Abstract
Integrins and OPN are potential mediators of blastocyst attachment to the endometrium to initiate implantation. The goals were to examine the temporal/spatial pattern of expression of integrins at the endometrial–placental interface of sheep encompassing Days 9 through 80 of gestation and determine if OPN co-localizes with integrins. Results show the following: (1) αv, α4, β1, β3 and β5 integrins at the apical surface of endometrial luminal epithelium (LE) from Days 11 through 16 of pregnancy that indicate a role for these integrins during implantation; (2) large, intermittent aggregates of αv, α4, α5, β1 and β5 integrins at the endometrial–placental interface from Days 20 through 55, suggesting adaptation to a localized tissue remodeling stage of placentation; and (3) integrin adhesion complexes (IACs) containing αv, α4, α5, β1 and β5 integrins precisely distribute at the apical surfaces of apposed endometrial LE and chorion along expanses of the interplacentomal endometrial–placental interface between Days 60 and 80 of gestation, suggesting engagement of these integrins with the ECM to stabilize adhesion between endometrial LE and chorion in response to the increasing mechanical stress on this interface by the increasing size of the fetus and volumes of fetal fluids. An advancement is the clear co-localization of OPN and integrins at the endometrial–placental interface throughout gestation in sheep. The comprehensive nature of these results provide evidence that integrins potentially interact with OPN to play key roles in the mechanisms required for implantation and placentation throughout pregnancy in sheep and have implications concerning implantation and placentation in other species.
Introduction
From the early to mid-1990s, integrins emerged as potential mediators of blastocyst attachment to the endometrium to initiate implantation when it was established that the endometrial luminal epithelium (LE) of women expressed multiple integrin subunits during the peri-implantation period of pregnancy. Of these integrins, αv, α1, α2, α3, β3 and β4 were localized to the apical domains of endometrial LE cells by light microscopy. A bridging ligand hypothesis of blastocyst attachment to the endometrial LE was proposed in which bifunctional extracellular ligands interacted with integrin receptors on the apical surfaces of placental trophectoderm and endometrial LE (Lessey et al. 1992, Tabibzadeh 1992, Klentzeris et al. 1993, Aplin et al. 1994, Aplin 1996). Integrins are dominant glycoproteins in adhesion cascades. They comprise a ubiquitous family of cation-dependent, heterodimeric (one α subunit non-covalently linked to one β subunit), intrinsic transmembrane glycoprotein receptors that mediate cellular differentiation, motility, and adhesion (Hynes 1987, Albelda and Buck 1990, Ruoslahti et al. 1994). The central role of integrins in the implantation adhesion cascade stems from their ability to bind ECM ligands to mediate adhesion, cause cytoskeletal reorganization to stabilize adhesion, and transduce cellular signals through numerous signaling intermediates (Giancotti & Ruoslahti 1999, Gallant et al. 2005, Horton et al. 2015, Humphries et al. 2019). Bowen et al. (1996) were the first to establish the presence of multiple integrin subunits at the apical surfaces of the conceptus (embryo/fetus and associated placental membranes) trophectoderm and endometrial LE of a species with epitheliochorial placentation. Alpha 4, α5, αv, β1, β3 and β5 were localized to porcine implantation sites on Days 12 through 15 of gestation providing for the potential assembly of these subunits into the integrin receptors αvβ1, αvβ3, αvβ5, α4β1, and α5β1 (Bowen et al. 1996). Subsequently, the integrin subunits αv, α4, α5, β1, β3, and β5 were shown to be constitutively expressed on the conceptus trophectoderm and at the apical surface of the endometrial LE on Day 16 of gestation, during the attachment phase of implantation in sheep (Johnson et al. 2001). Further investigation of pregnant sheep established that the integrin subunits αv, α4, α5, β1, and β5 assembled into integrin adhesion complexes (IACs), a hallmark of activated integrins (Humphries et al. 2019), at the endometrial–chorioallantoic interface of interplacentomal regions of placentation from Days 40 through 120 of gestation (Burghardt et al. 2009). The apical expression of integrin subunits at the endometrial–placental interface of sheep likely contribute to the assembly of the integrin receptors αvβ3, αvβ1, αvβ5, α4β1, and α5β1 all of which bind to the matricellular protein osteopontin (OPN, also known as secreted phosphoprotein 1 (SPP1)). However, the expression of integrins has not been examined between Days 17 and 40 of gestation in sheep when the conceptus trophectoderm cells begin to fuse to form trophoblast giant cells that invade the endometrial LE and eventually replace the endometrial LE with a syncytium (Seo et al. 2019). Further, OPN has not been definitively localized to IACs during pregnancy in sheep.
OPN is an acidic member of the small integrin-binding ligand N-linked glycoprotein (SIBLING) family of extracellular matrix (ECM) proteins/cytokines that undergoes extensive posttranslational modification, including phosphorylation, glycosylation, and cleavage. The result is a versatile protein with multiple functions arising from its role as a mediator of cell–cell and cell–extracellular matrix communications that encompass developmental processes, immunological responses during inflammation and wound healing, and biomineralization. OPN binds to various integrin heterodimers via its Gly-Arg-Gly-Asp-Ser (GRGDS) amino acid sequence to effect cell adhesion, migration and proliferation (Denhardt & Guo 1993, Butler et al. 1996, Sodek et al. 2000). OPN is a constituent of the uterine-placental microenvironment of multiple species including mice (Nomura et al. 1988), humans (Brown et al. 1992), sheep (Johnson et al. 1999b ), pigs (Garlow et al. 2002), rabbits (Apparao et al. 2003), cattle (Kimmins et al. 2004), and goats (Joyce et al. 2005a). It is hypothesized to bridge integrins expressed on the apical surface of the conceptus trophectoderm to integrins expressed on the apical surface of endometrial LE cells to mediate conceptus attachment to the endometrium to initiate implantation in the majority of these species (Johnson et al. 2003). In sheep (reviewed in Johnson et al. 2014), as the lifespan of the CL is extended, the result of the actions of interferon tau secretion from elongating ovine conceptuses, the CL secrete progesterone. Progesterone then induces the synthesis and secretion of OPN from the endometrial glandular epithelium (GE) (Johnson et al. 2001). In sheep, the implantation cascade is initiated with down-regulation mucin 1 (Muc 1; the regulatory mechanism for down-regulation remains to be identified) on the endometrial LE surface to expose integrins on the endometrial LE to conceptus trophectoderm for interactions with OPN that mediate adhesion of conceptus trophectoderm to endometrial LE (Kim et al. 2010, Johnson et al. 2014). OPN continues to be expressed at the interface between the endometrium and placenta during later stages of pregnancy in sheep (Burghardt et al. 2009), but the temporal and spatial relationship between the expression of integrins and OPN during the latter stages of pregnancy are unclear.
Therefore, the goals of this study were to comprehensively examine the temporal and spatial patterns of expression of multiple integrin subunits at the endometrial–placental interface of sheep encompassing Days 9 through 80 of pregnancy and determine if OPN protein co-localizes with integrin subunit proteins at the endometrial–placental interface of sheep. Results suggest that integrins likely interact with OPN at the endometrial–placental interface on all days of pregnancy examined, but the functional role(s) of these interactions change. OPN-integrin interactions at the endometrial–placental interface of sheep can be classified into three stages: (1) cell–cell adhesion during conceptus attachment for implantation; (2) cell migration during the early stages of syncytialization of the endometrial and placental epithelia; and (3) mechanosensation and mechanotransduction in response to tensional forces applied through increases in fetal size and volumes of fetal fluids as pregnancy progresses (Bazer et al. 2012).
Materials and methods
Animal model
Mature Rambouillet ewes were observed daily for estrus (Day 0 = onset of estrus) in the presence of vasectomized rams and used in experiments after they had exhibited at least two estrous cycles of normal duration (16–18 days). All experimental and surgical procedures followed the Guide for the Care and Use of Agriculture Animals in Research Teaching and were approved by the Institutional Animal Care and Use Committee of Texas A&M University. Ewes were mated to intact rams three times at 12-h intervals beginning at onset of estrus. Pregnant ewes were ovariohysterectomized on either Day 9, 11, 13, 16, 20, 25, 35, 55, 60 or 80 of pregnancy (n = 4 ewes/day). Uteri from Days 9 and 13 were flushed with PBS, and pregnancy was verified by the recovery of an apparently normal conceptus in uterine flushes. Several sections (1–1.5 cm) of uterine wall from the middle of each uterine horn were fixed in 4% paraformaldehyde for 24 h and then embedded in Paraplast Plus (Oxford Labware, St Loius, MO) or snap frozen in Tissue-Tek OCT compound (Fisher Scientific).
Immunohistochemical analyses of integrins at the endometrial–placental interface
Immunohistochemical localization of integrin proteins in cross-sections of the interplacentomal endometrial–placental interface (paraffin-embedded thin sections of 5 μm) was performed as previously described (Joyce et al. 2005b) with six rabbit anti-integrin (human) antibodies (EMD Millipore). Trial experiments were performed using 1:100 and 1:200 dilutions of the antibodies. In this series of immunohistochemical localizations, we purposely focused on detecting large aggregates of integrins and OPN that we propose are clusters of the molecules due to receptor–ligand binding, activation and adhesion complex formation. Because these complexes are large, it allows detection of proteins with minimal concentration of antibody, and the 1:200 dilution was subsequently used for immunohistochemical detection of integrins and OPN. Although it is likely that some of the integrins are also present in other cell types within the uterus and placenta of pigs, the technique used for this study was not sensitive enough for us to make definitive statements on expression in other cell types. The antibodies included anti-αv (AB1930), anti-α4 (AB1924), anti-α5 (AB1949), anti-β1 (AB1952), anti-β3 (AB1932) and anti-β5 (AB1926). Antigen retrieval was performed using protease (0.5 mg/mL in PBS, Sigma-Aldrich) for αV, α4 and α5 antibodies and boiling citrate (0.01M sodium citrate buffer, pH 6.0) for β1 antibody. Antigen retrieval was not necessary for β3 and β5. Purified rabbit IgG was used as a negative control. A band of protein of between 95 and 140 kDa has previously been detected by Western blotting of ovine myometrial tissue (anti-α5 and anti-β1 immunoglobulins, data not shown). The anti-αv and anti-α4 immunoglobulins have been previously used to immunoprecipitate bands of protein of between 95 and 140 kDa in surface biotinylated ovine trophectoderm cells (Kim et al. 2010). The anti-β3 and anti-β5 immunoglobulins have been previously used to immunoprecipitate bands of protein of between 95 and 140 kDa in surface biotinylated porcine trophectoderm cells (Massuto et al. 2010). Immunoreactive proteins were visualized in sections using the Vectastain ABC kit (Vector Laboratories, Burlingame, CA) following the manufacturer’s instructions with 3,3′-diaminobenzidine tetrahydrochloride (Sigma–Aldrich) as the color substrate. Sections were counterstained with Harris modified hematoxylin. Coverslips were affixed using Permount mounting medium (Fisher Scientific). Images were taken using an Axioplan 2 microscope (Carl Zeiss) interfaced with an Axioplan HR digital camera.
Immunofluorescence analyses of OPN and integrins at the endometrial–placental interface
Immunofluorescence localization of OPN and integrins in serial cross-sections of the interplacentomal endometrial–placental interface was performed using tissues frozen in OCT (10 μm thickness) from Day 60 of pregnancy as previously described (Seo et al. 2019). Tissue sections were fixed in methanol at −20°C, and then washed in PBS. The sections were then blocked in 10% normal goat serum diluted in antibody dilution buffer (1× PBS containing 1% BSA and 0.1% Triton X-100) for 1 h at room temperature and then washed three times (5 min/wash) in PBS. The antibodies were then added to the cross-sections at a dilution optimized for each antibody and incubated overnight at 4°C in a humidified chamber. The antibodies used were a rabbit anti-OPN (1:500; AB10910; EMD Millipore), rabbit anti-αV (1:1000; AB1930; EMD Millipore), rabbit anti-α4 (1:1000; AB1924; EMD Millipore), rabbit anti-α5 (1:1000; AB1949; EMD Millipore), rabbit anti-β1 (1:1000; AB1952; EMD Millipore), rabbit anti-β3 (1:1000; AB1932; EMD Millipore), and rabbit anti-β5 (1:1000; AB1926; EMD Millipore). Tissue sections were then washed three times (5 min/wash) in PBS. Goat anti-rabbit IgG Alexa 594 (1:250; Life Technologies) was added to the tissue sections, and the sections were incubated for 1 h at room temperature. Tissue sections were then washed three times for 5 min/wash in PBS. Slides were counterstained with Prolong Gold Antifade reagent containing DAPI (Life Technologies) and coverslipped. Images were taken as previously described.
For dual immunofluorescence staining (rabbit anti-β1 integrin + mouse anti-E-cadherin (BD Biosciences; 610182; 1:500), rabbit anti-OPN + mouse anti-E-cadherin, rabbit anti-E-cadherin (GeneTex, Irvine, CA; GTX100443; 1:200) + mouse anti-proliferating nuclear cell antigen (PCNA; Abcam; ab29; 1:200), we followed the same procedures as described for normal immunofluorescence staining except that we added the two primary antibodies simultaneously on the first day and added the two secondary antibodies (goat anti-rabbit-Alexa Fluor 488-conjugated and goat anti-mouse-Alexa Fluor 594-conjugated) simultaneously on the second day.
Results
Immunolocalization of integrins at the endometrial–placental interface of sheep
To determine the temporal and spatial expression of integrin subunits at the interplacentomal endometrial–placental interface during sheep pregnancy, immunohistochemistry was conducted on paraffin-embedded thin sections from Days 9 through 80. Alpha v, α4, α5, β1 and β5 integrin subunits were detected in uterine and placental tissues (Figs 1, 2, 3, 4, 5 and 6), and the immunostaining pattern revealed that: (1) integrin subunits were not detectable at the endometrial–placental interface on Day 9; (2) integrin subunits presented a continuous thin layer at the apical surface of endometrial LE between Days 11 through 16; (3) integrin subunits were localized to large, infrequent, intermittent and disorganized aggregates, often in the shallow GE (sGE) near the mouths of glands at the endometrial–placental interface on Days 20 through 55; and (4) integrin subunits showed precisely distributed IACs at the apical surfaces of the apposed endometrial LE and placental chorion along extensive expanses of the endometrial–placental interface on Days 60 through 80 of gestation. Figure 6 illustrates the co-distribution αv, α4, α5, β1 and β5 integrin subunits to IACs of the same size and distribution in serial cross-sections of the interplacentomal endometrial–placental interface on Day 60 of gestation in sheep. Immunolocalization of the β3 integrin subunit differed from that for the other integrins. Similar to the other integrin subunits, β3 integrin was not detectable at the endometrial–placental interface on Day 9, but was then present as continuous thin layer at the apical surface of endometrial LE on Days 11 through 16 (data not shown here). However, the β3 integrin was not detected in endometrial or placental tissues from Days 20 through 80 of gestation, and β3 containing IACs were not detected throughout pregnancy (Fig. 6).
Immunohistochemical detection of the αv integrin subunit at the endometrial–placental interface of sheep between Days 9 and 80 of gestation. Alpha V expression is initially detected on Day 16 of pregnancy on the apical surface of endometrial luminal epithelial (LE; an arrow; 40×) cells. Large intermittent and disorganized aggregates of αv integrin form at the endometrial–placental interface beginning on Day 20 of pregnancy. These aggregates of the αv integrin grow in size and become more organized as pregnancy progresses. By Day 60 of pregnancy, paired, symmetric and aligned aggregates of αv integrin are detectable in a ‘train-track like’ pattern at the apical domains of the endometrial LE and chorionic epithelium (CE; 40×) which continue to Day 80 of pregnancy and likely represent in vivo integrin adhesion complexes (IACs). On Day 35 it is interesting to note that aggregates of αv integrin occur in epithelial cells around the mouths of the uterine glands (GE; 5×). A rabbit IgG was used as a negative control. Width of fields for 5× magnification is 1800 µm, for 10× magnification is 890 µm and for 40× magnification is 220 µm. Tr, trophectoderm; ST, stroma.
Citation: Reproduction 160, 4; 10.1530/REP-20-0273
Immunohistochemical detection of the α4 integrin subunit at the endometrial–placental interface of sheep between Days 9 and 80 of gestation. Alpha 4 expression is initially detected on Day 16 of pregnancy on the apical surface of endometrial luminal epithelial (LE; arrows; 40×) cells. Large intermittent and disorganized aggregates of α4 integrin form at the endometrial–placental interface beginning on Day 20 of pregnancy. These aggregates of the α4 integrin grow in size and become more organized as pregnancy progresses. By Day 60 of pregnancy, paired, symmetric and aligned aggregates of α4 integrin are detectable in a ‘train-track like’ pattern at the apical domains of the endometrial LE and chorionic epithelium (CE; 40×) which continue to Day 80 of pregnancy and likely represent in vivo integrin adhesion complexes (IACs). On Day 35 of pregnancy, aggregates of α4 integrin are detected in epithelia around the mouths of the uterine glands (GE; 5×). A rabbit IgG was used as a negative control. Width of fields for 5× magnification is 1800 µm, for 10× magnification is 890 µm and for 40× magnification is 220 µm. Legend: Tr, trophectoderm; ST, stroma.
Citation: Reproduction 160, 4; 10.1530/REP-20-0273
Immunohistochemical detection of the α5 integrin subunit at the endometrial–placental interface of sheep between Days 9 and 80 of gestation. Alpha 5 is initially detected on Day 16 of pregnancy in the cytoplasm and weakly on the apical surface of endometrial luminal epithelial (LE) cells. Large intermittent and disorganized aggregates of α5 integrin form at the endometrial–placental interface by Day 20 of pregnancy, but increase in size and become more organized as pregnancy progresses. By Day 60 of pregnancy, paired, symmetric and aligned aggregates of α5 integrin are detectable in a ‘train-track like’ pattern at the apical domains of the endometrial LE and chorionic epithelium (CE; 40×) which continue to Day 80 of pregnancy and likely represent in vivo integrin adhesion complexes (IACs). A rabbit IgG was used as a negative control. Width of fields for 10× magnification is 890 µm and for 40× magnification is 220 µm. Tr, trophectoderm; ST, stroma.
Citation: Reproduction 160, 4; 10.1530/REP-20-0273
Immunohistochemical detection of the β1 integrin subunit at the endometrial–placental interface of sheep between Days 9 and 80 of gestation. Beta 1 expression is initially detected on Day 16 of pregnancy on the apical surface of endometrial luminal epithelial (LE) cells. Large intermittent and disorganized aggregates of β1 integrin form at the endometrial–placental interface beginning by Day 20 of pregnancy, but then increase in size and become more organized as pregnancy progresses. By Day 60 of pregnancy, paired, symmetric and aligned aggregates of β1 integrin are detectable in a ‘train-track like’ pattern at the apical domains of the endometrial LE and chorionic epithelium (CE; 40×) which continue to Day 80 of pregnancy and likely represent in vivo integrin adhesion complexes (IACs). On Day 35 pregnancy, aggregates of β1 integrin are localized in epithelial cells around the mouths of the uterine glands (GE; 5×). A rabbit IgG was used as a negative control. Width of fields for 5× magnification is 1800 µm, for 10× magnification is 890 µm and for 40× magnification is 220 µm. Tr, trophectoderm; ST, stroma.
Citation: Reproduction 160, 4; 10.1530/REP-20-0273
Immunohistochemical detection of the β5 integrin subunit at the endometrial–placental interface of sheep between Days 9 and 80 of gestation. Beta 5 expression is detected by Day 11 of pregnancy on the apical surface of endometrial luminal epithelial (LE; arrows) and shallow glandular epithelial (sGE; arrows) cells. Large intermittent and disorganized aggregates of β5 integrin form at the endometrial–placental interface beginning on Day 20 of pregnancy. These aggregates of the β5 integrin increase in size and become more organized as pregnancy progresses. By Day 60 of pregnancy, paired, symmetric and aligned aggregates of β5 integrin are detectable in a ‘train-track like’ pattern at the apical domains of the endometrial LE and chorionic epithelium (CE; 40×) which continue to Day 80 of pregnancy and likely represent in vivo integrin adhesion complexes (IACs). On Day 35 of pregnancy, aggregates of β5 integrin are present in epithelial cells around the mouths of the GE (5×). A rabbit IgG was used as a negative control. Width of fields for 5× magnification is 1800 µm, for 10× magnification is 890 µm and for 40× magnification is 220 µm. Tr, trophectoderm; ST, stroma.
Citation: Reproduction 160, 4; 10.1530/REP-20-0273
Co-distribution of integrin subunits at the endometrial–placental interface on Day 60 of gestation. The αv, α4, α5, β1, and β5 integrin subunits co-distributed into aggregates, presumably in vivo integrin adhesion complexes (IACs) at the apical domains of the apposed endometrial luminal epithelial (LE) and chorionic epithelial (CE) cells in a ‘train-track like’ pattern in the same regions of the endometrial–placental interface (arrows). ITGB3 containing IACs were not observed (asterisks). A rabbit IgG was used as a negative control. Width of fields is 890 µm. ST, stroma; GE, glandular epithelium.
Citation: Reproduction 160, 4; 10.1530/REP-20-0273
Co-localization of integrin subunits and OPN
The aggregation of integrins at the interplacentomal endometrial–placental interface from Days 20 through 80 of gestation in sheep strongly suggests the presence of IACs. It has been hypothesized that ECM molecules such as OPN engage these integrins to induce IACs, but no ECM molecule has been demonstrated to localize to these presumed IACs. To determine if OPN is an ECM protein that may bind to and activate integrin receptors at the endometrial–placental interface, immunofluorescence microscopy was performed to co-localize OPN with integrin subunits in frozen Day 60 endometrial–placental tissues. Because both the OPN and the integrin antibodies were generated in rabbits, co-localization was demonstrated by individual protein staining in serial thin sections. OPN co-distributed with αv, α4, α5, β1 and β5 integrins to large aggregates at the endometrial–placental interface (Figs 7 and 8). Indeed, the pattern of immunostaining in serial sections was nearly identical for OPN and each of these integrin subunits. Consistent with results from immunohistochemistry using paraffin-embedded tissues, large aggregates of β3 integrin were not detected at the Day 60 endometrial–placental interface, although β3 was detected in a continuous thin layer at the apical surface of the endometrial LE, much like the other integrins were observed on Days 11 through 16, using immunofluorescence and frozen tissues (Fig. 8). In our experience the signal-to-noise ratio using immunofluorescence vs colorimetric immunohistochemistry results in slightly greater sensitivity.
Co-distribution of αv, α4 and α5 integrin subunits with OPN by immunofluorescence microscopy. Serial sections of endometrial–placental tissues from Day 60 of pregnancy were utilized to determine if integrin alpha subunits co-distributed with OPN at the endometrial–placental interface. Large aggregates of αv, α4 and α5 integrins and OPN are detected in the same regions of the endometrial–placental interface. This suggests that these integrin subunits are forming integrin adhesion complexes (IACs) that sequester OPN which serves as a bridging ligand between these two epithelial surfaces. A rabbit IgG was used as a negative control. Width of fields at 10× magnification is 890 µm and at 40× magnification is 220 µm. LE, luminal epithelium; CE, chorionic epithelium.
Citation: Reproduction 160, 4; 10.1530/REP-20-0273
Co-distribution of β1, β3 and β5 integrin subunits with OPN by immunofluorescence microscopy. Serial sections of endometrial–placental tissues from Day 60 of pregnancy were utilized to determine if integrin beta subunits co-distributed with OPN at the endometrial–placental interface. Large aggregates of β1 and β5 integrins and OPN are localized in the same regions of the endometrial–placental interface. This suggests that these integrin subunits are forming integrin adhesion complexes (IACs) that sequester OPN which serves as a bridging ligand between these two epithelial surfaces. Βeta 3 integrin is present at the interface, but not aggregated into IACs. Width of fields at 10× magnification is 890 µm and at 40× magnification is 220 µm. LE, luminal epithelium; CE, chorionic epithelium.
Citation: Reproduction 160, 4; 10.1530/REP-20-0273
Figure 9 illustrates the co-distribution of the β1 integrin and OPN to epithelial cells expressing PCNA. There are often regions of endometrium along the endometrial–placental interface near the openings of uterine glands that lack epithelial cells, but the sGE in uterine glands were always maintained. A subset of the epithelial cells at the mouths of the uterine glands, presumably sGE, proliferate with some of the epithelial cells expressing β1 integrin and some of the epithelial cells expressing OPN. The spatial distribution of β1 integrin and OPN in those epithelial cells at the mouths of the uterine glands is very similar (Fig. 9).
Top left panel shows double immunofluorescence staining for the β1 integrin subunit (β1, green) and E-cadherin (E-cad, red) at the endometrial–placental interface of sheep. The beta 1 integrin is detectable in epithelial cells at the mouths of the uterine glands (shallow GE, sGE). Top right panel shows double immunofluorescence staining for proliferating cell nuclear antigen (PCNA, red) and E-cad (green). Arrowheads indicate PCNA-positive cells. Some of the sGE cells are proliferating. Bottom panels show double immunofluorescence staining for osteopontin (OPN, green) and E-cad (red). OPN is expressed by some of the sGE cells. Note the loss of endometrial LE on both sides of the endometrial gland and the isolated endometrial LE cells, some of which are proliferating, to the right of the gland. We propose that the remaining endometrial epithelial cells are proliferating and will migrate and re-epithelialize the interplacentomal endometrial–placental interface that will eventually become epitheliochorial. Nuclei are stained with DAPI for histological reference. Width of fields at 10× magnification is 890 µm and at 40× magnification is 220 µm. LE, endometrial luminal epithelium; Tr, trophectoderm; Syn, syncytia; IPlac, interplacentomal region of sheep placentation.
Citation: Reproduction 160, 4; 10.1530/REP-20-0273
Discussion
Results of the present study suggest that multiple integrins interact with OPN at the endometrial–placental interface throughout gestation in sheep,and that the temporal and spatial patterns of expression for these integrins and OPN reflect dynamic changes in the histoarchitecture of placentation as endometrial and placental tissues remodel to adapt to changing physiological requirements. We propose that interactions between integrins and OPN at the endometrial–placental interface of sheep be classified into three stages: (1) cell–cell adhesion during conceptus attachment for implantation; (2) cell migration during the early stages of syncytialization of the endometrial and placental epithelia; and (3) mechanosensation and mechanotransduction in response to tensional forces applied through increasing fetal size and fetal fluid volumes as pregnancy progresses (Bazer et al. 2012).
As noted previously, the temporal and spatial patterns of expression of integrins and OPN suggest that these proteins mediate the attachment of the blastocyst/conceptus to the endometrial LE to initiate implantation in multiple species (Lessey et al. 1992, Tabibzadeh 1992, Klentzeris et al. 1993, Aplin et al. 1994, Aplin 1996). Functional evidence for roles of integrins and OPN in implantation is not lacking. Null mutation of the αv integrin subunit gene in mice leads to peri-implantation lethality (Hynes 1996), while functional blockade of αv integrin, using neutralizing antibody, reduces the number of implantation sites in mice and rabbits (Illera et al. 2000, 2003). When ovine and porcine conceptus trophectoderm cells are cultured with OPN-coated microspheres placed on the apical surfaces of the trophectoderm cells, IACs containing integrins and talin assembled at the surface of these cells around the OPN-coated beads (Johnson et al. 2001, Erikson et al. 2009). Further, studies using affinity chromatography followed by immunoprecipitation revealed that OPN binds specific integrins on cultured ovine and porcine trophectoderm cells and that both ovine and porcine trophectoderm cells adhere to OPN-coated culture plates and migrate through OPN-coated transwell insert filters in culture. In those studies, adhesion and migration was reduced if the RGD integrin binding sequence in OPN was mutated to an RAD sequence that cannot engage integrins, or in the presence of a linear RGD integrin blocking peptide, or when the αv integrin was knocked down in the trophectoderm cells using siRNA (Erikson et al. 2009, Kim et al. 2010). An important series of studies demonstrate that integrins and OPN mediate attachment of blastocysts to endometrial cell monolayer cultures and suggest that integrins and OPN are involved in initial trophoblast invasion during implantation (Kang et al. 2014, Berneau et al. 2019). Also, disruption of the OPN gene in mice results in increased early pregnancy loss and the pups born are significantly smaller than are pups for WT counterparts (Weintraub et al. 2004).
The present results suggest that three integrin receptors potentially assemble at the apical surface of endometrial LE to mediate conceptus attachment during the peri-implantation period of pregnancy in sheep. These receptors are αvβ3, αvβ5 and α4β1. During the peri-implantation period, integrin subunits αv, α4, α5, β1, β3 and β5 are expressed by endometrial LE. Alpha 5 integrin is present in the cytoplasm and only minimal protein is evident at the apical surface of LE cells (Fig. 3), but αv, α4, β1, β3 and β5 are all present at the apical surface of endometrial LE cells during the critical period of conceptus attachment for implantation. Interestingly, unlike the other integrins, β3 is limited exclusively to the apical surface of uterine LE during the peri-implantation period and is not expressed thereafter, suggesting that the β3 integrin may play a unique role during implantation in sheep. In addition, β5 integrin expression extends from the endometrial LE into the shallow GE (sGE) where it potentially interacts with trophectoderm papillae that are thought to serve as tethers against which forces necessary to generate elongation are applied and serve as sites of maximal uptake of nutrients in histotroph (Fig. 5). We hypothesize that the continuous thin layer of immunostaining for αv, α4, β1, β3 and β5 at the apical surface of endometrial LE from Days 11 through 16 represents a role for αvβ3, αvβ5 and α4β1 during conceptus elongation and the adhesion cascade of implantation.
Sheep have synepitheliochorial placentation in which fusion of trophectoderm cells and their invasion into the endometrial LE occurs. Trophectoderm cell fusion first appears between Days 14 and 16 of gestation in sheep (Wooding 1984). These trophoblast giant cells (TGCs) invade into the endometrial LE where some of the TGCs fuse with other TGCs to form an extended syncytia in the placentomes. However, other TGCs appear to engulf endometrial LE cells that are undergoing apoptosis and then migrate into the endometrial stroma leaving gaps in the endometrial LE layer (see the gaps in endometrial LE on Days 20 through 35 of gestation in Figs 1, 2, 3, 4, 5 and on Day 30 of gestation in Fig. 9) (Seo et al. 2019). These gaps are particularly prominent in the interplacentomal regions of placentation in sheep. Clearly this is a histologically complex and dynamic period of placentation in the sheep. There is extensive tissue remodeling at the endometrial–placental interface while increasing shear and tensional forces are applied by the growing fetus and increasing volumes of fetal fluids (Bazer et al. 2012).
In the present study, we observed large, infrequent, intermittent, and disorganized aggregates of αv, α4, α5, β1 and β5 integrin subunits, suggestive of IACs, assemble at the interface between the endometrium and placenta from Days 20 through 55. Some of these aggregates were found at the apical surfaces where endometrial LE remains attached to trophectoderm/chorion, but other IACs form in regions where the conceptus trophectoderm contacts the endometrial stroma and still other aggregates form at the base of endometrial LE. The size, composition, cell signaling activity and adhesion strength of IACs are force-dependent (Hynes 1987, Vogel 2006). These IACs act to transmit force placed on the cell by the ECM at sites of adhesion, or vice versa, and serve as signaling centers where cell signaling pathways involved in cell growth, proliferation, survival, gene expression, development, tissue repair, migration and invasion are activated (Humphries et al. 2019). We propose that the IACs present in these regions are crucial to maintaining contact between the endometrium and developing placenta.
As erosion of the endometrial LE continues, the sGE at the mouths of the uterine glands never degrade (Figs 1, 2, 3, 4, 5 and 9) and could serve as a stable reserve of epithelial stem cells to replace the endometrial LE that is lost during this period of placentation. Indeed a similar process occurs for replacement of the surface mucous cells of the stomach by stem cells that migrate from the necks of the gastric glands and for the replacement of enterocytes by stem cells that migrate from the crypts of Lieberkuhn in the intestines. Results of the present study support this idea because they indicate that the sGE proliferate, express integrins, and express OPN. OPN mRNA is not detectable at the endometrial–placental interface of sheep. OPN is only present in the endometrial GE (Johnson et al. 1999a). We propose that OPN is secreted from the endometrial GE and binds to the integrins expressed by the proliferating sGE. The sGE then utilize the interactions between integrins and OPN to migrate out of the endometrial GE and repopulate the endometrial LE in the interplacentomal regions of sheep during placentation. Certainly, the dynamic interactions between integrins and ECM are common modes of action to induce cell migration and formation of IACs. In this capacity, the IACs interact with the actin cytoskeleton to give the cell traction as it migrates along an ECM. At the leading edge of the migrating cells, there are nascent, immature, focal complexes formed that can then mature into IACs as the cells become stably attached to the ECM and more force is exerted on the focal complex (Hood & Cheresh 2002, Huttenlocker & Horwitz 2011). Further, OPN supports robust ovine and porcine trophectoderm cell haptotactic migration (Erikson et al. 2009, Kim et al. 2010). Finally, an important concept concerning Days 20 through 55 of ovine gestation is that the histological instability at the endometrial–placental interface during this period of placentation would also allow for OPN secreted by endometrial GE to spread throughout the interface between the placental chorion and endometrial LE to form an ECM reservoir for subsequent interactions with integrins expressed by both chorion and endometrial LE.
By Day 60 of pregnancy, the interplacentomal endometrial–placental interface stabilizes into a continuous seal between the endometrial LE and chorion except at the openings of uterine glands (reviewed in Johnson et al. 2018). Here the chorion never fuses with the endometrial LE; rather it forms a pocket referred to as an areola. Secretions from the endometrial GE are absorbed and transported across the areolae of the chorioallantoic placenta by fluid phase pinocytosis for release into the fetal circulation (Renegar et. al. 1982, Dantzer & Leiser 1993). The tall columnar cells of the areola are specialized to transport large macromolecules across the placenta, and it is essential that all histotroph be directed to these structures. Therefore, tight attachment between chorion and endometrial LE in non-areolar regions of the interplacentomal endometrial–placental interface must be present even though this interface is exposed to exponential increases in mechanical forces generated from the growing fetus and increases in volumes of fetal fluids (Bazer et al. 2012). We observed immunostaining for αv, α4, α5, β1 and β5 integrin subunits, suggestive of IACs, precisely distributed at the apical surfaces of apposed endometrial LE and chorion along extensive expanses of the endometrial–placental interface on Days 60 and 80 of gestation (Fig. 6). Each integrin subunit shows a very organized pattern of expression in which the integrin subunit localizes to IACs at the apical surfaces of both endometrial LE and chorionic epithelia, resulting in a gap between the apposed surfaces where adhesive ECM molecules could reside. Further aggregates for each integrin co-localize to the same regions of endometrial–placental interface (Fig. 6). We propose that the temporal and spatial formation of these mature IACs represent engagement of these integrins with the ECM to stabilize adhesion between endometrial LE and chorionic epithelium in response to the increasing mechanical stress being placed on this interface by the ever increasing size of the fetus and volumes of fetal fluids (Bazer et al. 2012). An important advancement of the present studies is the clear co-distribution of OPN and these multiple integrins within these large protein aggregates as constituents of IACs at the endometrial–placental interface of sheep (Figs 7 and 8). This is the first study to definitively co-localize OPN with large integrin subunit-containing IACs, suggesting that OPN is acting as a bridging ligand between these IACs to maintain contact between the endometrial LE and the chorionic epithelium.
In conclusion, the comprehensive nature of these results over the majority of ovine pregnancy provide evidence that integrins interacting with OPN play roles in implantation and placentation and provide insights into the mechanics of placental development in sheep. These results may have implications concerning implantation and placentation in other species that also exhibit prominent expression of integrins and OPN within endometrial and placental tissues.
Declaration of interest
Greg A Johnson is on the editorial board of Reproduction. Greg A Johnson was not involved in the review or editorial process for this paper, on which he is listed as an author. The other authors have nothing to disclose.
Funding
This project was supported by Agriculture and Food Research Initiative Competitive Fellowship Grant no. 2012-67011-19892 to J W F from the USDA National Institute of Food and Agriculture.
Author contribution statement
H S performed the immunofluorescence localization, image capture and assembly for Figures 8 and 9, and contributed to the interpretation of data and manuscript preparation. JWF performed the colorimetric immunohistochemistry and contributed to animal husbandry and manuscript preparation. RCB performed image capture for Figures 1–7 and contributed to the interpretation of data and manuscript preparation. FWB co-directed the study, contributed to animal husbandry, and contributed to the interpretation of data and manuscript preparation. GAJ co-directed the study, assembled Figures 1–7, and contributed to the interpretation of data and manuscript preparation.
References
Albelda SM & Buck CA 1990 Integrins and other cell adhesion molecules. FASEB Journal 4 2868–2880. (https://doi.org/10.1096/fasebj.4.11.2199285)
Aplin JD 1996 The cell biology of human implantation. Placenta 17 269–275. (https://doi.org/10.1016/s0143-4004(96)90050-8)
Aplin JD, Seif MW, Graham RA, Hey NA, Behzad F & Campbell S 1994 The endometrial cell surface and implantation. Expression of the polymorphic mucin MUC-1 and adhesion molecules during the endometrial cycle. Annals of the New York Academy of Sciences 734 103–121. (https://doi.org/10.1111/j.1749-6632.1994.tb21739.x)
Apparao KB, Illera MJ, Beyler SA, Olson GE, Osteen KG, Corjay MH, Boggess K & Lessey BA 2003 Regulated expression of osteopontin in the peri-implantation rabbit uterus. Biology of Reproduction 68 1484–1490. (https://doi.org/10.1095/biolreprod.101.001347)
Bazer FW, Spencer TE & Thatcher WW 2012 Growth and development of the ovine conceptus. Journal of Animal Science 90 159–170. (https://doi.org/10.2527/jas.2011-4180)
Berneau SC, Ruane PT, Brison DR, Kimber SJ, Westwood M & Aplin JD 2019 Characterization of osteopontin in an in vitro model of embryo implantation. Cells 8 432–445. (https://doi.org/10.3390/cells8050432)
Bowen JA, Bazer FW & Burghardt RC 1996 Spatial and temporal analysis of integrin and Muc-1 expression in porcine uterine epithelium and trophectoderm in vivo. Biology of Reproduction 55 1098–1106. (https://doi.org/10.1095/biolreprod55.5.1098)
Brown LF, Berse B, Van de Water L, Papadopoulos-Sergiou A, Perruzzi CA, Manseau EJ, Dvorak HF & Senger DR 1992 Expression and distribution of osteopontin in human tissues: widespread association with luminal epithelial surfaces. Molecular Biology of the Cell 3 1169–1180. (https://doi.org/10.1091/mbc.3.10.1169)
Burghardt RC, Burghardt JR, Taylor JD II, Reeder AT, Nguyen BT, Spencer TE & Johnson GA 2009 Enhanced focal adhesion assembly reflects increased mechanosensation and mechanotransduction along the maternal/conceptus interface during pregnancy in sheep. Reproduction 137 567–582. (htps://doi.org/10.1530/REP-08-0304)
Butler WT, Ridall AL & McKee MD 1996 Osteopontin. In Principals of Bone Biology, pp. 167–181. New York: Academic Press.
Dantzer V & Leiser R 1993 Microvascularization of regular and irregular areolae of the areola-gland subunit of the porcine placenta: structural and functional aspects. Anatomy and Embryology 188 257–267. (https://doi.org/10.1007/BF00188217)
Denhardt DT & Guo X 1993 Osteopontin: a protein with diverse functions. FASEB Journal 7 1475–1482. (https://doi.org/10.1096/fasebj.7.15.8262332)
Erikson DW, Burghardt RC, Bayless KJ & Johnson GA 2009 Secreted phosphoprotein 1 (SPP1, osteopontin) binds to integrin alphavbeta6 on porcine trophectoderm cells and integrin alphavbeta3 on uterine luminal epithelial cells, and promotes trophectoderm cell adhesion and migration. Biology of Reproduction 81 814–825. (https://doi.org/10.1095/biolreprod.109.078600)
Gallant ND, Michael KE & García AJ 2005 Cell adhesion strengthening: contributions of adhesive area, integrin binding, and focal adhesion assembly. Molecular Biology of the Cell 16 4329–4340. (https://doi.org/10.1091/mbc.e05-02-0170)
Garlow JE, Ka H, Johnson GA, Burghardt RC, Jaeger LA & Bazer FW 2002 Analysis of osteopontin at the maternal-placental interface in pigs. Biology of Reproduction 66 718–725. (https://doi.org/10.1095/biolreprod66.3.718)
Giancotti FG & Ruoslahti E 1999 Integrin signaling. Science 285 1028– 1032. (https://doi.org/10.1126/science.285.5430.1028)
Hood JD & Cheresh DA 2002 Role of integrins in cell invasion and migration. Nature Reviews: Cancer 2 91–100. (https://doi.org/10.1038/nrc727)
Horton ER, Byron A, Askari JA, Ng DHJ, Millon-Frémillon A, Robertson J, Koper EJ, Paul NR, Warwood S & Knight D et al.2015 Definition of a consensus integrin adhesome and its dynamics during adhesion complex assembly and disassembly. Nature Cell Biology 17 1577–1587. (https://doi.org/10.1038/ncb3257)
Humphries JD, Chastney MR, Askari JA & Humphries MJ 2019 Signal transduction via integrin adhesion complexes. Current Opinion in Cell Biology 56 14–21. (https://doi.org/10.1016/j.ceb.2018.08.004)
Huttenlocker A & Horwitz AR 2011 Integrins in cell migration. Cold Spring Harbor Perspectives in Biology 3 a005074. (https://doi.org/10.1101/cshperspect.a005074)
Hynes RO 1987 Integrins: a family of cell surface receptors. Cell 48 549–554. (https://doi.org/10.1016/0092-8674(87)90233-9)
Hynes RO 1996 Targeted mutations in cell adhesion genes: what have we learned from them? Developmental Biology 180 402–412. (https://doi.org/10.1006/dbio.1996.0314)
Illera MJ, Cullinan E, Gui Y, Yuan L, Beyler SA & Lessey BA 2000 Blockade of the alpha(v)beta(3) integrin adversely affects implantation in the mouse. Biology of Reproduction 62 1285–1290. (https://doi.org/10.1095/biolreprod62.5.1285)
Illera MJ, Lorenzo PL, Gui YT, Beyler SA, Apparao KB & Lessey BA 2003 A role for alphavbeta3 integrin during implantation in the rabbit model. Biology of Reproduction 68 766–771. (https://doi.org/10.1093/biolreprod/68.3.766)
Johnson GA, Spencer TE, Burghardt RC & Bazer FW 1999a Ovine osteopontin. I. Cloning and expression of mRNA in the uterus during the peri-implantation period. Biology of Reproduction 61 884–891. (https://doi.org/10.1095/biolreprod61.4.884)
Johnson GA, Burghardt RC, Spencer TE, Newton GR, Ott TL & Bazer FW 1999b Ovine osteopontin: II. Osteopontin and αvβ3 integrin expression in the uterus and conceptus during the peri-implantation period. Biology of Reproduction 61 892–899. (https://doi.org/10.1095/biolreprod61.4.892)
Johnson GA, Bazer FW, Jaeger LA, Ka H, Garlow JE, Pfarrer C, Spencer TE & Burghardt RC 2001 Muc-1, integrin and osteopontin expression during the implantation cascade in sheep. Biology of Reproduction 65 820–828. (https://doi.org/10.1095/biolreprod65.3.820)
Johnson GA, Burghardt RC, Bazer FW & Spencer TE 2003 Osteopontin: roles in implantation and placentation. Biology of Reproduction 69 1458–1471. (https://doi.org/10.1095/biolreprod.103.020651)
Johnson GA, Burghardt RC & Bazer FW 2014 Osteopontin: a leading candidate adhesion molecule for implantation in pigs and sheep. Journal of Animal Science and Biotechnology 5 56. (https://doi.org/10.1186/2049-1891-5-56)
Johnson GA, Bazer FW, Burghardt RC, Wu G, Seo H, Kramer AC & McLendon BA 2018 Cellular events during ovine implantation and impact for gestation. Animal Reproduction 15 843–855. (https://doi.org/10.21451/1984-3143-AR2018-0014)
Joyce MM, González JF, Lewis S, Woldesenbet S, Burghardt RC, Newton GR & Johnson GA 2005a Caprine uterine and placental osteopontin expression is distinct among epitheliochorial implanting species. Placenta 26 160–170. (https://doi.org/10.1016/j.placenta.2004.05.009)
Joyce MM, White FJ, Burghardt RC, Muñiz JJ, Spencer TE, Bazer FW & Johnson GA 2005b Interferon stimulated gene 15 (ISG15) conjugates to cytosolic proteins and is expressed at the uterine-placental interface throughout ovine pregnancy. Endocrinology 146 675–684. (https://doi.org/10.1210/en.2004-1224)
Kang YJ, Forbes K, Carver J & Aplin JD 2014 The role of the osteopontin-integrin αvβ3 interaction at implantation: functional analysis using three different in vitro models. Human Reproduction 29 739–749. (https://doi.org/10.1093/humrep/det433)
Kim J, Erikson DW, Burghardt RC, Spencer TE, Wu G, Bayless KJ, Johnson GA & Bazer FW 2010 Secreted phosphoprotein 1 binds integrins to initiate multiple cell signaling pathways, including FRAP1/mTOR, to support attachment and force-generated migration of trophectoderm cells. Matrix Biology 29 369–382. (https://doi.org/10.1016/j.matbio.2010.04.001)
Kimmins S, Lim HC & MacLaren LA 2004 Immuohistochemical localization of integrin alpha v and beta 3 and osteopontin suggests that they do not interact during embryo implantation in ruminants. Reproductive Biology and Endocrinology 2 19. (https://doi.org/10.1186/1477-7827-2-19)
Klentzeris LD, Bulmer JN, Tredosiewicz LK, Morrison L & Cooke ID 1993 Beta-1 integrin cell adhesion molecules in the endometrium of fertile and infertile women. Human Reproduction 8 1223–1230. (https://doi.org/10.1093/oxfordjournals.humrep.a138231)
Lessey BA, Damjanovich L, Coutifaris C, Castelbaum A, Albelda SM & Buck CA 1992 Interin adhesion molocules in the human endometrium. Correlation with the normal and abnormal menstrual cycle. Journal of Clinical Investigation 90 188–195. (https://doi.org/10.1172/JCI115835)
Massuto DA, Kneese EC, Johnson GA, Hooper NH, Burghardt RC, Ing NH & Jaeger LA 2010 Transforming growth factor beta (TGFB) signaling is activated during porcine implantation: proposed role for latency associated peptide-integrins at the conceptus-maternal interface. Reproduction 139 465–478. (https://doi.org/10.1530/REP-09-0447)
Nomura S, Wills AJ, Edwards DR, Heath JK & Hogan BL 1988 Developmental expression of 2ar (osteopontin) and SPARC (osteonectin) RNA as revealed by in situ hybridization. Journal of Cell Biology 106 441–450. (https://doi.org/10.1083/jcb.106.2.441)
Renegar RH, Bazer FW & Roberts RM 1982 Placental transport and distribution of uteroferrin in the fetal pig. Biology of Reproduction 27 1247‑1260. (https://doi.org/10.1095/biolreprod27.5.1247)
Ruoslahti E, Noble NA, Kagami S & Border WA 1994 Integrins. Kidney International: Supplement 44 S17–S22. (https://doi.org/10.1172/JCI114957)
Seo H, Bazer FW, Burghardt RC & Johnson GA 2019 Immunohistochemical examination of trophoblast syncytialization during early placentation in sheep. International Journal of Molecular Sciences 20 4530. (https://doi.org/10.3390/ijms20184530)
Sodek J, Ganss B & McKee MD 2000 Osteopontin. Critical Reviews in Oral Biology and Medicine 11 279–303. (https://doi.org/10.1177/10454411000110030101)
Tabibzadeh S 1992 Patterns of expression of integrin molecules in human endometrium during the menstrual cycle. Human Reproduction 7 876–882. (https://doi.org/10.1093/oxfordjournals.humrep.a137753)
Vogel V 2006 Mechanotransduction involving multimodular proteins: converting force into biochemical signals. Annual Review of Biophysics and Biomolecular Structure 35 459–488. (https://doi.org/10.1146/annurev.biophys.35.040405.102013)
Weintraub AS, Lin X, Itskovich VV, Aguinaldo JG, Chaplin WF, Denhardt DT & Fayad ZA 2004 Prenatal detection of embryo resorption in osteopontin-deficient mice using serial noninvasive magnetic resonance microscopy. Pediatric Research 55 419–424. (https://doi.org/10.1203/01.PDR.0000112034.98387.B2)
Wooding FB 1984 Role of binucleate cells in fetomaternal cell fusion at implantation in the sheep. American Journal of Anatomy 170 233–250. (https://doi.org/10.1002/aja.1001700208)
