kouhara H and Kasayama S (2002) RU486 antagonizes the inhibitory effect of peroxisome proliferator-activated receptor {alpha} on interleukin-6 production in vascular endothelial cells. J Steroid Biochem Mol Biol 81, 141–146.[CrossRef][ISI][Medline]

Submitted on October 13, 2004; resubmitted on January 14, 2005; accepted on January 19, 2005.