Intraluteal Prostaglandin Biosynthesis and Signaling Are Selectively ...

2 downloads 0 Views 4MB Size Report
Jun 27, 2012 - 2Correspondence: Joe A. Arosh, Department of Veterinary Integrative ...... Wiepz GJ, Wiltbank MC, Nett TM, Niswender GD, Sawyer HR.
BIOLOGY OF REPRODUCTION (2012) 87(4):97, 1–14 Published online before print 27 June 2012. DOI 10.1095/biolreprod.112.100438

Intraluteal Prostaglandin Biosynthesis and Signaling Are Selectively Directed Towards PGF2alpha During Luteolysis but Towards PGE2 During the Establishment of Pregnancy in Sheep1 JeHoon Lee,3,4 John A. McCracken,5 Jone A. Stanley,3,4 Thamizh K. Nithy,3,4 Sakhila K. Banu,3,4 and Joe A. Arosh2,3,4 3

Reproductive Endocrinology and Cell Signaling Laboratory, College of Veterinary Medicine and Biomedical Sciences, Texas A&M University, College Station, Texas 4 Department of Veterinary Integrative Biosciences, College of Veterinary Medicine and Biomedical Sciences, Texas A&M University, College Station, Texas 5 Department of Animal Science, University of Connecticut, Storrs, Connecticut INTRODUCTION

In ruminants, endometrial prostalgandin (PG) F2alpha causes functional luteolysis, whereas luteal synthesis of PGF2alpha is required for structural luteolysis. PGE2 is considered to be a luteoprotective mediator. Molecular aspects of luteal PGF2alpha and PGE2 biosynthesis and signaling during the estrous cycle and establishment of pregnancy are largely unknown. The objectives of the present study were 1) to determine the regulation of proteins involved in PGF2alpha and PGE2 biosynthesis, catabolism, transport and signaling in the corpus luteum (CL); 2) to investigate the transport of interferon tau (IFNT), PGF2alpha, and PGE2 from the uterus to the ovary through the vascular uteroovarian plexus (UOP); and 3) to compare the intraluteal production of PGF2alpha and PGE2 on Days 12, 14, and 16 of the estrous cycle and pregnancy in sheep. Our results indicate that luteal PG biosynthesis is selectively directed towards PGF2alpha at the time of luteolysis and towards PGE2 during the establishment of pregnancy. Moreover, the ability of the CL of early pregnancy to resist luteolysis is due to increased intraluteal biosynthesis of PGE2 and PGE2 receptor (PTGER) 2 (also known as EP2)- and PTGER4 (also known as EP4)-mediated signaling. We also found that IFNT protein is not transported through the UOP from the uterus to the ovary; in contrast, a large proportion of endometrial PGE2 is transported from the uterus to the ovary through the UOP. These results indicate that endometrial PGE2 stimulated by pregnancy is transported locally to the ovary, which increases luteal PGE2 biosynthesis and hence activates luteal PTGER2 and PTGER4 signaling, thus protecting the CL during the establishment of pregnancy in sheep.

Arachidonic acid (AA) is the primary precursor for the synthesis of prostaglandins (PGs). Cytosolic phospholipase A2 liberates AA from phospholipids. Prostaglandin-endoperoxide synthases 1 (PTGS-1) and 2 (PTGS-2) (also known as cyclooxygenases COX-1 and COX-2, respectively) convert AA into PGH2 [1]. Prostaglandin F synthases PTGFSAKR1B1 (earlier known as AKR1B5) and PTGFS-AKR1C3 convert PGH2 into PGF2a. Prostaglandin E synthases PTGES-1 (microsomal PGES-1), PTGES-2 (microsomal PGES-2), and PTGES-3 (cytosolic PGES) convert PGH2 into PGE2. Prostaglandin dehydrogenase (PGDH) catabolizes PGF2a and PGE2 into inactive metabolites 13,14-dihydro-15-keto-prostaglandin F2a (PGFM) and 13,14-dihydro-15-keto prostaglandin E2 (PGEM), respectively [2]. PGF2a and PGE2 are transported competitively across cell membranes by solute carrier organic anion transporter family, member 2A1 (SLCO2A1; also known as prostaglandin transporter [PGT]) [3]. PGF2a and PGE2 exert their biological effects via seven-transmembrane G proteincoupled receptors [4]. PGF2a acts through PGF2a receptor (PTGFR, also known as FP), and activation of PTGFR in turn activates protein kinase C (PKC) and Ca2 þ cell signaling pathways [4]. Multifaceted effects of PGE2 are meditated through PGE2 receptor (PTGER) 1, PTGER2, PTGER3, and PTGER4 (also known as EP1, EP2, EP3, and EP4, respectively) by integrating multiple cell signaling pathways [5, 6]. PTGER1 activates PKC and Ca2 þ pathways. PTGER2 and PTGER4 activate the protein kinase A (PKA) pathway. Activation of PTGER3A through PTGER3D produces a wide range of complex and opposite actions [4]. Recent studies indicate that PGE2 transactivates extracellular signal-regulated kinases (ERK1/2), AKT (also known as protein kinase B; PKB), nuclear factor-kappa B (NFjB), and b-catenin (CTNNB1) pathways through EP2 and EP4 in cancer [5, 6] and in endometriosis [7]. In ruminants, PGF2a is the luteolytic hormone, whereas PGE2 is considered to be a luteoprotective mediator [8–10]. At the time of natural luteolysis in ruminants, PGF2a is released in a pulsatile manner from the endometrium. A minimum of five 1-h pulses of PGF2a over a period of 48 h is required to cause complete functional (decline in progesterone) as well as structural (loss of luteal cells) luteolysis consistently in sheep [11]. Systemic administration of PGF2a during the midluteal phase of the estrous cycle in sheep increases luteal PGF2a production, which is inhibited by pretreatment with indomethacin, as measured in corpus luteum (CL) explant culture [12]. In sheep, administration of a luteolytic dose of PGF2a in vivo and

corpus luteum, establishment of pregnancy, estrous cycle, intraluteal prostaglandins, luteolysis, prostaglandins, ruminants, uterus 1

Supported by an Agriculture and Food Research Initiative Competitive Grant 2008-35203-19101 from the USDA National Institute of Food and Agriculture to J.A.A. and in part by USDA grant 2004-3520-14176 to J.A.M. 2 Correspondence: Joe A. Arosh, Department of Veterinary Integrative Biosciences, College of Veterinary Medicine & Biomedical Sciences, Mail Stop: TAMU 4458, Texas A&M University, College Station, TX 77843. E-mail: [email protected] Received: 7 March 2012. First decision: 2 April 2012. Accepted: 21 June 2012. Ó 2012 by the Society for the Study of Reproduction, Inc. eISSN: 1529-7268 http://www.biolreprod.org ISSN: 0006-3363

1

Article 97

Downloaded from www.biolreprod.org.

ABSTRACT

LEE ET AL.

treatment of luteal cells with PGF2a in vitro induces expression of PTGS-2 mRNA in luteal tissues/cells [13–15]. These findings suggest that endometrial PGF2a causes functional luteolysis and that luteal synthesis of PGF2a likely is an important component of a positive-feedback loop between the uterus and the CL during the process of structural luteolysis [16]. In ruminants, the CL of early pregnancy is more resistant to the luteolytic action of PGF2a [17–21] on Days 12–16, and the resistance is even greater when multiple embryos are present [18]. Injection of PGF2a into an ovarian artery or follicles of early pregnant sheep caused luteolysis in 28% and 17% of animals, compared with 78% and 83% in nonpregnant sheep, respectively [17, 19]. Exogenous estradiol at doses causing premature luteolysis in cyclic sheep is less effective in pregnant sheep [22]. Intraovarian administration of PGE2 counteracts the luteolytic actions of PGF2a [23]. PGE2 is secreted by both the conceptus [24–26] and the endometrium [26–28] from pregnant ewes in vitro. Intrauterine or intraovarian infusions of PGE2 in nonpregnant ewes extends the interestrous interval and reduces luteal sensitivity to both endogenously secreted and exogenously administered PGF2a [19, 23, 29–31]. Overall, the above-mentioned studies provide compelling evidence that, as a component of antiluteolytic mechanisms, PGE2 produced in the CL may counteract the luteolytic effect of both exogenous and endogenous PGF2a during early pregnancy in ruminants. However, the underlying molecular and cellular mechanisms are largely unknown. During the establishment of pregnancy, interferon tau (IFNT) secreted by the trophoblast of the conceptus inhibits endometrial pulsatile release of PGF2a and prevents luteolysis [32]. Intrauterine infusions of exogenous IFNT in cyclic sheep inhibits endometrial pulses of PGF2a by suppressing receptors for estrogen (ESR-1) and oxytocin (OXTR) [32]. Infusions of IFNT directly into the uterine vein maintained a functional CL in 80% of sheep for up to 32 days through as-yet-unidentified mechanisms [33, 34]. Experiments involving anastomosis of the uterine vein or the ovarian artery (OA) from the pregnant to the nonpregnant uterine horn indicated that both luteolytic and luteoprotective mediators need to be transported locally from the utero-ovarian vein (UOV) to the OA via the utero-ovarian plexus (UOP) in sheep and cattle [35–37]. Moreover, embryo/ conceptus transfer and hysterectomy experiments indicated that the luteolytic and luteoprotective mechanisms are locally mediated between the uterus and the CL of the ipsilateral ovary and do not act systemically in sheep [38–40]. Early studies demonstrated that during the establishment of pregnancy in sheep, one or more factors from the conceptus or gravid uterus reach the ovary locally through the UOP and protect the CL from luteolysis [10, 21, 35, 36, 39–42]. However, such a factor has not been positively identified. The objectives of the present study were 1) to determine the regulation of proteins involved in PGF2a and PGE2 biosynthesis, catabolism, transport and signaling in the CL; 2) to investigate the transport of IFNT, PGF2a, and PGE2 from the uterus to the ovary through the UOP; and 3) to compare the intraluteal production of PGF2a and PGE2 on Days 12, 14, and 16 of the estrous cycle and pregnancy in sheep.

tra-acetic acid (EDTA; Invitrogen Life Technologies, Inc.); rabbit polyclonal anti-human PTGS-1, PTGS-2, PTGES-1, PTGES-2, PTGES-3, PGDH, PTGFR, PTGER1, PTGER2, PTGER 3, PTGER4, and SLCO2A1 antibodies and PGF2a, PGE2, PGFM, and PGEM enzyme-linked immunosorbent assay (ELISA) kits (Cayman Chemicals); rabbit polyclonal anti-human PTGFSAKR1C3 and mouse monoclonal anti-human b-actin monoclonal antibodies, rabbit immunoglobulin (Ig) G or serum, PGH2, and Dulbecco modified Eagle medium/Ham F12 (DMEM/F12; Sigma-Aldrich); PTGFS-AKR1B1 (Abcam, Inc.); goat anti-rabbit or anti-mouse IgG conjugated with horse radish peroxidase antibody (Kirkegaard & Perry Laboratories); Vectastain Elite ABC kit (Vector Laboratories, Inc.); Blue X-Ray film (Phoenix Research Products); fetal bovine serum (Hyclone); tissue-culture dishes and plates (Corning, Inc.); intravenous Teflon catheters (Beckton Dickinson Infusion Therapy System, Inc.); heparin sodium (Abraxis Pharmaceutical Products); and blood collection tubes treated with EDTA (10.8 mg; BD Biosciences). ABTS peroxidase substrate (2, 2-azino-di-3-ethylbenzhiazoline-sulphonate crystallized ammonium salt, catalog no. 50-66-18) was purchased from KPL, Inc. Rabbit polyclonal anti-sheep IFNT antibody was a generous gift from Dr. Fuller Bazer (Texas A&M University). The other chemicals used were molecular biological grade from Fisher or Sigma-Aldrich.

MATERIALS AND METHODS

Protein Extraction

Materials

Total protein was isolated from CL tissues as described previously [46, 47]. Briefly, tissues were homogenized in buffer (50 mM Tris [pH 8.0], 10 mM EDTA, 1 mM diethyldithiocarbamic acid-[DE-DTC], and 0.1% Tween-20) and centrifuged at 30 000 3 g for 1 h at 48C. The homogenized tissue pellets were sonicated in sonication buffer (20 mM Tris [pH 8.0], 0.5 mM EDTA, 0.1 mM DE-DTC, 1 mM phenylmethylsulfonyl fluoride, and protease inhibitor cocktail tablets [complete EDTA-free, 1 tablet/50 ml; PhosStop, 1 tablet /10 ml], and

Animal Husbandry and Surgery

The reagents used in the present study were purchased from the following suppliers: prestained protein markers and Bio-Rad assay reagents and standards (Bio-Rad Laboratories); protran BA83 Nitrocellulose membrane (Whatman, Inc.); Pierce ECL Western blotting substrate (Pierce); protease inhibitor (Roche Applied Biosciences); antibiotic-antimycotic and Trypsin-ethylenediaminete-

2

Article 97

Downloaded from www.biolreprod.org.

Matured crossbred Suffolk ewes (Ovis aries) were observed daily for estrus in the presence of vasectomized rams. Ewes that had exhibited at least two estrous cycles of normal duration (17–18 days) were used in the present study. All experiments and surgical procedures were in accordance with the Guide for Care and Use of Agricultural Animals and approved by Texas A&M University’s Laboratory Animal Care and Use Committee. At estrus (Day 0), the ewes were bred to either an intact or a vasectomized ram. The ewes (n ¼ 4 per day) were randomly ovariohysterectomized on Days 12, 14, or 16 of the estrous cycle and on Days 12, 14, or 16 of pregnancy as described previously [43, 44]. On the day of surgery, a jugular vein was catheterized using a BD Angiocath 16 gauge x 3.25-inch intravenous Teflon catheter. Anesthesia was induced using diazepam (0.2 mg/kg i.v.; Abbott Laboratories) and ketamine (4 mg/kg i.v.; Ketaset; Fort Dodge). The ewes were intubated endotracheally and ventilated spontaneously. Anesthesia was maintained using isoflurane (0.5%–2.5%; IsoFlo; Abbott Laboratories) and oxygen as described previously [45]. A ventral midline laparotomy was performed, and the entire reproductive tract was exteriorized as described previously [45]. Blood samples (5 ml) were collected from the UOV and ovarian vein (OV) ipsilateral to the CL using a 20 gauge butterfly needle. The OA was isolated from the ovarian pedicle by blunt dissection. Each ewe was given heparin (10 000 IU/ewe i.v.) 10 min before the OA was sectioned and blood (5 ml) was collected. Plasma was harvested immediately from all blood samples, and indomethacin (100 lM) was added to prevent in vitro production of PGs. Ringer lactate was administered at the rate of 5 ml kg1 h1to replace fluid losses and to maintain systemic blood pressure. A heating pad was placed underneath the animal and used to maintain body temperature, which was monitored on a continuous basis. At the end of each experiment, the ewes were euthanized using Beuthanasia (MWI Veterinary Supply, Boise, ID). Ovariohysterectomy was then performed. The uterus was flushed with 20 ml of physiological saline, then centrifuged at 2000 rpm for 10 min, after which the supernatant (uterine flushing [UF]) was collected. Indomethacin (100 lM) was then added to prevent in vitro production of PGs, and the UF samples were stored at 208C until analysis. Pregnancy was confirmed on each day by the presence of a normal conceptus in the uterine lumen flushing as described previously [44]. The uterus was separated from the surrounding tissues and vasculature, and sections (thickness, ;0.5 cm) from the midportion of each uterine horn were fixed in fresh 4% buffered paraformaldehyde and processed for immunohistochemistry using standard procedures. The ovaries were collected, and the CLs were isolated. Two sections from the midportion of each CL were collected: One section was processed for explant culture as described below, and the other section was fixed in fresh 4% buffered paraformaldehyde and processed for immunohistochemistry using standard procedures. The remaining CL tissue was cut into small pieces, snap-frozen in liquid nitrogen, and stored at 808C for protein extraction.

LUTEAL PGF2a AND PGE2 BIOSYNTHESIS AND SIGNALING each assay and compared between assays. For the PGF2a assay, the sensitivity or minimal detection limit was 3.9 pg/ml; intra-assay CV was 9.4% and interassay CV 12.5%. For the PGE2 assay, the sensitivity or minimal detection limit was 7.8 pg/ml; intra-assay CV was 4.2% and interassay CV 12.4%. For the PGFM assay, the sensitivity or minimal detection limit was 23 pg/ml; intraassay CV was 6.8% and interassay CV 8.8%. For the PGEM assay, the sensitivity or minimal detection limit was 0.39 pg/ml; intra-assay CV was 5.4% and interassay CV 7.2%.

1.0% Tween-20) using a Microson ultrasonic cell disruptor (Microsonix Incorporated), then centrifuged at 15 000 3 g for 15 min at 48C, after which the supernatants (total protein) were stored at 808C until analyzed. Total protein concentrations were determined using the Bradford method [48] and a Bio-Rad Protein Assay kit.

Western Blot Analysis Total protein samples (75 lg) were resolved using 7.5%, 10%, or 12.5% SDS-PAGE, and Western blot analysis was performed as described previously [46, 47]. The blots were incubated with one of the following primary antibody for overnight at 48C: PTGS-1, PTGS-2, PTGES-1, PTGES-2, PTGES-3, AKR1B1-PGFS, AKR1C3-PGFS, PGDH, PTGFR, PTGER1, PTGER2, PTGER4, or SLCO2A1 antibody was used at 1:250 dilution, and b-actin was used at 1:8000 dilution. Then, the blots were washed and incubated with goat anti-rabbit or anti-mouse IgG conjugated with horse radish peroxidase secondary antibody at 1:10 000 dilution for 1 h at room temperature. Chemiluminescent substrate was applied according to the manufacturer’s instructions (Pierce Biotechnology). The blots were exposed to Blue X-Ray film, and densitometry of autoradiograms was performed using an Alpha Imager (Alpha Innotech Corporation).

For Western blot analysis, UF (2 ll), plasma (2 ll) from UOV or OA, and protein (75 lg) from CL tissue lysates were resolved in 12.5% SDS-PAGE, and Western blot analysis was performed as described previously [46, 47]. The blots were incubated with rabbit polyclonal anti-sheep IFNT antibody (1:5000 dilution) overnight at 48C. The blots were then washed and incubated with goat anti-rabbit IgG conjugated with horse radish peroxidase secondary antibody at 1:10 000 dilution for 1 h at room temperature. Chemiluminescent substrate was applied according to the manufacturer’s instructions. The blots were exposed to Blue X-Ray film, and densitometry of autoradiograms was performed using an Alpha Imager.

Immunohistochemistry

IFNT Assay by Indirect ELISA

Paraffin sections (thickness, 5 lm) were used for immunohistochemical localization of the proteins using a Vectastain Elite ABC kit according to the manufacturer’s protocols and as described previously [7, 46, 49]. Endogenous peroxidase activity was removed by fixing sections in 0.3% hydrogen peroxide in methanol. Tissue sections were blocked in 10% goat serum for 1 h at room temperature. Incubation with the primary antibodies for PTGS-2, PTGES-1, PTGES-3, PTGFS-AKR1B1, PTGFS-AKR1C3, PGDH, PTGFR, PTGER1, PTGER2, PTGER4, or SLCO2A1 (1:125 dilution) was performed overnight at 48C. The tissue sections were further incubated with the secondary antibody (goat anti-rabbit IgG biotinylated) for 45 min at room temperature. For the negative control, serum or IgG from respective species with reference to the primary antibody at the respective dilution was used. Digital images were captured using a Zeiss Axioplan 2 Research Microscope (Carl Zeiss) with an Axiocam HR digital color camera. The intensity of staining for each protein was quantified using Image-ProPlus 6.3 image processing and analysis software according to the manufacturer’s instructions (Media Cybernetics, Inc.). The detailed methods for quantification are given in the instruction guide, ‘‘The Image-Pro Plus: The proven solution for image analysis.’’ In brief, six images of at 4003 magnification were captured randomly without hot-spot bias in each tissue section per animal. Integrated optical density of immunostaining was quantified under RGB mode as published recently [50]. Numerical data were expressed as the least square mean 6 SEM. This technique is more quantitative than conventional blind scoring systems, and the validity of the quantification was reported previously by our group [7, 46, 49, 50].

The assay was performed as described previously [53] with little modification. The UF sample was diluted 1:50, and plasma samples from UOV and OA were diluted 1:1 in EIA buffer (1 M phosphate solution containing 1% bovine serum albumin, 4 M sodium chloride, 10 mM EDTA, and 0.1% sodium azide). Standard curves were set up by diluting recombinant ovine IFNT at concentrations from 0 to 500 ng/ml in EIA buffer. On Day 1, the 96-well plate was coated with antigen (50 ll of diluted UF or plasma samples) and incubated at 48C overnight. On Day 2, the unbound antigen in each well was removed, and the wells were washed three times with 200 ll of washing buffer (4 M phosphate solution with 0.05% Tween-20, overall buffer pH 7.4). The remaining protein-binding sites in the coated wells were blocked by 100 ll of blocking buffer (5% nonfat dried milk in Tris-buffered saline with 0.1% Tween-20) and incubated for 2 h at room temperature. The wells were then washed three times with 200 ll of washing buffer. Rabbit anti-sheep polyclonal antibody (100 ll) at 1:5000 dilution in 2% milk was added to each well and incubated overnight at 48C. On Day 3, the wells were washed four times (1 min for each washing) with 200 ll of washing buffer. Horseradish peroxidaseconjugated goat anti-rabbit IgG at 1:2000 in 2% milk was then added to each well and incubated for 2 h at room temperature. The wells were then washed four times (1 min for each washing) with 200 ll of washing buffer. Then, 100 ll of ABTS peroxidase substrate was added, the plate was incubated at room temperature for 1 h and read using an ELISA plate reader (Synergy 4; Bio Tek Instruments, Inc.) at a single wavelength at 405 nm.

CL Explant Culture

Statistical analyses were performed using general linear models of the Statistical Analysis System (SAS Institute, Inc.). Data were checked for normality or homogeneity of variance before analyzing the data statistically. Day (12, 14, or 16) and status (estrous cycle vs. pregnancy) interactions on expression of various proteins were tested using repeated measures for multivariate ANOVA. Comparison of means was tested by Wilks lambda or orthogonal contrast tests. Effects of Day 16 of the estrous cycle or pregnancy on cell-specific expression of various proteins in luteal cells were analyzed using one-way ANOVA. Comparison of means was tested by Tukey honestly significant difference test. Effects of day of estrous cycle or pregnancy on PGF2a and PGE2 in UF, UOV, and OA and on PGF2a, PGFM, PGE2, and PGEM in OV and on PGF2a and PGE2 levels at various time points in CL explant culture were analyzed using repeated measures for split-plot ANOVA. Numerical data are presented as least squares means 6 SEM. Statistical significance was considered to be P , 0.05. The statistical model accounted for sources of variation, including treatments, replicates, and ewes as appropriate.

IFNT Protein Measurement by Western Blot Analysis

3

The CL tissue (;100 mg) was cut into 1-mm fragments and cultured in 1 ml of plain DMEM/F12 in 5% CO2 and 95% air at 378C in a 24-well tissueculture plate. At 0 min, PGH2 at 0, 25, 50, 75, 100, or 200 lg/ml was added. At 90 min, the culture medium was collected, and concentrations of PGF2a and PGE2 were measured by ELISA as described below.

Intraluteal PGF2a and PGE2 Extraction Intracellular PGs from luteal tissues were extracted as described previously [51] with little modifications. Fresh samples of luteal tissues (100 mg) were homogenized in ice-cold Tris-buffer (pH 7.4) containing indomethacin (100 lm) using a polytron homogenizer on ice. The homogenate was centrifuged at 50 000 rpm for 1 h at 48C, after which the supernatant (intraluteal lysate) harvested and the concentrations of PGF2a and PGE2 in the lysates were assayed by ELISA as described below.

RESULTS

ELISA of PGF2a, PGE2, PGFM, and PGEM

Expression and Regulation of PGF2a and PGE2 Metabolic Enzymes in the CL

The concentrations of PGF2a, PGE2, PGFM, or PGEM were measured in the plasma of UF, UOV, and OA as well as in luteal tissue extract and CL explant culture medium using commercially available kits (Cayman Chemicals) according to the manufacturer’s instructions and as described previously [43, 49, 52]. Intra- and interassay coefficients of variation (CVs) were determined at multiple time points on the standard curve as described by the manufacturer in

We determined temporal regulation of proteins involved in PGF2a and PGE2 biosynthesis and catabolism in the CL on Days 12, 14, and 16 of the estrous cycle (CY) and pregnancy (PX) (Fig. 3

Article 97

Downloaded from www.biolreprod.org.

Statistical Analyses

LEE ET AL.

Downloaded from www.biolreprod.org. FIG. 1. Temporal expression of PGF2a and PGE2 metabolic enzymes in the ovine CL on Days 12, 14, and 16 of the CY and PX (n ¼ 4 for each day). A) Western blot analysis. b-Actin protein (ACTB) was measured as an internal control. B–J) Densitometry was performed using an Alpha Imager, with values expressed as integrated optical density. The numerical data are expressed as the least square mean 6 SEM. Statistical significance (*P , 0.05) was determined for the difference in the expression levels of a given protein between the CY versus PX on a particular day or days. See Materials and Methods and Results for more details.

CY and PX. Importantly, its expression was decreased (P , 0.05) on Day 16 of the CY, but it was sustained on Day 16 of PX (Fig. 1D). PTGES-2 protein was expressed constantly on Days 12–16 of the CY and PX (Fig. 1E). Expression of PTGES-3 protein was

1). Results indicated that PTGS-1 protein was barely detectable on Days 12–16 of the CY or PX (Fig. 1B). PTGS-2 protein was expressed but was not regulated on Days 12–16 of the CY or PX (Fig. 1C). PTGES-1 protein was expressed on Days 12–16 of the 4

Article 97

LUTEAL PGF2a AND PGE2 BIOSYNTHESIS AND SIGNALING

undetectable on Days 12–16 of the CY, but it was highly (P , 0.05) expressed on Days 12–16 of PX (Fig. 1F). PTGFSAKR1B1 protein was abundantly expressed on Days 12–16 of the CY and PX; importantly, its expressions was increased (P , 0.05) on Day 16 of the CY compared to that of PX (Fig. 1G). PTGFS-AKR1C3 protein was expressed without modulation on Days 12–16 of the CY and PX (Fig. 1H). PGDH protein was not expressed or not delectable on Days 12–16 of the CY; by contrast, its expression was induced (P , 0.05) on Days 12–16 of PX (Fig. 1I). Our results further indicated that ratio between AKR1B1PTGFS:PTGES-1 increased (P , 0.05) on Days 14–16 of the CY, whereas ratio between PTGES-1 and AKR1B1-PTGFS increased (P , 0.05) on Days 14–16 of PX (Fig. 1J). Collectively, these results indicate that luteal PG biosynthetic machinery is selectively directed towards PGF2a at the time of luteolysis and towards PGE2 during the establishment of pregnancy in sheep.

PTGER4, and PGT SLCO2A1 proteins in the CL on Days 12, 14, and 16 of the CY and PX (Fig. 2). Results indicated that expression of PTGFR protein was not regulated on Days 12–16 by the CY or PX (Fig. 2B). PTGER1 protein was abundantly expressed on Days 12–14 of the CY and PX, but its expression was decreased (P , 0.05) on Day 16 of PX compared that of the CY (Fig. 2C). PTGER2 protein was expressed on Days 12–16 of the CY and PX. Importantly, its expression was decreased (P , 0.05) on Day 16 of the CY, but it was sustained on Day 16 of PX (Fig. 2D). PTGER3 protein was expressed at a very low level on Days 12–16 of the CY, and its expression was marginally increased on Days 12 and 14 of PX (Fig. 2E). PTGER4 protein was not expressed or undetectable on Days 12–16 of the CY. In contrast, its expression was induced (P , 0.05) on Days 12– 16 of PX (Fig. 2F). SLCO2A1 protein was expressed but not regulated on Days 12–16 of the CY or PX (Fig. 2G). These results together indicate that luteal PTGER2- and PTGER4mediated PGE 2 signaling is activated at the time of establishment of pregnancy, whereas it is suppressed at the time of luteolysis in sheep.

Expression and Regulation of PGF2a and PGE2 Receptors in the CL We determined temporal regulation of PGF2a receptor PTGFR and PGE2 receptors PTGER1, PTGER2, PTGER3, 5

Article 97

Downloaded from www.biolreprod.org.

FIG. 2. Temporal expression of PGF2a and PGE2 receptors and transporter in the ovine CL on Days 12, 14, and 16 of the CY and PX (n ¼ 4 for each day). A) Western blot analysis. b-Actin protein (ACTB) was measured as an internal control. B–G) Densitometry was performed using an Alpha Imager, with values expressed as integrated optical density. The numerical data are expressed as the least square mean 6 SEM. *P , 0.05 for the difference in the expression levels of a given protein between the CY versus PX on a particular day or days. See Materials and Methods and Results for more details.

LEE ET AL.

Intraluteal Production of PGF2a and PGE2

Cellular Localization of PGF2a and PGE2 Metabolic Enzymes and Receptors in the CL

We determined luteal production and secretion of PGF2a and PGE2 and their metabolites PGFM and PGEM, respectively, into ovarian venous blood on Days 14 and 16 of the CY and PX. Results (Fig. 5) indicated that ovarian venous plasma concentrations of PGF2a were higher (P , 0.05) on Days 14 and 16 of the CY compared to that of PX. Importantly, concentrations of PGE2 were higher (P , 0.05) on Days 14 and 16 of PX compared to that of the CY. The concentrations of PGFM were higher (P , 0.05) on Days 14 and 16 of PX compared to that of the CY; however, such regulation was not found for PGEM. Our results further indicated that the PGF2a:PGE2 ratio was higher (P , 0.05) on Days 14 and 16 of the CY, whereas the PGE2:PGF2a ratio was higher (P , 0.05) on Days 14 and 16 of PX. In addition, we isolated intraluteal lysates and measured the concentrations of PGF2a and PGE2 on Day 16 of the CY and PX. Results (Fig. 6) indicated that intraluteal production of PGF2a was higher (P , 0.05) on Day 16 of the CY, whereas intraluteal PGE2 production was higher (P , 0.05) on Day 16

Based on the results obtained by Western blot analysis, we determined cellular localization of PGF2a and PGE2 biosynthetic, catabolic, and signaling proteins by immunohistochemistry in the CL on Days 16 of the CY and PX. Results (Figs. 3 and 4) indicated that PTGS-2, PTGES-1, PTGES-3, PTGFSAKR1B1, PTGFS-AKR1C3, PGDH, PTGFR, PTGER1, PTGER2, PTGER4, and SLCO2A1 proteins were abundantly expressed in the large luteal cells (LLCs) and expressed to some degree in small luteal cells (SLCs). Spatial expressions of PTGES-1, PTGES-3, PGDH, PTGER2, and PTGER4 proteins in LLCs and SLCs were increased (P , 0.05) on Days 16 of PX compared to that of the CY. By contrast, spatial expressions of PTGFS-AKR1B1 and PTGER1 proteins were decreased (P , 0.05) on Day 16 of PX compared to that of the CY. These results indicate that a shift in the expression of proteins involved in PGE2 and PGF2a biosynthesis, catabolism, and signaling occur in both LLCs and SLCs at the time of luteolysis and during the establishment of pregnancy in sheep. 6

Article 97

Downloaded from www.biolreprod.org.

FIG. 3. Cellular localization of PGF2a and PGE2 metabolic enzymes (1) and receptors and transporter (2) in the ovine CL on Day 16 of the CY and PX (n ¼ 4 for each day). Immunohistochemistry was performed using Vectastain Elite ABC kit, and representative photomicrographs are shown. Original magnification 3400. See Materials and Methods and Results for more details.

LUTEAL PGF2a AND PGE2 BIOSYNTHESIS AND SIGNALING

Downloaded from www.biolreprod.org. FIG. 4. Densitometry of relative spatial expression of proteins (shown in Fig. 3) was quantified using Image-Pro Plus, with values expressed as integrated optical density (IOD). The numerical data are expressed as the least square mean 6 SEM. *P , 0.05 for the expression of PTGES-1, PTGES-3, PTGFSAKR1B1, PGDH, PTGER2, or PTGER4 protein in luteal cells on Day 16 of the CY versus PX. See Materials and Methods and Results for more details.

of PX. Moreover, PGF2a:PGE2 ratio was higher (P , 0.05) on Day 16 of the CY, whereas the PGE2:PGF2a ratio was higher (P , 0.05) on Day 16 of PX. Furthermore, we conducted CL explant culture in which CL tissue slices were incubated with PGH2, and the conversion of

PGH2 into PGF2a and PGE2 was measured. Results indicated that luteal tissues from Day 16 of CY (Fig. 7A) had a higher capacity to convert PGH2 to PGF2a (P , 0.05), whereas luteal tissues from Day 16 of PX (Fig. 7B) had a higher capacity to convert PGH2 to PGE2 (P , 0.05) in a dose-dependent 7

Article 97

LEE ET AL.

Downloaded from www.biolreprod.org.

FIG. 5. Ovarian venous plasma concentrations of PGF2a, PGE2, PGFM, and PGEM on Days 14 and 16 of the CY and PX (n ¼ 4 for each day). a) Concentrations of PGF2a on Days 14 and 16 of the CY versus PX. b) Concentrations of PGFM on Days 14 and 16 of the CY versus PX. c) Concentrations of PGE2 on Days 14 and 16 of the CY versus PX. The numerical data are expressed as the least square mean 6 SEM. a–cP , 0.05. See Materials and Methods and Results for more details.

manner. Collectively, these results (Figs. 5–7) indicate that CL secretes more PGF2a at the time of luteolysis, but it secretes more PGE2 during the establishment of pregnancy. Transport of IFNT and PGF2a and PGE2 from the Uterus to the CL At the time of establishment of pregnancy, the CL appears to be rescued and protected by factors that originate from the

FIG. 7. A) Selective conversion of PGH2 into PGF2a and PGE2 by the Day 16 CL of the CY. a) Dose response (0 or 25 vs. 50, 75, 100, or 200 lm) of PGH2 on production of PGF2a. b) Dose response (0 or 25 vs. 50, 75, 100, or 200 lm) of PGH2 on production of PGE2. *P , 0.05 for selective conversion of PGF2a compared to PGE2 at the given dose of PGH2 by the Day 16 CL of the CY. B) Selective conversion of PGH2 into PGF2a and PGE2 by the Day 16 CL pregnancy (PX). cDose response (0 or 25 vs. 50, 75, 100, or 200 lm) of PGH2 on production of PGF2a. dDose response (0 or 25 vs. 50, 75, 100, or 200 lm) of PGH2 on production of PGE2. **P , 0.05 for selective conversion of PGE2 compared to PGF2a by the Day 16 CL of PX. The numerical data are expressed as the least square mean 6 SEM. See Materials and Methods and Results for more details.

FIG. 6. Intraluteal concentrations of PGF2a and PGE2 on Day 16 of the CY and PX (n ¼ 4 for each day). a) Concentrations of PGF2a on Day 16 of the CY versus PX. b) Concentrations of PGE2 on Day 16 of the CY versus PX. c) PGF2a:PGE2 ratio on Day 16 of the CY versus PX. d) PGE2:PGF2a ratio on Day 16 of the CY versus PX. The numerical data are expressed as the least square mean 6 SEM. a–dP , 0.05. See Materials and Methods and Results for more details.

8

Article 97

LUTEAL PGF2a AND PGE2 BIOSYNTHESIS AND SIGNALING

Downloaded from www.biolreprod.org. FIG. 8. Transport of IFNT protein from the uterus to the CL via the UOP on Days 12, 14, or 16 of the CY and PX (n ¼ 4 for each group). A and B) Western blot analysis. b-Actin was measured in CL tissues as internal control. The numbers 1 and 2 denote representative animals. C–G) Indirect ELISA. **P , 0.05 for concentration of IFNT protein in the UF of the CY versus PX. *P , 0.05 for concentration of IFNT protein in the UOV or OA versus UF on Days 12, 14, and 16 of pregnancy. The numerical data are expressed as the least square mean 6 SEM. See Materials and Methods and Results for more details.

9

Article 97

LEE ET AL.

DISCUSSION In ruminants, PGF2a is the luteolytic hormone, whereas PGE2 is considered to be a luteoprotective mediator [8, 9]. Endometrial PGF2a causes functional luteolysis, but luteal PGF2a is necessary for structural luteolysis. During early pregnancy, PGE2 produced by the CL in response to endometrial PGE2 may counteract the luteolytic effect of both exogenous and endogenous PGF2a. However, the underlying molecular and cellular mechanisms of luteal PGF2a and PGE2 biosynthesis, catabolism, and signaling are largely unknown. The present results indicate that the ratio between PTGFSAKR1B1 and PTGES-1 or PTGES-3 is higher on Days 14–16 of the estrous cycle. In contrast, the ratio between PTGES-1 or PTGES-3 and PTGFS-AKR1B1 is higher on Days 14–16 of pregnancy. To evaluate the functional aspects of these enzymatic changes, we measured luteal production and secretion of PGF2a and PGE2. The ratio of PGF2a to PGE2 in ovarian venous plasma is higher on Days 14 and 16 of the estrous cycle, whereas the ratio of PGE2 to PGF2a is higher on Days 14 and 16 of pregnancy. Analyses of luteal lysates indicate that intraluteal production of PGF2a is higher on Day 16 of the estrous cycle; in contrast, intraluteal PGE2 production is increased on Day 16 of pregnancy. Data from CL explant cultures indicate that luteal tissues from Day 16 of the estrous cycle selectively convert PGH2 to PGF2a, whereas luteal tissues from Day 16 of pregnancy selectively convert PGH2 to

FIG. 9. Transport of PGF2a and PGE2 from the uterus to the CL via the UOP on Days 12, 14, or 16 of the CY and PX (n ¼ 4 for each group). A) PGF2a and PGE2 concentrations in UF. B) PGF2a and PGE2 concentrations in uterine vein (UOV) plasma. C) PGF2a and PGE2 concentrations in OA plasma. a–c) Concentrations of PGF2a on Day 16 of the CY in (a) UF, (b) UOV, and (c) OA. d and e) Concentrations of PGF2a on Days 14 or 16 of PX (d) UF and (e) UOV. f–h) Concentrations of PGE2 on Days 14 or 16 of PX (f) UF, (g) UOV, and (h) OA. *P , 0.05 for concentrations of PGF2a on Days 14 or 16 of PX versus the CY. **P , 0.05 for concentrations of PGE2 on Days 14 or 16 of PX versus the CY. The numerical data are expressed in least square mean 6 SEM. a–hP , 0.05. See Materials and Methods and Results for more details.

gravid uterus; these factors need to be transported from the gravid uterus to the ovary through the UOP. As detailed in the Introduction, the potential candidates are IFNT, PGE2, and PGF2a. We first determined the transport of the conceptus-derived factor IFNT from the uterus to the ovary. Western blot analyses 10

Article 97

Downloaded from www.biolreprod.org.

(Fig. 8, A and B) indicated that IFNT protein was abundant (P , 0.05) in UF on Days 12, 14, and 16 of PX but was not detectable in UF of the CY. IFNT protein was not detectable (P , 0.05) in the UO, OA, and CL on Days 12, 14, and 16 of PX and CY. To confirm this observation, we measured IFNT protein using the more sensitive and quantitative ELISA. Results (Fig. 8, C–H) indicated that IFNT protein was abundant (P , 0.05) in UF but not detectable (P , 0.05) in UOV and OA on Days 12, 14, and 16 of PX. As expected, IFNT protein was not detectable (P , 0.05) in UF, UOV, and OA on Days of 12, 14, and 16 of the CY. These quantitative and qualitative analyses indicate that the IFNT protein is not transported from the uterus to the CL through the UOP during the establishment of pregnancy in sheep. Next, we determined transport of PGE2 and PGF2a from the gravid uterus to the CL. Results (Fig. 9) indicated that the concentrations of PGE 2 in UF, UOV, and OA were approximately 21.1-, approximately 32.3-, and approximately 10.2-fold higher (P , 0.05) on Days 14–16 of PX compared to that of the CY. The concentrations of PGF2a in the UF, UOV, and OA were approximately 3.7-, approximately 1.6-, and approximately 0.8-fold higher (P , 0.05) on Days 14–16 of PX compared to that of the CY. The estimated rate of PGE2 transport from the uterus to the UOV on Day 16 of PX versus CY was approximately 92.1% versus approximately 24.1%, and from the UOV to the OA, transport was approximately 12.2% versus approximately 1.6%, respectively. The estimated rate of PGF2a transport from the uterus to the UOV on Day 16 of PX versus CY was approximately 37.4% versus approximately 83.8%, and from the UOV to the OA, transport was approximately 1.7% versus approximately 3.3%, respectively. The PGE2:PGF2a ratio in the UOV was approximately 72-fold higher (P , 0.05), and in the OA was approximately 115-fold higher (P , 0.05), on Day 16 of PX versus CY. These results indicate that PGE2 is more efficiently transported from the uterus to the ovary through the UOV and the OA during the establishment of pregnancy. In contrast, PGF2a is preferentially transported at the time of luteolysis.

LUTEAL PGF2a AND PGE2 BIOSYNTHESIS AND SIGNALING

11

Article 97

Downloaded from www.biolreprod.org.

SLCO2A1 protein is readily available to transport PGF2a or PGE2 from luteal cells. The present results also indicate that the expression of the FP protein in the CL on Days 12–16 is not regulated by the estrous cycle or pregnancy. It has been shown that parenteral administration of PGF2a decreased PTGFR mRNA within 12– 24 h on Days 10–14 of the estrous cycle in sheep [60, 61]. Activation of FP in turn activates PKC, inositol 1,4,5triphosphate (IP3), and Ca2 þ cell signaling pathways [4] and interacts with multiple intracellular cell signaling pathways in the luteal cells in sheep and cows [13, 62, 63]. The present results support the previous findings that PTGFR numbers or PTGFR mRNA in the CL are similar in cyclic and early pregnant sheep and that the resistance of the CL of early pregnancy to exogenous PGF2a is apparently not due to a change in PTGFR receptor numbers/concentration [64]. In addition, our results indicate that luteal PTGER2- and PTGER4-mediated PGE2 signaling is activated during the establishment of pregnancy, whereas it is suppressed at the time of luteolysis. PTGER2 and PTGER4 activate cAMP and PKA pathways and transactivates ERK1/2, AKT, NFjB, and CTNNB1 pathways in multiple cell types [5–7], which in turn activates cell survival pathways and suppresses apoptotic pathways. Thus, activation of PTGER2 and PTGER4 may be one of the critical mechanisms protecting the CL from luteolytic challenges during early pregnancy in sheep. Our future studies will determine the direct interactions between the effects of PTGER2/PTGER4 on luteal cell survival and resistance of the CL of early pregnancy to PGF2a. Expression of PTGER3 protein is marginally increased on Days 12 and 14 of pregnancy compared to that of the estrous cycle. Four PTGER3 isoforms (PTGER3A through PTGER3D) are produced by alternative splicing of the Cterminal. Each PTGER3 isoform showed different efficiency in activation of the Gq, Gs, and Gi proteins as well as inhibition or stimulation of adenylate cyclase and cAMP [4]. Activation of PTGER3A decreases cAMP, PTGER3B and PTGER3C increases cAMP, and PTGER3D decreases cAMP and increases IP3 [4]. The commercially available PTGER3 antibody used in the present study recognized N-terminal but not C-terminal splices. Thus, specific PTGER3 isoforms were not detectable in the present study. However, a recent study in cows showed that expression of PTGER3B mRNA increased in the CL in response to PGE2, which was associated with maintenance of the CL [65]. It is possible that PTGER2, PTGER4, and PTGER3B could activate or share the common cAMP-PAK intracellular pathways in the ovine CL. However, this possibility is yet to be examined by receptor-specific functional studies. The present results also show that expression of EP1 protein in the CL is decreased on Day 16 of pregnancy. Stimulation of PTGER1 is known to activate PKC, IP3, and Ca2 þ cell signaling pathways [4]. The observed decrease in expression of PTGER1 protein on Day 16 of pregnancy in the CL suggests that luteal PGE2 signaling is specifically directed towards the cAMP-PKA pathways through PTGER2, PTGER4, and PTGER3B subtypes during the establishment of pregnancy in sheep. Infusion of IFNT into a uterine vein of sheep maintains a functional CL in 80% of treated animals for up to 32 days through as-yet-unidentified mechanisms [33, 34]. Our current finding that IFNT protein is not present in uterine venous blood and in the CL on Days 14–16 of pregnancy in sheep is contrary to the recent reports by Bott et al. [33] and Oliveira et al. [34], who reported high levels of IFNT in uterine venous blood on Day 15 of pregnancy in sheep. The reason for this discrepancy is most likely the methodology used. Earlier pioneering work

PGE2. Collectively, these results indicate that luteal PG biosynthesis is selectively directed towards PGF2a at the time of luteolysis and towards PGE2 during the establishment of pregnancy. The present results support previous findings that PTGES-1 protein is highly expressed in the CL at the time of establishment of pregnancy in cows [46] and that the ratio of PTGES-1 to PTGFS-AKR1B1 mRNA is increased in luteal tissue on Day 12 of pregnancy compared to that of the estrous cycle in sheep [54]. Earlier studies reported that the administration of PGF2a on Day 11 or 12 of the estrous cycle induced expression of PTGS-2 mRNA in luteal tissues at 1–4 h, but it returned to basal levels at 12 and 24 h in sheep [50]. The present data indicate that PTGS-2 protein is constantly expressed in CL on Days 12–16 of the estrous cycle and pregnancy. Expression of PTGS-2 is required for both PGF2a and PGE2 biosynthesis. PTGES-1 is functionally coupled to PTGS-2 and controls more than 90% of total PGE2 production. PTGES-2 is preferentially coupled to constitutive PTGS-1. PTGES-3 is coupled to both PTGS-1 and PTGS-2 under different physiological circumstances [55, 56]. The present results indicate that the PTGS-2 and PTGFS-AKR1B1 pathway is involved in luteal PGF2a biosynthesis at the time of luteolysis and that the PTGS2 and PTGES-1/PTGES-3 pathway is involved in luteal PGE2 biosynthesis during the establishment of pregnancy. The net luteal production of PGF2a versus PGE2 is not only regulated by biosynthetic enzymes PTGS-2, PTGFS-AKR1B1, PTGES-1, or PTGES-3 but also by the catabolic enzyme PGDH. Our results indicate that the PGDH protein is highly expressed on Days 12–16 of pregnancy but not expressed on Days 12–16 of the estrous cycle in sheep. To determine the functional aspects of luteal PGDH in vivo, we measured the concentrations of PGFM and PGEM in ovarian venous plasma. The concentrations of PGFM are higher on Days 14 and 16 of pregnancy compared to that of the estrous cycle. Surprisingly, the concentration of PGEM is very low and does not appear to be regulated. Three isoforms of PGDH are identified. These PGDH isoforms are differentially expressed in human placental tissues [57], and their functions are not completely understood. The action of PGDH on PGF2a catabolism is well supported in the ovine CL [58]. However, PGDH action on PGE2 catabolism is largely unknown. Our result suggests that the PGDH protein detected at 29 kDa in the ovine CL catabolizes only PGF2a into PGFM, and not PGE2 into PGEM, during early pregnancy. We believe that other PGDH isoforms that are yet to be characterized or discovered may control the catabolism of PGE2 to PGEM in the ovine CL. Our present results support previous findings that PGDH mRNA expression is increased in the CL on Days 13–14 of pregnancy compared to that of the estrous cycle in sheep and that this increase is critical for catabolism of PGF2a by the CL of early pregnancy [58]. Collectively, these results indicate that intraluteal PGF2a production at the time of luteolysis is governed by an increase in the ratio of PTGFS-AKR1B1 to PTGES-1 and PTGES-3, whereas the increased intraluteal PGE2 production during the establishment of pregnancy is governed by an increase in the ratio of PTGES-1 and PTGES-3 to PTGFS-AKR1B1 plus increased PGF2a catabolism by PGDH. Both PGF2a and PGE2 synthesized within the CL need to be transported out of luteal cells and act through their specific receptors, PTGFR or PTGER, respectively, to produce their physiological effects. SLCO2A1 competitively transports PGF2a and PGE2 with equal affinity [59]. The present results indicate that expression of SLCO2A1 protein in the CL is not regulated on Days 12–16 by the estrous cycle or pregnancy. This constant expression of SLCO2A1 suggests that the

LEE ET AL.

the time of establishment of pregnancy in ruminants, as has been proposed in pigs [76]. Transport of PGE2 from the uterus to the UOV and from the UOV to the OA appears to depend on the net concentration of PGE2 in the uterine lumen and the UOV, respectively. These results suggest that PGE2 is transported from the uterus to the ovary selectively or preferentially and is not regulated by endocrine-exocrine secretory mechanisms at the time of luteolysis and establishment of pregnancy in sheep. It is possible that SLCO2A1 [43, 49, 59], either individually or in coordination with other transporters, such as ATP-binding cassette subfamily C member 4 (ABCC4, also known as MRP4) [77], could regulate this endocrine versus exocrine transport of PGF2a and the preferential or selective transport of PGE2 in ruminant endometrium. The current understanding of PG transport mechanisms is not enough to explain the selective transport of PGE2 and PGF2a from the uterus to the ovary in ruminants. Thus, more functional studies are needed to understand this complex and important mechanism. In summary, the results of the present study indicate that 1) luteal PG biosynthesis is selectively directed towards PGF2a at the time of luteolysis and towards PGE 2 during the establishment of pregnancy, 2) the ability of the CL of early pregnancy to resist luteolysis is likely due to increased intraluteal biosynthesis of PGE2 and PTGER2 and PTGER4 signaling, 3) IFNT protein is not transported from the uterus to the CL/ovary through the UOP, 4) a large proportion of endometrial PGE2 is transported from the uterus to the CL/ ovary through the UOP, and 5) endometrial PGE2 stimulated by pregnancy appears to increase luteal PGE2 biosynthesis and, hence, activate PTGER2- and PTGER4-mediated intracellular mechanisms, thus rescuing and protecting the CL during the establishment of pregnancy in sheep/ruminants. The underlying mechanisms that regulate selective luteal PGE2 biosynthesis and signaling during the establishment of pregnancy are not presently known. Our future studies will investigate these proposed direct interactions among IFNT, endometrial/conceptus-derived PGE2, luteal PGE2 biosynthesis, and luteal PTGER2 and PTGER4 signaling as well as their effects on the survival/resistance of the CL of early pregnancy. ACKNOWLEDGMENT We are very grateful to Dr. Fuller W. Bazer, Department of Animal Sciences, Texas A&M University, for providing IFNT antibody. We thank Dr. Thomas E. Spencer, Department of Animal Sciences, Texas A&M University, for assistance with animal husbandry (current address: Department of Animal Sciences, Washington State University). We also thank Dr. Robert C. Burghardt and the Image Analysis Laboratory at the College of Veterinary Medicine and Biomedical Sciences, Texas A&M University, for technical service/advice.

REFERENCES 1. Smith WL, Dewitt DL. Prostaglandin endoperoxide H synthases-1 and -2. Adv Immunol 1996; 62:167–215. 2. Tai HH, Ensor CM, Tong M, Zhou H, Yan F. Prostaglandin catabolizing enzymes. Prostaglandins Other Lipid Mediat 2002; 68–69:483–493. 3. Schuster VL. Molecular mechanisms of prostaglandin transport. Annu Rev Physiol 1998; 60:221–242. 4. Narumiya S, Sugimoto Y, Ushikubi F. Prostanoid receptors: structures, properties, and functions. Physiol Rev 1999; 79:1193–1226. 5. Cha YI, DuBois RN. NSAIDs and cancer prevention: targets downstream of COX-2. Annu Rev Med 2007; 58:239–252. 6. Castellone MD, Teramoto H, Williams BO, Druey KM, Gutkind JS. Prostaglandin E2 promotes colon cancer cell growth through a Gs-axinbeta-catenin signaling axis. Science 2005; 310:1504–1510. 7. Banu SK, Lee J, Speights VO Jr, Starzinski-Powitz A, Arosh JA. Selective inhibition of prostaglandin E2 receptors EP2 and EP4 induces apoptosis of human endometriotic cells through suppression of ERK1/2, AKT, NFkB

12

Article 97

Downloaded from www.biolreprod.org.

by Godkin et al. [66] indicated that intrauterine infusion of 125Ilabeled ovine trophoblast protein-1 (oTP-1, later named IFNT) into Day 12 nonpregnant ewes is retained within the uterus. Only very small amounts of subunits of [125I]oTP-1, but not intact [125I]oTP-1 protein, entered the maternal vasculature [66]. Moreover, IFNT is a relatively large protein (19 kDa), and it could not be transferred locally to the ovary via the UOP. Our present quantitative (ELISA) and qualitative (Western blot) analyses confirm that IFNT protein is not transported from the uterus to the CL through the UOP at the time of establishment of pregnancy in sheep, which is in agreement with the earlier pioneering study by Godkin et al. [66]. Furthermore, embryo/conceptus transfer and hysterectomy experiments indicate that the luteolytic and luteoprotective mechanisms are locally mediated between the uterus and the CL of the ipsilateral ovary and do not act systemically in sheep [38–40]. This local action of the embryonic antiluteolytic/ luteoprotective factor effectively precludes any direct action of a protein transported systemically from the gravid uterus to the CL in ruminants [21]. Several authors [34, 67–70] have reported an increase in IFN-stimulated genes in peripheral tissues (leukocytes, liver, and CL) of pregnant sheep and cows. However, because of the established local effect between the uterus and the CL in early pregnancy, the above-reported systemic effects on peripheral expression of interferonstimulated genes appeared not to be critical for establishment of pregnancy. Given the facts discussed above, the reports by Bott et al. [33] and Oliveira et al. [34] on the release of IFNT from the uterus to the uterine vein at the time of establishment of pregnancy in sheep are still controversial; however, these studies open up a new area of research. Thus, further work will be required to determine the physiological significance of uterine vein infusions of recombinant ovine IFNT and the possible mechanisms by which such parenteral administrations of IFNT may affect luteal function in sheep. The concentrations of both PGE2 and PGF2a are increased in the uterine venous plasma at the time of establishment of pregnancy in sheep [10, 20, 21, 27, 71–74]. Therefore, we measured the transport of PGF2a and PGE2 from the uterus to the ovary through the UOP on Days 12–16 of the estrous cycle and pregnancy. Our results indicate that PGF2a is preferentially transported at the time of luteolysis from the uterus to the ovary. By contrast PGE2 is preferentially transported at the time of establishment of pregnancy from the uterus to the ovary. Moreover, the PGE2:PGF2a ratio is 72-fold higher in the UOV and 115-fold higher in the OA on Day 16 of pregnancy compared to that of the estrous cycle. Our present data are in agreement with several previous findings that the secretion of PGE2 from the uterus into the UOV increases during the period of luteal resistance in early pregnancy in sheep [20, 21, 27, 71– 75]. PGE2 is a lipid-soluble mediator with a small molecular weight of 0.35 kDa, is structurally similar to PGF2a, and can be easily transported locally from the uterus to the ovary [21] through PGT-mediated mechanisms in the UOP [52, 59] and, thus, exert a luteoprotective effect through multiple mechanisms. The intrauterine concentration of PGF2a is significantly higher compared to that of PGE2 on Day 16 of the estrous cycle and pregnancy. Importantly, approximately 85% of PGF2a is transported from the uterus to the UOV at the time of luteolysis, whereas only approximately 35% of PGF2a is transported from the uterus to the UOV at the time of establishment of pregnancy. This observation suggests that PGF2a is secreted/transported from the endometrium to the UOV (endocrine secretion) at the time of luteolysis and from the endometrium to the uterine lumen (exocrine secretion) at

LUTEAL PGF2a AND PGE2 BIOSYNTHESIS AND SIGNALING

8. 9.

10. 11.

12.

13.

14.

15.

16.

18.

19. 20.

21.

22. 23.

24.

25. 26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

39. 40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

50.

51.

52.

53. 54.

13

for establishment and maintenance of pregnancy. Anim Reprod Sci 2004; 82–83:537–550. Bott RC, Ashley RL, Henkes LE, Antoniazzi AQ, Bruemmer JE, Niswender GD, Bazer FW, Spencer TE, Smirnova NP, Anthony RV, Hansen TR. Uterine vein infusion of interferon tau (IFNT) extends luteal life span in ewes. Biol Reprod 2010; 82:725–735. Oliveira JF, Henkes LE, Ashley RL, Purcell SH, Smirnova NP, Veeramachaneni DN, Anthony RV, Hansen TR. Expression of interferon (IFN)-stimulated genes in extrauterine tissues during early pregnancy in sheep is the consequence of endocrine IFN-tau release from the uterine vein. Endocrinology 2008; 149:1252–1259. Mapletoft RJ, Del Campo MR, Ginther OJ. Unilateral luteotropic effect of uterine venous effluent of a gravid uterine horn in sheep. Proc Soc Exp Biol Med 1975; 150:129–133. Mapletoft RJ, Ginther OJ. Adequacy of main uterine vein and the ovarian artery in the local venoarterial pathway for uterine-induced luteolysis in ewes. Am J Vet Res 1975; 36:957–963. Mapletoft RJ, Lapin DR, Ginther OJ. The ovarian artery as the final component of the local luteotropic pathway between a gravid uterine horn and ovary in ewes. Biol Reprod 1976; 15:414–421. Moor RM, Hay MF, Short RV, Rowson LE. The corpus luteum of the sheep: effect of uterine removal during luteal regression. J Reprod Fertil 1970; 21:319–326. Moor RM, Rowson LE. The corpus luteum of the sheep: effect of the removal of embryos on luteal function. J Endocrinol 1966; 34:497–502. Moor RM, Rowson LE, Hay MF, Caldwell BV. The corpus luteum of the sheep: effect of the conceptus on luteal function at several stages during pregnancy. J Endocrinol 1969; 43:301–307. Mapletoft RJ, Del Campo MR, Ginther OJ. Local venoarterial pathway for uterine-induced luteolysis in cows. Proc Soc Exp Biol Med 1976; 153: 289–294. Moor RM, Rowson LE. Local maintenance of the corpus luteum in sheep with embryos transferred to various isolated portions of the uterus. J Reprod Fertil 1966; 12:539–550. Banu SK, Lee J, Satterfield MC, Spencer TE, Bazer FW, Arosh JA. Molecular cloning and characterization of prostaglandin transporter in ovine endometrium: role of mitogen activated protein kinase pathways in release of prostaglandin F2 alpha. Endocrinology 2008; 149:219–231. Simmons RM, Satterfield MC, Welsh TH Jr, Bazer FW, Spencer TE. HSD11B1, HSD11B2, PTGS2, and NR3C1 expression in the periimplantation ovine uterus: effects of pregnancy, progesterone, and interferon tau. Biol Reprod 2010; 82:35–43. McCracken JA, Carlson JC, Glew ME, Goding JR, Baird DT, Green K, Samuelsson B. Prostaglandin F2 identified as a luteolytic hormone in sheep. Nat New Biol 1972; 238:129–134. Arosh JA, Banu SK, Chapdelaine P, Emond V, Kim JJ, MacLaren LA, Fortier MA. Molecular cloning and characterization of bovine prostaglandin E2 receptors EP2 and EP4: expression and regulation in endometrium and myometrium during the estrous cycle and early pregnancy. Endocrinology 2003; 144:3076–3091. Banu SK, Lee J, Speights VO, Starzinski-Powitz A, Arosh JA. Cyclooxygenase-2 regulates survival, migration and invasion of human endometriotic cells through multiple mechanisms. Endocrinology 2008; 149:1180–1189. Bradford MM. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 1976; 72:248–254. Banu SK, Lee J, Stephen SD, Nithy TK, Arosh JA. Interferon tau regulates PGF2 alpha release from the ovine endometrial epithelial cells via activation of novel JAK/EGFR/ERK/EGR-1 pathways. Mol Endocrinol 2010; 24:2315–2330. Lee J, Banu SK, Subbarao T, Starzinski-Powitz A, Arosh JA. Selective inhibition of prostaglandin E2 receptors EP2 and EP4 inhibits invasion of human immortalized endometriotic epithelial and stromal cells through suppression of metalloproteinases. Mol Cell Endocrinol 2011; 332: 306–313. Tiano HF, Loftin CD, Akunda J, Lee CA, Spalding J, Sessoms A, Dunson DB, Rogan EG, Morham SG, Smart RC, Langenbach R. Deficiency of either cyclooxygenase (COX)-1 or COX-2 alters epidermal differentiation and reduces mouse skin tumorigenesis. Cancer Res 2002; 62:3395–3401. Lee J, McCracken JA, Banu SK, Rodriguez R, Nithy TK, Arosh JA. Transport of prostaglandin F(2alpha) pulses from the uterus to the ovary at the time of luteolysis in ruminants is regulated by prostaglandin transporter-mediated mechanisms. Endocrinology 2010; 151:3326–3335. Lo WC, Summers PM. In vitro culture and interferon-tau secretion by ovine blastocysts. Anim Reprod Sci 2002; 70:191–202. Costine BA, Inskeep EK, Blemings KP, Flores JA, Wilson ME.

Article 97

Downloaded from www.biolreprod.org.

17.

and b-catenin pathways and activation of intrinsic apoptotic mechanisms. Mol Endocrinol 2009; 23:1291–1305. McCracken JA, Custer EE, Lamsa JC. Luteolysis: a neuroendocrinemediated event. Physiol Rev 1999; 79:263–323. Niswender GD, Juengel JL, Silva PJ, Rollyson MK, McIntush EW. Mechanisms controlling the function and life span of the corpus luteum. Physiol Rev 2000; 80:1–29. Weems CW, Weems YS, Randel RD. Prostaglandins and reproduction in female farm animals. Vet J 2006; 171:206–228. McCracken JA, Custer EE, Schreiber DT, Tsang PC, Keator CS, Arosh JA. A new in vivo model for luteolysis using systemic pulsatile infusions of PGF(2alpha). Prostaglandins Other Lipid Mediat 2012; 97:90–96. Rexroad CE Jr, Guthrie HD. Prostaglandin F2 alpha and progesterone release in vitro by ovine luteal tissue during induced luteolysis. Adv Exp Med Biol 1979; 112:639–644. Anderson LE, Wu YL, Tsai SJ, Wiltbank MC. Prostaglandin F(2alpha) receptor in the corpus luteum: recent information on the gene, messenger ribonucleic acid, and protein. Biol Reprod 2001; 64:1041–1047. Tsai SJ, Wiltbank MC. Prostaglandin F2alpha induces expression of prostaglandin G/H synthase-2 in the ovine corpus luteum: a potential positive feedback loop during luteolysis. Biol Reprod 1997; 57: 1016–1022. Tsai SJ, Wiltbank MC. Prostaglandin F2alpha regulates distinct physiological changes in early and midcycle bovine corpora lutea. Biol Reprod 1998; 58:346–352. Wiltbank MC, Ottobre JS. Regulation of intraluteal production of prostaglandins. Reprod Biol Endocrinol 2003; 1:91–102. Inskeep EK, Smutny WJ, Butcher RL, Pexton JE. Effects of intrafollicular injections of prostaglandins in non-pregnant and pregnant ewes. J Anim Sci 1975; 41:1098–1104. Nancarrow CD, Evison BM, Connell PJ. Effect of embryos on luteolysis and termination of early pregnancy in sheep with cloprostenol. Biol Reprod 1982; 26:263–269. Pratt BR, Butcher RL, Inskeep EK. Antiluteolytic effect of the conceptus and of PGE2 in ewes. J Anim Sci 1977; 45:784–791. Silvia WJ, Niswender GD. Maintenance of the corpus luteum of early pregnancy in the ewe. III. Differences between pregnant and nonpregnant ewes in luteal responsiveness to prostaglandin F2 alpha. J Anim Sci 1984; 59:746–753. Silvia WJ, Niswender GD. Maintenance of the corpus luteum of early pregnancy in the ewe. IV. Changes in luteal sensitivity to prostaglandin F2 alpha throughout early pregnancy. J Anim Sci 1986; 63:1201–1207. Kittok RJ, Britt JH. Corpus luteum function in ewes given estradiol during the estrous cycle or early pregnancy. J Anim Sci 1977; 45:336–341. Henderson KM, Scaramuzzi RJ, Baird DT. Simultaneous infusion of prostaglandin E2 antagonizes the luteolytic action of prostaglandin F2alpha in vivo. J Endocrinol 1977; 72:379–383. Charpigny G, Reinaud P, Tamby JP, Creminon C, Guillomot M. Cyclooxygenase-2 unlike cyclooxygenase-1 is highly expressed in ovine embryos during the implantation period. Biol Reprod 1997; 57: 1032–1040. Hyland JH, Manns JG, Humphrey WD. Prostaglandin production by ovine embryos and endometrium in vitro. J Reprod Fertil 1982; 65:299–304. Lacroix MC, Kann G. Comparative studies of prostaglandins F2 alpha and E2 in late cyclic and early pregnant sheep: in vitro synthesis by endometrium and conceptus effects of in vivo indomethacin treatment on establishment of pregnancy. Prostaglandins 1982; 23:507–526. Ellinwood WE, Nett TM, Niswender GD. Maintenance of the corpus luteum of early pregnancy in the ewe. II. Prostaglandin secretion by the endometrium in vitro and in vivo. Biol Reprod 1979; 21:845–856. Marcus GJ. Prostaglandin formation by the sheep embryo and endometrium as an indication of maternal recognition of pregnancy. Biol Reprod 1981; 25:56–64. Magness RR, Huie JM, Hoyer GL, Huecksteadt TP, Reynolds LP, Seperich GJ, Whysong G, Weems CW. Effect of chronic ipsilateral or contralateral intrauterine infusion of prostaglandin E2 (PGE2) on luteal function of unilaterally ovariectomized ewes. Prostaglandins Med 1981; 6: 389–401. Pratt BR, Butcher RL, Inskeep EK. Effect of continuous intrauterine administration of prostaglandin E2 on life span of corpora lutea of nonpregnant ewes. J Anim Sci 1979; 48:1441–1446. Reynolds LP, Stigler J, Hoyer GL, Magness RR, Huie JM, Huecksteadt TP, Whysong GL, Behrman HR, Weems CW. Effect of PGE1 on PGF2 alpha-induced luteolysis in nonbred ewes. Prostaglandins 1981; 21: 957–972. Spencer TE, Burghardt RC, Johnson GA, Bazer FW. Conceptus signals

LEE ET AL.

55.

56. 57. 58.

59.

60. 61.

63. 64. 65.

66. Godkin JD, Bazer FW, Roberts RM. Ovine trophoblast protein 1, an early secreted blastocyst protein, binds specifically to uterine endometrium and affects protein synthesis. Endocrinology 1984; 114:120–130. 67. Chen Y, Green JA, Antoniou E, Ealy AD, Mathialagan N, Walker AM, Avalle MP, Rosenfeld CS, Hearne LB, Roberts RM. Effect of interferontau administration on endometrium of nonpregnant ewes: a comparison with pregnant ewes. Endocrinology 2006; 147:2127–2137. 68. Gifford CA, Racicot K, Clark DS, Austin KJ, Hansen TR, Lucy MC, Davies CJ, Ott TL. Regulation of interferon-stimulated genes in peripheral blood leukocytes in pregnant and bred, nonpregnant dairy cows. J Dairy Sci 2007; 90:274–280. 69. Han H, Austin KJ, Rempel LA, Hansen TR. Low blood ISG15 mRNA and progesterone levels are predictive of non-pregnant dairy cows. J Endocrinol 2006; 191:505–512. 70. Yankey SJ, Hicks BA, Carnahan KG, Assiri AM, Sinor SJ, Kodali K, Stellflug JN, Ott TL. Expression of the antiviral protein Mx in peripheral blood mononuclear cells of pregnant and bred, non-pregnant ewes. J Endocrinol 2001; 170:R7–R11. 71. Rawlings NC, Hyland JH. Prostaglandin F and E levels in the conceptus, uterus and plasma during early pregnancy in the ewe. Prostaglandins 1985; 29:933–951. 72. Silvia WJ, Ottobre JS, Inskeep EK. Concentrations of prostaglandins E2, F2 alpha and 6-keto-prostaglandin F1 alpha in the utero-ovarian venous plasma of nonpregnant and early pregnant ewes. Biol Reprod 1984; 30: 936–944. 73. Vincent DL, Inskeep EK. Role of progesterone in regulating uteroovarian venous concentrations of PGF2 alpha and PGE2 during the estrous cycle and early pregnancy in ewes. Prostaglandins 1986; 31:715–733. 74. Lewis GS, Jenkins PE, Fogwell RL, Inskeep EK. Concentrations of prostaglandins E2 and F2 alpha and their relationship to luteal function in early pregnant ewes. J Anim Sci 1978; 47:1314–1323. 75. Vincent DL, Meredith S, Inskeep EK. Advancement of uterine secretion of prostaglandin E2 by treatment with progesterone and transfer of asynchronous embryos. Endocrinology 1986; 119:527–529. 76. Bazer FW, Thatcher WW. Theory of maternal recognition of pregnancy in swine based on estrogen controlled endocrine versus exocrine secretion of prostaglandin F2alpha by the uterine endometrium. Prostaglandins 1977; 14:397–400. 77. Lacroix-Pepin N, Danyod G, Krishnaswamy N, Mondal S, Rong PM, Chapdelaine P, Fortier MA. The multidrug resistance-associated protein 4 (MRP4) appears as a functional carrier of prostaglandins regulated by oxytocin in the bovine endometrium. Endocrinology 2011; 152: 4993–5004.

14

Article 97

Downloaded from www.biolreprod.org.

62.

Mechanisms of reduced luteal sensitivity to prostaglandin F2alpha during maternal recognition of pregnancy in ewes. Domest Anim Endocrinol 2007; 32:106–121. Thoren S, Jakobsson PJ. Coordinate up- and down-regulation of glutathione-dependent prostaglandin E synthase and cyclooxygenase-2 in A549 cells. Inhibition by NS-398 and leukotriene C4. Eur J Biochem 2000; 267:6428–6434. Kudo I, Murakami M. Diverse functional coupling of prostanoid biosynthetic enzymes in various cell types. Adv Exp Med Biol 1999; 469:29–35. Lindstrom T, Bennett P. Transcriptional regulation of genes for enzymes of the prostaglandin biosynthetic pathway. Prostaglandins Leukot Essent Fatty Acids 2004; 70:115–135. Silva PJ, Juengel JL, Rollyson MK, Niswender GD. Prostaglandin metabolism in the ovine corpus luteum: catabolism of prostaglandin F(2alpha) (PGF(2alpha)) coincides with resistance of the corpus luteum to PGF(2alpha). Biol Reprod 2000; 63:1229–1236. Banu SK, Arosh JA, Chapdelaine P, Fortier MA. Molecular cloning and spatio-temporal expression of the prostaglandin transporter: a basis for the action of prostaglandins in the bovine reproductive system. Proc Natl Acad Sci U S A 2003; 100:11747–11752. Rueda BR, Botros IW, Pierce KL, Regan JW, Hoyer PB. Comparison of mRNA levels for the PGF(2alpha) receptor (FP) during luteolysis and early pregnancy in the ovine corpus luteum. Endocrine 1995; 3:781–787. Tsai SJ, Anderson LE, Juengel J, Niswender GD, Wiltbank MC. Regulation of prostaglandin F2 alpha and E receptor mRNA by prostaglandin F 2 alpha in ovine corpora lutea. J Reprod Fertil 1998; 114:69–75. Davis JS, Rueda BR. The corpus luteum: an ovarian structure with maternal instincts and suicidal tendencies. Front Biosci 2002; 7: d1949–1978. Wiltbank MC, Shiao TF, Bergfelt DR, Ginther OJ. Prostaglandin F2 alpha receptors in the early bovine corpus luteum. Biol Reprod 1995; 52:74–78. Wiepz GJ, Wiltbank MC, Nett TM, Niswender GD, Sawyer HR. Receptors for prostaglandins F2 alpha and E2 in ovine corpora lutea during maternal recognition of pregnancy. Biol Reprod 1992; 47:984–991. Weems YS, Bridges PJ, Jeoung M, Arreguin-Arevalo JA, Nett TM, Vann RC, Ford SP, Lewis AW, Neuendorff DA, Welsh TH Jr, Randel RD, Weems CW. In vivo intra-luteal implants of prostaglandin (PG) E1 or E2 (PGE1, PGE2) prevent luteolysis in cows. II: mRNA for PGF2alpha, EP1, EP2, EP3 (A-D), EP3A, EP3B, EP3C, EP3D, and EP4 prostanoid receptors in luteal tissue. Prostaglandins Other Lipid Mediat 2012; 97: 60–65.