Hum. Reprod. Advance Access originally published online on June 12, 2008
Human Reproduction 2008 23(9):1993-2000; doi:10.1093/humrep/den205
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Comparative protein-profile analysis of implanted versus non-implanted human blastocysts
1 Fundacion IVI, Instituto Universitario IVI, Valencia University, C/Guadassuar 1 bajo, Valencia 46015, Spain 2 Department of Experimental Biology, University of Jaén, Jaén, Spain
3 Correspondence address. Tel: +34 963455560; Fax: +34 963455512; E-mail: fdominguez{at}ivi.es
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionFundingAcknowledgementsReferences |
|---|
BACKGROUND: New approaches for non-invasive embryo-quality assessment are among the major goals in Reproductive Medicine. We hypothesize that the detection of changes in the protein profile of the culture media in which blastocysts are cultured could be a potential indicator of the viability of the embryo and, thus, a useful tool for selecting the more appropriate blastocysts to be transferred.
METHODS: Using protein-array technology, we analysed the protein profile corresponding to 24 h conditioned media of blastocysts that implanted versus those that did not implant. A statistical approach was followed to compare each of these media versus a medium in the absence of blastocysts (control medium). In addition, a gene ontology functional analysis—including those proteins showing a statistical difference among conditions—was performed, and a network with the predicted functional partners and corresponding relationships was obtained.
RESULTS: The soluble TNF receptor 1 and IL-10 increased significantly and MSP-
, SCF, CXCL13, TRAILR3 and MIP-1β decreased significantly when the protein profile of the blastocyst culture medium was compared with the control medium. CXCL13 (BLC) and granulocyte-macrophage colony-stimulating factor was also decreased significantly in the implanted blastocyst media compared with that in media from the non-implanted counterparts with a similar morphology. None of the proteins included in the array was increased significantly in the implanted blastocyst-conditioned media.
CONCLUSIONS: The differences identified in the protein profile of the culture media in the presence of implanted versus non-implanted blastocysts can be considered a new non-invasive approach in the search for new tools to diagnose blastocyst viability.
Key words: implantation/secretome/implantome/protein array
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionFundingAcknowledgementsReferences |
|---|
A major issue in the field of human reproductive is the selection of appropriate embryos for transfer to the uterus. To date, the method used to select the best cleavage embryo or blastocyst has been morphological assessment, but this selection method is subjective (Guerif et al., 2007
). In fact, implantation rates in Assisted Reproductive Technologies have not improved in the last decade (Nygren et al., 2006).
Technological advances in translational research have enabled non-invasive determination of the proteomic and metabolic status of an embryo. Recently, proteomic analysis of human and mouse embryos/blastocysts has been conducted in two studies using time-of-flight mass spectrometry (Katz-Jaffe et al., 2005, 2006). Nevertheless, no information is available concerning the secretome (proteins secreted/consumed) of the human blastocyst and whether it may correlate with implantation ability (implantome).
The aim of the present study was to identify the differences in the protein profile of the human blastocyst using protein-array technology. We investigated proteomic profiles from both morphologically normal blastocysts as well as functionally viable blastocysts.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionFundingAcknowledgementsReferences |
|---|
Study design
This study was registered with number NCT00505115 [ClinicalTrials.gov] in the Clinical Trial web (www.clinicaltrials.gov), and it received institutional review-board approval by the Ethics Committee of IU-IVI (Valencia, Spain). We developed a single blastocyst transfer protocol, which is the basis for the collection of blastocyst-conditioned media from the last 24 h of the development of a single blastocyst. The blastocyst is transferred to the mother, where it either implants (positive pregnancy test) or does not implant (negative pregnancy test). We collected up to 20 samples (50 µl each) of conditioned media from implanted blastocysts and 15 samples (50 µl each) of conditioned media from non-implanted blastocysts. We also collected 10 samples (50 µl each) of control media cultured under the same conditions as the samples, but with no blastocyst, and these were pooled in groups of five because of the low amounts of protein that the samples contained.
To gather the differential protein abundance data and to carry out the statistical analysis, we compared four pools of samples taken from implanted blastocysts (a total of 20 blastocysts analysed in pools of five samples, n = 4) with three pools of samples obtained from non-implanted blastocyst-conditioned media (a total of 15 blastocysts analysed in pools of five samples, n = 3). To prepare the protein profile of the human blastocysts, we compared the four implanted blastocyst pools (n = 4) versus the control media (a total of 10 control-media samples in pools of five samples, n = 2).
Human embryo culture
After fertilization assessment, all zygotes were rinsed and cultured in HTF medium. On Day 2, embryos were assessed for cleavage stage. Culture samples containing 50 µl of CCM medium (Vitrolife AB, Sweden) were placed in a culture dish and overlaid with oil. On Day 3, embryos were transferred from HTF medium to CCM medium until transfer (Day 5 or 6). Single blastocyst transfer was performed when a good-quality embryo (in terms of trophoectoderm and inner-cell-mass (ICM) morphology) was available (Schoolcraft et al., 1999). Control and blastocyst-conditioned media (50 µl) were collected and frozen at –80°C until needed. The mean age of the women in the implantation group (n = 20) was 29.9, whereas the mean age of those in the non-implanted group (n = 15) was 29.7. The embryo morphology of each group was comparable in terms of ICM and trophoectoderm morphology. Furthermore, the number of suboptimal blastocysts (see Table I) was 40% in the non-implantation group and 0% in the implantation group, according to a previously described classification (Schoolcraft, 1999).
|
Protein array
The conditioned media obtained from single blastocysts that implanted versus those that did not were compared using ChemiarrayTM (Chemicon International, USA). Arrays VI and VII contained 120 proteins that could be simultaneously compared between two given conditions (see Table II). This system has a detection-range limit of 10–250 000 pg/ml, and the variation of duplicates ranged from 0 to 10% in duplicate experiments.
|
Briefly, five samples (50 µl each) of media taken from implanted blastocysts of similar morphology, five samples (50 µl each) of media taken from non-implanted blastocysts with a similar morphology or five control-media samples (50 µl of CCMTM) were collected, pooled and diluted to 1 ml with blocking buffer (one-fourth dilution). ChemiarrayTM membranes were blocked for 2 h at room temperature prior to media incubation. Each pool of media (implanted versus non-implanted) was incubated with two different membranes (arrays VI and VII) for 2 h at room temperature. After the sample incubation and washing, diluted biotin-conjugated anti-cytokine primary antibody was applied to each membrane for 2 h at room temperature. After washing, the diluted horse-radish peroxidase-streptavidin was added to the membranes for 2 h at room temperature, treated with the ECL detection system and exposed to Kodak X-omat film.
Image-densitometry analyses for all membranes were performed using the Image J software (http://rsb.info.nih.gov/ij/). Spot intensity was quantified and relative expression levels between proteins were calculated in each experiment (implanted, non-implanted and controls). For all membranes, the background signal was subtracted using Image J tools, and the spot intensity was quantified as arbitrary densitometry units using positive spots to determine the relative values. Given the small number of replicates, only the proteins with a positive relative value in all the arrays studied were included in the statistical analysis (see below).
Statistical and bioinformatic analysis
Statistical analysis was performed using the R software (http://www.r-project.org/) and the appropriate Bioconductor packages (http://www.bioconductor.org/) run under R (see below).
To remove all the possible sources of variation of a non-biological origin between arrays, densitometry values between arrays were transformed to the logarithmic scale (log2) and normalized using the quantile normalization function implemented in the Bioconductor ‘limma’ package. In other words, intensities were scaled so that each array had the same average value to compensate for the technical variation that is introduced while conducting the experiment. In our experiments, each array was processed using the pools of five biological samples. We analysed the data from two controls, four implanted replicates and three non-implanted replicates. Statistically significant differences between groups were identified using the Student’s t-test. Even though applying the Student’s t-test with such a limited number of samples is problematic and the resulting statistical significance is not robust, and the mean and the standard deviation could be easily biased by outliers, we made this statistical test as a rough filter to narrow down the list of proteins that could be most relevant to include in future investigations. Moreover, with the aim of correcting the raw P-values and to ascertain the false discovery rate (FDR), a multiple hypothesis test was performed using the Bioconductor ‘multitest’ package.
In addition, we used the densitometry values corresponding to the four proteins, which were identified to show a lower P-value after the Student’s t-test (when comparing blastocyst-conditioned media with control media, and implanted blastocysts versus non-implanted blastocysts, respectively), to ascertain how the data structure was affected by the normalization procedure. To search for the relationship patterns that may support the statistical Student’s t-test, we also plotted the densitometry values against the principal components (data not shown). The densitometry values of these proteins in both the control media and the blastocyst-conditioned media were clearly related to the second principal component before normalization and also to the first principal component after normalization of the quantiles. This situation indicated that both were well-structured groups and that the normalization procedure led to a better grouping of samples. Well-structured groups were also detected when the densitometry values of the four proteins with the lowest P-values in the implanted versus non-implanted blastocysts comparison were plotted against the principal components, especially after quantile normalization (data not shown).
A gene ontology (GO) functional annotation was also performed. Gene symbols of the up- and down-regulated proteins were imported into The Database for Annotation, Visualization and Integrated Discovery (DAVID; http://david.abcc.ncifcrf.gov/); we selected the biological processes showing an FDR < 0.05 as significant terms. Moreover, gene symbols corresponding to differentially expressed proteins were also sent to the Search Tool for the Retrieval of Interacting Genes/Proteins (STRING; http://string.embl.de/) to build a network using edge information from three separate forms of evidence: databases, experiments and text mining. We used 0.7 (high confidence) as the value for edge confidence provided by STRING, and 20 as the maximal number of predicted integrators.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionFundingAcknowledgementsReferences |
|---|
The comparison of blastocyst-conditioned media with control media revealed the protein profile of the human blastocyst prior to implantation. Raw P-values, the FDR and the fold change for each protein detected in the array are shown in Table III. It is noteworthy that proteins such as CXCL13 (BCL), stem-cell factor (SCF), macrophage-stimulating protein-alpha (MSP-
), TRAILR3 and MIP-1β were significantly decreased (raw P < 0.05) in the conditioned media (Fig. 1). In contrast, soluble TNF receptor 1 (sTNFR1) and IL-10 were significantly increased in the media (raw P < 0.05) where the blastocyst was present (Fig. 1). Moreover, CXCL13, SCF and sTNFR1 showed an FDR < 0.05 after the multiple hypothesis testing; thus, the change of these proteins between the two conditions can be considered as no false positives by chance.
|
|
We then compared the conditioned media of implanted blastocysts versus non-implanted blastocysts in an attempt to describe a footprint for implantation ability (Fig. 1 and Table IV). A comparative analysis revealed that the granulocyte-macrophage colony-stimulating factor (GM-CSF), CXCL13 (BCL), IGF-1, IL-1R, Eotaxin-3 and NT-4 were decreased in media from implanted blastocysts compared with that from unimplanted blastocysts (Table IV). However, the only proteins that we found significantly reduced in the implantome were CXCL13 and GM-CSF (raw P < 0.05). Only CXCL13 presented an FDR < 0.05 after multiple hypothesis testing. None of the proteins investigated was found to increase significantly and, therefore, no secreted proteomic markers for embryo viability were found.
|
To understand the pathways involved in human blastocyst implantation, we performed a GO functional analysis taking into account all the proteins detected as statistically regulated in our arrays. The results showed that these proteins were significantly included in GO categories (FDR < 0.05 as provided by DAVID) mainly related to an immune response towards the surrounding environment: ‘response to wounding, response to external stimulus and response to stress’ (MSP-
, IL10, sTNFR1, GM-CSF, MIP-1β and CXCL13); ‘response to pest, pathogen or parasite and response to other organisms’ (IL10, sTNFR1, GM-CSF, MIP-1β and CXCL13); ‘inflammatory response’ (IL10, sTNFR1, MIP-1β and CXCL13); ‘cell communication’ (TRAILR3, IL10, SCF, sTNFR1, GM-CSF, MIP-1β and CXCL13); ‘immune response’ (IL10, sTNFR1, GM-CSF, MIP-1β and CXCL13); ‘taxis’ and ‘chemotaxis’ (IL10, MIP-1β and CXCL13); ‘defence response’ (IL10, sTNFR1, GM-CSF, MIP-1β and CXCL13); ‘locomotory behaviour’ (IL10, MIP-1β and CXCL13); ‘response to biotic stimulus’ (IL10, sTNFR1, GM-CSF, MIP-1β and CXCL13); and ‘organism physiological process’ (MSP-, IL10, sTNFR1, GM-CSF, MIP-1β and CXCL13). In addition, using as seed nodes the differentially detected proteins described above, we made the first network, including a set of the predicted functional partners (and their relationships) that could be considered at least a partial component of the human blastocyst secretome and implantome (Fig. 2).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionFundingAcknowledgementsReferences |
|---|
Using a commercially available membrane protein array containing 120 proteins, we identified the protein profile of the culture media in the presence of the blastocyst, and in particular, of the implanted blastocyst. Statistical analysis revealed that the secretome of the human blastocyst consumes, metabolizes or binds CXCL13 (BCL) SCF, MSP-
, TRAILR3, and MIP-1B. It is not clear which mechanisms are responsible for the down-regulation of these proteins, but they may be internalized, bound to their ligands, or may simply be metabolized by the blastocyst. The fact is that we noted a significantly reduced abundance of these proteins in the media. It bears noting, taking into account that 120 proteins were studied, that only soluble TNF-receptor 1 (sTNFr1) and IL-10 were found to increase by the presence of the human blastocyst. The comparison of implanted versus non-implanted blastocyst-conditioned media is characterized by a decrease in GM-CSF and CXCL13 in the presence of the implanted blastocyst. The ability to assess the protein profile of an individual embryo would lead to a better understanding of the cellular function at specific embryogenesis stages. Furthermore, such an approach would help us to improve embryo-culture media and to identify interactions between the blastocysts and the maternal uterine epithelium prior to implantation.
In our study, we have pooled-conditioned media samples in groups of five because of the limited effect of a single blastocyst in the conditioned media. We are aware that pooling samples could imply a decreased sensitivity in the protein arrays performed. However, we could not use single embryo-conditioned medium to analyse the protein changes given the lack of high-throughput protein arrays to date and the limitation of the initial sample (50 µl of conditioned media). With the future arrival of high-throughput protein arrays, we will be able to increase the sensitivity of these techniques in order to analyse the single and complete secretome and implantome of the human blastocyst.
In our case, and even though the numbers of proteins in the arrays were around two orders of magnitude lower than the number of probes that are usually tested simultaneously in DNA-microarrays, we also performed a multiple hypothesis test to correct the original raw P-value of the increased occurrence of false positive results by chance through testing multiple proteins simultaneously. Although new high-resolution methods are being developed that may help us to validate the present results, CXCL13, SCF and sTNFR1 could, therefore, be considered significant proteins that change between the experimental groups, as described above.
SCF mRNA has been detected in human preimplantation embryos (Sharkey et al., 1995), and also in human endometrium (Kauma et al., 1996, 1999). The binding of SCF to its receptor, c-kit, stimulates trophoblast outgrowth in mice blastocysts (Mitsunari et al., 1999). MSP-, also called hepatocyte growth factor-like (HGFL), has a dual function through its receptor RON tyrosine kinase. In vitro, the activation of the tyrosine kinase RON receptor by HGFL results in epithelial-cell dissociation, migration and matrix invasion (Comoglio et al., 1999; Danilkovitch et al., 2000; Wang et al., 2003). Although in vitro studies have clearly demonstrated that both epithelial and stromal cells attach to and invade intact peritoneum (Witz et al., 2002, 2003), the specific factors involved in these processes remain unknown. HGFL also behaves as a trophic cytokine that prevents apoptosis (Matsuzaki et al., 2005). In addition, activation of the RON tyrosine kinase receptor by HGFL may assist in implantation, thereby improving trophoblast function and viability (Hess et al., 2003). In fact, a deletion of the entire receptor proves lethal in mouse embryos (Uehara et al., 1995). Ron –/– embryos are viable through the blastocyst stage of development, but they fail to survive beyond the peri-implantation period (Muraoka et al., 1999). Based on this evidence, we hypothesize that the blastocyst needs the presence of both proteins in either the culture medium or the uterine cavity for proper development.
The only known function of the soluble sTNFR1 is to antagonize and buffer circulating TNF-. There is some evidence that sTNFR1 exerts immunoregulatory functions by inducing apoptosis in monocytes through reverse signalling via TNF-alpha (Waetzig et al., 2005). The interaction net of proteins in our study includes TNF and LTA, which are involved in cell death and apoptosis, and also includes caspase precursors (caspases 2, 8 and MCH-4), TRAP-3, FADD and the TRADD proteins that interact with these three proteins: LTA, TNF and sTNFR1 (Fig. 2). We hypothesize that the sTNFR1 increased by the human embryo might favour immunotolerance and induce endometrial epithelial apoptosis at the implantation site, as described in humans (Galan et al., 2000) and rodents (Kamijo et al., 1998).
In the second part of our study, we compared the protein patterns of viable and non-viable embryos. We identified two proteins that had decreased in viable embryos, GM-CSF and CXCL13 (BCL). GM-CSF is a well-studied factor that prevents ICM apoptosis in humans (Sjoblom et al., 2002) and mice (Karagenc et al., 2005), and promotes human blastocyst development in vitro (Sjoblom et al., 1999). GM-CSF treatment affects resistance to freezing/thawing and re-expansion of murine blastocysts (Papayannis et al., 2007), and it is not only crucial for embryo development, but also for implantation (Robertson, 2007). Clinically, it has been associated with a better reproductive outcome in patients with implantation failure undergoing co-culture with autologous endometrial cells (Spandorfer et al., 1998). This evidence indicates that GM-CSF is an essential protein in blastocyst metabolism and that it could be a prospective biomarker for blastocyst viability.
Notably, CXCL13 (B-cell-attracting homing chemokine) has not been described either in embryo development or in the implantation process. Several functions have been described for this protein in the immune system, such as natural antibody production and mucosal immunity (Ansel et al., 2002), although its functions in the reproductive field are still unknown.
The GO functional analysis, the network of relationships between these proteins, and their predicted functional partners (Fig. 2) indicate that a competent blastocyst must activate pathways to change the immunity of the surrounding environment by secreting mainly sTNFr1, which also facilitates local apoptosis and starts epithelial degradation (activating caspases and pro-apoptotic molecules). The implanting blastocyst must also communicate with maternal cells by binding or metabolizing the essential proteins secreted by the endometrium, such as GM-CSF or the CXCL13 chemokine, to begin the implantation process. Whether or not these proteins could be used for embryo selection in IVF clinical practice needs to be addressed with further research that focuses on these two proteins.
To summarize, by comparing the protein profile of conditioned media from single blastocysts with clinical implantation versus non-implanted blastocysts, we describe for the first time the protein profile based on (commercially available) protein arrays of the human blastocyst as a potential non-invasive tool to investigate blastocyst viability.
|
Funding |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionFundingAcknowledgementsReferences |
|---|
The participation of FJE in this work was partially supported by a grant from Junta de Andalucía to the Systems Biology Unit at University of Jaén (BIO-302).
|
Acknowledgements |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionFundingAcknowledgementsReferences |
|---|
The authors thank two anonymous reviewers for comments which have improved the final version of the manuscript.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionFundingAcknowledgementsReferences |
|---|
Ansel KM, Harris RB, Cyster JG. CXCL13 is required for B1 cell homing, natural antibody production, and body cavity immunity. Immunity (2002) 16:67–76.[CrossRef][Web of Science][Medline]
Comoglio PM, Tamagnone L, Boccaccio C. Plasminogen-related growth factor and semaphorin receptors: a gene superfamily controlling invasive growth. Exp Cell Res (1999) 253:88–99.[CrossRef][Web of Science][Medline]
Danilkovitch A, Donley S, Skeel A, Leonard EJ. Two independent signaling pathways mediate the antiapoptotic agent of macrophage-stimulating protein on epithelial cells. Mol Cell Biol (2000) 20:2218–2227.
Galan A, O’Connor JE, Valbuena D, Herrer R, Remohi J, Pampfer S, Pellicer A, Simon C. The human blastocyst regulates endometrial epithelial apoptosis in embryonic adhesion. Biol Reprod (2000) 63:430–439.
Guerif F, Le Gouge A, Giraudeau B, Poindron J, Bidault R, Gasnier O, Royere D. Limited value of morphological assessment at days 1 and 2 to predict blastocyst development potential: a prospective study based on 4042 embryos. Hum Reprod (2007) 22:1973–1981.
Hess KA, Waltz SE, Chan EL, Degen SJ. Receptor tyrosine kinase Ron is expressed in mouse reproductive tissues during embryo implantation and is important in trophoblast cell function. Biol Reprod (2003) 68:1267–1275.
Kamijo T, Rajabi MR, Mizunuma H, Ibuki Y. Biochemical evidence for autocrine / paracrine regulation of apoptosis in cultured uterine epithelial cells during mouse embryo implantation in vitro. Mol Hum Reprod (1998) 4:990–998.
Karagenc L, Lane M, Gardner DK. Granulocyte-macrophage colony-stimulating factor stimulates mouse blastocyst inner cell mass development only when media lack human serum albumin. Reprod Biomed Online (2005) 10:511–518.[Web of Science][Medline]
Katz-Jaffe MG, Linck DW, Schoolcraft WB, Gardner DK. proteomic analysis of mammalian preimplantation embryonic development. Reproduction (2005) 130:899–890.
Katz-Jaffe MG, Gardner DK, Schoolcraft WB. Proteomic analysis of individual human embryos to identify novel biomarkers of development and viability. Fertil Steril (2006) 85:101–107.[CrossRef][Web of Science][Medline]
Kauma S, Huff T, Krystal G, Ryan J, Takacs P, Turner T. The expression of stem cell factor and its receptor, c-kit in human endometrium and placental tissues during pregnancy. J Clin Endocrinol Metab (1996) 81:1261–1266.[Abstract]
Kauma S, Bae-Jump V, Walsh S. Hepatocyte growth factor stimulates trophoblast invasion: a potential mechanism for abnormal placentation in preeclampsia. J Clin Endocrinol Metab (1999) 84:4092–4096.
Matsuzaki S, Canis M, Pouly JL, Dechelotte P, Okamura K, Mage G. The macrophage stimulating protein/RON system: a potential novel target for prevention and treatment of endometriosis. Mol Hum Reprod (2005) 11:345–349.
Mitsunari M, Harada T, Tanikawa M, Iwabe T, Taniguchi F, Terakawa N, Mitsunari M, Harada T, Tanikawa M, Iwabe T, et al. The potential role of stem cell factor and its receptor c-kit in the mouse blastocyst implantation. Mol Hum Reprod (1999) 5:874–879.
Muraoka RS, Sun WY, Colbert MC, Waltz SE, Witte DP, Degen JL, Degen SJ. The Ron/STK receptor tyrosine kinase is essential for peri-implantation development in the mouse. J Clin Invest (1999) 103:1277–1128.[Web of Science][Medline]
Nygren K, Andersen AN, Felberbaum R, Gianaroli L, de Mouzon J. ESHRE’s European IVF Monitoring (EIM). On the benefit of assisted reproduction techniques, a comparison of the USA and Europe. Hum Reprod (2006) 21:21.
Papayannis M, Eyheremendy V, Sanjurjo C, Blaquier J, Raffo FG. Effect of granulocyte-macrophage colony stimulating factor on growth, resistance to freezing and thawing and re-expansion of murine blastocysts. Reprod Biomed Online (2007) 14:96–101.[Web of Science][Medline]
Robertson SA. GM-CSF regulation of embryo development and pregnancy. Cytokine Growth Factor Rev (2007) 18:287–298.[CrossRef][Web of Science][Medline]
Schoolcraft WB, Gardner DK, Lane M, Schlenker T, Hamilton F, Meldrum DR. Blastocyst culture and transfer: analysis of results and parameters affecting outcome in two in vitro fertilization programs. Fertil Steril (1999) 72:604–609.[CrossRef][Web of Science][Medline]
Sharkey AM, Dellow K, Blayney M, Macnamee M, Charnock-Jones S, Smith SK. Stage-specific expression of cytokine and receptor messenger ribonucleic acids in human preimplantation embryos. Biol Reprod (1995) 53:974–981.[Abstract]
Sjoblom C, Wikland M, Robertson SA. Granulocyte-macrophage colony-stimulating factor promotes human blastocyst development in vitro. Hum Reprod (1999) 14:3069–3076.
Sjoblom C, Wikland M, Robertson SA. Granulocyte-macrophage colony-stimulating factor (GM-CSF) acts independently of the beta common subunit of the GM-CSF receptor to prevent inner cell mass apoptosis in human embryos. Biol Reprod (2002) 67:1817–1823.
Spandorfer SD, Barmat LI, Liu HC, Mele C, Veeck L, Rosenwaks Z. Granulocyte macrophage-colony stimulating factor production by autologous endometrial co-culture is associated with outcome for in vitro fertilization patients with a history of multiple implantation failures. Am J Reprod Immunol (1998) 40:377–381.[Medline]
Uehara Y, Minowa O, Mori C, Shiota K, Kuno J, Noda T, Kitamura N. Placental defect and embryonic lethality in mice lacking hepatocyte growth factor/scatter factor. Nature (1995) 373:702–705.[CrossRef][Web of Science][Medline]
Waetzig GH, Rosenstiel P, Arlt A, Till A, Brautigam K, Schafer H, Rose-John S, Seegert D, Schreiber S. Soluble tumor necrosis factor (TNF) receptor-1 induces apoptosis via reverse TNF signaling and autocrine transforming growth factor-beta1. FASEB J (2005) 19:91–93.
Wang MH, Wang D, Chen YQ. Oncogenic and invasive potentials of human macrophage-stimulating protein receptor, the RON receptor tyrosine kinase. Carcinogenesis (2003) 24:1291–1300.
Witz CA, Cho S, Montoya-Rodriguez IA, Schenken RS. The alpha(2)beta(1) and alpha(3)beta(1) integrins do not mediate attachment of endometrial cells to peritoneal mesothelium. Fertil Steril (2002) 78:796–803.[CrossRef][Web of Science][Medline]
Witz CA, Cho S, Centonze VE, Montoya-Rodriguez IA, Schenken RS. Time series analysis of transmesothelial invasion by endometrial stromal and epithelial cells using three-dimensional confocal microscopy. Fertil Steril (2003) 79:770–778.[CrossRef][Web of Science][Medline]
Submitted on September 4, 2007; resubmitted on April 28, 2008; accepted on May 6, 2008.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  M.G. Katz-Jaffe, S. McReynolds, D.K. Gardner, and W.B. Schoolcraft The role of proteomics in defining the human embryonic secretome Mol. Hum. Reprod., May 1, 2009; 15(5): 271 - 277. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Dominguez, A. Pellicer, and C. Simon The Human Embryo Proteome Reproductive Sciences, February 1, 2009; 16(2): 188 - 190. [Abstract] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||