Reproductive Biology and Endocrinology Open Access Prostaglandin F2-alpha Receptor (fpr) Expression on Porcine Corpus Luteum Microvascular Endothelial Cells (pcl-mvecs)

Background: The corpus luteum (CL) is a transient endocrine gland and prostaglandin F2-alpha is considered to be the principal luteolysin in pigs. In this species, the in vivo administration of prostaglandin F2-alpha induces apoptosis in large vessels as early as 6 hours after administration. The presence of the prostaglandin F2-alpha receptor (FPr) on the microvascular endothelial cells (pCL-MVECs) of the porcine corpus luteum has not yet been defined. The aim of the study was to assess FPr expression in pCL-MVECs in the early and mid-luteal phases (EL-p, ML-p), and during pregnancy (P-p). Moreover, the effectiveness of prostaglandin F2-alpha treatment in inducing pCL-MVEC apoptosis was tested.


Background
The corpus luteum (CL) is a transient endocrine gland essential for the regulation of ovarian cycles as well as for establishing and maintaining pregnancy. Prostaglandin F 2α (PGF 2α ) is considered the principal luteolysin in domestic animals and acts on target cells [1] by means of a specific plasma membrane-associate receptor, prostaglandin F2-alpha receptor (FPr). The FPr was detected on steroidogenic luteal cells [2][3][4][5] but its presence on other target cells could not be excluded.
Our previous data [6] showed that, in a porcine CL regression model, cloprostenol (a PGF 2α synthetic analogue) first induces apoptosis of luteal endothelial cells then apoptosis of steroidogenic cells and suggest that the structural regression of the CL may begin with an early response of the endothelial compartment to PGF 2α . This hypothesis relies on the assumption that FPrs are present on endothelial cells (ECs). The expression of the FPr was described by RT-PCR in bovine ECs derived from CLs [7]; on the contrary, FPr mRNA has been identified only on swine steroidogenic cells by an in situ study [8]. Therefore, to date, the presence of the FPr on luteal ECs is not fully clarified, perhaps due to the different techniques utilized.
The porcine CL is not sensitive to the luteolytic stimulus until day 12 of the cycle [9][10][11]. Luteal PGF 2α binding sites increase during CL formation reaching a maximum on day 13 of the oestrous cycle, concurrently with the onset of luteal sensitivity to PGF 2α [3,12]. Limited data exist confirming that an FPr increase is responsible for the acquisition of luteolytic sensitivity.
The importance of the interaction between luteal cells and ECs was demonstrated both in in vivo and in vitro experiments in bovine and swine species [13][14][15][16][17].
The interest in EC biology increased in parallel with the optimization of successful protocols for their isolation and culture, allowing the study of EC involvement in many different physiological or pathological processes. EC isolated from large vessels or from tissue microvasculature show a morphological heterogeneity [18,19] and significant functional differences [20]. Additional differences have been observed among EC isolated from microvessels of different organs [21,22] as well as from different regions of the same organ [23][24][25]. These differences are mainly due to the local micro-environment (e.g., the presence of matrix proteins, soluble cytokines, growth factors and paracrine signals) and to cell-to-cell communication.
In this context, the isolation and characterization of pCL-MVECs may lead to a better insight into endocrine gland physiology and contribute to explaining some physiological differences during the CL lifespan.
In order to obtain pure MVEC cultures from CLs at different stages (early and mid-luteal phases and during pregnancy), the initial aim of the present study was to define a reliable isolation method. Thus, our subsequent aims were to investigate the presence of the FPr, both at the mRNA and the protein level, on pCL-MVEC and to verify if PGF 2α treatment is effective in inducing pCL-MVEC apoptosis.

pCL-MVEC culture set-up
Animals and corpora lutea collection Twelve prepubertal Large White gilts (94 ± 1.47 Kg body weight) were injected i.m. with 1250 iu (international units) of equine chorionic gonadotropin (eCG; Folligon-Intervet, Holland) and, 60 h later, with 750 iu of human chorionic gonadotropin (hCG; Corulon-Intervet) to synchronize ovulation. The animals were randomly assigned to three groups (n. = 4). Four animals were artificially inseminated at 40/44 h post-hCG treatment and pregnancy was determined after 30 days by ultrasonography. The ovaries were surgically removed from all animals: 4 at the early luteal phase (EL-p: 6 days after ovulation), 4 at the mid-luteal phase (ML-p: 12 days after ovulation) and 4 during the pregnancy phase (P-p: 60 days of pregnancy).
All animals were housed and used according to EEC animal care guidelines. The experimental procedures had previously been submitted to and approved by the Ethical Committee of Bologna University.

Isolation and culture conditions of Microvascular Endothelial Cells (MVECs)
The ovaries were immediately taken to the laboratory in ice-cold DPBS (Dulbecco Phosphate Buffer Saline; Cambrex Bio Sciences, Verviers, Belgium) and washed several times in sterile DPBS; the corpora lutea (CLs) were then isolated under sterile conditions. All CLs from each animal (14.1 ± 5.9) were cut into small fragments (~1 mm 3 ) with a razor blade and minced together with a pestle and mortar. Small aliquots (100 µg) of homogenate were immediately frozen in liquid nitrogen for RNA extraction, and protein and progesterone (P 4 ) analyses. The remaining homogenate was washed (3 times) by adding sterile DPBS (1:3, w:v) and centrifuged (500 × g, 5 min). The final pellet was resuspended in sterile DPBS containing collagenase (2 mg/ml; type IA [420 U/mg]; Sigma Chemical Company, St. Louis, MO, USA), BSA (4 mg/ml; Fraction V, Sigma) and DNase I (0.02 mg/ ml; [1.87 U/mg]; Sigma). After 20 min on a rocking platform (200 rpm) at 37°C, enzymatic digestion was stopped by adding FBS (Foetal Bovine Serum, Gibco-Invitrogen, Paisley, UK) (1:1, v/v); the suspension was filtered (pore sizes: 280 µm; Sigma) and placed on ice, then centrifuged (800 × g, 20 min) and resuspended in DPBS-BSA (0.1%; w/v) at a final concentration of 5 × 10 5 cells/ ml (Thoma chamber counting). The pCL-MVECs were then positively selected by an immunomagnetic separation protocol using Dynabeads ® coated with anti-CD31 monoclonal antibody (Invitrogen), according to the manufacturer's instructions. Briefly, the cells were mixed with Dynabeads ® to give a ratio of 5 × 10 5 cells/4 × 10 8 beads/ ml and were then incubated for 20 min at 4°C on a rocking platform.
The last resuspension was carried out in EGM™2-MV medium (Microvascular Endothelial Cell Medium-2, Cambrex); cell viability was determined using the Trypan Blue exclusion test and viable cell density was calculated using a Thoma chamber.

Progesterone assay
Tissue P 4 levels were determined as previously described by Ribeiro et al. [35]. A small amount of frozen CL homogenate (for each animal) was further homogenised by Ultra Turrax in PBS (0.1 g/ml) in an ice bath. An aliquot (20 µl) was extracted with petroleum ether (1 h at RT on a rotary shaker) and centrifuged. The ether was then collected and dried under a N 2 stream. The dried ether extract was resuspended in PBS (1 ml), diluted 1:50 and an aliquot of 50 µl was then assayed using a validated Radio Immuno Assay (RIA) [36]. The P 4 results are expressed as ng/mg tissue. The intra-and inter-assay coefficients of variation were 5.2 % and 9 %, respectively.
The cells were counterstained with propidium iodide (PI) and examined under an epifluorescence microscope (Eclipse E600, Nikon, Japan) with fluorescein (FITC) and tetramethylrhodamine (TRITC) filters using a Nikon digital camera. A minimum of 100 cells was evaluated in each experimental sample.
To evaluate the presence of contaminant steroidogenic luteal cells, a PCR Real time amplification for Steroidogenic acute regulatory protein (StAR) [37] was conducted on pCL-MVEC cultures, as described below.

RNA extraction and Real time PCR
Total RNA was extracted from endothelial cells (pCL-MVECs and pAECs) using RNeasy Mini Kits 50 (Qiagen Sciences Inc, MD, USA) and was treated with an RNasefree DNase set (Qiagen) according to the manufacturer's instruction. Total RNA was extracted from tissue CL homogenates using a TRIZOL isolation reagent (Invitrogen) according to the manufacturer's instructions.
Purified RNAs were resuspended in 25 µl of RNAse-free water and spectrophotometrically quantified (A260 nm); their quality was checked by gel electrophoresis on 2% agarose gel. One microgram of total RNA was reverse-tran-scribed to cDNA using an iScript cDNA Synthesis Kit (Bio-RAD Laboratories Inc., California, USA) for a final volume of 20 µl. Transcription reactions without reverse transcriptase were performed in order to check for possible DNA contamination.
Swine primers (FPr, StAR, HPRT [Hypoxanthine phosphoribosyltransferase]) were designed to span one intron, to avoid genomic DNA amplification, using Beacon Designer 2.07 Software (Premier Biosoft International, Palo Alto, CA, USA). Primer sequences, expected PCR product lengths and sequence accession numbers are shown in Table 1.
Real-time quantitative PCRs were performed in the iCycler Thermal Cycler (Bio-RAD) using SYBR green I detection. A master-mix of the following reaction components was prepared to indicate end-concentrations: 0.5 µl forward primer (0.2 µM), 0.5 µl reverse primer (0.2 µM), 9 µl water and 12.5 µl IQ SYBR Green BioRad Supermix (Bio-RAD Laboratories Inc.). cDNA (2.5 µl) was added to the master mix (22.5 µl). All samples were assayed in duplicate. The real-time PCR protocol employed was the following: initial denaturation for 3 min at 95°C, 40 cycles at 95°C for 15 sec and 60°C for 30 sec, followed by a melting step from 55° to 95°C with a rate of 0.5°C/s. The housekeeping gene HPRT data were used to normalize the RNA amount in all samples.
Real-time efficiency for each primer set was acquired by amplification of a standardized cDNA dilution series. The specificity of the amplified PCR products was verified by melting curve analyses and an agarose gel electrophoresis run. The housekeeping gene HPRT was used to normalize the amount of RNA. FPr mRNA expression was calculated as ∆C T (HPRT C T -FPr C T ) [38] for all samples.

Protein extraction and Western Blot
The protein contents of CL homogenates and endothelial cell lysates (sonication in PBS, 1 × 10 6 cells/200 µl) were determined by a protein assay kit (Sigma). Protein aliquots (20 µg) were separated on NuPage 10 % Bis-Tris Gel (Invitrogen) for 50 min at 200 V and were then electrophoretically transferred onto a nitrocellulose membrane. The blots were washed in PBS, and protein transfer was checked by staining the nitrocellulose membranes with 0.2 % (w/v) Ponceau Red and gels with Comassie Blue. Non-specific protein binding on the nitrocellulose membrane was blocked with 5 % (w/v) milk powder in PBS-T20 (phosphate buffer saline -0.1 % (v/v) Tween-20) for 1 h at RT. The membranes were then incubated with a 1:500 dilution of a rabbit anti-FPr (NLS 1049, Novus Biologicals Inc, Littleton, CO, USA) in TBS-T20 (20 mM Tris-HCl, pH 7.4; 500 mM NaCl; 0.1 % Tween-20) with 0.5 % milk overnight (O/N) at 4°C. After several washings with PBS-T20, the membranes were incubated with the secondary biotin-conjugated antibody (1:80 000) and then with a 1:1000 dilution of an anti-biotin horseradish peroxidase (HRP)-linked antibody. Western blots were developed using a chemiluminescent substrate (Pierce Biotechnology Inc, Rockford, IL, USA) according to the manufacturer's instructions. The membranes were stripped and reprobed by rabbit anti-HPRT (1:250; FL218, Santa Cruz Biotechnology Inc, Santa Cruz CA, USA) in order to normalize the results. The relative protein content was determined by the density of the resultant bands and expressed in arbitrary units (AU) relative to the HPRT content, using Quantity One Software (Bio-Rad).

Immunofluorescent staining
The immunolocalization of the FPr in pCL-MVEC and pAEC cultures was performed by indirect immunofluorescence. Cells grown in slide chambers for 24 h were fixed with acetone and methanol (1:1; v/v) for 10 min at -20°C and washed several times with PBS. To better identify the FPr at the plasma membrane level, immunofluorescent staining was also performed without previous sample fixation. Slides were blocked with 10 % FBS in PBS for 1 h at RT and incubated O/N at 4°C with anti-FPr diluted 1:100 in 10 % FBS in PBS or without a primary antibody (negative control). After several washings in PBS, an FITC-conjugated secondary antibody was added (1:800 at RT for 1 h).
The cells were then counterstained with PI and examined under an epifluorescence microscope equipped with appropriate filters. A minimum of 100 cells were evaluated for each experimental sample.

Statistical analysis
Data were analyzed by one-way analysis of variance (ANOVA); the significant differences were analyzed by the Duncan's test (SPSS Inc, Chicago, IL, USA). A probability of P < 0.05 was considered significant. Data represent mean ± SD (n = 4).
We were able to obtain primary cultures of pCL-MVECs from all animals. The primary cultures reached confluence (about 10 6 cells/flask) in 7.6 ± 2.7 days (min 4, max 12); all cultures derived from EL-p CLs reached confluence faster (5 ± 1 days) than ML-p-and P-p-MVECs.

Characterization of pCL-MVEC cultures
Identification of the endothelial origin was confirmed by CD31 and factor VIII immunofluorescence. As expected, due to the isolation method used, all cells were positively stained for CD31. The factor VIII distribution varied among MVEC phenotypes, as reported on bovines [30], exhibiting a sometimes perinuclear, diffuse or granular cytoplasmatic pattern while some pCL-MVECs were negative (Fig. 1C). No fluorescence was detected in the absence of a primary antibody.
StAR mRNA was not detectable in any pCL MVECs.

Real time quantification of the FPr
FPr mRNA expression was calculated as ∆C T (HPRT C T -FPr C T ). As the HPRT C T were comparable among samples, increased ∆C T values represent an increase in FPr mRNA expression.
FPr mRNA expression levels in CL homogenates and in pCL-MVEC confluent cultures are shown in Fig. 2 and Fig.  3, respectively.
The relative amount of FPr mRNA in luteal tissue significantly increased from EL-to ML-, reaching a maximum in the P-phase ( Fig. 2A). FPr mRNA was detectable in all pCL-MVEC cultures, showing the highest expression in EL-p derived MVECs; FPr mRNA was also detectable in pAECs (Fig. 3A).
ML-p pCL-MVECs analysed at different culture times showed that FPr mRNA was highly expressed in freshly isolated ML-p pCL-MVEC then the expression was significantly reduced after an ON culture and reached the minimum level at confluent stage. The hormonal treatment had no regulative effect on the expression of FPr mRNA expression (Fig 4).
EL-p-derived pCL-MVECs showed a slight but significant increase in the FPr level (Fig. 3B).
The FPr was diffusely distributed in the cytoplasm of all pCL-MVECs and in pAECs (Fig. 5A-D). Moreover, P-p pCL-MVECs showed a clear perinuclear labeling pattern (Fig. 5C). Positive staining was also observed along the plasma membrane and was more evident in non-fixed cells (Fig. 5E). No fluorescence was detected in the absence of an primary antibody.

Discussion
In this study, we describe a reliable method for isolating and setting up primary cultures of pCL-MVECs. A positive selection with the employment of anti-CD-31-coupled magnetic beads was carried out twice. In order to support the attachment of MVECs, a basal medium optimized for bovine lung-derived MVECs supplemented with growth factors (hEGF, VEGF, hFGF, Hydrocortisone) was used until day 4 of the culture. Rare fibroblasts and parenchymal cells were observed in the culture before applying the second positive selection procedure. Cells reached confluence in a specific medium for human endothelial cells supplemented with 15% FBS. The time required to reach confluence varied for each pCL-MVEC culture; however, we observed that EL-p-derived pCL-MVECs proliferated faster than the others. This was in agreement with the high mitotic index of ECs observed during CL formation [40][41][42]. Confluent pCL-MVEC monolayers showed a heterogeneous morphology as documented in bovine CL-MVECs [23,28]. Various cell phenotypes were present with a predominance of cobble-stone-like and spindle- shaped cells in all pCL-MVECs without any relationship between the morphological types and CL phases analyzed, as reported for bovines [39]. The endothelial origin of the cells was confirmed by traditional endothelial cell marker expression. Moreover, the absence of StAR mRNA in the EC cultures indicated the absence of steroidogenic cell contamination in the confluent MVEC cultures.
FPr expression in CL tissues was found in all the stages analyzed with an increasing trend from the EL-to the MLphase reaching the maximum during pregnancy, both at mRNA and protein levels in agreement with data already published for the porcine [3] and the bovine species [43].
Our study documents, for the first time, the presence of the FPr on MVECs isolated from swine corpora lutea. We observed that the degree of expression of FPr mRNA varied among the pCL-MVEC cultures in relation to the different CL phases, the highest in the EL-p, decreasing in the ML-p and reaching its lowest level in the P-phase. At the protein level, the presence of the FPr was evidenced in all EC cultures. The amount of FPr protein was very similar among the MVECs derived from ML-p and P-p CLs and pAECs, and was significantly higher on MVECs derived from EL-p CLs. Our data, showing the presence of the FPr on MVECs at early stages testify to the fact that the refractoriness in this species was not be related to the absence of the receptor and rather that other mechanisms must be considered.
The FPr expression level in freshly isolated, overnight and long-term cultured MVECs, was gradually reduced reaching its minimum value at the confluent stage, in agreement with a recent study [31] showing that ECs modify their phenotypes when removed from the microenvironment. Nevertheless, in our model, steroid supplementation (P 4 and E 2 ) resembling the in vivo hormonal environment was not able to regulate FPr expression. Further studies may be required to establish if and how culture conditions may influence FPr expression as well as to investigate if the receptor is functional on pCL-MVECs.
The FPr presence on porcine MVECs is in agreement with data obtained for bovine CLs [7] and contrast with those obtained for swine CLs in an in situ mRNA study [8], thus confirming the hypothesis that the discrepancies among these studies are related to the different techniques used.
Finally, the immunolocalization of the FPr on MVECs showed positive staining at plasma membrane level as well as a diffuse perinuclear and cytoplasmatic localization in all cultures. Perinuclear FPr immunoreactivity may be an interaction between the outer membrane of the nuclei and the endoplasmic reticulum where nascent FP protein is formed before being transported to the plasma membrane. The perinuclear staining of P-p CL-MVECs was more intense than others, suggesting an intensive synthesis/glycosilation of the receptor. All these findings are consistent with the distribution of the FPr on ECs of human ocular tissue [44] and on the rat myometrium [45] as well as other prostaglandin receptors on porcine cerebral endothelial cells [46].
In our model, PGF 2α failed to induce an apoptotic response in all pCL MVEC cultures, in agreement with other in vitro studies conducted on bovines [30,47]. These results contrast with in vivo experiments which indicated endothelial cell apoptosis as an early hallmark of PGF 2αinduced luteal regression in many species [5,6,48]. Therefore, we can affirm that PGF 2α does not directly induce apoptosis of porcine MVECs in our in vitro model, but the involvement of other molecules should be investigated [47,49,50].

Authors' contributions
All the authors participated in the experimental design and collected biological material. AZ, CB and TR isolated and optimized the MVEC cultures. CB and TR carried out characterization of the MVECs. AZ and LAR carried out RNA extraction and real-time RT-PCR. CB performed the Western blots. MF performed the immunostaining of the MVECs, apoptosis and statistical analysis. MLB was responsible for animal care and surgical procedures. MLB conceived and supervised the study. AZ, MF and MLB wrote the manuscript. All authors read and approved the final manuscript.