Abstract
Prostaglandins (PGs) are important lipid mediators regulating various reproductive processes in many species. In pigs, the expression pattern of PGE2 and PGF2α metabolic enzymes and the regulatory mechanism controlling PGE2 and PGF2α levels in the uterus during pregnancy are not completely understood. This study determined endometrial expression of the genes (PLA2G4A, PTGS1, PTGS2, PTGES, PTGES2, PTGES3, AKR1B1, CBR1, and HPGD) involved in PGE2 and PGF2α metabolism during the estrous cycle and pregnancy and measured levels of PGE2 and PGF2α in uterine endometrial tissues and uterine flushings at the time of conceptus implantation in pigs. Except PTGES3, expression of the genes studied changed in a pregnancy-stage-specific manner, and localization of PTGES, AKR1B1, CBR1, and HPGD mRNAs were cell-type specific in the uterine endometrium. Levels of both PGE2 and PGF2α in uterine endometrial tissues and uterine lumen were higher on Day 12 of pregnancy than those of the estrous cycle and affected by different morphology of spherical and filamentous conceptuses. Furthermore, we determined that endometrial expression of AKR1B1, known to encode a PGF2α synthase in other species, was increased by estrogen and interleukin-1beta and that AKR1B1 exhibited PGF2α synthase activity in the porcine uterine endometrium. These results in pigs indicate that the PGE2 and PGF2α metabolic enzymes are expressed stage specifically in the endometrium during pregnancy and regulate the abundance of PGE2 and PGF2α in the uterus at the time of implantation and that AKR1B1 may act as a major PGF synthase in the endometrium during early pregnancy.
Introduction
During the peri-implantation period of pregnancy, the conceptus signals its presence to the mother to escape luteolysis and prolong the life span of the corpus luteum (CL) beyond the estrous cycle [1]. This physiological process is known as maternal recognition of pregnancy, which provides for continuous secretion of progesterone by the CL for the establishment and maintenance of pregnancy. In pigs, the conceptus elongates morphologically from spherical to tubular and to filamentous forms between Days 10 and 12 of pregnancy and secretes estrogen that functions as the signal for maternal recognition of pregnancy and a cytokine, interleukin-1β (IL1B) [2]. The mechanism of maternal recognition of pregnancy induced by estrogen has been explained by endocrine and exocrine theory [3]. The concept of the theory is that the uterine endometrial cells differentially secrete prostaglandin F2α (PGF2α), which causes luteolysis in the CL in either an endocrine or an exocrine manner. In cyclic pigs, the uterine endometrium secretes PGF2α in an endocrine manner into the uterine vasculature, which allows it to be transported to the ovary to cause luteolysis on Days 15 and 16 of the estrous cycle. However, in pregnant pigs, the endometrium responds to estrogen produced by conceptuses on Days 11 and 12 to secrete PGF2α in an exocrine manner into the uterine lumen, where it is sequestered to exert its biological actions in the uterus and/or metabolized to prevent luteolysis [2, 3].
Levels of PGE2 as well as PGF2α increase in the uterine lumen on Days 11–14 of pregnancy in pigs [4]. Intrauterine infusion of PGE2 during the diestrus period delayed the decline of plasma progesterone levels and extended CL function [5] and maintained the plasma progesterone levels longer than PGF2α alone when infused with PGF2α into the cycling gilts [6]. This suggests that PGE2 has a luteotrophic effect and protects CL against the luteolytic action of PGF2α [5, 6]. Furthermore, the PGE2:PGF2α ratio produced in cultured endometrial stromal cells obtained from pregnant pigs was higher than that from cyclic pigs [7]. Based on these findings, it is proposed that prevention of luteolysis during the maternal recognition of pregnancy is also mediated by an increase in the PGE2:PGF2α ratio in response to estrogen secreted by conceptuses [8–10]. Although there is no reported evidence for a definitive role of PGs in conceptus development in pigs, it has been shown that PGs of conceptus and/or endometrial origin function in conceptus implantation by modulating endometrial gene expression in mice, humans, and sheep [11, 12]. Further, treatment of pregnant gilts and sows with indomethacin, an inhibitor of PG synthesis, leads to embryonic death and low pregnancy rates [13]. Therefore, regulation of PGE2 and PGF2α production in the uterine endometrium at the time of implantation may be critical for modulating luteolysis, maternal recognition of pregnancy, and conceptus development. However, the mechanism regulating the secretion of endometrial/conceptus PGs and the role(s) of endometrial and uterine luminal PGE2 and PGF2α during the implantation period are not fully understood in pigs.
Synthesis of PGE2 and PGF2α in the cell involves sequential actions of several enzymes, including phospholipase A2 (PLA2G4A; also known as cytosolic phospholipase A2), PG-endoperoxide synthases (PTGS1 and PTGS2), and specific terminal PG synthases (PGE synthase and PGF synthase) [14, 15]. PGE2 can be converted to PGF2α by carbonyl reductase 1 (CBR1; also called PG-9-ketoreductase), which can also act to convert PGE2 and PGF2α to inactive PG metabolites [16]. There are three isoforms reported for PGE synthase (PTGES, PTGES2, and PTGES3). PGs are catabolized mainly by hydroxyprostaglandin dehydrogenase (HPGD) to terminate their action [17]. Expression of HPGD in the uterine endometrium has been reported for humans [18], mice [19], and cows [20] but not pigs. Expression of PTGS1, PTGS2, PTGES, and CBR1 has been reported in the uterine endometrium during early pregnancy in pigs [9, 21–23], but expression of these during the later stage of pregnancy and other enzymes involved in PGE2 and PGF2α synthesis and metabolism during pregnancy has not been completely studied in pigs.
Several enzymes classified as members of the aldo-keto reductase (AKR) superfamily are known to have PGF synthase activity [15], but which one of these is the major enzyme for PGF synthase has not been determined in pigs. It has been suggested that an AKR1C subclass is a PGF synthase responsible for PGF2α production in the porcine endometrium [9]; however, the expression pattern of this enzyme does not seem to correlate with the levels of uterine endometrial and luminal PGF2α during the estrous cycle and early pregnancy. Interestingly, recent studies have shown that AKR1B1 is the major PGF synthase responsible for PGF2α synthesis from PGH2 in the bovine and human uterine endometrium [24–26]. It has been reported that AKR1B1 is expressed in the porcine uterine endometrium, proposing that AKR1B1 may be involved in glucose metabolism [27]. But its function in PGF2α synthesis and the regulatory mechanism of AKR1B1 expression have not been well determined in pigs.
Therefore, to understand the regulatory mechanism of PGE2 and PGF2α production in the uterine endometrium during pregnancy in pigs, we determined 1) expression of genes responsible for PG synthesis and catabolism in the uterine endometrium during the estrous cycle and pregnancy and early stage conceptuses; 2) cell-specific expression of PTGES, AKR1B1, CBR1, and HPGD in the uterine endometrium; 3) levels of PGE2 and PGF2α in the uterine endometrium and lumen; 4) regulation of AKR1B1 mRNAs in the uterine endometrium; and 5) characterization of AKR1B1 as a PGF synthase in the uterine endometrium at the time of implantation.
Materials and Methods
Animals and Tissue Preparation
All experimental procedures involving animals were conducted in accordance with the Guide for Care and Use of Research Animals in Teaching and Research and approved by the Institutional Animal Care and Use Committee of Yonsei University. Sexually mature crossbred female gilts were assigned randomly to either cyclic or pregnant status. The reproductive tracts of gilts were obtained immediately after slaughter on either Day 12 (n = 5) or Day 15 (n = 3) of the estrous cycle and either Day 12 (n = 4), 15 (n = 5), 30 (n = 4), 60 (n = 3), 90 (n = 3), or 114 (n = 4) of pregnancy. Pregnancy was confirmed by the presence of apparently normal filamentous conceptuses in uterine flushings on Days 12 and 15 and the presence of embryos and placenta on the later days of pregnancy. Additional tissues from the endometrium with spherical conceptuses were obtained on Day 11 of pregnancy (n = 3). Uterine flushings were obtained by introducing and recovering 50 ml PBS (pH 7.4) after hysterectomy (25 ml/uterine horn). The flushings were clarified by centrifugation (3000 × g for 10 min at 4°C), aliquoted, and frozen at −80°C until analyzed.
Endometrium, dissected free of myometrium, was collected from the middle portion of each uterine horn, snap frozen in liquid nitrogen, and stored at −80°C for RNA extraction. For in situ hybridization analysis, cross sections of endometrium were fixed in 4% paraformaldehyde in PBS (pH 7.4) for 24 h and then embedded in paraffin as previously described [28].
Explant Cultures
Endometrium from gilts on Day 12 of the estrous cycle was dissected from the myometrium and placed into warm phenol red-free Dulbecco modified Eagle medium/F-12 culture medium (DMEM/F-12; Sigma, St. Louis, MO) containing penicillin G (100 IU/ml) and streptomycin (0.1 mg/ml) as described previously [29], with some modifications. The endometrium was minced with scalpel blades into small pieces (2–3 mm3), and aliquots of 500 mg were placed into T25 flasks with serum-free modified DMEM/F-12 containing 10 μg/ml insulin (Sigma), 10 ng/ml transferrin (Sigma), and 10 ng/ml hydrocortisone (Sigma). Endometrial explants were cultured immediately after mincing in the presence of ethanol (control), E2 (50 ng/ml; Sigma), P4 (3 ng/ml; Sigma), E2 + P4, E2 + P4 + ICI182 780 (ICI; an estrogen receptor antagonist; 200 ng/ml; Tocris Bioscience, Ellisville, MO), or E2 + P4 + RU486 (RU; a progesterone receptor antagonist; 30 ng/ml; Sigma) for 24 h with rocking in an atmosphere of 5% CO2 in air at 37°C. To determine the effects of IL1B on expression of endometrial genes, explant tissues were treated with 0, 1, 10, or 100 ng/ml IL1B (Sigma) in the presence of both E2 (50 ng/ml) and P4 (3 ng/ml) at 37°C for 24 h.
To determine whether PGF2α secretion decreased after blocking AKR1B1 activity, explant tissues were treated with ethyl 1-benzyl-3-hydroxy-2(5H)-oxopyrrole-4-carboxylate (EBPC; Tocris Bioscience), a specific blocker for AKR1B1 [30]. Explant tissues were treated with 0, 1, 10, or 20 ng/ml EBPC in the presence of both E2 (50 ng/ml) and P4 (3 ng/ml) at 37°C for 24 h. Explant tissues were then harvested, and total RNA or protein was extracted for real-time RT-PCR or immunoblot analysis, respectively, to determine expression levels for AKR1B1 mRNAs and AKR1B1 proteins. These experiments were conducted using endometrium from three gilts on Day 12 of the estrous cycle, and treatments were performed in triplicate using tissues obtained from each of the three gilts.
Total RNA Extraction and Cloning of Genes Involved in PG Synthesis and Catabolism
Total RNA was extracted from endometrial and conceptus tissues using TRIzol reagent (Invitrogen Life Technology, Carlsbad, CA) according to the manufacturer's recommendations. The quantity of RNA was assessed spectrophotometrically, and integrity of RNA was validated following electrophoresis in 1% agarose gel.
Four micrograms of total RNA were treated with DNase I (Promega, Madison, WI) and reverse transcribed using SuperScript II Reverse Transcriptase (Invitrogen) to obtain cDNAs. The cDNA templates were then diluted 1:4 with sterile water and amplified by PCR using Taq polymerase (Takara Bio, Shiga, Japan). The PCR conditions and sequences of primer pairs are listed in Table 1. The PCR products were separated on 2% agarose gel and visualized by ethidium bromide staining. The identity of each amplified PCR product was verified by sequence analysis after cloning into the pCRII vector (Invitrogen).
Summary of PCR primer sequences used for RT-PCR and real-time RT-PCR and expected product sizes.
| Gene name | Primer sequences (5′ → 3′)a | Product size (bp) | GenBank accession no. |
|---|---|---|---|
| PLA2G4A | F: TAAAGGAGTGCTATGTCTTCAAACC | 172 | XM_003130393 |
| R: GGGTCATCAAAGATATCAAAATCAG | |||
| PTGS1 | F: CAACACTTCACCCACCAGTTCTTC | 99 | AF207823 |
| R: TCCATAAATGTGGCCGAGGTCTAC | |||
| PTGS2 | F: TCGACCAGAGCAGAGAGATGAGAT | 134 | AF207824 |
| R: ACCATAGAGCGCTTCTAACTCTGC | |||
| PTGES | F: AGAGCATGAAAACCATCACTCC | 249 | NM_001038631 |
| R: CTCAAGGACATTCTGTCAGGTTC | |||
| PTGES2 | F: ATACCAGTACAAGACGTGTCCATTC | 195 | XM_001927936 |
| R: ATGATGACAGAGGAGTCATTCAGTT | |||
| PTGES3 | F: CAAATGATTCCAAGCATAAAAGAAC | 235 | XM_001929413 |
| R: GGTAAATCTACATCCTCATCACCAC | |||
| AKR1B1 | F: ATGTGTACCAGAACGAGAACGAG | 192 | NM_001001539 |
| R: GGATAAGGTAGAGGTCCAGGTAGTC | |||
| CBR1 | F: AGGATCTATGCCAGAAAACTGAGAG | 198 | NM_214073 |
| R: GTCAGTAACAAACTGACCATGAGG | |||
| HPGD | F: GAACCTCAGAAGACTCTTTTCATCC | 217 | NM_001190248 |
| R: TTTGCTTACTCATGTAATCCAAACC | |||
| RPL7 | F: AAGCCAAGCACTATCACAAGGAATACA | 172 | NM_001113217 |
| R: TGCAACACCTTTCTGACCTTTGG |
| Gene name | Primer sequences (5′ → 3′)a | Product size (bp) | GenBank accession no. |
|---|---|---|---|
| PLA2G4A | F: TAAAGGAGTGCTATGTCTTCAAACC | 172 | XM_003130393 |
| R: GGGTCATCAAAGATATCAAAATCAG | |||
| PTGS1 | F: CAACACTTCACCCACCAGTTCTTC | 99 | AF207823 |
| R: TCCATAAATGTGGCCGAGGTCTAC | |||
| PTGS2 | F: TCGACCAGAGCAGAGAGATGAGAT | 134 | AF207824 |
| R: ACCATAGAGCGCTTCTAACTCTGC | |||
| PTGES | F: AGAGCATGAAAACCATCACTCC | 249 | NM_001038631 |
| R: CTCAAGGACATTCTGTCAGGTTC | |||
| PTGES2 | F: ATACCAGTACAAGACGTGTCCATTC | 195 | XM_001927936 |
| R: ATGATGACAGAGGAGTCATTCAGTT | |||
| PTGES3 | F: CAAATGATTCCAAGCATAAAAGAAC | 235 | XM_001929413 |
| R: GGTAAATCTACATCCTCATCACCAC | |||
| AKR1B1 | F: ATGTGTACCAGAACGAGAACGAG | 192 | NM_001001539 |
| R: GGATAAGGTAGAGGTCCAGGTAGTC | |||
| CBR1 | F: AGGATCTATGCCAGAAAACTGAGAG | 198 | NM_214073 |
| R: GTCAGTAACAAACTGACCATGAGG | |||
| HPGD | F: GAACCTCAGAAGACTCTTTTCATCC | 217 | NM_001190248 |
| R: TTTGCTTACTCATGTAATCCAAACC | |||
| RPL7 | F: AAGCCAAGCACTATCACAAGGAATACA | 172 | NM_001113217 |
| R: TGCAACACCTTTCTGACCTTTGG |
F, forward; R, reverse.
Summary of PCR primer sequences used for RT-PCR and real-time RT-PCR and expected product sizes.
| Gene name | Primer sequences (5′ → 3′)a | Product size (bp) | GenBank accession no. |
|---|---|---|---|
| PLA2G4A | F: TAAAGGAGTGCTATGTCTTCAAACC | 172 | XM_003130393 |
| R: GGGTCATCAAAGATATCAAAATCAG | |||
| PTGS1 | F: CAACACTTCACCCACCAGTTCTTC | 99 | AF207823 |
| R: TCCATAAATGTGGCCGAGGTCTAC | |||
| PTGS2 | F: TCGACCAGAGCAGAGAGATGAGAT | 134 | AF207824 |
| R: ACCATAGAGCGCTTCTAACTCTGC | |||
| PTGES | F: AGAGCATGAAAACCATCACTCC | 249 | NM_001038631 |
| R: CTCAAGGACATTCTGTCAGGTTC | |||
| PTGES2 | F: ATACCAGTACAAGACGTGTCCATTC | 195 | XM_001927936 |
| R: ATGATGACAGAGGAGTCATTCAGTT | |||
| PTGES3 | F: CAAATGATTCCAAGCATAAAAGAAC | 235 | XM_001929413 |
| R: GGTAAATCTACATCCTCATCACCAC | |||
| AKR1B1 | F: ATGTGTACCAGAACGAGAACGAG | 192 | NM_001001539 |
| R: GGATAAGGTAGAGGTCCAGGTAGTC | |||
| CBR1 | F: AGGATCTATGCCAGAAAACTGAGAG | 198 | NM_214073 |
| R: GTCAGTAACAAACTGACCATGAGG | |||
| HPGD | F: GAACCTCAGAAGACTCTTTTCATCC | 217 | NM_001190248 |
| R: TTTGCTTACTCATGTAATCCAAACC | |||
| RPL7 | F: AAGCCAAGCACTATCACAAGGAATACA | 172 | NM_001113217 |
| R: TGCAACACCTTTCTGACCTTTGG |
| Gene name | Primer sequences (5′ → 3′)a | Product size (bp) | GenBank accession no. |
|---|---|---|---|
| PLA2G4A | F: TAAAGGAGTGCTATGTCTTCAAACC | 172 | XM_003130393 |
| R: GGGTCATCAAAGATATCAAAATCAG | |||
| PTGS1 | F: CAACACTTCACCCACCAGTTCTTC | 99 | AF207823 |
| R: TCCATAAATGTGGCCGAGGTCTAC | |||
| PTGS2 | F: TCGACCAGAGCAGAGAGATGAGAT | 134 | AF207824 |
| R: ACCATAGAGCGCTTCTAACTCTGC | |||
| PTGES | F: AGAGCATGAAAACCATCACTCC | 249 | NM_001038631 |
| R: CTCAAGGACATTCTGTCAGGTTC | |||
| PTGES2 | F: ATACCAGTACAAGACGTGTCCATTC | 195 | XM_001927936 |
| R: ATGATGACAGAGGAGTCATTCAGTT | |||
| PTGES3 | F: CAAATGATTCCAAGCATAAAAGAAC | 235 | XM_001929413 |
| R: GGTAAATCTACATCCTCATCACCAC | |||
| AKR1B1 | F: ATGTGTACCAGAACGAGAACGAG | 192 | NM_001001539 |
| R: GGATAAGGTAGAGGTCCAGGTAGTC | |||
| CBR1 | F: AGGATCTATGCCAGAAAACTGAGAG | 198 | NM_214073 |
| R: GTCAGTAACAAACTGACCATGAGG | |||
| HPGD | F: GAACCTCAGAAGACTCTTTTCATCC | 217 | NM_001190248 |
| R: TTTGCTTACTCATGTAATCCAAACC | |||
| RPL7 | F: AAGCCAAGCACTATCACAAGGAATACA | 172 | NM_001113217 |
| R: TGCAACACCTTTCTGACCTTTGG |
F, forward; R, reverse.
Quantitative Real-Time RT-PCR
To analyze expression of transcripts in the uterine endometrium, real-time RT-PCR was performed using the Applied Biosystems StepOnePlus System (Applied Biosystems, Foster City, CA) using the SYBR Green method. Complementary DNAs were synthesized from 4 μg total RNA isolated from different uterine endometrial tissues, and newly synthesized cDNAs (total volume of 21 μl) were diluted 1:4 with sterile water and then used for PCR. The Power SYBR Green PCR Master Mix (Applied Biosystems) was used for PCR reactions. The final reaction volume of 20 μl included 2 μl of cDNA, 10 μl of 2× Master mix, 2 μl of each primer, and 4 μl of dH2O. PCR conditions and sequences of primer pairs are listed in Table 1. The results are reported as expression relative to that detected on Day 12 of the estrous cycle or that detected in control explant tissues after normalization of the transcript amount to the endogenous RPL7 control by the 2−ΔΔCT method [31].
Nonradioactive In Situ Hybridization
The nonradioactive in situ hybridization procedure was performed as described previously with some modifications [32]. Sections (5 μm thick) were rehydrated through successive baths of xylene, 100% ethanol, 95% ethanol, diethylpyrocarbonate (DEPC)-treated water, and DEPC-treated PBS. Tissue sections were boiled in citrate buffer (pH 6.0) for 10 min. After washing in DEPC-treated PBS, they were digested using 5 μg/ml Proteinase K (Sigma) in TE (100 mM Tris-HCl, 50 mM EDTA, pH 7.5) at 37°C. After postfixation in 4% paraformaldehyde, tissue sections were incubated twice for 15 min each in PBS containing 0.1% active DEPC and equilibrated for 15 min in 5× SSC. The sections were prehybridized for 2 h at 68°C in hybridization mix (50% formamide, 5× SSC, 500 μg/ml herring sperm DNA, 250 μg/ml yeast tRNA). Sense and antisense riboprobes for each gene were generated using partial cDNAs cloned into pCRII vectors by linearizing with appropriate restriction enzymes and labeling with digoxigenin (DIG)-UTP using a DIG RNA Labeling kit (Roche, Indianapolis, IN). The probes were denatured for 5 min at 80°C and added to the hybridization mix. The hybridization reaction was carried out overnight at 68°C. Prehybridization and hybridization reactions were performed in a box saturated with a 5× SSC 50% formamide solution to avoid evaporation, and no coverslips were used. After hybridization, sections were washed for 30 min in 2× SSC at room temperature, 1 h in 2× SSC at 65°C, and 1 h in 0.1× SSC at 65°C. Probes bound to the section were detected immunologically using sheep anti-DIG Fab fragments covalently coupled to alkaline phosphatase and nitro blue tetrazolium chloride/5-bromo-4-chloro-3-indolyl phosphate (toluidine salt) as a chromogenic substrate, according to the manufacturer's protocol (Roche).
Protein Isolation and Immunoblot Analysis
Endometrial explant tissues were homogenized in lysis buffer (1% Triton X-100, 0.5% Nonidet P-40, 150 mM NaCl, 10 mM Tris, 1 mM EDTA, 0.2 mM Na3VO3, 0.2 M PMSF, and 0.5 μg/ml NaF) at a ratio of 100 mg tissue:1 ml buffer, and cellular debris was removed by centrifugation (16 500 × g for 5 min). Protein concentrations in endometrial lysates were determined using the Bradford protein assay (Bio-Rad Laboratories, Richmond, CA) with bovine serum albumin as the standard. Twenty micrograms of endometrial protein lysate were loaded in each lane and electrophoresed on 12% SDS-polyacrylamide gels followed by electrotransfer onto nitrocellulose membranes. Nonspecific binding was blocked with 5% (w/v) fat-free milk in Tris-buffered saline with 0.1% (v/v) Tween-20 (TBST) buffer for 1 h at room temperature. The blot was incubated overnight at 4°C with 0.5 μg/ml rabbit polyclonal anti-AKR1B1 antibody (Abcam, Cambridge, U.K.) diluted in 2% milk-TBST. The blot was washed in TBST at room temperature three times for 10 min each, incubated with peroxidase-conjugated goat anti-rabbit secondary antibody (1:20 000 dilution; Jackson Laboratories, West Grove, PA) for 1 h at room temperature, and rinsed again for 30 min at room temperature with TBST. Immunoreactive proteins were detected by chemiluminescence (SuperSignal West Pico; Pierce Chemical Co., Rockford, IL) according to the manufacturer's recommendations, using x-ray film (Agfa-Gevaert). Blots were reblotted with rabbit polyclonal anti-β-actin (ACTB) antibody (1:5000 dilution; Sigma) to assess for consistent loading. The integrated optical density of AKR1B1 and ACTB bands in the immunoblots was quantified by scanning densitometry using an HP1210 model scanner (HP, Seoul, Korea) and a GelPro analyzer (Media Cybernetics, Silver Spring, MD). Values are presented as the ratio of each AKR1B1 signal to the corresponding ACTB signal.
Immunohistochemical Analysis
To determine which type of cells in the porcine endometrium expresses proteins, immunohistochemistry was applied. Sections (5 μm thick) were deparaffinized and rehydrated in an alcohol gradient. Tissue sections were washed with PBS, and the peroxidase block was performed with 0.5% H2O2 in methanol for 30 min. Tissue sections were then blocked with 10% normal goat serum for 30 min at room temperature. Rabbit polyclonal anti-PTGES antibody (1:200; Youngin Frontier, Seoul, Korea), anti-AKR1B1 (1:80; Abcam), anti-CBR1 (1:100; Sigma), or anti-HPGD (1:400; Lifespan Biosciences, Seattle, WA) was added and incubated overnight at 4°C in a humidified chamber. For each tissue tested, normal rabbit IgG was substituted for primary antibody and served as a negative control. Tissue sections were washed with PBS. The biotinylated goat anti-rabbit secondary antibody (Vector Laboratories, Burlingame, CA) was added and incubated for 1 h at room temperature. Following washes with PBS, the streptavidin peroxidase conjugate (Invitrogen) was added to the tissue sections, and tissue sections were incubated for 10 min at room temperature. The sections were washed with PBS, and the 3-amino-9-ethylcarbozole in N,N-dimethylformamide color development substrate (Invitrogen) was added to the tissue sections, which were then incubated at room temperature. The tissue sections were washed in water, counterstained with Mayer hematoxylin, and coverslipped.
Analysis of Levels of PGE2 and/or PGF2α in Uterine Endometrial Tissues, Uterine Flushings, and Explant Tissue Culture Supernatants
Levels of PGE2 and/or PGF2α in the endometrial tissue lysates, uterine flushings, and explant tissue culture supernatants were measured using enzyme immunoassay (EIA) kits (Cayman Chemical, Ann Arbor, MI), according to manufacturer's instructions. Assay sensitivity for PGE2 was 8.7 pg/ml, and the intra- and interassay coefficients of variation were 3.7% and 8.3%, respectively. Assay sensitivity for PGF2α was 3.0 pg/ml, and the intra- and interassay coefficients of variation were 9.0% and 9.1%, respectively. The concentrations of PGs in the endometrial tissues were standardized per total protein content, and the amounts of PGE2 and PGF2α in the uterine flushings are total recoverable amounts of PGs. The amounts of PGF2α in explant culture supernatants were standardized per total explant tissue weight.
Statistical Analysis
Data from real-time RT-PCR for genes during the estrous cycle and pregnancy were subjected to least-squares ANOVA using the General Linear Models procedures of SAS (Cary, NC). As sources of variation, the model included day, pregnancy status (cyclic or pregnant, Days 12 and 15 postestrus), and their interactions to evaluate steady-state levels of mRNAs and effects of treatment and animal to evaluate effects of steroid hormones on AKR1B1 mRNA. Preplanned orthogonal contrasts (control vs. E2, control vs. P4, E2 vs. E2 + P4, E2 + P4 vs. E2 + P4 + ICI, and E2 + P4 vs. E2 + P4 + RU) were used to test for effects of treatments in the explant cultures. Effect of day of pregnancy (Days 12, 15, 30, 60, 90, and 114) for data from real-time RT-PCR for expression of PG metabolic enzymes and data from IL1B or EBPC dose-response studies were analyzed by least-squares regression analysis. Data are presented as least-squares means with SEM. Data from endometrial and uterine luminal PGE2 and PGF2α levels, ratios of PGE2 to PGF2α levels, and mRNA levels of PG synthases in the uterus with different conceptus morphology were subjected to the Student t-test procedures of SAS and are presented as means with SEM. A P-value of 0.05 or less was considered significant, whereas a P-value of 0.05–0.10 was considered a trend toward significance.
Results
Expression of mRNAs for Genes Involved in PG Metabolism in the Uterine Endometrium During the Estrous Cycle and Pregnancy in Pigs
To determine the steady-state levels of mRNAs for the genes involved in PG metabolism in the porcine uterine endometrium, we performed real-time RT-PCR analysis (Fig. 1). Expression of PLA2G4A, PTGS1, PTGS2, PTGES, PTGES2, PTGES3, and CBR1 was not affected by day, pregnancy status, or day × pregnancy status interaction on Days 12 and 15 postestrus. However, expression of AKR1B1 mRNA was affected (P < 0.001) by day, pregnancy status, and day × pregnancy status interaction on Days 12 and 15 postestrus, and expression of AKR1B1 mRNA was greater on Day 12 of pregnancy than on Day 12 of the estrous cycle (day × status, P < 0.001). However, HPGD expression was greater on Day 12 of the estrous cycle than Day 12 of pregnancy (day × status, P < 0.01).
Expression of PLA2G4A (A), PTGS1 (B), PTGS2 (C), PTGES (D), PTGES2 (E), PTGES3 (F), AKR1B1 (G), CBR1 (H), and HPGD (I) mRNAs in the porcine uterine endometrium during the estrous cycle and pregnancy. Endometrial tissue samples from cyclic and pregnant gilts were analyzed by real-time RT-PCR, and data are reported as expression relative to that detected on Day 12 of the estrous cycle after normalization of the transcript amount to the endogenous RPL7 control. Data are presented as least squares means with standard error.
During pregnancy, steady-state levels of PTGS1 (cubic effect of day, P < 0.001), PTGS2 (cubic effect of day, P < 0.05), PTGES (quadratic effect of day, P < 0.05), PTGES2 (quadratic effect of day, P < 0.001), AKR1B1 (cubic effect of day, P < 0.05), CBR1 (quadratic effect of day, P < 0.001), and HPGD (quadratic effect of day, P < 0.001) mRNAs changed during pregnancy. Interestingly, levels of AKR1B1 mRNA change in a biphasic manner during pregnancy, being highest on Day 12 of pregnancy and during late pregnancy, and changes in PTGS2 mRNA were also biphasic with higher expression during early and late pregnancy. Expression of PTGS1, PTGES, PTGES2, CBR1, and HPGD mRNAs were high during late pregnancy, but the expression of PLA2G4A and PTGES3 mRNAs was not affected by day of pregnancy.
Localization of mRNAs for Genes Involved in PG Metabolism in the Uterine Endometrium During the Estrous Cycle and Pregnancy in Pigs
To determine the cell types(s) expressing PG metabolic enzymes in the uterine endometrium, we performed in situ hybridization analysis. We focused on cellular localization of PTGES, AKR1B1, CBR1, and HPGD mRNAs (Fig. 2). Expression of PTGES mRNA was localized primarily to uterine luminal epithelia (LE) and glandular epithelia (GE) and chorionic membrane (CM) during pregnancy. AKR1B1 mRNAs were detected in the uterine LE on Day 12 of pregnancy and uterine LE and CM during mid- to late pregnancy with stronger signal intensity in the CM. CBR1 and HPGD mRNAs were localized primarily to uterine LE during mid- to late pregnancy and rarely detectable during early pregnancy.
In situ hybridization analysis of PTGES, AKR1B1, CBR1, and HPGD mRNAs in the uterine endometria from the estrous cycle and pregnancy in pigs. PTGES mRNA was localized primarily to epithelial cells in the endometrium and chorionic membrane during pregnancy. AKR1B1 mRNAs were detected in the luminal epithelial cells on Day 12 of pregnancy and chorionic epithelial cells during mid- to late pregnancy. CBR1 and HPGD mRNAs were localized primarily to uterine luminal epithelial cells during mid- to late pregnancy. Representative uterine sections from Days 12, 60, and 90 of pregnancy hybridized with a DIG-labeled sense PTGES, AKR1B1, CBR1, or HPGD cRNA probes (Sense) as a negative control are shown. D, day; C, estrous cycle; P, pregnancy; LE, luminal epithelial cell; GE, glandular epithelial cell; St, stroma; CM, chorionic membrane. Bar = 100 μm.
Localization of Proteins for Genes Involved in PG Metabolism in the Uterine Endometrium During the Estrous Cycle and Pregnancy in Pigs
Next, we determined cellular localization of PTGES, AKR1B1, CBR1, and HPGD proteins in the uterine endometrium by immunohistochemistry (Fig. 3). PTGES proteins were localized to uterine LE and GE and to CM, which is similar to that for PTGES mRNA. AKR1B1 proteins were detected in the uterine LE during early pregnancy and uterine LE and CM during mid- to late pregnancy. CBR1 and HPGD were localized primarily to uterine LE during mid- to late pregnancy. AKR1B1, CBR1, and HPGD proteins were localized to both the cytoplasm and the nucleus.
Immunohistochemical analysis of PTGES, AKR1B1, CBR1, and HPGD proteins in the uterine endometria from the estrous cycle and pregnancy in pigs. PTGES proteins were localized primarily to endometrial epithelial cells and chorionic membrane during pregnancy. AKR1B1 proteins were localized in the luminal epithelial cells on Day 12 of pregnancy and chorionic epithelial cells during mid- to late pregnancy. CBR1 and HPGD proteins were localized to luminal epithelial cells during mid- to late pregnancy. A higher magnification image of AKR1B1, CBR1, and HPGD proteins is shown in inset on Day 60 of pregnancy. D, day; C, estrous cycle; P, pregnancy; LE, luminal epithelial cell; GE, glandular epithelial cell, St, stroma; CM, chorionic membrane. Bars = 100 μm and 50 μm in inset.
Expression of mRNAs for Genes Involved in PG Metabolism in Conceptuses During Early Pregnancy
To determine whether conceptuses also express genes involved in PG metabolism, we performed RT-PCR using conceptus tissues from Days 12 and 15 of pregnancy (Fig. 4). Except for CBR1 mRNA, all genes studied were expressed in conceptuses during early pregnancy.
Expression of PLA2G4A, PTGS1, PTGS2, PTGES, PTGES2, PTGES3, AKR1B1, CBR1, and HPGD mRNAs in conceptuses from Days (D) 12 and 15 of pregnancy by RT-PCR. RPL7 was used as a positive control. Except for CBR1 mRNA, all PG metabolic enzymes tested were detected in conceptuses. RTase +/−, with (+) or without (−) reverse transcriptase; M, molecular marker; D12P Endo, endometrium on Day 12 of pregnancy; D15P Endo, endometrium on Day 15 of pregnancy; D12 Con, Day 12 conceptus; D15 Con, Day 15 conceptus.
Amounts of PGE2 and PGF2α in Uterine Endometrial Tissues and Uterine Lumen on Day 12 of the Estrous Cycle and Pregnancy
Having determined that PG metabolic enzymes are expressed in the uterine endometrium during pregnancy and in the conceptus during early pregnancy, we next questioned whether the levels of expression of PG metabolic enzymes reflect production of PGE2 and PGF2α in the uterine endometrium and in the uterine lumen during early pregnancy. Thus, we measured levels of PGE2 and PGF2α in the endometrial tissue lysates and total recoverable amounts of PGE2 and PGF2α in uterine flushings on Day 12 of the estrous cycle and pregnancy by EIA (Fig. 5).
Levels of PGE2 (A and B) and PGF2α (C and D) in uterine endometrial tissues and uterine lumen on Day 12 of the estrous cycle and pregnancy and the ratio of PGE2 to PGF2α in the uterine endometrial tissues (E) and uterine lumen (F). Endometrial tissue extracts and uterine flushings were obtained on Day 12 of the estrous cycle and the amounts of PGE2 and PGF2α determined by enzyme immunoassay. The PGE2 and PGF2α concentrations in the endometrial tissues were standardized per total protein content, and the amounts of PGE2 and PGF2α in the uterine flushings are total recoverable amounts of PGs. The ratios of PGE2 to PGF2α in the uterine endometrial tissues and uterine flushings were determined for uterine endometrial tissues and uterine flushings, respectively. Data were analyzed by the Student t-test procedures of SAS and presented as means with standard errors (*P < 0.05).
The abundance of PGE2 was greater in both endometrial tissues (P < 0.05) and uterine flushings (P = 0.078) on Day 12 of pregnancy than on Day 12 of the estrous cycle (Fig. 5, A and B). Similarly, amounts of PGF2α were also higher in both the endometrial tissues (P < 0.01) and uterine flushings (P < 0.05) on Day 12 of pregnancy than on Day 12 of the estrous cycle (Fig. 5, C and D).
We next determined differences in the ratio of PGE2 to PGF2α in the uterine endometrium and uterine lumen on Day 12 of the estrous cycle and pregnancy. Interestingly, the ratio of PGE2 to PGF2α in the uterine endometrial tissues on Day 12 of the estrous cycle was higher than that on Day 12 of pregnancy (P < 0.05; Fig. 5E), whereas the ratio of PGE2 to PGF2α in uterine flushings on Day 12 of pregnancy was higher than that on Day 12 of the estrous cycle (P = 0.069; Fig. 5F).
Analysis of Levels of PTGES, PTGES2, PTGES3, and AKR1B1 mRNAs in the Uterine Endometrium and Levels of PGE2 and PGF2α in the Uterine Endometrium and Uterine Lumen with Spherical and Filamentous Conceptuses
Because secretion of E2 and IL1B is significantly higher for filamentous than spherical conceptuses [4, 33] and AKR1B1 mRNA levels in the uterine endometrium were increased by E2 in explant cultures in this study, we next determined if expression of PGE synthases PTGES, PTGES2, and PTGES3 and PGF synthase AKR1B1 and amounts of PGE2 and PGF2α in the uterine endometrium and uterine lumen were affected by the presence of conceptuses with different morphologies. Levels of PTGES, PTGES2, PTGES3, and AKR1B1 mRNAs were analyzed in the uterine endometrium with spherical conceptuses from Day 11 of pregnancy and with filamentous conceptuses from Day 12 of pregnancy using real-time RT-PCR (Fig. 6, A–D). Expression of PTGES mRNA was lower in endometria with filamentous conceptuses than in endometria with spherical conceptuses (P < 0.001; Fig. 6A), whereas expression of PTGES3 and AKR1B1 mRNAs was greater for endometria with filamentous conceptuses (P < 0.01 for both PTGES3 and AKR1B1; Fig. 6, C and D). PTGES2 mRNA levels were different (Fig. 6B).
Expression of endometrial PTGES (A), PTGES2 (B), PTGES3 (C), and AKR1B1 (D) mRNAs, levels of PGE2 (E and F) and PGF2α (G and H) in the uterine endometrial tissues and uterine lumen and the ratio of PGE2 to PGF2α in the uterine endometrial tissues (I) and uterine lumen (J) in the uterine endometrium associated with spherical and filamentous conceptuses. Levels of endometrial PTGES, PTGES2, PTGES3, and AKR1B1 mRNAs and the amounts of PGE2 and PGF2α in endometrial tissues and uterine flushings were analyzed by real-time RT-PCR and enzyme immunoassay, respectively. Abundance of PTGES, PTGES2, PTGES3, and AKR1B1 mRNAs is presented as expression relative to that in endometria with spherical conceptuses after normalization of the transcript amount to RPL7. The PGE2 and PGF2α concentrations in the endometrial tissues were standardized per total protein content, and the amounts of PGE2 and PGF2α in the uterine flushings are total recoverable amounts of PGs. The ratios of PGE2 to PGF2α in uterine endometrial tissues and uterine flushings were determined by amounts of PGE2 and PGF2α in the uterine endometrial tissues and uterine flushings, respectively. Data were analyzed by the Student t-test procedures of SAS and are presented as means with standard errors (*P < 0.05).
Next, we analyzed amounts of PGE2 and PGF2α in endometrial tissue lysates and total recoverable amounts of PGE2 and PGF2α in uterine flushings from the uterine endometrium with spherical and filamentous conceptuses by EIA (Fig. 6, E–H). PGE2 levels were higher in the uterine endometrium (P < 0.001) and in uterine flushings (P = 0.065) with filamentous conceptuses than those with spherical conceptuses (Fig. 6, E and F). The amounts of PGF2α were higher in both endometrial tissues (P < 0.001) and uterine flushings (P = 0.083) of uteri with filamentous as compared to spherical conceptuses (Fig. 6, G and H).
We further analyzed the ratio of PGE2 to PGF2α in the uterine endometrial tissues and in uterine flushings from uteri with spherical and filamentous conceptuses. As shown in Figure 6, I and J, the ratio of PGE2 to PGF2α levels was not different in uterine endometrial tissues (P > 0.10) and uterine flushings (P > 0.10) from uteri with spherical as compared to filamentous conceptuses.
Effects of E2, P4, and IL1B on AKR1B1 Expression in Uterine Endometrium from Day 12 of the Estrous Cycle
Because expression levels of AKR1B1 mRNAs in the uterine endometrium were highest on Day 12 of pregnancy, we further analyzed factors that regulate AKR1B1 expression. Expression of many endometrial genes is regulated by E2 and IL1B of conceptus origin and/or P4 from CL at the time of implantation; therefore, we hypothesized that E2, IL1B, and/or P4 affect expression of AKR1B1 in the uterine endometrium. We treated endometrial explant tissues from Day 12 of the estrous cycle with control; E2; P4; E2 + P4; E2 + P4 + ICI, an estrogen receptor antagonist; or E2 + P4 + RU, a progesterone receptor antagonist, and with 0, 1, 10, or 100 ng/ml of IL1B in the presence of both E2 and P4. As shown in Figure 7A, AKR1B1 mRNA abundance increased in response to E2 (control vs. E2; P < 0.001) but not to P4 (control vs. P4; P > 0.10). The E2-induced increase in AKR1B1 mRNA was inhibited by ICI, an estrogen receptor antagonist (E2 + P4 vs. E2 + P4 + ICI; P < 0.001). There was no effect of P4 on AKR1B1 mRNA levels when P4 was combined with E2 (E2 vs. E2 + P4; P > 0.10), and E2-induced AKR1B1 mRNA levels were not affected by the presence of RU, a progesterone receptor antagonist (E2 + P4 vs. E2 + P4 + RU; P > 0.10). As shown in Figure 7B, AKR1B1 mRNA levels were affected by increasing doses of IL1B (linear effect of dose, P < 0.05).
Effect of steroid hormones (A) and IL1B (B) on AKR1B1 mRNA levels in porcine uterine endometrial explant cultures. Endometrial explants from gilts on Day 12 of the estrous cycle were cultured in DMEM/F-12 in the presence of control (C), E2 (E; 50 ng/ml), P4 (P; 3 ng/ml), E2 + P4 (PE), E2 + P4 + ICI (I; 200 ng/ml, an estrogen receptor antagonist) (PEI), or E2 + P4 + RU (R; 30 ng/ml, a progesterone receptor antagonist) (PER) at 37°C for 24 h or with 0, 1, 10, or 100 ng/ml IL1B in the presence of both E2 (50 ng/ml) and P4 (3 ng/ml) at 37°C for 24 h. Abundance of mRNA expression determined by real-time RT-PCR analyses is relative to that for AKR1B1 mRNAs in the control group of endometrial explants after normalization of transcript amounts to RPL7 mRNA. Data are presented as least-squares means with standard errors (*P < 0.05). For each treatment, all experiments were repeated in triplicate with endometrium from each of three gilts.
Effects of EBPC on PGF2α Production in the Uterine Endometrium from Day 12 of the Estrous Cycle
Since AKR1B1 is a major PGF synthase in human and cow endometria [24–26] and its levels of expression were greater on Day 12 of pregnancy than Day 12 of the estrous cycle (Fig. 1G), we determined if AKR1B1 also acts as a PGF synthase in the uterine endometrium during the peri-implantation period of pregnancy in pigs. We treated endometrial explant tissues from Day 12 of the estrous cycle with increasing doses of EBPC, an inhibitor of AKR1B1 activity, in the presence of both E2 and P4 and measured PGF2α concentrations in culture supernatants by EIA. Increasing doses of EBPC decreased production of PGF2α in endometrial explant cultures in a dose-dependent manner (linear effect of dose, P < 0.01; Fig. 8A). To confirm that the decrease in PGF2α in culture supernatants was not caused by a decrease in expression of AKR1B1 mRNA and AKR1B1 protein, we analyzed for AKR1B1 mRNA and AKR1B1 protein in endometrial explant tissues by real-time RT-PCR and immunoblot analysis, respectively. The expression of AKR1B1 mRNA and AKR1B1 protein was not affected by EBPC treatment (P > 0.10; Fig. 8, B and C).
![Effect of EBPC, a blocker of AKR1B1, on amounts of PGF2α in explant culture supernatants (A), AKR1B1 mRNA levels (B), and AKR1B1 protein levels (C) in porcine uterine endometrial explant cultures. Endometrial explants from gilts on Day 12 of the estrous cycle were cultured in DMEM/F-12 with 0, 1, 10, or 20 ng/ml EBPC (ethyl 1-benzyl-3-hydroxy-2[5H]-oxopyrrole-4-carboxylate) in the presence of both E2 (50 ng/ml) and P4 (3 ng/ml) at 37°C for 24 h. The amounts of PGF2α in explant culture supernatants were analyzed by enzyme immunoasay and standardized per total amount of explant tissue. Levels of AKR1B1 mRNA and protein in endometrial explant tissues were determined by real-time RT-PCR and immunoblot analysis, respectively. Abundance of mRNA expression determined by real-time RT-PCR analyses is relative to that for AKR1B1 mRNAs in the control group of endometrial explants (0 μM EBPC) after normalization of transcript amounts to RPL7 mRNA. The ratio of AKR1B1 protein density to ACTB density was obtained by scanning densitometry. Data are presented as least-squares means with standard errors. These experiments were conducted using endometria from three gilts on Day 12 of the estrous cycle, and treatments were performed in triplicate using tissues obtained from each of the three gilts.](https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/biolreprod/90/5/10.1095_biolreprod.113.114926/1/m_i0006-3363-90-5-99-f08.jpeg?Expires=1712723788&Signature=n0OGzYHM8MC3j4L07mG98Y3mbzWdLRdV9MBoWjF-FiI2CBh1SMasLgxROXn3zz6gOxgsn5PYUwH17qvif1l6CWI6bXUuAAQPb5f1KwPXSNU51tUynAVF~L224tDpeSkpcaYfy8gm6g1z13rjBG1ZCCX94CYOw45-NPEVtsuxEpbb0wbY2Ic~Mj-OI~MWakgssO2qcu~YBEHcEnL7W0v3MqwCIHIxVZ6cko3PoDvQCnY5lNc39iaLmy9PhA1ENt6ittfv66GWaH~y5V~4ze-7KduuFABW~GJAQAwAr~XkGenIM1XyxstbbcKtTCDDLjC2x-FVepjYF-xxd4s7nU6j3w__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA)
Effect of EBPC, a blocker of AKR1B1, on amounts of PGF2α in explant culture supernatants (A), AKR1B1 mRNA levels (B), and AKR1B1 protein levels (C) in porcine uterine endometrial explant cultures. Endometrial explants from gilts on Day 12 of the estrous cycle were cultured in DMEM/F-12 with 0, 1, 10, or 20 ng/ml EBPC (ethyl 1-benzyl-3-hydroxy-2[5H]-oxopyrrole-4-carboxylate) in the presence of both E2 (50 ng/ml) and P4 (3 ng/ml) at 37°C for 24 h. The amounts of PGF2α in explant culture supernatants were analyzed by enzyme immunoasay and standardized per total amount of explant tissue. Levels of AKR1B1 mRNA and protein in endometrial explant tissues were determined by real-time RT-PCR and immunoblot analysis, respectively. Abundance of mRNA expression determined by real-time RT-PCR analyses is relative to that for AKR1B1 mRNAs in the control group of endometrial explants (0 μM EBPC) after normalization of transcript amounts to RPL7 mRNA. The ratio of AKR1B1 protein density to ACTB density was obtained by scanning densitometry. Data are presented as least-squares means with standard errors. These experiments were conducted using endometria from three gilts on Day 12 of the estrous cycle, and treatments were performed in triplicate using tissues obtained from each of the three gilts.
Discussion
The novel findings of this study with pigs are that 1) genes involved in PGE2 and PGF2α metabolism are differentially expressed in a pregnancy-stage- and cell-type-specific manner in the uterine endometrium during pregnancy; 2) conceptuses during early stages of pregnancy express genes involved in PG metabolism, except CBR1; 3) AKR1B1 expression in the uterine endometrium is increased by estrogen and IL1B of conceptus origin; and 4) AKR1B1 has PGF synthase activity in the uterine endometrium. This study also confirmed previous findings that levels of both PGE2 and PGF2α in uterine endometrial tissues and uterine lumen are higher on Day 12 of pregnancy than on Day 12 of the estrous cycle and tht levels of both PGE2 and PGF2α in uterine endometrial tissues and uterine lumen increase with conceptus elongation. This is a comprehensive report characterizing expression of genes determining PGE2 and PGF2α production in the uterine endometrium during pregnancy and provides initial evidence that AKR1B1 is a PGF synthase in the uterine endometrium during the implantation period in pigs.
PGs are synthesized by coordinate actions of multiple enzymes within the cell, and the local actions of PGs are terminated by catabolic enzymes within the cells affected by PGs [14, 15]. Thus, tight regulation of PG levels is critical for normal female reproductive functions. The importance of PG functions in the female reproductive tract during pregnancy is well known. PTGS1-deficent mice show delayed parturition [34], and disruption of the PTGS2 gene in mice results in multiple reproductive problems in ovulation, fertilization, implantation, and decidualization [35]. In humans, dysregulated endometrial PG production is associated with menstrual cycle disorders, infertility, and uterine malignancies [35–38]. PGs regulate endometrial gene expression and conceptus development during the implantation period in sheep [11]. In pigs, inhibition of PG synthesis causes pregnancy failure before the implantation process [13]. Although there are some reports investigating expression of PG-metabolizing enzymes, such as PTGS1, PTGS2, and PTGES, in the uterine endometrium during early pregnancy in pigs [9, 21, 22], results of the present study provide a more comprehensive evaluation of expression of genes for enzymes involved in PGE2 and PGF2α metabolism in the uterine endometrium during early and late stages of pregnancy.
Results of the present study showed that genes responsible for PGE2 and PGF2α metabolism are expressed in the uterine endometrium during pregnancy with specific expression patterns; PLA2G4A and PTGES3 are expressed constitutively, while PTGS1, CBR1, HPGD, PTGES, and PTGES2 are highly expressed during late stages of pregnancy, and PTGS2 and AKR1B1 are expressed in a biphasic pattern, with expression being high during early and late stages of pregnancy. PGF2α is essential for initiation of parturition and induction of luteolysis to regress CL and eliminate their production of progesterone, and PGE2 plays an important role in production of endometrial cytokines and myometrial contractions [39, 40]. In mice, disruption of the Ptgs1 gene delays parturition with neonatal death and persistent secretion of progesterone due to failure of luteolysis [34], and disruption of the Hpgd gene results in early labor and an early increase in secretion of PGF2α [41]. Thus, overall expression patterns of expression of some enzymes responsible for metabolizing PGE2 and PGF2α are high at mid- to late pregnancy in the endometrium, indicating that actions of PGE2 and PGF2α may be critical for placental/fetal development and preparation for parturition. Interestingly, levels of CBR1 expression increased more than one million-fold during mid- to late pregnancy as compared to early pregnancy. CBR1 converts PGE2 to PGF2α and is also involved in inactivation of PGE2 and PGF2α and their conversion to PG metabolites [16]. Since levels of endometrial PGF2α do not increase until term (Seo et al., unpublished data) and CBR1 has PG-inactivating reductase activity [16], it is likely that CBR1 and HPGD, both expressed in uterine LE during mid- to late pregnancy, contribute to rapid inactivation of PGE2 and PGF2α synthesized through the actions of PTGS1, PTGS2, and PGE and PGF synthases during those periods of pregnancy. The detailed functions of PGE2 and PGF2α produced during mid- to late pregnancy are not fully understood, and this is an interesting area for further study.
Cellular localization of the genes regulating PGE2 and PGF2α metabolism in the porcine endometrium showed a unique cell-specific expression pattern. Since PTGS1 and PTGS2 mRNAs are localized primarily to endometrial epithelial cells in pigs [21, 22], we focused on localization of expression of other genes that have not been studied. PTGES mRNA and protein are localized primarily to uterine LE and GE and CM during pregnancy. Expression of AKR1B1 mRNA and protein was exclusive to uterine LE on Day 12 of pregnancy and to uterine LE and CM during mid- to late pregnancy with strong expression in CM. CBR1 and HPGD mRNAs and proteins were localized primarily to uterine LE with high levels of expression only during mid- to late pregnancy. These findings were somewhat different from the results shown in other species. In mice, Cbr1 mRNA is localized to uterine GE until Day 4 of pregnancy but not after embryo implantation [19], and endometrial HPGD expression is detected in uterine GE in cattle [20] and placental trophoblast in humans [42], suggesting that the mechanism to regulate PGE2 and PGF2α metabolism at the maternal-fetal interface during pregnancy may be species specific.
In this study we determined that amounts of both PGE2 and PGF2α in uterine endometrial tissues and uterine flushings are higher on Day 12 of pregnancy than the estrous cycle and in uteri with filamentous conceptuses as compared to spherical conceptuses. These results are in agreement with the report that amounts of PGE2 and PGF2α in utero-ovarian vein increase transiently on Days 11–13 of pregnancy [8], suggesting that production of both PGE2 and PGF2α in the uterus increases during this period. In addition, we found that the amounts of endometrial PGE2 were similar to those of PGF2α on Day 12 of pregnancy, while total recoverable amounts of PGE2 in uterine flushings were much higher than those of PGF2α on Day 12 of pregnancy, although direct comparisons were not made between the levels of PGE2 and PGF2α in endometrial tissues and uterine flushings. These results indicate that levels of both PGE2 and PGF2α produced by the endometrium on Day 12 of pregnancy are similar, but amounts of PGE2 produced by conceptuses may be much higher than for PGF2α. Indeed, levels of PGE2 in conceptus tissues are higher than those of PGF2α on Day 12 of pregnancy [43], and levels of PGE2 produced in vitro by conceptuses recovered from Day 10 of pregnancy were higher than those of PGF2α in pigs [44]. In this study, expression of PGE synthases, PTGES, PTGES2, and PTGES3 in conceptuses on Days 12 and 15 of pregnancy were also confirmed.
Because it has been proposed that an increase in the ratio of PGE2 to PGF2α in utero-ovarian vein and in uterine flushings may be important for the establishment of pregnancy in pigs [7–10], we measured the ratio of PGE2 to PGF2α levels in uterine endometrial tissues and uterine flushings on Day 12 of the estrous cycle and pregnancy. Results showed that the ratio of PGE2 to PGF2α in uterine flushings was higher on Day 12 of pregnancy than the estrous cycle, whereas the ratio in endometrial tissues was lower on Day 12 of pregnancy than the estrous cycle. The lower ratio of PGE2 to PGF2α in endometrial tissues but the higher ratio in utero-ovarian vein and uterine flushings on Day 12 of pregnancy as compared to the estrous cycle suggests that the rate of elimination of PGE2 from the endometrium into uterine lumen or into the utero-ovarian vein is much higher than for PGF2α on Day 12 of pregnancy. Also, these results indicate that there is likely an active mechanism to transport uterine luminal or endometrial PGE2 produced by conceptuses and endometrium into the utero-ovarian vein. Indeed, we found that prostaglandin transporters ABCC4 (ATP-binding cassette, subfamily C, member 4; also called MRP4) and SLCO2A1 (solute carrier organic anion transporter family, member 2A1; also called PGT) are expressed in the uterine endometrium on Day 12 of pregnancy and that their expression is up-regulated by IL1B in the uterine endometrium [45]. IL1B is a major cytokine produced by filamentous pig conceptuses [33]. If the increased ratio of PGE2 to PGF2α in the uterine lumen and in utero-ovarian vein or simply the increased amounts of PGE2 and reduced amounts of PGF2α in utero-ovarian vein are critical for preventing luteolysis and ensuring the establishment of pregnancy, the cellular and molecular mechanisms responsible for actions of PGE2 in the ovary remain to be clarified. Nevertheless, it is clear that production and secretion of both PGE2 and PGF2α increase in the uterine endometrium and conceptuses during the peri-implantation period of pregnancy.
Expression of PGE synthases PTGES, PTGES2, and PTGES3 and PGF synthase AKR1B1 and levels of PGE2 and PGF2α in the uterine endometrium and uterine lumen were affected by the presence of conceptuses and stage of development of the conceptuses, which is spherical or filamentous morphology. In particular, expression of PTGES3 and AKR1B1 increased in the endometrium with filamentous conceptuses as compared with spherical conceptuses, and estrogen or other factors from filamentous conceptuses may be responsible for increased levels of PGE2 and PGF2α in the uterine endometrium and uterine lumen. Since conceptuses undergo a dramatic morphological change from spherical to filamentous forms at the time of implantation and filamentous conceptuses secrete E2 and IL1B [4, 33] to signal maternal recognition of pregnancy in pigs, these results suggest that factors, such as E2 and IL1B, derived from conceptus affect the endometrial production of PGE2 and PGF2α. The ratios of PGE2 to PGF2α in uterine endometrial tissues and uterine flushings were not different between uteri with spherical and filamentous conceptuses. These results suggest that relative amounts of PGE2 to PGF2α produced by the endometrium and conceptuses were not different due to morphology of conceptuses. Porcine conceptuses do produce PGE2 and PGF2α from Day 7 of pregnancy [44], and levels of PGE2 increase in the uterine lumen, beginning with spherical conceptuses [4].
Results of the current study showed that AKR1B1 was expressed in uterine endometrial epithelial cells and that it had PGF synthase activity. It has been suggested that an AKR1C family enzyme is a PGF synthase responsible for endometrial PGF2α production at the time of maternal recognition of pregnancy in pigs [9]. However, PGF synthase activity of the AKR1C family enzyme was not determined in that study [9], and expression of the enzyme did not increase in the uterine endometrium in response to estrogen or IL1B [46]. Thus, those results do not demonstrate that the enzyme is responsible for increased levels of PGF2α production in uterine endometria at the time of maternal recognition of pregnancy and suggest that other PGF synthases responsible for PGF2α production may exist in the uterine endometrium at the time of implantation in pigs. Interestingly, AKR1B1 is a major PGF synthase responsible for PGF2α synthesis from PGH2 in bovine and human uterine endometria [14–16]. In addition, Ross et al. [27] reported that AKR1B1 is expressed in the uterine endometrium at the time of implantation and speculated that AKR1B1 may be involved in glucose metabolism. Thus, we hypothesize that AKR1B1 is a major PGF synthase responsible for estrogen-induced PGF2α production in the uterine endometrium at the time of conceptus implantation in pigs. Indeed, results of the present study showed that endometrial AKR1B1 expression on Day 12 of pregnancy was more than 50-fold higher than on Day 12 of the estrous cycle, which is the time when the levels of PGF2α dramatically increase in the uterine lumen of pregnant pigs. This is in agreement with the results of Samborski et al. [47] as determined by RNA-Seq analysis. In addition, AKR1B1 expression in explant culture studies was increased by estrogen as well as IL1B, and the levels of PGF2α in explant culture supernatants were reduced by blocking AKR1B1 enzymatic activity. These results indicate that AKR1B1 is a major PGF synthase induced by estrogen and IL1B of conceptus origin and responsible for increased PGF2α production in the uterine endometrium during the period of maternal recognition of pregnancy in pigs. Remarkably, AKR1B1 expression was barely detectable on Day 15 of the estrous cycle, when levels of luteolytic PGF2α increase in the utero-ovarian vein in pigs [8]. Thus, it is likely that different types of PGF synthase(s) other than AKR1B1 may be responsible for production of luteolytic PGF2α in the uterine endometrium during late diestrus, but this needs to be further elucidated.
In conclusion, this study in pigs showed that the PGE2 and PGF2α metabolic enzymes are expressed in the endometrium during pregnancy in a cell-type and stage of pregnancy-specific manner. The abundance of both PGE2 and PGF2α increase in the uterine endometrium and uterine lumen at the time of implantation, and AKR1B1 acts as a major PGF synthase in the endometrium during early pregnancy. These results suggest that PGE2 and PGF2α actions at the maternal-fetal interface are essential and tightly regulated for the establishment and maintenance of pregnancy. Although the molecular and cellular mechanisms of actions of PGE and PGF to effect luteolysis during the estrous cycle and to effect functions at the maternal-fetal interface during pregnancy are not completely understood, results of this study advance understanding of the expression of genes for synthesis and metabolism of PGE2 and PGF2α.
Acknowledgment
We thank Dr. Fuller W. Bazer, Texas A&M University, for critical reading and valuable suggestions for the manuscript.
References
Author notes
Supported by the Next Generation BioGreen 21 Program (#PJ009610), Rural Development Administration, and by the Yonsei University Research Fund of 2012 (H. Seo), Republic of Korea.
