|
|
||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Pregnancy |
State Key Laboratory of Reproductive Biology, Institute of Zoology, Chinese Academy of Sciences, Beijing 100080, China
| ABSTRACT |
|---|
|
|
|---|
1,
5,
6, ß1, and ß4) and their ECM ligands, such as laminin (LN), type IV collagen (Col IV), and fibronectin (FN), at the maternal-fetal interface during Gestational Days 15, 25, 50, and 100 and at full term in 20 pregnant rhesus monkeys. Immunohistochemistry and in situ hybridization revealed that a relatively high expression of integrins occurred in trophoblast cells at Gestational Day 15, with the peak level occurring at Day 25. The expression level decreased from Day 50 to term. Along the invasive pathway, expression levels of integrin
1,
5, and ß1 subunits were gradually elevated from the proximal to distal column, reaching peak level in the trophoblast shell, but were reduced in those invasive extravillous cytotrophoblast (EVCT) cells in contact with the decidua. Integrin
1,
5, ß1, and ß4 subunits were also highly expressed in decidual stromal cells and moderately expressed in the maternal epithelium and endothelium. Immunoreactive FN, LN, and Col IV were distributed in EVCT and decidual stromal cells and part of the uterine epithelial and endothelial cells. These data suggest that the correlated expression of integrins and their ECM ligands at the maternal-fetal interface might be involved in regulation of cell proliferation and differentiation and the counterbalanced invasion-accelerating and invasion-restraining processes in trophoblast cells during the early stage of pregnancy.
decidua, implantation, placenta, pregnancy, trophoblast
| INTRODUCTION |
|---|
|
|
|---|
The interaction of integrins with their corresponding ECM ligands plays an important role in embryo development, morphogenesis, and tumorigenesis [46]. Usually, fibronectin (FN) targets the integrin
5ß1 and
3ß1 subunits, laminin (LN) interacts with
6ß4,
1ß1, and
3ß1, and type IV collagen (Col IV) binds to
1ß1 and
2ß1 [7]. Integrins link with ECM through the recognition of Arg-Gly-Asp peptides and then lead to active focal adhesion kinase and trigger the early step of the signal transduction cascade [8, 9]. The role of integrins and ECM in embryo development has been inferred from gene knockout experiments. For example, mice in which the integrin
1 gene had been deleted underwent normal development but had a specific defect in cell adhesion [10]. Homozygous null mutation of the
5 gene led to defects in posterior trunk and yolk sac mesodermal structures by Day 9 of gestation, and the embryo died on Days 1011 [11]. Integrin ß1 mutant mice died at the early stage of postimplantation [12]. Careful examination of implantation sites suggested the importance of the interaction between trophoblastic ß1 integrin and maternal ECM in placentation. FN knockout mice also died during the embryonic stage [13]. Targeted deletion of the LAMC1 gene that encodes the LN
1 subunit caused failure in formation of the basement membrane, leading to earlier embryonic lethality [14]. Col IV may play an important role in the maintenance of pregnancy [15].
The fetal-maternal biochemical dialogue that occurs during embryo implantation and subsequent development has been extensively investigated in rodents; the available data suggest that the interaction of integrins and ECM facilitates the adhesion to and invasion of the uterine endometrium by trophoblast cells [16]. However, because of the limited availability of pathology materials, the related events in humans, especially those occurring at the very early stages of pregnancy, remain unclear despite considerable research efforts. Nonhuman primates share many physiological processes with humans and consequently have been presumed to be good animal models for the investigation of relevant regulatory mechanisms. In an early in vivo study in cynomolgus monkeys (Macaca fascicularis), immunoreactive FN, LN, and Col IV were detected in the distal column of anchoring villi and extravillous cytotrophoblast (EVCT) [1719]. Fazleabas et al. [20] found an increased amount of integrin
1,
3,
6, ß1, and
vß3 in decidualized stromal cells of baboons, which also occurs in humans. High levels of LN and Col IV were also widely distributed throughout the endometrium, including the implantation site. FN was most evident at the implantation site and corresponded to the expression of integrin
5 in invading cytotrophoblast cells. Douglas et al. [21] observed that integrin
vß3/ß5 was expressed in villous cytotrophoblast (CTB) cells and in those cells that had invaded blood vessels of macaque deciduas. Thus, integrin
vß3/ß5 might play an important role in mediating the invasion of the maternal vascular endothelium by EVCT cells. To the best of our knowledge, all available information in pregnant primates was obtained by immunohistochemical assay and only during a specific period of pregnancy. There is little information available on the transcriptional changes of integrins and their corresponding ligands on both sides of the primate maternal-fetal interface during the entire process of gestation.
In this study, we used in situ hybridization and immunohistochemistry to investigate the expression pattern and the temporal and spatial alteration of integrin
1,
5,
6, ß1, and ß4 subunits and the heterodimer
5ß1 and their corresponding ECM ligands (FN, LN, and Col IV) in rhesus monkeys during various periods of pregnancy.
| MATERIALS AND METHODS |
|---|
|
|
|---|
Twenty mature female rhesus monkeys (Macaca mulatta) were housed at the Non-Human Primate Animal Center of the Institute of Science and Technology of Family Planning (Fujian, China). Animals were allowed to mate from 5 days before until 4 days after the day of ovulation, which was estimated from records of the previous two or three normal menstrual cycles. The presumed day of ovulation was considered Day 0 of gestation, and from Day 13 on, type B ultrasound diagnosis was performed to determine the actual gestational day, mainly based on the size of the uterus and conceptus. At Days 15, 25, 50, and 100 and at term, hysterectomies was performed on pregnant monkeys (
3 animals/stage of pregnancy). The project was approved by the local ethics committee at the State Key Laboratory of Reproductive Biology, Institute of Zoology, Chinese Academy of Sciences.
Each sample with an intact maternal-fetal interface was washed twice with PBS buffer and fixed immediately in 4% paraformaldehyde (PFA) at 4°C for 16 h. Fixed tissues were then gradually dehydrated in ethanol and embedded in paraffin. Sections 6 µm thick were collected on Super Frost+ glass slides (Menzel-Gläser, Braunschweig, Germany).
Immunohistochemistry
Paraffin-embedded sections were deparaffinized, rehydrated, and then incubated in 10 mM citrate buffer (pH 6.0) at 95°C for 15 min. After immersion in 1% hydrogen peroxide, the sections were incubated with various primary antibodies at 4°C overnight. Antibodies used in this study were rabbit polyclonal antibodies against FN (F-3648, 1:500) and LN (L-9393, 1:100) (Sigma International, St. Louis, MO); mouse monoclonal antibodies against Col IV (MAB 1910, 1:200), integrin ß1 subunit (MAB 1951, 1:500), integrin
5ß1 (MAB 1969, 1:50); and rabbit polyclonal antibody against integrin
1 (AB 1934, 1:200) (Chemicon International, Temecula, CA). The negative control was assessed by replacing the primary polyclonal antibodies with the nonimmune rabbit serum at the same concentration and replacing the monoclonal antibodies with the preabsorbed antibodies. The purified antigens used were human Col IV (12168-019; Life Technologies, Rockville, MD) and integrin
5ß1 (CC1026; Chemicon). Final visualization was achieved using Envision Kits (DAKO, Carpinteria, CA). Counterstaining with hematoxylin was carried out before slides were mounted. Staining intensity was evaluated under a microscope (Olympus, Tokyo, Japan) by three observers who did not know the stage of gestation from which samples had been obtained and semiquantified visually as absent (-), weak (±), moderate (+), strong (++), or very strong (+++).
Preparation of Digoxigenin-Labeled Probes
Digoxigenin (DIG)-labeled RNA probes were used for in situ hybridization. Total RNA was isolated from human chorionic villi at 67 wk of gestation. Specific cDNA fragments of integrin
1, ß1, and
5 were amplified by reverse transcription polymerase chain reaction with primer sets designed according to sequences from GenBank (Table 1). The pGEM-T easy vector (Promega, Madison, WI) was used to reconstruct the recombinant plasmids containing these fragments. After confirmation by sequencing, the recombinant plasmids were linearized with SalI or NcolI (Promega) and purified with QIAquick Spin Columns (Qiagen, Valencia, CA), respectively, before being used as templates for in vitro transcription reactions to synthesize DIG-labeled RNAs according to the manufacturer's instruction provided with the DIG-RNA Labeling System (Enzo Diagnostics, Indianapolis, IN). The transcription mixture (20 µl) contained 1 µg template cDNA, 2 µl 10x nucleotide triphosphate labeling mixture, 1 µl RNase inhibitor, and 2 µl SP6 or T7 RNA polymerase. The transcription reaction was performed at 37°C for at least 2 h, and template cDNA was then digested by RNase-free DNase. The reaction was stopped by adding 0.2 M EDTA. DIG-labeled sense and antisense RNA probes were stored at -80°C at a concentration of 0.1 µg/µl.
|
In Situ Hybridization
In situ hybridization was performed as previously described [22]. Paraffin-embedded sections were routinely deparaffinized and rehydrated. After treatment with 0.2 N HCl for 15 min, slides were denatured at 70°C in 2x saline sodium citrate (SSC) for 15 min and then digested with 4 µg/ml of proteinase K (Gibco, Rockville, MD) for another 15 min. Postfixation were performed in 4% PFA at room temperature for 10 min, followed by incubation twice in PBS containing 0.1% active diethyl pyrocarbonate (Fluka, Seelze, Germany) for 15 min. Acetylation was then carried out in Tris-EDTA-acetate buffer containing 0.5% acetic anhydride for 10 min with subsequent equilibration in 5x SSC for 15 min. The slides were prehybridized for 4 h at 58°C in prehybridization buffer (50% formamide, 20 mM Tris-HCl, 50 mM EDTA, 0.5 mg/ml tRNA Coli, 100 mM dithiothreitol) and further hybridized for 18 h at 58°C in fresh hybridization buffer containing 1 ng/µl antisense probes. After washing in consecutive baths of 2x SSC and 0.1x SSC at 65°C for 1 h each, slides were blocked with Tris-HCl buffer containing 0.5% blocking reagent (Boehringer Mannheim, Mannheim, Germany) and then incubated at room temperature for 2 h with alkaline phosphatase-coupled anti-DIG antibody (dilution 1:500). Color development was performed in buffer II (100 mM Tris-HCl, 100 mM NaCl, 50 mM MgCl2, pH 9.5) containing 4.5 µl nitroblue tetrazolium and 3.5 µl 5-bromocresyl-3-indolylphosphate (Boehringer Mannheim). Nonspecific staining was removed by rinsing with 95% ethanol. Slides were mounted in resin after dehydration with ethanol and xylene. Negative controls was processed by replacing antisense probes with sense probes in hybridization buffer. Results were assessed based on the evaluations of three observers who did not know the stage of gestation from which samples had been obtained.
| RESULTS |
|---|
|
|
|---|
Expression of Integrins at the Maternal-Fetal Interface During Pregnancy
Spatial expression of integrins on Gestational Day 25
On Gestational Day 25, integrated structure at the maternal-fetal interface, including villi with cytotrophoblast column, shells, and invading EVCT cells in maternal blood vessels, was observed. Expression of integrins at this interface was extensive. In situ hybridization revealed that the integrin
1 and ß1 subunits were expressed in villous CTB cells, syncytiotrophoblast (STB) cells, mesenchyme, and endothelium and in column CTB and invasive EVCT cells (Fig. 1, A and B). Maternal epithelium, endothelium, and decidual stromal cells also transcribed integrin
1 and ß1 subunits at different levels. Along the invasive pathway, the level of
1 and ß1 mRNA increased in order from villous CTBs to proximal column CTBs to distal column CTBs to EVCT cells in the trophoblast shell (Fig. 1, A and B) but decreased significantly in those EVCT cells that were adjacent to decidual cells (Fig. 2G).
|
|
Messenger RNA of the
5 subunit was detected in villous CTBs, STBs, and mesenchyme and in column CTBs and most EVCT cells distributed at the maternal-fetal interface. Along the invasive pathway, the expression level increased slightly from villous CTBs to column CTBs and EVCT cells in the trophoblast shell (Fig. 1C) but was reduced in EVCT cells in contact with maternal decidual cells (Fig. 2B). Uterine epithelium, part of the maternal blood vessel endothelium, and decidual stromal cells also transcribed
5 subunit mRNA.
Integrin
6 and ß4 mRNA could be detected in villous CTBs and some STBs. In addition,
6 subunit was also expressed in some column CTBs, and ß4 subunit was expressed in EVCT cells but at a lower levels than in villous CTBs (Fig. 1, G and H).
Immunoreactive
1 and ß1 subunits were also detected in the cytoplasm and cell membrane of villous CTBs, STBs, mesenchyme, column CTBs, and EVCT cells. The spatial pattern of protein concentration change was the same as that of mRNA expression. In maternal tissue, positive staining was observed in the epithelium, vascular endothelium, and decidual stromal cells. Staining in maternal decidual stromal cells was relatively stronger than that in trophoblast cells (Fig. 3, C and D).
|
Immunohistochemical assay also revealed positive staining for
5ß1 in the cell membrane and cytoplasm of column CTBs, invasive EVCT cells, and villous mesenchyme, whereas staining was negative in villous CTBs and STBs. Uterine epithelium, vascular endothelium, and decidual stromal cells also showed positive staining for integrin
5ß1. Immunoreactivity of integrin
5ß1 in trophoblast cells displayed an alteration pattern along the invasive pathway similar to that of
5 subunit mRNA levels, as revealed by in situ hybridization (data not shown).
Temporal alteration of integrin expression levels at different stages of gestation
Transcripts of
1 and ß1 subunits could be detected as early as Gestational Day 15. In trophoblast cells, the highest expression level was present on Day 25. From Day 50 to full term, expression of
1 and ß1 subunits declined gradually in EVCT cells. The alteration of ß1 mRNA in the maternal epithelium coincided with the same process in EVCT cells. In decidual stromal cells, levels of the ß1 subunit remained high from Day 15 to Day 100, finally declining at full term (Fig. 2, GK). The expression level of the
1 subunit in stromal cells showed no significant changes during the process of gestation (data not shown).
The mRNA level of the
5 subunit increased from Gestational Day 15 to Day 25 and then declined from Day 50 on. Expression of this subunit could hardly be detected in most EVCT cells from Day 100 to full term.
Little mRNA of the
6 subunit was detected in EVCT cells during gestation, whereas the expression of ß4 was detected from Day 15 on. The expression level of this subunit peaked at Day 25 and then declined significantly from Day 50 on, falling to a much lower level at term. In maternal decidual cells, expression of
6 and ß4 subunits showed no significant changes during pregnancy (data not shown).
Immunoreactivity of the
1 and ß1 subunits exhibited a pattern similar to that for expression of the corresponding mRNA. Immunostaining of
5ß1 resembled the changes in
5 mRNA expression. In maternal decidual cells, immunoreactive staining for
5ß1 was slightly stronger at Day 25 (Fig. 2, AE). Integrin immunoreactivity is summarized in Table 2.
|
Distribution of FN, LN, and Col IV at the Maternal-Fetal Interface During Pregnancy
Spatial distribution of FN, LN, and Col IV on Day 25 of gestation Distribution patterns of immunoreactive FN and Col IV at the maternal-fetal interface were similar. Signals for both kinds of ECM ligand were positive in the cytoplasm and extracellular region from the distal column CTBs of anchoring villi to invasive EVCT cells. Staining was most intensive in many EVCT cells in the trophoblast shell but was significantly reduced in invasive EVCT cells that were in contact with the uterine endothelium (Figs. 2M and 3B). Few invasive EVCT cells with positive staining were observed wandering in the maternal decidua. Villous CTBs and STBs were not stained, but mesenchyme and villous endothelial cells were moderately stained. On the maternal side of the interface, strong staining of FN and mild staining of Col IV were observed in the cytoplasm and extracellular region of decidual stromal cells (Fig. 3B). Some vascular endothelial cells adjacent to the junctional zone also showed positive immunoreaction for the two kinds of ECM ligand. The difference in distribution of FN and Col IV was that Col IV was also found in basement membranes of villi and some uterine epithelial cells (Fig. 3B), but FN was absent from these areas.
Immunoreactivity of LN was extensively in the cytoplasm and extracellular regions of column CTBs and invasive EVCT cells. The strength increased along the invasive pathway, with the peak level observed in the trophoblast shell and those EVCT cells invading into maternal blood vessels (Fig. 3A). Positive staining was also found in villous CTBs, some STBs, mesenchyme, and basement membranes of villi. On the maternal side of the interface, LN was detected in some uterine epithelial and vascular endothelial cells, but no staining was observed in most decidual stromal cells (Fig. 3A).
Temporal alteration of FN, LN, and Col IV at different stages of gestation Strong immunoreactivity for FN and Col IV in EVCT cells could be observed as early as Day 15 and peaked on Day 25. Staining intensity began to decrease from Day 50 on. From Day 100 to full term, the signal decreased to a much lower level. In maternal decidual cells, staining for FN increased gradually from Day 15 to Day 25 and remained at a high level until Day 100. At full term, immunoreactivity for FN decreased (Fig. 2, MQ). Immunoreactivity for Col IV in decidual cells was extremely weak compared with that in EVCT cells, and no significantly alteration was observed during gestation.
From Day 15 to Day 25, the immunoreactivity of LN in EVCT cells was relatively strong and then declined significantly from Day 50 on. By full term, most EVCT cells, except those distributed in maternal blood vessels, did not stain. ECM immunostaining intensity is summarized in Table 2.
| DISCUSSION |
|---|
|
|
|---|
1,
5,
6, ß1, and ß4 subunits at the maternal-fetal interface of the rhesus monkey during the whole process of gestation at both the protein and mRNA levels. The results obtained by in situ hybridization were highly consistent with those obtained by immunohistochemistry.
In the rhesus monkey, the blastocyst attaches to the maternal endometrium on Day 9 of gestation. Trophoblast cells penetrate the uterine epithelium shortly after this attachment has taken place. Up to about Day 15, the conceptus can be detected by ultrasound diagnosis. The early stage of pregnancy lasts for about 50 days, during which implantation occurs and an intimate connection between fetus and uterus is well established. Days 50100 can be considered the middle stage of pregnancy. Our data revealed strong expression of various integrins and their corresponding ECM ligands, such as integrin
5ß1 and FN, and integrin
1ß1 and LN and Col IV, as early as Day 15. Expression increased to peak levels at Day 25, when large numbers of invasive EVCT cells were clustered at the maternal-fetal interface. Thus, expression of ECM proteins in macaque placenta occurs at the very early stage of gestation, i.e., on Day 15. Blankenship et al. [17, 18] studied the distribution of immunoreactive Col IV, FN, and LN in macaque placenta and found relatively strong positive staining for the three matrix proteins in the distal column on Gestational Day 22 until Day 53, consistent with our data. Wang et al. [23] also reported the expression of matrix metalloproteinase (MMP) 2, MMP-9, and MMP-14 in trophoblast cells of rhesus monkey during this stage. Previously, we reported that LN, FN, and vitronectin (VN) promoted migration of cultured human CTB cells with induced expression of MMP-9, tissue inhibitor of metalloproteinase (TIMP) 1, and TIMP-3 [24]. Rhesus monkey EVCT cells might also possess such an autoregulatory mechanism in controlling cell invasion. Based on our observation, the period between Day 15 to Day 25 seems the most important time for the invasion of the maternal endometrium by trophoblasts in rhesus monkeys. The upregulated integrins and ECM in EVCT cells probably facilitate adhesion of the conceptus to the uterine epithelium during this period. However, the interaction between integrins and ECM in EVCT might induce the expression and secretion of certain MMPs that destroy the ECM of maternal tissue and thus promote invasion of EVCT cells into the uterine endometrium and blood vessels. After Day 50 when placental circulation is established and implantation is completed, the adhesive and invasive potential of trophoblast cells decreases and they consequently display lower levels of integrins and ECM ligands. Blankenship et al. [18] also demonstrated the decreased immunoreactivity for Col IV, FN, and LN in macaque placenta after Day 53. The production of certain kinds of ECM ligands and integrins in column would contribute to cell proliferation and differentiation and to tissue reconstruction during the early stage of gestation. This hypothesis has received support from other research [25, 26].
Our data indicate that along the invasive pathway, the expression pattern of integrins at the maternal-fetal interface of the rhesus monkey changed gradually from
6ß4 in villous CTBs to large amounts of
1ß1 and
5ß1 in invading EVCT cells. Damsky et al. [27] and Aplin [28] reported the same integrin switching pattern in the human placenta during the first trimester, and thus differentiation into the extravillous lineage may be indicated by cell surface features, including upregulation of integrin
5ß1 and
1ß1 and loss of
6ß4. Using an in vitro invasion model, Damsky et al. [29] further demonstrated that interactions involving LN or Col IV and their
1ß1 integrin receptor promoted invasion of EVCT cells, whereas interactions between FN and integrin
5ß1 inhibited invasion. They also suggested that integrin switching observed during differentiation may have significant functional consequences for CTB invasion and that counterbalanced invasion-accelerating and invasion-restraining adhesion mechanisms may coordinately exist in differentiating CTBs. In the present study, all three kinds of ECM ligands, including FN, LN, and Col IV, were detected in EVCT cells, similar to those observed in human placenta [27, 30] Therefore, the upregulated expression of integrin
1ß1 and
5ß1 and the corresponding ECM ligands in EVCT cells of the rhesus monkey might also contribute to counterbalanced regulation of the trophoblast invasion into the endometrium during early pregnancy. These results indicate that some similar mechanisms for the regulation of trophoblast invasion may exist in humans and rhesus monkeys, although a number of distinct differences in placental features, such as formation of epithelial plaque and secondary placenta, depth of penetration, and degree of decidual reaction, have been documented [31].
As trophoblast cells invade the endometrium, they are confronted with various basement membranes and matrix substrates in maternal tissue. LN was widely distributed in the uterine epithelium, the maternal vascular endothelium, and the epithelial basement membranes and blood vessels. In addition, relatively weak immunoreactivity for LN and Col IV and intensive FN immunoreactivity was found in decidual stromal cells. These data are consistent with previous findings in baboons [20]. In contrast, in the human placenta during the first trimester, extensive expression of Col IV, LN, and FN was recognized in the pericellular regions of decidual stromal cells [30, 32]. Considering the invasion-accelerating role mediated by the interaction of Col IV and LN with the integrin
1ß1 receptor [29], the low level of LN and Col IV observed in rhesus monkey decidual stromal cells might jointly limit the invasion of trophoblast cells. Concurrently, the level of FN in decidual stromal cells increased from Day 15 to Day 25 and remained high until Day 100. Meanwhile, the expression of integrin
5ß1 in EVCT cells was high from Day 15 to Day 25 and declined from Day 50 on. Considering the invasion-inhibitory role mediated by the binding of FN and integrin
5ß1 [29], the interaction between FN in decidual stromal cells and integrin
5ß1 in EVCT cells might restrain the further migration of EVCT cells into the maternal decidua. All of these integrins and ECM ligands in EVCT cells were downregulated when they were in contact with the maternal decidua. Blankenship et al. [18] also demonstrated that immunostaining for Col IV, FN, and LN decreased abruptly at the fetal-maternal interface in pregnant macaques. Therefore, there may be some regulatory factors at the rhesus monkey maternal-fetal interface that limit the depth of trophoblast invasion, partly by downregulating the expression of adhesive molecules and ECM proteins in EVCT cells after contact with the decidua. The regulatory signals are very likely derived from the maternal side, which may be why trophoblast cell penetration into the decidua is much lower and wandering invasive trophoblast cells are so scarcely distributed in rhesus monkey decidua compared with that of humans. However, even with shallow implantation, pre-eclampsia rarely occurs in the rhesus monkey. Many EVCT cells with high levels of expression of LN and integrin
1ß1 were present in uterine blood vessels. Moreover, the epithelioid EVCT cells were also observed along the wall of uterine blood vessels that might have replaced the vascular endothelial cells. Therefore, the reconstruction of uterine blood vessels seemed fully successful, avoiding pre-eclampsia. However, using the explant model of placental anchoring site development, Aplin et al. [33] suggested that integrin
5ß1-FN interactions contributed significantly to anchorage of the placenta to the uterine extracellular matrix. In rhesus monkeys, high levels of FN at the maternal-fetal interface might also play an important role in anchoring villi to maternal tissue through binding with integrin
5ß1 on the surfaces of EVCT cells. The decreased expression of these molecules at full term may play a part in facilitating placental detachment.
EVCT cells invading into maternal blood vessels always possessed strong LN immunoreactivity but weak or negative Col IV and FN reactivity. However, integrin
1ß1, the corresponding receptor of LN and Col IV, was expressed by these cells. Previous studies also revealed the expression of MMP-2 and MMP-9 in these cells during pregnancy in rhesus monkeys [34]. This group of EVCT cells differs considerably from others distributed at the maternal-fetal interface. Establishment of placental circulation is a critical event during early pregnancy, involving adhesion of invading trophoblast cells to the vascular endothelium and basement membrane, followed by penetration of the blood vessels. Endovascular trophoblast cells replace the endothelial cells of the maternal spiral arteries to produce a low-resistance circulation supplying the placental bed [35]. High levels of LN and integrin
1ß1 receptor in endovascular trophoblast cells might autoregulate the expression of certain MMPs to enhance the degradation of the vascular basement membrane. However, LN is also distributed in the maternal vascular endothelium and basement membrane, which may facilitate the adhesion and migration of trophoblast cells toward the vascular lumina.
The spatial and temporal changes in transcription of integrin subunits
1,
5,
6, ß1, and ß4 in the rhesus monkey endometrium were also first revealed in this study by in situ hybridization. Uterine epithelial cells expressed the integrin ß1 subunit, whereas only some of the cells were positive for the
1 subunit. Integrin
1, ß1, and
5 subunits were also transcribed in some of the endothelial cells. These molecules might mediate cell-cell interactions between trophoblasts and uterine epithelial and endothelial cells, thus facilitating the degradation of the epithelium and maternal blood vessels. High expression levels of all of these integrin subunits were exhibited by decidual stromal cells, and their expression patterns were in accordance with those determined by immunohistochemical assay in baboons [20] and similar to those found in humans [36, 37]. Interactions between ECM and integrin receptors trigger an early signal transduction cascade, which results in expression of special genes and affects cell proliferation, differentiation, and survival [8, 9]. Therefore, the existence of integrins and their ECM ligands in decidual cells suggests their involvement in the autoregulation of decidual cell behaviors, such as occurrence of the decidual reaction and limitation of trophoblast invasion by producing a certain amount of TIMPs. Our data demonstrate the coherent alteration pattern of integrin ß1 and FN in decidual stromal cells and the expression of MMPs and TIMPs in decidual cells [23]. Previous findings have indicated that the interaction of FN with integrin
5ß1 could induce the expression and regulate the activity of MMP-9 [38]. All these data support our hypothesis.
This study is the first to document the spatial and temporal expression patterns of integrins and their corresponding ECM ligands at the maternal-fetal interface in rhesus monkeys during the entire course of gestation. Our findings suggest the existence of certain differences between rhesus monkeys and humans in adhesion and invasion of trophoblast into uterine endometrium during early gestation and subsequent placental development.
| ACKNOWLEDGMENTS |
|---|
| FOOTNOTES |
|---|
2 Correspondence: Yan-Ling Wang and Yun-Shang Piao, State Key Laboratory of Reproductive Biology, Institute of Zoology, Chinese Academy of Sciences, 25 Bei Si Huan Xi Rd., Haidian, Beijing 100080, China. FAX: 86 10 62529248; wangyl{at}panda.ioz.ac.cn and lmc{at}panda.ioz.ac.cn ![]()
Received: 5 January 2003.
First decision: 31 January 2003.
Accepted: 9 April 2003.
| REFERENCES |
|---|
|
|
|---|
5 integrin-deficient mice. Development 1993 119:1093-1105[Abstract]
6ß4 in human trophoblast and its loss from extravillous cells. Placenta 1993 14:203-215[CrossRef][Medline]This article has been cited by other articles:
![]() |
O. Shynlova, S. J. Williams, H. Draper, B. G White, D. J MacPhee, and S. J Lye Uterine Stretch Regulates Temporal and Spatial Expression of Fibronectin Protein and Its Alpha 5 Integrin Receptor in Myometrium of Unilaterally Pregnant Rats Biol Reprod, November 1, 2007; 77(5): 880 - 888. [Abstract] [Full Text] [PDF] |
||||
![]() |
S X Bai, Y L Wang, L Qin, Z J Xiao, R Herva, and Y S Piao Dynamic expression of matrix metalloproteinases (MMP-2, -9 and -14) and the tissue inhibitors of MMPs (TIMP-1, -2 and -3) at the implantation site during tubal pregnancy Reproduction, January 1, 2005; 129(1): 103 - 113. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. Cheng, Y. Li, Y. Omoto, Y. Wang, T. Berg, M. Nord, P. Vihko, M. Warner, Y.-S. Piao, and J.-A. Gustafsson Differential Regulation of Estrogen Receptor (ER){alpha} and ER{beta} in Primate Mammary Gland J. Clin. Endocrinol. Metab., January 1, 2005; 90(1): 435 - 444. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |