Temporal and spatial expression of adrenomedullin and its receptors in the porcine uterus and peri-implantation conceptuses

Abstract

Adrenomedullin (ADM) is an evolutionarily conserved multifunctional peptide hormone that regulates implantation, embryo spacing, and placentation in humans and rodents. However, the potential roles of ADM in implantation and placentation in pigs, as a litter-bearing species, are not known. This study determined abundances of ADM in uterine luminal fluid, and the patterns of expression of ADM and its receptor components (CALCRL, RAMP2, RAMP3, and ACKR3) in uteri from cyclic and pregnant gilts, as well as conceptuses (embryonic/fetus and its extra-embryonic membranes) during the peri-implantation period of pregnancy. Total recoverable ADM was greater in the uterine fluid of pregnant compared with cyclic gilts between Days 10 and 16 post-estrus and was from uterine luminal epithelial (LE) and conceptus trophectoderm (Tr) cells. Uterine expression of CALCRL, RAMP2, and ACKR3 were affected by day (P < 0.05), pregnant status (P < 0.01) and/or day x status (P < 0.05). Within porcine conceptuses, the expression of CALCRL, RAMP2, and ACKR3 increased between Days 10 and 16 of pregnancy. Using an established porcine trophectoderm (pTr1) cell line, it was determined that 10⁻⁷ M ADM stimulated proliferation of pTr1 cells (P < 0.05) at 48 h, and increased phosphorylated mechanistic target of rapamycin (p-MTOR) and 4E binding protein 1 (p-4EBP1) by 6.1- and 4.9-fold (P < 0.0001), respectively. These novel results indicate a significant role for ADM in uterine receptivity for implantation and conceptus growth and development in pigs. They also provide a framework for future studies of ADM signaling to affect proliferation and migration of Tr cells, spacing of blastocysts, implantation, and placentation in pigs.

Our study suggests that adrenomedullin plays a significant role in uterine receptivity and conceptus growth and development, via the stimulation of cell proliferation and activation of MTORC1 signaling pathway in the porcine conceptus trophectoderm.

Introduction

Advances in modern swine production have led to increased litter sizes from females with greater prolificacy [1–3]. In the United States, the average litter size of female pigs has increased from 7.35 to 10.50 between 1969 and 2001 (USDA 2001) [4, 5]. This is due primarily to selection for ovulation rate in sows and semen quality in boars, to achieve fertilization rates exceeding 95% [6]. However, the average current litter size of pigs seems stable at 10.79 (USDA 2019) [4] because large litters in commercial breeding herds often increase the birth of “runt” piglets due to intrauterine growth restriction (IUGR; defined as impaired growth and development of fetuses and their organs) leading to lower rates of survival of neonatal piglets [7–16]. Among all mammals, the pig exhibits the most severe naturally occurring IUGR (15–25%) [8, 17, 18] due to the increased fetal crowding during gestation and limited uterine capacity [5, 6, 19–22]. Uterine capacity is the result of the combined effects of uterine, placental, and fetal functions established as early as Days 12–30 of gestation [21, 22] and continuing to Days 30–40 of gestation, the critical period of placentation [23–25]. Unlike primates and rodents with hemochorial placenta, ungulates (e.g. pigs and ruminants) have epitheliochorial placentae that require rapid morphological changes in conceptuses (embryo/fetus and its extra-embryonic membranes) as they elongate during the peri-implantation period of pregnancy. Porcine conceptuses, in particular, undergo rapid transitions from spherical (Days 10 and 11, 9–10 mm), to ovoid and/or tubular (Day 11, 11–50 mm), to expanding filamentous forms on Day 12 (100 mm), Days 13 to 14 (100–200 mm) and Days 14 to 16 (800–1000 mm), and implant to the uterus between Days 13 and 20 of gestation [26–29]. This process is highly dependent on the composition of histotroph [30–35], i.e. secretions from luminal (LE), superficial glandular (sGE), and glandular (GE) epithelia, as well as selective transport of nutrients into the uterine lumen, including enzymes, growth factors, adhesion proteins, cytokines, hormones, amino acids, and glucose [36]. Notably, 30% of porcine conceptuses arrest during the same time frame, increasing to 40% by Day 30 of gestation [6]. Thus, understanding the fundamental roles of conceptus- and uterine-secreted factors that govern synchronous uterine and conceptus development required for implantation, establishment of pregnancy, and placental development is a prerequisite for developing strategies to increase placental and fetal growth to improve overall health and survivability of piglets in utero and during the neonatal period of life.

Adrenomedullin (ADM) is an evolutionarily conserved multifunctional peptide involved in a wide variety of physiological processes including angiogenesis and cardiovascular homeostasis [37, 38]. It was first isolated from human phaeochromocytoma, i.e. neuroendocrinal tumor of adrenal gland medullae, hence its name, and demonstrated to increase cyclic adenosine monophosphate (cAMP) production in platelets of rats [39]. ADM is comprised of 52 amino acids in humans and pigs (with only one amino acid difference; Supplemental Figure S1), and 50 amino acids in mice and rats. ADM initiates signal transduction through receptor complexes consisting of the G-protein coupled receptor calcitonin-receptor-like receptor (CALCRL) and one of the receptor activity-modifying proteins (RAMPs) that include RAMP2 and RAMP3. Heterodimerization of RAMP2 or RAMP3 with CALCRL results in specific ADM receptors: ADM receptor 1 (ADM₁; CALCRL/RAMP2) and ADM receptor 2 (ADM₂; CALCRL/RAMP3) [40, 41]. Combinations of RAMP1 with CALCRL, however, form calcitonin gene-related peptide (CGRP) receptors, which can also bind ADM with much lower affinity [42]. On the other hand, atypical chemokine receptor 3 (ACKR3) is a decoy receptor serving as a cell-autonomous molecular rheostat to dampen ADM signaling [43]. ADM and its receptors are highly expressed in reproductive tissues of humans and rodents, including the uterine endometrium [44], fetal membranes/placentae [45], stromal macrophages [46], and trophoblast cells [47–50]. In mice, ADM and its receptors are expressed in uterine LE by gestational day 0.5, and ADM is expressed by both conceptus trophectoderm, and uterine LE and stromal cells at implantation sites during the peri-implantation period of pregnancy [51]. Due to the embryonic lethality in homozygous Adm-null mice, heterozygous Adm^+/− (50% ADM expression) female mice have been studied [52] and found to have a significantly reduced rate of pregnancy success compared to wild type females, despite having normal rates of ovulation and fertilization [53, 54]. The implantation sites in pregnant Adm^+/− females are spaced abnormally leading to crowding of implantation sites, increased rates of embryo loss, and reduced prolificacy [54]. The lower pregnancy rate persists even when wild-type conceptuses are transferred to Adm^+/− females, suggesting that reduced maternal ADM is responsible for defects in uterine receptivity, implantation, and/or placentation [54]. Moreover, up-regulated ADM and its receptors not only induce the expression of nitric oxide synthase (NOS) in endothelial cells of rats and release of nitric oxide, a gaseous vasodilator to increase blood flow [55, 56]; but also enhance differentiation of rat Tr cells via phosphorylation of the mechanistic target of rapamycin (MTOR) [30]. There are two MTOR complexes: MTORC1 (MTOR, LST8, Raptor) and MTORC2 (MTOR, LST8, and Rictor). Previous studies have shown that NOS is required for ovine conceptus development [57, 58]; whereas MTOR is the master regulator of conceptus development in sheep and pigs, i.e. Tr cell proliferation and expression of mRNAs via MTORC1, and migration and cytoskeletal reorganization of Tr cells via MTORC2 [30, 59–62]. Thus, we hypothesized that ADM and its receptors increase in porcine uterine endometria and conceptuses during the peri-implantation period of pregnancy and have a functional role for uterine receptivity as well as conceptus growth and development. We will test this hypothesis by determining temporal and cell-specific changes in the expression of ADM and associated receptors (CALCRL, RAMP2, RAMP3, and ACKR3) in uteri of cyclic and pregnant gilts, and conceptuses. Results of this study indicated that ADM and associated receptors were expressed abundantly in porcine uteri and conceptuses during peri-implantation period of pregnancy, and that ADM increased the proliferation of porcine trophectoderm (pTr1) cells via activation of the MTORC1 cell signaling pathway.

Material and methods

All experimental and surgical procedures were in compliance with the Guide for Care and Use of Agriculture Animals in Research and Teaching and approved by Institutional Animal Care and Use Committee of North Carolina State University.

Animals and sample collection

Sexually mature gilts (F1 cross of Yorkshire X Landrace sows and Duroc boars) were maintained at the North Carolina State University Educational Swine Unit. Gilts were observed daily for signs of estrus (Day 0) and exhibited at least two estrous cycles of normal duration (18–21 days) prior to being used in experiments. Cyclic gilts were assigned randomly to be hysterectomized on either Day 10, 11, 12, 13, 14, or 15 of the estrous cycle (n = 3–6 gilts/day), while pregnant gilts, they were bred via artificial insemination at 12 and 24 h after detection of estrus, and assigned randomly to be hysterectomized on either day 10, 11, 12, 13, 14, 15, or 16 of pregnancy (n = 6 gilts/day). After hysterectomy, each uterine horn of all gilts was flushed with 20 ml sterile phosphate-buffered saline (PBS, pH 7.2). Pregnancy was confirmed by the presence of morphologically normal conceptuses in the uterine flushing. Uterine flushings were clarified by centrifugation (3000 x g for 30 min at 4°C), and part of the conceptuses were fixed in fresh 4% paraformaldehyde (prepared in PBS, pH 7.2) for 24 h and then in 70% ethanol for 24 h. The fixed tissues were dehydrated through a graded series of alcohol to xylene and embedded in Paraplast-Plus (Sigma-Aldrich, St. Louis, MO, USA). Sections (~1 cm) from the mid-portion of each uterine horn of the cyclic and pregnant gilts were fixed in fresh 4% paraformaldehyde for eventual embedding in Paraplast-Plus as described above. The remaining endometrium (physically dissected from myometrium as described previously [63]), and conceptuses were frozen in liquid nitrogen, and stored at −80°C for subsequent RNA extraction.

Quantitative detection for pig ADM

Porcine ADM (pADM) in uterine flushings was determined using an enzyme-linked immunosorbent assay according to the manufacturer’s protocol (LS-F6084, Lifespan Biosciences, Seattle, WA, USA). Intra-assay coefficient of variation (CV) for the ADM assay was <10%, while the inter-assay CV was <12%. The minimum detectable concentration for ADM was 5.21 pg/ml. Measurements were carried out using a SpectraMax iD3 Multi-Mode microplate reader (Molecular Devices, San Jose, CA, USA). All samples were assayed in duplicate.

RNA isolation

Frozen tissue was homogenized in 1 ml of TRIzol reagent (Thermo Fisher) using a Bead Mill 24 homogenizer (Thermo Fisher), two times at 4.5 m/s for 30 s, and rested intermittently for 20 s on ice. Homogenates were centrifuged for 10 min at 12,500 g in 4°C to pellet cellular debris. The supernatant was transferred to a 1.5 ml centrifuge tube and mixed with 200 μl of 1-bromo-3-chloropropane by manually shaking for 20 s. The tube was then incubated at room temperature for 3 min, and centrifuged 18 min at 12,500 g at 4°C. The aqueous phase was carefully removed, placed into a new tube, and mixed with 200 μl of chloroform by shaking the tubes for 20 s. Samples were rested for 3 min at room temperature and subsequently centrifuged at 21,000 g for 18 min at 4°C. Approximately 500 μl of the aqueous layer was transferred to a new tube and mixed with equal parts of 70% ethanol. This mix was filtered in columns from the RNeasy Mini kit (Qiagen, Valenica, CA, USA). Columns were washed once with 700 μl of RW1 buffer (catalog no. 1053394; Qiagen) and three times with 500 μl of RPE buffer (catalog no. 1018013; Qiagen). The RNA was eluted with RNase free water. The quantity and quality of total RNA were determined using spectrometry and denaturing agarose gel electrophoresis, respectively.

Quantitative real-time PCR analyses

RNA was reverse transcribed into cDNA using the Moloney Murine Leukemia Virus (M-MLV; Thermo Fisher) according to the manufacturer’s instructions. Quantitative RT-PCR (qRT-PCR) was performed using the CFX Connect Real-Time PCR Detection System (Bio-Rad, Hercules, CA, USA) and 1) the SsoAdvanced™ Universal SYBR® Green Supermix (catalog no. 1725274; Bio-Rad) with oligonucleotide primers synthesized by Integrated DNA Technologies (IDT; Coralville, IA, USA), or 2) the SsoAdvanced™ Universal Probes Supermix (catalog no. 1725284; Bio-Rad) with Taqman probes (Applied Biosystems). Delta–delta Ct values were calculated using 18 S control amplification results to acquire relative mRNA levels per sample. Information for all of the primers is provided in Table 1. Taqman probes of RAMP1 (Ss06942043_m1), RAMP3 (Ss03382547_u1) and 18S (4319413E) were purchased from Applied Biosystem to target porcine mRNAs of RAMP1 (NM_214199.2), RAMP3 (NM_214089.1), and eukaryotic 18S rRNA, a reliable normalization gene for qRT-PCR in eukaryote including pigs [64].

Table 1.

Summary of primer sequences for quantitative RT-PCR and expected product sizes.

Gene	RefSeq gene	Primers 5′ → 3′	Product, bp	Source
ADM	NM_214107.1 Validated	F: GGCCTGCCCAGACTGTCATT R: GGTAGCGCTTGACTCGGATG	102	Designed
CALCRL	NM_214095.1 Validated	F: ATCCATGGCCCGATTTGTGC R: AGGTCGCCATGGAATCAGCA	189	Designed
RAMP2	NM_214082.1 Validated	F: CCTGGCTCAGCATTTTCCCAC R: AATCGTGCCAGCAAAGTTGGG	95	Designed
ACKR3	XM_003133759.4 Validated	F: AACAACGAGACCTACTGCCG R: GCAGTAGAAGACGGCGATGA	129	Designed

Primers were designed to overlap exon–exon boundaries and validated.

Immunohistochemical analyses

Immunohistochemical (IHC) localization of ADM, CALCRL, and RAMP2 protein in porcine uterine and conceptus tissues (~5 μm) was performed as described previously [57, 65, 66]. Goat anti-ADM polyclonal immunoglobulin G (IgG; AF6108; R&D systems, Minneapolis, MN, USA), rabbit anti-CALCRL polyclonal IgG (NLS6731; Thermo Fisher, Waltham, MA, USA), and rabbit anti-RAMP2 polyclonal IgG (PA5–21953; Invitrogen, Waltham, MA, USA) were used at dilutions of 1:200, 1:500, and 1:500, respectively. Antigen retrieval was performed using antigen unmasking solution (H-3300; Vector Laboratories, Burlingame, CA, USA) for ADM, CALCRL, and RAMP2. Purified nonrelevant goat or mouse IgG was used as a negative control to replace the primary antibody at the same final concentration. For ADM, ADM-overexpressed (ADM^OE/OE) mouse uterine tissue at Day 4.5 (pseudo-pregnant day 4.5) of pseudopregnancy and ADM-deleted (ADM^−/−) mouse placenta were used as positive and negative controls, respectively. Immunoreactive proteins were visualized in sections using the Vectastain ABC Kit (PK-6100; Vector Laboratories), following the manufacturer’s instructions, and 3,3′-diaminobenzidine tetrahydrochloride (D8001; Sigma-Aldrich) was used as the color substrate. Sections were counterstained with hematoxylin before dehydrating and affixing coverslips with Permount. Digital images of uteri and conceptuses were captured using an Axioplan 2 microscope with an Axiocam HR camera and Axiovision 4 software (Carl Zeiss, Thornwood, NY, USA).

RNA in situ hybridization analyses

RNAscope in situ hybridization (ISH; Advanced Cell Diagnostic, Newark, CA, USA) was performed according to the manufacturer’s instructions using paraformaldehyde-fixed uterine and conceptus tissues (~5 μm) as well as customized probes Ss-ADM (819541), Ss-CALCRL (819561), Ss-RAMP2 (819551), Ss-RAMP3 (857371), and Ss-ACKR3 (819571) for porcine ADM, CALCRL, RAMP2, RAMP3, and ACKR3, respectively. A porcine peptidylprolyl isomerase B probe (Ss-PPIB; 428591) and a bacterial DapB gene probe (310043) were used as positive and negative controls, respectively. Following hybridization, slides were washed and probe binding visualized using the HD 2.5 Red Detection Kit (322360-USM). Sections were briefly counterstained with hematoxylin before dehydrating and affixing coverslips with Permount.

Cell culture

An established spontaneously immortalized pTr1 primary cell line (pTr1; a kind gift from Dr Robert C. Burghardt, Texas A&M University, College Station, TX) from Day-12 porcine conceptuses was used in the present in vitro studies as described previously [30, 67, 68]. The pTr1 cells exhibit numerous properties of pTr cells in vivo [67]. The cells were cultured in complete medium (CM; Dulbecco modified Eagle medium/Nutrient Mixture F-12, DMEM/F-12; Gibco BRL, Grand Island, NY), with 10% fetal bovine serum (FBS; Gibco BRL), 50 U/ml penicillin, 50 μg/ml streptomycin, 0.1 mM each for nutritionally nonessential amino acids, 1 mM sodium pyruvate, 2 mM glutamine, and 4 μg/ml insulin. The medium was replaced every 2 days. When the density of cells in the dishes reached about 80% confluence, subcultures of cells were prepared at a ratio of 1:3, and frozen stocks of cells were preserved at each passage. For the experiments, cultures of pTr1 cells (between passages 8 and 20) were grown in CM to 20–30% confluences in a 24-well plate (Costar no. 3524; Corning, Corning, NY, USA) or Lab-Tek II 4-well chamber slides (154,534; Sigma-Aldrich). Cells were serum- and insulin-starved for 24 h in customized medium, further deprived of arginine for additional 6 h, and then treated with porcine ADM (pADM; 010-11; Phoenix Pharmaceuticals, Inc., Burlingame, CA, USA) at indicated concentrations in basal medium (BM; DMEM/F-12 with 5% FBS and 1 ng/ml insulin). The experiments were replicated three times independently.

Proliferation assay

The pTr1 cells were subcultured (1 × 10⁴ cells/0.4 ml/well) in 24-well plates (Costar#3524; Corning) in CM until the monolayer reached 30% confluence and then switched to serum- and insulin-free customized medium for 24 h starvation. After additional 6 h of arginine deprivation as arginine can induce endogenous ADM production (S. Paudel and X. Wang; unpublished results), cells (n = 8 wells per treatment) were then cultured in 0.4 ml BM with pADM at various doses. For proliferation assay with MTOR inhibitor, pTr1 cells were pre-incubated with 50 nM rapamycin for 6 h, and then subjected to pADM treatment at 10⁻⁷ M. Cell numbers were determined after 48 and/or 96 h of incubation as described previously [30, 58, 60, 69]. Briefly, medium was removed from cells by vacuum aspiration, cells were fixed in 50% ethanol for 30 min, and fixative was removed by vacuum aspiration. Fixed cells were stained with Janus Green B in PBS (pH 7.2) for 3 min at room temperature. The stain was immediately removed using a vacuum aspirator, and the whole plate was sequentially dipped into water and destained by gentle shaking. The remaining water was removed by shaking, after which stained cells were immediately lysed in 0.5 N HCl and absorbance readings were taken at 595 nm using a microplate reader. As described previously [69], cell numbers were calculated from absorbance readings using the following formula: cell number = (absorbance – 0.00462)/0.00006926.

Quantitative immunocytochemistry

The pTr1 cells were seeded onto Lab-Tek II 4-well chamber slides (154,534; Sigma-Aldrich). After serum and insulin starvation for 24 h and arginine deprivation for extra 6 h, cells were treated with or without pADM (10⁻⁷ M) in BM medium. After 2 h, cells were fixed at −20°C with methanol for 10 min and rinsed with 0.02 M of PBS containing 0.3% Tween (PBST) for 5 min. The cells were blocked in 5% normal donkey serum for 2 h at room temperature, rinsed, and then immunofluorescence staining was performed overnight at 4°C using the primary antibody. Primary antibodies included rabbit anti-phospho-MTOR (p-MTOR) polyclonal IgG (#2971; Cell Signaling), and rabbit anti-phospho-eukaryotic translation initiation factor 4E binding protein 1 (p-4EBP1) polyclonal IgG (#9455; Cell Signaling) at a dilution of 1:100. Purified nonrelevant rabbit IgG was substituted for the primary antibody as a negative control. Cells were then incubated with secondary antibody, i.e. Alexa Flour 568 donkey anti-rabbit IgG (A10042; Invitrogen, Madison, WI) at a 1:250 dilution for 1 h at room temperature, and then rinsed in PBST and overlaid with Prolong Gold Antifade with DAPI. Images were captured using a Zeiss Axioplan 2 microscope with an Axiocan HR camera and Axiovision 4 software (Carl Zeiss Microscopy, Thornwood, NY, USA). Image acquisition was conducted under the same settings between the control and pADM-treated groups. Signals were quantified by Image J software (Version 1.47, National Institutes of Health) using standardized procedure as described previously [58, 70, 71]. Briefly, the cell region of interest (ROI; number of pixels for the selected cell) was defined by the Freehand selection tool. The next step was to split the image into the three color channels (RGB merge/split function) to gain one image per channel and then obtain the integrated density value (IDV; sum of the intensity of the pixels for the selected cells) of each molecule (i.e. p-MTOR and p-4EBP1) depending on fluorescent intensity of secondary antibodies. In parallel, average signal per pixel for a region selected just beside the cell was measured as background signal for subtraction. Finally, the corrected total cell fluorescence [CTCF; arbitrary units (a.u.)] was calculated based on the following formula: CTCF = IDV − (ROI × background signal), indicating the level of protein expression per cell analyzed.

Statistical analysis

The total recoverable amount of ADM was calculated by multiplying concentrations in the uterine flushing by volume of uterine flushing. The normality of data and homogeneity of variance were tested using the Shapiro–Wilk test and Brown–Forsythe test in Statistical Analysis System, respectively (version 8.1; SAS Institute). Data were analyzed by least squares one-way analysis of variance and post hoc analysis (the Fisher least significant difference) with each gilt/conceptus as an experimental unit. The effects of day, status (pregnant versus cyclic), and their interaction were also determined. All analyses were performed using SAS. Data are expressed as means with SEM. P < 0.05 was considered statistically significant.

Results

Recoverable amounts of ADM in uterine flushings

As illustrated in Figure 1, total recoverable ADM in uterine flushings was 9.1-fold greater for pregnant (10,558 ± 3165 pg) than cyclic gilts (1157 ± 449 pg) between Days 10 and 16 of pregnancy (status, P < 0.0001; day x status, P < 0.0001). In cyclic gilts, total ADM in uterine flushings increased 13.1-fold between Days 10 (157 ± 39 pg) and 14 (2064 ± 991 pg), and remained elevated through Day 16 (linear effect of day, P < 0.001). In pregnant gilts, total ADM in uterine flushings increased 7.8-fold between Days 10 (3518 ± 750 pg) and 16 (27,330 ± 4655 pg) of gestation (quadratic effect of day, P < 0.0001).

Figure 1.

Total recoverable ADM in porcine uterine flushings from cyclic and pregnant gilts. Effects of day, pregnancy status, and interaction (day-by-pregnancy status were significant (P < 0.0001). ^*P < 0.05; ^**P < 0.01; ^***P < 0.001; ^****P < 0.0001. n = 3–6 per day of the estrous cycle and n = 4 per day of pregnancy. Data are presented as means ± SEM.

Effects of day and status of the estrous cycle and pregnancy on ADM, CALCRL, RAMP1, RAMP2, RAMP3, and ACKR3 mRNAs in porcine endometria

Quantitative RT-PCR analyses revealed temporal changes in steady-state expression of mRNAs for ADM, CALCRL, RAMP1, RAMP2, RAMP3, and ACKR3 in uterine endometria during the estrous cycle and early pregnancy in gilts (Figure 2 and Supplemental Figure S2A). The expression of ADM mRNA was greater for pregnant than for cyclic gilts between Days 10 and 16 of pregnancy (status, P < 0.05; day x status, P < 0.05; Figure 2A). In cyclic gilts, endometrial expression of ADM mRNA increased 5.8-fold (linear effect of day, P < 0.05) between Days 10 and 16. In pregnant gilts, it increased 44.6-fold (quadratic effect of day, P < 0.01) between Days 10 and 16 of gestation. Endometrial CALCRL mRNA was not affected by day (P = 0.48) in cyclic gilts, but increased 3.0-fold (quadratic effect of day, P < 0.01) in pregnant gilts between Days 10 and 16 of gestation (Figure 2B). The expression of endometrial RAMP2 mRNA was not affected by day in either cyclic (P = 0.17) or pregnant (P = 0.31) gilts; however, it increased 2.4-fold (status, P < 0.01) in endometria of all pregnant compared to all cyclic gilts (Figure 2C). The expression of endometrial RAMP3 mRNA was not affected (P > 0.10) by day or pregnant status (Figure 2D). As the component of CGRP receptor with much lower affinity to bind ADM, RAMP1 mRNA was weakly expressed (Ct value range between 33 and 38) in the porcine endometria and further decreased by 68.1% (status, P < 0.01) in endometria of all pregnant compared to all cyclic gilts (Supplemental Figure S2A). Interestingly, as a cell-autonomous molecular rheostat to dampen ADM signaling, the abundance of ACKR3 mRNA was 40.5% (status, P < 0.01) less in endometria of pregnant compared to cyclic gilts (Figure 2E). In cyclic gilts, endometrial expression of ACKR3 mRNA was not affected by day (P = 0.61); however, endometrial expression of ACKR3 mRNA decreased 67.6% (quadratic effect of day, P < 0.05) between Days 10 and 12 and remained low to Day 16 of gestation.

Figure 2.

Steady-state levels of ADM (A), CALCRL (B), RAMP2 (C), RAMP3 (D), and ACKR3 (E) mRNAs in endometria of cyclic and pregnant gilts. ^*P < 0.05; ^****P < 0.0001. n = 3–6 per day of the estrous cycle and n = 6 per day of pregnancy. Data are presented as means ± SEM.

Effects of day of pregnancy on ADM, CALCRL, RAMP1, RAMP2, RAMP3, and ACKR3 mRNAs in porcine conceptuses

To elucidate whether ADM may have a functional role in growth and development of porcine conceptuses, we investigated temporal changes in steady-state mRNA for ADM, CALCRL, RAMP1, RAMP2, RAMP3, and ACKR3 in porcine conceptuses (Figure 3). Intriguingly, ADM mRNA increased (P < 0.05) in porcine conceptuses between Days 10 and 16 of gestation, with two peaks, i.e. 4.2- and 3.4-fold increases at Days 12 and 15, respectively, as compared to the one at Day 10 of gestation (Figure 3A). As components of the major ADM receptor (ADM₁), the expression of mRNAs for both CALCRL and RAMP2 increased (P < 0.05) by 13.4- and 5.5-fold, respectively, in porcine conceptuses between Days 10 and 16 of gestation (Figure 3B and C). In addition, RAMP3 mRNA was expressed in porcine conceptuses but no differences were detected between Days 10 and 16 of gestation (Figure 3D). The RAMP1 mRNA was only weakly detected (Ct value range between 33 and 38) in Days-12, −15, and −16 conceptuses even though it increased (P < 0.01) by 9.9-fold between Days 12 and 16 of gestation (Supplemental Figure S2B). The abundance of ACKR3 mRNA was low (P < 0.05) in porcine conceptuses between Days 10 and 14, compared to increased 4.4-fold (P < 0.05) at Day 16 of gestation (Figure 3E).

Figure 3.

Steady-state levels of ADM (A), CALCRL (B), RAMP2 (C), RAMP3 (D), and ACKR3 (E) mRNAs in porcine conceptuses during the peri-implantation period of pregnancy. Effects of gestational day were significant (P < 0.05). Different superscript letters denote significant (P < 0.05) differences. n = 3 for gestational day 10; and n = 6 for gestational days 12 and 16. Data are presented as means ± SEM.

Localization of ADM mRNA and protein in porcine endometria and conceptuses

ISH and IHC analyses were used to detect ADM mRNA and protein in a cell-specific manner in uteri of cyclic and pregnant gilts, as well as conceptuses (Figures 4). In cyclic gilts, ADM mRNA was weakly detectable in uterine LE and GE on Days 10, 11, 14, and 15 of the estrous cycle (Figure 4A). In pregnant gilts, ADM mRNA in uterine LE and GE was detectable between Days 10 and 12, and visually abundant only in uterine LE between Days 13 and 16 of pregnancy (Figure 4A). In porcine conceptuses, ADM mRNA abundance was weak on Days 10 and 11, but visually stronger in trophectoderm (Tr) and extraembryonic endoderm (En) between Days 12 and 16 of pregnancy (Figure 4A).

Figure 4.

In situ hybridization (A) and immunohistochemical (B) analyses of ADM mRNA and protein in uteri from cyclic and pregnant gilts and conceptuses. For in situ hybridization, a porcine peptidylprolyl isomerase B probe (Ss-PPIB; 428,591) and a bacteria DapB gene probe (310043) served as the positive and the negative controls, respectively. For immunohistochemistry, a goat IgG (gIgG) and ADM^−/− murine placenta (mPlacenta) at gestational day 13.5 served as negative controls; whereas ADM overexpressed (ADM^OE/OE) murine uterus at pseudo-pregnant day 4.5 served as positive control. n = 3. C, cyclic; P, pregnant; LE, luminal epithelium; GE, glandular epithelium; S, Stromal cells; Tr, trophectoderm; En, extraembryonic endoderm; NC, negative control; PC, positive control. Bar = 100 μm.

Unlike mRNA levels, immunoreactive ADM protein was detectable in uterine LE and GE, as well as stromal cells of cyclic and pregnant gilts (Figure 4B). However, ADM protein in uterine LE was visually greater for pregnant than cyclic gilts between Days 12 and 16. In addition, ADM protein was detectable in conceptuses between Days 10 and 11, but visually abundant in conceptus Tr and En between Days 12 and 16 of pregnancy (Figure 4B). Together, the expression of ADM mRNA and protein increased significantly in cells of the uterus (particularly in LE) and in conceptus tissues during early pregnancy.

Localization of CALCRL, RAMP2, RAMP3, and ACKR3 mRNAs and/or proteins in porcine endometria and conceptuses

Next, we investigated the localization of ADM receptors ADM₁ (CALCRL/RAMP2) and ADM₂ (CALCRL/RAMP3), and associated component ACKR3 at mRNA and/or protein in uteri of cyclic and pregnant gilts and in conceptuses (Figures 5–8). CALCRL mRNA and protein were detectable in uterine LE, GE, and stromal cells of cyclic and pregnant gilts, as well as Tr and En of conceptuses (Figure 5). In cyclic gilts, CALCRL mRNA and protein were expressed weakly in uterine LE, GE, and stromal cells between Days 10 and 11, undetectable in uteri between Days 12 and 13, and then visually abundant in the uteri between Days 14 and 15. On the other hand, CALCRL mRNA and protein were strongly expressed in sGE and GE between Days 10 and 11, and in uterine LE between Days 12 and 16 of pregnancy. In peri-implantation conceptuses, CALCRL mRNA and protein were weakly expressed in Tr between Days 10 and 11, but visually abundant in Tr and En between Days 12 and 16 of pregnancy.

Figure 5.

In situ hybridization (A) and immunohistochemical (B) analyses of CALCRL mRNA and protein in uteri from cyclic and pregnant gilts and conceptuses. For in situ hybridization, a porcine peptidylprolyl isomerase B probe (Ss-PPIB; 428,591) and a bacteria DapB gene probe (310043) served as the positive and the negative controls, respectively. For immunohistochemistry, a rabbit IgG (gIgG) served as negative control. n = 3. C, cyclic; P, pregnant; LE, luminal epithelium; GE, glandular epithelium; sGE, superficial glandular epithelium; S, Stromal cells; Tr, trophectoderm; En, extraembryonic endoderm; NC, negative control; PC, positive control. Bar = 100 μm.

Figure 8.

In situ hybridization analyses of ACKR3 mRNA in uteri from cyclic and pregnant gilts and conceptuses. A porcine peptidylprolyl isomerase B probe (Ss-PPIB; 428,591) and a bacteria DapB gene probe (310043) served as the positive and the negative controls, respectively. n = 3. C, cyclic; P, pregnant; LE, luminal epithelium; GE, glandular epithelium; S, Stromal cells; Tr, trophectoderm; En, extraembryonic endoderm; NC, negative control; PC, positive control. Bar = 100 μm.

Both RAMP2 mRNA and protein were expressed weakly in uterine LE, GE, and stromal cells on Days 10, 11, 14, and 15, but were undetectable between Days 12 and 13 of the estrous cycle in gilts (Figure 6). In contrast, the expression of RAMP2 mRNA and protein was visually abundant in uterine LE, GE, and stromal cells between Days 10 and 16 of pregnancy. Furthermore, RAMP2 mRNA and protein were expressed in Tr and En of conceptuses and visually abundant between Days 12 and 16 of pregnancy.

Figure 6.

In situ hybridization (A) and immunohistochemical (B) analyses of RAMP2 mRNA and protein in uteri from cyclic and pregnant gilts and conceptuses. For in situ hybridization, a porcine peptidylprolyl isomerase B probe (Ss-PPIB; 428,591) and a bacteria DapB gene probe (310043) served as the positive and the negative controls, respectively. For immunohistochemistry, a rabbit IgG (gIgG) served as negative control. n = 3. C, cyclic; P, pregnant; LE, luminal epithelium; GE, glandular epithelium; S, Stromal cells; Tr, trophectoderm; En, extraembryonic endoderm; NC, negative control; PC, positive control. Bar = 100 μm.

Due to the limited sources of antibodies, localization of RAMP3 and ACKR3 genes were only detected at the mRNA level via ISH (Figures 7 and 8). RAMP3 mRNA in uterine LE, GE, and stroma was detectable on Days 11, 14 and 15 of the estrous cycle and visually abundant between Days 13 and 16 of pregnancy (Figure 7). In porcine conceptuses, RAMP3 mRNA was expressed in Tr and En and visually abundant between Days 12 and 16 of pregnancy (Figure 7). In addition, ACKR3 mRNA was expressed weakly in uterine LE and GE between Days 12 and 13, but visually abundant at Days 10, 11, 14, and 15 of the estrous cycle in gilts (Figure 8). In pregnant gilts, ACKR3 mRNA in uterine LE was abundant on Day 10, and expressed weakly between Days 11 and 16 of gestation; whereas the expression of ACKR3 mRNA in uterine GE and stromal cells was strong on Day 10, remained visually elevated through Day 15, and undetectable to Day 16 (Figure 8). In porcine conceptuses, ACKR3 mRNA was detectable in Tr between Days 14 and 16 of pregnancy.

Figure 7.

In situ hybridization analyses of RAMP3 mRNA in uteri from cyclic and pregnant gilts and conceptuses. A porcine peptidylprolyl isomerase B probe (Ss-PPIB; 428,591) and a bacteria DapB gene probe (310043) served as the positive and the negative controls, respectively. n = 3. C, cyclic; P, pregnant; LE, luminal epithelium; GE, glandular epithelium; S, Stromal cells; Tr, trophectoderm; En, extraembryonic endoderm; NC, negative control; PC, positive control. Bar = 100 μm.

ADM stimulates proliferation and the MTORC1 signaling pathway in pTr1 cells

To determine whether ADM play a functional role on growth and development of porcine conceptuses, we determined effects of ADM on the proliferation of pTr1 cells after 48 h of incubation. ADM increased the proliferation of pTr1 cells (P < 0.05) at 48 h and 10⁻⁷ M was the optimal dose among tested ones for maximum stimulation of cell proliferation (by 1.8-fold at 48 h; P < 0.05; Figure 9A). Next, we evaluated the abundance of p-MTOR and p-4EBP1 using quantitative immunocytochemical analyses. After 2 h of incubation, p-MTOR increased 6.1-fold (P < 0.0001) in ADM-treated (10⁻⁷ M) pTr1 cells compared to control cells (Figure 9B and C). As one of the downstream effectors for MTORC1, the abundance of p-4EBP1 also increased 4.9-fold (P < 0.0001) in response to ADM treatment (Figure 9B and C). To further determine whether ADM stimulates the proliferation of pTr1 cells requires MTORC1, MTOR inhibitor rapamycin (50 nM; [58]) was employed. After 48 and 96 h of incubation, ADM at 10⁻⁷ M increased pTr1 cell proliferation by 2.0- and 3.7-fold (P < 0.05; Figure 9D), respectively. When rapamycin was added, ADM-derived proliferative effects were completely inhibited (P < 0.05; Figure 9D).

Figure 9.

Porcine ADM (pADM) stimulates proliferation via activation the MTORC1 signaling pathway in porcine conceptus trophectoderm (pTr1) cells. (A) The pTr1 cells (n = 8 wells) were seeded at 30% confluence in 24-well plates. After serum and insulin starvation for 24 h followed by arginine deprivation for extra 6 h, cells were cultured with the indicated doses of pADM. Cell numbers were determined after 48 h of incubation. Data are expressed as a percentage relative to nontreated control at 48 h. Effects of pADM dosage were significant (P < 0.05). (B and C) The pTr1 cells (n = 3 wells) were seeded at 30% confluence onto Lab-Tek II four-well chamber slides. After serum and insulin starvation for 24 h followed by arginine deprivation for extra 6 h, cells were treated with or without pADM (10⁻⁷ M). Quantitative immunocytochemical analyses demonstrated an increased abundance of phosphorylated MTOR (p-MTOR) and 4EBP1 (p-4EBP1) in pTr1 cells treated with pADM for 2 h. a.u., arbitrary units. Bar = 100 μm. (D) The pTr1 cells (n = 4 wells) were pre-incubated with 50 nM rapamycin, and then subjected to pADM treatment at 10⁻⁷ M. Cell numbers were determined after 48 and 96 h of incubation. Data are presented as means ± SEM. Different superscript letters denote significant differences (P < 0.05) among groups. ^****P < 0.0001 vs. control (CON).

Discussion

ADM is a highly conserved peptide hormone required for intra-uterine spacing of blastocysts and angiogenesis during early pregnancy in rodents [51–54, 72]. Results from human studies, however, have not been as consistent [53]. Two studies found that ADM levels in human umbilical plasma and amniotic fluid are inversely correlated with birth weight and length of babies [73, 74]; whereas two other studies showed no difference in ADM levels between small for gestational age and appropriate for gestational age infants [75, 76]. Thus, it is necessary to further determine with certainty how ADM alterations may be involved in the pathogenesis of IUGR. Given that pigs are also the litter-bearing species and exhibit the most severe naturally occurring IUGR (15–25%) [8, 17, 18] due to the increased fetal crowding during gestation and limited uterine capacity [5, 6, 19–22], it is imperative that the physiological role of ADM in the control of uterine capacity and conceptus development in pigs be established. This is the first report of changes in ADM and its associated components (CALCRL, RAMP2, RAMP3, and ACKR3) during the estrous cycle and peri-implantation period of pregnancy in pigs; as well as its role in stimulating proliferation of conceptus trophectoderm cells via activation of the MTORC1 cell signaling pathway.

In this study, total amounts of ADM in uterine fluids are reported rather than concentrations in uterine flushings because a method has not been developed to estimate the actual amounts of fluid in the uterine lumen due to the rapid exchange of water. The results indicate that significant increases in amounts of ADM in the uterine lumen are coordinate with rapid growth, development, and elongation of porcine conceptuses between Days 12 and 16 of pregnancy. The changes in recoverable ADM were much greater in pregnant than in cyclic gilts, which indicates that mechanisms for production and secretion of ADM into the uterine lumen are likely influenced by regulatory molecules from conceptuses that, in turn, probably require ADM for successful growth and development.

Next, we investigated the sources of ADM using qRT-PCR, ISH, and IHC analyses. ADM mRNA was expressed mainly by uterine LE and sporadically by uterine GE, but not by uterine stromal cells during early pregnancy. However, at the protein level, ADM was detectable in stromal cells, indicating the transport of ADM from maternal blood into the uterine tissues and cells. Even though the expression of ADM mRNA increased significantly in both cyclic and pregnant porcine uteri between Days 10 and 16, the increase was greater during pregnancy. This suggests that uterine ADM expression may be progesterone (P₄)-induced and estradiol (E₂)-stimulated as conceptus E₂ is the pregnancy recognition signaling in pigs. On the other hand, ADM mRNA and protein were also expressed in the porcine conceptus Tr and En and most abundant between Days 12 and 16 of pregnancy, with the greatest expression on Days 12 and 15, respectively. This indicates that not only endometria but also conceptuses contribute to ADM increases in the uterine lumen during the peri-implantation period of pregnancy. Also, it may suggest a significant role for ADM in the initiation of conceptus elongation and implantation as porcine conceptuses elongate most rapidly from Day 12 to Day 16 and undergo implantation.

ADM signals through its receptor complexes ADM₁ (CALCRL/RAMP2) and/or ADM₂ (CALCRL/RAMP3) on cells [40, 41]. CALCRL is a G-protein coupled receptor whereas RAMPs contribute to CALCRL translocation toward the plasma membrane. Results from qRT-PCR analyses demonstrated significant increases in CALCRL mRNA in the endometrium during the peri-implantation period of pregnancy in gilts, but expression was not affected by day of the estrous cycle. This suggests a significant role of ADM signaling in uterine functions during early pregnancy in pigs. Based on qRT-PCR, ISH and/or IHC analyses, the significant increases in RAMP2 (between Days 10 and 16) and RAMP3 (between Days 13 and 16) in uteri of pregnant gilts, further suggests that ADM₁ and ADM₂ play important, but different roles in ADM signaling during the peri-implantation period of pregnancy. Interestingly, there was a discrepancy between the expression of RAMP2 mRNA and protein on Days 11 and 15 of pregnancy as expression at the mRNA level was greater, but RAMP2 protein was less abundant in uterine GE and stromal cells on Day 11, and uterine LE on Day 15 (Figure 6), perhaps due to posttranslational modification (e.g. modified protein folding and glycosylation patterns). Unlike ADM that is expressed primarily by uterine LE, the expression of CALCRL, RAMP2, and RAMP3 was greatest in uterine LE, GE, and stromal cells of pregnant gilts, which is a prerequisite for ADM to exert its functional role(s) during early pregnancy. Meanwhile, as a decoy receptor to dampen ADM signaling [43], the expression of ACKR3 was not affected by day of the estrous cycle, but decreased in endometria of pregnant gilts between Days 10 and 16 of pregnancy. The decrease in expression of the decoy receptor ACKR3 was most notable in uterine LE and is consistent with cellular mechanisms to increase ADM signaling as part of enhancing the window of receptivity to implantation. These results suggest that ADM may also play a functional in uterine receptivity to implantation during peri-implantation period of pregnancy in pigs.

Notably, the components of ADM receptors, i.e. CALCRL, RAMP2, and RAMP3 were expressed in the porcine conceptus Tr and En during the peri-implantation period of pregnancy, suggesting both paracrine and autocrine effects of ADM on growth and development of porcine conceptuses. Meanwhile, the significant increases in CALCRL and RAMP2, but not RAMP3, particularly between Days 14 and 16 of pregnancy further indicate that (1) ADM₁ (CALCRL/RAMP2) is the regulatory receptor of ADM in the conceptuses in response to pregnancy and (2) there may be a positive correlation between ADM signaling and conceptus behavior of Tr in terms of elongation, migration and adhesion as porcine conceptus elongates from mid- (100–200 mm) to late-filamentous (800–1000 mm) forms, and implant between Days 14 and 16 of pregnancy [26–28]. In addition, as the CGRP receptor component, RAMP1 were barely detectable in both endometria and conceptuses, suggesting its irrelevance in ADM signaling during the peri-implantation period of pregnancy in pigs. Interestingly, ACKR3 was also detectable in Tr between Days 14 and 16 of pregnancy. Recent reports indicate that RAMP3 can mediate rapid recycling of ACKR3 and enhance angiogenesis in the retina postnatally [77]. Understanding the precise mechanisms by which expression of ACKR3 serves as non-signaling receptor to control the functional dosage of ADM in growth and development of porcine conceptuses in future research.

Finally, to validate the significance of ADM on growth and development of conceptuses in pigs, we investigated effects of porcine ADM on proliferation and MTORC1 cell signaling in pTr1 cells based on the recoverable amounts of ADM in porcine uterine flushings, as well as a previous study with trophoblast stem cells of rats [78]. ADM stimulated proliferation of pTr1 cells with 10⁻⁷ M ADM being the optimal dose. ADM activated MTORC1 cell signaling via increases in phosphorylation of MTOR and 4EBP1. When MTOR inhibitor rapamycin was added, ADM-driven proliferation of pTr1 cells was completely abrogated, suggesting that ADM stimulates proliferation in pTr1 cells via activation of the MTORC1 cell signaling pathway. Research beyond the scope of this study is necessary to understand the intermediate components of ADM mediated cell signaling and expression of transcription factors downstream of the MTORC1 signaling pathway.

In summary, this is the first report of effects of pregnancy status, days of gestation, and days of estrous cycle on changes in amounts of recoverable ADM in the porcine uterine luminal fluids, as well as expression of mRNAs and proteins associated with functional aspects of ADM and its receptor complexes in both uterine and conceptus tissues. We have also provided evidence for a functional role of ADM to induce the proliferation of pTr1 cells via activation of the MTORC1 cell signaling pathway. These results are highly relevant to reproduction in pigs that are a litter-bearing species with a high incidence of early embryonic death (30–40%), naturally occurring intrauterine growth restriction (15–25%), and frequent occurrences of stillborn piglets (3–9%) [8, 11, 15, 17, 18, 79–84]. Because ADM is a multifunctional regulatory peptide hormone known to influence the spacing of blastocysts and angiogenesis, understanding its precise roles and mechanisms of action during early pregnancy in pigs is a prerequisite for improving reproductive efficiency in swine. Therefore, future studies will further dissect the functional role of ADM signaling on peri-implantation conceptus growth and development, as well as pre-implantation spacing of blastocysts and uterine capacity using in vitro and in vivo loss-of-function studies with pigs. The correlation between concentrations of ADM in reproductive fluids (e.g. serum, amniotic and allantoic fluids) and reproductive health status throughout the gestation will also be investigated in pigs.

Grant Support: This research was supported by the Hatch Project 1020014 from the USDA National Institute of Food Agriculture, Faculty Research and Professional Development Award 2020-2571, and Research and Innovation Seed Funding Award 2021-1946 from North Carolina State University.

Data availability

The data underlying this article will be shared on reasonable request to the corresponding author.