Roles of prostaglandin F2alpha and hydrogen peroxide in the regulation of Copper/Zinc superoxide dismutase in bovine corpus luteum and luteal endothelial cells

Background Prostaglandin F2alpha (PGF) induces luteolysis in cow by inducing a rapid reduction in progesterone production (functional luteolysis) followed by tissue degeneration (structural luteolysis). However the mechanisms of action of PGF remain unclear. Reactive oxygen species (ROS) play important roles in regulating the luteolytic action of PGF. The local concentration of ROS is controlled by superoxide dismutase (SOD), the main enzyme involved in the control of intraluteal ROS. Thus SOD seems to be involved in luteolysis process induced by PGF in cow. Methods To determine the dynamic relationship between PGF and ROS in bovine corpus luteum (CL) during luteolysis, we determined the time-dependent change of Copper/Zinc SOD (SOD1) in CL tissues after PGF treatment in vivo. We also investigated whether PGF and hydrogen peroxide (H2O2) modulates SOD1 expression and SOD activity in cultured bovine luteal endothelial cells (LECs) in vitro. Results Following administration of a luteolytic dose of PGF analogue (0 h) to cows at the mid-luteal stage, the expression of SOD1 mRNA and protein, and total SOD activity in CL tissues increased between 0.5 and 2 h, but fell below the initial (0 h) level at 24 h post-treatment. In cultured LECs, the expression of SOD1 mRNA was stimulated by PGF (1–10 microM) and H2O2 (10–100 microM) at 2 h (P<0.05). PGF and H2O2 increased SOD1 protein expression and total SOD activity at 2 h (P<0.05), whereas PGF and H2O2 inhibited SOD1 protein expressions and total SOD activity at 24 h (P<0.05). In addition, H2O2 stimulated PGF biosynthesis at 2 and 24 h in bovine LECs. Overall results indicate that, SOD is regulated by PGF and ROS in bovine LECs. SOD may play a role in controlling intraluteal PGF and ROS action during functional and structural luteolysis in cows.


Background
Prostaglandin F 2α (PGF) from the uterus or from the ovary is responsible for the regression of the corpus luteum (CL) in mammals [1,2]. In vivo studies in cows demonstrated that intramuscular injections of PGF analogues given in the mid-luteal stage induce an acute decrease in progesterone (P 4 ) production (functional luteolysis) followed by tissue degradation and a decrease in size of the CL (structural luteolysis) [2]. On the other hand, in vitro studies showed that direct treatment of pure populations of luteal steroidogenic cells (LSCs) with PGF does not inhibit basal P 4 production by the large LSCs, and stimulates P 4 production by the small LSCs and by a mixture of large and small LSCs [3,4] suggesting that PGF action differs in each type of luteal cells. The exact mechanism involved in the luteolytic cascade initiated by PGF remains unclear. Interestingly, several studies reported the presence of PGF receptors in luteal endothelial cells (LECs) and that LECs respond to PGF stimulation [5,6], suggesting that LECs are a target for PGF, and that LECs play roles in the local mechanism of CL regression induced by PGF.
Reactive oxygen species (ROS) including hydrogen peroxide (H 2 O 2 ), superoxide anion (O 2 -) and hydroxyl radical (OH -) have been implicated in the luteolytic process [7]. In rats, PGF induces a decrease in serum concentrations of P 4 in association with increasing generation of superoxide anion and H 2 O 2 in the luteal tissue [8]. Treatment of LSCs with PGF induces ROS production and apoptosis [9]. Moreover, H 2 O 2 was found to stimulate PGF production in human endometrial stromal cells [10]. We recently observed that an injection of PGF induces a transient (1-2 h) increase in the partial pressure of oxygen (pO 2 ) in ovarian venous blood [11], and that the pO 2 of venous blood is higher in the ovarian vein than in the jugular vein in cow. Therefore, the luteal microenvironment seems to be exposed to high O 2 condition, especially during the short period of time (1-2 h) following PGF treatment.
Superoxide dismutase (SOD) acts as an antioxidant enzyme in nearly all cells exposed to oxygen. SOD catalyzes the dismutation of superoxide into oxygen and H 2 O 2 . Mammalian cells have three forms of SOD. Copper/Zinc (SOD1) is present in the cytosol, nucleus and the intermembrane space of mitochondria; Manganese SOD (SOD2) is a manganese-containing enzyme that is present in the mitochondrial matrix; Extracellular SOD (SOD3) is a secreted copper-containing protein that is found in the extracellular matrix of tissues [12]. SOD1 dismutates superoxide radicals resulting from cellular oxidative metabolism into H 2 O 2 [13] represents 80% of the total SOD found in the rat CL [14]. In addition, a decrease in intracellular SOD activity was accompanied by a decrease of P 4 production by rat CL [15]. However, the local mechanisms controlling the luteolytic actions of PGF in LECs remain unknown. We hypothesized that SOD, the main enzyme involved in the control of intraluteal ROS, is differently regulated at the time of functional (2 h) and structural (24 h) luteolysis.
To test the above hypothesis, we determined the timedependent changes of SOD in CL tissues after PGF treatment in vivo and the influence of PGF and H2O2 on SOD1 expression and SOD activity in bovine LECs in vitro.

Collection of CL tissue during PGF-induced luteolysis
The animal procedures were approved by the local institutional animal care and use committee of Polish Academy of Sciences in Olsztyn, Poland (Agreement No. 5/2007, 6/2007 and 88/2007). Healthy normally cycling Polish Holstein Black and White cows were used for collection of CL. Estrus was synchronized in the cows by two injections of an analogue PGF (Dinoprost, Dinolytic; Pharmacia & Upjohn, Belgium) with an 11-days interval. Ovulation was determined by a veterinarian via transrectal ultrasonography examination. Then, CLs were collected by Colpotomy technique using a Hauptner's effeninator (Hauptner and Herberholz, Solingen, Germany) on day 10 post-ovulation, i.e., just before administration of a luteolytic dose of analogue PGF (Dinoprost, Dinolytic; Pharmacia & Upjohn, Belgium) (0 h), and at 0.5, 2, 12 and 24 h post-treatment (n=4 per timepoint). CL tissues were dissected from the ovary and then stored at −80°C until the SOD1 mRNA, protein and total SOD activity analysis.

Bovine LEC isolation and cell culture
LECs were isolated from five CLs at the mid-luteal phase (days 8-12 of the estrous cycle) [16] using magnetic beads as previously described [17] and recently validated in our laboratory [6,18]. Briefly, magnetic tosylactivated beads (Dynabeads M-450, 140.04; Dynal ASA, Oslo, Norway) were coated with 0.15 mg/ml lectin from Bandeiraea simplicifolia (BS-1; L2380; Sigma-Aldrich, St. Louis, MO, USA), which specifically binds the glycoproteins expressed by bovine LECs [17]. Luteal cell (LC) suspension was mixed with beads at a concentration of 4 × 10 8 beads/ml, and incubated for 20 min at 4°C on a rocking platform. The cells were pooled and cultured in endothelial cell (EC) growth medium (MV 2; C22121; Promo Cell, Heidelberg, Germany) at 37°C in a humidified atmosphere of 5% CO 2 in air. Only colonies with a homogeneous cell population were removed with a pipette and cultured in collagen-coated 25 cm 2 culture flasks (690175; Greiner Bio-One, Frickenhausen, Germany). The cultures and passages were repeated until a homogeneous population of pure LECs was obtained. The LECs used in the present study were previously confirmed to be positively stained with rabbit anti-human von Willebrand factor (vWF, F3520; Sigma-Aldrich), isolated LECs expresses CD31 but not 3β-hydroxysteroid dehydrogenase (3β-HSD) mRNA as reported previously [6,17,18]. Experiments were performed on confluent cultures and the cells (LECs) were from passage 2-5.

Cell viability test
LECs cultured in 96-well plates were exposed to H 2 O 2 (0.1 -1000 μM) for the final 24 h of culture. The cell viability was determined by Dojindo Cell Counting Kit including WST-1 (Dojindo, Kumamoto, Japan; No. 345-06463). Briefly, WST-1, a kind of MTT [3-(4,5-dimethyl-2 thiazolyl)-2,5-diphenyl-2 H-tetrazolium bromide], is a yellow tetrazolium salt that is reduced to formazan by live cells containing active mitochondria. The culture medium was replaced with 100 μl D/F without phenol red medium-BSA, and a 10-μl aliquot (0.3% WST-1, 0.2 mM 1-methoxy PMS in PBS, pH 7.4) was added to each well. The cells were then incubated for 4 h at 38°C. The absorbance (A) was read at 450 nm using a microplate reader (Bio-Rad, Hercules, CA; Model 450). Percentage of cytotoxicity was determined by subtracting the mean A of H 2 O 2 -treated wells (A test ) from the mean A of untreated wells (A control ) and then dividing by the mean A of untreated wells (A control ). The mean A of wells in the absence of the cells was subtracted from the mean A of all experimental wells. The percent cytotoxicity was calculated as 100 × (A control -A test )/(A control ).

Dose-dependent effect of PGF and H 2 O 2 on SOD1 mRNA expression at 2 h in vitro
To determine whether LECs were responsive to increased concentrations of PGF and H 2 O 2 in the present culture system, LECs cultured in 24-well plates were incubated for 2 h with or without 0.1-10 μM PGF or 1-100 μM H 2 O 2 (n=4 experiments). The cells were then collected and stored at −80°C for the analysis of SOD1 mRNA expression.

SOD1 protein and total SOD activity at 2 h and at 24 h in vitro
To examine whether SOD is differently regulated by PGF and H 2 O 2 at the time of functional and structural luteolysis, LECs cultured in 75-cm 2 culture flasks were incubated for 2 h (mimicking functional luteolysis) and 24 h (mimicking structural luteolysis) with or without PGF (1 μM) or H 2 O 2 (10 μM). The samples were then used for analysis of SOD1 protein expression and total SOD activity (n=4 experiments). The concentrations of PGF and H 2 O 2 were chosen based on the result of SOD1 mRNA expression in vitro, and the effect of H 2 O 2 on cell viability

PGF production in vitro
To determine the dose-dependent effects of H 2 O 2 on PGF production, LECs cultured in 24-well plates were exposed to H 2 O 2 (1-100 μM) for 2 h or 24 h. After incubation, the conditioned media were collected in 1.5 ml tubes containing 5 μl of a stabilizer solution (0.3 M EDTA, 1% (w/v) acid acetyl salicylic, pH 7.3) and frozen at −30°C until the PGF assay (n=4 experiments).

Reverse transcript-PCR
Total RNA was prepared from the CL tissue or cultured bovine LECs using TRI reagent according to the manufacturer's direction (TRI Reagent RNA isolation protocol, © 2008 Ambion, Inc). One microgram of total RNA of each sample was reverse transcribed using a Super-Script First-Strand Synthesis System for RT-PCR (11904-018; Invitrogen), and the reaction mixture was used in each PCR together with the appropriate oligonucleotide primer pairs. The primer for SOD and beta-actin (ACTB) were designed and characterized as described previously [19]. Primer for SOD1 was: forward: AAGGCCGTCTGCGTGCTGAA; reverse: CAGGTCTC-CAACATGCCTCT (accession No: M81129; product: 240 bp). Primer for ACTB was: Forward: CGGCATT-CACGAAACTACC; Reverse: ATCAAGTCCTCGGCCA-CAC (accession No: AY141970; product: 536).
The RT-PCRs were conducted with the house-keeping gene ACTB as an internal standard. ACTB primer was added at the appropriate cycle number by the "primmerdropping method" as described by Wong et al. [20] with modification [21]. The PCRs were carried out using TaKaRa Taq (R001A; Takara Bio Inc., Shiga, Japan) and a thermal cycler (iCycler; Bio-rad Laboratories, Hercules, CA, USA). The PCR conditions were as follow: activation of DNA polymerase for 20 sec at 95°C, annealing for 1 min at 60°C, and extension for 1 min at 70°C, follow by final extension for 5 min at 72°C. The number of cycles was 27 for SOD and 23 for ACTB. Two-fifths aliquot of each reaction mixture was electrophoresed on a 1.5% agarose gel containing ethidium bromide with a known standard (100-bp ladder, New England Biolabs Inc., Beverely, MA, USA; #N3231S) and photographed under ultraviolet illumination. The integrated density was determined by ImageJ software (Windows version of NIH Image, http://rsb.info. nih.gov/nih-image/, National Institutes of Health). Relative density was quantified by normalization of the integrated density of each corresponding β-actin.

Total SOD activity assay
Total SOD activity was determined in CL tissues collected at 0, 0.5, 2, 12, and 24 h after PGF injection and in the LECs at the end of a 2-h or 24-h incubation period using SOD Assay Kit-WST (S311-08; Dojindo Laboratories, Kumamoto, Japan). Total SOD activity was calculated using a concurrently run SOD standard curve, and expressed as inhibition rate or percentage of control (raw data on total SOD activity was normalized based on protein concentrations, units per mg of cellular protein).

Determination of PGF concentration
The concentration of PGF in the culture medium was determined by enzyme immunoassay (EIA) as described previously [18]. The PGF standard curve ranged from 15.625 to 4000 pg/ml, and the median effective dose (ED 50 ) of the assay was 250 pg/ml. The intra-and inter-assay coefficients of variation were 7.4 and 11.6%, respectively. The cross-reactivities of the antibody were 100% for PGF, 3.0% for PGD2, 1.1% for PGI, 0.15% for PGE2, and < 0.03% for PGA2. The DNA content, estimated using the spectrophotometric method by Labarca & Paigen [23], was used to standardize the PGF concentrations.

Statistical analysis
Data of SOD1 mRNA, protein level and total SOD activity were obtained in four separate experiments. PGF concentration and cell viability were performed in triplicate samples for each experimental group of LECs or CL tissues (in vivo). The statistical significances of differences in the amounts of SOD1 mRNA, protein levels, total SOD activity, cell viability and differences in PGF concentrations were analyzed using two-way analysis of variance (ANOVA) with repeated-measures or one-way ANOVA followed by Fisher's protected least-significant difference (PLSD) procedure as multiple comparison tests. All values were expressed as the mean ± SEM of four separate experiments. A level of P<0.05 was considered to be statistically significant.

SOD1 mRNA, protein expression and total SOD activity during PGF-induced luteolysis in vivo
Injection of a luteolytic dose of PGF increased the expression of SOD1 mRNA ( Figure 1A), SOD1 protein ( Figure 1B) and total SOD activity ( Figure 1C) from 0.5 to 2 h in bovine CL tissues, but decreased at 24 h (P<0.05).

SOD1 protein and total SOD activity at 2 h and at 24 h in vitro
PGF and H 2 O 2 significantly stimulated SOD1 protein expression ( Figure 3A) and total SOD activity ( Figure 3B) at 2 h (P<0.05) in cultured LECs. However, PGF and H 2 O 2 decreased SOD1 protein expression ( Figure 3C) and total SOD activity ( Figure 3D) at 24 h (P<0.05) in cultured LECs.

PGF production in vitro
In cultured LECs, PGF production was simulated by 100 μM H 2 O 2 , but not by lower concentrations at 2 h ( Figure 4A) and 24 h ( Figure 4B) (P<0.05).

Discussion
The present study demonstrated that administration of a luteolytic dose of PGF increased the expression of SOD1 in the CL tissue between 0.5 and 2 h post-treatment, but decreased at 24 h in vivo. Furthermore, in vitro studies examining the effects of PGF and H 2 O 2 on SOD expression in LECs, showed a similar pattern with increased SOD at 2 h followed by an inhibition at 24 h. These results suggest that SOD is temporally regulated by PGF and ROS during luteolysis in cattle. The dose of H 2 O 2 used in the present study to examine SOD1 protein and total SOD activity did not affect significantly the viability of LECs.
Previous in vivo studies in cows demonstrated that injection of a luteolytic dose of PGF induces a transient (1-2 h) increase in the partial pressure of oxygen (pO 2 ) in the ovarian venous blood [11]. In addition, it has been shown that a functional PGF receptor (FPr) is present in bovine LECs, and that PGF acutely increased ROS production in these cells [6]. Moreover, the rat CL produces significant amounts of ROS [24] and increases ROS (H 2 O 2 ) generating capacity within a few hours after injection of a luteolytic dose of PGF [8,9]. In the present study, PGF stimulated the expressions of SOD1 mRNA, protein, and total SOD activity at 2 h in bovine CL tissue and LECs. The temporal pattern of SOD expression suggests a role for ROS and SOD during the beginning functional luteolysis in cows. On the other hand, SOD acts as an antioxidant enzyme, and together with catalase, protects the endothelium of a variety of tissues against ROS [25]. Taken together, SOD and ROS seem to be involved in the luteolytic cascade induced by PGF in bovine CL.
ROS have been implicated in luteolysis due to their ability to increase uterine PGF [10], to decrease P 4 biosynthesis [15] and to induce apoptosis in LSCs [26]. The intraluteal production of ROS is regulated by SOD [15]. SOD catalyzes the dismutation of superoxide into oxygen and H 2 O 2 [12]. SOD1 inhibition increases the steady-state levels of superoxide [27].  in total SOD activity and a decrease in catalase activity during CL regression in mice [28]. Since H 2 O 2 has the capacity to increase PGF production in bovine LSCs [29] and LECs (the present study), as well as to induce apoptosis in bovine LSCs and human umbilical vein endothelial cells [29,30], increased levels of H 2 O 2 may be crucial for luteal regression. The above findings also suggest that a local increase of ROS within the bovine CL facilitates the local production of PGF. Moreover, the concomitant increase of SOD and ROS may be due to the strong stimulatory effect of ROS on SOD in response to the increasing levels of PGF within the CL.
Although luteolytic PGF is derived from the uterus in many species, including ewes [31] and cows [2], a considerable amount of PGF is also synthesized by the CL [32] and LECs represent an important source of PGF [18]. Previous studies have reported that PGF increases the production of ROS in rats [9,24] and cow in vivo [11]. Interestingly, ROS has been demonstrated to stimulate PGF production in the CL of rats [33], cows [29] and human [10]. In the present study, H 2 O 2 stimulated PGF production in cultured bovine LECs. The above results suggest the presence of a positive feedback loop between PGF and ROS in the bovine CL during luteolysis. Also, the increase of intraluteal PGF induced by ROS seems to be crucial for promotion of luteal regression in cow.
The present study demonstrates that PGF and H 2 O 2 inhibited the expression of SOD1 protein in bovine LECs cultured for 24 h. The inhibitory effects of PGF and H 2 O 2 on SOD expression at 24 h contrasted with their stimulatory effects at 2 h. The increase in SOD observed at 2 h seems to be the result of an acute stimulatory effect of PGF on ROS production by LECs. However, the mechanism by which PGF and H 2 O 2 inhibit the expression of SOD1 at 24 h of incubation is unclear. SOD catalyzes the dismutation of superoxide into H 2 O 2 and oxygen, to maintain low-state levels of superoxide [12]. Therefore, a reduction of SOD by PGF and H 2 O 2 at 24 h may enhance intraluteal ROS or superoxide radical accumulation for the promotion of structural luteolysis [29,30]. In support of such an idea, an accumulation of superoxide radicals and a decrease in SOD levels are associated with the inhibition of luteal P 4 secretion [7] and apoptotic cell death [34].
Furthermore, SOD1 protein was localized in LECs (Additional file 1: Figure S1) and other types of luteal cells (Additional file 2: Figure S2). A robust SOD1 protein expression was detected in bovine CL tissue after PGF treatment (Additional file 2: Figure S2). These findings suggest that not only SOD1 in LECs but also in other types of luteal cells including LSCs are regulated during PGF induced luteolysis. Further studies are needed to clarify the relative contribution of LSCs in total luteal SOD during luteolysis in cow.

Conclusions
SOD is regulated by PGF and ROS in bovine LECs. SOD may play roles in controlling intraluteal PGF and ROS action during functional and structural luteolysis in cows.

Additional files
Additional file 1: Figure S1. Immunohistochemical examination of SOD1 in bovine luteal tissue. For localization of SOD1, bovine corpus luteum tissues was fixed with 10% phosphate buffer formalin, embedded with paraffin and cut into 4 micrometers of thickness. For antigen retrieval, sections were incubated in Tris-EDTA buffer (pH 9.0) for 15 min at 98 0 C. Normal horse serum blocking solution was used for inhibition of nonspecific bindings. The slides were then incubated with or without (negative control) SOD1 antibody raised in goat (Santa Cruz; sc-8637) at a dilution rate of 1/200. After washing, sections were incubated with biotinylated rabit anti goat IgG serum as the second antibody (1/4000). Horseradish peroxidase (HRP)-conjugated ABC (Vector Laboratory Inc., Burlingame, CA, USA) was applied to the section at room temperature for 30 min. The binding sites were visualized using 0.02% 3,3 'diaminobenzidine tetrahydrochloride (DAB) in 50 mM Tris-HCl (pH 7.4) containing 0.02% H 2 O 2 . After immunohistochemical staining, the sections were lightly counterstained with Mayer's hematoxylin. The sections were washed in distilled water, dehydrated in a graded series of ethanol, and cleared in xylene, coverslipped and observed under light field microscope. Immunohistochemical representative pictures of SOD1 were shown. Picture A showed the negative control (magnification: 200x). Picture B (magnification: Figure 4 Effect of H 2 O 2 on PGF production in LECs. Dosedependent effects of H 2 O 2 on PGF production in cultured bovine LECs. The cells were cultured with H 2 O 2 (1, 10 and 100 μM) for 2 h ( Figure 4A) and 24 h ( Figure 4B). All values represent mean ± SEM of four separate experiments. Asterisks indicate significant differences compared with untreated cells (P<0.05).