|
|
||||||||
Articles |
a Department of Obstetrics and Gynecology, New York University School of Medicine, New York, New York 10016
b Department of Physiology, University of Cambridge, Cambridge, United Kingdom
c Laboratory for Pregnancy and Newborn Research, Department of Physiology, Cornell University, Ithaca, New York 14853-6401
| ABSTRACT |
|---|
|
|
|---|
| INTRODUCTION |
|---|
|
|
|---|
In humans, the concentrations of glucocorticoid (GC) and corticotropin-releasing hormone (CRH) in amniotic fluid and maternal and fetal plasma increase in association with parturition [810]. Plasma levels of biologically active (free) CRH were high in women who delivered preterm and were low in those women delivering postterm [11]. Since CRH enhances prostaglandin production in fetal membranes, placenta, and decidua [12], and given that the placenta is the major source of plasma CRH during pregnancy [13], these results are consistent with the hypothesis that altered placental CRH production plays a role in the mechanisms that control the length of gestation [11]. GC treatment increases CRH levels in placental and fetal membrane cells [14], suggesting an important role of GC in the events surrounding parturition.
Epithelial and mesenchymal cells in human fetal membranes carry out extracellular matrix (ECM) protein synthesis [15]. The amnion epithelial cells, the innermost cell layer of fetal membranes, in conjunction with underlying fibroblasts, are the major cell types in amnion responsible for the synthesis of FNs and collagens at this site [15]. Similarly, ECM protein synthesis is carried out by epithelial cells (i.e., cytotrophoblasts) and mesenchymal cells in human placenta [16]. However, the patterns of cytotrophoblast FFN expression reflect their state of differentiation and invasive potential, with FFN restricted to extravillous cytotrophoblasts of the anchoring villi and cytotrophoblastic shell [17, 18]. Previous results obtained in our laboratory indicate that GC treatment reduced the expression of FFN, laminin, and collagen III in primary cultures of cytotrophoblasts and amnion epithelial cells [19, 20]. The GC-mediated reduction in cytotrophoblast FN expression is mediated through a short-lived protein [21], and stands in sharp contrast to the stimulatory effect of GC on FN expression that is observed in most cell types [22, 23].
The purpose of the present study was to elucidate the effects of GC treatment on FFN expression in the predominant ECM-synthesizing epithelial and stromal cell types from human amnion and placenta using primary cell culture models. In addition, we tested whether the patterns of ECM protein regulation observed in human amnion cells were conserved in other primate species, using the baboon as an example.
| MATERIALS AND METHODS |
|---|
|
|
|---|
Tissue culture media and dexamethasone (DEX) were obtained from Sigma (St. Louis, MO). Bovine sera were obtained from Gemini Bio-Products (Calabasas, CA). Laboratory plasticware was obtained from Falcon, Becton-Dickinson Labware (Lincoln Park, NJ). ITS+, a mixture containing insulin, transferrin, and selenium, was purchased from Collaborative Research-Becton Dickinson (Bedford, MA). Ultraspec used for RNA isolation was purchased from Biotecx Laboratories (Houston, TX). [32P]dCTP was from New England Nuclear (Boston, MA). Plasmids containing cDNAs to fibronectin and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) were from the American Type Culture Collection (Rockville, MD). Other reagents used in tissue culture and Northern blotting were from previously described sources [19, 21].
Methods
Cell culture Amnion epithelial cells were isolated from approximately 5 g of human or baboon amnion tissue by digestion with 5% trypsin [20]. This treatment removes amnion epithelial cells from their basal lamina without disrupting underlying collagenous connective tissue. After digestion with trypsin, amnion fibroblasts were released from membrane fragments by subsequent treatment with 0.25% collagenase [24]. The ratio of isolated epithelial cell to fibroblast was approximately 10:1 for both human and baboon preparations, and cell viability was approximately 90% as judged by trypan blue exclusion. Approximately 50 x 106 and 15 x 106 amnion epithelial cells were obtained per 5 g of human and baboon tissue, respectively.
Cytotrophoblasts were isolated from approximately 90 g of human villous tissue at term after trypsin digestion and centrifugation on continuous Percoll gradients as we have described [19]. Placental fibroblasts were isolated from 510 g of villous tissue after treatment with 0.25% collagenase on the basis of our modification [21] of procedures originally described by Fant and Nanu [25]. Fibroblasts between passage 2 and 7 were used for experiments.
After isolation, amnion and placental cells were plated at a density of 14 x 105 cells per 24-well dish for ELISA studies and 210 x 106 cells per 10-cm dish for Northern blot analysis in culture medium containing a 1:1 mixture of phenol red-free Ham's F-12:Dulbecco's modified Eagle's medium containing 10% charcoal-stripped calf serum (i.e., SCS medium) and ITS+ (a supplement utilized to obtain a final concentration of insulin of 6.25 µg/ml, transferrin 6.25 µg/ml, selenious acid 6.25 ng/ml, BSA 1.25 mg/ml, and linoleic acid 5.35 µg/ml) [19, 21] in the presence and absence of 100 nM DEX. After the indicated time in culture, conditioned media were saved for ELISA, and protein and RNA were extracted from adherent cells.
Cells were maintained at 37°C in SCS medium in a humidified atmosphere of 5% CO2 and 95% air.
FFN ELISA Levels of FFN in culture media were measured by an ELISA using FDC-6 monoclonal antibody according to information provided by the manufacturer (Adeza Biomedical, Sunnyvale, CA) as we have previously described [19, 26]. The concentration of FFN in culture media was determined in triplicate wells and was normalized to cell protein using the DC protein assay from Bio-Rad Laboratories (Hercules, CA).
Northern blotting RNA was extracted from placental and amnion cells using UltraSpec RNA (Biotecx) as we have described [21]. Approximately 20 µg total RNA was separated on a 1% agarose gel containing formaldehyde. After transfer, FN and GAPDH mRNAs were detected using 32P-labeled cDNA probes. Autoradiographic signals were quantitated using the Electrophoresis Documentation and Analysis System 120 and Digital Science 1D Image Software (Eastman Kodak, Rochester, NY).
Statistics Representative results for human and baboon amnion epithelial cells and fibroblasts and human cytotrophoblasts shown in Figures 1, 2, 4, and 5 are from 1 tissue representing 3 or 4 tissues from different subjects. Representative results for placental fibroblasts depict data obtained from 1 cell passage in 1 experiment representing 3 identically conducted experiments from 3 different cell passages. For the analysis of cumulative data shown in Figure 3 and the text accompanying Figure 4, the percentage difference in FFN levels between control and DEX-treated cultures was determined in 3 or 4 independent experiments; the average percentage difference was calculated; and a t-test then determined whether this value differed from 0. Results are expressed as a mean ± SEM. Statistical analysis was carried out using Student's t-test (SigmaStat software; Jandel, San Rafael, CA).
|
|
|
Human subjects and animal care Human amnion (n = 6) and placental tissue (n = 4) were obtained from pregnancies with normally grown, singleton fetus delivered by cesarean section at term. The study protocol was approved by the institutional review board committee at New York University School of Medicine. Baboon amnion tissue was obtained from animals delivered in the third trimester by cesarean section (n = 6) in the Laboratory for Pregnancy and Newborn Research Laboratory at Cornell University in compliance with institutional protocols for animal care as previously described [7]. Baboon amnions were shipped on ice to New York University School of Medicine for cell isolation. The time from tissue procurement to initiation of cell isolation was approximately 6 h.
| RESULTS |
|---|
|
|
|---|
Primary cultures of amnion epithelial cells and fibroblasts isolated from human and baboon term amnion were maintained for up to 9 days in SCS medium with and without 100 nM DEX, and levels of FFN in culture media were determined by ELISA. Results from representative experiments are shown in Figures 1 and 2. We observed that DEX treatment down-regulated FFN levels in human and baboon amnion epithelial cells to 540% of control levels. Conversely, DEX treatment increased FFN levels in amnion fibroblasts to 120220% of control levels on Days 6 and 8.
Statistical analysis of cumulative data obtained from 4 experiments indicated that DEX treatment significantly decreased FFN levels in human and baboon amnion epithelial cells to 35 ± 9% and 37 ± 4% of control levels, respectively (P < 0.01, Fig. 3). Conversely, DEX treatment increased FFN levels in human and baboon fibroblast cultures to 156 ± 12% and 193 ± 37% of control values, respectively (P < 0.05). These results suggest that GC differentially regulates FFN expression in epithelial and mesenchymal cells isolated from human and nonhuman primate amnion.
Effects of DEX Treatment on FFN Levels in Placental Cells
To test the effects of DEX on placental FFN expression, cytotrophoblasts and fibroblasts were isolated from a human term placenta and were maintained in SCS medium with and without 100 nM DEX for the indicated times (Fig. 4). Levels of FFN in cytotrophoblast culture media were down-regulated to 530% of control levels by a 2- or 4-day treatment with DEX. Since cytotrophoblasts differentiate to syncytiotrophoblasts after 72 h in culture [26, 27], this indicated that DEX suppressed FFN levels in both cytotrophoblasts and syncytiotrophoblasts. In sharp contrast, DEX treatment increased FFN expression in placental fibroblasts to 130220% of control levels between 3 and 10 days of culture (Fig. 4). Statistical analysis of cumulative data from 3 independent experiments revealed that DEX treatment significantly decreased FFN expression in cytotrophoblasts to 20 ± 6% of the level control (P < 0.01) and increased FFN levels in placental fibroblasts to 226 ± 2% of control (P < 0.01).
GC-Mediated Regulation of FN mRNA in Human Amnion and Placental Cells
Consistent with the FFN protein data described above, DEX treatment down-regulated expression of FN mRNA to approximately 30% of control levels in amnion epithelial cells and cytotrophoblasts as normalized to levels of GAPDH mRNA (Fig. 5). Conversely, DEX treatment up-regulated levels of FN mRNA to approximately 250% of control levels in amnion and placental fibroblasts (Fig. 5). Statistical analysis of cumulative data from 34 independent experiments revealed that DEX treatment significantly decreased levels of FN mRNA in amnion epithelial cells and cytotrophoblasts to 27 ± 13% of control (P < 0.01) and 16 ± 3% of control (P < 0.01), respectively. Conversely, analysis of cumulative mRNA results also revealed that DEX treatment significantly increased levels of FN mRNA in amnion fibroblasts to 288 ± 55% of control levels (P < 0.01). We noted that DEX treatment mediated a more modest increase in levels of FN mRNA in placental fibroblasts to 161 ± 47% of control levels that was not statistically significant (P = 0.26). In summary, our results indicate that GC treatment promotes differential effects on FN expression in the major epithelial and fibroblast cell types of amnion and placenta, and that these patterns of GC-mediated regulation are conserved in the human and nonhuman primate.
|
| DISCUSSION |
|---|
|
|
|---|
In light of the similar structure of human and nonhuman primate amnion [28], we used cell culture techniques developed for human amnion cell isolation [24] for the isolation of amnion epithelial cells and fibroblasts from baboon amnion. Treatment of whole amnion with trypsin gently removes amnion epithelial cells from the amnion basal lamina as essentially a homogeneous preparation [29, 30]. Collagenase treatment of epithelial cell-denuded amnion releases underlying mesenchymal cells [24]. We observed similar effects of GC on FN expression in amnion cells isolated from baboon and human tissue, suggesting that patterns of GC responsiveness are conserved in human and nonhuman primates. The observation of lower cellular yields and FN production in baboon amnion cells may be attributed to the modest delay in tissue processing or may represent interspecies differences.
It is likely that amnion epithelial cells and fibroblasts contribute to the synthesis of major ECM protein components of the amnion basal lamina (laminin, FN, collagen IV) [15]. It is likely that the synthesis of interstitial collagens in amnion (types I and III) is a primary function of amnion mesenchymal cells [24]. It is suggested that decreased levels of the collagen cross-linking enzyme lysyl oxidase [31], reflecting a lower density of mesenchymal cells in fetal membranes late in gestation, plays an important role in the reduction in tensile strength of fetal membranes that accompanies their premature rupture. Although in vitro results must be interpreted conservatively, our data do suggest that when GC levels are high in the third trimester [8], amnion mesenchymal cells will most likely play a more important role in FFN synthesis at this site. Furthermore, GC effects on FFN expression in amnion may be responsible in part for elevated cervico-vaginal levels of FFN observed in women at risk for preterm delivery [4], as well as the parturition-associated increase in FFN expression in fetal membranes [6, 7]. Our results may have important clinical implications in the face of the rising use of antenatal GC for the enhancement of fetal lung maturity [32]. GC therapy may alter the patterns of ECM protein expression in the amnion as well as tensile strength and may increase the risk of premature rupture of the fetal membranes.
Our previous results suggest that GC effects on FFN expression in cytotrophoblasts were not mediated by CRH [33], and therefore it is likely that parturition-dependent changes in FFN in placenta and fetal membranes are direct actions of GC and not of CRH. The GC effects on fetal membrane and placental ECM protein synthesis will be influenced by the tissue and cell type-specific expression of GC receptor and activity of 11ß-hydroxysteroid dehydrogenase (the enzyme that interconverts active cortisol and inactive cortisone), known components of placenta and fetal membranes [3436]. It is not clear whether these determinants of GC action are altered in activity or distribution in association with parturition.
In conclusion, our results show that GC differentially modulates FFN expression in the major epithelial and mesenchymal cell types isolated from human and baboon fetal membranes and placenta, suggesting that GC may play an important role in remodeling of the ECM at these sites in association with parturition.
|
| ACKNOWLEDGMENTS |
|---|
| FOOTNOTES |
|---|
1 These studies were supported in part through NIH grant HD 33909 (S.G.) and HD 21350 (P.W.N.). ![]()
2 Correspondence: Seth Guller, Dept. OB/GYN, NYU School of Medicine, 550 First Ave., New York, NY 10016. FAX: 212 263 5742; gulles01{at}mcrcr.med.nyu.edu ![]()
Accepted: January 14, 2000.
Received: September 27, 1999.
| REFERENCES |
|---|
|
|
|---|
5ß1. Biol Reprod 1999; 60:828838.This article has been cited by other articles:
![]() |
M.-J. Lee, Z. Wang, H. Yee, Y. Ma, N. Swenson, L. Yang, S. S. Kadner, R. N. Baergen, S. K. Logan, M. J. Garabedian, et al. Expression and Regulation of Glucocorticoid Receptor in Human Placental Villous Fibroblasts Endocrinology, November 1, 2005; 146(11): 4619 - 4626. [Abstract] [Full Text] [PDF] |
||||
![]() |
M.-J. Lee, Y. Ma, L. LaChapelle, S. S. Kadner, and S. Guller Glucocorticoid Enhances Transforming Growth Factor-{beta} Effects on Extracellular Matrix Protein Expression in Human Placental Mesenchymal Cells Biol Reprod, May 1, 2004; 70(5): 1246 - 1252. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |