Deregulation of the endometrial stromal cell secretome precedes embryo implantation failure.

STUDY QUESTION: Is implantation failure following ART associated with a perturbed decidual response in endometrial stromal cells (EnSCs)? SUMMARY ANSWER: Dynamic changes in the secretome of decidualizing EnSCs underpin the transition of a hostile to a supportive endometrial microenvironment for embryo implantation; perturbation in this transitional pathway prior to ART is associated with implantation failure. WHAT IS KNOWN ALREADY: Implantation is the rate-limiting step in ART, although the contribution of an aberrant endometrial microenvironment in IVF failure remains ill defined. STUDY DESIGN, SIZE, DURATION: In vitro characterization of the temporal changes in the decidual response of primary EnSCs isolated prior to a successful or failed ART cycle. An analysis of embryo responses to secreted cues from undifferentiated and decidualizing EnSCs was performed. The primary clinical outcome of the study was a positive urinary pregnancy test 14 days after embryo transfer. PARTICIPANTS/MATERIALS, SETTING, METHODS: Primary EnSCs were isolated from endometrial biopsies obtained prior to IVF treatment and cryopreserved. EnSCs from 10 pregnant and 10 non-pregnant patients were then thawed, expanded in culture, subjected to clonogenic assays, and decidualized for either 2 or 8 days. Transcript levels of decidual marker gene [prolactin (PRL), insulin-like growth factor binding protein 1 (IGFBP1) and 11ß-hydroxysteroid dehydrogenase (HSD11B1)] were analysed using real-time quantitative PCR and temporal secretome changes of 45 cytokines, chemokines and growth factors were measured by multiplex suspension bead immunoassay. The impact of the EnSC secretome on human blastocyst development was scored morphologically; and embryo secretions in response to EnSC cues analyzed by multiplex suspension bead immunoassay. MAIN RESULTS AND THE ROLE OF CHANCE: Clonogenicity and induction of decidual marker genes were comparable between EnSC cultures from pregnant and non-pregnant group groups (P > 0.05). Analysis of 23 secreted factors revealed that successful implantation was associated with co-ordinated secretome changes in decidualizing EnSCs, which were most pronounced on Day 2 of differentiation: 17 differentially secreted proteins on Day 2 of decidualization relative to undifferentiated (Day 0) EnSCs (P < 0.05); 11 differentially secreted proteins on Day 8 relative to Day 2 (P < 0.05); and eight differentially secreted proteins on Day 8 relative to Day 0 (P < 0.05). By contrast, failed implantation was associated with a disordered secretome response. Blastocyst development was compromised when cultured for 24 h in medium conditioned by undifferentiated EnSCs when compared to decidualizing EnSCs. Analysis of the embryo microdroplets revealed that human blastocysts mount a secretory cytokine response to soluble decidual factors produced during the early (Day 2) but not late phase (Day 8) of differentiation. The embryo responses to secreted factors from decidualizing EnSCs were comparable between the pregnant and non-pregnant group (P > 0.05). LARGE SCALE DATA: Not applicable. LIMITATIONS, REASONS FOR CAUTION: Although this study uses primary EnSCs and human embryos, caution is warranted when extrapolating the results to the in vivo situation because of the correlative nature of the study and limited sample size. WIDER IMPLICATIONS OF THE FINDINGS: Our finding raises the prospect that endometrial analysis prior to ART could minimize the risk of treatment failure. STUDY FUNDING AND COMPETING INTEREST(S): This work was supported by funds from the Biomedical Research Unit in Reproductive Health, a joint initiative of the University Hospitals Coventry & Warwickshire NHS Trust and Warwick Medical School, the University of Nottingham and Nurture Fertility, and the National Medical Research Council, Singapore (NMRC/BNIG14NOV023), the "Instituut voor Innovatie door Wetenschap en Technologie" (IWT, Flanders, Belgium), the "Fonds voor Wetenschappelijk Onderzoek" (FWO, Flanders, Belgium) and the "Wetenschappelijk Fonds Willy Gepts" (WFWG, UZ Brussel). The authors have declared that no conflict of interest exists.