| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
The Jones Institute for Reproductive Medicine, Department of Obstetrics and Gynecology (S.M., J.-G.H., S.O.) and Department of Pathology (J.D.), Eastern Virginia Medical School, Norfolk, Virginia 23507; and Second Department of Obstetrics and Gynecology (G.N.), Aretaieio University Hospital, Athens 16675, Greece
Address all correspondence and requests for reprints to: Sergio Oehninger, M.D., Ph.D., The Jones Institute for Reproductive Medicine, 601 Colley Avenue, Norfolk, Virginia 23507. E-mail:oehninsc{at}evms.edu.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The blastocyst and endometrium generate a variety of factors that appear to be of functional significance during implantation (4). However, it is likely that the molecular cross-talk among those factors involves many more yet- unknown molecules. Indeed it is more realistic to view the process of implantation as a condition of equilibrium in the up-regulation and down-regulation of diverse endometrial genes under control of steroid hormones and probably other local (paracrine and autocrine) regulatory factors (5, 6, 7, 8).
In women undergoing in vitro fertilization and embryo transfer (IVF–ET), implantation may fail, even when embryos of apparent good quality (based upon cleavage and morphological assessments) are transferred in either gonadotropin-stimulated or estrogen (E2)/progesterone (P4)- supplemented cycles for egg donation or cryopreserved-thawed ET cycles. Thus, although we have now become quite adept at many of the steps of the IVF–ET procedure, including controlled ovarian hyperstimulation (COH), oocyte retrieval, sperm preparation, gamete in vitro culture, and ET, more information is needed as to the optimization of embryo quality and efficiency of implantation both at cellular and molecular levels (9). On average, approximately 70% of apparently healthy embryos transferred in utero vanish, giving no signs of trophoblastic attachment and production of human chorionic gonadotrophin (hCG) (10). This brings up the question of embryonic vs. endometrial factors as determinants of implantation success/failure.
It has been speculated that a possible cause for the rates of implantation observed in IVF cycles might be an impairment of endometrial receptivity due to high concentration of sex steroids resulting from COH (11). Histological advancement of endometrial glandular/stromal compartments has been a common feature of IVF cycles stimulated with gonadotropins, with and without use of adjuvant therapy with GnRH agonists (12, 13). Furthermore, recent data suggested that the use of GnRH antagonists in some COH regimens might result in compromised implantation rates (14).
The initial step of blastocyst attachment to the luminal epithelium of the endometrial surface is thought to involve the formation of specialized cellular apical protrusions called pinopodes (10). The formation of pinopodes follows a specific pattern: In a normal cycle, a pronounced cell bulging appears on d 19, and microvilli fuse to form a smooth, slender membrane projection arising from the entire cell apex (developing pinopodes). On d 20 and 21, the microvilli are absent, and the membrane protrudes and folds maximally (fully developed pinopodes). On d 22, bulging decreases, and small tips of microvilli reappear on the membrane, which appear wrinkled (regressing pinopodes). By d 23, the pinopodes have largely disappeared. Fully developed pinopodes last no more than 2 d (10), and they have been proposed as ultrastructural markers of the window of implantation (15). There is some evidence that E2 and P4 are involved in the regulation of pinopodes formation and maturation via action on specific E2 receptor (ER) and P4 receptor (PR) (16). Recently, it has been shown that down-regulation of PR is concurrent with development of pinopodes in the endometrium at the time of implantation (17).
Understanding the molecular events underlying the development and maintenance of a receptive endometrium is fundamental if we are to understand and further improve the success of embryo implantation during IVF therapy. In the pregenomic era, a one-by-one approach has been adopted to investigate genes expressed in the window of implantation. A genomic-wide approach using microarray technology now allows us to investigate global gene expression during the implantation window of IVF-ET cycles.
Few studies have addressed endometrial gene expression during the luteal phase and putative window of implantation in natural cycles. As expected, multiple differential gene expression was found, comparing the follicular vs. secretory phases of the cycle as well as the early vs. the midluteal phases (18, 19, 20, 21). The functional significance of these findings, however, remains to be established at the cellular and protein/molecular levels. To the best of our knowledge, gene expression has not been previously examined during COH cycles as performed for IVF treatment.
The objectives of this prospective and randomized clinical study were: 1) to investigate the endometrial gene expression profiles during the putative window of implantation of state-of-the-art COH protocols, i.e. cycles stimulated with recombinant FSH (rFSH), using either GnRH agonist or GnRH antagonist, with and without P4 supplementation of the luteal phase; 2) to compare gene expression profiles of COH cycles with those of temporally matched natural cycles; and 3) to relate gene expression profiles with other structural and functional features of the endometrium, including histological appearance, pinopodes, and ER and PR content. To eliminate the potential bias associated with individual ovarian responses to COH and uterine factors seen in infertile patients, we studied a homogenous group of healthy, fertile, normally cycling women participating in our egg donation program (egg donors).
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
A total of 15 female subjects participated in these studies. These were women participating in our egg donation program, ranging in age between 24 and 32 yr, with regular menstrual cycles and previously confirmed ovulation, and who were of proven fertility. All donors had a normal uterus as assessed by transvaginal ultrasonography. Donors were screened and enrolled in the egg donation program, following the guidelines of the American Society for Reproductive Medicine as previously described (22). All volunteers signed an informed consent approved by the Institutional Review Board at Eastern Virginia Medical School. Donors were requested to use condom contraception during the preceding and study cycles.
Studied groups and timing of endometrial biopsies
Fifteen subjects, who were active donors in our egg donation program and who agreed to participate, were randomized (using sealed envelopes) to the following three groups: 1) group I, donors undergoing natural cycles (n = 5); 2) group II, donors (n = 7) who were stimulated with rFSH and GnRH antagonist adjuvant therapy. Donors received either Follistim (Organon Inc., West Orange, NJ) and ganirelix (Antagon, Organon Inc.) or Gonal-F (Serono Laboratories, Randolph, MA) and cetrorelix (Cetrotide, Serono Laboratories). To assess the effect of exogenous P4 supplementation, five donors in this group received micronized P4 (Prometrium, 200 mg intravaginally three times daily; Solvay, Marietta, GA) starting 2 d after oocyte retrieval (IIa), whereas two donors did not receive any P4 supplementation (IIb). 3) group III, donors (n = 3) who were stimulated using down-regulation with a long GnRH agonist protocol (Leuprolide Acetate, Lupron, Tap Pharmaceuticals, Abbott Park, IL) and rFSH (Gonal F, Serono Laboratories), with supplementation of the luteal phase with micronized P4 as described for group II. This was performed to compare the impact of the use of a GnRH antagonist vs. down-regulation with a GnRH agonist.
Ovarian stimulation protocols were carried out following previously published guidelines (22). Briefly, the starting gonadotropin dose was 150–225 IU/d, with the dose being adjusted in an individual fashion using a step-down regimen. The GnRH antagonist (0.25 mg/d) was started when the leading follicle(s) reached 14 mm. The GnRH agonist was commenced on d 21 of the preceding luteal phase (0.5 mg/d), dropped to 0.25 mg/d at menses, and continued until hCG administration. Ten thousand units of hCG were given im in all COH cycles.
Donors underwent an endometrial biopsy on d 21 (during the putative window of implantation) of temporally matched natural or COH cycles. Daily assessment of the urinary LH surge beginning cycle d 10 was performed by the volunteers (natural cycles), using a commercial available ovulation predictor kit (Ovuquik One Step; Quidel, San Diego, CA). Natural cycle endometrial biopsies were performed on d 21 (LH +8, where LH = 0 is the day of the urinary LH surge and LH +1 is the day of ovulation or d 14) (18, 20) All COH cycles were gonadotropin-stimulated cycles using rFSH preparations only and with hCG administered to trigger ovulation (22). COH endometrial biopsies were also performed on equivalent d 21, established as hCG d +9 (where the day of egg retrieval is d 14 or equivalent to the day of ovulation in a natural cycle) (16).
Endometrial tissue specimens
Endometrial biopsies were performed using a Pipelle catheter (Unimar, Bridgeport, CT) under sterile conditions, from the uterine fundus. Extreme care was taken to reassure that enough tissue was obtained from each biopsy. Each sample was divided into three portions, one of which was fixed in 10% formalin and processed for histologic evaluation (hematoxylin-eosin, H-E) and immunohistochemical analysis for ER and PR. A second portion was processed for pinopodes detection using scanning electron microscopy (SEM). Samples for SEM were fixed in a solution containing 2.5% (wt/vol) glutaraldehyde solution in a sodium cacodylate buffer (0.1 mol/liter, pH 7.3). A third portion was immediately (within a minute) frozen in liquid nitrogen for subsequent RNA isolation.
Histologic evaluation and immunohistochemistry for ER and PR
For endometrial dating, 4-µm sections stained with H-E and periodic acid–Schiff stain were evaluated. All endometrial biopsies were coded and evaluated blinded by an experienced pathologist (J.-G.H.) according to the histopathological criteria of Noyes et al. (23). All the biopsies were also used for immunocytochemical analysis. The biopsy samples were embedded in paraffin and cut into 4-µm sections. The sections were dewaxed in Bioclear (Bio-Optica, Milan, Italy) and rehydrated in decreasing concentrations of ethanol and, finally, distilled water. Sections were pretreated in 0.01 mol/liter citric buffer in a microwave oven. Endogenous peroxidases were subsequently blocked by 3% hydrogen peroxide in methanol for 10 min. Sections were washed in PBS, covered in 100 µl blocking serum, and incubated for 30 min. Normal horse serum was used as blocking serum for PR (A and B) and ER. The sections were then incubated overnight with the primary antibody at 4 C.
The ER antibody, reactive with both 
and ß ER, was a second-generation monoclonal mouse antihuman ER antibody (08–1149; Zymed Laboratories, Inc., San Francisco, CA). The monoclonal mouse antihuman PR A and B antibody (MAI-410) was purchased from Affinity Bioreagents, Inc. (Golden, CO). Normal mouse IgG was used as negative controls for ER and PR. Slides were washed in PBS with 0.1% Triton X100. Biotinylated horse antimouse antibody (Vector Laboratories, Burlingame, CA) was used as a secondary antibody for PR and ER. After incubation for 30 min and washing, the sections were incubated with a freshly prepared solution of horseradish peroxidase-avidin-biotin complex (Vectastain ABC Elite; Vector Laboratories) for 30 min. After washing, the site of bound enzyme was visualized, when incubated with the enzyme in the presence of 3,3-diaminobenzidine in H2O2 (DAB Kit; Vector Laboratories), a chromogen that produces a brown insoluble precipitate when incubated with enzyme. Slides were washed in PBS with 0.1% Triton X100. The sections were counterstained with hematoxylin and dehydrated before mounting with Pertex (Histolab, Gothenburg, Sweden).
Nuclear ER and PR expression were evaluated in glandular and stroma endometrial compartments. Staining was evaluated semiquantitatively by using a grading system. The number of stained cells was graded on a scale of: 0 = 0 stained cells, + = 1–25% stained cells, 2+ = 26–50% stained cells, 2++ = 51–75% stained cells, and 2 + 2+ = 76–100% stained cells (16). One experienced observer, who was blinded to the identity of the slides, performed all the assessments (J.-G.H.). After completion of the study, the same observer reexamined the slides to ensure reproducibility of the semiquantitative assessment.
SEM
The specimens were dehydrated in an acetone series, dried in a critical point drier using carbon dioxide, mounted on the specimen holder, coated with gold, and examined under a Stereoscan 360 SEM (Cambridge Instruments, Cambridge, UK). From each biopsy, between six to eight tissue fragments, each measuring approximately 2 mm in thickness and 10 mm in length, were evaluated. A minimum of an aggregated 4–5 mm2 of well-preserved epithelial surface was required for the analysis. This was performed to increase the likelihood that the observations were representative, because the endometrium may show different morphology from one area to another. Pinopodes were defined qualitatively as: none, developing, fully developed, or regressing-disappearing (10). These evaluations were performed with coded tissue and read blindly by an experienced investigator (G.N.) (15).
RNA preparation/target preparation/array hybridization and scanning
For microarray analysis, each endometrial biopsy (n = 15) was processed individually for microarray hybridization at Genome Quebec Innovation Centre, McGill University (Montreal, Quebec, Canada) (24) (samples were not pooled). Total RNA was isolated using Trizol reagent (Life Technologies, Carlsbad, CA) following the supplier’s protocol. RNA was then further purified using the RNeasy total RNA clean up protocol (Qiagen, Valencia, CA). The integrity of the RNA samples (RNA QC procedure) was assessed by using the 2100 bioanalyzer (Agilent, Santa Clara, CA), running an aliquot of the RNA samples on the RNA 6000 Nano LabChip (Agilent). The Bioanalyser results are indicative of RNA integrity, based on a procedure that involves the separation of RNA into different sized fragments by capillary electrophoresis molecular sieving, with results shown on an electropherogram. Although the presence of small-size fragments is indicative of sample degradation, the presence of two prominent peaks marking the 18S and 28S ribosomal RNA is indicative of good mRNA quality. RNA quantity was determined with the Genesys 5 spectrophotometer. The same methods were used to verify the fragment sizes and quantity of RNA after amplification. Probe synthesis, hybridization, and scanning were done according to the Affymetrix protocol (Affymetrix, Sunnyvale, CA).
A single round of linear amplification of RNA samples was performed by the In Vitro Transcription T7-promoter method (25) with the Bioarray High-Yield RNA transcript labeling kit (Enzo Diagnostics, Farmingdale NY). The T7 primer was added to the total RNA in the very first step following the initial RNA QC procedure, before RT. The primer-RNA mixture was denatured for 10 min at 70 C and then chilled on ice for 2 min. Annealing of the T7 primer to RNA was performed by mixing 100 pM T7-(T)24 primer (Genset, 5'-GGCCAGTGAATTGTAATACGACTCACTATAGGGAGGCGG-(deoxythymidine)24–3') with 10 µg RNA sample in a 10-µl vol and incubating at 37 C for 3 min. First-strand cDNA synthesis was performed using 2 µl Superscript II reverse transcriptase (Invitrogen Life Technologies, Carlsbad, CA) in a 20-µl reaction vol containing 10 µM dithiothreitol, 500 µM of each deoxynucleotide triphosphate, and 1x first-strand buffer (all from Invitrogen Life Technologies) for 60 min at 42 C, resulting in the incorporation of the T7-promoter sequence (primer bound) into first-strand cDNA. Second-strand synthesis was performed by adding to the first-strand reaction mix: 40 U DNA polymerase I (Invitrogen Life Technologies), 10 U Escherichia coli DNA ligase, (Invitrogen Life Technologies), and 2 U ribonuclease H (MBI Fermentas, Hanover, MD) in a final reaction vol of 150 µl containing 1x second-strand buffer (Invitrogen Life Technologies). The reaction mixture was incubated at 16 C for 2 h. Ten units of T4 DNA polymerase (MBI Fermentas) were added and incubated at 16 C for 5 min, followed by the addition of 10 µl of 0.5-M EDTA (Sigma, The Woodlands, TX).
After second-strand synthesis, the cDNA was purified by phenol chloroform extraction with Phase-Lock tubes (Eppendorf, Westbury, NY), precipitated, and redissolved in 20 µl nuclease-free water. The purified cDNA was used to generate the biotinylated cRNA probe with the Bioarray High Yield RNA transcript labeling kit (Enzo Diagnostics) as indicated by the supplier. The probe synthesis reaction was performed at 37 C for 5 h, with occasional agitation using T7 RNA Polymerase. The labeled cRNA was then purified using the RNeasy total RNA clean up protocol (Qiagen), eluted in 60 µl nuclease-free water, and quantified by spectrophotometry. An aliquot of the purified cRNA was analyzed on RNA 6000 Nano LabChip (Agilent) to verify the integrity and size distribution. Twenty micrograms of cRNA were fragmented by heating at 94 C for 35 min in 1xfragmentation buffer (40 mM Tris-acetate, pH 8.1; 100 mM KOAc; 30 mM MgOAc), to reduce the probe length. The hybridization mixture was prepared by mixing 15 µg of the biotinylated probe cRNA with control oligonucleotide B2 (final concentration, 50 pM; Affymetrix), herring sperm DNA (final concentration, 0.1 mg/ml; Invitrogen Life Technologies), acetylated BSA (final concentration, 0.5 mg/ml; Invitrogen Life Technologies) in a final vol of 300 µl of 1x N-morpholinoethane sulfonic acid (MES) hybridization buffer (100 mM MES, 1 M NaCl, 20 mM EDTA, 0.01% Tween 20; all reagents from Sigma). The hybridization mixture was denatured for 10 min at 99 C, incubated for 5 min at 45 C, and spun for 5 min in a benchtop microcentrifuge. The microarray was warmed to room temperature and prehybridized with 200 µl of 1x hybridization buffer for 10–20 min at 45 C. The prehybridization solution was removed, and 200 µl of the hybridization mix was added to the array. The array and probe fragments were incubated at 45 C overnight (16–20 h) in a rotating oven (Affymetrix).
For the experiment, we used the HG_U95Av2 Array, containing 12,686 human genes and expressed sequence tags (ESTs). After hybridization, the hybridization cocktail was removed from the chip and stored at –80 C. The chip was immediately placed in the Affymetrix GeneChip Fluidics Station 400 (Affymetrix). In total, 10 low-stringency washes (63 SSPE, 0.01% Tween 20, 0.005% Antifoam) and 4 high-stringency washes (100 mM MES, 0.1 M NaCl, 0.01% Tween 20, 50 C) were performed (all reagents from Sigma). The array was then incubated with streptavidin/phycoerythrin stain (Molecular Probes, Eugene, OR), followed by 10 low-stringency washes. The array was incubated with biotinylated antistreptavidin antibody (Vector Laboratories) and washed again with 15 low-stringency washes. Specifically bound probe was detected by placing the array in the Agilent GeneArray scanner 2500 (Affymetrix). The scanned images were analyzed using the Microarray Analysis Suite, version 5.0 (Affymetrix). Detailed information on the gene array system is available at www.affymetrix.com and at genomequebec.mcgill.ca.
Validation of the gene expression data by quantitative real-time RT-PCR (Q-RT-PCR)
To validate microarray findings, we quantified the expression of some genes using one-step Q-RT-PCR on the LightCycler (Roche, Indianapolis, IN). The selected genes for validation were either randomly selected (procollagen-lysin and kallikrein-11) or selected after linear discriminant analysis (LDA) was performed (optineurin, see below). Primers were designed using LightCycler Primer design software (Roche) as follows: procollagen-lysine, 5'-CCCGAGTGTGAGTTCTAC-3' (forward primer), 5'-GCCCTTGTCTCGAAAGC-3' (reverse primer); kallikrein-11, 5'-GAGAAGACGCGGCTAC-3' (forward primer), 5'-GATGTCATTGCGGTGG-3' (reverse primer); and optineurin, 5'-AGTGGAGACTGTTCTCGTGGACCC-3' (forward primer), 5'-CAGCACCGCATCAGAGAATTG-3' (reverse primer).
Real-time one-step RT-PCR was performed as described elsewhere (26). We ran all assays using RNA Master SYBR Green I as indicated. The relative quantification was performed by crossing point curve analysis using SYBR Green I. Serial dilutions (from 1:10 to 1:10,0000) of a positive control were created for each gene. Anticipation of the crossing points was associated with higher gene expression as reflected on the serial dilutions, allowing the creation of the standard curve.
The reaction mixture was prepared in a controlled access reagent preparation room. The reaction mixture consisted of 7.6 µl LightCycler RNA Master SYBR Green I, 1.4 µl Mn(OAc)2, 1.2 µl of each primer (0.2 µM), and 8.9 ml H2O. The carousel containing the sealed capillaries was centrifuged and placed in the LightCycler. The standard cycling protocol consisted of: DNA denaturation at 95 C, annealing at 57 C for 10 sec, and extension at 72 C for 10 sec at the transition rate of 20 C per sec. A total of 47 cycles were performed. This protocol was used for all genes with minimal differences. Melting curve analysis was performed by bringing temperature from 95 C to 45 C for 60 sec and raising it to 75 C at the same transition rate of 20 C per second.
E2 and P4 serum levels
In COH cycles, blood was drawn on the day of the hCG administration and on the day of hCG +9. In natural cycles, blood was drawn on the urinary LH surge day and on day LH +8. Hormone levels were measured with a microparticle enzyme immunoassay [MEIA-IMX (microenzyme immunoassay, IMX); Abbott Laboratories, Abbott Park, IL]. The interassay coefficients of variations were 8.2% and 7.3% for E2 and P4, respectively. The intraassay coefficients of variation were 6.1% and 5.6% for E2 and P4, respectively. The lower limits of sensitivity were E2 = 25 pg/ml and P4 = 0.2 ng/ml. The regression equations to convert RIA to IMX are as follows: IMX E2 = 1.26 x RIA – 1.5 and IMX P4 = 1.23 x RIA – 1.5.
Statistical analysis of pinopodes, histological, and ER-PR characteristics
Statistical analysis comparing natural vs. COH cycles was performed using the Wilcoxon rank-sum test. P values < 0.05 were considered significant.
Statistical analysis of microarray data
The statistical analysis of the microarray data was independently performed by a bioinformatic group (Incogen, Williamsburg, VA). Multiple, pair-wise group comparisons were performed using public software, http://www.stat.stanford.edu/
tibs/SAM/ for the significant analysis of microarrays (SAM) (27). The software performed 1000 random permutations of group labels on the original data set to calculate the false discovery rate on the basis of t-statistics or score (difference between the group means in the units of SD), assuming equal group variance. Thus, significant genes were selected independent of their expression levels when their score was above the specified threshold 
for the difference between the observed score (t-statistics for true labels), and the average expected score (t-statistics for the randomly permuted data). In addition, SAM allows setting the detection threshold for the fold change in gene expression, which is the ratio of the mean expression levels for this gene in the groups under comparison. A q value was assigned to each detectable gene in the array. This q value is similar to the familiar P value, measuring the false discovery rate at which a gene is called significant (27). A median false significant number refers to the median falsely called genes (false-positive calls).
After SAM, LDA (28) was performed for the three groups of patients, i.e. natural cycles (group I), GnRH antagonist (group II), and GnRH agonist (group III), to elucidate relationships between the changes in gene expression levels and group membership of the samples. Classification was performed on the basis of distance to the closest mean measured in the units of variance (Mahalanobis distance). LDA is a mathematical technique of finding a linear transformation for original variables (gene expressions) that provides the largest between-group separation. Discriminant coordinates, constructed by LDA, were orthogonal linear combinations of original variables selected to maximize the ratio of between-groups to within-group covariance, such that the first coordinate would point in the direction of the most significant t test. We applied LDA to the set of genes found to be significant by SAM, to determine a subset of genes that is most useful for discrimination and, potentially, classification. The discriminating power of the individual genes was estimated on the basis of their loadings in discriminant coordinates as well as by their ability to provide less than one sample misclassification error in leave-one-out cross-validation (28). LDA was applied to increase the confidence level of potential functional relevance for the genes preselected by SAM.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
)All endometrial biopsies performed in natural cycles (n = 5) confirmed development as d 21 of the standardized menstrual cycle both by pinopodes appearance and H-E dating. Conversely, in all COH cycles (n = 10), pinopodes maturation was advanced (regressing-disappearing characteristics or d 22–23) compared with natural cycles (fully developed or d 21, n = 5) (P < 0.05). Regressing and disappearing pinopodes were observed in COH cycles, whereas fully developed pinopodes were present in all natural cycles biopsies. Furthermore, there was no difference, in terms of pinopodes maturation, between the different COH protocols (GnRH antagonist or agonist, with or without P4 supplementation). However, P4 supplementation produced other visible structural changes consisting of thickening and flattening of the microvilli formatting ruffles, especially at the cell borders.
|
Results of serum P4 and E2 levels (Table 2)
As expected, we found supraphysiological and significantly higher E2 and P4 levels in COH, compared with natural cycles (P < 0.05). There were no differences in the steroid levels on the day that biopsies were performed (matched d 21) between COH cycles accomplished with GnRH antagonist with P4 supplementation (group IIa), GnRH antagonist without P4 supplementation (group IIb), or GnRH agonist with P4 supplementation (group III) (P > 0.1).
|
and Figs. 1–3)
Global analysis.
Global gene expression profiles were analyzed by microarray technology comparing the expression patterns of COH cycles and natural cycles during the putative window of implantation. The data were analyzed using the SAM software as described in Subjects and Methods. Of the 12,686 human genes and ESTs represented on the Affymetrix Genechip HG_U95Av2 Array, 3,615 (29%) were recorded as present in all 15 endometrial biopsies.
|
|
|
|
shows the genes whose expression was found to be significantly different among the groups as well as the fold changes using pair-wise comparison analysis by SAM software. With a median false-positive number of 1.7, 12 genes/ESTs were found to be significantly different (0.4% of the genes expressed in all biopsies). Six genes (prosaposin, nonmetastatic cells 2, serine proteinase inhibitor, tyrosine 3-monooxygenase, IGF binding protein (IGFBP)-5, and procollagen-lysine) were up-regulated, whereas three genes (splicing factor arginine/serine, wingless-type mouse mammary tumor virus, and ATPase H+ transporting) and three ESTs were down-regulated. Figure 1 shows the SAM plot for this comparison. The y-axis plots the value of the observed score (distance between means in the units of SD for a particular gene) for each gene, and the x-axis plots the expected score (the average random score obtained for this gene with random permutations of group labels for all replica) for each gene; the diagonal line shows where the genes with random score (false positives statistical significance) would fall. The distance between dashed lines is the Delta threshold that was applied for detection of false positives.
Applying the same strategy, we investigated whether gene expression profiles differed in the endometrium during the putative window when comparing the temporally matched natural cycles (n = 5, d 21 or LH +8) vs. COH cycles accomplished with a long protocol using a GnRH agonist and rFSH with supplementation of the luteal phase with P4 (n = 3, d 21 or hCG +9). With a median false-positive number less than 1, six genes (0.16% of the genes expressed in all biopsies) were found to be expressed differently in this specific comparison. Five genes were up-regulated [leukocyte Ig-like receptor (LILR), optineurin, nonmetastatic cells 2, CD63 antigen, and procollagen type III N-endopeptidase], whereas one gene was down-regulated (rTS ß protein) (Table 3 and Fig. 2).
Impact of the use of GnRH agonists vs. GnRH antagonists.
We also investigated concordant genes during the window of implantation (hCG + 9) in COH cycles stimulated either with GnRH agonist and rFSH supplemented with P4 (n = 3) or GnRH antagonist and rFSH supplemented with P4 (n = 5) (Table 3). In this comparison, 13 genes were found to be significantly different [major histocompatibility complex (MHC) class II, MHC class I, IL 10, phosphorylase kinase ß, COX17 homolog, RAB1A, selenoprotein P, annexin A7, sorting nexin 7, related RAS viral, putative prostate cancer tumor suppressor, and UDP-glucose pyrophosphorylase 2]. All these genes/EST were up-regulated in the GnRH agonist group.
Impact of the use of luteal P4 supplementation. We investigated the effect of micronized P4 supplementation during the luteal phase among GnRH antagonist-treated subjects. We compared COH cycles stimulated with rFSH and GnRH antagonist with luteal phase supplementation vs. COH cycles stimulated in the same way but without supplementation of the luteal phase. With a false-positive value of approximately 0.6, no genes were found to be expressed in a significantly different way.
LDA: identification of genes of putative maximal functional significance.
LDA comparisons were used to reformulate the pair-wise comparisons into three-way comparisons to further identify a subset of genes that might have functional significance. Four genes contributed the best discriminating gene expression profile and were identified in the comparison of natural cycles (group I), GnRH antagonist-stimulated cycles (group IIa), and GnRH agonist-stimulated cycles (group III), the latter groups receiving P4 supplementation. Those genes were: optineurin, procollagen (type III) N-endopeptidase, sorting nexin 7, and COX 17 (Fig. 3).
Gene classification.
We further carried out an unsupervised cluster analysis of the genes found to be significantly different in the various comparisons (Table 3). Such analysis allowed us to classify the genes into different categories/families according to their function. Different databases (National Center for Biotechnology Informatio/PubMed/OMIM/Affimetrix database/Gene Cards) were used for this purpose (18, 29).
Validation of gene expression using quantitative Q-RT-PCR (Fig. 4)
To validate the findings of the microarray analysis, we quantified the expression pattern of three genes by Q-RT-PCR. The selected genes were: procollagen-lysine, optineurin, and kallikrein-11. Results corroborated the regulation profiles observed in the microarray study.
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Our data confirmed previous results indicating that gonadotropin-stimulated COH cycles resulted in a structural-histological advancement of the endometrium (12, 13, 14, 30, 31, 32, 33). However, other reports indicated that hCG and/or P4 supplementation resulted in in-phase biopsies in some cases (34). In our study, when compared with temporally matched d 21 natural cycles, all COH biopsies demonstrated 1–2 d advancement of pinopodes characteristics and stromal development with variable but minimal degrees of glandular/stromal dyssynchrony. Furthermore, COH cycles depicted down-regulation of ER and PR content, also corroborating previous results (16) (Table 1).
We expected that the changes observed in COH cycles, including cellular and histological characteristics (advancement) and steroid receptor content (down-regulation), would be associated with differences in endometrial gene expression, most likely secondary to the supraphysiological steroid milieu. Alternatively, gene expression differences might be due to a direct effect of the GnRH agonist or antagonist on the endometrium (35, 36, 37). Notwithstanding the major differences in the steroid internal milieu of the COH and natural cycles, only small changes in gene expression were observed. Altogether, for all COH cycles compared with the natural cycle controls, there were 18 genes/ESTs differentially expressed, with a range of 1.55 to +3.40-fold changes (Table 3 and Figs. 1 and 2).
Several proteins previously identified by immunohistochemistry have been postulated as putative markers of endometrial receptivity (4, 9). Some of them might be functionally relevant during implantation, including members of the integrin family (adhesion molecules) (38), leukemia inhibitory factor (39), and glycodelin-A (40) (immunomodulator) among others. Moreover, a synchronous expression of pinopodes and ß3 integrin has been reported during the window of implantation (41). Although some studies have revealed increased expression of some of these proteins (by immnocytochemistry) and pinopodes advancement (by SEM) during COH cycles, when compared with natural cycles, others were not able to confirm these differences (35, 42, 43, 44, 45). All the genes responsible for these products were present in the microarrays for all biopsies in the COH and natural cycle controls performed herein. Nevertheless, their expression was not significantly different.
Our results bring up several possibilities. First, during the COH cycles, gene expression changes may have occurred earlier than on the day of the biopsy (d 21), and as a consequence, we observed the structural/functional impact of gene expression modifications without being able to demonstrate greater changes in the profiles of the responsible genes at the point in time analyzed. Second, the effects of COH regimens on the endometrium may manifest at the late luteal phase, after the biopsy was taken, but still on time to influence the placentation process. In such a rapidly evolving tissue that undergoes day-by-day changes as the endometrium during the secretory phase, the timing could be very important in understanding the impact of gene expression in endometrial physiology. Third, another possibility is that the demonstrated changes in gene expression may truly reflect the magnitude of the observed ultraestructural-cytological modifications, i.e. small changes in gene expression do result in 1–2 d cellular-histological advancement. Fourth, it could be possible that products expressed in a given cell type (i.e. epithelium, glands, or stroma) may undergo changes that are of insufficient amplitude to be detected in the entire tissue. This has been addressed by Carson et al. (20) as an important shortcoming of the microarray technique. Although all biopsies studied herein were obtained using a similar technique, and H-E confirmed a full endometrial depth, we are now conducting studies using microdissection techniques to try to overcome this problem. Fifth, it could be argued that morphological and functional modifications might be due to posttranscriptional changes.
It is our belief that the small difference in gene expression and histological modifications documented in the COH cycles may, indeed, not represent a true functional difference because implantation rates are not too dissimilar in natural vs. COH cycles (46, 47). The rate of implantation expected for natural reproduction has been estimated as 20% (46). IVF statistics from the United States for the year 2001 revealed a live birth per ET value of 33%, however, as the result of transfer of multiple embryos (48). Although there are no available figures for implantation rate, it is generally estimated that rates of implantation range from 20–30%. Recent reports of selective single ET have revealed an implantation rate of 20–30% (47, 49, 50).
Twelve genes were differently expressed in COH cycles accomplished with rFSH and GnRH antagonist adjuvant therapy with supplementation of the luteal phase (group II) when compared with natural cycles (group I) (Table 3 and Fig. 1). Three of these genes have been previously related to some endometrial/embryo implantation function (51, 52, 53). However, although the differences observed in these three genes were significant, the fold change in expression was relatively small (1.3 to 2.0).
One of such genes, prosaposin, was 1.3-fold unregulated. Studies in animal models have indicated that prosaposin has an important role in intracellular and extracellular glycosphingolipid metabolism or transport in the uterine environment and appeared to be steroid dependent (51, 54). Another gene, IGFBP-5, was 2-fold up-regulated in the COH cycles. IGFBPs regulate the mitogenic and metabolic actions of IGFs by inhibiting or, in some cases, enhancing the receptor binding of IGF (52). IGFBPs have an important role in the regulation of stroma-glandular relation and embryo- endometrial communication (55, 56). We speculate that this gene is up-regulated in the COH/GnRH antagonist cycles, due to the higher estrogenic milieu. The third gene, procollagen-lysine (PLOD3), was also up-regulated. This gene product belongs to the lysyl hydroxylase family, and catalyzes the hydroxylations of proline and lysine in collagen. Collagens are a major component of the extracellular matrix (ECM), and the composition of the ECM is essential for embryo implantation (53). We have validated the up-regulation of this gene using Q-RT-PCR.
Importantly, within this COH group (group II), we did not see any differences in gene expression profiles or any variation in pinopodes characteristics, H-E dating or ER and PR content when comparing the two rFSH preparations (Gonal F and Follistim) or the two GnRH antagonists used (ganirelix and cetrorelix) (intragroup analysis of subjects in group II, data not shown).
Next, we compared gene expression profiles between COH cycles accomplished with GnRH agonist and rFSH plus supplementation of the luteal phase with P4 (group III) vs. natural cycles (group I). Six genes were found to be significantly differently expressed (Table 3 and Fig. 2). Of these six genes, two have been previously reported as involved in embryo implantation (LILR and procollagen type III N-endopeptidase) (57, 58).
One of such genes, LILRB1, was up-regulated (3.4-fold). This gene belongs to a family of immunoreceptors expressed predominantly on endometrial monocytes and B cells. They have been identified as receptors for class I HLA and transduce inhibitory signals to leukocytes upon binding of these ligands (57). The other gene, procollagen type III N-endopeptidase, was up-regulated 1.5-fold in this COH group. It is essential for the synthesis of the collagen that forms the fibrous scaffold of the ECM of the predecidual endometrium (58).
Some early clinical trails that compared the use of GnRH antagonists vs. the more standard long GnRH agonist protocols showed a trend for a lower pregnancy rate in the GnRH antagonist group (59, 60). Questions about the impact of the use of antagonists on COH cycles have been raised (36, 61, 62). We therefore asked the question of whether gene expression profiles might be significantly different when comparing GnRH antagonist- vs. GnRH agonist-treated cycles. Thirteen genes were found differently expressed between these two groups, and all of them were up-regulated (1.42- to 2.10-fold change) in the GnRH agonist-treated group (group III) (Table 3).
Of these 13 genes, three genes have been previously related to the implantation process (63, 64). Two of them, MHC class I and II, were up-regulated approximately 1.5-fold compared with the GnRH antagonist cycles. Genes of the MHC are related to the survival of the early embryo, playing an important role in immune tolerance (63). Cytokines play an important role in the regulation of MHC. IL-10 is an anti-inflammatory cytokine that is related to the regulation of MCH, and it was up-regulated 1.55-fold in the GnRH agonist group.
A recent meta-analysis indicated the need of luteal phase supplementation in COH cycles accomplished with GnRH agonist and gonadotropins (65). However, the utility of luteal phase supplementation in GnRH antagonist cycles has been less examined. Recently, Beckers et al. (66) reported the necessity of luteal support in such COH cycles. Therefore, we examined whether P4 supplementation resulted in differential gene expression during the window of implantation in COH cycles accomplished with a GnRH antagonist. We were unable to identify any gene differentially expressed in cycles treated with a GnRH antagonist that were supplemented or not with P4. When we analyzed P4 levels in supplemented cycles vs. those cycles without P4 supplementation, we found that serum P4 levels (and also E2 levels) were similar on the day of the biopsies (Table 1). This might explain the lack of effect on gene expression profile of P4 vs. no P4 supplementation in the GnRH antagonist cycles. Nevertheless, because only two patients were studied in the nonsupplemented group, more studies are needed to verify these findings. P4 supplementation altered the structure of cell microvilli as seen in SEM, raising the possibility of nongenomic effects of P4 concentrations.
LDA identified four genes that contributed the best discriminating expression profile and that were identified in the comparison of natural cycles (group I), GnRH antagonist-stimulated cycles (group II), and GnRH agonist-stimulated cycles (group III), the latter groups receiving P4 supplementation (Fig. 3). The following genes contributed to the best discriminating gene expression profile: optineurin, procollagen (type III) N-endopeptidase, sorting nexin 7, and COX 17. Of interest, most of the genes identified by the LDA were significantly affected by GnRH agonist treatment (group III). These results might indicate that the endometrial gene expression could be more effected on COH cycles accomplished with a GnRH agonist.
Of the four genes identified by LDA, optineurin (originally named FIP-2) appeared to have the highest potential functional significance, on the basis of the loadings in discriminant coordinates as well as its ability to provide less than one sample misclassification error in leave-one-out cross-validation. Optineurin was 2.10-fold up-regulated in the COH agonist group compared with natural cycles. This gene has not been previously involved in endometrial receptivity. It is a 74-kDa protein implicated in signal transduction of the TNF pathway and has been implicated in membrane traffic regulation, cellular morphogenesis, and apoptosis (67). Our Q-RT-PCR results confirmed the microarray profile. Procollagen (type III) N-endopeptidase was the only gene identified by LDA to be previously related to implantation (58). It has to be stressed, however, that the functional significance of the genes identified by the LDA needs to be further established.
We summarize our findings as follows: First, COH cycles depicted structural and functional changes when compared with natural cycles (i.e. we observed the expected histological advancement and steroid receptor down-regulation) that were associated with significant, but small, changes in gene expression. Second, there were significant changes in gene expression when comparing cycles using a GnRH agonist vs. a GnRH antagonist (in the absence of other structural or functional changes). Some of these genes have been cited as relevant for implantation. We speculate that the observed differences in gene expression in COH cycles may not have a major impact on implantation in IVF-ET cycles. Nevertheless, we are conducting follow-up studies to determine the functional significance, if any, of those genes identified by LDA.
|
Footnotes |
|---|
Abbreviations: COH, Controlled ovarian hyperstimulation; E2, estrogen; ECM, extracellular matrix; ER, E2 receptor; EST, expressed sequence tag; ET, embryo transfer; hCG, human chorionic gonadotropin; H-E, hematoxylin-eosin; IGFBP, IGF binding protein; IMX, microenzyme immunoassay; IVF, in vitro fertilization; LDA, linear discriminant analysis; LILR, leukocyte Ig-like receptor; MES, N-morpholinoethane sulfonic acid; MHC, major histocompatibility complex; P4, progesterone; PR, P4 receptor; Q-RT-PCR, quantitative real-time RT-PCR; rFSH, recombinant FSH; SAM, significant analysis of microarrays; SEM, scanning electron microscopy.
Received March 30, 2004.
Accepted July 27, 2004.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
v ß 3 and 4 ß 1 integrins in the endometrial surface epithelium of normally menstruating women during the implantation window. J Reprod Med 48:355–361[Medline]
This article has been cited by other articles:
 |
|
 |
|
  A. Allegra, A. Marino, F. Coffaro, A. Lama, G. Rizza, P. Scaglione, F. Sammartano, A. Santoro, and A. Volpes Is there a uniform basal endometrial gene expression profile during the implantation window in women who became pregnant in a subsequent ICSI cycle? Hum. Reprod., October 1, 2009; 24(10): 2549 - 2557. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Haouzi, S. Assou, K. Mahmoud, S. Tondeur, T. Reme, B. Hedon, J. De Vos, and S. Hamamah Gene expression profile of human endometrial receptivity: comparison between natural and stimulated cycles for the same patients Hum. Reprod., June 1, 2009; 24(6): 1436 - 1445. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. A. Horcajadas, P. Minguez, J. Dopazo, F. J. Esteban, F. Dominguez, L. C. Giudice, A. Pellicer, and C. Simon Controlled Ovarian Stimulation Induces a Functional Genomic Delay of the Endometrium with Potential Clinical Implications J. Clin. Endocrinol. Metab., November 1, 2008; 93(11): 4500 - 4510. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Franchi, J. Zaret, X. Zhang, S. Bocca, and S. Oehninger Expression of immunomodulatory genes, their protein products and specific ligands/receptors during the window of implantation in the human endometrium Mol. Hum. Reprod., July 1, 2008; 14(7): 413 - 421. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H.-F. Xia, J. Sun, Q.-H. Sun, Y. Yang, and J.-P. Peng Implantation-associated gene-1 (Iag-1): a novel gene involved in the early process of embryonic implantation in rat Hum. Reprod., July 1, 2008; 23(7): 1581 - 1593. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. S. Macklon, M. H. van der Gaast, A. Hamilton, B. C. J. M. Fauser, and L. C. Giudice The Impact of Ovarian Stimulation With Recombinant FSH in Combination With GnRH Antagonist on the Endometrial Transcriptome in the Window of Implantation Reproductive Sciences, April 1, 2008; 15(4): 357 - 365. [Abstract] [PDF]
|
|
|
|
 |
|
 J.A. Horcajadas, A. Pellicer, and C. Simon Wide genomic analysis of human endometrial receptivity: new times, new opportunities Hum. Reprod. Update, January 1, 2007; 13(1): 77 - 86. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H.-C. Ruan, X.-M. Zhu, Q. Luo, A.-X. Liu, Y.-L. Qian, C.-Y. Zhou, F. Jin, H.-F. Huang, and J.-Z. Sheng Ovarian stimulation with GnRH agonist, but not GnRH antagonist, partially restores the expression of endometrial integrin beta3 and leukaemia-inhibitory factor and improves uterine receptivity in mice Hum. Reprod., October 1, 2006; 21(10): 2521 - 2529. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. L. Kelley, K. L. Kind, M. Lane, R. L. Robker, J. G. Thompson, and L. J. Edwards Recombinant human follicle-stimulating hormone alters maternal ovarian hormone concentrations and the uterus and perturbs fetal development in mice Am J Physiol Endocrinol Metab, October 1, 2006; 291(4): E761 - E770. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Strowitzki, A. Germeyer, R. Popovici, and M. von Wolff The human endometrium as a fertility-determining factor Hum. Reprod. Update, September 1, 2006; 12(5): 617 - 630. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. A. Horcajadas, A. M. Sharkey, R. D. Catalano, J. R. A. Sherwin, F. Dominguez, L. A. Burgos, A. Castro, M. R. Peraza, A. Pellicer, and C. Simon Effect of an Intrauterine Device on the Gene Expression Profile of the Endometrium J. Clin. Endocrinol. Metab., August 1, 2006; 91(8): 3199 - 3207. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. S. Macklon, R. L. Stouffer, L. C. Giudice, and B. C. J. M. Fauser The Science behind 25 Years of Ovarian Stimulation for in Vitro Fertilization Endocr. Rev., April 1, 2006; 27(2): 170 - 207. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Mirkin, M. Arslan, D. Churikov, A. Corica, J.I. Diaz, S. Williams, S. Bocca, and S. Oehninger In search of candidate genes critically expressed in the human endometrium during the window of implantation Hum. Reprod., August 1, 2005; 20(8): 2104 - 2117. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. A White and L. A Salamonsen A guide to issues in microarray analysis: application to endometrial biology Reproduction, July 1, 2005; 130(1): 1 - 13. [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 |
