Trophoblast attachment to the endometrial epithelium elicits compartment-specific transcriptional waves in an in-vitro model.

RESEARCH QUESTION: Which are the early compartment-specific transcriptional responses of the trophoblast and the endometrial epithelium throughout early attachment during implantation? DESIGN: An endometrial epithelium proxy (cell line Ishikawa) was co-cultured with spheroids of a green fluorescent protein (GFP) expressing trophoblast cell line (JEG-3). After 0, 8 and 24 h of co-culture, the compartments were sorted by fluorescence-activated cell sorting; GFP+ (trophoblast), GFP- (epithelium) and non-co-cultured control populations were analysed (in triplicate) by RNA-seq and gene set enrichment analysis (GSEA). RESULTS: Trophoblast challenge induced a wave of transcriptional changes in the epithelium that resulted in 295 differentially regulated genes involving epithelial to mesenchymal transition (EMT), cell movement, apoptosis, hypoxia, inflammation, allograft rejection, myogenesis and cell signalling at 8 h. Interestingly, many of the enriched pathways were subsequently de-enriched by 24 h (i.e. EMT, cell movement, allograft rejection, myogenesis and cell signalling). In the trophoblast, the co-culture induced more transcriptional changes and regulation of a variety of pathways. A total of 1247 and 481 genes were differentially expressed after 8 h and from 8 to 24 h, respectively. Angiogenesis and hypoxia were over-represented at both stages, while EMT and cell signalling only were at 8 h; from 8 to 24 h, inflammation and oestrogen response were enriched, while proliferation was under-represented. CONCLUSIONS: Successful attachment produced a series of dynamic changes in gene expression, characterized by an overall early and transient transcriptional up-regulation in the receptive epithelium, in contrast to a more dynamic transcriptional response in the trophoblast.

Transcriptomic analysis of the interaction of choriocarcinoma spheroids with receptive vs. non-receptive endometrial epithelium cell lines: an in vitro model for human implantation.

PURPOSE: Several in vitro systems have been reported to model human implantation; however, the molecular dynamics of the trophoblast vs. the epithelial substrate during attachment have not been described. We have established an in vitro model which allowed us to dissect the transcriptional responses of the trophoblast and the receptive vs. non-receptive epithelium after co-culture. METHODS: We established an in vitro system based on co-culture of (a) immortalized cells representing receptive (Ishikawa) or non-receptive (HEC-1-A) endometrial epithelium with (b) spheroids of a trophoblastic cell line (JEG-3) modified to express green fluorescent protein (GFP). After 48 h of co-culture, GFP+ (trophoblast cells) and GFP- cell fractions (receptive or non-receptive epithelial cells) were isolated by fluorescence-activated flow cytometry (FACS) and subjected to RNA-seq profiling and gene set enrichment analysis (GSEA). RESULTS: Compared to HEC-1-A, the trophoblast challenge to Ishikawa cells differentially regulated the expression of 495 genes, which mainly involved cell adhesion and extracellular matrix (ECM) molecules. GSEA revealed enrichment of pathways related to cell division, cell cycle regulation, and metabolism in the Ishikawa substrate. Comparing the gene expression profile of trophoblast spheroids revealed that 1877 and 323 genes were upregulated or downregulated when co-cultured on Ishikawa substrates (compared to HEC-1-A), respectively. Pathways favorable to development, including tissue remodeling, organogenesis, and angiogenesis, were enhanced in the trophoblast compartment after co-culture of spheroids with receptive epithelium. By contrast, the co-culture with less receptive epithelium enriched pathways mainly related to trophoblast cell proliferation and cell cycle regulation. CONCLUSIONS: Endometrial receptivity requires a transcriptional signature that determines the trophoblast response and drives attachment.

Vaginal microbiota profile at the time of embryo transfer does not affect live birth rate in IVF cycles with donated oocytes.

RESEARCH QUESTION: What is the relationship between the vaginal microbiota profile at the time of embryo transfer and live birth rates in women undergoing IVF/intracytoplasmic sperm injection (ICSI) with donated oocytes? DESIGN: One hundred and fifty Caucasian women receiving donated oocytes were prospectively included in the study from March 2017 to January 2018. Samples of vaginal fluid were taken immediately before transfer of a fresh single blastocyst and genomic DNA (gDNA) was extracted. Bacterial load as well as the presence of four lactobacilli (L. crispatus, L. gasseri, L. jensenii and L. iners) and species associated with bacterial vaginosis (Gardnerella vaginalis, Atopobium vaginae, Mycoplasma hominis and Prevotella spp. - here collectively termed BVB) were determined by quantitative polymerase chain reaction. Vaginal microbiota profiles for each patient were characterized and correlated with reproductive results. RESULTS: Although bacterial load was variable, a majority of samples were dominated by a single species (80.7%, 121/150). Most samples (76.7%, 115/150) were dominated by Lactobacillus spp., while 23.3% (35/150) were dominated by bacteria associated with bacterial vaginosis. The distribution of microbiota profiles among women who achieved a live birth and women who did not was similar (P = 0.43). Interestingly, we found a significantly higher proportion of samples dominated by L. crispatus- in women achieving live birth compared with those who did not (P = 0.021); this correlation was also statistically significant for biochemical pregnancy (P = 0.039) and clinical pregnancy (P = 0.015). CONCLUSIONS: Our data suggest that bacterial vaginosis-like vaginal microbiota at the time of embryo transfer does not directly affect the live birth rate.