E. Alsat

Differential expression of the TEF family of transcription factors in the murine placenta and during differentiation of primary human trophoblasts in vitro

We describe the molecular cloning of murine (m) Transcriptional Enhancer Factor (TEF)‐5 belonging to the TEF family of transcription factors. We show that mTEF‐5 is specifically expressed in trophoblast giant cells and other extra‐embryonic structures at early stages of development. At later stages, mTEF‐5 is specifically expressed in the labyrinthine region of the placenta and in several embryonic tissues. We further show that the other mTEFs are differentially expressed in extraembryonic structures and in the mature placenta. Interestingly, human (h)TEF‐5 is specifically expressed in the differentiated syncytiotrophoblast of the human term placenta and its expression is upregulated during the differentiation of cytotrophoblasts to syncytiotrophoblast in vitro, whereas that of hTEF‐1 is down‐regulated. Together with previous results describing hTEF‐binding sites in the human placental lactogen‐B gene enhancer, these novel observations support a role for hTEF‐5 in the regulation of this gene. We further propose that the hTEF factors may play a more general role in placental gene regulation and development. Dev. Dyn. 1998; 212:423–436.

Ultrastructural visualization of the internalization of low density lipoprotein by human placental cells

SummaryLow density lipoproteins (LDL) were conjugated to colloidal gold to visualize the route for internalization of LDL in the cultured cells of human term placenta. Cells were obtained from placental villi (caesarian section) by a standard trypsin-DNase dispersion method followed in some cases by a Percoll gradient centrifugation step. Employing electron microscopy it was observed that after 3 days of culture, cells obtained by trypsin-DNAse dispersion were a mixture of macrophages, mononucleated cells and large multinucleated cells. When the cells were incubated for 3 days after the Percoll purification, essentially multinucleated cells identical to the syncytiotrophoblast were present. The number of LDL receptor was increased by preincubation in medium with lipoprotein depleted serum. In binding experiments cells incubated at 4° C for 2 h with medium containing gold LDL conjugates showed gold LDL attached to the plasma membrane without characteristic localization. After incubation with gold LDL at 37° C for various times, the three cellular types showed ligand internalization. Gold LDL endocytosis involved first coated pits but also uncoated plasmalemmal invaginations. Then gold LDL was further observed in coated and non coated vesicles, smooth walled endosomes, multivesicular bodies and tubular vesicles. Lastly free gold particles were observed in lysosome like dense bodies. These results prove the internalization of gold LDL conjugates by human cultured placental cells, particularly by syncytiotrophoblast like multinucleated cells. This accumulation of LDL (the major cholesterol carrying protein in humans) is recognised to be responsable for the exogenous cholesterol supply indispensable to the progesterone biosynthesis and cellular growth of the placenta.

Low-density lipoprotein binding sites in the microvillous membranes of human placenta at different stages of gestation

Microvillous membranes isolated from early gestation placentas (8-12 weeks of amenorrhoea) and from mid-term placentas (20-22 weeks of amenorrhoea) were used to study the specific binding of low-density lipoprotein (LDL) to the trophoblast. The purity of the microvillous preparations has been assessed by electron microscopy and by their enrichment in two membrane markers, 5-nucleotidase and alkaline phosphatase. Evidence was presented demonstrating the existence of saturable binding sites for [125I]LDL in placental microvilli as early as 6 weeks of pregnancy. The apparent KD values for these binding sites have been determined by Scatchard analyses to be 6.98 +/- 0.83 and 6.57 +/- 0.81 micrograms protein LDL/ml, for early gestation and mid-term preparations, respectively. This apparent KD value was unaffected by a pretreatment of the membranes by heparin, as indicated by the mean values of 7.13 +/- 0.89 and 6.97 +/- 0.75 micrograms protein LDL/ml obtained for immature microvilli preincubated with or without heparin, respectively. Large variations of binding capacity were observed in each gestational age group and no significant difference was found between them. These results indicate that the LDL binding sites of the human placenta, located on the microvillous membranes, (i) are present as early as the 6th week of pregnancy, and (ii) display the same high affinity and specificity for LDL as those of the term trophoblast.

The stimulatory action of the prostaglandins on the production of oestrogens by the human placenta perfused in vitro

n The effect of prostaglandins (PGs) on estrogen production by the human placenta perfused in vitro is investigated in this study. 5 placentas obtained by natural delivery at term were perfused within the 1st 30 minutes following delivery. 10 mg of testosterone was added at the beginning of the experiment and varying amounts of PGF2alpha, PGE1, and PGE2 at the end of the 1st hour of perfusion (maintained for 2 to 2-1/2 hours). Fluorimetric measurements of estrone and estradiol were done following chromatography on a celite column of 50 ml samples of perfusate taken every 30 minutes. Placentas perfused in a similar manner but receiving 10 mg testosterone only were used as controls. The results show that addition of 1 mg of PGE2 (3 x 10 -6 M) or PGF2alpha significantly increased the level of estrogens in the perfusion fluid compared with the control placentas. It is possible that this stimulatory effect facilitates uterine contractions during parturition. The increase appeared to be of long duration and to have a dose-response relationship. However, the perfusion of the whole organ appears not to be a very good model for studying the quantitative effect of a stimulator or the relationship between different doses used and responses obtained because of the variation of 1 placenta to another. To investigate the possibility that PGs are possible mediators between human chorionic gonadotropins or luteinizing hormones and adenyl-cyclase, the authors conducted experiments where PGE1 and PGF2alpha were added to minced placentas together with DB cyclic AMP or NADPH generating factors. Only in the 2nd case was an additive effect observed, suggesting that this effect on estrogen production is brought about by the stimulation of adenyl-cyclase observed in placental homogenates by investigators Satoh and Ryan.n

Increase in expression and activity of thrombomodulin in term human syncytiotrophoblast microvilli.

A comparative study of thrombomodulin (TM), a potent natural anticoagulant, was performed in first trimester and term human placentae. Immunoreactive TM was observed on fetal vascular endothelium and syncytiotrophoblast at both gestational ages. Staining was stronger in term than in early placentae, particularly along the microvillous apical membrane of the syncytiotrophoblast. Similarly, a higher level of TM mRNA was detected by RT-PCR (P<0.02) and Northern blot analysis in extracts of whole term placentae. The localization of TM on syncytial microvilli was confirmed by electron microscopy after immunogold labelling. When isolated microvilli were compared at both gestational ages; a significant 2.3-fold increase in TM protein was observed in term microvilli as compared to first trimester microvilli by Western blot analysis (P<0.005) and ELISA (P<0.05). This higher level of TM in term microvilli was associated with an increase in its ability to activate protein C, from 3.7 +/- 1.2 to 8.7 +/- 4.2 mOD/min/microg protein +/- s.d. (P<0.01) in first trimester and term microvilli, respectively. The modulation of biologically active TM at the syncytial membrane exposed to maternal blood according to the length of gestation suggests that TM may be involved both in maternal haemostasis within the intervillous spaces, and also in the trophoblast differentiation process.

Characterization of specific low-density lipoprotein binding sites in human term placental microvillous membranes

Purified microvillous membranes prepared from normal term human placenta were studied for their ability to bind specifically low-density lipoprotein (LDL). Electron microscopic examination of the membrane preparations revealed essentially microvilli-like structures, and the enzyme analyses a 14-17-fold enrichment in the membrane markers 5-nucleotidase and alkaline phosphatase. The binding of [125I]LDL was dependent on time, temperature, pH and protein concentration; it was saturable with a low capacity (130.4 +/- 22.2 ng/mg of membrane protein) and presented a high affinity (apparent Ka 6.12 +/- 1.32 micrograms protein per ml). These high-affinity binding sites were specific for LDL (high-density lipoprotein induced less competition than unlabelled LDL) and were sensitive to pronase digestion. Unlike the binding of LDL to other tissues, the [125I]LDL binding to microvillous membranes did not require divalent cations. The presence of specific LDL receptors on the placental microvillous membranes, located at the effective site of exchange between the maternal blood and the placental tissue, supports the concept that human placenta utilizes LDL-cholesterol for its progesterone synthesis.

Regulation of steroidogenesis in the human placenta.

Abstract Large quantities of progesterone and ocstrogens are synthesized during human pregnancy and the principal steps of biosynthesis occurring in the placenta are quite well-known. The progesterone synthesis is difficult to study because it derives from the important endogenous pool of free and esterified cholesterol, and the only enzymatic system studied is the 3β HSDH 4ene → 5ene isomerase. The placenta acquires during the course of pregnancy a gradually increasing power of aromatization. LH. HCG and some PG stimulate the oestrogen biosynthesis, this effect being probably mediated by cAMP. Otherwise some enzymes such as 5-ene-3β sulphatase and 5-ene-3β HSDH are directly inhibited by some endogenous steroids. Significant differences have been observed in the characteristics of both enzymatic activities according to the use of C 19 or C 21 substrates, with higher affinities ( K M ) tor the latter ones. Moreover some progestative agents, such as chlormadinone acetate, induce a decrease of the oestrogen biosynthesis by a 5-ene-3β sulphatase deficiency which could spontaneously occur in pregnancies with male fetus. This X-linked recessive character has no consequence on the fetal development, but is frequently associated with an ichthyosis.

Identification of specific binding sites for acetylated low density lipoprotein in microvillous membranes from human placenta

The ability of microvillous membranes isolated from human placenta to specifically bind human low density lipoprotein (LDL) modified by acetic anhydride has been investigated. The presence of saturable high affinity binding sites specific for [125I]acetyl-LDL was demonstrated. Scatchard analysis of the binding data, obtained at 4 degrees C, revealed a single class of sites with a mean KD value of 3.63 +/- 1.16 micrograms acetyl-LDL protein/ml, and a maximal binding capacity of 335.1 +/- 148.8 ng acetyl-LDL protein/mg of membrane protein. At 37 degrees C, the binding capacity was increased, while the KD value was not modified. The specificity of these binding sites was assessed by competition studies: unlabelled acetyl-LDL were effective competitors, whereas native LDL, VLDL and HDL3 were ineffective. Conversely, unlabelled acetyl-LDL failed to prevent the binding of native [125I]LDL to placental microvilli. The [125I]acetyl-LDL binding was partially inhibited (about 35%) by dextran sulfate and fucoidin, and was abolished by a pretreatment of the microvillous membranes with pronase. The binding sites specific for acetyl-LDL are present during all the gestation and are distinctly different from the binding sites for native LDL, previously characterized in placental microvilli. These 2 types of binding sites may be related to the high amount of cholesterol required by the human placenta for progesterone synthesis and trophoblastic growth.

Placental sulfatase deficiency: Clinical and biochemical study of 16 cases

Clinical and biochemical data of 16 typical cases of placental sulfatase deficiency have been observed. In vivo loading tests with DHA-S allowed us to make a prenatal diagnosis. In vitro experiments gave confirmation, showing zero or virtually zero placental sulfatase activity towards delta 5P or DHA sulfates Aromatase activities, when tested, were normal or more often less than standard values, the latter showing themselves rather large individual variations. All pregnancies were associated with the delivery of male neonates in good health but 3. The 15 living babies have been developing normally since then. These results, together with those reported in the literature, suggest that placental sulfatase deficiency is under control of an X-linked recessive character, this being supported by the recent observation of such a disorder in two sisters simultaneously pregnant. As to the high frequency problem of cesarian section, pointed out by several authors, we cannot conclude, from our own observations, that the defect has an obvious influence on the good outcome of labor, as 10 out of the 16 women delivered vaginally near term.

In vitro Aromatization of Androgens into Estrogens in Placental Insufficiency

We perfused in vitro the placentas of the mothers of 3 small for gestational age (SGA) and 3 adequate for gestational age (AGA) infants. After a systemic injection of 10 mg of dehydroepiandrosterone sulfate (DHEA-S), we measured the time course of the levels of DHEA-S, unconjugated DHEA, delta 4-androstenedione (delta 4-A), testosterone (T), estrone (E1) and estradiol (E2) in perfusate. We observed a significant accumulation of delta 4-A + T and a concomitant low increase of E1 + E2 in the placental fluid of SGA infants. We conclude that the aromatase activity could be decreased in some cases of placental insufficiency. This could, at least in part, explain the accumulation of circulating androgens in the mothers of some SGA infants following DHEA-S loading test observed in vivo.