| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
research-article |
Departments of Molecular Anatomy3 and Obstetrics and Gynecology,4 Nippon Medical School, Tokyo 113-8602, Japan
Department of Physiology and Cell Biology,5 Ohio State University, Columbus, Ohio 43210
Department of Obstetrics and Gynecology,6 Jichi Medical University, Tochigi 329-0498, Japan
Second Department of Pathology,7 Faculty of Medicine, Miyazaki University, Miyazaki 889-1692, Japan
ABSTRACT
Chorionic villi in the human placenta serve as essential structures in fetomaternal exchanges. According to the embryology and placentology literature, during the first trimester, the cytotrophoblast (CTB) layer that is subjacent to the syncytiotrophoblast (STB) and supported by a basal lamina is nearly complete, but later, it becomes discontinuous. In the present study, we investigated the structural integrity of the CTB layer in the normal villous tree by advanced microscopy techniques using an antibody to hepatocyte growth factor (HGF) activator inhibitor type 1 (SPINT1), a potent inhibitor of HGF activators expressed exclusively on villous CTB. In full-term placenta, the cell surface of the CTB layer was spread over the basal lamina but was not interrupted. Morphometric analysis showed that throughout the villous tree, 80% of the continuity of the CTB layer of full-term placenta and 90% of that of first-trimester placenta were preserved. Gestation was accompanied by unique structural change in the basal domain of the trophoblast layer. The initially cuboidal-shaped CTB cells were transformed to flat cells with many cellular processes that, together with those of the adjacent STB, eventually covered the trophoblast basal lamina in a complex network of interdigitations. In addition, the expression levels of SPINT1, ST14, HGF, and MET mRNAs in the villous tree increased over the course of gestation. These results suggest that the structural integrity of the SPINT1-positive CTB layer may play an important role in villous differentiation and in maintenance of the villous tree via the HGF signaling system during gestation.
placenta, trophoblast
In the human placenta, the trophoblast is essential to fetomaternal exchanges. This specialized layer, which covers the surface of the chorionic villous tree, consists of two distinct layers: the syncytiotrophoblast (STB) layer, which is in contact with maternal blood in the maternal intervillous space, and the cytotrophoblast (CTB; Langhans' cells) layer, which is subjacent to the STB and is supported by a basal lamina. The CTB is the source of dividing cells that differentiate into STB, which is postmitotic. Together, these two layers separate a core of mesenchymal connective tissue and fetal blood vessels from the intervillous space. According to the embryology and placentology literature, during the first trimester, the inner CTB layer, consisting of a complete, single layer of cuboidal cells, shares about the same volume as the outermost STB layer. As pregnancy progresses, the STB is the predominant layer, whereas the CTB becomes discontinuous [1, 2]. Eventually, some areas of the STB layer come in close contact with fetal blood vessels, resulting in the formation of a vasculosyncytial membrane. These gestation-dependent structural changes of the chorionic villi greatly facilitate the fetomaternal transport of gases, ions, and molecules.
The discontinuous appearance of the CTB layer is not caused by a decline in the number of CTB cells during gestation. Stereologic analysis to estimate trophoblast growth in three-dimensional human placental villi revealed that the total number of CTB cells increases steadily until term [3]. According to the current theory, during the later stages of pregnancy, the CTB layer cannot keep up with the rapid branching and expansion of the villous surface, resulting in the interruption of the CTB layer below the still-intact STB layer [1]. However, very little is known about the structural alterations of the CTB even during normal gestation.
One of the difficulties in studying the growth and dynamics of the villous CTB layer is a shortage of specific markers for CTB cells. Hepatocyte growth factor (HGF) activator inhibitor type 1 (SPINT1) is a membrane-associated, Kunitz-type serine proteinase inhibitor that initially was identified as a potent inhibitor of HGF activator (HGFAC) [4]. The human placenta expresses high levels of SPINT1 mRNA [4], and immunohistochemical studies have shown that SPINT1 is expressed exclusively on the CTB of chorionic villi but, apparently, not on STB or extravillous trophoblasts [5]. Pötgens et al. [6] used a cell sorting method involving a specific mouse monoclonal antibody (mAb) to SPINT1 (clone C76/18) to isolate highly pure, villous CTB cells without contamination from the STB or extravillous trophoblast. Thus, SPINT1 is considered to be a novel specific marker of the CTB in human placenta.
In the present study, we investigated the structural integrity of the CTB layer in normal human placenta by advanced microscopy techniques employing a specific antibody to SPINT1. In testing the generally accepted theory that the CTB layer becomes discontinuous during gestation, we found that the cell surface of this layer is spread over the basal lamina but is not interrupted in full-term placenta. In addition, we examined the expression of SPINT1, HGFAC, ST14 (previously known as matriptase), HPN (also known as hepsin), PRSS8 (also known as prostasin), HGF, and MET (also known as HGF receptor) mRNAs in human chorionic villi during gestation by quantitative RT-PCR. Our data provide fresh insight regarding the differentiation and growth of the human chorionic villous tree during gestation.
Human first-trimester and full-term placentas were obtained according to a protocol approved by the Nippon Medical School Hospital Ethics Committee. Five first-trimester placental tissues (8–14 wk of gestation) were obtained from legal abortions and processed within 20 min. Six full-term tissue samples (37–38 wk of gestation) from uncomplicated cesarean deliveries were processed as soon as possible following delivery (within 20 min).
Morphometric Analysis Using Immunofluorescence Microscopy
Immunofluorescence microscopy using cryostat sections was carried out as described previously [7]. Briefly, placental tissues were fixed for 2 h at 22°C with 4% paraformaldehyde in 0.1 M sodium cacodylate buffer (pH 7.4) containing 5% sucrose. The samples were washed with the same buffer, embedded in Tissue-Tek O.C.T. compound (Sakura Finetechnical, Tokyo, Japan), placed in aluminum-foil molds, flash-frozen in liquid nitrogen, and then stored at –80°C until use. Tissue sections (thickness, 5 µm) were prepared with a Microm HM 550 cryostat (Microm, Walldorf, Germany), mounted on round glass coverslips (diameter, 13 mm; no. 1 thickness; Matsunami, Osaka, Japan), coated with 2% 3-aminopropyltriethoxy-silane (Sigma Chemical Co., St. Louis, MO), and then allowed to air dry.
To block nonspecific protein-binding sites, the sections were washed three times in PBS and then incubated in 1% BSA and 5% normal goat serum in PBS for 30 min at 22°C. The sections were then incubated with two primary antibodies, mouse anti-SPINT1 mAb (clone C76/18, 2.0 µg/ml) [8, 9] and rabbit anti-collagen type IV (anti-COL4, 0.5 µg/ml; Rab Biochemicals, Berlin, Germany), for 30 min at 37°C and, subsequently, with two secondary antibodies, Alexa Fluor 488-labeled goat anti-mouse IgG and 594-labeled goat anti-rabbit IgG (10 µg/ml; Molecular Probes, Eugene, OR), again for 30 min at 37°C. The sections were counterstained with 4',6-diamidino-2-phenylindole dihydrochloride (DAPI) and mounted in ProLong anti-photobleaching medium (Molecular Probes) on glass microscope slides. Control sections received the same treatment, with the exception that the primary antibody was either omitted or replaced with purified nonimmune IgG. Fluorescence and differential interference contrast (DIC) images were obtained with a BX60 microscope (Olympus, Tokyo, Japan) equipped with a Spot RT SE6 CCD camera (Diagnostic Instruments, Sterling Heights, MI) and processed with the MetaMorph image-analysis system (Universal Imaging, Downingtown, PA). Figures were compiled using PhotoShop CS software (Adobe Systems, San Jose, CA).
The relative continuity of the CTB layer in the human placenta (rC) was determined by immunofluorescence microscopy followed by measurement of the fluorescence signal using the MetaMorph system. Briefly, the immunofluorescence and DIC images were merged, after which the collagen type IV (COL4)-defined basal lamina was delineated using the MetaMorph Traced Line function. After the created outline (width, one pixel) was defined with the Paint Region function followed by segmentation using the Binary function, the outline width was dilated from 1 to 10 pixels using the Morphology Filters function. The pixel area of the SPINT1-positive outline, indicating the CTB layer in the selected outline, was measured using the Integrated Morphometry Analysis function. The rC was defined as follows: rC = (C1/C2) x 100, where C1 is the pixel area of the SPINT1-positive outline and C2 is the total pixel area of the selected outline of the basal lamina. For the terminal, intermediate, and stem villi of full-term placenta, 140, 45, and 30 randomly captured profiles, respectively, were analyzed, whereas 40 randomly captured profiles were analyzed for the villi of first-trimester placenta.
Three-Dimensional Imaging under Confocal Microscopy
Serial confocal images of cryostat sections (thickness, 50 µm) were obtained with an IX81 microscope equipped with a FV1000 confocal scan unit and processed with an FV10-ASW image-analysis system (Olympus). Sections were double-immunostained with anti-SPINT1 and anti-Fc gamma receptor IIb (anti-FCGR2B, Ab 163.96, 1:3200 dilution) [10] as described above. Anti-FCGR2B was used as a placental endothelial marker [10]. Optical sections were acquired in 1-µm steps and perpendicular to the z-axis (microscope optical axis) using a triple 488-nm multi-Ar, 543-nm HeNe, and 405-nm LD laser system.
Ultrahigh-Resolution Immunofluorescence Microscopy
Ultrathin cryosections were prepared as described previously [11, 12]. Briefly, placental tissues were fixed as described above, solidified with gelatin, infiltrated with 2.3 M sucrose, and mounted on specimen pins designed to fit a cryo-ultramicrotome (Leica, Wetzlar, Germany). The samples were then both frozen and stored in liquid nitrogen until sectioned. Ultrathin cryosections (thickness, 50–100 nm) were cut with a Leica ultramicrotome EM UC6b equipped with an FC6 cryounit and then transferred to coverslips.
For double-labeling, ultrathin cryosections were incubated for 30 min at 37°C with the primary antibodies mouse anti-SPINT1 mAb and goat anti-keratin 7 antibody (anti-KRT7; catalog no. sc-17116, 1.0 µg/ml; Santa Cruz Biotechnology, Santa Cruz, CA). The blocking solution consisted of 1% BSA in PBS. The sections were incubated for 30 min at 37°C with the secondary antibodies Alexa Fluor 488-labeled donkey anti-mouse IgG and Alexa Fluor 594-labled donkey anti-goat IgG (each at 10 µg/ml; Molecular Probes). Double labeling of anti-SPINT1 and anti-COL4 also was performed as described above. The tissue sections were counterstained with DAPI. After immunostaining, images of the ultrathin cryosections were processed as described above.
Conventional transmission-electron microscopy. Human placentas were fixed for more than 8 h at 22°C by incubation in a mixture of 2.5% glutaraldehyde and 2% paraformaldehyde in 0.1 M sodium cacodylate buffer (pH 7.4) containing 0.05% CaCl2. Subsequently, the tissue was postfixed for 1 h at 4°C in 2% osmium tetroxide in 0.1 M sodium cacodylate buffer (pH 7.4) containing 1.6% potassium ferrocyanide and then rinsed three times in distilled water before dehydration in a graded series of ethanol. The specimens were treated with propylene oxide and embedded in Epon 812 resin. Thin sections were cut, mounted on formvar-coated grids, and contrast-stained with lead citrate and uranyl acetate before examination under a JEM-1010 (JEOL Ltd., Akishima, Japan) or H-7500 (Hitachi High-Technologies Co., Tokyo, Japan) transmission-electron microscope.
Scanning-electron microscopy with NaOH treatment. Human placentas were fixed as described above. Cryostat sections (thickness, 20–50 µm) on 2% 3-aminopropyltriethoxy silane-coated coverslips also were prepared as described above. The sections were then fixed again with 2% glutaraldehyde in PBS at 4°C overnight in a 24-well plate (Sumitomo Bakelite Co., Tokyo, Japan), and immersed in a 6 N NaOH aqueous solution for 20 min at 60°C to remove extracellular matrix components, such as collagen fibrils and basal lamina [13]. After NaOH treatment, the samples were washed three times in PBS. Because most placental sections came off the coverslips during the PBS wash, they were carefully collected into 1.5-ml microcentrifuge tubes. The sections were conductive-stained with 2% osmium tetroxide for 1 h at 4°C, rinsed in distilled water, dehydrated through a graded series of ethyl alcohol, treated with t-butyl alcohol, frozen for 10 min at 4°C, and then evaporated in a vacuum chamber [14]. The freeze-dried samples were mounted on aluminum stubs with adhesive carbon tape, coated with platinum-palladium in a Hitachi E102 ion sputter coater, and then observed in a Hitachi S-3000N scanning-electron microscope at an accelerating voltage of 15 kV.
First-trimester and full-term placental tissues were analyzed by RT-PCR. Total RNA was extracted from these samples using the Isogen reagent (Nippon Gene, Tokyo, Japan) according to the manufacturer's instructions. Complementary DNA was synthesized using 500 ng of total RNA, the ExScript RT reagent kit, and a random primer (Takara Bio, Shiga, Japan). The primers for SPINT1 (Takara Primer Set ID: HA038486), HGFAC (HA004501), ST14 (HA051736), HPN (HA004727), PRSS8 (HA008017), HGF (HA032509), MET (HA031954), and YWHAZ (HA038050) were purchased from the Real Time support system (Takara Bio). The RT-PCR was carried out on an Applied Biosystems 7300 real-time PCR system (Foster City, CA) with a master mix of SYBR green I premix ExTaq (Takara Bio) according to the manufacturer's instructions. The thermal cycle profile was 10 sec at 95°C, followed by 40 cycles of 5 sec at 95°C and 34 sec at 60°C. Product amplification specificity was determined by melting-curve analysis. As an internal standard, each individual sample was normalized to its tyrosine 3-monooxygenase/tryptophan 5-monooxygenase activation protein, zeta polypeptide (YWHAZ) mRNA content [15]. The ratio of SPINT1, HGFAC, ST14, HPN, PRSS8, HGF, or MET mRNA to YWHAZ mRNA was calculated to adjust for variations in the RT-PCR.
Morphometric Analysis Using Immunofluorescence Microscopy of SPINT1 Expression in Human Placental Villi
In villi of first-trimester human placenta, SPINT1 was exclusively expressed on the CBT (Fig. 1A) and was barely detectable, if at all, in the STB layer (Fig. 1B). These immunofluorescence microscopy results are consistent with earlier findings obtained by 3,3'-diaminobenzidine-based immunohistochemistry [5]. Within the villous tree of full-term placenta, the distribution of SPINT1 appeared to be somewhat uneven at low magnification, albeit SPINT1 was, indeed, present (Fig. 1C). At higher magnification, the SPINT1-defined CTB layer was thinner, but its continuity was maintained even in terminal villi (Fig. 1D). The STB, stromal cells, and fetal endothelial cells were hardly immunostained with anti-SPINT1 in full-term placenta, and no fluorescence signal above background was detected in control sections (data not shown). The three-dimensional distribution of SPINT1 in the CTB of full-term placenta also was examined by laser-scanning confocal microscopy. A series of optical sections (z series) acquired from cryostat sections showed the continuity of the SPINT1-defined CTB layer even in terminal villi of full-term placenta (Fig. 1E). To investigate the structural integrity of the CTB layer in terminal, intermediate, and stem villi of full-term placenta, the relative continuity of the SPINT1-defined CTB layer was determined using the MetaMorph image-analysis system (Fig. 2). The results showed that the continuity of the CTB layer in full-term placenta was maintained, in that approximately 80% of the structural integrity was preserved at all three anatomic locations examined. No significant differences were observed in the continuity of the CTB layer at the three sites. In first-trimester placenta, approximately 90% of the structural integrity remained (Fig. 2). It also should be noted that no differences were found in the continuity of the CTB layer between first-trimester and full-term placentas.
|
|
Ultrahigh-Resolution Immunofluorescence Microscopy of the Subcellular Distribution of SPINT1 in Full-Term Placental Villi
The precise distribution of the CTB in full-term human placental villi was analyzed by ultrahigh-resolution immunofluorescence microscopy (UHR-IFM) of ultrathin cryosections (Fig. 3). Both STB and CTB layers were readily identifiable by immunostaining of keratin 7 (KRT7), a widely used trophoblast marker [16, 17] (Fig. 3A). Double-immunostaining with SPINT1 and KRT7 allowed the detection of SPINT1 on the CTB but not on the STB, stromal cells, or fetal endothelial cells. In tangential views of the juxtanuclear area of CTB cells, KRT7 was predominantly distributed just beneath the cell membrane, and most fluorescence signals indicated that SPINT1 was mainly localized just outside of the KRT7-positive area (Fig. 3A). This observation is in good agreement with the properties of the mAb, in which the epitope of clone C76/18 was shown to recognize an extracellular domain of SPINT1 [9]. The peripheral areas of the CTB cells were found to be extremely thin. The KRT7 most likely stained two layers (Fig. 3B): an outer layer, containing KRT7 from the STB, and an inner layer, most likely consisting of KRT7 from both the STB and CTB. In these peripheral areas, SPINT1 was exclusively distributed in the inner layer, as visualized with anti-KRT7. Moreover, the distribution of the fluorescence signal showed little overlap between SPINT1 and KRT7 in the inner layer. Horizontal views of the CTB layer in full-term terminal villi showed that the SPINT1-positive CTB cells exhibited many cellular processes (Fig. 3C). It also should be emphasized that the STB was SPINT1-negative.
|
Double-immunolabeling of SPINT1 and COL4 was used to assess whether SPINT1 was present in the basal lamina supporting the STB-CTB lining of full-term villi (Fig. 3D). The fluorescence signal for SPINT1 was barely colocalized with that for COL4 and was located primarily just above the basal lamina labeled with anti-COL4 (Fig. 3D, inset). These results from UHR-IFM strongly suggest that SPINT1 is expressed exclusively on the cell surface of the CTB.
Electron Microscopy of Full-Term Placental Villi
Electron microscopy revealed the fine structure of the basal domain of the trophoblastic layer (Fig. 4). The plasma membrane infoldings of the STB and CTB were widely spread over the basal lamina and often observable in tangential views of full-term terminal villi (Fig. 4A). The basal surfaces of the CTB and STB layers were three-dimensionally revealed by NaOH treatment followed by scanning-electron microscopy (Fig. 4, B and C). Terminal villi were treated with NaOH to remove the basal laminas and extracellular matrix components from the villi, thus enabling direct observation of the basal domain of the trophoblast layer, which is in direct contact with the basal lamina. In full-term placental villi, portions of the CTB layer were very thin but nonetheless identifiable (Fig. 4B). Numerous fine, irregular process-like structures in the basal domain facing the basal lamina were present (Fig. 4C). The cell processes from both the CTB and adjacent STB were intricately interdigitated and covered the basal lamina.
|
The histochemical findings in the integrity of the CTB layer simultaneously indicate the unique distribution of SPINT1; it is expressed throughout the chorionic villous tree during gestation. Little information, however, is available for the transcription level of SPINT1 within the villous tree. To provide additional information of SPINT1 in developing placenta, we examined de novo transcription of SPINT1 and genes related to the regulation of proteolytic activation of HGF by real-time PCR. The expression of mRNA transcripts of SPINT1, HGFAC, ST14, HPN, PRSS8, HGF, and MET during gestation was investigated. The SPINT1 mRNA expression in full-term placenta was approximately 2.5-fold higher than in first-trimester placenta (Fig. 5). During gestation, the expression patterns of ST14, HGF, and MET mRNAs were similar to that of SPINT1 mRNA (Fig. 5). Although PRSS8 mRNA was obvious, no differences were observed in its expression level between first-trimester and full-term placentas (Fig. 5). The HFGAC mRNA was detectable in full-term placental samples by RT-PCR but not by real-time PCR (data not shown), which suggested that HGFAC is expressed at very low levels in this tissue [5]. The HPN mRNA was undetectable in placental samples by real-time PCR (data not shown).
|
In the present study, we analyzed the structural integrity of the CTB layer in human full-term placenta. The results provide further insight regarding the morphogenesis of the chorionic villous tree. Examination of the CTB layer by advanced microscopy techniques, including UHR-IFM, showed that it becomes thinner but largely maintains its continuity during gestation. It has long been accepted that as chorionic villi differentiate during the course of gestation, the CTB layer cannot keep up with the rapid branching and expansion of the villous tree, resulting in the interruption of this layer [1, 2]. The CTB cells are found beneath 20% of the villous STB in full-term placenta [1]. Earlier stereologic analysis showed that STB volume per CTB nucleus is constant during gestation, despite a decline in the number of CTB nuclei per unit STB surface area [3, 18]. Theories regarding the incompleteness of the CTB layer during late gestation have relied on these stereologic studies in which nuclei were counted. The current theories, however, have not taken into account the dramatic change in shape of the CTB cells during late gestation compared to the first-trimester. Our approach more accurately reveals the structural integrity of the CTB layer, because SPINT1 immunocytochemistry shows the entire outline of the CTB cells. The results showed that more than 80% of the continuity of the CTB layer is preserved throughout the villous tree of human full-term placenta and that differences in the structural integrity of the CTB layer between first-trimester and full-term placentas are less than 10%, because the structural integrity of the former is approximately 90% (Fig. 2). The earlier, quantitative results permit an alternative interpretation—that the CTB is extended concomitantly with the expanding villous surface of the STB and that the thickness of the CTB greatly declines during gestation. According to this viewpoint, our findings are in good agreement with those of the earlier studies, and the incompleteness of the CTB layer during late gestation thus has been overestimated. A new scheme explaining the morphogenesis of the CTB layer of the human chorionic villous tree during gestation is presented in Figure 6.
|
The UHR-IFM of ultrathin cryosections greatly facilitated efforts to determine the exact subcellular distribution of SPINT1 in the complex villous structure of human placenta. This new technique was developed for immunofluorescence microscopy studies of the subcellular distribution of and interactions between specific macromolecules in complex structures in vivo [11, 12, 19, 20]. In UHR-IFM, the z-axis resolution of placental samples is 100 nm or less, because it is equal to the thickness of the ultrathin cryosections. In the present study, the extraordinary thinness of the sections minimized false colocalization of SPINT1 with other macromolecular markers (e.g., COL4) (Fig. 3D) in the z-dimension and clearly revealed that SPINT1 is located on the surface of CTB cells. The UHR-IFM was previously employed to investigate the mechanisms of IgG transfer from maternal to fetal blood in human placenta, which led to the identification of a novel Fc
receptor-defined, IgG-containing organelle in placental endothelial cells [10].
Our conclusions depend on the use of a new specific marker of the CTB, SPINT1, which is a membrane-associated, Kunitz-type serine proteinase inhibitor [4]. Recently, a splice variant (SPINT1_isoform 1 [SPINT1_i1]), which contained an extra 16-amino-acid insertion after the first Kunitz domain, also was reported [21]. No significant differences were observed in tissue distribution, enzyme activity, or specificity between SPINT1_i1 [21] and SPINT_i2 [4], and both are potent inhibitors of HGFAC, ST14, and HPN [8, 22–24]. The HGFAC is a factor XII-like serine proteinase that plays a critical role in the activation of pro-HGF/SF in injured tissue and in cancer tissue [25–27]. A multidomain transmembrane serine protease of the S1 trypsin-like family, ST14 activates pro-HGF/SF and pro-urokinase-type plasminogen activator in addition to protease-activated receptor 2 [24, 28, 29]. The ST14 has been implicated in several cellular events, such as epithelial migration, remodeling of the extracellular matrix, cell differentiation, and oncogenesis in epithelial cancers [30]. An activator of HGF, HPN is highly expressed in prostate and ovarian cancers as well as in normal liver [22, 23]. Thus, SPINT1 is considered to be an important regulatory molecule for HGF activators in the pericellular microenvironment [25, 31, 32]. Although little is known regarding the function of placental SPINT1, Tanaka et al. [33] used Spint1 (also known as HAI-1)-deficient mice to demonstrate a physiologic role for the inhibitor during early placental development. In murine placenta, SPINT1 is present on the STB and chorionic trophoblast in the labyrinth, which in the human placenta fundamentally corresponds to chorionic villi containing SPINT1-positive CTB cells. They found that the branching morphogenesis of the labyrinth layer is severely impaired in the Spint1–/– placenta, resulting in the failure of placental development and function. Consequently, Spint1–/– mice exhibit embryonic lethality at 10.5 days postcoitum. The results from Spint1-knockout mice indicate that SPINT1 plays an important role in the early stage of placental development. Interestingly, the expression levels of mouse Spint1 and human SPINT1 mRNAs in placentas increase over the course of gestation [33] (Fig. 5). Even in terminal villi, which are the final branches of the villous tree during late gestation, CTB cells continue to express SPINT1 mRNA (data not shown). Thus, SPINT1 expression and synthesis likely occur throughout the human chorionic villous tree during late gestation. In addition, SPINT1-related HGF activators (i.e., HGFAC and ST14), HGF, and MET are present in the chorionic villous tree of human full-term placenta [5, 34–36] (Fig. 5), and villous differentiation associated with angiogenesis persists, even at term [37, 38]. The HGF stimulates the branching morphogenesis of human villous CTB [39, 40] and human placental angiogenesis [40]. Taking all these findings into consideration, we hypothesize that the structural integrity of the SPINT1-positive CTB layer plays an important role in regulating the efficient localization and concentration of HGF activators on and around the basal surface of the CTB layer. This function of the inhibitor allows villous differentiation and maintenance of the chorionic villous tree via the HGF signaling system during gestation.
Recently, SPINT1 was found to be a potent inhibitor not only of the HGF activators described above but also of PRSS8, a glycosylphosphatidylinositol-anchored serine protease that is highly expressed in human and mouse placentas [31, 41] (Fig. 5). In human airway epithelium, PRSS8 activates epithelial sodium channels [42], although its physiologic functions in the placenta remain unclear. The PRSS8/SPINT1 system may have another important role in placental development in addition to regulation of the HGF signaling pathway.
Another aspect of the present study was the shape transformation of CTB cells. As a component of the placental barrier, this property is germane to the mechanism of fetomaternal exchange. Changes in the placental barrier, in the form of a decrease in the thickness and number of layers, occur in parallel with villous differentiation. As noted above, the CTB layer has been considered to play a reduced role in maintaining the placental barrier during late gestation; however, the present results suggest that the extremely thin layer of CTB cells is involved in barrier maintenance. It should be emphasized that we do not interpret our findings as evidence for the complete interruption of communication between the STB layer and fetal endothelium by the CTB layer during late gestation. The initially cuboidal-shaped CTB cells are transformed to flat cells with many long cellular processes that, together with those of STB, eventually cover the trophoblast basal lamina in a complex network of interdigitations. This unique structural change of the basal domain of the trophoblast layer may be involved in the regulation of molecular transport for fetomaternal exchange, an important function of the placental barrier.
Although our data challenge currently held ideas about the morphogenesis of the CTB layer in human placenta, the functional roles of this layer during late gestation remains to be elucidated. In this context, it would be of interest to use SPINT1, as a specific marker, to examine the CTB in pathological conditions that affect villous development and vascular morphogenesis, such as intrauterine growth retardation and preeclampsia. These types of experiments should be carried out concomitantly with conditional knockout studies in animal models. Further studies in our laboratories are aimed at investigating the physiologic functions of the CTB in late gestation.
ACKNOWLEDGMENTS
We thank Yukihiro Furuyama of Molecular Devices Corp., Yukiko Hachiya, Sachiko Fujii, and Takuji Kosuge for technical support. We also are indebted to Drs. Clark L. Anderson and Jean-Luc Teillaud for providing anti-FCGR2B antibody.
FOOTNOTES
1Supported in part by grants-in-aid (nos. 16390479 and 16659457 to T.T. and no. 17791140 to M.M.) from the Ministry of Education, Culture, Sports, Science, and Technology of Japan. 
Correspondence: 2 Toshihiro Takizawa, Department of Molecular Anatomy, Nippon Medical School, 1-1-5 Sendagi, Tokyo 113-8602, Japan. FAX: 81 3 5685 3052; e-mail: t-takizawa{at}nms.ac.jp
Received: 29 July 2006.
First decision: 31 August 2006.
Accepted: 3 October 2006.
REFERENCES
R-defined, IgG-containing organelle in placental endothelium. J Immunol 2005; 175:2331–2339
and CD31 in human placental endothelial cells. J Electron Microsc 2006; 55:107–112This article has been cited by other articles:
|
|
 |
|
  H. Satoh, E. Kiyota, Y. Terasaki, T. Sawamura, K. Takagi, H. Mizuta, and M. Takeya Expression and Localization of Lectin-like Oxidized Low-density Lipoprotein Receptor-1 (LOX-1) in Murine and Human Placentas J. Histochem. Cytochem., August 1, 2008; 56(8): 773 - 784. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
