| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Pregnancy |
Endocrine Sciences3
School of Biological Sciences,4 University of Manchester, Manchester, M13 9PT, United Kingdom
Academic Unit of Obstetrics & Gynaecology,5 Human Development and Reproductive Health Academic Group, University of Manchester, St Mary's Hospital, Manchester, M13 OJH, United Kingdom
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2-macroglobulin, and all three inhibitors attenuated the proteolysis of IGFBP-1 by MMPs. The N-terminal sequence analysis of the fragments revealed that the enzymes cleave IGFBP-1 at 145Lys/Lys146, resulting in a small (9-kDa) C-terminal peptide of IGFBP-1. These findings suggest cleavage of IGFBP-1 as a novel mechanism in the control of placental development by matrix metalloproteases.
decidua, growth factors, placenta, trophoblast
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The insulin-like growth factors (IGFs) are known regulators of fetal growth. Ablation of either the IGF-I or IGF-II gene reduces birth weight to 60% that of normal littermates [4, 5], and elimination of both genes accentuates the growth-restricted phenotype. Interestingly, Igf-II-null mice have small placentas, indicating that the IGFs may also influence fetal growth through a role in promoting normal placental development. Recently, this has been confirmed by studies in which targeting disruption of the IGF-II gene solely to the placenta resulted in pups with low birth weight [6].
Because IGF-II has both mitogenic and metabolic actions, it has the potential to influence many aspects of placental development and function. However, mRNA localization studies demonstrate abundant IGF-II expression in the trophoblastic columns of anchoring villi, particularly in those cells at the leading edge of the column [7], suggesting that IGF-II promotes trophoblast migration into the endometrium. This concept is supported by in vitro studies demonstrating that in primary first-trimester explant cultures [8], monolayer wounding [9], or trans-Matrigel barrier assays [10], trophoblast migration is enhanced by IGF-II.
The activity of IGF is controlled by a family of six binding proteins (IGFBPs) [11]. At the fetal/maternal interface, IGFBP-1 is the most relevant member, because it is a major secretory product of decidua [12, 13]. Additionally, IGFBP-1 is involved in the control of trophoblast migration, yet its precise role remains controversial, with reports of IGFBP-1-mediated enhancement or restraint of trophoblast migration in different in vitro assays [10, 14]. We have previously shown that decidua produces a protease that can cleave IGFBP-1 into fragments incapable of binding IGF [15]. Partial characterization by substrate zymography and inhibitor profile studies have suggested the involvement of a matrix metalloproteinase (MMP). Several members of this enzyme family are present at the fetal/maternal interface [16, 17], and MMPs and their tissue inhibitors (TIMPs) are important for controlling trophoblast migration through their effect on extracellular matrix [18–21]; in the present study, we investigate whether MMPs may also affect placental development through proteolysis of IGFBP-1.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Samples of normal human first-trimester decidua parietalis (8–12 wk of gestation) were obtained from a total of eight women undergoing elective surgical termination of pregnancy. Each woman gave informed consent, and the collection of tissue was approved by the local ethical committee. All traces of villous material were removed with the aid of a dissecting microscope and then, in accordance with our previously described protocol [22], the PBS-washed decidual tissue was incubated, with agitation, at 37°C with 0.2% hyaluronidase and 0.25% collagenase (Sigma, Dorset, U.K.) for 2 h. The resulting suspension was filtered initially through a 100-µm nylon sieve to remove undigested tissue fragments and then through a 40-µm sieve to retain whole glands and cell aggregates as previously shown for endometrium. Cells were then resuspended in 25% Percoll (Pharmacia, Uppsala, Sweden) and layered over 60% Percoll. After centrifugation at 670 x g for 30 min, cells at the 25%/60% interface were collected, washed three times in PBS, counted, and plated at 0.5 x 106 cells/well in six-well plates. Cells were then maintained for 96 h in Dulbecco modified Eagle medium, 10% fetal calf serum, 100 µg/ml of streptomycin, 100 U/ml of penicillin, and 100 ng/ml of medroxyprogesterone acetate at 37°C in 5% CO2.
Western Ligand and Immunoblot Analysis
Samples (conditioned medium or recombinant peptides) were subjected to SDS-PAGE (10%) and transferred to nitrocellulose. Membranes for Western immunoblot analysis were blocked with 3% BSA/PBS for 1 h and then probed with a monoclonal antibody to IGFBP-1 (0.5 µg/ml; 6303; a kind gift of Medix Biochemica, Kauniainen, Finland) or rabbit polyclonal antibodies to MMP-2, MMP-3, MMP-9, TIMP-1, TIMP-2, or TIMP-3 (1 µg/ml; Chemicon, Hampshire, U.K.) or 2-macroglobulin (0.68 µg/ml; Sigma) overnight at 4°C. Blots were then incubated with either anti-rabbit or anti-mouse horseradish peroxidase-linked antibodies (1:3000 dilution; Amersham, Buckinghamshire, U.K.) for 1 h at room temperature. Immunoreactivity was visualized by staining with diaminobenzidine (Sigma).
Ligand blots were incubated in 0.5% sodium azide, 1% Nonidet-P40, and PBS for 30 min at 4°C; blocked with 1% BSA, 0.15 M NaCl, and 0.5% Tween for 1 h at room temperature; and then probed with [125I]IGF-I (150 000 cpm/ml) for 3 h. Membranes were then washed with 0.15 M NaCl and exposed to film for 72 h at –70°C.
Immunocytochemistry
Primary decidual cells were plated into eight-well chamber slides and cultured for as long as 4 days before fixing in methanol for 30 min at 4°C. Slides were incubated with DAKO (Cambridgeshire, U.K.) protein block for 30 min and then with PBS or the antibodies against IGFBP-1, MMP-2, MMP-3, or MMP-9 (1 µg/ml) or 2-macroglobulin (13.6 µg/ml) for 2 h at room temperature. Controls included comparable concentrations of alternative primary antibodies (mouse monoclonal antibody to CD23 [MHM6; DAKO] or mouse monoclonal immunoglobulin G1 of unknown specificity [DAKO]) or serum from an unimmunized rabbit, and all were negative. Slides were probed with fluorescein-conjugated secondary antibodies, and nuclei were stained using propidium iodide. Images were collected on a Bio-Rad laser confocal microscope (Bio-Rad Laboratories, Hercules, CA). The cells were plated at high density and grew as a multilayer; fluorescence emission is detected from nuclei that are not centered in the optical plane. This accounts for the apparent variation in nuclear brightness visible in the micrographs presented in Figure 1.
|
IGFBP-1 Proteolysis
Proteolysis of IGFBP-1 was achieved using an adaptation of our previously published protocol [15]. Briefly, recombinant human IGFBP-1 (0.5–10 µg; a kind gift of Genentech, South San Francisco, CA) was incubated with medium conditioned by human decidual cells (1 mg of protein) or recombinant MMP-3 or MMP-9 (1 µg; Chemicon) in PBS and 0.5 mM CaCl2 for 18 h at 37°C. The reaction was stopped by the addition of gel loading buffer and 5-min incubation at 100°C, and the samples were subjected to SDS electrophoresis (15% resolving gel) and IGFBP-1 Western blot analysis as described above. In some experiments, an inhibitor (TIMP-1, TIMP-2, or 2-macroglobulin) was also included in the reaction mixture at a 1:1 molar ratio with enzyme.
Analysis of IGFBP-1 Fragments by N-Terminal Microsequencing
Fragments of IGFBP-1 were generated using the method described above. Samples were loaded onto a degassed SDS polyacrylamide (10%) gel and separated using a Tris/tricine buffer containing the free-radical scavenger sodium thioglycolate (0.1 mM) to reduce N-terminal blocking. Proteins were transferred to polyvinylidene difluoride membranes by electroblotting in 10 mM 3-[cyclohexylamino]-1-propanesulfonic acid and 10% MeOH (pH 11.0), stained with Coomassie blue (0.1% in 1% acetic acid and 40% MeOH), allowed to air-dry, and then cut from the membrane for sequencing using an Applied Biosystems 476A protein sequencer (Applied Biosystems, Foster City, CA). Fragments were generated using medium conditioned by two different decidual cell preparations, and each fragment was sequenced in duplicate.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Decidual cells isolated from tissue obtained at termination of pregnancy during the first trimester were characterized by immunocytochemistry for IGFBP-1 (Fig. 1A), which is a major secretory product. The IGFBP-1 staining appeared as an array of streaks that resolved at higher magnification (not shown) into spots apparently representing a vesicular compartment in the decidual cell secretory pathway. Our earlier work demonstrated that these cells produce a protease capable of cleaving IGFBP-1 and that this enzyme may belong to the family of MMPs [15]. Decidual cells and conditioned medium were screened for the presence of MMP-2, MMP-3, and MMP-9, because these have a molecular mass coinciding with the enzymes identified by substrate zymography. Figure 1 demonstrates that both cells and conditioned medium were immunoreactive when an antibody specific for MMP-3 was used. The distribution of MMP-3 was clearly distinct from that observed for IGFBP-1, localizing to more closely clustered groups of intracellular vesicles. Figure 1B also shows that in addition to the active form of MMP-3 (45 kDa), proforms (59/57 kDa) are also present in decidual conditioned medium. We were unable to detect MMP-2 or MMP-9 by either immunocytochemistry or Western immunoblot analysis (Fig. 1B).
MMP-3 and MMP-9 Can Proteolyze IGFBP-1
Incubation of recombinant IGFBP-1 with MMP-3 resulted in fragments with apparent molecular masses of 21, 17.5, 14.5, 12.5, and 9 kDa (Fig. 2A). These fragments displayed a migration profile similar to those observed when IGFBP-1 was incubated with medium conditioned by decidual cells (Fig. 2A). Although we were unable to demonstrate decidual cell production of MMP-9, we investigated whether this enzyme could also proteolyze IGFBP-1, both because others have reported its presence at the maternal/fetal interface [23] and because Figure 2A shows cleavage generating a series of peptide products with molecular masses similar to those generated by MMP-3 proteolysis. None of the proteolytic fragments was capable of binding [125I] IGF-I (Fig. 2B).
|
IGFBP-1 Proteolysis Is Regulated by Inhibitors
Medium conditioned by first-trimester decidual cells was screened by Western immunoblot analysis for the presence of TIMP-1, TIMP-2, and TIMP-3, and the results are depicted in Figure 3A. We identified both TIMP-1 and TIMP-2, but TIMP-3 was not apparent (data not shown).
|
We have recently shown that IGFBP-1 in plasma can be protected from proteolysis by association with the homotetrameric glycoprotein 2-macroglobulin [24]. Using immunohistochemistry and Western immunoblot analysis (Fig. 3B), we have found that 2-macroglobulin is abundantly produced by human first-trimester decidual cells. Moreover, 2-macroglobulin attenuates the proteolysis of IGFBP-1 by MMP-3 (Fig. 3C).
The MMP-3 cleavage of IGFBP-1 was also inhibited by TIMP-2, resulting in decreased production of the lower-molecular-mass proteolytic fragments (Fig. 3C). Activity of MMP-9 was reduced by TIMP-2, but no evidence of significant inhibition by TIMP-1 or 2-macroglobulin was found.
Identification of Cleavage Site
Peptides generated by proteolysis of IGFBP-1 with MMP-3, MMP-9, or decidual cell-conditioned medium were analyzed by N-terminal protein sequencing to identify cleavage sites. The majority of fragments (21, 17.5, 14.5, and 12.5 kDa) had the same N-terminal sequence, which was identical to that of intact IGFBP-1. However, the 9-kDa peptide had the N-terminal sequence KWKEPCRIEL, which corresponds to residues 146–155 of IGFBP-1 (Fig. 4), indicating that both MMP-3 and MMP-9 cleave IGFBP-1 at the 145Lys/Lys146 bond. Cleavage of IGFBP-1 at this site would be predicted to produce peptides of 21 and 9 kDa, corresponding with the apparent size of two of the fragments on SDS-PAGE/immunoblot analysis. The data suggest that the N-terminal, 21-kDa polypeptide is subject to further C-terminal proteolysis to produce peptides of 17.5, 14.5, and 12.5 kDa.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Our previous work [15] demonstrated that medium conditioned by first-trimester human decidual cells contains a protease capable of degrading IGFBP-1. Substrate zymography and inhibitor characterization suggested MMP activity, and in the present study, we have identified two enzymes, MMP-3 and MMP-9, that generate IGFBP-1 fragments of 21 and 9 kDa in a manner remarkably similar to the activity found in decidual cell-conditioned medium.
In other cell systems, MMPs have been shown to cleave IGFBPs [25, 26], and ADAM-12, a disintegrin metalloprotease produced by trophoblast, is reported to have activity against IGFBP-3 [27, 28]. To our knowledge, however, this is the first demonstration that MMPs are involved in the regulation of IGFBP-1 at the maternal/fetal interface. This phenomenon may have been undetected in other studies because of the widespread use of IGFBP-1 from amniotic fluid, which also contains high levels of 2-macroglobulin (data not shown). We have recently shown that IGFBP-1 can bind 2-macroglobulin and is thus protected from proteolysis [24]. In the present study, we report that 2-macroglobulin is produced by human decidual cells and that this protease inhibitor is also effective in protecting IGFBP-1 against MMP-3. Mice lacking 2-macroglobulin and its close relative, murinoglobulin-1, show anomalously deep trophoblast invasion in midgestation and smaller pups, suggesting that disturbing the balance between proteolysis and its inhibition and growth factor bioavailability at the maternal/fetal interface has important effects on both placenta and fetus [29].
Proteolysis is thought to be a mechanism for regulating the bioactivity of IGFBPs, because cleavage of other binding proteins results in fragments with functional properties that differ from those of the intact protein [30]. We have found that proteolysis of IGFBP-1 is also of physiological significance, because the fragments are unable to bind ligand [15], which may have consequences for IGF bioavailability as well as for the effect of IGFBP-1 on trophoblast migration.
Overexpression of decidual IGFBP-1 in mice produces significant changes in placental structure, suggesting altered allocation of trophoblast to different lineages during development. This provides clear evidence in support of a role for maternal IGFBP-1 in regulation of trophoblast differentiation [31]. IGFBP-1 was first suggested as a regulator of cell motility by Jones et al. [32], who demonstrated that an RGD sequence-dependent interaction with 5ß1 integrin stimulates migration of Chinese hamster ovary cells. The role of IGFBP-1 in regulating human trophoblast migration, however, is uncertain. Enhanced migration and invasion of a transformed trophoblast cell line are observed in response to IGFBP-1 [9], and this is dependent on the interaction of RGD and 5ß1 integrin and on subsequent activation of focal adhesion kinase and mitogen-activated protein kinase pathways [33]. In contrast, experiments with primary cultured trophoblasts have given rise to a hypothesis that trophoblast migration may be impeded by the paracrine production of IGFBP-1 in the decidualized endometrium [14].
We now suggest that these seemingly conflicting data could be explained by IGFBP-1 proteolysis. The 9-kDa, C-terminal fragment of IGFBP-1 retains the RGD sequence, and as a result of cleavage, conformational change in the region of this motif might alter integrin affinity relative to the intact molecule, enabling IGFBP-1 to behave as either an agonist or an antagonist in a manner analogous to the switch caused by changes in RGD peptide concentration [34]. Indeed, recent work using immobilized RGD sequences has demonstrated that flanking residues are important in determining binding affinity for integrins and, in consequence, specific inhibitory activity in cell attachment and spreading assays [35]. The 5ß1 integrin is acquired by cytotrophoblasts differentiating down the invasive pathway [36], and the importance of this integrin for the control of trophoblast migration is well established [37].
The MMPs are also necessary for cell motility and invasion by virtue of their ability, directly or indirectly, to degrade extracellular matrix. We now propose that an additional mode of action might include proteolysis of IGFBP-1. Others have demonstrated MMP-3 and MMP-9 in maternal decidua during implantation in rodent pregnancies [38, 39], and MMP-9 has been immunolocalized in decidua obtained during the first trimester [40]. In the present study, we observed production of MMP-3 by first-trimester decidual cells; however, MMP-2 and MMP-9 were not apparent. Both MMP-3 and MMP-9 are also produced by cytotrophoblast cells [23], and their temporal and spatial localization identifies them as possible regulators of IGFBP-1 activity [41, 42]. This may well be a reciprocal interaction, because IGFBP-1 is reported to increase the gelatinolytic activity of cytotrophoblasts [43], probably as a result of stimulating MMP-3 [44]. However, IGFBP-1 has also been shown to increase trophoblast secretion of TIMP-1 [43], one of the known inhibitors of MMP activity. In accordance with previous in vivo studies, we identified decidual cell production of both TIMP-1 and TIMP-2 [45– 47]—although TIMP-3 was not apparent—and we demonstrate that like 2-macroglobulin, TIMPs, and particularly TIMP-2, attenuate the proteolysis of IGFBP-1 by MMP-3.
The effect of IGFBP-1 at the maternal/fetal interface will therefore be dependent on the balance between IGFBP-1 production, MMP activation, and the presence and activity of the inhibitors 2-macroglobulin and TIMPs in the varying microenvironments of the interacting cells (Fig. 5). Altering the interaction between the components involved in regulating trophoblast migration could compromise placental development and, hence, fetal growth.
|
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Correspondence: Melissa Westwood, Endocrine Sciences, University of Manchester, Stopford Building, Oxford Road, Manchester, M13 9PT, U.K. FAX: 44 161 275 5958; melissa.westwood{at}man.ac.uk
Received: 11 September 2003.
First decision: 29 September 2003.
Accepted: 15 March 2004.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
5ß1 integrin and inhibits cytotrophoblast invasion into decidualized endometrial stromal cultures. Growth Horm IGF Res 1998 8:21-31[Medline]2-Macroglobulin: A new component in the insulin-like growth factor/ insulin-like growth factor binding protein-1 axis. J Biol Chem 2001 276 41668-416742-Macroglobulin controls trophoblast positioning in mouse implantation sites. Placenta 2003 24:912-921[CrossRef][Medline]5ß1 integrin by means of its Arg-Gly-Asp sequence. Proc Natl Acad Sci U S A 1993 90:10553-105575ß1 integrin via mitogen activated protein kinase pathway. J Clin Endocrinol Metab 2001 86:2484-24935ß3 by low antagonist concentrations. J Cell Sci 2001 114:1545-1553[Abstract]5ß1. Biol Reprod 1999 60:828-838This article has been cited by other articles:
|
|
 |
|
  G. S. Butler, R. A. Dean, E. M. Tam, and C. M. Overall Pharmacoproteomics of a Metalloproteinase Hydroxamate Inhibitor in Breast Cancer Cells: Dynamics of Membrane Type 1 Matrix Metalloproteinase-Mediated Membrane Protein Shedding Mol. Cell. Biol., August 1, 2008; 28(15): 4896 - 4914. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Minici, F. Tiberi, A. Tropea, M. Orlando, M. F. Gangale, F. Romani, S. Campo, A. Bompiani, A. Lanzone, and R. Apa Endometriosis and human infertility: a new investigation into the role of eutopic endometrium Hum. Reprod., March 1, 2008; 23(3): 530 - 537. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. V. Cockle, N. Gopichandran, J. J. Walker, M. I. Levene, and N. M. Orsi Matrix Metalloproteinases and Their Tissue Inhibitors in Preterm Perinatal Complications Reproductive Sciences, October 1, 2007; 14(7): 629 - 645. [Abstract] [PDF]
|
|
|
|
|
|
W.-H. Zhou, M.-R. Du, L. Dong, X.-Y. Zhu, J.-Y. Yang, Y.-Y. He, and D.-J. Li Cyclosporin A increases expression of matrix metalloproteinase 9 and 2 and invasiveness in vitro of the first-trimester human trophoblast cells via the mitogen-activated protein kinase pathway Hum. Reprod., October 1, 2007; 22(10): 2743 - 2750. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Piccard, P. E. Van den Steen, and G. Opdenakker Hemopexin domains as multifunctional liganding modules in matrix metalloproteinases and other proteins J. Leukoc. Biol., April 1, 2007; 81(4): 870 - 892. [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]
|
|
|
|
 |
|
 M. Ii, H. Yamamoto, Y. Adachi, Y. Maruyama, and Y. Shinomura Role of Matrix Metalloproteinase-7 (Matrilysin) in Human Cancer Invasion, Apoptosis, Growth, and Angiogenesis Experimental Biology and Medicine, January 1, 2006; 231(1): 20 - 27. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Kabir-Salmani, Y. Shimizu, K. Sakai, and M. Iwashita Posttranslational modifications of decidual IGFBP-1 by steroid hormones in vitro Mol. Hum. Reprod., September 1, 2005; 11(9): 667 - 671. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Sala, S. Capaldi, M. Campagnoli, B. Faggion, S. Labo, M. Perduca, A. Romano, M. E. Carrizo, M. Valli, L. Visai, et al. Structure and Properties of the C-terminal Domain of Insulin-like Growth Factor-binding Protein-1 Isolated from Human Amniotic Fluid J. Biol. Chem., August 19, 2005; 280(33): 29812 - 29819. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Kaushal, A. H. Heald, K. W. Siddals, M. S. Sandhu, D. B. Dunger, J. M. Gibson, and N. J. Wareham The Impact of Abnormalities in IGF and Inflammatory Systems on the Metabolic Syndrome Diabetes Care, November 1, 2004; 27(11): 2682 - 2688. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) and the 