An immunohistological comparison of the secretory capacity of villous and extravillous trophoblast in the human placenta

The isolation and characterization of a novel telomerase immortalized first trimester trophoblast cell line, Swan 71

Studies using first trimester trophoblast cells may be limited by the inability to obtain patient samples and/or adequate cell numbers. First trimester trophoblast cell lines have been generated by SV40 transformation or similar methods, however, this approach is known to induce phenotypic and karyotypic abnormalities. The introduction of telomerase has been proposed to be a viable alternative for the immortalization of primary human cells. To investigate whether telomerase-induced immortalization might be a more feasible approach for the generation of first trimester trophoblast cell lines, we isolated primary trophoblast cells from a 7-week normal placenta and infected the cells with human telomerase reverse transcriptase (hTERT), the catalytic subunit of telomerase. Although this hTERT-infected first trimester trophoblast cell line, which we have named Swan 71, has been propagated for more than 100 passages, it still has attributes that are characteristic of primary first trimester trophoblast cells. The Swan 71 cells are positive for the expression of cytokeratin 7, vimentin and HLA-G, but do not express CD45, CD68 or the Fibroblast Specific Antigen (FSA), CD90/Thy-1. In addition, we also demonstrated that the Swan 71 cells secrete fetal fibronectin (FFN) as well as low levels of human Chorionic Gonadotrophin (hCG). Moreover, the Swan 71 cells exhibit a cytokine and growth factor profile that is similar to primary trophoblast cells and are resistant to Fas, but not TNF-alpha-induced apoptosis. This suggests that the Swan 71 cells may represent a valuable model for future in vitro trophoblast studies.

Trophoblast invasion

The establishment of a pregnancy depends upon mechanisms of forming close contacts between embryonic/fetal and maternal tissues. The starting point is the attachment of the blastocyst to the uterine wall, and this critical step is followed by a series of different morphogenetic events leading to placentation. These processes depend on the invasive properties of extra-embryonic trophectoderm-derived cells that show their highest expression in species with haemochorial placentation, i.e. Insectivores, rodents, and primates including the human. In this review we will concentrate upon the human with occasional reference to other species.

Early growth response-1 mediates up-regulation of telomerase in placenta

Telomerase is thought to play a very important role in oncogenesis. It is also believed to wind back the "mitotic clock" which leads to ageing and enable permanent cell division. We evaluated telomerase activity in chorionic tissues, with particular attention to the early growth response-1 (EGR-1) gene, the importance of what was recently shown by Khachigian et al. We started our study by evaluating the relationship between activation of transcription of the human telomerase reverse transcriptase (hTERT) gene and EGR-1 gene. For this purpose, we first evaluated telomerase activity using the villous cancer cell lines JAR and JEG-3. We then demonstrated that EGR-1 plays an important role in activation of the transcription of hTERT by luciferase assay using hTERT promoter constructs. As a result of further computer analysis, we discovered a site postulated to be an EGR-1 consensus binding site at -273 to -281 in the hTERT promoter region. With forced expression of EGR-1, an increase in hTERT protein concentration was detected on Western blot analysis, while marked high expression of hTERT mRNA was observed by reverse transcriptase polymerase chain reaction. Furthermore, we evaluated the expression of EGR-1 and hTERT at the mRNA level in the placenta during the first, second and third trimesters of pregnancy and in patients with preeclampsia. Expression of EGR-1 and hTERT in the chorion increased in the first trimester of pregnancy and decreased later. Increased expression was noted in the placenta of patients with preeclampsia. The present findings suggest that EGR-1 plays an important role in activating the transcription of hTERT, showing that activation of the transcription of hTERT by EGR-1 is involved in the trophoblast growth mechanism.

Human chorionic gonadotropin expression in human trophoblasts from early placenta: comparative study between villous and extravillous trophoblastic cells

Human trophoblast differentiates into two pathways: extravillous cytotrophoblasts (EVCT) that invade the uterus wall and villous cytotrophoblasts (VCT) that fuse to form the syncytiotrophoblast (ST) involved in placental exchanges and endocrine function. It is established that hCG is produced and secreted by the ST into the maternal compartment where it plays a key endocrine role and stimulates ST formation in an autocrine manner. Herein, we investigated hCG expression in early placentas by immunohistochemistry using different antibodies. We then compared hCG secretion by primary cultures of VCT and EVCT isolated from the same first trimester human chorionic villi. In situ hCG was immunodetected in EVCT all along their invasive differentiating pathway except in cells near the stromal core of the proximal column. hCG expression was confirmed in vitro by immunocytochemistry and hCG secretion quantified in cell supernatants. Interestingly, whereas hCG secretion increased during VCT differentiation into ST (from 60 to 350UI/L/microg DNA), EVCT secretion remained constant and at a high level during the same culture period (160UI/L/microg DNA). Our data demonstrated that in addition to the ST, invasive EVCT also expressed and secreted high levels of hCG, suggesting a specific paracrine and/or autocrine role for hCG from EVCT origin.

Applications of monoclonal anti-human inhibin alpha subunit in endometrial curettings

AIMS

Using archival material, we studied the immunoreactivity and utility of monoclonal anti-human inhibin alpha subunit in the identification of chorionic villi (CV) and trophoblastic subpopulations in endometrial curettings (EC) from patients who had intra-uterine, ectopic, molar and, particularly, probable intra-uterine pregnancies. We also compared its expression with those of betaHCG, HPL and CAM 5.2.

METHODS

The four groups of EC investigated included: Group 1, 15 patients with intra-uterine pregnancies (IUP); Group 2, 15 patients with tubal pregnancies (TP); Group 3, 15 patients with hydatidiform moles (HM); and Group 4, 20 patients with purported history of intra-uterine pregnancies (PIUP). Positive and negative control cases were from Groups 1 and 3 and Group 2, respectively. The test cases were from Group 4. Immunohistochemistry was performed on each case testing for expression of inhibin alpha, betaHCG, HPL and CAM 5.2.

RESULTS

Trophoblastic populations, which included syncytiotrophoblast (ST), cytotrophoblast (CT) and intermediate trophoblast (IT), were absent in all 15 negative control cases (Group 2). The 30 positive control cases (Groups 1 and 3) revealed the following: (a) ST, CT and IT were identified in all cases and were positive for CAM 5.2, (b) inhibin alpha, betaHCG and HPL (except one case) were reactive for all cases with ST, but not CT, and (c) IT positivity for betaHCG, HPL and inhibin alpha was 67, 80-93 and 100%, respectively. From the 20 test cases (Group 4), the findings were: (a) CT was absent in all cases, (b) scattered ST cells, which were identified only in 10 cases, were positive for all antibodies, (c) scattered IT cells were present in 17 cases and showed 100% CAM 5.2 positivity, and (d) IT positivity for betaHCG, inhibin alpha and HPL was 58.8% (10/17), 76.5% (13/17) and 82.4% (14/17), respectively. Background staining was observed in 22 of 65 cases (33.8%) stained with betaHCG and HPL; half of these cases came from Group 3. Inhibin alpha and CAM 5.2 staining did not show this problem.

CONCLUSIONS

We suggest that inhibin alpha is a useful antibody in diagnosing IUP and HM and in documenting intra-uterine gestations in cases with PIUP because it is a sensitive marker in immunolabelling IT and ST. Combined application of inhibin alpha and CAM 5.2 might be more useful than betaHCG and HPL because the latter showed background staining in one third of the cases.

Developmental changes in the ploidy of mouse implanting trophoblast cells in vitro

Shortly after the onset of implantation, polar mouse trophoblast cells proliferate and give rise to the ectoplacental cone, constituted by two distinct cell populations: undifferentiated, diploid cells and giant cells. Giant cells characteristically exhibit exaggerated dimensions and polyploid nuclei. In this study, we employ ectoplacental cones as a dynamic source of trophoblast giant cells to analyze cell proliferation, cell death, and ploidy under in vitro conditions. Our results show that DNA synthesis and the increase in the cell number are relevant only during the first 24 h of culture. Subsequently, DNA synthesis still occurs, mainly in the giant cell compartment, while the number of cells gradually decreases. Cell death by injury and apoptosis was also observed in the non-giant cell compartment of the ectoplacental cone. These findings suggest that the first 24 h of culture are crucial to the mitotic activity of the ectoplacental cone cells that gradually ceases, favoring the endoreduplication process. The DNA synthesis index during the subsequent experimental intervals emphasizes accumulation of DNA for the polyploidization. There was clear correlation between DNA content and nuclear dimension. The ploidy values for the trophoblast giant cells varied from 2C up to 368C in the giant cells, but were not as expressive as those known from in vivo conditions, probably due to the absence of regulatory factors specific to the embryonic-maternal interface. In situ hybridization and histochemistry for the nucleolus-organizing region showed that trophoblast nuclei have only two marker signals, indicative of a typical polytenic process. This present study elucidates important aspects of trophoblast behavior and provides new information on trophoblast physiology in vivo and in vitro.

Identification and regulation of the IGFBP-4 protease and its physiological inhibitor in human trophoblasts and endometrial stroma: evidence for paracrine regulation of IGF-II bioavailability in the placental bed during human implantation

The IGF family plays an important role in implantation and placental physiology. IGF-II is abundantly expressed by placental trophoblasts, and IGF binding protein (IGFBP)-4, a potent inhibitor of IGF actions, is the second most abundant IGFBP in the placental bed, expressed exclusively by the maternal decidua. Proteolysis of IGFBP-4 results in decreased affinity for IGF peptides, thereby enhancing IGF actions. In the current study, we have identified the IGFBP-4 protease and its inhibitor in human trophoblast and decidualized endometrial stromal cell cultures, and we have investigated their regulation in an effort to understand control of IGF-II bioavailability at the placental-decidual interface in human implantation. IGFBP-4 protease activity was detected in conditioned media (CM) from human trophoblasts and decidualized endometrial stromal cells using (125)I-IGFBP-4 substrate. Identification of the IGFBP-4 protease as pregnancy-associated plasma protein-A (PAPP-A) was confirmed by specific immunoinhibition and immunodepletion of the IGFBP-4 protease activity with specific PAPP-A antibodies. The IGFBP-4 protease activity was IGF-II-dependent in trophoblast CM. In decidualized stromal CM, PAPP-A/IGFBP-4 protease activity was also IGF-II-dependent, but was evident only when IGF-II was added in molar excess of the predominant IGFBP in decidualized stromal cell CM, IGFBP-1, supporting bioavailable IGF-II as a key cofactor of IGFBP-4 proteolysis by PAPP-A. Cultured first and second trimester human trophoblasts (n = 5) secreted PAPP-A into CM with mean +/- SEM levels of 172.4 +/- 32.8 mIU/liter.10(5) cells, determined by specific ELISA. PAPP-A in trophoblast CM (n = 3) and did not change in the presence of IGF-II (1-100 ng/ml). Cultured human endometrial stromal cells (n = 4) secreted low levels of PAPP-A (6.25 +/- 3.6 mIU/liter.10(5) cells). A physiological inhibitor of PAPP-A, the proform of eosinophil major basic protein (proMBP), was detected in trophoblast CM at levels of 1853 +/- 308 mIU/liter.10(5) cells, determined by specific ELISA, and was nearly undetectable in CM of human endometrial stromal cells. Upon in vitro decidualization of endometrial stromal cells with progesterone, PAPP-A levels in CM increased nearly 9-fold without a concomitant change in proMBP. In contrast to the experiments with trophoblasts, IGF-II and the IGF analogues, Leu(27) IGF-II, and Des (1-6) IGF-II, resulted in a dose-dependent decrease of PAPP-A levels in decidualized endometrial stromal CM by 70-90%, and a dose-dependent increase in proMBP of 14- to 41-fold. The data demonstrate conclusively that the IGF-II-dependent IGFBP-4 protease of human trophoblast and decidual origin is PAPP-A. Furthermore, the differential regulation of decidual PAPP-A and proMBP by insulin-like peptides supports a role for trophoblast-derived IGF-II as a paracrine regulator of these maternal decidual products that have the potential to regulate IGF-II bioavailability at the trophoblast-decidual interface. Overall, the data underscore potential roles for a complex family of enzyme (PAPP-A), substrate (IGFBP-4), inhibitor (proMBP), and cofactor (IGF-II) in the placental bed during human implantation.

Expression of serine proteinase inhibitor PP5/TFPI-2/MSPI decreases the invasive potential of human choriocarcinoma cells in vitro and in vivo

OBJECTIVE

PP5/TFPI-2/MSPI is a Kunitz-type serine proteinase inhibitor with broad inhibitory spectra, abundantly produced by placenta and detected in the blood of pregnant women. Expression of PP5/TFPI-2/MSPI is exclusively detected in syncytiotrophoblasts of placenta, but is barely detectable in choriocarcinoma cells, a trophoblast-derived malignant tumor. Chromosome 7, in which the PP5/TFPI-2/MSPI gene is localized, is frequently lost in various types of tumors. We attempted to elucidate the relation between PP5/TFPI-2/MSPI expression and the malignant properties of choriocarcinoma cells.

METHODS

Human choriocarcinoma cells, JAR, were transfected with either a human PP5/TFPI-2/MSPI expression vector or an empty vector, and stable clones were obtained. Messenger RNA expression, protein secretion/localization, growth rate, and plating efficiency were evaluated. In vitro migration and invasive activity were determined by transwell chamber experiments. In vivo tumor growth was evaluated by the subcutaneous injection of cells to nude mice and followed by histological examination.

RESULTS

Expression of mRNA and protein of PP5/TFPI-2/MSPI were confirmed, and a high producing clone and a low producing clone were chosen for further analysis. The majority of secreted PP5/TFPI-2/MSPI protein was revealed to associate with the extracellular matrix. Expression of PP5/TFPI-2/MSPI did not affect the growth and migration of the tumor cells, but enhanced their plating efficiency. Its expression significantly inhibited invasion through the Matrigel. Invasive growth into the subcutaneous muscle layer was not evident in the nude mouse tumors of the PP5/TFPI-2/MSPI-expressing cells.

CONCLUSION

PP5/TFPI-2/MSPI-expressing choriocarcinoma cells showed suppressed potential of invasion in vitro and in vivo. It is suggested that loss or suppression of PP5/TFPI-2/MSPI expression may result in the acquisition of invasiveness in choriocarcinoma cells.

Studies of mesenchymal cells from 1st trimester human placenta: expression of cytokeratin outside the trophoblast lineage

A fibroblast cell strain that expressed cytokeratins 8 and 18 was isolated from explanted first trimester placenta. Its properties were consistent with an origin in the villous fibroblast-myofibroblast lineage, including expression of vimentin, smooth muscle alpha-actin and fibroblast surface protein. The cells grew rapidly in vitro, and exhibited tightly aligned bipolar morphology at confluence, absence of multi-nucleated cells, lack of secreted chorionic gonadotrophin, and absence of HLA G, placental alkaline phosphatase and pregnancy-specific beta-1 glycoprotein (SP1). On these criteria it was concluded they were not trophoblasts. On the basis of their morphology, cytokeratin expression and absence of CD34 and endoglin it was concluded they were not endothelial cells. They lacked desmin and smooth muscle myosin and so were not vascular smooth muscle cells. Further mesenchymal cell isolates were studied to determine the generality of these findings. Phenotypic heterogeneity was a consistent characteristic, but cytokeratin-positive cells were always present both in first trimester and term strains. Desmin was absent from almost all the cells isolated using the protocols employed, despite its occurrence in a significant subpopulation of cells in the villous stroma. Cytokeratins 8 and 18 can be observed in the stromal compartment of both first trimester and term placental villi. Cytokeratin has hitherto been regarded as a highly reliable marker for cells of the trophoblast lineage in vitro. These observations suggest that care should be taken in characterization of placental cell isolates; trophoblasts should be identified by the presence of cytokeratin 7 in preference to cytokeratin 8/18. The functional significance of cytokeratin expression in placental mesenchymal cells remains to be established.

Are human placental bed giant cells merely aggregates of small mononuclear trophoblast cells? An ultrastructural and immunocytochemical study

The ultrastructure of placental bed giant cells in early human pregnancies of 7-12 weeks gestational age is described. Their nature and function was further characterized by confocal immunofluorescence microscopy of paraffin sections labelled for cytokeratin, gap junction connexins (CX) 32 or 43, and placental hormones, alpha-human chorionic gonadotrophin (alpha-HCG) and human placental lactogen (HPL). Placental bed giant cells were observed with two phenotypes; as single large trophoblast cells containing one or more nuclear profiles in a voluminous cytoplasm, and as cell aggregates comprising mononuclear trophoblast cells in close apposition separated by narrow intercellular spaces. Cells within the aggregates are attached to one another by desmosomes, and also possess gap junctions as shown by immunolabelling for CX32 and CX43. By contrast, gap junctions were absent in the true multinucleated giant cells. Organelles present within the cytoplasm of the giant cells and their immunoreactivity for HPL and alpha-HCG suggest protein synthesis.

Localization of the cellular expression of inhibin in trophoblastic tissue

AIMS

Inhibin is a peptide hormone which is normally produced by ovarian granulosa cells and which inhibits the release of follicle stimulating hormone from the pituitary gland, thus acting as a modulator of folliculogenesis. Serum inhibin levels are higher during pregnancy than during the normal menstrual cycle and the placenta is thought to be a source of circulating inhibin. Previous studies have yielded conflicting results as to the cellular localization of inhibin in the placenta and the aim of the present study was to investigate the immunohistochemical localization of the hormone in placental tissue. We also wished to investigate whether inhibin could be demonstrated in choriocarcinoma and in non-gestational trophoblastic tissue.

MATERIALS AND RESULTS

Immunohistochemical staining was performed using a monoclonal antibody against the alpha subunit of human inhibin. Specimens included in the study were intrauterine products of conception (n = 36), extrauterine products of conception (n = 4), decidualized endometrium (n = 15), extrauterine decidualized tissue (n = 3), hydatidiform mole (n = 5), uterine choriocarinoma (n = 2) and testicular embryonal carcinoma with syncytiotrophoblast giant cells (n = 6). In cases of products of conception, including hydatidiform mole, there was consistent strong positive staining of syncytiotrophoblast but no staining of cytotrophoblast with anti-inhibin. Staining with anti-inhibin highlighted trophoblastic cells within the placental bed. In a small number of cases there was focal weak positive staining of decidua. There was positive staining of the two cases of uterine choriocarcinoma and of syncytiotrophoblast giant cells in the six cases of testicular embryonal carcinoma.

CONCLUSIONS

The study shows that immunohistochemically detectable inhibin alpha subunit in placental tissue is mainly localized within syncytiotrophoblast although in some cases there is also positive staining of decidua. Production of inhibin by these cells may account for raised serum levels during pregnancy. Inhibin can also be demonstrated in choriocarcinoma and in nongestational trophoblastic tissue. Inhibin is a sensitive marker of syncytiotrophoblast and staining with this antibody may prove useful in the diagnosis of choriocarcinoma and in the demonstration of trophoblastic cells in germ cell tumours.

Attachment and differentiation in vitro of trophoblast from normal and preeclamptic human placentas

OBJECTIVE

Trophoblast from preeclamptic patients shows impairment of various functions, including restricted invasive behavior of extravillous trophoblast. In this light the effect of different matrix components on attachment and differentiation of primary trophoblast cultures derived from normal and preeclamptic pregnancies was investigated.

STUDY DESIGN

Trophoblast was isolated from placentas of normotensive (n = 5) and preeclamptic patients (n = 5) and cultured up to 7 days on LabTek slides precoated with fibronectin, laminin, or vitronectin. Attachment was evaluated 24 hours after plating, the degree of syncytialization was evaluated, and slides were immunocytochemically stained for cytokeratin, vimentin, human chorionic gonadotropin, and human placental lactogen.

RESULTS

Trophoblast from placentas of preeclamptic patients showed a significantly lower attachment on fibronectin and vitronectin compared with controls. Diminished multinuclear cell formation was found on uncoated and laminin-coated slides in preeclamptic cases. No difference was found in the percentage of human chorionic gonadotropin- and human placental lactogen-positive cells between groups.

CONCLUSIONS

Lower trophoblast attachment on fibronectin and vitronectin was observed in preeclamptic pregnancies, which may reflect differences in expression of matrix receptors. Lower syncytialization of trophoblast in this group indicates an intrinsic defect in differentiation, but otherwise no differences were found in differentiation between normotensive and preeclamptic patients.

Isolation and characterization of human placental trophoblast subpopulations from first-trimester chorionic villi

A method for the simultaneous preparation of highly enriched human placental trophoblast populations (villous and extravillous) from first-trimester placental villi (5 to 12 weeks) by using sequential trypsinization, percoll gradient centrifugation, and negative selection with anti-CD9 immunomagnetic separation is described. The purification method resulted in the isolation of four distinct trophoblast populations identified on the basis of morphology and phenotyping: (i) mononuclear villous cytotrophoblast cells which, through differentiation, become committed to syncytium formation; (ii) an extravillous trophoblast population which appeared as a "crazy pavement" and, with subsequent subculturing, differentiated morphologically to mononuclear cells; (iii) an extravillous trophoblast fraction which fused to form multinucleated trophoblast giant cells; and (iv) floating intermediate extravillous trophoblast cells which fused together to form cell clumps and which further differentiated to a mononuclear anchoring intermediate extravillous trophoblast. Short-term cultures of the freshly isolated cell fractions consisted of heterogeneous trophoblasts at different differentiation stages as determined by their varied biochemical and morphological properties. All the isolated trophoblast populations expressed the cytokeratin intermediate filament and the epithelium-specific cell-cell adhesion molecule E-cadherin. The isolated villous trophoblasts in culture expressed integrins alpha 6 and beta 4 and reduced levels of beta 1 subunits, whereas the proliferating extravillous trophoblast cultures expressed alpha 1, alpha 3, and alpha 5 and high levels of beta 1 integrin subunits, vitronectin receptor (alpha V beta 3/beta 5), and major histocompatibility complex class 1 molecules. Furthermore, the isolated trophoblast populations secreted metalloproteases (such as type IV collagenases [mainly 72- and 92-kDa enzymes, i.e., gelatinases A and B]) and urokinase plasminogen activator, as evaluated by substrate gel zymography. This method of isolation should facilitate in vitro studies of trophoblast proliferation, differentiation, invasion, virus interactions, cytokenesis, and immunology.

Polytene chromosomes in mammalian cells

This article deals with the structural and functional organization of polytene chromosomes in mammals. Based on cytophotometric, autoradiographic, and electron microscopic data, the authors put forward a concept of nonclassic polytene chromosomes, with special reference to polytene chromosomes in the mammalian placenta. In cells with nonclassic polytene chromosomes, two phases of the polytene nucleus cycle are described, such as the endointerphase (S phase) and endoprophase (G phase). The authors generalize that the main feature of nonclassic polytene chromosomes is that forces binding the sister chromatids are much weaker than in the Diptera classic polytene chromosomes. This concept is confirmed by comparative studies of human, mink, and fox polytene chromosomes. The final step of the trophoblast giant cell differentiation is characterized by a transition from polyteny to polyploidy, with subsequent fragmentation of the highly polyploid nucleus into fragments of low ploidy. Similarities and dissimilarities of pathways of formation and rearrangement of nonclassic polytene chromosomes in mammals, insects, plants, and protozoans are compared. The authors discuss the significance of polyteny as one of the intrinsic conditions for performance of the fixed genetic program of trophoblast giant cell development, a program that provides for the possibility of a long coexistence between maternal and fetal allogenic organisms during pregnancy.

Characteristics of trophoblast cells migrating from first trimester chorionic villus explants and propagated in culture

We developed a method of propagating pure first trimester human trophoblast cells growing out of primary explants of mechanically derived chorionic villus fragments (Yagel et al, 1989; Graham et al, 1992). We have now extensively characterized these cells during their initial outgrowth and in long-term culture, employing a variety of markers and techniques as outlined below. By double label immunofluorescence using epithelial (cytokeratin) and mesenchymal (vimentin) cell markers, we identified the chorionic villus migrant cell populations as pure trophoblast (39 per cent of outgrowths) or a mixture of trophoblast and fibroblast (61 per cent). Further phenotyping of the pure trophoblast outgrowths by double label immunostaining using anti-cytokeratin antibody and a panel of other primary antisera revealed that these cells exhibit a variety of markers characteristic of extravillous invasive trophoblast cells in situ: insulin-like growth factor (IGF)-II, NDOG-5, proliferating cell nuclear antigen (PCNA), human leucocyte antigen framework antigen (W6/32) and a distinct set of integrins including alpha 1, alpha 3, alpha 5, alpha v and beta 1 subunits and alpha v beta 3/beta 5 vitonectin receptor. They were negative for alpha 6 and beta 4 integrin subunits. Immunogold electron microscopy of explants grown on type IV collagen gel revealed the production of conventional and oncofetal types of fibronectin by mononucleate trophoblast cells and human placental lactogen by multinucleate cells. Immunolabelling, flow cytometry and immunoprecipitation revealed that this phenotypic profile was retained with complete fidelity in the long-term culture; thus, trophoblasts migrating out of first trimester chorionic villus explants and their propagated progeny belong to the invasive extravillous trophoblast of the placenta.

Corticotrophin-releasing factor immunostaining is present in placenta and fetal membranes from the first trimester onwards and is not affected by labour or administration of mifepristone

BACKGROUND AND OBJECTIVES

Corticotrophin releasing factor (CRF) is present in the human placenta and fetal membranes. Placental CRF content and plasma CRF concentrations rise throughout gestation and fall rapidly after delivery. The regulation of CRF production from the placenta is poorly understood. The objective of this study was to use the antiprogestin, mifepristone, to determine whether progesterone has a regulatory effect on CRF production in the first trimester of pregnancy.

PATIENTS

Women undergoing first trimester (gestation 5-12 weeks) therapeutic abortion (by suction curettage with and without the synthetic PGE1 analogue, gemeprost (16,16-dimethyl-trans-delta 2-PGE1 methyl ester) vaginally 2-4 hours prior to the procedure; or with 600 mg mifepristone 48 hours prior to receiving 1 mg gemeprost vaginally), second trimester therapeutic abortion (600 mg mifepristone, 1 mg gemeprost), in association with preterm delivery (gestation 25-34 weeks) and at term (gestation 35-42 weeks) by spontaneous delivery, induced labour or elective Caesarean section.

MEASUREMENTS

Immunohistochemical localization of CRF and quantification of CRF content by radioimmunoassay of tissue extracts, in human placenta and fetal membranes.

RESULTS

CRF was immunolocalized to the syncytiotrophoblast cells of the placenta at all stages of gestation from 5 to 42 weeks. In the fetal membranes CRF immunoreactivity was localized in the epithelial and subepithelial cells of the amnion, some cells of the reticular and cellular layers of the chorion, and in decidual stroma. This pattern was seen in all tissues studied. Pretreatment with prostaglandins, mifepristone or both during the first trimester did not alter the distribution or the intensity of the CRF immunostaining. Placental CRF content rose throughout gestation but, consistent with the immunostaining results, was unaffected by the administration of mifepristone or by labour.

CONCLUSIONS

CRF is localized in the syncitiotrophoblast cells of the placenta and is clearly present early in the first trimester of pregnancy. The lack of an effect of mifepristone or mode of delivery suggests that syncytiotrophoblast produces CRF constitutively throughout pregnancy.

Giant and binucleate trophoblast cells of mammals

The cellular origin, structure, and function of trophoblastic giant cells (GC) and binucleate cells (BNC) are reviewed. Mammals in which these cells have received the greatest attention include rodents, rabbits, and humans (GCs), and ruminants and equids (BNCs). In almost all cases these cells arise from the cytotrophoblast. All are large cells and contain either two diploid nuclei (BNCs), multiple nuclei (human placental bed GCs), or single nuclei with amplified DNA content (rodent and rabbit GCs). Giant and binucleate cells typically exhibit the capacity for migration or invasion, although the degree of migratory activity varies between species. While most end up within, or at the interface with, endometrial tissue, in some instances the GCs or BNCs contribute directly to the interhemal membrane of the placenta. Hormone production is a property which most GC-BNC populations have in common. Lactogen or gonadotropin has been documented in almost all cells of this type examined to date, and in some animals they are also steroidogenic (e.g., rats and sheep). In spite of some common features, both structural and functional differences remain and it is suggested that use of terms such as mononuclear giant cells, multinucleate giant cells, and binucleate cells be continued rather than assuming that these cells are all members of a single trophoblastic subtype.

In vitro differentiation and ultrastructure of human extravillous trophoblast (EVT) cells

Tissue explants of anchoring villi from the first trimester placentae cultured on extracellular matrix (Matrigel) give rise to EVT cells in vitro. This study was designed to address two issues important for further application of the described in vitro model: first, were the observed EVT cells derived by cell proliferation in vitro and second, what is the degree of homology between the in vivo and the in vitro differentiated EVT cells. The cultures (tissue and matrix) were prepared for light and electron microscopic (EM) examinations. Semi-thin sections from Spurr epoxy resin-embedded tissue were used to 'pop-off' the selected area for EM examination. Cell proliferation in vitro was assessed immunohistochemically using proliferative cell nuclear antigen (PCNA) antibodies. Since positive hPL immunostaining has been consistently demonstrated in the invasive subpopulation of EVT cells from placental bed in situ, hPL staining was used as a marker of EVT cell differentiation in vitro. It has been demonstrated that PCNA antibodies immunostained nuclei of cytotrophoblast cells from cell column at the base of the anchoring villi, indicating that these cells expressed proliferative activity in vitro. Cytotrophoblast proliferation resulted in the formation of the flattened zone of cell outgrowths at the tip of anchoring villi. Cells from the distal layer of the cell column detached gradually and migrated into the surrounding matrix. These cells appeared as individual, round-shaped EVT cells with smooth surface cell membrane. Their cytoplasm was rich in glycogen and contained large lipid droplets and flattened cisternae of the RER. Positive PCNA immunostaining, along with the presence of mitotic figures, indicated that EVT cells in vitro retained the ability for cell proliferation. As a result of cell proliferation and migration, the number of EVT cells increased during the culture period of 4 days. EVT cell glycogen content and lipid stores decreased progressively as they migrated into the matrix. Individual EVT cells, as well as EVT cell clusters, became surrounded by the clear zone of digested matrix. Some cells started to express strong positive staining with hPL antibodies as soon as they had migrated outside the villous explant. By day 4 of culture, a small percentage of EVT cells (about 5-10%) ceased to migrate, firmly attached to the substratum and appeared as irregular shaped cells with filopodia-like projections. Their cytoplasm contained dilated cisternae of RER, a small number of glycogen granules and bundles of actin-like filaments located in the cytoplasm inside the plasma membrane.(ABSTRACT TRUNCATED AT 400 WORDS)

Variant choriocarcinoma (BeWo) cells that differ in adhesion and migration on fibronectin display conserved patterns of integrin expression

Starting from the BeWo choriocarcinoma cell line, two stable variant cell lines (epi and lc) were isolated. Epi cells displayed an epithelioid colony morphology while lc were fibroblastoid. lc cells attached and spread on fibronectin-coated surfaces at significantly lower density of fibronectin than epi or the parent cell line. lc also migrated more efficiently to fibronectin in a trans-filter assay than either epi or parent cells. Integrin expression by the cell lines was investigated by flow cytometry and immunoprecipitation from surface-labelled cells with a panel of subunit-specific antibodies. Integrins alpha 2 beta 1, alpha 5 beta 1, alpha v beta 1 and alpha 6 beta 4 were detected in each case, and levels of expression were identical in the two variant lines. Anti-functional antibodies were used to probe the role of integrins in fibronectin- and vitronectin-mediated adhesion. Complete inhibition of adhesion to fibronectin was observed with anti-beta 1 antibody, and partial inhibition with anti-alpha 5, suggesting that integrin alpha 5 beta 1 is mainly responsible for the interaction. Adhesion to vitronectin was inhibitable using anti-alpha v and anti-beta 1 antibodies, suggesting that integrin alpha v beta 1 is active in these cells as a vitronectin receptor. There was a correlation between the altered morphology of the variant cells and alterations in the distribution of integrin alpha 6 beta 4 and laminin in monolayer cultures. The results support the idea that fibronectin may mediate the migratory behaviour of extravillous trophoblast in vivo. Switch to a more migratory phenotype may be mediated by the selective activation of integrins and altered interaction with basement membrane.

Expression of c-erbB-2 protein and epidermal growth factor receptor in normal tissues of the female genital tract and in the placenta

The c-erbB-2 (HER-2/neu) protein is a membrane glycoprotein growth factor receptor that has molecular homology with the epidermal growth factor receptor (EGFR). To investigate the relationship between the expression of c-erbB-2 protein and EGFR in the tissues of the human female genital tract and in the placenta, we examined the immunohistochemical reactivity of monoclonal antibodies against both of these proteins. In the müllerian-derived genital tract, epithelial cells of the fallopian tube, endometrium, and endocervix showed reactivity for c-erbB-2 protein, whereas reactivity for EGFR was distributed mainly in the stromal cells throughout the menstrual cycle and during pregnancy. In addition, the staining intensity for EGFR in the endometrial stroma increased with its decidualization. In the exocervical squamous epithelium, basal cells were c-erbB-2 protein-negative and EGFR-positive, but the more differentiated squamous cells of the intermediate layer were c-erbB-2 protein-positive and EGFR-negative. In the placental tissues, cytotrophoblasts and syncytiotrophoblasts of the chorionic villi were c-erbB-2 protein-negative and EGFR-positive. In contrast, intermediate trophoblasts in the extravillous space were c-erbB-2 protein-positive and EGFR-negative. Thus, there is an inverse relationship between the expression of c-erbB-2 protein and EGFR in the tissues of the female genital tract and in the placenta. This suggests that there may be a regulatory mechanism(s) for the expression of both proteins that is associated with the differentiation and/or function of cells in the female genital tract and the placenta.

HPL-positive infiltrating trophoblastic cells in normal and abnormal pregnancy

In a series of 24 pregnant women, placental bed biopsies were performed in the third trimester at cesarean section. All the resulting specimens contained infiltrating trophoblast with both small and giant cells, and eight also contained vascular trophoblast. On immunoperoxidase staining for HPL, some small interstitial trophoblastic cells were positive in 12 cases. Some cells of the vascular intramural trophoblast and practically all cells of the vascular intraluminal trophoblast were positive. Seven cases were normal pregnancies whereas 17 were complicated by arterial hypertension and/or fetal growth retardation. A significant correlation between abnormal pregnancy and absence of HPL-positive interstitial cells in the placental bed biopsy was found. This probably indicates a diminished overall number of HPL-positive interstitial cells in the group of abnormal pregnancies and might reflect some defect of interstitial trophoblast. Such a defect may play a role in the arrest of the physiological changes of pregnancy in spiral arteries, which has been described in pre-eclampsia and in many cases of idiopathic fetal growth retardation.

Immunohistological demonstration of intermediate trophoblast in the diagnosis of uterine versus ectopic pregnancy: a retrospective survey and results of a prospective trial

Immunohistological demonstration of human placental lactogen (hPL) in non-villous, mononuclear intermediate trophoblastic cells may be of routine diagnostic value, when chorionic villi are absent in endometrial curettings from patients suspected of miscarriage of an intrauterine pregnancy. The histological presence and distribution of hPL was investigated in endometrial curettings from 90 patients studied retrospectively (47 had ectopic pregnancies, 14 miscarriages, and 29 legal abortions), and a consecutive, prospective series of 50 patients (40 had miscarriages and 10 had ectopic pregnancies) without chorionic villi in their endometrial curettings. Non-specific background staining was not a problem. The retrospective survey disclosed that hPL was a highly sensitive marker of intrauterine pregnancy (sensitivity = 0.98). In the prospective series, the predictive value of positive staining for hPL for intrauterine pregnancy was 1.00, and the sensitivity of hPL, as an indicator of uterine gestation, was 0.62. In absence of specific hPL-staining, the risk of ectopic pregnancy was about 50%. The immunohistochemical demonstration of hPL is a useful tool for identifying patients who are suspected of having had a miscarriage, but for whom evidence in the form of chorionic villi in endometrial curettings is lacking.

Corticotrophin-releasing hormone production by the placenta and fetal membranes

Maturation of the fetal hypothalamic-pituitary-axis may be an important factor in the stimulation of the onset of parturition. Recent studies have found that corticotrophin-releasing hormone (CRH), the hypothalamic component of this axis is also present in the placenta and the fetal membranes. We review the evidence demonstrating that glucocorticoids stimulate the production of CRH by these tissues towards term, and this is manifest by increases in CRH concentrations in maternal and fetal plasma and in placental extracts, and by increased abundance of CRH mRNA in the placenta. We discuss how CRH secretion by the placenta and fetal membranes is controlled. We hypothesize that CRH may be an important signal in initiating of the onset and maintenance of term labour and may also be a factor in the etiology of premature labour, through stimulation of the fetal HPA axis, and through paracrine/autocrine interactions within the placenta, fetal membranes and decidua.

Immunocytochemical distribution and localization of 15-hydroxyprostaglandin dehydrogenase in human fetal membranes, decidua, and placenta

Biochemical studies have shown the presence of type I oxidized nicotinamide-adenine dinucleotide-dependent 15-hydroxyprostaglandin dehydrogenase in human fetal membranes, decidua, and placenta. However, the localization of prostaglandin dehydrogenase within these tissues is not known. Because the distribution of prostaglandin dehydrogenase may affect the concentration of prostaglandins that reach the myometrium, we used immunocytochemistry to localize immunoreactive prostaglandin dehydrogenase in fetal membranes and placenta. We also examined whether this distribution changed with labor. Tissues were collected at term elective cesarean section or after term spontaneous labor and delivery, were fixed, embeded, and sectioned at 5 microns. Immunoreactive prostaglandin dehydrogenase was determined with a polyclonal primary antibody to human placental prostaglandin dehydrogenase and visualized with the avidin-biotin procedure. Epithelial and epithelium-derived cells were identified by positive staining with a polyclonal antikeratin primary antibody. Cytokeratin staining was observed in amniotic epithelium, trophoblast layer of chorion, invading trophoblast in decidua, and all subsets of trophoblast in the placenta. Immunoreactive prostaglandin dehydrogenase was localized to the trophoblast layer of chorion, invading trophoblast in decidua, and in syncytiotrophoblast and intermediate trophoblast but not cytotrophoblast in the placenta. In chorion, approximately 50% to 60% of the nonvacuolated trophoblast cells stained positively for prostaglandin dehydrogenase. There was no change in the localization of immunoreactive prostaglandin dehydrogenase in any tissue in association with labor.

Histological study of the materno-embryonic interface in spontaneous abortion

In a histological study of 184 specimens of complete spontaneous abortion, the following points were delineated. In cases of anomaly or death of the conceptus, there was a reduced trophoblastic penetration into the decidua and into the spiral arteries where physiological changes were limited or absent. Trophoblastic proliferation within the columns and the outer shell was limited with frequent disruption or even disappearance of the shell. These findings seem to be related to the untimely initiation of blood flow in the intervillous space which in turn is associated with arrest of pregnancy and eventual expulsion.

Comparative study of placental protein 19, human chorionic gonadotrophin and pregnancy-specific beta 1-glycoprotein as immunohistochemical markers for extravillous trophoblast in pregnancy and trophoblastic disease

PP19, a new placental tissue protein, has alpha 1-beta 1 electrophoretic mobility, a molecular weight of 36,500 and 3.9% carbohydrate. To study immunocytochemical PP19 localization in extravillous trophoblast, we obtained formalin-fixed specimens from extravillous tubal pregnancy at gestational weeks (GW) 7-9 (12 blocks); four early intrauterine pregnancies at GW 7-13 (12 blocks); four late pregnancies at GW 28-38 complicated with intramural uterine myoma, placenta increta and abruptio placenta (8 blocks); four invasive complete moles (9 blocks); and seven primary and metastatic gestational choriocarcinomas (12 blocks). Immunohistochemical staining was done for PP19, pregnancy-specific beta 1-glycoprotein (SP1) and human chorionic gonadotrophin (hCG) using the indirect-labeled antibody method [purified PP19 (Lot no. 225/242) and antibody against PP19 (Lot no. 632ZA) prepared by H. Bohn, antibodies against hCG (Behringwerke, Marburg, FRG) and SP1 (Dakopatts, Copenhagen, Denmark)]. In both early and late intrauterine pregnancies, the extravillous syncytiotrophoblastic cell (XST) showed positive staining for hCG and SP1 in the cytoplasm, as well as for PP19, which stained more intensively in the nucleus than in the cytoplasm. The three proteins were not seen in the evtravillous cytotrophoblastic cell (XCT) in the trophoblastic cell column and shell. The interstitial cytotrophoblast-like cell (ICT), which infiltrated into the decidua and myometrium, and their blood vessels, was immunoreactively positive for PP19 but negative for hCG and SP1 with the exception of SP1-positive ICT in the myometrium in late pregnancy. XST and ICT in the endosalpinx of tubal pregnancy stained for all three proteins.(ABSTRACT TRUNCATED AT 250 WORDS)

Subinvolution of the uteroplacental arteries in the human placental bed

Subinvolution of the uteroplacental arteries of the placental bed is a recognized cause of post partum haemorrhage causing significant morbidity. Whilst the physiological changes in these arteries during pregnancy and the part played by endovascular trophoblast migration are well documented, the sequence of events during involution and the pathophysiology of subinvolution are unknown. Using immunohistochemical techniques we have studied uteroplacental arteries in the placental bed in 25 cases of post partum haemorrhage and compared the subinvoluted vessels with normally involuted vessels. Non-involuted vessels were present in 22 test cases; these abnormal vessels were filled with thrombus and no endothelial lining was detected. Extravillous perivascular trophoblast was usually present in the walls of these abnormal vessels and in some cases was seen in an endovascular position. Subinvolution of placental site vessels may represent an abnormal interaction between maternal uterine cells and fetal trophoblast.

The human placental bed: histology, immunohistochemistry and pathology

There has been much interest recently on local intra-uterine materno-fetal interactions particularly in the placental bed where cellular relationships between mother and fetus are at their most intimate. While few histopathologists are expected to interpret formal placental bed biopsy specimens, confrontation with tissue from this site is common following abortion, post-partum haemorrhage or molar gestation. This review gives an account of recent advances in our knowledge of the histology, immunohistochemistry and pathology of the placental bed. It focuses particularly on extravillous trophoblast populations and their relationship to maternal cells and emphasizes the importance of vascular changes.

Immunohistochemical localization of HLA antigens and placental proteins (alpha hCG, beta hCG CTP, hPL and SP1 in villous and extravillous trophoblast in normal human pregnancy: a distinctive pathway of differentiation of extravillous trophoblast

Immunohistochemical localization of HLA antigens and placental proteins (alpha hCG, beta hCG CTP, hPL and SP1) in villous and extravillous trophoblast at various stages of normal human gestation were studied, using hysterectomy specimens. In the chorionic villi, the capacity for synthesizing placental proteins seemed to develop in parallel with the morphological change from mononuclear cells to multinucleated syncytiotrophoblast and no villous trophoblast expressed HLA antigens. In contrast, extravillous trophoblast, including the multinucleated trophoblastic cells at the deciduomuscular junction, expressed HLA-A, -B, and -C, and their capacity for synthesizing placental proteins did not seem to correspond with the degree of morphological change: the location of alpha hCG, beta hCG CTP and SP1 was restricted to mononuclear trophoblast in the superficial decidua, while hPL was present extensively in extravillous trophoblast. These findings strongly suggest that extravillous trophoblast possesses many distinctive biological features and differentiates in an independent manner. Mononuclear trophoblast forming the cell columns was also positive for HLA-A, -B, and -C, and no placental protein was demonstrated in these cells; this, together with previous morphological observations, may indicate the germinative nature of these cells.

Detection of oncomodulin, an oncodevelopmental protein in human placenta and choriocarcinoma cell lines

Oncomodulin was detected by immunocytochemical means in human placenta and found to occur in the cytotrophoblastic shell during implantation, in Langhans type villar cytotrophoblastic cells, and in extravillar cytotrophoblast or intermediate trophoblast. The presence of oncomodulin during first and early second trimester was confirmed by immunoradiometric assay (IRMA). In term placenta oncomodulin appeared in intermediate trophoblast cells, and was largely absent from the villi which lack cytotrophoblast. Furthermore, oncomodulin was abundant in the choriocarcinoma cell line BeWo, and detectable in JAR but not JEG-3. The function of this oncodevelopmental calcium-binding protein in normal development or in neoplasia is not known.

Increases of plasma eosinophil major basic protein levels late in pregnancy predict onset of labor

Major basic protein (MBP), the core of the eosinophil granule, is a potent toxin for parasites and mammalian cells; it also causes histamine release from mast cells and basophils. Plasma levels of MBP are increased in all pregnant women, and MBP has been localized by immunofluorescence to placental X cells and placental-site giant cells. Analyses of plasma MBP levels during pregnancy showed a striking correlation between increases late in the third trimester and spontaneous onset of labor. In one case with premature labor, the plasma MBP level also was increased coincidentally with labor. In contrast, in cases of prolonged gestation in which labor was induced, there were no late increases in MBP levels. Also, uterine contraction during spontaneous labor or oxytocin-induced labor did not cause elevations of plasma MBP. Measurement of plasma MBP levels in pregnancy may be useful for the prediction of labor and for the development of new means of intervention for premature labor and prolonged gestation.

Placenta accreta: an immunohistological study of trophoblast populations

Trophoblast populations in four cases of placenta accreta were characterized using antibodies directed against cell membrane antigens, placental hormonal products and low-molecular-weight cytokeratins in standard immunoperoxidase techniques. The results obtained with antibody to syncytiotrophoblast membrane (rabbit anti-StMPM), antibody to an epithelial membrane antigen (HMFGI) and a cytokeratin marker (CAM 5.2) appeared identical to those reported for normal term placental tissues. Similarly the localization of human placental lactogen (hPL), human chorionic gonadotrophin (hCG) and pregnancy-specific beta 1-glycoprotein (SP1) within trophoblast populations in placenta accreta was identical to their reported distribution in term placenta. However, increased reactivity at the villous-maternal junction was demonstrated with NDOGI, an antibody raised against term syncytiotrophoblast membrane and directed against hyaluronic acid. NDOGI reactivity at this site is normally maximal during early placental development and is virtually absent in the third trimester. The results suggest that placenta accreta does not arise through excessive trophoblast invasiveness or proliferation and the absence of decidua is of more importance in the pathogenesis. Trophoblast may regulate its development at an unfavourable intramyometrial implantation site by the production of hyaluronic acid.

Comparison of different antibody preparations against pregnancy-associated plasma protein-A (PAPP-A) for use in localization and immunoassay studies

Four antibody preparations against pregnancy-associated plasma protein (PAPP-A) were compared in order to find an explanation for the contradictory results published on tissue localization, clinical usefulness and biological function of PAPP-A. One of the preparations studied was a rabbit anti-PAPP-A antiserum which has been offered for general scientific use (Bischof et al. 1979). Only the IgG fraction of anti-PAPP-A antisera which appeared to be monospecific and had been further absorbed with fetal connective tissue gave specific uniform staining of the cytoplasm of the syncytiotrophoblast exclusively. Circulating PAPP-A could not be detected by RIA employing this IgG preparation in the non-pregnant state, or before 18 days after conception. Circulating PAPP-A could be detected in all seven pregnant women studied within 4 weeks after conception. Identical results were obtained with a commercially available IgG fraction against PAPP-A.

Immunohistochemical location of HPL, SP1 and beta-HCG in normal placentas of varying gestational age

Sixty-four placentas at various gestational ages were examined by immunohistochemical stains for HPL, SP1 and beta-HCG according to a modified PAP method (Sternberger 1970). Syncytiotrophoblast cell layer was identified as the main site of synthesis. Extravillous immunohistochemical reactions for HPL and SP1 (but not for beta-HCG) were found in X-cells of the basal plate and in the intervillous trophoblast islands. These cell types would thus seem to be derived from trophoblast. Hofbauer-cells of villous connective tissue stained specifically for beta-HCG apparently because of HCG phagocytosis. The intensity of staining for HPL, SP1 and beta-HCG was evaluated semiquantitatively in the syncytiotrophoblast cell layer at various gestational weeks. The maximum of staining for beta-HCG was found in the uniform syncytiotrophoblast layer of immature intermediate villi in early pregnancy (7-14 weeks), HPL and SP1 had their peak of staining reaction at 32-36 weeks of gestation. Increasing maturation results in a subspecialisation of the villous surface: epithelial plates (allowing feto maternal exchange) and nucleated portions of syncytiotrophoblast (which are the main site of hormone production and endocrine activity).

The placental bed biopsy: review from three European centers

This review derives from extensive experience with the placental bed biopsy technique in three centers over the last 30 years. A placental bed biopsy, usually taken at cesarean section, must include basal decidua and subjacent myometrium from the central zone of the placental site. Attention is drawn specifically to the sampling errors and to the pitfalls in morphologic interpretation of tissues, both maternal and fetal, that are continuously changing throughout the course of pregnancy. The features of the normal placental bed and of vascular lesions in pathologic pregnancies are briefly reviewed. Extension and elaboration of the technique and its more widespread use could contribute to the elucidation of many of the unresolved problems in human pregnancy.

Immunohistochemical aspects of immunological cross-reaction and masking of epitopes for localization studies on pregnancy-associated plasma protein A

The influence of antibody absorption procedures and proteolytic pre-treatment of formaldehyde-fixed placental tissue on the localization of pregnancy-associated plasma protein A by immunoperoxidase technique was examined. Apparently monospecific IgG fraction of the anti-plasma protein applied directly on fixed tissue resulted in staining of connective tissue and a thin apical rim of the syncytiotrophoblast. Further absorption of the antibody with foetal connective tissue abolished this staining reaction. Pre-treatment of the fixed placental tissue with trypsin prior to application of the antibody, which had been absorbed with connective tissue, resulted in staining within the cytoplasm of the syncytiotrophoblast exclusively. Identical staining was seen when this IgG preparation was used directly on frozen placental tissue. The results point to the importance of the specificity of the antibody preparations and of proteolytic unmasking of epitopes when fixed tissues are used for localization studies of pregnancy-associated plasma protein A by immunoperoxidase technique.