Binding of human isotransferrin variants to microvillous and basal membrane vesicles from human term placenta

Transferrin Sialylation in Smoking and Non-Smoking Pregnant Women with Intrauterine Growth Restriction

Transferrin (Tf) is a glycosylated protein responsible for transporting iron. Various sialylation levels of Tf are observed during physiological and pathological processes. We studied if the changes in iron stores as well as tobacco smoke may have an impact on foetal development and in consequence lead to intrauterine growth restriction (IUGR). In the third trimester of pregnancy, lower levels of 4-sialoTf isoform and higher levels of 5-sialoTf were observed in the serum of non-smoking women with IUGR in comparison to the control group. On the day of labour, level of 2-sialoTf was significantly lower and level of 3-sialo was Tf higher in the serum of non-smoking women. Level of 4-sialo was found lower in the serum of smoking women with IUGR than in the control group. The observed changes may suggest a connection between iron stores, transport of iron to the foetus and foetal development.

Localisation of proteins of iron metabolism in the human placenta and liver

Two anatomical sites that are important in human iron metabolism are the liver and placenta. Liver macrophages recycle iron from erythrocytes, and the placenta transfers iron from the mother to the fetus. The cellular distribution of proteins involved in iron transport in these two sites was studied. Transferrin receptor-1 (TfR1) and Ferroportin (FPN) expression was found on the placental syncytiotrophoblast (STB) and were polarised such that TfR1 was on the apical maternal-facing membrane and FPN was on the basal fetal-facing membrane, consistent with unidirectional iron transport from mother to fetus. Ferritin was strongly expressed in the stroma, suggesting that fetal tissue can store and accumulate iron. HFE was on some parts of the basal STB and, where present, HFE clearly colocalised with FPN but not TfR1. In the stroma, both HFE and FPN were present on CD68+ Hofbauer macrophage cells. In liver, the location of HFE is controversial. Using four mouse monoclonals and two polyclonal sera we showed that the pattern of HFE expression mirrored the distribution of CD68+ macrophage Kupffer cells. FPN was also most strongly expressed by CD68+ Kupffer cells. These findings contribute to understanding how iron is transported and stored in the human placenta and liver.

Effects of chronic hypoxia in vivo on the expression of human placental glucose transporters

Birth weight is reduced and the risk of preeclampsia is increased in human high altitude pregnancies. There has been little work to determine whether hypoxia acts directly to reduce fetal growth (e.g. reduced blood flow and oxygen delivery), or via changes in functional capacities such as nutrient transport. We therefore investigated the expression of a primary nutrient transporter, the GLUT1 glucose transporter and two in vitro markers of hypoxia (erythropoietin receptor, EPO-R, and transferrin receptor, TfR) in the syncytial microvillous (MVM) and basal membrane fractions (BMF) of 13 high (3100 m) and 12 low (1600 m) altitude placentas from normal term pregnancies. Birth weight was lower at 3100 m than at 1600 m despite similar gestational age, but none of the infants were clinically designated as fetal growth restriction. EPO-R, TfR and GLUT1 were examined by immunoblotting and maternal circulating erythropoietin and transferrin by ELISA. EPO-R was greater on the MVM (+75%) and BMF (+25%) at 3100 m. TfR was 32% lower on the MVM at 3100 m. GLUT1 was 40% lower in the BMF at 3100 m. Circulating EPO was greater at high altitude, while transferrin was similar, and neither correlated with their membrane receptors. BMF GLUT1 was positively correlated with birth weight at high, but not low altitude. In this in vivo model of chronic placental hypoxia, syncytial EPO-R increased as expected, while nutrient transporters decreased, opposite to what has been observed in vitro. Therefore, hypoxia acts to reduce fetal growth not simply by reducing oxygen delivery, but also by decreasing the density of nutrient transporters.

Endocytic and Transcytotic Processes in Villous Syncytiotrophoblast: Role in Nutrient Transport to the Human Fetus

The supply of nutrients to the developing fetus is a major function of the human hemochorial placenta, a placenta type in which the fetal chorion is in direct contact with the maternal blood. At term, nutrients have to be transported across two cell layers in chorionic villi, the syncytiotrophoblast (STB) and fetal endothelial cells. The STB is a continuous syncytium covering the entire surface of chorionic villi. This polarized epithelium is specialized in exchange processes and membrane trafficking between the apical membrane facing the maternal blood and the basal membrane facing the fetal endothelium. To meet placental and fetal requirements, the STB selectively takes up and transports a variety of nutrients, hormones, growth factors and cytokines and also transfers passive immunity to the fetus by receptor-mediated transcytosis. In this review in vivo and in vitro systems currently used to study STB functions are discussed and the potential mechanisms of transplacental IgG, iron, lipoprotein and glucose transport are presented. As revealed in this article, the placenta is a tissue where intensive cell biological research is required to unravel endocytic trafficking pathways in a highly specialized cell such as the STB.

N-glycans of human amniotic fluid transferrin stimulate progesterone production in human first trimester trophoblast cells in vitro

AIMS

During pregnancy, the placenta produces a variety of steroid hormones and proteins. Several of these substances have been shown to exert immunomodulatory effects. Progesterone is thought to mediate some of these effects by regulating uterine responsiveness. The aim of this study was to clarify the effect of amniotic fluid transferrin and its N-glycans on the release of progesterone by first trimester trophoblast cells in vitro.

METHODS

Cytotrophoblast cells were prepared from human first trimester placentae by trypsin-DNAse dispersion of villous tissue followed by a percoll gradient centrifugation and depletion of CD45 positive cells by magnetic cell sorting. Trophoblasts were incubated with varying concentrations (50-300 microg/ml) of transferrin from human amniotic fluid and serum as well as with N-glycans obtained from amniotic fluid transferrin. Culture supernatants were assayed for progesterone by enzyme-immunometric methods.

RESULTS

The release of progesterone increased in amniotic fluid transferrin- and N-glycan-treated trophoblast cell cultures compared to untreated trophoblast cells. There was no stimulating effect of serum transferrin on the progesterone production of trophoblast cells.

CONCLUSIONS

The results suggest that amnion-transferrin and especially its N-glycans modulate the endocrine function of trophoblasts in culture by up regulating progesterone secretion.

Human amniotic fluid glycoproteins expressing sialyl Lewis carbohydrate antigens stimulate progesterone production in human trophoblasts in vitro

BACKGROUND

Progesterone is thought to mediate immune modulator effects by regulating uterine responsiveness. The aim of the study was to clarify the effect of transferrin and glycodelin A (former name PP14) as sialyl Lewis X-expressing glycoproteins on the release of progesterone by trophoblast cells in vitro.

METHODS

Cytotrophoblast cells were prepared from human term placentas by standard dispersion of villous tissue followed by a Percoll gradient centrifugation step. Trophoblasts were incubated with varying concentrations (50-300 microg/ml) of human amniotic fluid- and serum-transferrin as well as with glycodelin A. Culture supernatants were assayed for progesterone, human chorionic gonadotropin (hCG) and cortisol by enzyme immunometric methods.

RESULTS

The release of progesterone is increased in amniotic fluid transferrin- and glycodelin A-treated trophoblast cell cultures compared to untreated trophoblast cells. There is no relation between transferrin and the hCG or cortisol production of trophoblast cells.

CONCLUSION

The results suggest that sialyl Lewis carbohydrate antigen-expressing amniotic fluid glycoproteins modulate the endocrine function of trophoblasts in culture by upregulating progesterone production.

Glycodelin and amniotic fluid transferrin as inhibitors of E-selectin-mediated cell adhesion

Human amniotic fluid contains a variety of glycoproteins. Several of these substances have been shown to exert immunomodulatory effects. Glycodelin, previously known as placental protein 14, is one of these glycoproteins. It has a unique carbohydrate configuration, consistent with fucosylated LacdiNAc structures that are very unusual for mammals. Oligosaccharides with fucosylated LacdiNAc antennae have previously been shown to block selectin-mediated cell adhesion. Another glycoprotein, human transferrin, is also present in amniotic fluid in relatively high concentrations. This transferrin shows a different glycosylation compared with serum transferrin. Amniotic fluid transferrin carries sialylated Lewis X antigens. Glycodelin and transferrin were isolated from amniotic fluid and for comparison from serum of pregnant women by chromatographic methods. The purified proteins were used as ligands to block E-selectin-mediated HepG2 cell adhesion. Two types of binding assays with distinct receptor accommodations (immobilised E-selectin and activated HUVECs) were used to quantify inhibition efficiencies of the different proteins. We found that glycodelin is a strong inhibitor with a 10(3)-fold potency compared to the monovalent tetrasaccharide sialyl Lewis X whereas the potency of transferrin is rather low.

Identification and localization of divalent metal transporter-1 (DMT-1) in term human placenta

The mechanism by which iron is transported from mother to fetus is incompletely understood. Whereas transferrin receptor (TfR) is responsible for iron uptake from maternal serum by the syncytiotrophoblast, the proteins responsible for intracytoplasmic transport and for delivery to the fetal serum remain unknown. The aim of this study was to determine whether the recently characterized endosomal membrane iron transporter, divalent metal ion transporter-1 (DMT-1), is expressed in human syncytiotrophoblast, and whether its cellular localization would support roles for cytoplasmic and placental-fetal iron transport. Six micron sections of frozen, term human placenta were assessed immunohistochemically using a polyclonal antibody to rat DMT-1 and a monoclonal antibody to human TfR. DMT-1 was found both in the cytoplasm and at the junction of the fetal (basal) membrane and fetal vessels, while TfR was localized predominantly to the maternal (apical) side of the syncytiotrophoblastic membrane. Double staining demonstrated no overlap between the two proteins on the apical membrane and minimal areas of overlap in the cytoplasm. We postulate that the syncytiotrophoblast takes up diferric transferrin from serum via TfR, subsequently incorporating the transferrin : TfR complex via endosomes. Subsequent transport of iron out of the endosome and across the basal membrane to the fetus may occur via DMT-1.

Accumulation and release of iron in polarly and non-polarly cultured trophoblast cells isolated from human term placentas

We investigated the usefulness of membrane grown human term trophoblast cells in transferrin-mediated iron transfer studies. We showed that diferric transferrin is taken up both at the microvillous and at the basal membrane by means of receptor-mediated endocytosis. Uptake from the microvillous side is predominant. This corresponded with a much higher expression of transferrin receptors at the microvillous membrane as compared to the basal one. Iron appeared to accumulate in the cell. Accumulation was higher when transferrin was supplied at the microvillous side. Transfer of iron could not be assessed because uptake of transferrin by the cells was much less than passive diffusion of transferrin through the cell-free filter. The observation of iron accumulation was unexpected for a transfer epithelium. Could it be that part of the iron taken up by the cells is rapidly released whereas the remaining part accumulates? In this case the rate of iron uptake should be higher than the rate of iron accumulation. This question was assessed with non-polarly cultured trophoblast cells. We showed that like in polar cells iron accumulated in ferritin. A new experimental design enabled us to demonstrate that indeed the rate of transferrin-mediated iron is in excess over iron accumulation. We thus provide evidence for a mechanism that enables rapid transfer of iron across the syncytiotrophoblast cell layer.

Non-transferrin iron uptake by trophoblast cells in culture. Significance of a NADH-dependent ferrireductase

This study shows that trophoblast cells in culture are able to take up 59Fe from both Fe(III)nitrilotriacetate (NTA) and Fe-ascorbate. Fe in the presence of ascorbate is assumed to be Fe(III) in equilibrium with Fe(II). Kinetic parameters for non-transferrin iron uptake are determined from initial rate experiments, yielding Vmax=366 pmol/mg protein/5 min and Km=0.96 microM for Fe(III)NTA and Vmax=4043 pmol/mg protein/5 min and Km= 1.3 microM for Fe-ascorbate. Since trophoblast cells in culture reduce extracellular Fe(III)CN, and uptake of 59Fe from Fe-ascorbate is higher than that from Fe(III)NTA, it is suggested that reduction of Fe(III) precedes uptake. Uptake of 59Fe from both Fe-ascorbate and Fe(III)NTA is inhibited by Fe(II)chelator ferrozine and membrane-impermeable Fe(III)CN, further supporting this hypothesis. Studies with microvillous membrane vesicles (MMV) and basal membrane vesicles (BMV) reveal the presence of a NADH-dependent ferrireductase. Reduction of Fe(III)CN follows Michaelis-Menten kinetics, both with respect to [NADH] and [Fe]. NADPH is ineffective as electron donor. The rate of Fe(III)CN reduction by BMV is 2.5 times higher compared to MMV, while Km values for Fe(III)CN and NADH are not significantly different. These results reveal that a transmembrane NADH-dependent ferrireductase plays a role in uptake of non-transferrin iron. The possibility that this enzyme system is involved in iron transfer across the basal membrane is discussed.