|
|
||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Female Reproductive Tract |
a Divisions of Developmental Biology
b Hematology/Oncology, Cincinnati Children's Research Foundation, Cincinnati, Ohio 45229
| ABSTRACT |
|---|
|
|
|---|
female reproductive tract, implantation, placenta, trophoblast, uterus
| INTRODUCTION |
|---|
|
|
|---|
Interactions between the early embryo (blastocyst) and the uterus at the time of implantation are complex and not completely understood [611]. The complex interaction between the uterus and the blastocyst requires coordinate expression of interacting proteins. Proteins essential for implantation that are expressed in the uterus include heparin-binding epidermal growth factor (EGF), leukemia inhibitory factor, COX-2, specific integrins, and metalloproteinase inhibitors. In the blastocyst, essential proteins include heparan sulfate proteoglycans (i.e., perlecan); adhesion-promoting activities, including trophinin, tastin, and bystin; EGF-receptors ErbB1 and ErbB4; and metalloproteinases. Other EGF family members, such as amphiregulin, ß-cellulin, epiregulin, and heregulins, may also function during this process. Cytokines such as interleukin-1 are critical to the process. Mouse genetic models have illustrated this complexity and highlight the redundancy of function of many components of the implantation process, both in the embryo and in the uterus.
Recently, hepatocyte growth factor (HGF), which specifically activates the receptor tyrosine kinase, c-Met, has been shown to be essential during normal human and mouse trophoblast function [12, 13]. In the mouse, ablation of HGF results in a significantly smaller placenta and embryonic death by E15.5 because of the presence of fewer and less invasive trophoblast cells in the labyrinth [12]. In the human, HGF stimulates motility and invasion of trophoblast cells, and pregnancies complicated by preeclampsia exhibit decreased production and secretion of HGF in placental villi [13]. Therefore, receptor tyrosine kinase activation via specific ligand binding can be critical to early events in embryonic development.
Ron, a proto-oncogene product that is homologous to the HGF receptor, c-Met, is activated by HGF-like protein (HGFL), which is also known as macrophage-stimulating protein [14, 15]. Ron is a heterodimeric receptor consisting of a 45-kDa
chain and a 150-kDa ß chain linked via disulfide bonds [14, 16, 17]. To date, Ron expression appears to be ubiquitous, with transcripts being expressed in a majority of examined tissues, including skin, colon, lung, liver, monocytes, kidney, peripheral and central nervous systems, and bone marrow [14, 16, 18]. HGFL belongs to the kringle-containing protein family that includes HGF, plasminogen, and prothrombin [19]. HGFL is synthesized in hepatocytes and secreted into the bloodstream, and it circulates at relatively high concentrations (
400 ng/ml) in the serum as a single proprotein [15, 1921]. On cleavage, HGFL is converted into the mature ligand (
80 kDa) composed of an
chain with four kringle domains necessary for protein-protein interactions and a ß chain containing an enzymatically inactive, serine protease-like domain [1922].
Ron activation via HGFL stimulation elicits numerous physiological functions under normal and abnormal conditions. For example, HGFL stimulation increases motility of human nasal cilia [23], induces production of cytokines from bone marrow megakaryocytes [24], causes shape changes in mouse peritoneal macrophages [25], induces in vitro macrophage spreading [26], induces bone resorption via osteoclasts [27], and inhibits cytokine- or endotoxin-induced expression of inducible nitric oxide synthase [28]. Overexpression of Ron or of Ron containing point mutations in the tyrosine kinase domain, which induce constitutive receptor activation, can elicit an oncogenic phenotype following introduction into NIH3T3 cells [29]. Such upregulation and mutations of Ron have been found in various human cancers, such as breast carcinomas, hepatocellular carcinomas, gastric cancer cell lines, and colon malignancies [3033]. Deletion of the extracellular domain of Ron yields a mutated form of the receptor that has been found to confer susceptibility to Friends virus-induced erythroleukemia [34].
To explore additional physiological functions that Ron may induce during mammalian growth and development, animals deficient for the entire Ron receptor have been produced. Ron -/- embryos are viable through the blastocyst stage of development, but they fail to survive past the peri-implantation period [35]. In situ hybridization studies also demonstrated the presence of Ron transcripts within the trophectoderm cells surrounding the inner cell mass of E3.5 embryos [35]. Combining the results from the null mice analyses and the results demonstrating that overexpression of Ron promotes an invasive phenotype suggests that Ron may be necessary for embryo implantation and sustained trophoblast viability during early embryonic development. Therefore, the goals of the present study were to obtain a more detailed Ron expression pattern in reproductive tissues of mice during implantation and to test the hypothesis that Ron activation may aid in implantation by way of trophoblast function. Thus, reverse transcription-polymerase chain reaction (RT-PCR), in situ hybridization, immunohistochemical, cellular invasion, and cell survival assays were performed using wild-type murine tissues and trophoblast cell lines.
Our results demonstrate the presence of Ron and HGFL transcripts in various female reproductive tissues, specifically in the trophoblast cells of the ectoplacental cone (EPC) and the giant cell layer both during and after implantation. Additionally, HGFL increased invasion through an extracellular matrix and enhanced survival of mouse trophoblast cell lines expressing Ron. These results suggest that Ron may be a critical protein regulating trophoblast function during implantation.
| MATERIALS AND METHODS |
|---|
|
|
|---|
Animal protocols used in these studies were approved by Cincinnati Children's Research Foundation Institutional Animal Committee on Laboratory Animal Care and were performed according to NIH Animal Care and Use of Laboratory Animal Guidelines. Adult female and male Black Swiss mice (age 56 wk) were purchased from Taconic (Germantown, NY). Mice were housed in microisolator cages in rooms with constant temperature and humidity. The animals were fed standard chow and water ad libitum while exposed to a 14L:10D photoperiod. Timed matings were performed after hormonally priming females with 5 IU of eCG followed 48 h later by 5 IU of hCG to promote folliculogenesis and ovulation. Immediately following hCG administration, one female was placed in a cage housing one male. The presence of a vaginal plug the following day was considered to be 0.5 days of pregnancy.
Materials
The eCG was purchased from Professional Compounding Centers of America, Inc. (Houston, TX). Bovine serum albumin fraction V, porcine kallikrein, soybean trypsin inhibitor, 2-mercaptoethanol, and hCG were purchased from Sigma-Aldrich (St. Louis, MO). F-12 and RPMI tissue culture medium, glutamine, sodium pyruvate, Trypan blue, dNTPs, oligo(dT), Superscript II reverse transcriptase, and TRIzol reagent were purchased from Gibco/BRL Life Technologies (Rockville, MD). Fluorescein In Situ Cell Death Detection Kit was purchased from Roche Diagnostics, Inc. (Indianapolis, IN). Matrigel and transwell cell culture inserts were purchased from Becton Dickinson (Bedford, MA). Diff-Quick Stain Kit was purchased from Dade Behring, Inc. (Newark, DE). Recombinant human HGFL was purchased under the name of recombinant human macrophage-stimulating protein from R&D Systems, Inc. (Minneapolis, MN). Lab-Tek II chamber slides with covers were purchased from Nalge Nunc International (Naperville, IL). Mouse anti-Ron
chain monoclonal antibody was purchased from Transduction Laboratories (Lexington, KY). Recombinant mouse MSP R/Fc chimera was purchased from R&D Systems. Normal donkey and goat serum, Cy2-conjugated goat anti-mouse immunoglobulin G, and peroxidase-conjugated donkey anti-mouse F(ab')2 fragment were purchased from Jackson ImmunoResearch Laboratories, Inc. (West Grove, PA). Immu-mount was purchased from Thermo Shandon (Pittsburgh, PA). Pierce SuperSignal Western Detection kit was purchased from Pierce Endogen (Rockford, IL). Zymogram precast 10% (w/v) gels and zymography renaturation, sample and renaturation buffers, and prestained broad-range molecular weight markers were purchased from BioRad (Hercules, CA).
Cell Culture
Dr. Joan Hunt (University of Kansas Medical Center, Kansas City, KS) generously provided the Swiss Webster mouse trophoblast cell lines SM9-1, SM9-2, and SM-10. Chinese hamster ovary (CHO) cells and murine erythroleukemia (MEL) cells were purchased from the American Type Culture Collection (Manassas, VA). All trophoblast cell lines and MEL cells were maintained in medium (RPMI containing 60 µg/ml of gentamicin, 2 mM L-glutamine, 1 mM sodium pyruvate, and 50 µM 2-mercaptoethanol) with constant humidity at 37°C with 5% CO2. SM9-1 and SM-10 cells were maintained in medium with 10% (v/v) fetal bovine serum, whereas the SM9-2 cells were maintained in RPMI medium with 20% fetal bovine serum.
RNA Isolation
Individual tissue culture plates (10 cm) confluent with each trophoblast cell line were rinsed with sterile PBS. Cells were scraped from the plates, pelleted, aspirated, and snap-frozen in liquid nitrogen. Additionally, mouse tissues were removed from killed adult animals (age, 78 wk) and snap-frozen in liquid nitrogen. Cells and tissues were used immediately or stored at -80°C until further use. Total RNA was isolated from frozen cell pellets and frozen mouse tissues using TRIzol reagent according to the manufacturer's protocol. The RNA was resuspended in nuclease-free water and quantitated on a Bio-Tek Microquant spectrophotometer (Winooski, VT) using KC Junior Analytical Software (MWG Biotech Inc., High Point, NC) and then stored at -80°C.
Reverse Transcription-Polymerase Chain Reaction
Following quantitation, approximately 1 µg of RNA was reverse transcribed using Superscript II reverse transcriptase and oligo(dT) according to the manufacturer's protocol. The resulting cDNA (2 µl) was used in separate PCRs for the detection of Ron or HGFL expression. Amplification was performed in 25-µl volumes with the final PCR buffer (Fisher Scientific, Pittsburgh, PA) and primer concentrations as follows: 200 ng each primer, 200 µm each dNTP, 0.51 U Taq DNA polymerase, 1.5 mM MgCl2, and 10 mM Tris-HCl. An initial denaturing step of 95°C for 2 min was followed by amplification for 30 cycles with an annealing temperature of 65°C for both mouse Ron and HGFL, followed by a final extension time of 9 min at 72°C. Amplified products were separated on a 2% (w/v) agarose gel. The forward 5'-CACCCTTTGTGGCTCCAACTTC-3' (nucleotides 54725493 on the coding strand of the mouse Ron gene [16]) and reverse 5'-TTCCATTGACCAGCACAGCTCG-3' (nucleotides 63046325 of the complementary strand of the mouse Ron gene [16]) primers yielded a 404-base pair (bp) expression product or 854-bp genomic product specific for mouse exons 57 of the Ron extracellular domain. The forward 5'-TGATTTGCCTGCCTCCTGAAC-3' (nucleotides 51215142 on the coding strand of the mouse HGFL gene [20]) and reverse 5'-CTCACTCTCTTGTATGTGTCCTCGG-3' (nucleotides 53705394 of the complementary strand of the mouse HGFL gene [20]) primers yielded a 178-bp expression product or 274-bp genomic product specific for exons 1617 of mouse HGFL. Mouse beta actin was analyzed by RT-PCR, and served as a control for RNA integrity, using the forward 5'-TGACAGCATTGCTTCTGTGTAAATT-3' (nucleotides 15671591 of the mouse beta actin cDNA [36]) and the reverse 5'-ATTGGTCTCAAGTCAGTGTACAGGC-3' (nucleotides 17601784 of the mouse beta actin cDNA [36]) primers that yielded a 218-bp expression and/or genomic product because of the primers annealing in the 3'-untranslated region of the gene.
Immunocytochemistry
Tissue samples were fixed overnight in 4% (v/v) paraformaldehyde and dehydrated in 70% (v/v) ethanol. Following fixation, tissue samples were embedded in paraffin and sectioned (thickness, 57 µm) using a microtome. Tissue sections were dewaxed in xylenes and rehydrated for 2 min each through a 100%, 95%, and 75% standard ethanol series. Immunocytochemistry for the sectioned tissue samples and trophoblast cell lines that were maintained at very low confluency in Lab-Tek II chambers overnight at 37°C and 5% CO2 were performed in a similar manner. Following fixation with 4% paraformaldehyde, samples were washed with PBS and blocked in 5% (v/v) normal goat serum and 1% (w/v) bovine serum albumin for 30 min. Samples were incubated with anti-Ron antibody for 45 min at room temperature and washed three times with PBS. Finally, the samples were incubated with Cy2-conjugated secondary antibody at room temperature for 1 h. After washing briefly with PBS, samples were cover-slipped, observed, and photographed using either a Zeiss stereoscope or confocal microscope.
The specificity of the antibody used for these experiments was tested using a recombinant mouse Ron chimeric protein containing the
chain of mouse Ron. The recombinant protein effectively competed for binding with the Ron-specific antibody as determined by Western blot analysis (data not shown).
In Situ Hybridization Analysis
Primers that were similar to those used in the RT-PCR analysis were used to amplify a Ron-specific cDNA. The primers differed in that XbaI and XhoI restriction enzyme sites were added to either end of the amplified expression product that corresponded to exons 57 of the extracellular domain of the mouse Ron mRNA (nucleotides 18762279 of the mouse cDNA [16]). Following amplification, the cDNA was ligated into BlueScript SK. T7 and T3 polymerases were used to produce 35S-labeled sense and antisense riboprobes for in situ hybridization studies. Implanting embryos within the uterine horns (E6.57.5 and E12.0) of adult female Black Swiss mice were isolated, fixed in 4% paraformaldehyde, infused with 30% (w/v) sucrose, frozen, and sectioned. Hybridization of probes to tissue sections was performed as previously described [16, 37]. Slides were exposed for 6 wk.
Matrigel Invasion
Transwell cell culture inserts with 0.8-µm pores were coated with 75 µl of Matrigel (3.4 mg/ml) and incubated for 50 min at 37°C with 5% CO2. Once the Matrigel had solidified, kallikrein-treated conditioned medium from HGFL-expressing CHO cells, kallikrein-treated conditioned medium from nontransfected CHO cells, RPMI medium only, or RPMI medium with 10% FBS was added to the cell culture inserts and wells of a 24-well tissue culture plate. Three wells were used for each stimulant. Approximately 6.510 x 103 SM9-1 trophoblast cells were added to each well and incubated for 24 and 48 h. Following incubation, excess Matrigel was removed from the transwell inserts using a cotton swab. The cells attached to the bottom of the transwell membrane were fixed and stained using the Diff-Quick Stain Kit. Cells in 10 separate fields in each of the duplicate wells were counted and averaged.
SM-10 Cell Survival
SM-10 cells (1 x 106) were plated into individual wells of a six-well tissue culture plate. Cells were serum starved by incubating in RPMI medium with 0.5% FBS for 24 h at 37°C with 5% CO2, after which RPMI medium only, RPMI medium containing 400 ng/ml of recombinant human HGFL, or RPMI medium containing 10% FBS were added to the wells. To determine the survival rate of the SM-10 cells after stimulation, the numbers of floating versus attached cells were counted. To ensure that floating cells were dead, following stimulation the supernatant was centrifuged gently, and the cells were then recovered and stained with 0.4% (w/v) Trypan blue for 5 min. Cells excluding the dye were scored as viable. Wells were analyzed in triplicate to assess the effect of each stimulant.
TUNEL Analysis
SM-10 cells (2.5 x 105) were added to individual wells of a Lab-Tek II Chamber slide and incubated overnight at 37°C with 5% CO2. Cells were serum starved for 24 h by incubating in RPMI medium containing 0.5% FBS for 24 h. Following starvation, cells were stimulated with RPMI only, RPMI containing 10% FBS, or RPMI containing 400 ng/ml of recombinant human HGFL for 24 h. Cells undergoing cell death were identified by TUNEL analysis using a Fluorescein In Situ Death Detection Kit. Briefly, after stimulation for 24 h, cells were rinsed with sterile PBS and fixed for 1 h in 4% paraformaldehyde. Slides were rinsed with PBS and incubated in permeabilization solution (0.1% [w/v] Triton X-100 and 0.1% [w/v] sodium citrate) for 2 min on ice. Slides were rinsed twice with PBS. The TUNEL reaction mixture was then added and incubated in the dark in a humidified chamber at 37°C for 1 h. Negative controls incubated in TUNEL reaction mixture without terminal transferase were included on each slide. Glass cover slips were mounted onto the slides using aqueous antifade mounting medium. Cells were examined and photographed using a Zeiss fluorescence microscope (B&B Microscopes, LTD, Warrendale, PA).
Statistical Analysis
The average number of cells that invaded the Matrigel were compared by using a one-way analysis of variance on ranks followed by a pairwise multiple comparison using the Student-Newman-Keuls method to determine significance at P < 0.0005.
The percentage of surviving (attached) SM-10 cells per well was determined by dividing the number of attached cells by the total number of cells observed (attached plus floating). Because the percentages were not normally distributed, a transformation was taken, and the analysis was performed on arc-sine transformed values. A two-way analysis of variance with group, time, and interaction between group and time was performed using the general linear model procedure in SAS (Cary, NC) version 8.2. Bonferroni multiple comparison was used to determine statistical significance at P < 0.0001.
| RESULTS |
|---|
|
|
|---|
To establish a detailed temporal and spatial expression pattern for Ron and HGFL, our initial studies focused on determining which adult mouse reproductive tissues possessed these transcripts by using RT-PCR and in situ hybridization analyses. As previously reported, Ron and HGFL transcripts were detected in adult mouse colon and liver (Fig. 1, A and B, lanes 5 and 6 [14, 21]). Virgin uterus expresses Ron and not HGFL mRNA (Fig. 1, A and B, lane 1). Placenta (E12 and E16), epididymis, and testis expressed both Ron and HGFL (Fig. 1, A and B, lanes 2, 3, 7, and 8). The virgin cervix was the only tissue tested that did not express Ron but did express HGFL (Fig. 1, A and B, lane 4). No PCR products caused by genomic contamination were observed in the presence or absence of reverse transcriptase (data not shown). The beta actin RT-PCR reactions served as a control for RNA integrity.
|
In situ hybridization analysis was performed on several of the tissues that were positive for the presence of Ron as demonstrated by RT-PCR analysis. The trophoblast cells, within the EPC of the implanting embryo, expressed Ron as early as E7.5 (Fig. 2B). As implantation and development progresses, Ron expression is still seen throughout the giant trophoblast cells of the E12 placenta (Fig. 2D, arrow). Because the trophectoderm cells will give rise to the highly invasive trophoblast cells as well as to the extraembryonic membranes of the early implanting and developing embryo, these results support the earlier studies that demonstrated Ron expression in the trophectoderm cells surrounding the inner cell mass of embryos [35]. Furthermore, these results suggest that Ron may play a role during the process of embryo implantation.
|
Immunohistochemistry of E7.5 implanting embryos was performed to determine if Ron protein was present within the maternal and/or embryonic tissue during implantation. Consistent with the in situ hybridization analysis, Ron protein was highly expressed in the trophoblast cells of the EPC of the developing placenta (Fig. 3, arrow) and at low levels in the surrounding uterus.
|
Ron Expression in Mouse Trophoblast Cell Lines
Murine trophoblast cell lines were employed to examine the putative function of Ron in more detail. The RT-PCR analysis was performed to determine whether the mouse trophoblast cells lines (SM9-1, SM9-2, and SM-10) expressed the Ron receptor. These cell lines were generated in the laboratory of Dr. Joan Hunt (University of Kansas Medical Center) and have been extensively characterized [38]. The MEL cells were used as a positive control for Ron expression (Fig. 4A, lane 4). Expression of Ron mRNA was detected in all three cell lines (Fig. 4A). The MEL and SM9-2 cells were the only cell lines tested that expressed both Ron and HGFL (Fig. 4, A and B, lanes 2 and 4). The RNA integrity is indicated using RT-PCR for actin (Fig. 4C).
|
Immunocytochemical analysis was performed to determine where the Ron protein was expressed in the SM9-1, SM9-2, and SM-10 cell lines. After detection of Ron protein with an antibody specific for the extracellular
chain of Ron, localization of Ron protein was determined by using confocal microscopy. Ron protein is present in all three cell lines and is localized to the cell edges as well as to the stress fibers (Fig. 5). Control slides were probed with secondary antibody only and demonstrated no specific signal above background (Fig. 5, A, C, and E).
|
Activation of Ron by HGFL Affects Invasion
To test the hypothesis that activation of Ron affects trophoblast function, cellular invasion through an extracellular matrix was measured after stimulating SM9-1 trophoblast cells with either kallikrein-treated conditioned medium from HGFL-expressing CHO cells, kallikrein-treated conditioned medium from nontransfected CHO cells, or RPMI medium only (Fig. 6). Because kallikrein treatment has been shown to functionally activate HGFL, the conditioned media was treated with kallikrein to cleave the recombinant HGFL protein into a functional protein [22, 28]. The SM9-1 cells were chosen for this experiment, because these cells not only express Ron protein but have a moderately high level of invasive ability [38]. Stimulation with HGFL results in similar levels of invasion over 24 h compared to the other two treatments. However, after 48 h of incubation in serum-free medium or kallikrein-treated conditioned medium from nontransfected CHO cells, a decrease is seen in the number of invading and surviving cells compared to those cells stimulated with HGFL (Fig. 6). This decrease may result from cell death, because cells that die during the time course would detach from the membrane.
|
SM-10 Cell Survival
Although SM-10 cells express high levels of Ron protein, these cells exhibit a very low invasive and proliferative phenotype [38]. This suggests that Ron activation via HGFL stimulation might not be necessary for invasion but might aid in some other physiological function of these cells. Therefore, SM-10 cell viability was assessed after stimulation with recombinant HGFL for 648 h. Trypan blue staining confirmed that less than 10% of the floating cells were viable (data not shown). In serum-free medium, SM-10 cells have a significant decrease in the percentage of viable cells compared to SM-10 cells maintained in serum-containing medium or in medium containing recombinant HGFL (Fig. 7). These results suggest that HGFL may aid in cell survival of trophoblast cells during implantation.
|
To test whether HGFL could block SM-10 cells from undergoing apoptosis or necrosis, a TUNEL assay was performed (Fig. 8, AD). SM-10 cells incubated for 24 h in the presence of serum or recombinant HGFL exhibited fewer TUNEL-positive cells compared to those cells incubated in serum-free medium (Fig. 8E).
|
| DISCUSSION |
|---|
|
|
|---|
While screening a panel of male and female reproductive tissues via RT-PCR, all tissues tested were found to express Ron and/or HGFL transcripts. The presence of HGFL and Ron in the testes and epididymis may be important for proper sperm function and motility. Interestingly, Ron has been found to be expressed on the colon villi of mice as well as on the ciliated epithelia of the human mucociliary transport apparatus [16, 23]. In fact, stimulation with HGFL increased the beat frequency of the cilia found in the mucociliary transport apparatus [23]. Because flagella and cilia functions are similar and have been well characterized, it will be interesting to see if HGFL stimulation may affect sperm motility in mice. Although to our knowledge no direct evidence has been shown, it has been demonstrated that Ron and HGFL are present in the testes and epididymis of the rat, and it has been hypothesized that the interactions between this receptor and ligand may be involved in spermatogenesis [39].
The in situ hybridization and immunohistochemical assays performed in this report are, to our knowledge, the first to characterize the temporal and spatial distribution of Ron during the process of mouse implantation. Trophoblast cells surrounding the implanting embryo, in the EPC and in mouse trophoblast cell lines, were found to express Ron transcripts and protein. It should be noted that differential expression of Ron mRNA was found in the three mouse trophoblast cell lines when compared to actin controls, but at the protein level, Ron expression appeared to be equal. Thus, the presence of more transcripts did not equate to higher levels of Ron protein. Although HGFL transcripts were found only in the placenta (E12 and E16) and the SM9-2 cell line, the presence of HGFL protein has already been established in virgin and pregnant uteri [35]. The presence of HGFL protein in the virgin uterus, which lacks HGFL transcripts, is most likely caused by the extremely high levels of HGFL found in the serum [15, 21]. Therefore, even if locally produced HGFL protein is absent, circulating HGFL protein is present that may activate the Ron receptor during these critical developmental time-points. It should be noted that mice deficient for HGFL are phenotypically normal and do not exhibit any reproductive problems [40]. Constitutive expression of alternatively spliced forms of Ron or the presence of alternative ligands for Ron are two possible explanations being investigated for the difference between Ron- and HGFL-deficient mice [34].
The cellular events leading to the migration and stringently regulated, aggressive invasion of the trophoblast cells surrounding the developing blastocyst have been compared to pathways involved in the process of tumor metastasis [41]. Cytokines, growth factors, and hormones aid in the activation of cellular receptors on trophoblast and cancer cell surfaces, resulting in the production of proteases that aid in matrix metabolism during invasion. For example, activation of the tyrosine kinase receptor c-Met results in increased invasion and motility of both trophoblast cells and colonic cancer epithelial cells by way of upregulation of matrix metalloproteinase production [13, 42]. Additional matrix-degrading proteins produced after receptor activation and implicated in metastasis and the process of embryo implantation include serine proteases, cysteine proteases, and matrix metalloproteinases [43]. Thus, as implantation begins, the giant cells are responsible for invading into the maternal epithelium by secreting these various proteases, and the EPC gives rise to the cyto- and spongiotrophoblast cells that are responsible for angiogenesis and vascularization of the placenta [44]. One hypothesis is that the presence of Ron transcripts and protein in the EPC and giant cells during early implantation may be necessary for the production of proteases that aid in optimal invasion. HGFL was found to stimulate trophoblast invasion through an extracellular matrix, which may have been the result of an upregulation in protease production. In fact, gelatin zymography of media collected from SM9-1 cells stimulated with kallikrein-treated, HGFL-conditioned medium demonstrated a small yet visible increase in a 72-kDa protease, which may correspond to matrix metalloproteinase-2 (data not shown).
The role of heparin-binding EGF and its receptors ErbB1 and ErbB4 in implantation are well established [811]. Recent reports indicate that Ron can heterodimerize with other receptors, including Met [45, 46], and in fact, our laboratory has recently shown that Ron forms heterodimers with the EGF receptor (Peace et al., personal communication). This interaction could occur during the time of implantation and may explain why the Ron ligand, HGFL, is not absolutely required for this process to occur [40].
The presence of Ron in only the giant cell layer at E12.0 suggests a putative role for Ron activation during trophoblast survival. As gestation advances and the placenta becomes fully mature, the invasive phenotype of these trophoblast cells is downregulated. Instead, the main role for giant cells becomes the production of communication factors and hormones such as placental lactogen-1 and proliferin to aid in maintenance of the placenta and, in turn, sustain the ongoing pregnancy [4, 5, 47, 48]. Figure 7 clearly demonstrates that HGFL stimulation enhances survival of trophoblast cells in culture. Although the exact mechanism of this enhanced survival is not understood, TUNEL analysis demonstrated increased DNA fragmentation in the absence of HGFL. Thus, future studies will encompass experiments to aid in determining if this effect on survival is caused by decreasing or inhibiting necrotic or apoptotic events.
Because the main cells expressing the Ron transcripts and protein following implantation were the giant trophoblast cells, future studies will employ cell lines that are composed of this type of trophoblast cell. An optimal cell line to use would be the Rcho-1 cells, because these are a combination of trophoblast stem cells that can be manipulated to differentiate into a homogenous population of giant cells [49]. Although the above results combined with previous studies from our laboratory have determined that Ron activation aids in implantation, Ron has additional functions during development and adulthood in mammals. Therefore, to generate Ron -/- adult mice, conditional knockout strategies or tetraploid aggregation with wild-type embyros could be employed to delete this receptor after the implantation stage.
In conclusion, the studies presented here suggest important physiological functions for Ron during normal development. These results also support the hypothesis that Ron activation via HGFL stimulation aids in implantation, possibly by affecting early and late trophoblast function.
| ACKNOWLEDGMENTS |
|---|
| FOOTNOTES |
|---|
2 Correspondence: Sandra J.F. Degen, Division of Developmental Biology, Cincinnati Children's Research Foundation, 3333 Burnet Ave., Cincinnati, OH 45229. FAX: 513 636 4317; sandra.degen{at}cchmc.org ![]()
Received: 31 July 2002.
First decision: 23 August 2002.
Accepted: 18 October 2002.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
F. Dominguez, B. Gadea, F. J. Esteban, J. A. Horcajadas, A. Pellicer, and C. Simon Comparative protein-profile analysis of implanted versus non-implanted human blastocysts Hum. Reprod., September 1, 2008; 23(9): 1993 - 2000. [Abstract] [Full Text] [PDF] |
||||
![]() |
B. E. Peace, K. Toney-Earley, M. H. Collins, and S. E. Waltz Ron Receptor Signaling Augments Mammary Tumor Formation and Metastasis in a Murine Model of Breast Cancer Cancer Res., February 15, 2005; 65(4): 1285 - 1293. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |