| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Pregnancy; |
Prince Henry's Institute of Medical Research,3 Clayton, 3168 Victoria, Australia
Department of Obstetrics and Gynecology,4 Monash University, Clayton, 3168 Victoria, Australia
ABSTRACT
Embryo implantation is a complex process involving blastocyst attachment to the endometrial epithelium and subsequent trophoblast invasion of the decidua. We have previously shown that the chemokines CX3CL1 and CCL14 are abundant in endometrial vasculature, epithelial, and decidual cells at this time, and that their receptors, CX3CR1 and CCR1, are present on invading human trophoblasts. CX3CL1 and CCL14 promote trophoblast migration. We hypothesized that these endometrial chemokines promote trophoblast migration by regulating adhesion molecules and extracellular matrix (ECM) components on the trophoblast, similar to mechanisms used in leukocyte trafficking. Trophoblast cells (AC1M-88) used previously showed a marked increase in adhesion to fibronectin following treatment with CX3CL1 and CCL14. Alterations in trophoblast adhesion and ECM following chemokine stimulation were examined using pathway-specific oligo-arrays and quantitative real-time RT-PCR. More than 30 genes were affected by CX3CL1 treatment, and 15 genes were found to be regulated by CCL14 treatment. Real-time RT-PCR quantitation revealed significant changes in the mRNA transcripts of alpha-catenin (CTNNA1), extracellular matrix protein 1 (ECM1), osteopontin (SPP1), integrin alpha 6 (ITGA6), matrix metalloproteinase 12 (MMP12), and integrin beta 5 (ITGB5) following chemokine treatment. Several of these genes have previously been implicated in implantation. Immunohistochemistry confirmed the presence of integrin alpha 6 and SPP1 protein in first-trimester human implantation sites. The temporal and spatial expression of chemokines, their receptors, adhesion, and ECM at the maternal-fetal interface emphasizes an important role in the controlled directional migration of trophoblasts through the maternal decidua. For the first time, this study demonstrates the direct effects of CX3CL1 and CCL14 on trophoblast adhesion molecules and ECM, suggesting mechanisms by which trophoblast cells migrate during early pregnancy.
cytokines, placenta, pregnancy
The early stages of human embryo implantation depend on a series of tightly controlled key interactions between maternal and fetal cells. Trophoblast cells undergo unique differentiation processes as they progress to an invasive phenotype. The temporal and spatial expression of specific extracellular matrix (ECM) and adhesion molecules is crucial during embryo implantation, from the initial attachment of the blastocyst to the uterine epithelium, to the subsequent invasion of the trophoblast, particularly cytotrophoblast invasion of the maternal decidua and vasculature [1–3]. The precise mechanisms regulating the radical changes in the cell-cell and cell-matrix interactions remain largely unknown.
Chemokines, a family of small chemotactic cytokines initially recognized for roles in leukocyte chemotaxis, have been implicated in a number of reproductive events, such as ovulation, menstruation, embryo implantation, and parturition and disease conditions, like endometriosis and cancer [4–8]. Similarities between trophoblast invasion and leukocyte chemotaxis suggest common systems may be in play during these events. Adhesion molecules are fundamental mediators shared by these processes. Cytotrophoblast cell migration can be stimulated or inhibited by the manipulation of adhesion molecules on the cell surface in vitro [9]. Trophoblast invasion is an integrin-dependent process [10] that proceeds in the presence of a chemokine-cytokine-rich microenvironment derived from endometrial cells and predominantly the epithelium, decidua, leukocytes, and vasculature [5, 11, 12]. Radical changes in adhesion molecule and ECM expression at the maternal-fetal interface during embryo implantation [13–16] are thought to be vital for the attachment of the blastocyst, formation and maintenance of cell column structures in the early stages of placentation, and regulating trophoblast migration and invasion. Key changes in their expression profiles are thought to be regulated by hormones, proteolytic enzymes, growth factors, and cytokines/chemokines [2, 4, 17, 18].
Cell migration requires a series of cellular changes, including polarization, protrusion, traction, and retraction. These events require low-affinity adhesion to allow quick attachment and release of the cell [19]; effectively, the cell uses surface adhesion molecules to crawl along the ECM. Specific ECM and adhesion molecules are regulated during cell migration, including integrins, selectins, and cadherins, whose actions are also tightly regulated during implantation [13, 20–22].
Chemokines are key regulators of cell migration, and in leukocytes they alter the expression of cellular adhesion molecules and ECM components that sequentially elicit key changes in the architecture of a cell, allowing migration. A variety of chemokine receptors have been identified on both human blastocyst and on trophoblast cells [23–25]. CX3CL1 (also known as fractalkine) exists as both a membrane-anchored adhesion molecule that can capture/coordinate leukocyte migration in an integrin- and selectin-independent manner, and as a soluble chemotactic peptide that cleaves from the cell surface. CCL14 (also known as HCC-1) belongs to the CC chemokine family. Its mature propeptide is a low-affinity agonist of CCR1 that is converted to a high-affinity agonist of CCR1 and CCR5 on proteolytic processing by serine proteases. Previously, we identified that both chemokines are produced maximally by the human endometrium at the time of embryo implantation and during early pregnancy, predominantly by decidualized stroma and epithelial cells [11], and their receptors are present on invasive extravillous trophoblast (EVTs) [24]. In addition, we demonstrated the migration of trophoblasts in response to chemokines CX3CL1, CCL14, and CCL4 [24]. Endometrial cell-conditioned media also stimulated trophoblast migration, and this was attenuated by neutralizing antibodies to CX3CL1 and CCL4 [24]. Promotion of trophoblast migration by chemokines and endometrial cell-conditioned medium indicates an important involvement of chemokines in maternal-fetal communication. Taken together, these findings suggest that mechanisms similar to those used for leukocyte migration may be important during embryo implantation.
In this study we investigated the effects of the chemokines CX3CL1 and CCL14 on ECM and adhesion molecule expression. Trophoblast adhesion to different matrices was assessed in vitro using adhesion assays following chemokine treatment. Alterations in trophoblast ECM and adhesion molecules were examined using pathway-specific oligo-arrays, and quantitative real-time RT-PCR (qRT-PCR) verified changes in five candidate molecules: CTNNA1, ECM1, SPP1, MMP12, and ITGB5. The location of integrin 
6 and osteopontin (SPP1) protein on trophoblasts in first-trimester implantation sites further supports a role for these chemokines during embryo implantation. These studies provide potential mechanisms by which chemokines regulate trophoblast migration during implantation.
Trophoblast Cell Culture Conditions
The human choriocarcinoma-primary trophoblast hybrid AC1M-88 cell line (purchased from Deutsche Sammlung von Mikroorganismen und Zellkulturen) was selected for use in this study based on previous findings [24] that revealed abundant expression of chemokine receptors by AC1M-88 cells and significant trophoblast migration in response to recombinant human chemokines CX3CL1 and CCL14. The AC1M-88 cell line was cultured as previously reported [24] in RPMI with 10% charcoal-stripped fetal calf serum (CS-FCS) for two to three passages following thawing. Once 80% confluent, cells were serum starved for 24 h in serum-free RPMI containing transferrin (10 µg/ml), sodium selenite (25 ng/ml), linoleic acid (4.7 µg/ml), BSA (1 mg/ml; all from Sigma Diagnostics, St. Louis, MO) and insulin (5 µg/ml; Actrapid; Novo-Nordisk, Sydney, Australia), hereafter referred to as RPMI/TSL-I. Cells were then cultured with RPMI/TSL-I without (control) or with 10 nM (carrier-free) recombinant human (rh) chemokine CX3CL1 (fractalkine) or CCL14 (HCC-1; R & D Systems Inc., Minneapolis, MN) for a further 22 h, after which cells were washed in PBS without calcium or magnesium (Ca2+/Mg2+; PBS–) and harvested with a mild 5 mM EDTA/PBS– solution. Cells were used as detailed below in an adhesion assay or for subsequent RNA extraction.
Alterations in trophoblast adhesion following chemokine treatment were assessed using CytoMatrix Screening Kit, ECM105 adhesion assays (Chemicon, Temecula, CA). In brief, AC1M-88 cells were cultured and treated as detailed above. They were then detached, washed twice with sterile PBS, resuspended in PBS–, counted, and diluted to a final concentration of 106 cells/ml. A total of 100 µl of the diluted cell suspension was added to each well of plates that were coated with various matrix substrates (fibronectin, laminin, vitronectin, and collagen IV). Following incubation for 1.5 h at 37°C in a CO2 incubator, the wells were gently washed three times with PBS containing Ca2+/Mg2+, and then 0.2% crystal violet in 10% ethanol (100 µl) was added to each well and the plate was incubated for 5 min at room temperature. Stain was then gently removed from the wells, and wells were washed three times with PBS–. The cell-bound stain was solubilized for visualization by the addition of 100 µl solubilization buffer (a 50/50 mix of 0.1 M NaH2PO4, pH 4.5, and 50% ethanol) to each well and incubated at room temperature for 5 min. Adhesion was determined based on the absorbance of the stain measured at 540–570 nm. Bovine serum albumin-coated wells were included for each treatment in the assay as negative controls. The entire experiment was carried out three times (with n = 7 wells per treatment in each experiment).
Total RNA was extracted from the cultured cells using an RNeasy MiniKit (Qiagen GmbH, Hilden, Germany) according to manufacturer's instructions. All samples were treated with RNase-free DNase (Ambion) to remove any genomic DNA contamination, and were analyzed by spectrophotometry to determine RNA concentration, yield, and purity. Any samples with ratios of A260/280 < 1.7 or A230/280 > 1 were purified through RNeasy spin columns (Qiagen) according to manufacturer's instructions and reanalyzed. A total of 1 µg purified RNA was then run on a 1% agarose (Roche, Castle Hill, Australia) gel to ensure integrity of rRNA subunits. RNA concentrations were further analyzed by Ribogreen fluorescence RNA assay (Molecular Probes, Eugene, OR).
Extracellular Matrix and Adhesion Molecule Oligo-Array
Screening of alterations in ECM and adhesion molecule expression following treatment with CX3CL1 and CCL14 was conducted using human pathway-focused oligo-microarrays (GEarray; SuperArray Bioscience Corp., Frederick, MD) containing oligo probes to 96 ECM and adhesion molecules. RNA samples from control or chemokine-treated AC1M-88 were labeled and amplified using the TrueLabeling-AMP 2.0 kit (GA-030). A final concentration of 6 µg of labeled cRNA was applied to each array and hybridized overnight (18 h) at 60°C. Following stringent washing, positive cRNA binding was detected by incubation with CDP-Star (SuperArray), a chemiluminescent substrate. Membranes were exposed to x-ray film for a range of times (between 2 sec and 5 min) to ensure quantitation during the linear phase of the reaction. Arrays were performed concurrently, and the entire experiment was carried out in triplicate. The x-ray films were scanned at high resolution and analyzed using GEArray Expression Analysis Suite Software (GEArray). A 2-fold increase or decrease in expression was used as a cutoff for analysis of the oligo-array data.
Real-time RT-PCR was used to verify and more accurately quantitate mRNA changes following chemokine treatment, as previously detailed [5]. Real-time RT-PCR was performed using a Light Cycler (Roche). Standards were generated by conventional PCR. RNA samples (from n = 3 cultures) were reverse transcribed as described previously [5] and in triplicate and were analyzed for efficiency and reproducibility of the RT reaction by real-time PCR for the ribosomal 18S subunit. The 18S concentrations (pg/ml) were assessed between triplicates, and the intraassay variability of the RT and PCR steps was evaluated. Within triplicates, samples outside 15% variability of the average 18S concentration were excluded as outliers. Otherwise, triplicates were pooled, creating cDNA reaction products that are close representations of the initial mRNA population. The cDNA pools were diluted 1:5, aliquoted, and stored at –80°C to prevent freeze thawing. The cDNA samples used for 18S real-time RT-PCR were used to assess CTNNA1, ECM1, SPP1, ITGA6, MMP12, and ITGB5 expression by trophoblast cells following chemokine treatment (see Table 1 for primer sequences and PCR conditions). Polymerase chain reaction amplification of 4 µl diluted cDNA was performed for each molecule in glass capillaries in a master mix (Roche). Negative controls were performed by omission of RT. All samples to be compared were included within the same run in duplicate, and the entire PCR experiment was carried out in triplicate.
|
First-Trimester Tissue Collection
Ethical approval was obtained from appropriate Institutional Ethics Committees for all tissue collections, and written informed consent was obtained from all subjects. Human first-trimester implantation sites (containing decidua and placenta; n = 5) were collected by curettage prior to termination of pregnancy (gestation 8–12 wk) by vacuum aspiration.
SPP1 was localized in the first-trimester implantation sites (n = 5) using a mouse monoclonal antibody raised against mouse recombinant SPP1 (AKm2A1; SC-21742; Santa Cruz Biotechnologies). Antibody specificity has previously been established on human tissue [26]. In brief, serial paraffin sections (2 µm) were dewaxed in Histosol (Sigma Chemical) and rehydrated through descending grades of alcohol to dH2O. Sections were microwaved at high power (1000 W) in 0.01 mol/l sodium citrate buffer (pH 6.0) for 2 x 5 min and then incubated in the hot buffer for a further 20 min. Endogenous peroxidase activity was quenched with 3% H2O2 in dH2O for 10 min at room temperature. Nonspecific binding was blocked by incubation with a nonimmune block (10% normal horse serum [NHS; H0146; Sigma]), 2% normal human serum ("in-house") in 0.1% Tween 20 (Bio-Rad, North Ryde, Australia). Primary antibody was applied at 1 µg/ml overnight (16–18 h) at 4°C, and biotinylated horse anti-mouse IgG (Dako, Botany, Australia) was applied for 30 min, followed by avidin/biotin conjugated with horseradish peroxidase (Dako). Positive localization of SPP1 protein was identified by the application of the peroxidase substrate 3, 3'-diaminobenzidine (Dako). Tissue sections were counterstained with Harris hematoxylin, dehydrated through ascending grades of ethanol, and mounted. Negative controls were included for each section where mouse immunoglobulin (Dako) was substituted at matching concentration to the primary antibody.
Integrin 6 protein localization was examined using an affinity-purified goat polyclonal anti-human integrin 6 antibody (SC-6596 C-18; Santa Cruz Biotechnologies). The immunostaining protocol was similar to that used for SPP1 with the primary antibody also at 1 µg/ml, except that biotinylated horse anti-goat IgG (Vector Laboratories, Burlingame, CA) was used as the secondary antibody. Negative controls were included for each section where goat immunoglobulin (Dako) was substituted at matching concentration to the primary antibody.
ECM1 protein localization was examined using an affinity-purified sheep polyconal anti-human ECM1 antibody (AF3937; R & D Systems). The immunostaining protocol was similar to that used for SPP1, except that no antigen retrieval was necessary, the nonimmune blocking solution contained 10% normal rabbit and 2% human serum, the primary antibody was applied at 2 µg/ml, and the biotinylated rabbit-anti-sheep IgG (Vector Laboratories) was used as the secondary antibody. Negative controls were included for each section where sheep immunoglobulin (Dako) was substituted at matching concentration to the primary antibody.
Immunohistochemistry for cytokeratin was conducted on five serial (2-µm) sections of first-trimester human implantation sites to confirm trophoblast cell identity, as described previously [24]. Cytokeratin was used on alternating sections with SPP1, integrin 6, and ECM1.
Results for the adhesion assays and quantitative RT-PCR are expressed as mean fold change ± SEM for each treatment compared with control. Statistical analysis was performed using nonparametric Mann-Whitney test (P < 0.05 was taken as significant) using PRISM version 3.00 for Windows (GraphPad, SanDiego, CA).
Effect of CX3CL1 and CCL14 on Trophoblast Adhesion
AC1M-88 human trophoblast cells were used to determine the effects of the chemokines CX3CL1 and CCL14 on trophoblast cell adhesion. Trophoblast cells treated with rhCX3CL1 (10 nM) showed a 4-fold increase in adhesion to fibronectin in comparison to control trophoblast cells (P < 0.05; RPMI/TSL-I alone; Fig. 1). CCL14 (10 nM) similarly increased trophoblast adhesion to fibronectin 3-fold (P < 0.05; Fig. 1). Trophoblast adhesion to three other matrices (laminin, vitronectin, and collagen IV) was also assessed following chemokine treatment, but no significant changes in binding to these substrates were observed (data not shown).
|
ECM and Adhesion Pathway-Specific Oligo-Array Data
The effect of rh chemokines CX3CL1 and CCL14 on the expression profile of ECM and adhesion molecules on the trophoblast cell line AC1M-88 was assessed using human pathway-focused oligo-microarrays. Of the 96 genes represented on the array expression, 33 molecules were regulated by rhCX3CL1 treatment; 32 were upregulated, and 1 was downregulated (Table 2). CCL14 altered the expression of 15 genes, with an increase in the expression of 6 and reduction in the expression of 9 (Table 3).
|
|
Quantitative Real-Time RT-PCR Analysis
Real-time quantitative RT-PCR was conducted to verify that the ECM and adhesion molecules identified by oligo-gene array are indeed expressed by trophoblast cells and to provide a more accurate quantitative analysis of the expression of individual genes following chemokine treatment. Trophoblast mRNA expression levels were quantitated by comparison with a purified PCR-generated cDNA standard of known concentration for each molecule examined. Consistent with the findings from the array, human CTNNA1 mRNA levels were significantly enhanced in the trophoblast cells, with an increase greater than 3.5-fold following treatment with rhCX3CL1, but no effect with CCL14 (Fig. 2A). Similarly, assessment of ECM1 mRNA levels revealed a large, almost 6-fold increase in trophoblast ECM1 mRNA following treatment with rhCX3CL1 but not rhCCL14 (Fig. 2B), consistent with the array (Tables 2 and 3). Quantification of SPP1 mRNA revealed a significant reduction in expression following treatment with rhCX3CL1 but not rhCCL14 (Fig. 2C). Likewise, human ITGA6 mRNA levels were significantly reduced more than 2.5-fold in the trophoblast cells following treatment with rhCX3CL1, but no significant changes were observed with rhCCL14 treatment (Fig. 2D). This finding was not consistent in changes observed from the array. Human MMP12 mRNA levels were significantly enhanced (greater than 3-fold) in the trophoblast following treatment with both rhCX3CL1 and CCL14 (Fig. 2E), which is comparable with the results from the array analysis (Tables 2 and 3). Assessment of ITGB5 mRNA in the trophoblast following treatment with rhCX3CL1 and CCL14 revealed an increase in ITGB5 levels greater than 2-fold (Fig. 3F), identical to the results obtained from the array (Tables 2 and 3).
|
|
Immunolocalization of SPP1, ECM1, and Integrin 6 in First-Trimester Human Implantation Sites
Localization of SPP1 at the fetomaternal interface was assessed on tissue sections from first-trimester human implantation sites. Immunohistochemistry revealed immunoreactive SPP1 protein was localized to the anchoring villous syncytiocytotrophoblast layer (Fig. 3A) and the underlying cytotrophoblast cell column (Fig. 3A), whereas immunoreactive leukocytes were observed within the stromal compartment of the villous tip and in the endometrial decidua (Fig. 3, A and B and the panel to the right of B). Weak staining was observed on decidualized stromal cells (Fig. 3B). SPP1 protein was seen in the EVTs in the tissue surrounding the maternal arteries (Fig. 3C) and in endovascular EVTs (vEVTs) penetrating the vessel wall (Fig. 3C, arrow). Trophoblast identity was confirmed by cell morphology and serial staining for cytokeratin (Fig. 3D). Immunohistochemistry for integrin 6 was similar to that of SPP1, with strong staining in the syncytiotrophoblast layer (Fig. 3E) and the cytotrophoblast cell column (Fig. 3F), and reduced staining intensity at the distal portion of the cell column (Fig. 3F). Further reduction in staining occurred in the extravillous cytotrophoblasts surrounding and penetrating the maternal vasculature (Fig. 3G). Serial staining for cytokeratin ensured trophoblast identity and presence (Fig. 3H).
Immunohistochemistry for ECM1 in first-trimester implantation sites revealed that ECM1 protein was present on interstitial EVTs (iEVTs) and localized to the ECM of decidualized stromal cells (Fig. 3I), but not to nondecidualized stroma (Fig. 3J). Endometrial glands (Gl) and vasculature (V) were negative for ECM1 staining. ECM1 immunoreactivity was also observed in a subpopulation of trophoblast cells in the vessel wall (Fig. 3K) in the syncytiocytotrophoblast layer (Fig. 3L) and in the adjacent cell column (Fig. 3L), as confirmed by serial cytokeratin staining (Fig. 3M).
The identification of a number of chemokines and their receptors at the maternal-fetal interface suggested roles for chemokines in regulating key events during implantation [4]. Previous studies showed increased trophoblast migration toward CX3CL1 and CCL14 in vitro [24], and neutralization of chemokines in endometrial-conditioned media showed a significant reduction in the trophoblast migration observed. These observations raised the question, "Does this migration use mechanisms similar to those used in leukocyte recruitment?" The current study demonstrates for the first time the effects of recombinant human chemokines CX3CL1 and CCL14 on trophoblast adhesive and ECM-binding properties.
A significant increase in trophoblast adhesion to fibronectin was observed following chemokine treatment, clearly indicating a role for chemokines in the regulation of trophoblast adhesion. Subsequent application of pathway-specific ECM and adhesion molecule gene arrays revealed the relative abundance and changes in adhesion molecule and ECM gene expression in trophoblasts following treatment with CX3CL1 and CCL14. Quantitation of alterations in mRNA transcripts of CTNNA1, ECM1, SPP1, ITGA6, MMP12, and ITGB5 with chemokine treatment confirmed the gene array analysis findings. Localization of immunoreactive integrin 6 and SPP1 in human first-trimester implantation sites confirmed their presence on migrating EVTs in vivo. Such collective analysis of chemokine-induced changes to adhesion and ECM molecule expression is vital to understanding the mechanisms underlying trophoblast migration, as these molecules act in combination to facilitate the multiple steps involved in cell migration.
During decidualization under the influence of progesterone, stromal cells cease synthesis of collagen VI and increase fibronectin production [27]. Such production of a specialized ECM network at the human maternal-fetal interface and rich in fibronectin, laminin, and collagen IV has been shown to support trophoblast adhesion and migration during implantation [27]. Interestingly, treatment of the trophoblast cell line AC1M88 with both CX3CL1 and CCL14 substantially increased trophoblast adhesion to fibronectin, but these chemokines were without effect on trophoblast adhesion to laminin, collagen IV, and vitronectin. The increase seen in adhesion to fibronectin, however, indicates a direct role for chemokines in regulating trophoblast adhesion.
Chemokines play important roles in promoting cell migration by regulating cellular adhesion molecules and ECM, hence triggering key changes in the architecture of a cell that promote migration. Pathway-specific oligo-arrays showed altered mRNA expression of more than 30 adhesion and ECM molecules after treatment with rhCX3CL1, and of 15 genes following CCL14 treatment. Although much redundancy occurs within the chemokine family, this study revealed variable regulation by each chemokine on trophoblast adhesion and ECM expression profile.
Quantitative real-time RT-PCR analysis was used to verify changes in representative candidate molecules identified by array. CTNNA1, which was increased almost 4-fold in trophoblast cells by CX3CL1, had been previously shown as regulated along with CX3CL1 during the elongation and implantation of ovine trophoblast [28]. Alpha-catenin enables the E-cadherin/β-catenin complex to anchor to the actin cytoskeleton, which is important during the initial stages of cell migration and is also important in trophectoderm formation [29]. The presence of both CX3CL1 and -catenin during implantation across different species and the interaction demonstrated here indicate a role for this network during the early stages of implantation.
ECM1 mRNA was also increased following treatment with rhCX3CL1. ECM1 is a secreted protein involved in controlling the production of basement membrane proteins and is of likely importance in angiogenesis and tissue remodeling. Immunoreactive ECM1 protein was localized to the ECM of decidualized stroma and to a variety of trophoblast subpopulations in first-trimester implantation sites. This study describes for the first time the presence of ECM1 in the pregnant endometrium. During embryogenesis, ECM1 was expressed predominantly by the blood vessels, whereas in the adult human it is most highly expressed in the heart and placenta [30]. Its overexpression in a variety of cancers is associated with the acquisition of a metastatic phenotype and altered blood vessel formation [31]. Recently, ECM1 was proposed to be a novel mediator of angiostasis, as it inhibited the angiogenic sprouting of murine brain microvascular cells in vitro [32]. Since ECM1 appears to regulate proangiogenic and antiangiogenic signals, it is likely that its production by endovascular trophoblasts may contribute to regulation of spiral arteriole remodeling during implantation.
By contrast, SPP1 mRNA levels were decreased in trophoblast cells following CX3CL1 treatment. SPP1 has multiple functions, acting as a cytokine and an adhesion molecule interacting with cell surface integrin and CD44 receptors. It is temporally and spatially regulated in the endometrium and is a component of endometrial glandular histotroph secretions [33]. It is also present on cytotrophoblasts in human first-trimester implantation sites [34], and its expression is increased in vitro by progesterone [35]. Regulation of SPP1 at the maternal-fetal interface is proposed as important in mediating trophoblast adhesion and paracrine signaling events during implantation [33]. Importantly in conditions where uncontrolled trophoblast invasion occurs, such as gestational trophoblastic disease, SPP1 levels are increased, particularly in trophoblasts at the invasion front [36], highlighting the need for tight regulation of trophoblast invasion during implantation. CX3CL1 may contribute to the equilibrium of this highly invasive process, ensuring that trophoblast invasion is tightly restrained.
A discrepancy was noted between the oligo-array and qRT-PCR findings for ITGA6 mRNA levels that were significantly reduced 2.6-fold with CX3CL1 treatment as assessed by qRT-PCR, which is a more accurate method of quantitation. The oligo-probes used on the array are small, and thus there is a higher chance of nonspecific binding than to primers designed for qRT-PCR; in this case, the integrin 6 primers had been previously published, and the PCR product was confirmed by sequencing. CX3CL1 is produced by maternal decidual cells, glandular and luminal epithelium, and endothelial and perivascular cells [11], whereas CX3CR1 is expressed on vEVT, the cell column, and syncytium. Immunohistochemical analysis from this and from previous studies [15, 37] demonstrated integrin 6 on trophoblast cells in the anchoring villi and the proximal portion of the cell column, with a loss in the distal cell column and migrating trophoblast. Thus, CX3CL1 is appropriately located to regulate integrin 6, which plays an important role in anchoring trophoblasts to the basement membrane, and its loss may indeed promote cell migration into the maternal tissue compartment [15]. In preeclamptic placenta, which is characterized by abnormally shallow trophoblast invasion, integrin 6 expression remains high in the invading extravillous trophoblast [38]. The decrease in integrin 6 expression observed here following CX3CL1 treatment suggests CX3CL1 may contribute to switching off integrin 6 expression on the trophoblast, thus promoting differentiation of cytotrophoblasts to an invasive migratory phenotype.
By contrast, stimulation of ITGB5 by both CX3CL1 and CCL14 was consistent between the array and qRT-PCR data. Integrin β5 forms a heterodimer, with the v integrin subunit facilitating binding to fibronectin, vitronectin, and fibrinogen. Integrin vβ5 has been implicated in cell migration and angiogenesis, and a role for integrin β5 during embryo implantation has been proposed due to its apical expression on glandular and luminal epithelial cells [39]. However, to date there has been no description of integrin β5 in trophoblast migration. The increase in ITGB5 levels seen here is coincident with increased trophoblast adhesion to fibronectin following chemokine treatment, thus providing a novel mechanism for trophoblast adherence and migration in the fibronectin-chemokine-rich microenvironment at the maternal-fetal interface.
Quantitation of MMP12 confirmed that both CX3CL1 and CCL14 increased MMP12 mRNA. MMP12 (macrophage elastase) was first described as an elastolytic metalloproteinase secreted by inflammatory macrophages [40]. However, it can also degrade a broad spectrum of other ECM components, including type IV collagen, fibronectin, laminin, vitronectin, proteoglycans, chondroitin sulfate, and myelin basic protein [41], many of which are abundant in the complex matrix at the maternal-fetal interface. MMP12 has multiple additional roles in inflammatory processes, including vascular remodeling and the ability to activate other MMPs, thus enhancing the proteolytic cascade. Human MMP12 also cleaves the D1 domain of PLAUR (plasminogen activator urokinase receptor, also known as u-PA receptor, u-PAR) demonstrating a crosstalk between MMP12 and the u-PA system that may enhance pericellular proteolysis important for cell migration [42]. MMP12 expression also correlates with the extent of endometrial adenocarcinoma invasion [43]. The active 45-kDa form of MMP12 is also present in BeWo (trophoblast cells) [44]. CX3CL1 and CCL14 produced by endometrial decidual and perivascular cells may thus act to increase trophoblast MMP12 production, facilitating degradation of ECM networks at the maternal-fetal interface.
Clear parallels exist between leukocyte extravasation, cancer invasion, and embryo implantation, all of which use chemokines as key players in tightly regulating cellular migration. This study demonstrates a potential role for chemokines in the directional migration and homing of trophoblasts within the decidua, particularly to maternal arteries. Together with previous studies, these data demonstrate mechanisms whereby chemokines might regulate trophoblast invasion and spiral arteriole remodeling during placentation.
In conclusion, this study demonstrates that the chemokines CX3CL1 and CCL14 regulate the expression of CTNNA1, ECM1, SPP1, ITGA6, MMP12, and ITGB5 in human trophoblast cells, and both CX3CL1 and CCL14 enhance trophoblast adhesion to fibronectin. A direct role can be predicted for CX3CL1 and CCL14 in regulating trophoblast migration via actions on ECM and adhesion molecules expressed at the maternal-fetal interface.
ACKNOWLEDGMENTS
The authors would like to thank Prof. Euan Wallace and Dr. Ursula Manuelpillai (Monash University, Department of Obstetrics and Gynaecology) for provision of first-trimester implantation site blocks.
FOOTNOTES
1Supported by the National Health and Medical Research Council of Australia (241000 and 143798), a Prince Henry's Institute of Medical Research postgraduate scholarship, and The Flew Foundation. 
Correspondence: 2Natalie J. Hannan, PHIMR, P.O. Box 5152, Clayton, 3168 Victoria, Australia. FAX: 61 3 9594 6125; e-mail: natalie.hannan{at}princehenrys.org
Received: 4 November 2007.
First decision: 5 December 2007.
Accepted: 11 March 2008.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
