| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Article |
Department of Gynecology and Obstetrics (I.Y.C.S., L.C.G.), Center for Research on Women’s Health and Reproductive Medicine, Stanford University Medical Center, Stanford, California 94305-5317; and Department of Molecular and Structural Biology (M.T.O., C.O.), University of Aarhus, 8000 Aarhus C, Denmark
Address all correspondence and requests for reprints to: Linda C. Giudice, M.D., Ph.D., Division of Reproductive Endocrinology and Infertility, Center for Research on Women’s Health and Reproductive Medicine, Department of Gynecology and Obstetrics, Stanford University Medical Center, Stanford, California 94305-5317. E-mail: giudice{at}stanford.edu.
Abstract
Pregnancy-associated plasma protein-A (PAPP-A) is a product of the placenta and decidua and is secreted into the maternal circulation during human pregnancy. It recently has been identified as an IGF binding protein (IGFBP)-4 protease. Presumed functions at the maternal-fetal interface are to proteolyze IGFBP-4 and thus increase IGF bioavailability locally in the placenta, to promote IGF-II-mediated trophoblast invasion into the maternal decidua, and to modulate IGF regulation of steroidogenesis and glucose and amino acid transport in the villous. Herein, we have investigated the possibility that IGFBP-4 proteolysis may occur on the trophoblast cell membrane, presumably to increase local bioavailable IGF for interactions with cognate IGF membrane receptors. Human trophoblasts were cultured, trophoblast plasma membranes were isolated and solubilized, and IGFBP-4 protease activity and PAPP-A immunoreactivity in the solubilized plasma membrane fraction were investigated. IGFBP-4 protease activity was detected in solubilized human trophoblast membranes, resulting in cleavage of recombinant human IGFBP-4 into 18- and 14-kDa fragments, detected by Western immunoblot analysis. This protease activity was dependent on the presence of IGF-II, and its metal ion dependence was demonstrated by inhibition of the protease by the metal chelators, EDTA and EGTA. The presence of PAPP-A in solubilized human trophoblast membranes was demonstrated by Western immunoblotting. Trophoblast membrane PAPP-A had a relative molecular weight of approximately 200 kDa and comigrated on (reducing) SDS-PAGE with recombinant human PAPP-A and PAPP-A secreted into media conditioned by cultured human trophoblasts. IGFBP-4 protease activity was not detected after immunodepletion of PAPP-A from the trophoblast membrane fraction with PAPP-A polyclonal antibodies, suggesting the identity of the membrane-derived IGFBP-4 protease as PAPP-A. Immunocytochemistry revealed PAPP-A on the cell membrane and in the cytoplasm of human trophoblasts in culture. Together, these data demonstrate the presence of an IGF-II- and metal-dependent IGFBP-4 protease activity in human trophoblast plasma membranes, identified as PAPP-A, which is well situated to proteolyze IGFBP-4 at the maternal-placental interface to facilitate IGF action at the villous surface and/or the invading extravillous cytotrophoblast.
PREGNANCY-ASSOCIATED PLASMA PROTEIN A (PAPP-A) is expressed by human placental trophoblasts (1, 2, 3, 4) and has long been recognized as a secreted placental product of unknown function that is a useful first trimester marker for Down’s syndrome (5). Recently, a function for PAPP-A as a protease for IGF binding protein (IGFBP)-4 has been identified (6). Identification of PAPP-A as an IGFBP-4 protease has been demonstrated in human fibroblasts (6), ovarian follicular fluid and granulosa cells (7, 8), media conditioned by human trophoblast and decidualized endometrial stromal cells (9), and human pregnancy serum (10, 11). The enzyme is IGF-II- and metal ion-dependent (6, 7, 8, 9, 10, 11, 12) and cleaves IGFBP-4, decreasing its affinity for IGF-II, presumably resulting in increased bioavailable (free) IGFs in these fluids and media. In pregnancy, more than 99% of PAPP-A circulates as a covalent, heterotetrameric 2:2 complex of 500 kDa with the proform of eosinophil major basic protein (proMBP; Ref. 13), a 50-kDa protein also of placental origin (3), which functions as an inhibitor of the proteolytic activity of PAPP-A (10). PAPP-A contains an elongated zinc-binding motif that is strictly conserved within the metzincins, a superfamily of zinc peptidases (14). During human pregnancy in the maternal circulation, elevated levels of PAPP-A and IGFBP-4 proteolysis are detected early in gestation (15, 16, 17). In a recent study, we demonstrated that the trophoblast and the maternal decidua are likely sources of the IGFBP-4 protease/PAPP-A in the circulation of pregnant women (9).
IGFs are mitogenic peptides that regulate cell proliferation and differentiation and are important for fetal and placental growth during pregnancy (see review in Ref. 18). IGFBP-4, the second most abundant IGFBP in the placental bed (19), is an inhibitor of IGF actions (20), and proteolysis of IGFBP-4 enhances IGF bioavailability (21). Thus, it is likely that PAPP-A plays an important role in regulating the availability of IGFs in pregnancy serum for placental growth (and therefore fetal growth) and regulating placental functions. Locally in the placental bed, where IGF-II is abundantly expressed by the trophoblast and IGFBPs by the decidua, IGF-II facilitates trophoblast invasion into the maternal decidua (22). IGF-I is important in syncytiotrophoblast steroidogenesis (23, 24) and glucose and amino acid uptake in the villous portion of the placenta (25). Thus, regulation of bioavailable IGFs is likely important for placental development and function. Bioavailability of IGFs at their cognate receptors requires free ligand, i.e. IGFs unbound from their binding proteins. In vitro studies demonstrate that proteolysis of IGFBP-4 occurs in select body fluids, conditioned medium, and serum, which decreases the affinity of IGFBP-4 for IGF-II and presumably provides increased free, bioavailable IGFs in solution. In addition, IGFBP proteolysis may occur on the cell membrane, further enhancing IGF availability locally to IGF membrane receptors for IGF action. We have considered that proteolysis of IGFBP-4 by PAPP-A may occur on the trophoblast cell membrane, and in the present study we investigated IGFBP-4 protease activity and PAPP-A in the solubilized plasma membrane fraction prepared from cultured human trophoblasts. The data strongly support the identity of membrane-bound PAPP-A as membrane-bound IGFBP-4 protease.
Materials and Methods
Human trophoblast cell culture
Human placentae (n = 16) were collected at 10–18 wk under a protocol approved by the Stanford University Panel on the Use of Human Subjects in Medical Research, after informed consent. Placentae were obtained at elective termination of genetically normal pregnancies. Trophoblasts were isolated as described (26). Briefly, trophoblast villi containing a mixed population of syncytiotrophoblasts and cytotrophoblasts, including extravillous trophoblasts, were scraped into tissue culture dishes containing DMEM, and the suspension was put on ice for 5 min. After the supernatant was decanted, digestion buffer (26) was added (vol/wt 3:1) and incubated with shaking at 37 C for 10 min. The digestion mixture was placed on ice for 5 min. After the supernatant was decanted, the digestion was repeated once more for 10 min at 37 C, followed twice by trypsin/EDTA (Life Technologies, Inc., Rockville, MD) treatment for 15 min at 37 C with shaking. The digested supernatant was collected after each trypsin treatment. The supernatant was filtered through a 40-µm nylon cell strainer (Falcon) and centrifuged for 5 min at 2000 revolutions per minute (rpm) in a Beckman Allegra 6R Centrifuge (Beckman Instruments, Fullerton, CA). The trophoblast pellet was collected and resuspended in 4 ml of culture medium (DMEM and 2% Nutridoma; Roche Molecular Biochemicals, Indianapolis, IN), layered onto a 15-ml Percoll gradient (10%, 25%, and 50%), and centrifuged for 30 min at 2700 rpm. The middle layer (
7 ml at the interphase between 25% and 50% Percoll) was collected, and trophoblasts were pelleted by centrifugation at 2000 rpm for 5 min. Cells were then washed twice with culture medium and plated at 8 x 105 cells/ml in 10-cm diameter tissue culture dishes. Cultures were more than 95% pure trophoblasts, as determined by cytokeratin immunocytochemistry (26).
Preparation of human placental trophoblast cell membranes
Trophoblast membranes were isolated as described (27, 28). Briefly, after 48 h of culture, trophoblasts were collected, washed twice in cold PBS, and lysed in 0.5 ml cold RIPA buffer (0.15 N NaCl, 0.5% Nonidet P-40, 0.1% SDS, 50 mM Tris, pH 7.6). After standing on ice for 30 min, the cell lysate was centrifuged at 800 rpm for 5 min. The supernatant was centrifuged at 8000 rpm for 20 min. The resultant supernatant fraction was further centrifuged for 60 min at 50,000 rpm in a Beckman Ultracentrifuge (Beckman Instruments). The opaque membrane pellet, reportedly containing more than 99% pure plasma membrane with little or no contamination from Golgi or endoplasmic reticulum (27), was resuspended in 0.2 ml 20 mM Tris (pH 7.6), 20 mM NaCl, 3 mM CaCl2. This fraction was used in all experiments.
IGFBP-4 protease assay
IGFBP-4 protease activity in the solubilized trophoblast membrane fraction was determined with IGFBP-4 as a substrate, using a modification of previously reported IGFBP-4 protease assays (6, 29). Briefly, a 25-µl reaction mixture, containing 20 mM Tris, pH 7.6, 3 mM CaCl2, IGF-II (100 ng; Bachem Bioscience, Inc., King of Prussia, PA), recombinant human (rh)IGFBP-4 (150 ng; Ref. 10), and the solubilized membranes (1–3 µg protein) were incubated for 16 h at 37 C. The positive control was the addition of term pregnancy serum (5 µl) in lieu of the trophoblast membranes, and the negative control was omission of the membrane preparation and addition of buffer alone. At the conclusion of the incubation, the reaction mixture was electrophoresed on reduced 10–20% SDS-PAGE followed by Western blotting of intact IGFBP-4 and IGFBP-4 fragments, using polyclonal human IGFBP-4 antibodies (Upstate Biotechnology, Inc. Lake Placid, NY). In some experiments, IGF-II was omitted, or different concentrations of IGF-II were added. In other experiments, the metal chelating agents, EDTA and EGTA (5 mM), were added. MultiMark Multi-Colored proteins (Invitrogen, Carlsbad, CA) were used as molecular weight standards.
Immunodepletion assay
Solubilized membranes were mixed with polyclonal PAPP-A antibodies (1:250; DAKO Corp. International, Carpinteria, CA), or with nonimmune IgG (negative control) for 6 h at 4 C, followed by absorption with protein G plus/protein A agarose beads (Oncogene Research Products, San Diego, CA) for 16 h on ice. This step was repeated once again for 6 h on ice. The resultant supernatants were pooled and used in the IGFBP-4 protease assay.
Western blot analysis of PAPP-A in solubilized trophoblast membranes
Solubilized trophoblast membranes (20 µg), media (30 µl) conditioned by cultured human trophoblasts (positive control), and 5 µg rhPAPP-A (positive control; Ref. 10) were subjected to electrophoresis on reducing 8–16% SDS-PAGE (Jules Biotechnology, Inc., Science Park, New Haven, CT). Western blotting onto nitrocellulose and immunovisualization were conducted with a monoclonal antibody to PAPP-A (clone no. 234-2; Ref. 30). Nonimmune IgG was used as the negative control to determine nonspecific adsorption.
Immunocytochemistry
Immunocytochemistry was performed using trophoblasts (14–16 wk) cultured for 2 d in medium described above. Cells were then washed twice in PBS and fixed for 2–5 min in methanol/acetone (1:1). PBS containing 3% nonfat milk was used for blocking, followed by incubation with monoclonal antibody clone 234-5 (Ref. 30 ; diluted 1:750) for 30 min at room temperature. Omission of the primary antibody was used as the negative control. After washing three times in PBS/3% nonfat milk, immunocytochemical staining was done with Vectastain ABC peroxidase system (Vector Laboratories, Inc.).
Results
IGFBP-4 protease activity in solubilized human trophoblast membranes
To determine whether human trophoblast membranes contain IGFBP-4 protease activity, solubilized trophoblast membrane fractions were incubated with rhIGFBP-4 (0.5 µM), in the absence and presence of IGF-II. Western immunoblot analysis (Fig. 1
) demonstrates the disappearance of intact 32-kDa rhIGFBP-4 and the appearance of lower molecular weight immunoreactive fragments (18 and 14 kDa) in the presence of the solubilized membrane fraction and added IGF-II (0.3 µM; Fig. 1, lanes b and c). IGFBP-4 proteolysis was not detected in the absence of added IGF-II (lane d). Intact rhIGFBP-4 is shown at time zero of the incubation (lane a) that proceeded overnight (16 h) before analysis of samples and immunoblotting with antibodies to IGFBP-4.
|
Some IGFBP proteases are metalloproteinases (21). The metal ion dependence of solubilized human trophoblast membranes was investigated using metal chelators in the IGFBP-4 protease assay. Figure 2 shows representative data from six experiments using six different membrane preparations and demonstrates inhibition of trophoblast membrane IGFBP-4 protease activity in the presence of the metal chelators, EDTA and EGTA (lanes c and d, respectively), compared with without addition of these chelators and in standard assay buffer (lane b).
|
Although it is tempting to attribute the IGFBP-4 protease activity in human trophoblast membranes to PAPP-A, it has not been demonstrated that PAPP-A is a membrane-bound enzyme in any system in which it has been detected. To address this, we investigated PAPP-A in solubilized human trophoblast membrane proteins by Western immunoblotting and using monoclonal antibodies to PAPP-A. Figure 3 shows immunoreactive PAPP-A in the solubilized trophoblast membrane fraction (lane a) with a relative molecular weight of approximately 200 kDa (with reduction). It comigrates with PAPP-A that is secreted into media conditioned by cultured human trophoblasts (lane b) and with rhPAPP-A (lane d). Nonpregnant human serum containing negligible levels of PAPP-A demonstrates minimal PAPP-A immunoreactivity (lane c) and served as a negative control. In addition, nonimmune IgG served as another negative control (data not shown).
|
To investigate whether the IGFBP-4 protease activity in solubilized trophoblast membrane proteins can be attributable to PAPP-A, we depleted PAPP-A from the trophoblast membrane fraction using PAPP-A polyclonal antibodies and then tested the PAPP-A immunodepleted fraction for IGFBP-4 protease activity. Figure 4 demonstrates complete IGFBP-4 proteolysis by the fraction after overnight incubation with rhIGFBP-4 (lane a) and when the fraction was processed with nonimmune IgG (control, lane b). IGFBP-4 protease activity was not detected (lane c) after PAPP-A polyclonal antibodies were used to deplete PAPP-A from the solubilized trophoblast membrane proteins.
|
Immunocytochemical studies (Fig. 5) revealed immunoreactive PAPP-A on the cell membranes and in the cytoplasm of human trophoblasts isolated and cultured from second trimester placentae. No nuclear staining was observed, and the nonimmune control (Fig. 5a) showed no appreciable immunoreactive PAPP-A on the cell membranes or in the cytoplasm. These data further support the localization of PAPP-A on trophoblast cell membranes.
|
In human placenta, IGFs play important roles in syncytiotrophoblast steroidogenesis (23, 24) and glucose and amino acid transport in the villi (25) and also in the invasion of the extravillous trophoblast into the maternal decidua (19, 22). IGF-I is expressed in cytotrophoblasts, but not syncytiotrophoblasts (19), in placental villi that bathe in maternal blood that contains high levels of IGF-I, IGF-II, and IGFBPs. IGF-II is highly expressed by the intermediate trophoblast, especially in the front invading into the maternal decidua (19), and on the maternal side, the decidua does not make IGF-II, but rather secretes high levels of IGFBPs, with IGFBP-1 being the predominant IGFBP, followed by IGFBP-4 (19). Maternal decidual IGFBPs mostly inhibit IGF actions and may have some IGF-independent effects on these trophoblast functions (20, 21). Proteolysis of IGFBPs by specific IGFBP proteases increases free, bioavailable IGFs, and regulation of this process in the placenta is likely to be important in regulating trophoblast function. We have previously shown that human trophoblasts secrete the IGFBP-4 protease, PAPP-A (9). Herein, we have demonstrated the presence of IGF-II-dependent and metal ion-dependent IGFBP-4 protease activity in solubilized human trophoblast plasma membranes and have identified this protease activity as PAPP-A. The concept of membrane-bound PAPP-A on the cell surface supports the hypothesis that proteolysis of IGFBP-4 may occur on the trophoblast cell membrane to enhance IGF action on its cognate signaling receptors on the trophoblast membrane for regulation of glucose and amino acid uptake and cellular invasion. In contrast, secreted (as opposed to membrane bound) PAPP-A may potentiate IGF actions by proteolysis of decidual-derived IGFBP-4 for paracrine actions on nearby cells, such as the decidual stromal cells or vascular cells, or may potentiate IGF actions for maternal tissue growth during pregnancy.
In our culture system, there are several trophoblast phenotypes represented, including syncytiotrophoblasts, cytotrophoblasts, and extravillous trophoblasts, from which the plasma membrane fraction was purified. Although this study has not specifically localized which cell type contains membrane-bound PAPP-A, in situ hybridization studies using term placentae demonstrate PAPP-A mRNA expression in syncytiotrophoblasts and extravillous cytotrophoblasts (3). In addition, immunohistochemical studies with first trimester and term placentae have also localized PAPP-A to syncytiotrophoblasts, extravillous trophoblasts, and the extracellular matrix in placental septae (3, 4). Early immunohistochemical studies on the cellular localization of PAPP-A were conflicting, likely due to different antisera used and the fact that available polyclonal antibodies to PAPP-A contained antibodies to both PAPP-A and proMBP (13). They demonstrated localization only to the apical surface of the syncytiotrophoblasts of 6-wk placentae, as well as term placentae (1, 2, 31, 32, 33), or localization to the syncytiotrophoblast and the extravillous, intermediate (invading) trophoblast (34, 35). Subsequent immunofluorescence studies, using proMBP-absorbed anti-PAPP-A polyclonal antibodies, localized immunoreactive PAPP-A to the extravillous trophoblast in the anchoring villi, as well as to the syncytiotrophoblast layer in the floating villi (3). No other tissue elements, such as villous stroma or fetal endothelial cells, showed positive staining. In culture, trophoblasts lose their polarity, and thus localization of PAPP-A to the basal or apical surface of the syncytiotrophoblast would not be possible in this study. However, earlier studies using immunohistochemistry have localized PAPP-A to the apical surface (see above) and to the brush border of the villous syncytiotrophoblast (36). Taking this together, we conclude that PAPP-A is likely membrane-associated on the syncytiotrophoblast as well as the extravillous trophoblast, although this still remains to be validated.
The data presented herein on identification of PAPP-A in solubilized trophoblast membranes and of immunoreactive PAPP-A on cell membranes of cultured human trophoblasts support the localization of this enzyme to the trophoblast plasma membrane (in addition to its known secretion into media conditioned by trophoblast cells and into the maternal circulation). PAPP-A contains a signal peptide that can target it for secretion from cells in which it is synthesized, and to our knowledge there is no transmembrane protein motif or consensus sequence in PAPP-A (14). Thus, it is likely that PAPP-A in solubilized human trophoblast membranes is adsorbed or adherent to the plasma membrane. Whether there is a receptor for PAPP-A is an interesting issue for which we have no data, but which we are currently investigating. In the current study, we found that membrane-associated PAPP-A requires IGF-II and Ca2+ as cofactors (Fig. 2
). Whether the kinetics of IGFBP-4 proteolysis in the presence of these cofactors is the same for PAPP-A on trophoblast plasma membrane and PAPP-A in solution (in media conditioned by these cells in culture or in maternal serum) is not known and is the subject of investigation in our laboratory.
Our findings of IGFBP-4 protease activity in the solubilized plasma membrane fraction from cultured human trophoblasts is the first demonstration of membrane-bound IGFBP-4 protease activity in any cell type. These findings underscore the importance of reassessing proteolytic events on the cell surface of a variety of cells for regulation of IGF bioavailability with regard to PAPP-A and for the numerous IGFBP proteases that have been described in body fluids, conditioned media of a variety of cells, and tissue extracts. This membrane platform for IGFBP proteolysis may be a more general phenomenon than previously appreciated.
Acknowledgments
We thank Dr. Carl R. Nash for providing the human placentae tissue samples.
Footnotes
This work was supported by a Collaborative Research Initiative of the National Institutes of Health Specialized Cooperative Centers Program in Reproductive Research through cooperative agreement HD 35789-05 (to L.C.G.) and by the Novo Nordic Foundation (to M.T.O., C.O.).
Abbreviations: IGFBP, IGF binding protein; PAPP-A, pregnancy-associated plasma protein A; proMBP, proform of eosinophil major basic protein; rh, recombinant human; rpm, revolutions per minute.
Received April 9, 2002.
Accepted July 22, 2002.
References
5ß1 the integrin in human cytrophoblasts and inhibits their invasion into decidualized endometrial stromal cells in vitro. Growth Horm IGF Res 8:21–31[Medline]
This article has been cited by other articles:
 |
|
 |
|
  R. Sood, J. L. Zehnder, M. L. Druzin, and P. O. Brown Gene expression patterns in human placenta PNAS, April 4, 2006; 103(14): 5478 - 5483. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. E. Murphy, R. Smith, W. B. Giles, and V. L. Clifton Endocrine Regulation of Human Fetal Growth: The Role of the Mother, Placenta, and Fetus Endocr. Rev., April 1, 2006; 27(2): 141 - 169. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Nie, Y. Li, K. Hale, H. Okada, U. Manuelpillai, E. M. Wallace, and L. A. Salamonsen Serine Peptidase HTRA3 Is Closely Associated with Human Placental Development and Is Elevated in Pregnancy Serum Biol Reprod, February 1, 2006; 74(2): 366 - 374. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Peng, Y. Huang, F. Jin, S.-W. Jiang, and A. H. Payne Transcription Enhancer Factor-5 and a GATA-Like Protein Determine Placental-Specific Expression of the Type I Human 3{beta}-Hydroxysteroid Dehydrogenase Gene, HSD3B1 Mol. Endocrinol., August 1, 2004; 18(8): 2049 - 2060. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. S. Laursen, M. T. Overgaard, K. Weyer, H. B. Boldt, P. Ebbesen, M. Christiansen, L. Sottrup-Jensen, L. C. Giudice, and C. Oxvig Cell Surface Targeting of Pregnancy-associated Plasma Protein A Proteolytic Activity. REVERSIBLE ADHESION IS MEDIATED BY TWO NEIGHBORING SHORT CONSENSUS REPEATS J. Biol. Chem., November 27, 2002; 277(49): 47225 - 47234. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
