| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Other Original Articles |
Department of Obstetrics and Gynecology, C. S. Mott Center for Human Growth and Development (R.E.L., R.K., D.R.A.), Wayne State University, Detroit, Michigan; Department of Molecular and Integrative Physiology (S.K.Da., S.K.De.), University of Kansas Medical Center, Kansas City, Kansas; and Department of Obstetrics and Gynecology (A.B., A.T.F.), University of Illinois, Chicago Illinois 60612-7313
Address all correspondence and requests for reprints to: Asgerally T. Fazleabas, Ph.D., Department of Obstetrics and Gynecology, University of Illinois at Chicago, 820 South Wood Street (M/C 808), Chicago, Illinois 60612-7313. E-mail: asgi{at}uic.edu
Abstract
The objectives of this study were to determine whether antiprogestin therapy or the infusion of human CG to mimic blastocyst transit in the baboon alters heparin-binding EGF-like growth factor expression during the window of implantation. During the menstrual cycle, heparin-binding EGF-like growth factor protein accumulation in the glandular epithelium was low in the proliferative phase and increased to maximal expression on d 5 and 10 postovulation. Stromal cells accumulated high levels of heparin-binding EGF-like growth factor in the proliferative phase, which decreased by d 5 postovulation. These transitional changes in both cell types were delayed when cycling baboons were treated with the antiprogestin ZK 137.316 during the luteal phase. The treatment with human CG had no effect on expression of heparin-binding EGF-like growth factor when compared with cycling baboons on d 10 postovulation and was comparable with that observed on d 18 and 22 of pregnancy. However, the superimposition of the antiprogestin with the human CG treatment also decreased expression in the epithelial cells. In summary, heparin-binding EGF-like growth factor accumulation in the epithelial glands is under the influence of progesterone and does not seem to be influenced by the paracrine secretion of trophoblast CG.
HEPARIN-BINDING EGF-like growth factor (HB-EGF), first identified in conditioned media from macrophage-like U937 cells, is synthesized as a transmembrane precursor that is enzymatically cleaved and released as a soluble 14–20 kD form (1). It is a potent mitogen for endothelial cells, keratinocytes, and smooth muscle cells and a chemoattractant for the latter (1). HB-EGF seems to play an important role in embryonic-maternal communication during implantation. In the ovariectomized mouse and rat HB-EGF mRNA and protein accumulates only in the epithelial cells after estradiol injection and in stromal cells after treatment with progesterone and estradiol (2, 3). The receptive mouse uterus produces HB-EGF on the luminal epithelium (LE) exclusively at sites adjacent to the blastocyst (4). The HB-EGF gene is first expressed by the LE on gestation d 4 at 1600 h, before hatching, suggesting that it is induced by the blastocyst. HB-EGF preferentially binds receptors HER 1 and 4, which are linked to blastocyst activation (5). In addition to blastocyst factors, HB-EGF also seems to be under estradiol control according to the finding that the LE adjacent to dormant blastocysts exhibits temporal and cell-specific expression of uterine HB-EGF mRNA after the addition of estradiol (4).
The regulated expression of HB-EGF in the human endometrium occurs across the menstrual cycle and implantation sites (6, 7, 8). HB-EGF expression is increased in stromal cells during the late proliferative phase and returns to baseline during the early secretory phase (7). In contrast to rodent glandular and LE, HB-EGF is maximally expressed during cycle d 18–24, which is associated with high peripheral estrogen and progesterone levels and low epithelial progesterone receptor expression (7, 9, 10). HB-EGF mRNA increases in glandular epithelium from d 18–22, as determined by in situ hybridization. This is consistent with reports using RNase protection assays, which show regulation of HB-EGF expression in LE by circulating levels of estrogen and progesterone (8). Culturing of human blastocysts with recombinant HB-EGF significantly increases the number of high-quality embryos and doubles the hatching rate (11). HB-EGF cell signaling accelerates the differentiation of mouse trophoblast cells in vitro to an adhesion competent state by inducing Ca 2+ influx and inducing activation of calmodulin and protein kinase C (12). These and other observations suggest that luminal epithelial-expressed HB-EGF plays a regulatory role in trophoblast differentiation beginning at the attachment stage.
The endometrial component of early implantation in the baboon, like humans, is under complex regulation by progesterone (13). Also, the infusion of human CG (hCG) that mimics blastocyst transport in the baboon has physiological effects on uterine endometrium during the interval of receptivity. These cellular responses support the concept that the blastocyst, as in the mouse, plays an active role in transforming the epithelial barrier into a localized gateway to the stromal cells (14). In primates it is unknown whether progesterone exerts a direct regulatory role for the normal expression of HB-EGF by the endometrium or whether paracrine blastocyst factors, as in the mouse, induce its localized expression. To further explore the role of progesterone regulation or paracrine blastocyst induction of HB-EGF expression in the endometrium during the implantation interval, baboons were treated with antiprogestin ZK 137.316 and/or infused with hCG. In situ hybridization and immunohistochemistry were performed to determine the spatiotemporal localization of HB-EGF mRNA and protein in baboon endometrial tissues under various treatment conditions and early implantation sites. Here, we show that progesterone action during the window of implantation alters HB-EGF accumulation in the epithelium indirectly through stromal cell regulation. Our results also reveal that, unlike the mouse, hCG infusion to mimic blastocyst transit does not result in altered HB-EGF accumulation in endometrial tissues.
Materials and Methods
Tissue collection
Uterine tissue was obtained at laparotomy from adult female
baboons (Papio anubis) as described previously
(14). All procedures were approved by the Animal Care
Committee of the University of Illinois at Chicago. The baboons used in
this study were divided into five groups (n = 3): group A,
cycling; group B, progesterone receptor (PR) antagonist (ZK 137.316)
treated; group C, pregnant; group D, hCG treated (simulated pregnant)
baboons; and group E, simulated pregnant baboons treated with the PR
antagonist (Fig. 1
). The animals in group
A were used to confirm the normal spatiotemporal of HB-EGF in the
endometrium in situ. The animals in group B were used to
confirm whether PR antagonism attenuates HB-EGF expression observed in
group A. The animals in group C were observed to determine the normal
expression pattern of HB-EGF in early implantation sites in
situ. The animals in group D were used to determine whether or not
HB-EGF expression in the endometrium is regulated by infused hCG and
compare these findings with those in group C. Animals in group E were
studied to determine the role of both hCG and antiprogestin treatment
on HB-EGF expression. Ovulation was detected in normally cycling female
baboons by measuring serum estradiol levels, beginning 7 d after
the first day of menses (Fig. 1A). The day of the estradiol surge is
day -1, d 0 is the day of the ovulatory LH surge, and d 1 is the day
of ovulation. Normally cycling animals were treated with ZK137.316 (1
mg/kg per body weight; Schering AG, Berlin, Germany)
daily im commencing the day following the LH surge and
continuing for 9 d (Fig. 1B). Animals in groups D and E were
treated with recombinant hCG in the absence or presence of ZK 137.316,
as described previously (Fig. 1, D and E; Refs. 14 and
15).
|
Uterine tissues were immersion-fixed in Bouin’s solution for 24 h at room temperature, dehydrated in graded ethanol, cleared in xylene, and embedded in paraffin. For histological analyses the tissue sections (6 µm) were stained with Gomori’s trichrome stain (16). Paraffin-embedded sections were cut at 4 µm, deparaffinized, and heated three times for 3 min in a microwave oven set at 80% power. Sections stained for HB-EGF were incubated with 5% rabbit serum for 20 min. Polyclonal goat antibody (IgG) raised against recombinant HB-EGF (R&D Systems, Minneapolis, MN) was used at a final concentration of 10 µg/ml. EGF receptor antibodies HER-1 and -4 were purchased from Oncogene Research Products (Cambridge, MA). For HER-1 staining the sections were subjected to antigen retrieval in 0.01 M sodium citrate buffer (pH 6.0) at 80 C for 30 min. For HER-4 staining the slides were treated with 0.05% saponin for 30 min at room temperature. HER-1 antibody was used at a 1:100 dilution, and the HER-4 antibody was used at 5 µg/ml. All antibodies were applied to the mounted tissue sections for 45 min at 25 C. Negative control sections were incubated with nonimmune goat IgG (Sigma, St. Louis, MO) or neutralized by adding 100 µg/ml recombinant HB-EGF (R&D Systems) to the antibody. The slides were then incubated for 10 min at 25 C with 1:100 biotinylated rabbit antigoat IgG (Jackson ImmunoResearch Laboratories, Inc., West Grove, PA), as appropriate. After rinsing with PBS, the bound antibodies were visualized using avidin-biotin complex with peroxidase (Vector Laboratories, Inc., Burlingame, CA).
Hybridization probe
A 658-bp (nucleotides 303–960, Accession no. M60278) fragment of the cDNA clone for human HB-EGF (7) was inserted into a PstI site of the pGEM-7Zf (+) vector (Promega Corp., Madison, WI) and used as a template for the SP6- (antisense) or T7-directed (sense) 35S-labeled cRNA probes. Probes had specific activities of about 2 x 109 dpm/µg.
In situ hybridization
In situ hybridization of HB-EGF mRNA was performed as described previously (17). Frozen sections of tissue (10 µm) from each specimen were mounted together on poly-L-lysine-coated slides, fixed in 4% paraformaldehyde, and acylated. Samples were incubated for 10 min in PBS containing 5 mM MgCl2, 0.25 M Tris, 0.1 M glycine (pH 7.0), then at 37 C in 50% formamide prepared in 2x SET [1x SET = 150 mM NaCl, 5 mM EDTA, 10 mM Tris-HCl (pH 8.0)]. Hybridization was carried out in a humidified chamber for 5 h at 42 C in 100 µl containing 2x SET; 10x Denhardt’s solution, 50% formamide, 100 mM DTT, 250 µg/ml yeast tRNA, 10% dextran sulfate, 0.2% each of BSA, ficoll and polyvinylpyrrolidone, and 2.0 x 106 cpm/ml of sense (control) or antisense [35S]-cRNA probe. After hybridization, the siliconized coverslips were removed by washing in 4x SSC. Slides were incubated at 37 C for 30 min with 20 µg/ml RNase A and 100 µg/ml BSA in 3x SET. After a 30-min wash in 0.2x SSC and 0.1% mercaptoethanol, the hybridized probe was detected by autoradiography using Kodak NTB-2 liquid emulsion and an exposure time of 5–10 d. The slides were lightly counterstained with hematoxylin and eosin.
Results
Immunohistochemical staining for HB-EGF in normally cycling
baboons showed diffuse HB-EGF accumulation within the epithelial and
stromal compartments during the late proliferative phase (Fig. 2a) and in the glands during the
mid-secretory (postovulatory d10) and declined thereafter (Fig. 2, b–d). HB-EGF accumulation was abundant in the cytoplasm and on the
apical basal cell surfaces of glandular and luminal epithelia during d
10 and 13 postovulation (PO) (Fig. 2, c and d, respectively). In
addition, HB-EGF expression was identified within the endothelium of
stromal arterioles. HB-EGF mRNA localized within endometrium is in
accordance with protein expression. In situ hybridization
signal was diffuse over the stroma in specimens collected from the late
proliferative phase (Fig. 2f) and localized over the glandular
epithelium during the secretory phase (Fig. 2h). Antiprogestin
treatment of normally cycling animals showed a general delay in HB-EGF
accumulation from a mid-secretory to a proliferative pattern. The
glandular epithelium from d 10 PO specimens following 9 d of
antiprogestin treatment showed a marked reduction in HB-EGF
accumulation, which was shifted primarily to the stromal cells (Fig. 3c). This pattern resembled that found in
stroma during the late proliferative phase. This observation was
confirmed by the expression of mRNA that exhibited a more diffuse
hybridization signal over both the glandular and stromal compartments
(Fig. 3c). The infusion of hCG did not alter HB-EGF accumulation in d
10 PO specimens, which remained primarily within the glandular
epithelium (Fig. 3f). The treatment of hCG and antiprogestin again
showed HB-EGF accumulation similar to that seen with antiprogestin
treatment alone. The protein and mRNA pattern of expression shows
colocalization of HB-EGF mRNA in the glandular epithelium in CG-treated
animals (Fig. 3f) and more diffuse stromal expression with the addition
of antiprogestin (Fig. 3i). In agreement with hCG-treated animals,
early implantation sites (d 18 and 25) showed accumulation of HB-EGF
around the glands (Fig. 3, j and m). HB-EGF mRNA also showed a diffuse
signal pattern over the stroma and glands (Fig. 3, l and o). HER1 gave
distinctive cell localization across the menstrual cycle with
accumulation in glandular epithelium during the late follicular phase
(Fig. 4a) and both stroma and epithelia
during the mid-secretory phase (Fig. 4b). HER1 primarily accumulated in
trophoblast cells in early implantation sites (Fig. 4, c and d). HER 4
was not identified in pregnant or nonpregnant specimens (data not
shown).
|
|
|
The synchronous development of the endometrium during early implantation in the nonhuman primate and rodent is highly dependent on embryonic-maternal signaling. The morphological transformation of the endometrium to a secretory stage requires the action of progesterone in an estrogen-exposed endometrium. Specifically, progesterone binding to its receptor (PR) promotes induction or repression of a cascade of response genes that prepare the epithelium for implantation (18, 19). There is growing evidence from nonhuman primates and rodents that the embryo also plays an active role by inducing the receptive epithelial phenotype (4, 16, 17, 20). Secretion of CG by the primate embryo not only results in rescue of the corpus luteum for continued progesterone support of the endometrium, but also directly interacts with CG receptors in the endometrium. This activation results in morphological and biochemical changes within the implantation-competent endometrium (14, 21).
Initially, we identified the maternal component that regulates HB-EGF expression during the implantation window in normally cycling and antiprogestin-treated animals. In accordance with results from the previous analysis of human uterine tissue (7), immunohistochemical staining and in situ hybridization analysis revealed elevated expression of HB-EGF in stromal and epithelial cells during the proliferative and secretory phases, respectively. As observed in human tissues, the same HB-EGF mRNA expression colocalized with protein in a cell-specific pattern reflecting de novo synthesis. The observed nonuniform pattern of immunohistochemical staining in both epithelial and stromal cell compartments is not uncommon (22) and has also been reported in the rhesus monkey (23). The opposing HB-EGF accumulation patterns within epithelial and stromal cells between d 5 and 10 PO, during peak progesterone production, imply a role for progesterone regulation. Specifically, progesterone plays a dominant role in suppressing HB-EGF accumulation in stromal cells during the secretory phase. The increased stromal cell HB-EGF expression in baboons treated with the PR antagonist ZK137.316 provides supporting evidence for suppressive regulation by progesterone. The delay in epithelial HB-EGF expression during the secretory phase, changing its expression to a proliferative phase pattern, is consistent with a morphological delay observed in ZK137.316-treated baboons and RU486-treated humans (15, 24). The direct regulatory role of progesterone for epithelial cells is more difficult to reconcile, because PR levels are diminished during the secretory phase (7, 10). The decreased HB-EGF accumulation in baboon endometrial epithelial cells after PR antagonist treatment would support the hypothesis that a paracrine factor from stromal cells regulates epithelial HB-EGF expression during the secretory phase. Coculture experiments using reconstituted epithelial and stromal cells isolated from estrogen and PR knockout mice have provided evidence for the paracrine regulation of uterine epithelial cells by stromal cell factors (25, 26). Alternatively, potential corpora luteum paracrine effects on epithelial cells in baboon after treatment with ZK137.316 have been observed (15).
HB-EGF binding to the blastocyst requires the presence of not only the EGF receptor but also surface heparan sulfate proteoglycan known to increase during the attachment competent stage (27, 28, 29). There are four known members of the EGF growth factor receptor tyrosine kinases (HER-1 and -4). HER-1, but not HER-4, was localized by immunohistochemistry in the baboon epithelium during the proliferative and secretory phases and in the stroma only during d 5 and 10 PO. This spatiotemporal pattern differs from the mouse, in which neither HER-1 nor -4 is expressed in the epithelium, but resembles the expression of HER-1 in the stroma during implantation (5). HB-EGF binds to NIH 3T3 cells over expressing HER-1 or -4, individually, but not to cells expressing only HER-2 or HER-3 (30). HB-EGF has been shown to have greater binding affinity for HER-1 than does EGF in smooth muscle cells (1). This differing affinity of HB-EGF for HER subtype activation results in different cell responses. NIH 3T3 cells expressing only HER-1 respond to HB-EGF with increased cell proliferation and chemotaxis, whereas cells expressing only HER-4 respond with an increase in chemotaxis without cell proliferation (30). The ability of HER-1 and -4 to transduce different cellular responses suggests that multiple homologous receptors provide additional levels of control for growth factors. Our observation of predominant epithelial expression of HER-1 during the proliferative phase, and to a lesser extent during the implantation interval, suggests its direct involvement in epithelial cell proliferation. Furthermore, high HER-1 expression levels in the stroma during the implantation interval suggests a role in stromal cell transformation into decidual cells.
Factors produced by the primate embryo modulate endometrial morphology and induce molecular changes that promote endometrial receptivity. In the rhesus monkey the endometrial physiology in the presence of an embryo replaced at in vitro fertilization is different than normally cycling animals, despite no differences in peripheral blood levels of estradiol or progesterone (31). In the baboon, infusing CG into the fallopian tube to mimic embryo transit results in a morphological response within the endometrium that includes the plaque reaction and molecular changes that resemble early pregnancy (14). The molecular changes include increased a smooth muscle actin, IGF-binding protein-1, cyclooxygenase-2, and glycodelin, all thought to play a pivotal role in early implantation (21, 32).
We set out to determine whether the embryonic paracrine factor CG also influenced HB-EGF expression within the uterus during early implantation. The localized expression of HB-EGF within the glandular and stromal compartments on d 10 PO in animals infused with hCG did not differ from nontreated animals. This finding was confirmed in normal pregnant baboons sampled during a comparable interval of gestation. The addition of antiprogestin to hCG-treated baboons revealed a HB-EGF accumulation pattern within the epithelium and stroma resembling that found in antiprogestin-treated only baboons. One possible explanation for these findings is that HB-EGF accumulates within the epithelial cells during the window of receptivity and early implantation under progesterone regulation with no apparent direct paracrine regulation by the embryo. However, the presence of CG indirectly regulates HB-EGF accumulation by rescuing progesterone secretion from the corpus luteum. This observation opposes the situation observed in the mouse, where HB-EGF signaling is the earliest known molecular interaction between the trophectoderm and the LE that is directly regulated by the attaching embryo (4).
Intense accumulation of HER-1, but not HER-4, was observed in baboon cytotrophoblasts on PO d 18 and 25 pregnancies. Human trophoblasts differentially express HER-1 and HER-2 depending on their differentiated state (33). In human first trimester pregnancies, HER-1 is most heavily expressed in proliferative (Ki-67 positive) villous and extravillous cytotrophoblast. Conversely, HER-2 is expressed on differentiated structures, such as extravillous syncytiotrophoblast and distally located extravillous cytotrophoblasts (Ki-67 negative). This differs from mouse in which all HER receptors are expressed by implantation-competent embryos (12). In the delayed pregnancy model, HER-1 gene expression is down-regulated and returns to the blastocyst after the administration of estrogen to initiate nidation similar to HB-EGF accumulation with in the endometrial epithelium (34). HER-4 translocates from the cytoplasm to the apical surface of trophoblast cells late on gestation d 4 and is the prime candidate for the high-affinity HB-EGF receptor (29). HB-EGF induces both intracellular Ca2+ signaling and accelerated trophoblast development to an adhesion-competent stage only after HER-4 translocation has occurred (12). These findings suggest that HER-1 plays a different spatiotemporal networking role in the primate than has been observed in the mouse and requires further study.
The key finding of this study is that expression of HB-EGF during the implantation interval is increased in epithelial cells, probably through a putative progesterone-dependent, stromal-derived paracrine factor, whereas it is suppressed in stromal cells under the direct influence of progesterone. This pattern of steroid regulation, which expresses little epithelial HB-EGF before blastocyst attachment, is similar to human but unlike the rodents. Importantly, HB-EGF expression in the LE at the time of attachment in both primates and rodents is under the control of stromal factors or the attaching blastocyst, respectively. These fundamental differences in HB-EGF and HER-1 and -4 regulation between groups provides insight into the different strategies rodents and primates use to facilitate early implantation. The mouse blastocyst regulation of HB-EGF expression locally may represent a strategy by which rodents can regulate multiple nidation site spacing within the bihorned uteri not required in primates (35). Furthermore, the expression of HER-1 receptor in primate trophoblasts manifests a proliferative phenotype compared with the mouse that primarily undergoes a differentiated phenotype by activation of HER-4. The accumulation of the HB-EGF receptor HER-1 in both the endometrium and the trophoblast cells suggests that HB-EGF has a wide range of effects throughout developing uterine and placental tissues. HB-EGF expression during early pregnancy is not directly regulated by CG, suggesting that other regulatory pathways are involved.
Acknowledgments
Footnotes
This work was supported by the Women’s Reproductive Health Research Center, funded by NIH Career Development Agreement HD98004 (to R.E.L.), and The National Cooperative Program for Markers of Uterine Receptivity for Blastocyst Implantation, funded by NIH Cooperative Agreements HD29964 and HD29968 (to A.T.F. and S.K.Da.). This work was presented at the 47th Annual Meeting of the Society for Gynecologic Investigation, March 22–25, 2000, Chicago, Illinois (Abstract 21).
Abbreviations: HB-EGF, Heparin-binding EGF-like growth factor; hCG, human CG; LE, luminal epithelium; PO, postovulation; SET, 1x SET = 150 mM NaCl, 5 mM EDTA, 10 mM Tris-HCl (pH 8.0).
Received October 30, 2000.
Accepted May 29, 2001.
References
smooth muscle actin
expression in the baboon (Papio anubis) uterus is associated
with glandular secretory activity and stromal cell transformation. Biol
Reprod 53:598–608[Abstract]
This article has been cited by other articles:
 |
|
 |
|
  D. R. Armant, B. A. Kilburn, A. Petkova, S. S. Edwin, Z. M. Duniec-Dmuchowski, H. J. Edwards, R. Romero, and R. E. Leach Human trophoblast survival at low oxygen concentrations requires metalloproteinase-mediated shedding of heparin-binding EGF-like growth factor Development, February 15, 2006; 133(4): 751 - 759. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Kumar, A. Brudney, Y.-P. Cheon, A. T. Fazleabas, and I. C. Bagchi Progesterone Induces Calcitonin Expression in the Baboon Endometrium Within the Window of Uterine Receptivity Biol Reprod, April 1, 2003; 68(4): 1318 - 1323. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Carver, K. Martin, I. Spyropoulou, D. Barlow, I. Sargent, and H. Mardon An in-vitro model for stromal invasion during implantation of the human blastocyst Hum. Reprod., February 1, 2003; 18(2): 283 - 290. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Chobotova, M.-E. Muchmore, J. Carver, H.-J Yoo, S. Manek, W. J. Gullick, D. H. Barlow, and H. J. Mardon The Mitogenic Potential of Heparin-Binding Epidermal Growth Factor in the Human Endometrium Is Mediated by the Epidermal Growth Factor Receptor and Is Modulated by Tumor Necrosis Factor-{alpha} J. Clin. Endocrinol. Metab., December 1, 2002; 87(12): 5769 - 5777. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Stavreus-Evers, L. Aghajanova, H. Brismar, H. Eriksson, B.-M. Landgren, and O. Hovatta Co-existence of heparin-binding epidermal growth factor-like growth factor and pinopodes in human endometrium at the time of implantation Mol. Hum. Reprod., August 1, 2002; 8(8): 765 - 769. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
