|
|
||||||||
a Departments of Obstetrics and Gynecology and
b Physiology and Biophysics, Division of Reproductive Endocrinology and Infertility, Olson Center for Women's Health, University of Nebraska Medical Center, Omaha, Nebraska 681984515
| ABSTRACT |
|---|
|
|
|---|
35-yr-old women with either tubal defects or mild endometriosis who were undergoing controlled ovarian hyperstimulation prior to oocyte retrieval for in vitro fertilization. TßRI and TßRII were immunoprecipitated from 100 000 g soluble and crude membrane fractions using receptor-specific antibodies and analyzed by Western immunoblotting, and the relative expression was quantitated from the luminographs. The gross morphology (embryo grade) of the preimplantation embryos developed in vitro was determined using a stereomicroscope. Both TßRI and TßRII are expressed in the soluble and membrane fractions of granulosa cells. Most notably, the zygote always developed into a grade 1 quality preimplantation embryo when the oocyte originated from a follicle that expressed a low amount of TßR protein in the granulosa cell membrane. Reduced expression of TßR in the granulosa cell membrane may form a mechanism critically regulating TGFß action on granulosa cells, and the latter in turn precisely control oocyte development, hence, the subsequent cleavage potential.
| INTRODUCTION |
|---|
|
|
|---|
| MATERIALS AND METHODS |
|---|
|
|
|---|
The use of human granulosa cells from the IVF clinic was approved by the Institutional Review Board, and the procedure of egg retrieval was standard for assisted reproductive technology. An average of 4.3 ± 0.6 follicles (
18 mm with a healthy oocyte) per subject were aspirated from a total of 18 women with a mean age
35. These women were healthy donors with either tubal defect, mild endometriosis, or pelvic adhesion and were undergoing controlled ovarian hyperstimulation prior to oocyte retrieval for IVF. Granulosa cells were harvested by ultrasound-guided transvaginal follicular aspiration. After pituitary desensitization by leuprolide acetate (hormonal suppression; Lupron; TAP Pharmaceutical, North Chicago, IL), follicular development was induced by FSH (Pergonal or Metrodin; Serono, Randolph, MA). Granulosa cells were collected from mature ovarian follicles that supported the development of a cumulus-oocyte complex and were determined by ultrasound to be
18 mm in diameter at the time of follicular aspiration. Each follicle was aspirated per vagina using an aspiration needle. After initial aspiration of the follicular fluid containing the oocyte, the cannula was flushed with ~5 ml medium, and the flushed medium was checked under the microscope for granulosa cells to ensure a thoroughly cleaned needle before aspirating another follicle.
Cumulus-oocyte complexes (COCs) were graded microscopically as early, transitional, intermediate, or mature. The amount of cumulus expansion around the oocyte was used to grade the complex. Oocytes with no cumulus expansion were assigned a grade of early, and those with full cumulus expansion were graded as mature. Complexes graded as transitional and intermediate exhibited some and almost full cumulus-oocyte expansion, respectively. Each COC was removed from the follicular aspirate using a 1-ml tuberculin syringe with a Unopette catheter attachment, washed several times in Hepes-buffered human tubal fluid (mHTF; Irvine Scientific, Santa Ana, CA), and then transferred to the insemination medium (HTF; Irvine Scientific) containing 7.5% synthetic serum substitute (SSS, Irvine Scientific). All COCs were cultured individually. At this time, each follicular aspirate was placed on ice until granulosa cell isolation took place (
1 h).
Granulosa cells from each follicle were handled separately. Using a syringe attached to a Unopette catheter, granulosa cells from each follicular aspirate were transferred into a Falcon 1007 Petri dish (Becton Dickinson, Lincoln Park, NJ) containing 6 ml protein-free mHTF and washed 10 times by repeat pipetting. The entire wash protocol was repeated twice; cells were pelleted at 800 x g for 2 min and stored at -80°C until immunoprecipitation and immunoblotting were performed.
Analysis of TßRI and TßRII
Granulosa cell pellets were sonicated in 200 µl ice-cold 10 mM Tris-HCl, pH 7.0, containing 10% glycerol, 0.25 M sucrose, 200 µM sodium vanadate, 1 mM PMSF, and 10 µl/ml of a protease inhibitor cocktail comprising aprotinin, soybean trypsin inhibitor, EDTA, leupeptin, antipain, and benzimidine-HCl; they were then centrifuged at 100 000 x g for 30 min to separate soluble and particulate fractions as described previously [11, 12]. The pellet was solubilized in 200 µl of single-strength RIPA buffer (PBS, pH 7.0, with 0.1% SDS, 1% Nonidet P-40, and 5% sodium deoxycholate) on ice for 30 min, and detergent-soluble membrane proteins were isolated at 26 000 x g for 20 min at 4°C. Protein content of both soluble and membrane fractions was determined by a silver-gold-based protein assay kit (ISS, Natick, MA). For both TßRI and TßRII, 100 µg of protein of the soluble fraction was mixed with ice-cold PBS and 100 µg of membrane protein was mixed with ice-cold single-strength RIPA to a final volume of 400 µl for immunoprecipitation. TßRI and TßRII proteins were immunoprecipitated essentially as described by Roy and Kole [12]. Briefly, the samples were clarified by mixing with 1 µg of nonimmune rabbit IgG and 20 µl protein A/G agarose (Pierce Chemical Co., Rockford, IL) at 4°C for 4 h; the agarose beads were removed by centrifugation at 2000 x g. Nonspecific immunoprecipitation was always undetectable, and no difference was observed after extension of the clarification up to 18 h. Therefore, a 4-h clarification was always used. Clarified samples were mixed overnight with 1 µg of either anti-TßRII or anti-TßRI antibody (polyclonal; Santa Cruz Biotechnology, Santa Cruz, CA) and 20 µl of protein A/G agarose (Pierce) at 4°C [12]. The specificity of the antibodies was validated by immunocytochemistry and Western immunoblotting [12]. The immunoprecipitates were rinsed three times with single-strength RIPA at 4°C, mixed with 20 µl SDS electrophoresis buffer, heated for 2 min in a boiling water bath, and fractionated in a 7.5% polyacrylamide gel. Fractionated proteins were electrotransferred to a Immobilon membrane (Millipore, Bedford, MA) and subjected to immunodetection using the same receptor-specific antibodies and enzyme chemiluminescence (Amersham Corp., Arlington Heights, IL) as described by Roy and Kole [11, 12].
Analysis of Luminographs
The x-ray films were scanned in a Molecular Dynamics (Sunnyvale, CA) laser densitometer to digitize the image generated by the immunoprecipitated receptors. The lane containing no sample was digitized and the values were subtracted from the sample data as background. The data were expressed as mean ± SEM (n =
3 for each grade of embryo) and analyzed by Student's t-test with 5% level of significance. Representative luminographs of TßRI and TßRII were digitized to obtain a quantitative relationship between cytosolic and membrane forms. A membrane:cytosol ratio for the optical densities (ODs) was derived and plotted against the embryo grades to evaluate any relationship between the receptor expression and embryo cleavage potential. Immunoprecipitation and immunoblotting were repeated three times using different follicle samples to ensure reproducibility.
IVF and Preimplantation Embryo Development
Each COC was inseminated within 46 h after follicular aspiration with 100 000200 000 sperm per egg per milliliter depending upon sperm quality and cumulus-oocyte expansion. Male factor was not a consideration for this study. Fertilization was determined at 1618 h postinsemination, and the zygotes were transferred to growth medium (HTF; Irvine Scientific) containing 15% SSS (Irvine Scientific). In vitro embryonic development continued for 4852 h. The resulting embryos were graded on a scale of 1 to 3, based on their morphology, and were either transferred to the uterus or the fallopian tube or cryopreserved. The criteria for embryo gradation are presented in Table 1; there were
3 embryos in each grade.
|
| RESULTS |
|---|
|
|
|---|
|
Anti-TßRII antibody precipitated an 87-kDa protein in the membrane as well as in the soluble fractions (Fig. 2A); however, the amount of membrane-associated receptor protein was barely visible on the luminograph for follicles giving rise to oocytes that developed into grade 1 preimplantation embryos (Fig. 2A). Whereas the amount of cytosolic form did not differ noticeably between groups representing different grades of cleavage-stage embryos, expression of the membrane-associated form increased significantly (data not shown), resulting in embryo grade-associated increases in the OD ratios (Fig. 3). The OD ratio appeared to plateau with grade 3 embryos (Fig. 3). Upon digitization, a significant difference (p < 0.05) in TßRII expression between the membrane and cytosolic compartment was evident only for follicles that produced eggs with grade I embryonic potential (Table 2). When the OD ratios of membrane to cytosol for TßRII were plotted against the grades of embryo, a grade-specific increase was observed up to grade 3, and a significant (p < 0.05) difference in OD ratios was observed for follicles corresponding to grade 1 and 2 embryos compared to the other grades (Fig. 3).
|
|
|
Hyperstimulated human granulosa cells also expressed a 75-kDa TßRI both in the cytosol and in the cell membrane (Fig. 2B); however, the membrane-associated form of TßRI expression showed an upward trend as the quality of cleavage-stage embryos declined (Table 2). Furthermore, granulosa cells of follicles that produced oocytes with grade 1 and grade 2 (Fig. 1) embryonic potential had less receptor protein in the membrane compared to other groups (Table 2). The expression of the cytosolic form was always greater than that of the membrane form. However, the OD ratio of membrane to cytosol for TßRI did not change appreciably until embryo quality reached grade 3, when a sharp increase was noted (Fig. 3). Moreover, a differential expression of TßRI and TßRII in the granulosa cells was observed when the OD ratio was correlated with the quality of the cleavage-stage embryos (Fig. 3).
| DISCUSSION |
|---|
|
|
|---|
The women in the present study received a GnRH analogue, Lupron, to down-regulate pituitary gonadotropin secretion. GnRH has been shown to attenuate TßR receptor and protein expression in leiomyomata [17], and the presence of TGFß immunoreactivity has been demonstrated in surgically induced endometriosis [18] and pelvic adhesions [19] in the rat; however, such information is not available for the human. Nevertheless, questions can arise whether women with pelvic adhesions or endometriosis produce substantial TGFß, which can affect ovarian follicular development, or whether GnRH influences ovarian TßR expression. In our study, the formation of grade 1 embryos was randomly distributed among the women regardless of the clinical conditions, and all women received a similar dose of GnRH. Therefore, it is unlikely that any of these factors have a significant effect on granulosa cell TßR expression. Moreover, the decrease in the level of membrane receptor was not due to classical ligand-mediated receptor down-regulation, because follicular fluid of women treated with exogenous gonadotropins contains very high levels of TGFß1 regardless of the clinical conditions mentioned in Materials and Methods (unpublished results).
The unique correlation between TßR expression in granulosa cell membrane and subsequent cleavage of the zygotes indicates that presentation of TßR in the granulosa cell membrane may form a key determinant in overall follicular and oocyte development. Both TßRI and TßRII must be expressed in the cell membrane in order for the biological action of TGFß to occur [20, 21]. A novel inverse relationship between membrane and cytosolic forms of ovarian TßRII during the estrous cycle has been demonstrated in the hamster [12]. The expression of TGFß1 and TGFß2 ligands has been demonstrated in the granulosa and theca cells of a variety of species [2227], including humans [15, 28], and their induction by gonadotropins has been documented [24, 26]. TGFß has a wide variety of actions on ovarian cells [3, 29, 30].
In typical controlled ovarian hyperstimulation protocol, exogenous FSH overrides the natural follicle selection process by recruiting many antral follicles for ovulation and by preventing apoptosis leading to supernumerary follicular development [3]. Therefore, follicles that would otherwise undergo atresia are stimulated to develop. Upon retrieval, oocytes may appear morphologically normal and undergo fertilization, yet many of the zygotes do not undergo normal cleavage in vitro. Healthy follicular development is a critical prerequisite for oocyte growth, and granulosa cell-oocyte communication determines the final outcome of the growth process [31]. Premature or overt differentiation of granulosa cells may severely alter the microenvironment long before the oocyte can respond to such change(s), resulting in oocytes with impaired physiology. These oocytes may fail to undergo normal cleavage after fertilization. Our results indicate such a potential.
Because TGFß affects granulosa cell function [29, 32] and its function requires membrane receptors [33], increased expression of TßRI and TßRII in granulosa cell membrane may result in enhanced TGFß action and premature alteration in cell function. We have also detected very high levels of TGFß1 in the follicular fluid from hyperstimulated follicles (unpublished data). Recently, Juneja et al. [34] have demonstrated that intrabursal injection of TGFß1 in the mouse results in a dose-dependent inhibition of hCG-induced ovulation. Moreover, TGFß1 has been shown to inhibit IVF and in vitro preimplantation development in mice [35]. One of the major mechanisms whereby antral follicles rapidly increase their diameter is increased accumulation of antral fluid, which contains secretions from differentiated granulosa cells [3]. Increased expression of TßR in granulosa cells of immature antral follicles may induce premature cell differentiation with resultant increase in antral fluid accumulation and follicular diameter. Moreover, oocytes in these large but otherwise immature follicles may not be biologically mature for normal postfertilization development, even though they can undergo fertilization in vitro. Fertilization of immature human oocytes has been documented [36]. Excess exogenous gonadotropins in typical IVF protocols may induce a similar situation. The functional significance of high levels of TßR in the soluble fraction is not yet known; however, it may reflect an important regulatory step in TGFß action on follicular cells. Unlike TGFß receptor type III (betaglycans), TßRI and TßRII are not secreted as truncated protein [37].
In summary, we have presented evidence for the presence of both type I and type II receptors for TGFß in hyperstimulated human granulosa cell membrane and their potential role in the quality of oocyte development. Future studies are needed to advance understanding of the hormonal regulation of TßR expression in human granulosa cells.
| FOOTNOTES |
|---|
2 Correspondence: Shyamal K. Roy, Departments of OB/GYN and Physiology/Biophysics, University of Nebraska Medical Center, BH 5005, 600 S. 42nd Street, Omaha, NE 681984515. FAX: 402 559 6164. ![]()
3 Current address: Amy M. Carlson, Department of Psychiatry, University of Colorado Health Sciences Center, Denver, CO 80220. ![]()
Accepted: July 13, 1998.
Received: February 7, 1998.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
J.L. Juengel and K.P. McNatty The role of proteins of the transforming growth factor-{beta} superfamily in the intraovarian regulation of follicular development Hum. Reprod. Update, March 1, 2005; 11(2): 144 - 161. [Abstract] [Full Text] [PDF] |
||||
![]() |
K.L. Larson, C.J. DeJonge, A.M. Barnes, L.K. Jost, and D.P. Evenson Sperm chromatin structure assay parameters as predictors of failed pregnancy following assisted reproductive techniques Hum. Reprod., August 1, 2000; 15(8): 1717 - 1722. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. K. Roy and D. M. Terada Activities of Glucose Metabolic Enzymes in Human Preantral Follicles: In Vitro Modulation by Follicle-Stimulating Hormone, Luteinizing Hormone, Epidermal Growth Factor, Insulin-Like Growth Factor I, and Transforming Growth Factor ß1 Biol Reprod, March 1, 1999; 60(3): 763 - 768. [Abstract] [Full Text] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |