Reproductive Biology and Endocrinology Open Access Phospholipase A2 Regulation of Bovine Endometrial (bend) Cell Prostaglandin Production

Background: Prostaglandins (PG), produced by the uterine endometrium, are key regulators of several reproductive events, including estrous cyclicity, implantation, pregnancy maintenance and parturition. Phospholipase A2 (PLA2) catalyzes the release of arachidonic acid from membrane phospholipids, the rate-limiting step in PG biosynthesis. The bovine endometrial (BEND) cell line has served as a model system for investigating regulation of signaling mechanisms involved in uterine PG production but information concerning the specific PLA2 enzymes involved and their role in regulation of this process is limited. The objectives of this investigation were to evaluate the expression and activities of calcium-dependent group IVA (PLA2G4A) and calcium-independent group VI (PLA2G6) enzymes in the regulation of BEND cell PG production.


Background
Prostaglandins, produced by the endometrial epithelium, are important regulators of several reproductive processes, including estrous cyclicity, implantation, pregnancy maintenance and parturition [1]. Prostaglandin (PG) biosynthesis is dependent on arachidonic acid (AA) release from membrane phospholipids catalyzed by phospholipase A 2 enzymes [reviewed in [2]]. Arachidonic acid is then metabolized to intermediate products, PGG 2 and PGH 2 , by a cyclooxygenase reaction and by a peroxidase reaction, respectively, both performed by cyclooxygenase (COX) -1 and/or -2. Prostaglandin H 2 is converted to bioactive prostaglandins, such as PGF 2α , PGE 2 , PGD 2 and PGI 2 , by terminal PG synthases, which may exhibit tissue specific distribution [3].
Bovine and ovine endometrial explants and epithelial cell cultures have proven to be functional models for analysis of pathways that regulate PG biosynthesis. Early studies used endometrial explants [4,5] or glandular endometrial epithelial cells [5][6][7] harvested from animals at late diestrus. More recent studies have utilized primary or early passage luminal epithelial (LE) cells collected from animals early in the cycle (days 1-4) because the luminal epithelium is the major site of endometrial PG production and these cells exhibit much better growth characteristics than LE cells collected during diestrus [8][9][10][11]. Results from experiments with explants and glandular or luminal epithelial cells are consistent; oxytocin stimulates PGF 2α and PGE 2 production and interferon-tau (IFNT) diminishes this response. Bovine endometrial epithelial cells produce greater quantities of PGF 2α than PGE 2 and the PGF 2α response to oxytocin stimulation is stronger. The cellular response to IFNT, alone, is biphasic. Low concentrations (< 1 μg/ml) of IFNT diminish basal PG production and high concentrations (>1 μg/ml) stimulate PG production [10]. Interestingly, both low and high concentrations of IFNT diminish oxytocin stimulated PG production. Agonist-stimulated PG production, by oxytocin or high concentrations of IFNT, is associated with increased expression of COX-2 mRNA and protein [9,10], prostaglandin E 2 synthase (PGES) [10], and prostaglandin F 2α synthase (PGFS) mRNA [11]. Attenuation of oxytocinstimulated PGF 2α production is associated with decreased COX-2 production [9,10] and alteration in the expression of terminal PGF synthase [11].
Numerous studies on the regulation of PG biosynthesis have been performed with the BEND, the bovine endometrial cell line [reviewed in [12]]. BEND cells have been described as spontaneously immortalized endometrial cells originally isolated from a cow on day 14 of the cycle [13]. They have been described to have both epitheliallike [12,13] and stromal-like [14] morphological characteristics. BEND cells produce significantly more (10-20 fold) PGE 2 than PGF 2α under both resting and stimulated conditions, and they respond to phorbol ester (phorbol 12-myristate 13-acetate, PMA, or phorbol 12,13 dibutyrate, PDBu) stimulation, but not oxytocin stimulation, with increased PG production [12,14,15]. Similar to primary endometrial epithelial cells, BEND cells increase expression of COX-2 mRNA [16], protein and PGFS [15] when PG production is stimulated and these responses are diminished by IFNT at concentration less than 5 μg/ml [15]. Unlike primary endometrial epithelial cells, BEND cells are not stimulated to produce PGs by high levels of IFNT alone and when treated with both PDBu and high levels of IFNT (>5-20 μg/ml) inhibitory activities of IFNT on PG production are ameliorated [15].
Despite the fact that phospholipase A 2 (PLA 2 ) catalyzes the release of arachidonic acid, the rate-limiting step in PG biosynthesis, most studies on the regulation of uterine PG production have focused on expression and activation of downstream enzymes such as cyclooxygenases, which catalyze the committed step, and PG synthases. The reports [6,9,16] that have included analysis of PLA 2 expression in domestic ruminant uterine PG production, focused on a single enzyme, cPLA 2α , (Group IVA PLA 2 or PLA2G4A). Interestingly, while none of these studies identified consistent significant changes in expression of protein or mRNA for PLA2G4A in association with alterations in PG production, non-specific PLA 2 inhibitors significantly diminished basal and agonist-induced PG production [6,9,17].
Mammalian PLA 2 s constitute a growing family of enzymes, currently containing about 20 members, that catalyze hydrolysis of fatty acids from the sn-2 position of phospholipids with the concomitant formation of lysophospholipid [18]. Based largely on structural and enzymatic properties, PLA 2 s have been grouped into four main subfamilies: low molecular weight secretory PLA 2 s (PLA2 Groups 1-3, 5, 10 and 12); cytosolic PLA2G4; calcium independent PLA2G6; and platelet activating factor (PAF) acid hydrolases (PLA2G7 and PLA2G8). The intracellular PLA2G4 and PLA2G6 enzymes are of particular interest in the regulation of uterine PG biosynthesis because their sites of action are the perinuclear membranes where downstream AA metabolizing enzymes reside.
Recently, we characterized PLA2G4A, PLA2G4C and PLA2G6 activity and expression in early passage bovine luminal endometrial epithelial cells [19]. It was observed that oxytocin stimulation of PGF 2α production was associated with increased expression and activity of PLA2G6 and IFNT diminished these responses. In addition, the PLA2G6 inhibitor, bromoenol lactone (BEL), abolished oxytocin stimulation of PGF 2α production. Results indicated that oxytocin stimulation of endometrial PGF 2α pro-duction is mediated, at least in part, through activation of PLA2G6.
Both BEND cells and primary cultures of endometrial epithelial cells have served as models for regulation of uterine prostaglandin biosynthesis and the mechanism by which IFNT modifies this process. Physiological differences between the two cell types have been identified [14]. The objective of the present study was to determine the role of PLA 2 enzymes in BEND cell PG production and relate these results to the reported differences between BEND cells and primary endometrial epithelial cells. Cell culture BEND cells were cultured and propagated by the method described by Staggs, et al. [13], with modest modifications. Briefly, cells were seeded into tissue culture dishes of the appropriate size for each experiment (see below) at a concentration of 0.5 × 10 5 cells per ml in culture medium (40% Ham's F-12, 40% MEM, 200 U insulin/L, 50 μg gentamicin, 10% FBS, 10% horse serum) at 37°C in a humidified atmosphere of 95% air and 5% CO 2 and culture medium was changed every other day. For all experiments, cells were grown to 80-90% confluence before application of treatments which were applied in triplicate and each experiment was repeated (n = 6) unless noted otherwise.

Western blot analysis
Western blot analysis was performed as described previously [20]. Briefly, cells were grown in 60 mm dishes and treated with vehicle (control), PDBu and IFNT (50 ng and 1000 ng), alone and in combination for 6 h as described above. Cells were lysed in RIPA buffer (50 mM Tris-HCl, 150 mM NaCl, 1% SDS, 0.5% sodium deoxycholate, 1 mM DTT, 100 μM PMSF, 1% NP-40), and 1 × protease inhibitor cocktail (Roche Applied Science, Indianapolis, IN). Lysates from triplicate treatment wells were combined, sonicated and centrifuged (10,000 g, 20 minutes) and supernatants were subjected to 10% SDS-polyacrylamide gel electrophoresis and transferred onto nitrocellulose membranes. Non-specific binding sites were blocked with 5% non-fat dry milk and membranes were incubated with antibodies to PLA2G4A, PLA2G4C, and PLA2G6 in 1:800, 1:400 and 1:1000 dilutions, respectively, overnight at 4°C. Membranes were rinsed 3 times and protein bands were visualized by enhanced chemiluminescence (Amersham Biosciences, Arlington Heights, IL). A blot was prepared from each of the replicated experiments.

H-arachidonic acid and H-linoleic acid released from intact cells
Intact cell PLA 2 assays were performed on cells prelabeled for 24 h with either 3 H-arachidonic acid ( 3 H-AA; 0.25 μCi/ml) or 3 H-linoleic acid ( 3 H-LA, 0.25 μCi/ml). Cells were washed three times with Hank Balanced Salt Solution (HBSS), equilibrated for 45 min and treatments applied for 1 hr. Cells were treated with 100 ng/ml PDBu in the presence and absence of 7.5 μM BEL or 0.4 μM PYR-1. Inhibitors were added 10 minutes prior to treatment with PDBU. Medium and lysed cells were harvested separately and % of total radioactivity released into the medium determined by scintillation counting.

PLA 2 activity in cellular homogenates
PLA 2 assays were performed as described previously [20]. Briefly, cells were washed and treated with either the PLA2G4A inhibitor, PYR-1 (0.4 μM) or the PLA2G6 inhibitor BEL (7.5 μM) for 3 hours. At the end of the incubation period, cells were washed with Ca ++ -free PBS containing 5 mM EGTA and 1 mM PMSF, lysed in homogenizing buffer (50 mM Tris HCl, pH 7.4, 0.5 mM dithiothreitol, 20% glycerol, 1 μg/ml leupeptin, 10 μg/ml aprotinin, 1 mM phenylmethylsulfonylchloride) and sonicated two times for 10 seconds on ice. The substrates 1-palmitoyl-2-[arachidonoyl] phosphatidylcholine (cold AA-PC) and 1- phosphatidlycholine ( 14 C-LA-PC) were used to assay for PLA 2 activity. The substrates were dried under nitrogen and resuspended by sonication in assay buffer (10 mM Hepes, pH 7.5) to a final optimum concentration of cold to radiolabeled substrate as determined previously. Calcium-dependent activity (primarily PLA2G4A) assays were performed with AA-PC and 14 C-AA-PC in the presence of 5 mM CaCl 2 and calcium-independent activity (primarily PLA2G6 and PLA2G4C) assays were performed using the same substrates but in the absence of CaCl 2 and the presence of EGTA (5 mM). Assays optimized for identification of PLA2G6 utilized LA-PC and 14 C-LA-PC in both the presence and absence of Ca ++ and presence of 400 μM Triton-X and 0.8 mM ATP which enhance the activity of this enzyme [29]. Experiments were terminated by addition of chloroform-methanol, 2:1 (v/v), the chloroform layer was extracted and lipids separated by thinlayer chromatography (hexane: diethyl ether: glacial acetic acid, 7:3:0.2). Lipids were visualized with I 2 vapor, the zones corresponding to fatty acid and phospholipids were excised and radioactivity determined by scintillation counting.

Statistical analysis
Data for PG production and PLA 2 activity are presented as means +/-SEM and subjected to least squares analysis of variance using the general linear models procedure of the Statistical Analysis System (SAS Institute, Cary, NC). Sources of variance included experiments, treatments and their interactions. Individual comparisons of means were made using Student-Neuman-Keuls test in which the independent variables were the treatments and the dependent variables were the levels of PG or PLA 2 activity produced. Differences were considered statistically significant when p < 0.05.

Prostaglandin production
Production of PGE 2 and PGF 2α are illustrated in Figures 1  and 2, respectively. BEND cells produced about 10-fold more PGE 2 than PGF 2α under resting conditions. Production of both PGs increased significantly (10-to 20-fold) in Prostaglandin E 2 production by BEND cells  response to PDBu stimulation, which had a greater effect on PGE 2 production. PYR-1, the PLA2G4A inhibitor, significantly diminished production of both PGs by resting cells and abolished the stimulatory effect of PDBu. Interestingly, BEL, the PLA2G6 inhibitor, stimulated production of both PGs in control cells and PGF 2α release in PDBu-treated cells. IFNT reduced both PGE 2 and PGF 2α production from resting cells and diminished PDBu stimulation of PG production. Conversely, IFNT did not significantly reduce BEL stimulation of PG production.

PLA 2 protein expression
Western blot analysis of cellular proteins from BEND cells demonstrated that the PLA2G4A antibody recognized a protein that migrated at ~110 kD and a very minor protein of slightly less mass (~107 kD) that may have been a breakdown product (Figure 3, top). The PLA2G6 antibody cross-reacted with a protein that migrated at ~85 kD (

H-arachidonic acid release and PLA 2 activity assays
Release of incorporated 3 H-arachidonic acid and 3 H-linoleic acid were used as measures of PLA 2 activity in intact cells because PLA2G4A is selective for arachidonic acid, whereas PLA2G6 is non-selective and will cleave both arachidonic acid and linoleic acid from membrane phospholipids. As can be seen in Figure 4, PDBu induced significant release of 3 H-arachidonic acid when compared to DMSO control. This effect was attenuated by the Group IVA PLA 2 inhibitor PYR-1, but not by the Group VI inhibitor, BEL. In contrast, PYR-1 had no effect on PDBu stimulation of 3 H-linoleic acid release ( Figure 5) but BEL inhibited this action.
As can be seen in Figure 6, BEND cells exhibited significant PLA 2 activity from cell lysates when 14 C-PC-AA was used as a substrate both in the presence or absence of calcium; however, PLA 2 activity was significantly greater in the presence of calcium than in all other groups. The PLA2G4A inhibitor, PYR-1, returned values to baseline in both the presence or absence of calcium.
Cellular lysates also exhibited significant PLA 2 activity when 14 C-PC-LA was used as a substrate and this activity was not different in the presence or absence of calcium ( Figure 7). The PLA2G6 inhibitor, BEL, attenuated activity in the presence of calcium and in the presence of EGTA.
BEND cell expression of PLA2G4A and PLA2G6

Discussion
The results from this investigation strongly indicate that PLA2G4A is the enzyme that liberates arachidonic acid for PG biosynthesis in BEND cells. In addition, the results suggest that stimulation of PG production is regulated at the level of PLA2G4A activity and expression. Several lines of evidence support the concept that PG production is regulated by PLA2G4A in BEND cells. One, the PLA2G4A inhibitor PYR-1, significantly diminished basal PG production and abolished PDBu stimulation of PG production and 3 H-AA release from intact cells. Two, stimulation of PG production by PDBu was associated with increased expression and activity of PLA2G4A. Three, the predominant PLA 2 activity in BEND cell lysates was calciumdependent. Also, BEND cells demonstrated significantly more activity when 14 C-PC-AA was used as a substrate when compared to 14 C-PC-LA, suggesting that the predominant isoform active in these cells is arachidonoylselective, a characteristic of PLA2G4A. Finally, reduction of PG production by IFNT was associated with diminished PLA2G4A expression.
In addition to the demonstrated PLA2G4A activity, BEND cells exhibited considerable PLA2G6 activity. The use of 14 C-PC-LA as a substrate, showed activity was not different in the presence or absence of calcium, consistent with calcium-independent activity. This activity was inhibited by BEL, suggesting activity of a PLA2G6 isoform. Similarly, BEL, but not PYR-1, inhibited incorporated 3 H-linoleic acid release from intact cells. However, BEL was not inhib-Linoleic acid release by BEND cells  PLA 2 activity assays using AA-PC as substrate Figure 6 PLA 2 activity assays using AA-PC as substrate. Ca ++dependent PLA 2 activity assays were performed on cellular lysates with AA-PC and 14 C-AA-PC in the presence of 5 mM CaCl 2 and Ca ++ -independent assays were performed using the same substrate but in the absence of Ca ++ and the presence of 5 mM EGTA. Columns with different superscripts are significantly different (p < 0.05). The PLA2G4A inhibitor, pyrroline-1 (PYR-1) inhibited Ca ++ -dependent activity. itory to BEND cell PG synthesis, an indication that PLA2G6 activation does not promote PG production in these cells. PLA2G6 has been shown to be involved in maintaining membrane homeostasis in some cell types [18] and it is feasible that it plays a similar role in BEND cells. Interestingly, inhibition of PLA2G6 activity with BEL resulted in increased production of both PGF2α and PGE2 and BEL was synergistic with PDBu in promoting production of PGF2α. In previous studies with primary or early passage luminal endometrial cells, we observed that BEL stimulated PGE2 production and inhibited PGF2α production [19]. Whereas BEND cell PGF2α and PGE2 production appears to be mediated by PLA2G4A, oxytocin-stimulated luminal endometrial cell PGF2α production was associated with activation of PLA2G6. In BEND cells, inhibition of PLA2G6 activity by BEL may result in greater concentrations of phospholipid substrate available for hydrolysis by PLA2G4A resulting in increased production of both PGF2α and PGE2.
Several studies have used BEND cells as a model system for investigation of signaling mechanisms involved in uterine cell PG production and IFNT inhibition of this process [reviewed in [12]]. Base on results from these studies, it was concluded that phorbol ester-stimulation of PG production and reduction of this effect, by IFNT, was mediated by positive and negative regulation of COX-2 expression. For example, Binelli et al. [16] reported that activation of the protein kinase C (PKC) pathway with PDBu increased COX-2 expression and PGF 2α production. Pru et al. [23] observed that PDBu treatment stimulated COX-2 mRNA expression within 30-60 minutes, while COX-2 protein and PG production increased at 3 hr. Cotreatment of cells with IFNT and PDBu results in diminished COX-2 mRNA, COX-2 protein and PG production. One study [16] observed that PDBu tended to increase (p < 1.0) PLA 2 protein expression and IFNT tended to diminish this response, but only after 12 hours of culture, but PLA 2 activity was not evaluated. In the present study, we observed an increase in PLA 2 at 6 hours, the only time point tested, and maximal rate of increase in PLA 2 activity at 3 hours. In preliminary studies, an hourly time course of PLA 2 activity was performed (not shown) and significant increases were observed in the first hour reaching a maximum rate of increase at 3 hours. These results indicate that increases in PLA 2 activity in response to PDBu is an early response, while the increase in PLA 2 expression is a delayed response. Based on these, and previous results [19], it is suggested that uterine PG production is regulated, at least in part, at the level of PLA 2 activity and expression.
Stimulation of BEND cells with PDBu results in the phosphorylation and activation of several components of mitogen activated protein kinase (MAPK) pathways [23,24]. It is well established in a variety of cell types that MAPKs phosphorylate and activate PLAG4A resulting in arachidonic acid release [2]. Arachidonic acid, in addition to being the major prostaglandin precursor, can act as a signaling molecule and affects intracellular Ca ++ concentrations [25]. Importantly, arachidonic acid up-regulates COX-2 expression and PG production in a variety of cell types [26,27]. Parent, et al. [28], demonstrated that exogenous arachidonic acid stimulates COX-2 expression followed by a 30-fold increase in PG production by bovine endometrial epithelial cells. In BEND cells, arachidonic acid enhanced PGF 2α production in PDBu-stimulated cells [29]. It is suggested that PDBu-activation of the PKC/ MAPK pathway results in increased activity and expression of PLA2G4A which liberates arachidonic acid from perinuclear membranes. Endogenous arachidonic acid then may serve both as substrate for, and inducer of, COX-2 which completes the committed step in PG biosynthesis.
Our results demonstrating phorbol ester stimulation of PLA2G4A activity and expression do not preclude the possibility of similar actions on COX-2 expression. Hughes-Fulford et al. [27] suggested that activation of PLA2G4A expression may involve signaling mechanisms similar to those of COX-2, since the genes have several identical promoter elements and the genes are on the same region of the same chromosome, in mice and humans. The genes for PLA2G4 and COX-2 are located in the same region of chromosome 16 in cattle [30] and review of the promoter region revealed several promoter elements in common. This raises the possibility that the two genes may be coregulated, a phenomenon observed by others [31].
The mechanism by which IFNT exerts negative regulation on PG production in BEND cells is unclear. Guzeloglu, et al, [24] investigated effects of IFNT on COX-2 mRNA stability and concluded that IFNT may diminish PDBu-stimulated PG production by accelerating COX-2 degradation, mediated through an unidentified transcription dependent mechanism. Thatcher, et al, [12] suggested that IFNT, acting through Type I IFN receptors, activates the Janus kinase (JAK)-signal transducer and activator of transcription (STAT) pathway in a manner that directly represses COX-2 expression. Our results indicate that IFNT diminishes PLA2G4A expression, which would, in turn, reduce arachidonic acid release potentially resulting in diminished COX-2 expression and PG production. As indicated above, COX-2 and PLA2G4A gene expression may be coregulated since the promoter region of each share common elements, including IFN response elements [12,32].
Results from the present study with BEND cells differ from results of our previous study with bovine luminal endometrial epithelial cells [19]. Similar to studies by others [14] BEND cells produced 10-20 fold more PGE 2 than PGF 2α , whereas luminal endometrial epithelial cells pro-duce greater quantities of the latter. PDBu stimulation of BEND cell PG production did not affect PLA2G6 expression or activity and the PLA2G6 inhibitor, bromoenol lactone, did not diminish BEND cell PG production. Conversely, stimulation of PGF 2α production with oxytocin in luminal endometrial epithelial cells increased PLA2G6 expression and activity and this response was attenuated by bromoenol lactone. In BEND cells, the PLA2G4A inhibitor, PYR-1, diminished PG production in unstimulated cells and abolished PDBu stimulation of PG production whereas it had little effect on luminal endometrial epithelial cell PG production (Ochs, Roberts, Elgayyer, Godkin and Tithof, unpublished observation).
Together, these results indicate that PG production by the two cell types may be regulated by different PLA 2 isotypes. Another possible explanation for the discordant results may be that stimulation with a phorbol ester may result in activation of signal transduction pathways different from those activated by oxytocin.

Conclusion
PGE2 and PGF2-alpha production by BEND cells is mediated by the activity and expression of PLA2G4A. Interferontau treatment diminished expression of PLA2G4A and PG production. BEND cells were shown to express PLA2G6 but, unlike primary or early passage luminal bovine endometrial cells, stimulation of PLA2G6 activity was not associated with increased PG production. Identification of PLA 2 enzymes as key regulators of PG production may lead to novel methods of regulating prostanoid production.