Decidual Differentiation of Stromal Cells Promotes Proprotein Convertase 5/6 Expression and Lefty Processing
http://www.100md.com
《内分泌学杂志》
Department of Obstetrics and Gynecology (M.T., A.M., S.T.), Stony Brook University, Stony Brook, New York 11794
Laboratory for Molecular Cell Biology (I.P., J.W.M.C.), Department for Human Genetics, University of Leuven and Flanders Interuniversity Institute for Biotechnology, B-3000 Leuven, Belgium
Department of Gynecology and Obstetrics (L.C.G., S.T.), Stanford University Medical Center, Stanford, California 94305
Department of Obstetrics and Gynecology (A.T.F.), University of Illinois at Chicago, Chicago, Illinois 60612
Department of Biological Sciences (D.D.C.), University of Delaware, Newark, Delaware 19707
Department of Laboratory Pathology and Medicine (D.G.K., C.B.), Division of Reproductive Endocrinology and Infertility, University of North Carolina, Chapel Hill, North Carolina 27599
Abstract
Lefty/Ebaf polypeptides, novel members of the TGF- superfamily, are involved in endometrial differentiation and embryo implantation. Recently, we showed that, during undisturbed estrous cycle, lefty is present in mouse uterine horn primarily in a precursor form. Here, we show that decidual differentiation of endometrial stroma leads to increased lefty (3.1- to 3.6-fold in vivo and 5- to 8-fold in vitro) and processing of its precursor primarily to its long form. This event occurs on d 5 of pregnancy, and is paralleled by proprotein convertase (PC)5/6 up-regulation (6-fold increase for PC5A and 3-fold increase for PC5B) in decidualized uterine horn, independent of embryo implantation. Among the known convertases, only PC5/6A processes lefty to its long form. Taken together, the findings show that decidualized differentiation of stroma, which is a prerequisite for embryo implantation, leads to processing of lefty by PC5/6A.
Introduction
LEFTY/EBAF POLYPEPTIDES, novel members of the TGF- superfamily, are involved in the formation of the embryonic lateral patterning and endometrial differentiation (1, 2). Members of the TGF- superfamily require processing for their activation, suggesting that cleavage is an essential step for lefty activation. Transfection of different cell lines with human lefty A and B form resulted in expression of a 42-kDa protein, which was proteolytically processed to release two polypeptides of 34 kDa (long form) and 28 kDa (short form) (3). The mutation of the consensus sequences for proprotein convertase (PC) cleavage in lefty protein identified and validated the lefty cleavage sites. The mutation of the sequence RGKR to GGKG (amino acids 74–77) and RHGR to GHGR (amino acids 132–135) prevented the proteolytic processing of lefty precursor to the 34- and 28-kDa forms, respectively (4). In pluripotent P19 cells, lefty did not induce Smad2 or Smad5 phosphorylation, Smad2/Smad4 heterodimerization, or nuclear translocation of the Smad2 or Smad4, but activated MAPK pathway in a time- and dose-dependent fashion (4). The 28-kDa, but not the 34-kDa, polypeptide induced MAPK activity. These data supported a molecular model of processing as a mechanism for regulation of lefty action(s).
In humans, lefty is expressed in endometrial stromal cells that gradually undergo the differentiated state of decidualization (2). Decidual cells adopt unique phenotypic, functional, biological, secretory, and biosynthetic features. Decidualization is promoted by the cAMP and protein kinase A pathways as well as, after an initial estradiol priming, by progesterone (P). Recent studies indicate that this process is associated with exquisitely controlled sequential gene regulation including kinetic reprogramming of genes associated with decreased G protein signaling, increased STAT pathway signaling, cell proliferation, structural protein changes, cellular differentiation, and secretory processes (5, 6). Among these changes, expression and secretion of prolactin and IGBP-1 have become the hallmarks of this differentiated cellular state. Decidualized cells produce extracellular matrix proteins including laminin, type IV collagen, fibronectin, and heparin sulfate proteoglycan. The kinetic reprogramming, as revealed by microarray analysis of human endometrial stromal cells forced to differentiate in vitro by cAMP, includes remarkable elevation in the expression of lefty (5, 6). The expression of lefty is increased from a low basal level by 3-fold within 2 h, and 123-fold within 36 h of receiving the decidualizing stimulus (6). Brar et al. validated the microarray data with Northern blot analysis and showed a steady increase of lefty over a period of 12 d when the decidual cells are fully developed (5).
Taken together, these findings suggest that lefty might be an integral component of the decidualization process. We were interested in extending these observations to a model system more amenable to experimental manipulation in vivo. Therefore, the current study was carried out to define precisely the expression and processing of lefty during decidualization. The results show that induction of decidualization leads to the conversion of lefty into its long form. This event, which occurs upon decidualization of stromal cells, is paralleled with the expression of PC5/6, the primary enzyme for proteolytic cleavage of the precursor form of lefty to its long form (4). These findings show that lefty expression and processing are intrinsic features of the differentiated decidualized state of stromal cells.
Materials and Methods
Materials
Chemicals were from Sigma-Aldrich (St. Louis, MO) or Fisher Scientific (Pittsburgh, PA). The avidin-biotin-peroxidase kit was from Vector Laboratories (Burlingame, CA). The goat polyclonal antibody to lefty peptide (M-20) mapped to the carboxyl terminus of mouse lefty and recombinant IGF binding protein 1 (IGFBP-1) were from Santa Cruz Biotechnology Inc. (Santa Cruz, CA). The recombinant lefty was from R&D Systems (Minneapolis, MN). The antibody to actin was from Abcam, Inc. (Cambridge, MA). Rabbit antibody to PC5/6 was from Alexis Biochemicals (Laüfelfingen, Switzerland). Outbred CF-1 and CD-1 mice were obtained from Charles River Laboratory (Wilmington, MA).
Sample preparation and analysis of estrous cycle and pregnancy
In this report, the expression of mouse lefty was examined in the endometria of mice (n = 6) during pregnancy. Each experiment was repeated at least twice to confirm reproducibility of data. Day 1 of pregnancy was considered as the day when the vaginal plug was detected. Approval of the institutional Animal Care and Use Committee of the University of Delaware was obtained for carrying out the removal of uterine horns during pregnancy.
Transfection
CHO cells were maintained in DMEM (Life Technologies, Inc., Rockville, MD) supplemented with 10% (vol/vol) fetal bovine serum (FBS; Life Technologies, Inc.) and 1% (wt/vol) antibiotic-antimycotic mixture (Life Technologies, Inc.). For transfection, cells were trypsinized with 0.05% (wt/vol) trypsin-EDTA (Life Technologies, Inc.) and seeded into six-well plates (Falcon, Franklin lakes, NJ) at a concentration of 2 x 105 cells per well. Cells were cultured in 1.5 ml of the culture medium, in the presence of CO2 at 37 C for about 16 h. Upon reaching 60% confluency, cells were cotransfected with cDNA of lefty A and with furin, PACE4, PC5/6A, PC5/6B, or PC7 cDNA, using Superfect Transfect Reagent (Qiagen, Valencia, CA) following the manufacturer’s protocol. In separate experiments, 293 cells were transfected with an expression PC5/6 vector. Empty vector served as control in these transfections. Lysate from these cells was used as a control for identification of PC5/6 on Western blotting. In transient transfections, serum-free medium was used.
Induction of decidualization in mouse uterine horn
Decidualization was induced in ovariectomized mice 4–5 d after ovariectomy. Mice received three daily sc injections of 17- estradiol (E2, 100 ng). After 4 d of rest, animals received three additional daily sc injections of E2 (10 ng) as well as sc injections of P (1 mg). Both E2 and P were delivered in 0.1 ml of sesame oil. On the third day of treatment with P, animals received the decidualizing signal, which was either trauma or oil. After anesthesia, an incision was made in the left flank and the left uterine horn was exposed. Trauma was produced by the insertion of a blunt 25-gauge needle seven times into the horn lumen. For decidualization by oil, 20 μl of sesame oil was injected into the proximal left uterine horn at the tubal junction. Animals received three additional daily sc injections of P (1 mg). Seventy-two hours later, both uterine horns were removed. The approval of the institutional Animal Care and Use Committee of Stony Brook University was obtained for carrying out the experiments on decidualization of mouse uterine horns.
Culture and decidualization of human endometrial stromal cells
HuF were isolated from decidua parietalis dissected from the placental membranes after normal vaginal delivery at term under an approved protocol by the University of Illinois at Chicago Committee on the Use of Human Subjects in Medical Research. Briefly, scraped cells were digested in 0.1% collagenase, 0.02% deoxynuclease in calcium- and magnesium-free Hanks’ balanced salt solution. Cells were plated in four 100-mm culture dishes (Becton Dickinson and Co. Labware, Franklin Lakes, NJ) and placed into an incubator at 37 C in presence of 5% CO2. The next day, plates were extensively washed with PBS to remove nonadherent (mainly decidual) cells. At confluence, cells were trypsinized and used for experiments in passage number 3–5. Cell purity was assessed by immunocytochemistry using antibodies against cytokeratin (Dako Corp., Carpenteria, CA) and vimentin (Zymed Laboratories, Inc., San Francisco, CA). The purity of the fibroblast cell preparations used was more than 95%. These cells were treated with medium alone and with medium supplemented with hormones [10–6 M medroxy-progesterone acetate (MPA) + 36 nM E2] or with cAMP (1 mM) for 14 d. The medium was comprised of DMEM supplemented with 2% (wt/vol) charcoal-stripped FBS serum and was changed every 2 d. FBS was omitted 48 h before the end of treatment. A total of 4 x 106 SHT290 cells (immortalized human endometrial stromal cells) was treated with medium and with medium supplemented with 10–6 M MPA and 10–8 M E2, or with cAMP (1 mM) for 13 d. The medium was comprised of DMEM, with high glucose, supplemented with 2% charcoal-stripped FBS and epidermal growth factor (20 ng/ml).
Human endometrial samples were collected from women with normal menstrual cycles after obtaining informed consent under an approved protocol by the Stanford University Committee on the Use of Human Subjects in Medical Research. Biopsies were obtained from subjects, 26–46 yr of age, who had regular menstrual cycles (28–35 d), were documented not to be pregnant, and had no history of endometriosis. Endometrial tissues were subjected to collagenase/hyaluronidase (Sigma-Aldrich) digestion for 2 h at 37 C. After digestion, the stroma was dispersed, although the glandular structures remained mostly intact. Stromal cells were separated from glands on a size basis. Stromal cells were centrifuged, and the resulting pellet was resuspended in a 4:1 ratio in DMEM-F12 (Life Technologies, Inc.) and DMEM/10% FBS. Cells were preplated in 10 cm2 standard culture plates for 1 h at 37 C in the presence of 9% CO2, and the medium was replaced with high-glucose DMEM/MCDB-105 medium with 10% charcoal-stripped FBS, insulin (5 μg/ml) (Sigma-Aldrich), gentamycin, penicillin, and streptomycin. Endometrial stromal cells were passaged two to three times before the experiments. The stromal cells used were 99% pure, devoid of lymphocytes, endothelial cells, and fibroblasts (5). Culture medium was changed every 2–3 d. For induction of decidualization, cultured stromal cells were pretreated with E2 (10 μM), and, once confluent, they were treated in serum-free media supplemented with 1 μM 8-bromo cAMP for 72 h. The concentration of IGFBP-1 was determined in conditioned media after cAMP treatment using an IGFBP-1 ELISA (DSL, Webster, TX).
Northern blot analysis
Total RNA (20 μg/lane) or poly A RNA (2 μg/lane) was fractionated in formaldehyde-agarose gel. After diffusion transfer to nitrocellulose filters and cross-linking (Stratalinker; Stratagene, La Jolla, CA), hybridizations were performed using denatured 32P-labeled cRNA of lefty, and cDNA probes for mouse fur, PC5/6, PACE4, and PC7 were used as described previously (12). Gels were stained with ethidium bromide for revealing the 18S and 28S ribosomal RNA and blots were stripped and probed for actin cDNA for normalization. Quantification was performed using Sigma Gel (Sigma-Aldrich) as well as the Typhoon 9400 (GE Healthcare) and ImageQuant software (Molecular Dynamics).
SDS-PAGE and Western blotting
Media were centrifuged at 7000 x g for 3 h and concentrated by about 12-fold using Centricon 10 (protein molecular size cut-off: 10,000 kDa; Amicon, Danvers, MA). The concentration of proteins in these samples was determined by Bio-Rad Protein Assay kit (Bio-Rad Laboratories, Hercules, CA). Concentrated conditioned media or cell lysates (50 μg protein per lane) were fractionated in a 12% denaturing gel together with prestained protein ladder (Life Technologies, Inc.) and were subsequently blotted onto nitrocellulose membrane in a Mini-Trans-Blot apparatus (Bio-Rad Laboratories). Blots were stained with the goat polyclonal antibody to lefty peptide (M-20) (1–2 μg/ml) or where indicated with rabbit antibody to PC5/6. The secondary antibody used was donkey antigoat IgG-HRP or goat antirabbit (Santa Cruz Biotechnology). Bands were detected by chemiluminescence as described by the manufacturer. Quantification was performed using Sigma Gel (Sigma-Aldrich).
Statistical analysis
Density of bands and ratios of precursor/long form of lefty were compared using t test. Differences were considered statistically significant when P < 0.05.
Results
Lefty proteins in mouse uterine horn during pregnancy
Transfection of different cell lines with human lefty A and B form resulted in expression of a 42-kDa protein, which was proteolytically processed to release two polypeptides of 34 kDa (long form) and 28 kDa (short form) (4). To examine for the presence of similar processed forms in mouse uterine horn, proteins from horns of pregnant mice were subjected to Western blot analysis for lefty. On d 1 of pregnancy, precursor and long form of lefty were equally present. However, on d 3, and particularly on d 5, of pregnancy, there was more long form of lefty compared with the precursor form (Fig. 1). On d 9, this was more notable close to the implantation site rather than in tissues intervening the implantation sites (Fig. 1), suggesting that processing of lefty occurs in decidualized stroma more efficiently around the implanting embryo.
Lefty in mouse endometrium during artificial decidualization
To determine whether lefty processing was due to the decidualization of stroma and whether it required presence of an embryo, decidualization was artificially induced in endometrium. We introduced oil, as the deciduogenic signal, into the left uterine horns of mice primed with steroid hormones, E2 and P. The left uterine horns, removed 3 d after receiving the decidualizing signal, were enlarged, and the distal parts of some right uterine horns were also enlarged, suggesting transfer of oil from the left to the right horn (Fig. 2). The introduction of Chicago blue dye has substantiated this trans-uterine horn flow (7). For this reason, the uterine horns of mice primed with hormones that did not receive the decidualizing signal, rather than the right horn of animals that received the deciduogenic stimulus, were used as controls. Trauma was used as a second deciduogenic signal in the left uterine horn. The left horns removed 3 d later were also enlarged (Fig. 2). The wet weight of the uterine horns injected with oil was significantly higher than the weights of horns that were traumatized (data not shown). The sections of uterine horns examined after hematoxylin and eosin staining showed the typical morphology of a decidualized stroma (data not shown). Western blot analysis for IGFBP-1 was carried out to confirm the decidualization of stroma in the uterine horns. IGFBP-1 was found in both uterine horns, but unless both horns were enlarged to the same extent, more IGFBP-1 immunoreactivity was present in the left compared with the right horns (Fig. 2).
Northern blot analysis showed both 2.4- and 1.5-kb bands of lefty mRNA in mouse uterine horns of both nondecidualized control and decidualized horns. As compared with control uterine horns, lefty was expressed at a significantly higher level (3.1- to 3.6-fold) in the decidualized horns irrespective of the type of the decidualizing signal (Fig. 3A). Although both precursor and long forms of lefty were present in the nondecidualized uterine horns primed with hormones E2 and P, the long form was the predominant form in the decidualized horns irrespective of the deciduogenic signal (Fig. 3B).
Expression of convertases in mouse uterine horn during artificial decidualization
Because members of the PC family cleave different TGF- proteins, the data suggested that processing of lefty might be due to expression of a convertase in the decidualized stroma. For this reason, Northern blot analysis for the relevant convertases including Furin, PC5/6, PC7, and PACE4, were carried out on the RNA isolated from the decidualized and control nondecidualized mouse uterine horns. Among these convertases, only the expression of PC5/6 was significantly increased in the decidualized horns (Fig. 4A). Both deciduogenic signals increased the expression of PC5/6 and both the A (short form) and B (long form) isoforms were significantly increased (6-fold increase for PC5A and 3-fold increase for PC5B). Western blot analysis was performed on the tissue lysates of the decidualized and nondecidualized uterine horn to validate increased PC5/6. The lysate of 293 cells transfected with PC5/6 served as a positive control for the identification of PC5/6, whereas the lysate from 293 cells transfected with empty vector served as a negative control. As shown in Fig. 4B, PC5/6 was increased in the decidualized uterine horn, whereas it was low to undetectable in control hormone-primed nondecidualized uterine horn.
Processing of lefty by PC5/6A
These findings suggested that lefty was being processed in the decidualized stroma by PC5/6. To directly show that PC5/6 was responsible for lefty processing, CHO cells were transfected simultaneously with lefty cDNA and cDNAs of various PCs including Furin, PACE4, PC5/6A, PC5/6B, and PC7. From these convertases only the PC5/6 was able to convert lefty to its long form. PC5/6A was more efficient than PC5/6B in the processing of lefty to its long form (Fig. 5).
Processing of lefty in endometrial stromal cells decidualized in vitro
To directly show that lefty was processed in decidualized cells irrespective of other in vivo signals, decidualization was induced in HuFs cells. These cells were derived from human placenta and have been shown to decidualize in vitro both with cAMP and with E2/P treatment (8). Both cAMP and E2/P led to the decidualization of these cells as confirmed by release of IGFBP-1 into the culture medium (Fig. 6A). Both decidualizing signals led to the accumulation of the long form of lefty in the cell lysate. However, the long form of lefty was present in the culture medium of cells decidualized with cAMP but not with E2/P (Fig. 6A). To further validate these findings, SHT290 cells, which were derived from human endometrial stromal cells and immortalized by transfection with telomerase were used. SHT290 stromal cells were very similar to the parental strain from which they were derived according to criteria of proliferation, karyotype, cellular localization of cytoskeletal markers and nuclear staining, and basal gene expression based on microarray analysis (9). These cells were shown to decidualize with E2/P treatment (9). These cells were decidualized in vitro with E2/P and with cAMP. Untreated cells served as controls. In these cells, only E2/P and not cAMP led to the processing of lefty (Fig. 6B). However, neither treatment led to the release of lefty to the culture medium. This suggested that culture condition or transfection with telomerase might have hampered the ability of these cells to respond to cAMP.
To show that lefty was also processed in response to cAMP, and was released into culture medium, primary cultures of human endometrial stromal cells derived from five different subjects were decidualized with cAMP. Northern blot analysis of these cells showed significant increase (5- to 8-fold) in lefty expression within the short duration of treatment (Fig. 7A). The quantity of the IGFBP-1 was determined in the culture media of these cells by ELISA. IGFBP-1 was present in the culture media of decidualized cells and not the culture media of untreated cells (Fig. 7B). Western blot analysis of culture media of nondecidualized cells showed the precursor form of lefty (Fig. 7B). On the other hand, decidualization of stromal cells led to accumulation of both long and short forms of lefty in the culture media (Fig. 7B).
Discussion
We previously showed that lefty was expressed in human endometrium at a high level immediately before and during menstrual bleeding (2). In situ hybridization showed that lefty was expressed in cells that were undergoing decidualization, a requisite for successful implantation (2). In human endometrium, decidualization occurs in the absence of the embryo. However, in mice, decidualization occurs only in the presence of an embryo or by application of a decidualizing signal to a properly primed endometrium. During pregnancy, by the time the embryo arrives into the mouse uterine cavity on d 4 and by d 5, there is evidence of decidualization of stroma. Here, we show that this decidualization is associated with an increase in the long form of lefty in mouse endometrium. However, presence of the embryo is not required for this event, because lefty is efficiently processed to its long form when uterine horn is artificially forced to decidualize with oil or trauma. Although both precursor and long forms of lefty are increased with hormonal priming alone, decidualization appears to be important for the efficient conversion of lefty to its long form.
Decidualization of mouse uterine horn as well as decidualization of various human endometrial stromal cells leads to increased lefty expression. These findings are consistent with the microarray data and Northern blot analysis, which show a steady level of increase of lefty as the decidual cells differentiate (5, 6). In vitro, cAMP is more efficient than E2 + P in the induction of decidualization in lefty expression in primary cultures of human endometrial stromal cells. Treatment of endometrial stromal cells with cAMP induces decidualization of cells, IGFBP-1 release into the culture medium, and lefty processing within 72 h; longer treatment with E2/P (12 d) are required for decidualization and lefty processing.
The processing of lefty protein and its secretion appear to be dependent on the decidualizing signal and cell type. In SHT290 cells, E2/P and not cAMP led to the processing of lefty, and in HuFs cells, lefty was released into the culture medium only after treatment with cAMP and not after E2/P-induced decidualization. Similarly, Meno et al. reported that the processing of lefty-1 was dependent on the cells that were transfected to express the protein (1). When transfected with lefty-1 cDNA, the 293T cells produced the lefty precursor but failed to release it into the culture medium (10). On the other hand, the BALB/3T3 cells produced a 32-kDa protein and released it along with a 25- to 27-kDa protein into the culture medium (10). These findings show that the processing and release of lefty proteins have special requirements and not all cells have the required machinery to carry out both of these functions.
Members of the TGF- family are synthesized as precursor proteins, which are proteolytically processed to release the bioactive polypeptides (11). Proteins of the TGF- superfamily are cleaved by members of the PC family of endoproteases (12, 13, 14, 15). Thus far, in mammals, seven members of this family of proteins have been identified. These include furin, PC1/3, PC2, PC4, PC5/6A and B, PACE4, and PC7 (16, 17). These Ca2+-dependent serine proteases cleave proteins mostly at substrates on the C-terminal side of arginine-X-X-arginine (R-X-X-R) or lysine-X-X-arginine (K-X-X-R) pairs of basic amino acids (18, 19, 20). Furin was the first convertase to be extensively characterized, including in two knockout mouse models (21, 22), and has been shown in vitro to be required for the proper processing of several TGF- pro-proteins. Furin-deficient LoVo cells fail to cleave TGF-1, whereas cells transfected with Furin regain the ability to properly process TGF-1 (13). Thus, specific convertases appear to be important in the processing of members of the TGF- family. Lefty has two cleavage sites, RGKR and RHGR. Because the mutation of the sequence RGKR to GGKG (amino acids 74–77) and RHGR to GHGR (amino acids 132–135) prevented the proteolytic processing of lefty precursor, these sequences are bona fide sites for the action of convertases (4).
Despite the clear implication of the PC in the processing of TGF-1, little is known about the processing of lefty by these proteases during decidualization. In this study, we show the importance of the PC family of endoproteases in lefty processing. Neither Furin nor PACE4 or PC7 were able to process lefty to its long form. Only PC5/6 was capable of this processing and the A isoform was the most efficient. A similar observation was made for pro-lactase-phlorizin-hydrolase, which is cleaved by PC5/6A but not PC5/6B (23). Also, the proinsulin-like growth factor-IA is processed less efficiently by PC5/6A than by PC5/6B (24). The PC5/6A and B share the same catalytic domain (23). Therefore, it is not clear why they seem to differ so drastically in their activity of processing lefty protein. The reason for this discrepancy is likely due to the different compartmentalization of the isoforms. PC5/6A and PC5/6B are sorted to different compartments (25). PC5/6A has been shown to be localized in the extracellular matrix (26) and in the large dense-core vesicle in the neuroendocrine cells, whereas PC5/6B is concentrated in the trans-Golgi (27). Furthermore, PC5/6A is a soluble protein, whereas PC5/6B, due to its C-terminal membrane anchor, is membrane-bound (28, 29, 30).
Interestingly, both PC5/6A and B are increased during decidualization in mouse uterine horns. However, more PC5/6A than the B isoform is present in decidua, suggesting a more prominent role of PC5/6A in the processing of lefty. These findings are consistent with previous reports of strong PC5/6 expression during pregnancy in differentiated decidua (31) and in human endometrial stromal cells decidualized in vitro by E2/P treatment (32). Wong et al. (33) showed PC5/6A expression in mouse uterine decidua from E5.5 through to E8.5 and in artificially decidualized stroma. PC5/6 appears to be required for decidualization and for embryo implantation (32, 34). Okada et al. (32) have recently shown that antisense morpholino to PC5/6 decreases decidualization of human endometrial stromal cells in culture as evidenced by reduced prolactin production. Intraluminal administration of antisense morpholino to PC5/6 inhibits implantation (34). The requirement of PC5/6 does not appear to be limited to mouse, because PC5/6 is greatly up-regulated both in rhesus monkey and humans during the phase of endometrial receptivity and at implantation (34). The increase in the processed form of lefty in uterine horns treated with hormones occurred in the absence of decidualization or PC5 expression (Fig. 2B). This suggests that PC5-independent pathways might exist for lefty processing or that lefty accumulates in cells due to decreased degradation and or secretion.
Taken together, the findings show that lefty is a primary target for PC5/6 action and that up-regulation of PC5/6 along with processing of lefty are essential features of the decidualization process.
Footnotes
This work was supported by grants from the National Institute of Child Health and Human Development/National Institutes of Health: 1U01HD43165-01 (to S.T.), U01HD29963 (to D.C.), and HD42298 (to L.C.G.), and RO1HD42280 (to A.T.F.) as part of the National Cooperative Program on Trophoblast-Maternal Tissue Interactions. J.C. is holder of a scholarship from the "Fonds voor Wetenschappelijk Onderzoek Vlaanderen."
First Published Online September 1, 2005
Abbreviations: E2, 17- Estradiol; FBS, fetal bovine serum; IGFBP-1, IGF binding protein 1; MPA, medroxy-progesterone acetate; P, progesterone; PC, proprotein convertase.
Accepted for publication August 23, 2005.
References
Meno C, Saijoh Y, Fujii H, Ikeda M, Yokoyama T, Yokoyama M, Toyoda Y, Hamada H 1996 Left-right asymmetric expression of the TGF -family member lefty in mouse embryos. Nature 381:151–155
Kothapalli R, Buyuksal I, Wu S-Q, Chegini N, Tabibzadeh S 1997 Detection of ebaf, a novel human gene of the TGF- superfamily; association of gene expression with endometrial bleeding. J Clin Invest 99:2342–2350
Ulloa L, Tabibzadeh S 2001 Lefty inhibits receptor-regulated Smad phosphorylation induced by the activated transforming growth factor- receptor. J Biol Chem 276:21397–21404
Ulloa L, Creemers JW, Roy S, Liu S, Mason J, Tabibzadeh S 2001 Lefty proteins exhibit unique processing and activate the MAPK pathway. J Biol Chem 276:21387–21396
Brar AK, Handwerger S, Kessler CA, Aronow BJ 2001 Gene induction and categorical reprogramming during in vitro endometrial fibroblast decidualization. Physiol Genomics 7:135–148
Tierney EP, Tulac S, Huang ST, Giudice LC 2003 Activation of the protein kinase A pathway in human endometrial stromal cells reveals sequential categorical gene regulation. Physiol Genomics 16:47–66
Tang M, Taylor HS, Tabibzadeh S 2005 In vivo gene transfer of lefty leads to implantation failure in mice. Hum Reprod 20:1772–1778
Strakova Z, Srisuparp S, Fazleabas AT 2000 Interleukin-1 induces the expression of insulin-like growth factor binding protein-1 during decidualization in the primate. Endocrinology 141:4664–4670
Barbier CS, Becker KA, Troester MA, Kaufman DG 2005 Expression of exogenous human telomerase in cultures of endometrial stromal cells does not alter their hormone responsiveness. Biol Reprod 73:106–114
Meno C, Ito Y, Saijoh Y, Matsuda Y, Tashiro K, Kuhara S, Hamada H 1997 Two closely-related left-right asymmetrically expressed genes, lefty-1 and lefty-2: their distinct expression domains, chromosomal linkage and direct neuralizing activity in Xenopus embryos. Genes Cells 2:513–524
Sha X, Brunner AM, Purchio AF, Gentry LE 1989 Transforming growth factor 1: importance of glycosylation and acidic proteases for processing and secretion. Mol Endocrinol 3:1090–1098
Roebroek AJ, Creemers JW, Pauli IG, Bogaert T, Van de Ven WJ 1993 Generation of structural and functional diversity in furin-like proteins in Drosophila melanogaster by alternative splicing of the Dfur1 gene. EMBO J 12:1853–1870
Dubois CM, Laprise MH, Blanchette F, Gentry LE, Leduc R 1995 Processing of transforming growth factor 1 precursor by human furin convertase. J Biol Chem 270:10618–10624
Cui Y, Jean F, Thomas G, Christian JL 1998 BMP-4 is proteolytically activated by furin and/or PC6 during vertebrate embryonic development. EMBO J 17:4735–4743
Chan SJ, Oliva Jr AA, LaMendola J, Grens A, Bode H, Steiner DF 1992 Conservation of the prohormone convertase gene family in metazoa: analysis of cDNAs encoding a PC3-like protein from hydra. Proc Natl Acad Sci USA 89:6678–6682
Seidah NG, Chretien M 1997 Eukaryotic protein processing: endoproteolysis of precursor proteins. Curr Opin Biotechnol 8:602–607
Taylor NA, Van De Ven WJ, Creemers JW 2003 Curbing activation: proprotein convertases in homeostasis and pathology. FASEB J 17:1215–1227
Hosaka M, Nagahama M, Kim WS, Watanabe T, Hatsuzawa K, Ikemizu J, Murakami K, Nakayama K 1991 Arg-X-Lys/Arg-Arg motif as a signal for precursor cleavage catalyzed by furin within the constitutive secretory pathway. J Biol Chem 266:12127–12130
Molloy SS, Bresnahan PA, Leppla SH, Klimpel KR, Thomas G 1992 Human furin is a calcium-dependent serine endoprotease that recognizes the sequence Arg-X-X-Arg and efficiently cleaves anthrax toxin protective antigen. J Biol Chem 267:16396–16402
Creemers JW, Siezen RJ, Roebroek AJ, Ayoubi TA, Huylebroeck D, Van de Ven WJ 1993 Modulation of furin-mediated proprotein processing activity by site-directed mutagenesis. J Biol Chem 68:21826–21834
Roebroek AJ, Umans L, Pauli IG, Robertson EJ, van Leuven F, Van de Ven WJ, Constam DB 1998 Failure of ventral closure and axial rotation in embryos lacking the proprotein convertase Furin. Development 125:4863–4876
Roebroek AJ, Taylor NA, Louagie E, Pauli I, Smeijers L, Snellinx A, Lauwers A, Van de Ven WJ, Hartmann D, Creemers JW 2004 Limited redundancy of the proprotein convertase furin in mouse liver. J Biol Chem 279:53442–53450
Keller P, Zecca L, Boukamel R, Zwicker E, Gloor S, Semenza G 1995 Furin, PC1/3, and/or PC6A process rabbit, but not human, pro-lactase-phlorizin hydrolase to the 180-kDa intermediate. J Biol Chem 270:25722–25728
Duguay SJ, Milewski WM, Young BD, Nakayama K, Steiner DF 1997 Processing of wild-type and mutant proinsulin-like growth factor-IA by subtilisin-related proprotein convertases. J Biol Chem 272:6663–6670
De Bie I, Marcinkiewicz M, Malide D, Lazure C, Nakayama K, Bendayan M, Seidah NG 1996 The isoforms of proprotein convertase PC5 are sorted to different subcellular compartments. J Cell Biol 135:1261–1275
Tsuji A, Sakurai K, Kiyokage E, Yamazaki T, Koide S, Toida K, Ishimura K, Matsuda Y 2003 Secretory proprotein convertases PACE4 and PC6A are heparin-binding proteins which are localized in the extracellular matrix. Potential role of PACE4 in the activation of proproteins in the extracellular matrix. Biochim Biophys Acta 1645:95–104
Xiang Y, Molloy SS, Thomas L, Thomas G 2000 The PC6B cytoplasmic domain contains two acidic clusters that direct sorting to distinct trans-Golgi network/endosomal compartments. Mol Biol Cell 11:1257–1273
Nakagawa T, Hosaka M, Torii S, Watanabe T, Murakami K, Nakayama K 1993 Identification and functional expression of a new member of the mammalian Kex2-like processing endoprotease family: its striking structural similarity to PACE4. J Biochem 113:132–135
Nakagawa T, Murakami K, Nakayama K 1993 Identification of an isoform with an extremely large Cys-rich region of PC6, a Kex2-like processing endoprotease. FEBS Lett 327:165–171
Lusson J, Vieau D, Hamelin J, Day R, Chretien M, Seidah NG 1993 cDNA structure of the mouse and rat subtilisin/kexin-like PC5: a candidate proprotein convertase expressed in endocrine and nonendocrine cells. Proc Natl Acad Sci USA 90:6691–6695
Rancourt SL, Rancourt DE 1997 Murine subtilisin-like proteinase SPC6 is expressed during embryonic implantation, somitogenesis, and skeletal formation. Dev Genet 21:75–81
Okada H, Nie G, Salamonsen LA 2005 Requirement for proprotein convertase 5/6 during decidualization of human endometrial stromal cells in vitro. J Clin Endocrinol Metab 90:1028–1034
Wong BS, Liu S, Schultz GA, Rancourt DE 2002 Subtilisin proprotein convertase-6 expression in the mouse uterus during implantation and artificially induced decidualization. Mol Reprod Dev 61:453–459
Nie G, Li Y, Wang M, Liu YX, Findlay JK, Salamonsen LA 2005 Inhibiting uterine PC6 blocks embryo implantation: an obligatory role for a proprotein convertase in fertility. Biol Reprod 72:1029–1036(Meiyi Tang, Anatoly Mikhailik, Ilse Paul)
Laboratory for Molecular Cell Biology (I.P., J.W.M.C.), Department for Human Genetics, University of Leuven and Flanders Interuniversity Institute for Biotechnology, B-3000 Leuven, Belgium
Department of Gynecology and Obstetrics (L.C.G., S.T.), Stanford University Medical Center, Stanford, California 94305
Department of Obstetrics and Gynecology (A.T.F.), University of Illinois at Chicago, Chicago, Illinois 60612
Department of Biological Sciences (D.D.C.), University of Delaware, Newark, Delaware 19707
Department of Laboratory Pathology and Medicine (D.G.K., C.B.), Division of Reproductive Endocrinology and Infertility, University of North Carolina, Chapel Hill, North Carolina 27599
Abstract
Lefty/Ebaf polypeptides, novel members of the TGF- superfamily, are involved in endometrial differentiation and embryo implantation. Recently, we showed that, during undisturbed estrous cycle, lefty is present in mouse uterine horn primarily in a precursor form. Here, we show that decidual differentiation of endometrial stroma leads to increased lefty (3.1- to 3.6-fold in vivo and 5- to 8-fold in vitro) and processing of its precursor primarily to its long form. This event occurs on d 5 of pregnancy, and is paralleled by proprotein convertase (PC)5/6 up-regulation (6-fold increase for PC5A and 3-fold increase for PC5B) in decidualized uterine horn, independent of embryo implantation. Among the known convertases, only PC5/6A processes lefty to its long form. Taken together, the findings show that decidualized differentiation of stroma, which is a prerequisite for embryo implantation, leads to processing of lefty by PC5/6A.
Introduction
LEFTY/EBAF POLYPEPTIDES, novel members of the TGF- superfamily, are involved in the formation of the embryonic lateral patterning and endometrial differentiation (1, 2). Members of the TGF- superfamily require processing for their activation, suggesting that cleavage is an essential step for lefty activation. Transfection of different cell lines with human lefty A and B form resulted in expression of a 42-kDa protein, which was proteolytically processed to release two polypeptides of 34 kDa (long form) and 28 kDa (short form) (3). The mutation of the consensus sequences for proprotein convertase (PC) cleavage in lefty protein identified and validated the lefty cleavage sites. The mutation of the sequence RGKR to GGKG (amino acids 74–77) and RHGR to GHGR (amino acids 132–135) prevented the proteolytic processing of lefty precursor to the 34- and 28-kDa forms, respectively (4). In pluripotent P19 cells, lefty did not induce Smad2 or Smad5 phosphorylation, Smad2/Smad4 heterodimerization, or nuclear translocation of the Smad2 or Smad4, but activated MAPK pathway in a time- and dose-dependent fashion (4). The 28-kDa, but not the 34-kDa, polypeptide induced MAPK activity. These data supported a molecular model of processing as a mechanism for regulation of lefty action(s).
In humans, lefty is expressed in endometrial stromal cells that gradually undergo the differentiated state of decidualization (2). Decidual cells adopt unique phenotypic, functional, biological, secretory, and biosynthetic features. Decidualization is promoted by the cAMP and protein kinase A pathways as well as, after an initial estradiol priming, by progesterone (P). Recent studies indicate that this process is associated with exquisitely controlled sequential gene regulation including kinetic reprogramming of genes associated with decreased G protein signaling, increased STAT pathway signaling, cell proliferation, structural protein changes, cellular differentiation, and secretory processes (5, 6). Among these changes, expression and secretion of prolactin and IGBP-1 have become the hallmarks of this differentiated cellular state. Decidualized cells produce extracellular matrix proteins including laminin, type IV collagen, fibronectin, and heparin sulfate proteoglycan. The kinetic reprogramming, as revealed by microarray analysis of human endometrial stromal cells forced to differentiate in vitro by cAMP, includes remarkable elevation in the expression of lefty (5, 6). The expression of lefty is increased from a low basal level by 3-fold within 2 h, and 123-fold within 36 h of receiving the decidualizing stimulus (6). Brar et al. validated the microarray data with Northern blot analysis and showed a steady increase of lefty over a period of 12 d when the decidual cells are fully developed (5).
Taken together, these findings suggest that lefty might be an integral component of the decidualization process. We were interested in extending these observations to a model system more amenable to experimental manipulation in vivo. Therefore, the current study was carried out to define precisely the expression and processing of lefty during decidualization. The results show that induction of decidualization leads to the conversion of lefty into its long form. This event, which occurs upon decidualization of stromal cells, is paralleled with the expression of PC5/6, the primary enzyme for proteolytic cleavage of the precursor form of lefty to its long form (4). These findings show that lefty expression and processing are intrinsic features of the differentiated decidualized state of stromal cells.
Materials and Methods
Materials
Chemicals were from Sigma-Aldrich (St. Louis, MO) or Fisher Scientific (Pittsburgh, PA). The avidin-biotin-peroxidase kit was from Vector Laboratories (Burlingame, CA). The goat polyclonal antibody to lefty peptide (M-20) mapped to the carboxyl terminus of mouse lefty and recombinant IGF binding protein 1 (IGFBP-1) were from Santa Cruz Biotechnology Inc. (Santa Cruz, CA). The recombinant lefty was from R&D Systems (Minneapolis, MN). The antibody to actin was from Abcam, Inc. (Cambridge, MA). Rabbit antibody to PC5/6 was from Alexis Biochemicals (Laüfelfingen, Switzerland). Outbred CF-1 and CD-1 mice were obtained from Charles River Laboratory (Wilmington, MA).
Sample preparation and analysis of estrous cycle and pregnancy
In this report, the expression of mouse lefty was examined in the endometria of mice (n = 6) during pregnancy. Each experiment was repeated at least twice to confirm reproducibility of data. Day 1 of pregnancy was considered as the day when the vaginal plug was detected. Approval of the institutional Animal Care and Use Committee of the University of Delaware was obtained for carrying out the removal of uterine horns during pregnancy.
Transfection
CHO cells were maintained in DMEM (Life Technologies, Inc., Rockville, MD) supplemented with 10% (vol/vol) fetal bovine serum (FBS; Life Technologies, Inc.) and 1% (wt/vol) antibiotic-antimycotic mixture (Life Technologies, Inc.). For transfection, cells were trypsinized with 0.05% (wt/vol) trypsin-EDTA (Life Technologies, Inc.) and seeded into six-well plates (Falcon, Franklin lakes, NJ) at a concentration of 2 x 105 cells per well. Cells were cultured in 1.5 ml of the culture medium, in the presence of CO2 at 37 C for about 16 h. Upon reaching 60% confluency, cells were cotransfected with cDNA of lefty A and with furin, PACE4, PC5/6A, PC5/6B, or PC7 cDNA, using Superfect Transfect Reagent (Qiagen, Valencia, CA) following the manufacturer’s protocol. In separate experiments, 293 cells were transfected with an expression PC5/6 vector. Empty vector served as control in these transfections. Lysate from these cells was used as a control for identification of PC5/6 on Western blotting. In transient transfections, serum-free medium was used.
Induction of decidualization in mouse uterine horn
Decidualization was induced in ovariectomized mice 4–5 d after ovariectomy. Mice received three daily sc injections of 17- estradiol (E2, 100 ng). After 4 d of rest, animals received three additional daily sc injections of E2 (10 ng) as well as sc injections of P (1 mg). Both E2 and P were delivered in 0.1 ml of sesame oil. On the third day of treatment with P, animals received the decidualizing signal, which was either trauma or oil. After anesthesia, an incision was made in the left flank and the left uterine horn was exposed. Trauma was produced by the insertion of a blunt 25-gauge needle seven times into the horn lumen. For decidualization by oil, 20 μl of sesame oil was injected into the proximal left uterine horn at the tubal junction. Animals received three additional daily sc injections of P (1 mg). Seventy-two hours later, both uterine horns were removed. The approval of the institutional Animal Care and Use Committee of Stony Brook University was obtained for carrying out the experiments on decidualization of mouse uterine horns.
Culture and decidualization of human endometrial stromal cells
HuF were isolated from decidua parietalis dissected from the placental membranes after normal vaginal delivery at term under an approved protocol by the University of Illinois at Chicago Committee on the Use of Human Subjects in Medical Research. Briefly, scraped cells were digested in 0.1% collagenase, 0.02% deoxynuclease in calcium- and magnesium-free Hanks’ balanced salt solution. Cells were plated in four 100-mm culture dishes (Becton Dickinson and Co. Labware, Franklin Lakes, NJ) and placed into an incubator at 37 C in presence of 5% CO2. The next day, plates were extensively washed with PBS to remove nonadherent (mainly decidual) cells. At confluence, cells were trypsinized and used for experiments in passage number 3–5. Cell purity was assessed by immunocytochemistry using antibodies against cytokeratin (Dako Corp., Carpenteria, CA) and vimentin (Zymed Laboratories, Inc., San Francisco, CA). The purity of the fibroblast cell preparations used was more than 95%. These cells were treated with medium alone and with medium supplemented with hormones [10–6 M medroxy-progesterone acetate (MPA) + 36 nM E2] or with cAMP (1 mM) for 14 d. The medium was comprised of DMEM supplemented with 2% (wt/vol) charcoal-stripped FBS serum and was changed every 2 d. FBS was omitted 48 h before the end of treatment. A total of 4 x 106 SHT290 cells (immortalized human endometrial stromal cells) was treated with medium and with medium supplemented with 10–6 M MPA and 10–8 M E2, or with cAMP (1 mM) for 13 d. The medium was comprised of DMEM, with high glucose, supplemented with 2% charcoal-stripped FBS and epidermal growth factor (20 ng/ml).
Human endometrial samples were collected from women with normal menstrual cycles after obtaining informed consent under an approved protocol by the Stanford University Committee on the Use of Human Subjects in Medical Research. Biopsies were obtained from subjects, 26–46 yr of age, who had regular menstrual cycles (28–35 d), were documented not to be pregnant, and had no history of endometriosis. Endometrial tissues were subjected to collagenase/hyaluronidase (Sigma-Aldrich) digestion for 2 h at 37 C. After digestion, the stroma was dispersed, although the glandular structures remained mostly intact. Stromal cells were separated from glands on a size basis. Stromal cells were centrifuged, and the resulting pellet was resuspended in a 4:1 ratio in DMEM-F12 (Life Technologies, Inc.) and DMEM/10% FBS. Cells were preplated in 10 cm2 standard culture plates for 1 h at 37 C in the presence of 9% CO2, and the medium was replaced with high-glucose DMEM/MCDB-105 medium with 10% charcoal-stripped FBS, insulin (5 μg/ml) (Sigma-Aldrich), gentamycin, penicillin, and streptomycin. Endometrial stromal cells were passaged two to three times before the experiments. The stromal cells used were 99% pure, devoid of lymphocytes, endothelial cells, and fibroblasts (5). Culture medium was changed every 2–3 d. For induction of decidualization, cultured stromal cells were pretreated with E2 (10 μM), and, once confluent, they were treated in serum-free media supplemented with 1 μM 8-bromo cAMP for 72 h. The concentration of IGFBP-1 was determined in conditioned media after cAMP treatment using an IGFBP-1 ELISA (DSL, Webster, TX).
Northern blot analysis
Total RNA (20 μg/lane) or poly A RNA (2 μg/lane) was fractionated in formaldehyde-agarose gel. After diffusion transfer to nitrocellulose filters and cross-linking (Stratalinker; Stratagene, La Jolla, CA), hybridizations were performed using denatured 32P-labeled cRNA of lefty, and cDNA probes for mouse fur, PC5/6, PACE4, and PC7 were used as described previously (12). Gels were stained with ethidium bromide for revealing the 18S and 28S ribosomal RNA and blots were stripped and probed for actin cDNA for normalization. Quantification was performed using Sigma Gel (Sigma-Aldrich) as well as the Typhoon 9400 (GE Healthcare) and ImageQuant software (Molecular Dynamics).
SDS-PAGE and Western blotting
Media were centrifuged at 7000 x g for 3 h and concentrated by about 12-fold using Centricon 10 (protein molecular size cut-off: 10,000 kDa; Amicon, Danvers, MA). The concentration of proteins in these samples was determined by Bio-Rad Protein Assay kit (Bio-Rad Laboratories, Hercules, CA). Concentrated conditioned media or cell lysates (50 μg protein per lane) were fractionated in a 12% denaturing gel together with prestained protein ladder (Life Technologies, Inc.) and were subsequently blotted onto nitrocellulose membrane in a Mini-Trans-Blot apparatus (Bio-Rad Laboratories). Blots were stained with the goat polyclonal antibody to lefty peptide (M-20) (1–2 μg/ml) or where indicated with rabbit antibody to PC5/6. The secondary antibody used was donkey antigoat IgG-HRP or goat antirabbit (Santa Cruz Biotechnology). Bands were detected by chemiluminescence as described by the manufacturer. Quantification was performed using Sigma Gel (Sigma-Aldrich).
Statistical analysis
Density of bands and ratios of precursor/long form of lefty were compared using t test. Differences were considered statistically significant when P < 0.05.
Results
Lefty proteins in mouse uterine horn during pregnancy
Transfection of different cell lines with human lefty A and B form resulted in expression of a 42-kDa protein, which was proteolytically processed to release two polypeptides of 34 kDa (long form) and 28 kDa (short form) (4). To examine for the presence of similar processed forms in mouse uterine horn, proteins from horns of pregnant mice were subjected to Western blot analysis for lefty. On d 1 of pregnancy, precursor and long form of lefty were equally present. However, on d 3, and particularly on d 5, of pregnancy, there was more long form of lefty compared with the precursor form (Fig. 1). On d 9, this was more notable close to the implantation site rather than in tissues intervening the implantation sites (Fig. 1), suggesting that processing of lefty occurs in decidualized stroma more efficiently around the implanting embryo.
Lefty in mouse endometrium during artificial decidualization
To determine whether lefty processing was due to the decidualization of stroma and whether it required presence of an embryo, decidualization was artificially induced in endometrium. We introduced oil, as the deciduogenic signal, into the left uterine horns of mice primed with steroid hormones, E2 and P. The left uterine horns, removed 3 d after receiving the decidualizing signal, were enlarged, and the distal parts of some right uterine horns were also enlarged, suggesting transfer of oil from the left to the right horn (Fig. 2). The introduction of Chicago blue dye has substantiated this trans-uterine horn flow (7). For this reason, the uterine horns of mice primed with hormones that did not receive the decidualizing signal, rather than the right horn of animals that received the deciduogenic stimulus, were used as controls. Trauma was used as a second deciduogenic signal in the left uterine horn. The left horns removed 3 d later were also enlarged (Fig. 2). The wet weight of the uterine horns injected with oil was significantly higher than the weights of horns that were traumatized (data not shown). The sections of uterine horns examined after hematoxylin and eosin staining showed the typical morphology of a decidualized stroma (data not shown). Western blot analysis for IGFBP-1 was carried out to confirm the decidualization of stroma in the uterine horns. IGFBP-1 was found in both uterine horns, but unless both horns were enlarged to the same extent, more IGFBP-1 immunoreactivity was present in the left compared with the right horns (Fig. 2).
Northern blot analysis showed both 2.4- and 1.5-kb bands of lefty mRNA in mouse uterine horns of both nondecidualized control and decidualized horns. As compared with control uterine horns, lefty was expressed at a significantly higher level (3.1- to 3.6-fold) in the decidualized horns irrespective of the type of the decidualizing signal (Fig. 3A). Although both precursor and long forms of lefty were present in the nondecidualized uterine horns primed with hormones E2 and P, the long form was the predominant form in the decidualized horns irrespective of the deciduogenic signal (Fig. 3B).
Expression of convertases in mouse uterine horn during artificial decidualization
Because members of the PC family cleave different TGF- proteins, the data suggested that processing of lefty might be due to expression of a convertase in the decidualized stroma. For this reason, Northern blot analysis for the relevant convertases including Furin, PC5/6, PC7, and PACE4, were carried out on the RNA isolated from the decidualized and control nondecidualized mouse uterine horns. Among these convertases, only the expression of PC5/6 was significantly increased in the decidualized horns (Fig. 4A). Both deciduogenic signals increased the expression of PC5/6 and both the A (short form) and B (long form) isoforms were significantly increased (6-fold increase for PC5A and 3-fold increase for PC5B). Western blot analysis was performed on the tissue lysates of the decidualized and nondecidualized uterine horn to validate increased PC5/6. The lysate of 293 cells transfected with PC5/6 served as a positive control for the identification of PC5/6, whereas the lysate from 293 cells transfected with empty vector served as a negative control. As shown in Fig. 4B, PC5/6 was increased in the decidualized uterine horn, whereas it was low to undetectable in control hormone-primed nondecidualized uterine horn.
Processing of lefty by PC5/6A
These findings suggested that lefty was being processed in the decidualized stroma by PC5/6. To directly show that PC5/6 was responsible for lefty processing, CHO cells were transfected simultaneously with lefty cDNA and cDNAs of various PCs including Furin, PACE4, PC5/6A, PC5/6B, and PC7. From these convertases only the PC5/6 was able to convert lefty to its long form. PC5/6A was more efficient than PC5/6B in the processing of lefty to its long form (Fig. 5).
Processing of lefty in endometrial stromal cells decidualized in vitro
To directly show that lefty was processed in decidualized cells irrespective of other in vivo signals, decidualization was induced in HuFs cells. These cells were derived from human placenta and have been shown to decidualize in vitro both with cAMP and with E2/P treatment (8). Both cAMP and E2/P led to the decidualization of these cells as confirmed by release of IGFBP-1 into the culture medium (Fig. 6A). Both decidualizing signals led to the accumulation of the long form of lefty in the cell lysate. However, the long form of lefty was present in the culture medium of cells decidualized with cAMP but not with E2/P (Fig. 6A). To further validate these findings, SHT290 cells, which were derived from human endometrial stromal cells and immortalized by transfection with telomerase were used. SHT290 stromal cells were very similar to the parental strain from which they were derived according to criteria of proliferation, karyotype, cellular localization of cytoskeletal markers and nuclear staining, and basal gene expression based on microarray analysis (9). These cells were shown to decidualize with E2/P treatment (9). These cells were decidualized in vitro with E2/P and with cAMP. Untreated cells served as controls. In these cells, only E2/P and not cAMP led to the processing of lefty (Fig. 6B). However, neither treatment led to the release of lefty to the culture medium. This suggested that culture condition or transfection with telomerase might have hampered the ability of these cells to respond to cAMP.
To show that lefty was also processed in response to cAMP, and was released into culture medium, primary cultures of human endometrial stromal cells derived from five different subjects were decidualized with cAMP. Northern blot analysis of these cells showed significant increase (5- to 8-fold) in lefty expression within the short duration of treatment (Fig. 7A). The quantity of the IGFBP-1 was determined in the culture media of these cells by ELISA. IGFBP-1 was present in the culture media of decidualized cells and not the culture media of untreated cells (Fig. 7B). Western blot analysis of culture media of nondecidualized cells showed the precursor form of lefty (Fig. 7B). On the other hand, decidualization of stromal cells led to accumulation of both long and short forms of lefty in the culture media (Fig. 7B).
Discussion
We previously showed that lefty was expressed in human endometrium at a high level immediately before and during menstrual bleeding (2). In situ hybridization showed that lefty was expressed in cells that were undergoing decidualization, a requisite for successful implantation (2). In human endometrium, decidualization occurs in the absence of the embryo. However, in mice, decidualization occurs only in the presence of an embryo or by application of a decidualizing signal to a properly primed endometrium. During pregnancy, by the time the embryo arrives into the mouse uterine cavity on d 4 and by d 5, there is evidence of decidualization of stroma. Here, we show that this decidualization is associated with an increase in the long form of lefty in mouse endometrium. However, presence of the embryo is not required for this event, because lefty is efficiently processed to its long form when uterine horn is artificially forced to decidualize with oil or trauma. Although both precursor and long forms of lefty are increased with hormonal priming alone, decidualization appears to be important for the efficient conversion of lefty to its long form.
Decidualization of mouse uterine horn as well as decidualization of various human endometrial stromal cells leads to increased lefty expression. These findings are consistent with the microarray data and Northern blot analysis, which show a steady level of increase of lefty as the decidual cells differentiate (5, 6). In vitro, cAMP is more efficient than E2 + P in the induction of decidualization in lefty expression in primary cultures of human endometrial stromal cells. Treatment of endometrial stromal cells with cAMP induces decidualization of cells, IGFBP-1 release into the culture medium, and lefty processing within 72 h; longer treatment with E2/P (12 d) are required for decidualization and lefty processing.
The processing of lefty protein and its secretion appear to be dependent on the decidualizing signal and cell type. In SHT290 cells, E2/P and not cAMP led to the processing of lefty, and in HuFs cells, lefty was released into the culture medium only after treatment with cAMP and not after E2/P-induced decidualization. Similarly, Meno et al. reported that the processing of lefty-1 was dependent on the cells that were transfected to express the protein (1). When transfected with lefty-1 cDNA, the 293T cells produced the lefty precursor but failed to release it into the culture medium (10). On the other hand, the BALB/3T3 cells produced a 32-kDa protein and released it along with a 25- to 27-kDa protein into the culture medium (10). These findings show that the processing and release of lefty proteins have special requirements and not all cells have the required machinery to carry out both of these functions.
Members of the TGF- family are synthesized as precursor proteins, which are proteolytically processed to release the bioactive polypeptides (11). Proteins of the TGF- superfamily are cleaved by members of the PC family of endoproteases (12, 13, 14, 15). Thus far, in mammals, seven members of this family of proteins have been identified. These include furin, PC1/3, PC2, PC4, PC5/6A and B, PACE4, and PC7 (16, 17). These Ca2+-dependent serine proteases cleave proteins mostly at substrates on the C-terminal side of arginine-X-X-arginine (R-X-X-R) or lysine-X-X-arginine (K-X-X-R) pairs of basic amino acids (18, 19, 20). Furin was the first convertase to be extensively characterized, including in two knockout mouse models (21, 22), and has been shown in vitro to be required for the proper processing of several TGF- pro-proteins. Furin-deficient LoVo cells fail to cleave TGF-1, whereas cells transfected with Furin regain the ability to properly process TGF-1 (13). Thus, specific convertases appear to be important in the processing of members of the TGF- family. Lefty has two cleavage sites, RGKR and RHGR. Because the mutation of the sequence RGKR to GGKG (amino acids 74–77) and RHGR to GHGR (amino acids 132–135) prevented the proteolytic processing of lefty precursor, these sequences are bona fide sites for the action of convertases (4).
Despite the clear implication of the PC in the processing of TGF-1, little is known about the processing of lefty by these proteases during decidualization. In this study, we show the importance of the PC family of endoproteases in lefty processing. Neither Furin nor PACE4 or PC7 were able to process lefty to its long form. Only PC5/6 was capable of this processing and the A isoform was the most efficient. A similar observation was made for pro-lactase-phlorizin-hydrolase, which is cleaved by PC5/6A but not PC5/6B (23). Also, the proinsulin-like growth factor-IA is processed less efficiently by PC5/6A than by PC5/6B (24). The PC5/6A and B share the same catalytic domain (23). Therefore, it is not clear why they seem to differ so drastically in their activity of processing lefty protein. The reason for this discrepancy is likely due to the different compartmentalization of the isoforms. PC5/6A and PC5/6B are sorted to different compartments (25). PC5/6A has been shown to be localized in the extracellular matrix (26) and in the large dense-core vesicle in the neuroendocrine cells, whereas PC5/6B is concentrated in the trans-Golgi (27). Furthermore, PC5/6A is a soluble protein, whereas PC5/6B, due to its C-terminal membrane anchor, is membrane-bound (28, 29, 30).
Interestingly, both PC5/6A and B are increased during decidualization in mouse uterine horns. However, more PC5/6A than the B isoform is present in decidua, suggesting a more prominent role of PC5/6A in the processing of lefty. These findings are consistent with previous reports of strong PC5/6 expression during pregnancy in differentiated decidua (31) and in human endometrial stromal cells decidualized in vitro by E2/P treatment (32). Wong et al. (33) showed PC5/6A expression in mouse uterine decidua from E5.5 through to E8.5 and in artificially decidualized stroma. PC5/6 appears to be required for decidualization and for embryo implantation (32, 34). Okada et al. (32) have recently shown that antisense morpholino to PC5/6 decreases decidualization of human endometrial stromal cells in culture as evidenced by reduced prolactin production. Intraluminal administration of antisense morpholino to PC5/6 inhibits implantation (34). The requirement of PC5/6 does not appear to be limited to mouse, because PC5/6 is greatly up-regulated both in rhesus monkey and humans during the phase of endometrial receptivity and at implantation (34). The increase in the processed form of lefty in uterine horns treated with hormones occurred in the absence of decidualization or PC5 expression (Fig. 2B). This suggests that PC5-independent pathways might exist for lefty processing or that lefty accumulates in cells due to decreased degradation and or secretion.
Taken together, the findings show that lefty is a primary target for PC5/6 action and that up-regulation of PC5/6 along with processing of lefty are essential features of the decidualization process.
Footnotes
This work was supported by grants from the National Institute of Child Health and Human Development/National Institutes of Health: 1U01HD43165-01 (to S.T.), U01HD29963 (to D.C.), and HD42298 (to L.C.G.), and RO1HD42280 (to A.T.F.) as part of the National Cooperative Program on Trophoblast-Maternal Tissue Interactions. J.C. is holder of a scholarship from the "Fonds voor Wetenschappelijk Onderzoek Vlaanderen."
First Published Online September 1, 2005
Abbreviations: E2, 17- Estradiol; FBS, fetal bovine serum; IGFBP-1, IGF binding protein 1; MPA, medroxy-progesterone acetate; P, progesterone; PC, proprotein convertase.
Accepted for publication August 23, 2005.
References
Meno C, Saijoh Y, Fujii H, Ikeda M, Yokoyama T, Yokoyama M, Toyoda Y, Hamada H 1996 Left-right asymmetric expression of the TGF -family member lefty in mouse embryos. Nature 381:151–155
Kothapalli R, Buyuksal I, Wu S-Q, Chegini N, Tabibzadeh S 1997 Detection of ebaf, a novel human gene of the TGF- superfamily; association of gene expression with endometrial bleeding. J Clin Invest 99:2342–2350
Ulloa L, Tabibzadeh S 2001 Lefty inhibits receptor-regulated Smad phosphorylation induced by the activated transforming growth factor- receptor. J Biol Chem 276:21397–21404
Ulloa L, Creemers JW, Roy S, Liu S, Mason J, Tabibzadeh S 2001 Lefty proteins exhibit unique processing and activate the MAPK pathway. J Biol Chem 276:21387–21396
Brar AK, Handwerger S, Kessler CA, Aronow BJ 2001 Gene induction and categorical reprogramming during in vitro endometrial fibroblast decidualization. Physiol Genomics 7:135–148
Tierney EP, Tulac S, Huang ST, Giudice LC 2003 Activation of the protein kinase A pathway in human endometrial stromal cells reveals sequential categorical gene regulation. Physiol Genomics 16:47–66
Tang M, Taylor HS, Tabibzadeh S 2005 In vivo gene transfer of lefty leads to implantation failure in mice. Hum Reprod 20:1772–1778
Strakova Z, Srisuparp S, Fazleabas AT 2000 Interleukin-1 induces the expression of insulin-like growth factor binding protein-1 during decidualization in the primate. Endocrinology 141:4664–4670
Barbier CS, Becker KA, Troester MA, Kaufman DG 2005 Expression of exogenous human telomerase in cultures of endometrial stromal cells does not alter their hormone responsiveness. Biol Reprod 73:106–114
Meno C, Ito Y, Saijoh Y, Matsuda Y, Tashiro K, Kuhara S, Hamada H 1997 Two closely-related left-right asymmetrically expressed genes, lefty-1 and lefty-2: their distinct expression domains, chromosomal linkage and direct neuralizing activity in Xenopus embryos. Genes Cells 2:513–524
Sha X, Brunner AM, Purchio AF, Gentry LE 1989 Transforming growth factor 1: importance of glycosylation and acidic proteases for processing and secretion. Mol Endocrinol 3:1090–1098
Roebroek AJ, Creemers JW, Pauli IG, Bogaert T, Van de Ven WJ 1993 Generation of structural and functional diversity in furin-like proteins in Drosophila melanogaster by alternative splicing of the Dfur1 gene. EMBO J 12:1853–1870
Dubois CM, Laprise MH, Blanchette F, Gentry LE, Leduc R 1995 Processing of transforming growth factor 1 precursor by human furin convertase. J Biol Chem 270:10618–10624
Cui Y, Jean F, Thomas G, Christian JL 1998 BMP-4 is proteolytically activated by furin and/or PC6 during vertebrate embryonic development. EMBO J 17:4735–4743
Chan SJ, Oliva Jr AA, LaMendola J, Grens A, Bode H, Steiner DF 1992 Conservation of the prohormone convertase gene family in metazoa: analysis of cDNAs encoding a PC3-like protein from hydra. Proc Natl Acad Sci USA 89:6678–6682
Seidah NG, Chretien M 1997 Eukaryotic protein processing: endoproteolysis of precursor proteins. Curr Opin Biotechnol 8:602–607
Taylor NA, Van De Ven WJ, Creemers JW 2003 Curbing activation: proprotein convertases in homeostasis and pathology. FASEB J 17:1215–1227
Hosaka M, Nagahama M, Kim WS, Watanabe T, Hatsuzawa K, Ikemizu J, Murakami K, Nakayama K 1991 Arg-X-Lys/Arg-Arg motif as a signal for precursor cleavage catalyzed by furin within the constitutive secretory pathway. J Biol Chem 266:12127–12130
Molloy SS, Bresnahan PA, Leppla SH, Klimpel KR, Thomas G 1992 Human furin is a calcium-dependent serine endoprotease that recognizes the sequence Arg-X-X-Arg and efficiently cleaves anthrax toxin protective antigen. J Biol Chem 267:16396–16402
Creemers JW, Siezen RJ, Roebroek AJ, Ayoubi TA, Huylebroeck D, Van de Ven WJ 1993 Modulation of furin-mediated proprotein processing activity by site-directed mutagenesis. J Biol Chem 68:21826–21834
Roebroek AJ, Umans L, Pauli IG, Robertson EJ, van Leuven F, Van de Ven WJ, Constam DB 1998 Failure of ventral closure and axial rotation in embryos lacking the proprotein convertase Furin. Development 125:4863–4876
Roebroek AJ, Taylor NA, Louagie E, Pauli I, Smeijers L, Snellinx A, Lauwers A, Van de Ven WJ, Hartmann D, Creemers JW 2004 Limited redundancy of the proprotein convertase furin in mouse liver. J Biol Chem 279:53442–53450
Keller P, Zecca L, Boukamel R, Zwicker E, Gloor S, Semenza G 1995 Furin, PC1/3, and/or PC6A process rabbit, but not human, pro-lactase-phlorizin hydrolase to the 180-kDa intermediate. J Biol Chem 270:25722–25728
Duguay SJ, Milewski WM, Young BD, Nakayama K, Steiner DF 1997 Processing of wild-type and mutant proinsulin-like growth factor-IA by subtilisin-related proprotein convertases. J Biol Chem 272:6663–6670
De Bie I, Marcinkiewicz M, Malide D, Lazure C, Nakayama K, Bendayan M, Seidah NG 1996 The isoforms of proprotein convertase PC5 are sorted to different subcellular compartments. J Cell Biol 135:1261–1275
Tsuji A, Sakurai K, Kiyokage E, Yamazaki T, Koide S, Toida K, Ishimura K, Matsuda Y 2003 Secretory proprotein convertases PACE4 and PC6A are heparin-binding proteins which are localized in the extracellular matrix. Potential role of PACE4 in the activation of proproteins in the extracellular matrix. Biochim Biophys Acta 1645:95–104
Xiang Y, Molloy SS, Thomas L, Thomas G 2000 The PC6B cytoplasmic domain contains two acidic clusters that direct sorting to distinct trans-Golgi network/endosomal compartments. Mol Biol Cell 11:1257–1273
Nakagawa T, Hosaka M, Torii S, Watanabe T, Murakami K, Nakayama K 1993 Identification and functional expression of a new member of the mammalian Kex2-like processing endoprotease family: its striking structural similarity to PACE4. J Biochem 113:132–135
Nakagawa T, Murakami K, Nakayama K 1993 Identification of an isoform with an extremely large Cys-rich region of PC6, a Kex2-like processing endoprotease. FEBS Lett 327:165–171
Lusson J, Vieau D, Hamelin J, Day R, Chretien M, Seidah NG 1993 cDNA structure of the mouse and rat subtilisin/kexin-like PC5: a candidate proprotein convertase expressed in endocrine and nonendocrine cells. Proc Natl Acad Sci USA 90:6691–6695
Rancourt SL, Rancourt DE 1997 Murine subtilisin-like proteinase SPC6 is expressed during embryonic implantation, somitogenesis, and skeletal formation. Dev Genet 21:75–81
Okada H, Nie G, Salamonsen LA 2005 Requirement for proprotein convertase 5/6 during decidualization of human endometrial stromal cells in vitro. J Clin Endocrinol Metab 90:1028–1034
Wong BS, Liu S, Schultz GA, Rancourt DE 2002 Subtilisin proprotein convertase-6 expression in the mouse uterus during implantation and artificially induced decidualization. Mol Reprod Dev 61:453–459
Nie G, Li Y, Wang M, Liu YX, Findlay JK, Salamonsen LA 2005 Inhibiting uterine PC6 blocks embryo implantation: an obligatory role for a proprotein convertase in fertility. Biol Reprod 72:1029–1036(Meiyi Tang, Anatoly Mikhailik, Ilse Paul)