| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Research Article |
Laboratories of Experimental Hormonology,3 Histology,4 and Pharmacology,5 Université Libre de Bruxelles, B-1070 Brussels, Belgium
ABSTRACT
During human pregnancy, the trophoblast layer is in direct contact with maternal albumin. In contrast to immunoglobulins, albumin does not cross the placental barrier. However, albumin affects the trophoblast placental lactogen and chorionic gonadotroph secretion. The present study investigated the interaction between albumin and syncytiotrophoblast using human term placental explants. Bovine serum albumin, labeled with either 125I or fluorescein isothio-cyanate, was taken up rapidly by placental explants. This process was temperature-sensitive. The internalized labeled BSA quickly outflowed from the tissue at the maternal side, largely without any major modification in molecular weight. Colchicine (1 mM), which disrupts the microtubule network, or cytochalasin B (40 µM), which disassembles filamentous actin, did not interfere with the placental transmembrane movements of labeled BSA. Megalin, clathrin, and caveolin 1 are three membrane proteins associated with albumin endocytosis in other tissues, but only megalin and clathrin were detected in the syncytiotrophoblast layer by immunohistochemistry. The uptake of labeled BSA into placental explants was not modified by 4,4'-diisothiocyanatostilbene-2,2'-disulfonic acid (1 mM) or 5-nitro-2-(3-phenylpropylamino)benzoic acid (100 µM), two pharmacological tools known to disturb megalin-mediated albumin endocytosis. By contrast, methyl-ß-cyclodextrin (10 mM) and chlorpromazine (1.4 mM), both of which disrupt the clathrin-mediated endocytotic system, significantly reduced the uptake of labeled BSA. These data suggest, to our knowledge for the first time, that maternal albumin is actively internalized into the human trophoblast according to an apical recycling pathway. This temperature-sensitive process does not depend on an intact cytoskeleton, but it is associated with a clathrin-mediated endocytotic system.
human chorionic gonadotropin, placenta, pregnancy, syncytiotrophoblast
INTRODUCTION
During human pregnancy, maternal albumin is the most abundant plasma protein (5%, w/v) in direct contact with the trophoblast layer. Yet, its transfer across the placental barrier from the maternal to the fetal circulation is considered to be negligible compared with that of proteins such as immunoglobulin G [1, 2]. However, albumin seems to be internalized into isolated trophoblastic cells [3]. The significance of such albumin movements and their underlying mechanisms remains unknown, but the possibility that trophoblasts express low-affinity albumin receptors has been evoked. Moreover, albumin recently was found to stimulate the in vitro release of two placental polypeptidic hormones, the placental lactogen (hPL) and hCG from term placental explants [4, 5]. Taken together, these observations suggest a physiological role for maternal albumin in the placenta.
At least three different mechanisms have been described for albumin endocytosis in other tissues. First, the albumin uptake and transport in the vascular endothelium involve caveolae-mediated potocytosis [6, 7]. In such a case, a few protein molecules bind specifically to the membrane receptor albumin-binding glycoprotein gp60 (or albondin) in caveolae. This, in turn, induces the formation of a vesicle, trapping more extracellular fluid-phase albumin. The bulk of internalized albumin will then be transcytosed to the basal cellular side. So far, in placenta, caveolin 1, which is a specific marker of the caveolae [8], has been reported to be located strictly in the fetal capillary endothelium but not in the trophoblast layer [9]. Second, albumin also has been shown to be transcytosed via a clathrin-mediated endocytic pathway (coated pits) across the epithelium of the lactating mouse mammary gland [10]. According to this transcytotic process, most of the albumin molecules avoid being routed to lysosomes and degraded. Whether the coated-pits system is present in the trophoblast layer remains to be substantiated. Third, albumin endocytosis is associated with the megalin receptor in the kidney proximal tubules [11, 12]. At that site, megalin acts essentially as a scavenger receptor, playing an important role in the tubular uptake of filtered plasma proteins in association with their degradation in the lysosomes [13, 14]. Interestingly, megalin has been shown to be present in the trophoblast, but its precise role remains to be characterized [15].
To define the interactions between albumin and the trophoblast layer, the present study, using human term placental explants, aimed to analyze the movements of labeled BSA in the trophoblast layer.
MATERIALS AND METHODS
The protocol of this investigation was approved by the Ethics Committee of the Faculty of Medicine of the Université Libre de Bruxelles (Belgium).
Tissue Preparation
Human placentas were obtained after vaginal delivery following normal pregnancy (37–41 wk of gestation) and immediately brought to the laboratory. Villous tissue free of visible infarct, calcification, or hematoma was sampled from at least five cotyledons at a distance midway between the chorionic and basal plates. These core parts of cotyledons were cut into multiple cubic segments (10- to 20-mm edge) and thoroughly rinsed with cold (4°C) Hanks medium (pH 7.4) containing 137 mM NaCl, 5 mM KCl, 1 mM CaCl2, 1 mM MgSO4, 0.3 mM Na2HPO4, 0.4 mM KH2PO4, and 4 mM NaHCO3. These small pieces were kept overnight under 100% O2 in an open system at 4°C in a Hepes-buffered salt solution (pH 7.4) with the following composition: 10 mM Hepes, 139 mM NaCl, 5 mM KCl, 1 mM MgCl2, and 4.2 mM glucose, supplemented with 0.1% (w/v) BSA (fraction V), 50 IU/ml of penicillin, and 50 µg/ml of streptomycin (Gibco BRL) [16]. The next morning, the segments were cut into smaller explants (2- to 5-mm edge) that were collected in a Petri dish containing cold Hanks medium. All reagents were of analytical grade and, unless indicated otherwise, were purchased from Sigma Chemical Co.
Albumin Uptake into Incubated Placental Explants
Placental explants were randomly distributed in vials containing 1 ml of a Hepes-buffered incubation medium (n = 2 explants/vial). This medium (pH 7.4), which was gassed under 100% O2 in an open system, was composed of 10 mM Hepes, 139 mM NaCl, 1 mM CaCl2, 5 mM KCl, 1 mM MgCl2, and 4.2 mM glucose. Incubation of placental explants started in a shaking water bath at 37°C for three consecutive, 60-min equilibration periods [16, 17]. Then, the kinetic pattern of albumin uptake into the tissue was studied by experimental incubation of the explants during 2, 5, 15, 30, or 60 min in the same incubation medium supplemented either with [125I]BSA (total count [TC], 
500 000 cpm/100 µl) or fluorescein (FITC-BSA, ratio fluorescein / BSA: 5-1) (0.15µg/µl, molecular Probes-VWR, Belgium). BSA was radiolabeled with 125I (Amersham, Biosciences, UK) using the chloramine-T method [18, 19]. Labeling albumin with 125I preserves the native protein conformation and its binding to potential receptors [20]. The experiment was ended by removing the explants from the incubation medium. They were washed (ten washings), at 4°C, first with an isotonic NaCl solution, secondly with the incubation medium supplemented with unlabeled BSA (0.5%, w/v) to remove extracellular radioactivity or fluorescence (last washings were checked for the absence of remaining radioactivity and fluorescence) [6, 21]. Explants were immediately counted for radioactivity in a g-counter (Packard Instruments, Zaventem, Belgium) or, when FITC-BSA was used, put into 500µl of H2O and sonicated. After centrifugation, supernatants were collected, and FITC-BSA was assayed using a fluorimeter (Versafluor; Bio-Rad). Calculations were based on a calibration curve with known concentrations of FITC-BSA diluted in the incubation medium. When [125I]BSA was used, the total amount of radiolabeled BSA inside placental explants (cpm/mg tissue) was expressed as a percentage of the total radioactivity per 100 µL of the incubation medium. Experiments were conducted at 37°C and 4°C.
To assess a potential inhibition of the [125I]BSA or FITC-BSA uptake into the placental explants, methyl-ß-cyclodextrin (10 mM) [6], genistein (200 µM) [8], chorpromazine hydrochloride (1.4 mM) [10], amiloride hydrochloride (500 µM) [22], 4,4'-diisothiocyanatostilbene-2,2'-disulfonic acid (DIDS; 1 mM) [23], 5-nitro-2-(3-phenylpropylamino)benzoic acid (NPPB; 100 µM) [23], colchicine (1 mM) [4], or cytochalasin B (40 µM) [4] were added to the incubation medium during the third hour of the equilibration period and during a 5- or 30-min experimental incubation period in presence of labeled BSA. When dimethyl sulfoxide (DMSO) was used to dissolve the pharmacological agent, the control medium was supplemented with an equivalent volume of DMSO.
The impact of various extracellular BSA concentrations and colloidal osmotic pressure on the labeled BSA uptake was evaluated by supplementing the medium with BSA (0.1%, 0.5%, and 5%, w/v) or with dextran (70 kDa; 4.5%, w/v) or polygelin (4%, w/v), both in presence of BSA (0.5%, w/v), during a 20-min experimental incubation period.
For these experiments, labeled BSA uptake was expressed as a percentage of the control value (100%).
Albumin Outflow from Incubated Placental Explants
Explants were distributed in vials (n = 2 explants/vial) containing 1 ml of the previously described Hepes-buffered incubation medium. After two 60-min equilibration periods in a shaking water bath thermostabilized at 37°C, explants were incubated for a 60-min loading period in the same medium supplemented with [125I]BSA (TC, 500 000 cpm/100 µl) or FITC-BSA (0.15 µg/µl). Then, explants were washed (10 washings) at 4°C with an isotonic NaCl solution and then with the incubation medium supplemented with BSA (0.5%) to remove extracellular radioactivity or fluorescence (last washings were checked for the absence of remaining radioactivity or fluorescence). Explants were incubated for 5, 10, 15, or 30 min in vials (n = 2 explants/vials) containing 1 ml of the Hepes-buffered incubation medium supplemented with BSA (0.5%, w/v). The experiment was ended by removing the explants from the incubation medium. An aliquot of each incubation medium sample was counted for radioactivity in the 
-counter or for fluorescence in the fluorimeter. The radioactivity or fluorescence content within the explants also was determined as previously described. Integrity of the outflowing labeled BSA was checked using gel filtration (Sephadex G25 column; Amersham Pharmacia Biotech).
The impact of various extracellular BSA concentrations and colloidal osmotic pressure on the labeled BSA outflow was evaluated by supplementing the medium with BSA (0.5%, 1%, 2%, and 5%) or with dextran (4.5%) or polygelin (4%), both in presence of BSA (0.5%), during a 20-min experimental incubation period.
Labeled BSA outflow was expressed as a percentage of the total [125I]BSA/FITC-BSA incorporated in the placental explants.
Albumin Outflow from Perifused Placental Explants
The study of [125I]BSA outflow from preloaded explants was performed as previously described for the measurement of 45Ca outflow [24]. Briefly, explants were distributed in vials (n = 2 explants/vial) containing the Hepes-buffered incubation medium. After a 90-min equilibration period in a shaking water bath thermostabilized at 37°C, explants were incubated for an additional 90-min period in the same medium supplemented with [125I]BSA (TC, 500 000 cpm/100 µl). The explants were then washed with a nonradioactive incubation medium to remove extracellular radiolabeled albumin [24]. Placental explants were put on a moist cellulose filter and then inserted in a perifusion chamber connected to two reservoirs through a three-way valve. Reservoirs containing the incubation media were kept at 37°C and gassed with 100% O2 in an open system. The perifusate was delivered at a constant rate (1 ml/min) by means of a peristaltic pump. The effluent was collected continuously in glass vials for successive 1-min periods over 60 min. Exposure to 5% BSA or 4.5% dextran (in the presence of 0.5% BSA) was performed between 14 and 39 min of the experimental period. An aliquot of each collected fraction was immediately counted for radioactivity in a -counter. At the end of the experiments, the radioactive content within the explants also was determined [24].
The amount of [125I]BSA released in each vial (collected at 1-min intervals) was expressed as a fractional outflow rate (% instantaneous explant content/1 min).
Immunohistochemical Detection of Megalin, Caveolin 1, and Clathrin Using the Avidin Biotin Technique
Freshly delivered and preserved placental explants either were fixed using paraformaldehyde (4%, w/v) in PBS with 40 mM NaH2PO4.H2O and 160 mM Na2HPO4.7H2O and then paraffin-embedded or were frozen in methylbutane (99%) at –80°C. Four-micrometer sections were cut from paraffin blocks on a microtome (Reichert-Jung), placed onto Superfrost slides (International Medical Products), and rehydrated. Ten-micrometer sections were cut from frozen blocks on a cryostat (model CM 3050; Leica) set at –18°C to –20°C, collected onto Superfrost slides, and fixed with paraformaldehyde (4%) in PBS buffer for 10 min.
Both rehydrated and frozen sections were bathed for 30 min in H2O2 (1%, v/v) in methanol (Merck) to quench endogenous peroxidase activity. This procedure was followed by 20-min incubation periods with, respectively, normal goat serum (1.5%, v/v; Vector Laboratories) in PBS buffer to block nonspecific immunoreactive sites, an avidin blocking kit (Vector Laboratories), and a biotin blocking kit (Vector Laboratories). The primary antibody was diluted in PBS buffer supplemented with 1.5% (v/v) normal goat serum. Sections were incubated overnight at 4°C. Working dilutions were 1:100, 1:200, 1:500, and 1:1000 for the polyclonal rabbit anti-caveolin 1 antibody (N20; Santa Cruz Biotechnology); 1:100, 1:500, and 1:1000 for the polyclonal rabbit anti-clathrin antibody (H-300; Santa Cruz Biotechnology); and 1:200 and 1:400 for the polyclonal rabbit anti-megalin antibody (kindly donated by M. Larsson, Uppsala University, Uppsala, Sweden). The next day, sections were incubated first for 30 min at room temperature with the biotinylated secondary antibody (goat anti-rabbit; Vector Laboratories) diluted 1:100 in PBS buffer and then with a preformed avidin and biotinylated horseradish peroxidase complex (Vectastain ABC kit; Vector Laboratories). Peroxidase activity was evidenced by 3,3'-diaminobenzidine (Biogenex) staining. No immunostaining could be observed when sections were incubated with normal rabbit serum instead of the primary antibody.
Detection of clathrin also was performed according to the same experimental procedure on placental explants incubated for 60 min in the Hepes-buffered medium supplemented with chorpromazine hydrochloride (1.4 mM).
Confocal Microscopy Examination of FITC-BSA Uptake into Syncytiotrophoblasts
Placental explants were preincubated during three 60-min periods in the Hepes-buffered medium in a shaking water bath thermostabilized at 37°C. Thereafter, they were incubated in the same medium supplemented with FITC-BSA (0.15 µg/µl) for 1, 2, 4, 6, 8, 10, or 30 min. The experiment was stopped by removing the tissue from the medium and washing it at 4°C with, first, an isotonic NaCl solution and, second, the Hepes-buffered medium supplemented with 0.5% BSA to remove extracellular fluorescence. Explants were frozen in methylbutane (99%) at –80°C. Ten-micrometer sections were cut from frozen blocks on a cryostat set at –18°C to –20°C, collected onto coverslips, and fixed with paraformaldehyde (4%) in PBS buffer for 4 min. The coverslips were mounted with an aqueous-based mounting medium (25%, w/v, polyvinyl alcohol solution with 0.14 M NaCl, 0.01 M KH2PO4/Na2HPO4, and 30%, v/v, glycerol) and then stored at –20°C until imaging [12]. Observations were performed under a Axiovert fluorescence microscope (Zeiss) coupled to a laser-scanning confocal microscope (MRC 1000; Bio-Rad) equipped with an argon-krypton laser and Laser-Sharp software (Bio-Rad).
Statistics
For all albumin uptake/outflow measurements, values (mean ± SEM) refer to experiments repeated with multiple cotyledons from a minimum of two placentas. The significance of differences observed between data means was assessed using an ANOVA followed by the adequate post-hoc test when necessary. Immunohistochemical and confocal microscopic examinations were performed in duplicate on at least two placentas.
RESULTS
Albumin Uptake into Placental Explants
Bovine serum albumin labeled either with 125I or with FITC was quickly internalized into the placental explants (Fig. 1). After a rapid initial phase, protein uptake increased regularly, indicating that the marker entered into the tissue at a constant rate. No sustained plateau was reached after a 60-min incubation period. Lowering the temperature from 37°C to 4°C significantly reduced the total amount of BSA trapped by placental explants (Fig. 1B).
|
The rapid internalization of FITC-BSA was confirmed by confocal microscopy (Fig. 2). After a 2-min incubation period, the probe was visualized in the syncytiotrophoblast, mainly in the apical submembrane zone facing the maternal side. After longer incubation periods (30 min), FITC-BSA was observed across the whole cytoplasmic compartment of the syncytiotrophoblast, but never in the villous stroma or in the fetal vessels.
|
Albumin Outflow from Placental Explants
Labeled BSA rapidly outflowed from preloaded placental explants (Fig. 3). After 30 min, 24% ± 2% of the total amount of marker initially taken up in the explants was recovered in the medium. Albumin outflow progressively slowed. Such findings (Fig. 3), together with the labeled BSA uptake data (Fig. 1), suggest that after a 30-min period, the whole process reaches a steady state.
|
Using gel filtration, the outflowing BSA was mainly intact after as much as 10 min of incubation (Fig. 4A). The supernatants collected after 15 and 30 min revealed 25% degradation (Fig. 4B). However, a more pronounced percentage (44% after 5 and 10 min, and 46% after 15 and 30 min) of labeled BSA remaining in the explants was found to be degraded (Fig. 4, C and D).
|
Modulation of Placental Albumin Uptake
After a 30-min incubation period, the uptake of labeled BSA into placental explants was significantly reduced (66% ± 9% vs. 100% ± 10% [control], P < 0.01) by methyl-ß-cyclodextrin (10 mM), a drug known to perturb both caveolae- and clathrin-dependent endocytosis (Fig. 5A). No significant effect was observed after a 5-min incubation time (data not shown). Labeled BSA uptake (5 or 30 min) was not modified by genistein (200 µM), which is a tyrosine kinase-inhibitor that blocks the caveolae endocytic system more specifically (Fig. 5B). By contrast, albumin uptake was impaired by chlorpromazine hydrochloride (1.4 mM), an inhibitor of the clathrin-dependent endocytic system (62% ± 7% vs. 100% ± 11% [control], P < 0.01) (Fig. 5C). An inhibitory effect was already observed after a 5-min incubation period (40.2% ± 2.1% vs. 100% ± 11% [control], P < 0.001, data not shown).
|
Confocal microscopy further indicated that in the presence of chlorpromazine hydrochloride (1.4 mM), FITC-BSA accumulated at the maternal side of the plasma membrane and faintly diffused across the cytoplasm. In the control tissue, after a similar 30-min incubation period, FITC-BSA stained the whole syncytiotrophoblast (Fig. 6).
|
Additional experiments aimed to clarify the influence of the cytoskeleton on the transmembrane movements of BSA. The presence in the medium of colchicine (1 mM), to induce microtubule disruption, or of cytochalasin B (40 µM), to provoke disassembly of filamentous actin, did not interfere with the placental uptake of albumin after 5 or 30 min (Fig. 5F).
Amiloride (500 µM), which is an agent that interferes with macropinocytosis, did not cause any change (at 5 or 30 min) of the BSA uptake into the placental explants (Fig. 5E).
Two chloride-channel inhibitors, DIDS (1 mM) and NPPB (100 µM), which reduce the megalin-mediated internalization of albumin, also failed to affect the labeled BSA uptake (Fig. 5D).
Increasing the extracellular BSA concentration from 0.1% to 0.5%, and then to 5%, did not modify the amount of labeled BSA accumulated in the syncytiotrophoblast (data not shown). Moreover, no change in the labeled BSA uptake was detected when the colloidal osmotic pressure was raised to a similar extent as obtained with BSA (5%) by adding dextran (4.5%, w/v) or polygelin (4%, w/v) in the medium (data not shown).
Modulation of Albumin Outflow from Placental Explants
The BSA outflow from preloaded and incubated placental explants was dose-dependent on the extracellular BSA concentration (+9% ± 4% for 1% BSA, not significant; +24% ± 4% for 2% BSA, P < 0.01; and +44% ± 3% for 5% BSA, P < 0.001) (Fig. 7A). However, addition of dextran (4.5%, w/v) or polygelin (4%, w/v) in the incubation medium to raise the colloidal osmotic pressure caused no change in the labeled BSA efflux (Fig. 7B). Similar results were obtained with preloaded and perifused placental explants (Fig. 7C). Indeed, in this experimental system, a rise in the extracellular BSA concentration from 0.5% to 5% provoked an immediate and sustained (25 min) stimulation of the [125I]BSA outflow, whereas exposure to dextran (4.5%) provoked only a 2-min, transient peak.
|
In addition, the impact of the cytoskeleton on BSA outflow from placental explants also was evaluated using colchicine (1 mM) and cytochalasin B (40 µM). Both agents had no effect on the labeled BSA outflow from incubated or perifused, preloaded placental explants (data not shown).
Caveolin 1, Clathrin, and Megalin Immunolocalization in Syncytiotrophoblasts
Immunohistochemical studies indicated that caveolin 1 was not located in the trophoblast layer but was present in the capillary endothelium (Fig. 8A). Clathrin, on the other hand, was specifically observed both at the plasma membrane facing the maternal side and in the cytoplasm of the syncytiotrophoblast (Fig. 8B). The fetal vascular endothelium was devoid of clathrin, and no immunostaining was detected in the surrounding muscular wall that was observed next to the villous tissue (Fig. 8B). No difference was noticeable in the immunostaining distribution when placental explants were treated with chlorpromazine hydrochloride (1.4 mM, data not shown). This experimental approach may be not sensitive enough for this purpose. The presence of megalin was located selectively at the plasma membrane (maternal side) of the syncytiotrophoblast, but not in the cytoplasm or the vascular endothelium (Fig. 8C).
|
DISCUSSION
Our data clearly indicated that BSA, labeled either with 125I or with FITC, was rapidly internalized into the placental explants. This process was temperature-sensitive. Albumin was first visualized by confocal microscopy at the apical submembrane zone of the syncytiotrophoblast. Thereafter, it progressively diffused through the whole cytoplasm, but it never reached the villous stroma or crossed the endothelium of fetal capillaries. The present findings further showed that within 30 min, a quarter of the labeled BSA captured by the explants outflowed from the trophoblast layer in the medium corresponding to the maternal compartment. This outflowing fraction was mainly intact albumin. By contrast, a part of the albumin left in the tissue was degraded (50%). This fraction of degraded albumin was not significantly different after a 5- or 10-min incubation period (the rapid initial phase of entry) or after a 15- or 30-min incubation period. All these results would indicate that maternal albumin is actively internalized in the placental explants and that, in the syncytiotrophoblast layer, the protein is either apically recycled into the maternal blood circulation or follows a cellular route, leading to its degradation.
As previously mentioned, at least three mechanisms have been reported for albumin endocytosis and might be involved in its placental uptake. The first is binding to the surface albumin-binding glycoprotein gp60 or albondin, which induces the internalization of albumin via the caveolae [6]. Such an endocytic system has been described in the vascular endothelium [6]. Using caveolin 1 as an indirect marker [8], we have detected caveolae in the fetal capillary endothelium of the placental tissue but not in the trophoblast, as previously reported [9]. Moreover, genistein, which is a tyrosine kinase inhibitor able to block caveolae internalization [8], did not affect labeled BSA uptake in placental explants. Thus, these results would exclude the association of this pathway with albumin transport into the syncytiotrophoblast. Methyl-ß-cyclodextrin disrupts cholesterol-rich caveolae [6, 8]. However, it significantly reduced the uptake of labeled BSA. The latter observation might result from the fact that the extraction of cholesterol with methyl-ß-cyclodextrin perturbs another endocytotic pathway, the clathrin-coated pits [25]. This endocytotic system is, indeed, a second pathway having been involved in albumin endocytosis, more precisely in its transport across the epithelium of the lactating mouse mammary gland [10]. By immunohistochemistry, we have detected clathrin in the trophoblast layer, at the plasma membrane, and in the cytoplasm. Our results also revealed that labeled BSA uptake into placental explants was sensitive to chlorpromazine hydrochloride, which is a cationic amphiphilic drug that inhibits the assembly of clathrin and protein AP2 at the coated pit [8, 10]. By confocal microscopy, FITC-BSA was visualized to accumulate at the maternal side of the plasma membrane in the presence of chlorpromazine, whereas FITC-BSA stained the whole cytoplasm in the control tissue. These different observations suggest that albumin uptake into the syncytiotrophoblast involves a clathrin-mediated pathway. This mechanism is associated with fluid-phase endocytosis. Such a mechanism is in agreement with the linear kinetics observed for the placental uptake.
In addition, our findings confirm the presence of megalin in the trophoblast layer, essentially at the plasma membrane side. This scavenger receptor plays a role in the renal tubular uptake of filtered plasma proteins, including albumin, and is associated with their degradation in the lysosomes [13, 14]. The megalin receptor also could participate in the placental albumin uptake, which would explain the presence of albumin fragments in placental explants. However, NPPB and DIDS, which are two chloride-channel inhibitors reported to block the megalin-mediated endocytosis of albumin [23], had no significant impact on the placental labeled BSA uptake. So, one may speculate that the megalin in the trophoblast layer would play only a minor role in the albumin uptake into the syncytiotrophoblast.
Besides these specific pathways, another fluid-phase endocytotic process, macropinocytosis, could participate in the massive albumin entry into the trophoblast layer. Amiloride, which is reported to highly reduce macropinocytosis [22, 26], did not cause any change in the placental uptake of labeled BSA. This lack of effect of amiloride ruled out the involvement of such a mechanism.
Actin and microtubule cytoskeleton play a key role in the endocytotic traffic in polarized epithelial cells. Microtubules and actin are, indeed, usually required for efficient transcytosis and delivery of proteins to late endosomes and lysosomes, whereas microtubules are important in apical recycling [27]. Colchicine, which induces microtubule disruption, or cytochalasin B, which provokes disassembly of filamentous actin, did not interfere with the placental uptake or outflow of labeled BSA. These findings indirectly suggest that an intact cytoskeleton is not required for transmembrane movements of albumin in the syncytiotrophoblast.
Taken together, the present data firmly suggest that maternal serum albumin is internalized into the human trophoblast layer and undergoes a temperature-sensitive and cytoskeleton-independent apical recycling. Albumin also appears to follow an intracellular route, leading to its degradation. The placental entry of the protein likely would involve mainly a clathrin-dependent endocytotic system. Some binding of albumin to megalin, immunolocalized at the trophoblast membrane, might explain why, at 4°C, the apparent placental capture of labeled albumin is reduced by 52% and not completely impaired. Indeed, cellular internalization, but not binding to a membrane receptor, can be blocked at that temperature.
A recycling process has already been described for immunoglobulin G in trophoblast-derived BeWo cells [28]. Immunoglobulin G is transferred across the human placenta during pregnancy, interacting with the specific receptor hFcRn. The trophoblast-derived BeWo cells exhibited both an apical-to-basolateral transcytosis and a temperature-sensitive apical immunoglobulin G recycling, which was not modified by agents perturbing the cytoskeleton. The transmembrane movements of albumin in the syncytiotrophoblast might be related to the latter pathway.
The significance of this maternal albumin recycling in the trophoblast layer remains to be elucidated, but the process could have physiological relevance. Indeed, raising the extracellular concentration of BSA in a physiological range did not cause any change in the uptake of labeled BSA into placental explants. Such a feature suggests that the internalization rate of labeled BSA (i.e., the rate of the internalization process) was not modified, even though larger amounts of albumin went into the tissue. This observation also confirms that the uptake of albumin would implicate mainly a fluid-phase internalization process than a receptor-mediated endocytosis. Labeled BSA outflow was increased under the same experimental conditions. One could imagine that such a high-capacity internalization for albumin into the syncytiotrophoblast could contribute to the maternal nutrition of the fetus by carrying essential fatty acids and/or other ligands. Extrapolation of our in vitro results to the in vivo situation suggests that in 24 h, 1.3-fold the total maternal plasma is recycled in the placenta. With the maternal plasma albumin concentration being equal to 3% during pregnancy, approximately 150 g of albumin would enter into the trophoblast layer per day.
Moreover, would these movements of albumin be linked to the placental hormone secretion? As previously described in vitro, increasing extracellular concentrations of BSA from 0.5% up to 5% provokes an immediate, dose-dependent stimulation of the hPL and hCG releases from placental explants [4]. This secretory response is partly reproduced by other colloidal agents, such as dextran and polygelin [5]. However, by contrast with albumin, the two latter molecules did not significantly modify the amount of albumin internalized or the outflow of labeled BSA from placental explants. Moreover, and as mentioned above, colchicine and cytochalasin B did not affect the labeled BSA recycling while inhibiting or potentiating, respectively, the placental secretory response to BSA [4]. Therefore, the in vitro albumin stimulation of hPL and hCG releases from placental explants would not be related to the transmembrane movements of albumin in the syncytiotrophoblast.
In conclusion, our data suggest, to our knowledge for the first time, that large amounts of the maternal serum albumin are internalized into the fetal trophoblast layer according to an apical recycling pathway, which is temperature-sensitive, and mainly related to a clathrin-mediated endocytotic system.
ACKNOWLEDGMENTS
We gratefully thank the nursing staff from the Erasme Hospital (Brussels, Belgium) for their help in collecting placentas. The polyclonal rabbit anti-megalin antibody was kindly donated by M. Larsson, Uppsala University.
FOOTNOTES
1 P.L. is Research Director of the National Fund for Scientific Research (F.N.R.S., Belgium). N.L. was supported by the "Lekime-Ropsy" and the "Suzanne Maraîte" Fundations (Belgium). 
2 Correspondence: Nathalie Lambot, Laboratory of Experimental Hormonology, Université Libre de Bruxelles–Campus Erasme, CP 626, 808 Route de Lennik, B-1070 Brussels, Belgium. FAX: 32 02 555 6356; nlambot{at}ulb.ac.be
Received: 5 December 2005.
First decision: 30 December 2005.
Accepted: 22 February 2006.
REFERENCES
2-macroglobulin and albumin by human term placental syncytiotrophoblast in vitro. J Cell Biochem 1998 68:427-435[CrossRef][Medline]This article has been cited by other articles:
|
|
 |
|
  C. H. van den Akker, H. Schierbeek, T. Rietveld, A. Vermes, J. J Duvekot, E. A. Steegers, and J. B van Goudoever Human fetal albumin synthesis rates during different periods of gestation Am. J. Clinical Nutrition, October 1, 2008; 88(4): 997 - 1003. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. J. Mizejewski Physiology of Alpha-Fetoprotein as a Biomarker for Perinatal Distress: Relevance to Adverse Pregnancy Outcome Experimental Biology and Medicine, September 1, 2007; 232(8): 993 - 1004. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. D Vandre, W. E Ackerman IV, D. A Kniss, A. K Tewari, M. Mori, T. Takizawa, and J. M Robinson Dysferlin Is Expressed in Human Placenta But Does Not Associate with Caveolin Biol Reprod, September 1, 2007; 77(3): 533 - 542. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
