Endometrial Organoids as a Tool for Species Comparison of Endometrial Function

Jessica C. Edge¹, Elton JR. Vasconcelos² and Niamh Forde¹

¹Discovery and Translational Sciences Department, Leeds Institute of Cardiovascular and Metabolic Medicine, School of Medicine, University of Leeds, Leeds, LS2 9JT, United Kingdom

²LeedsOmics, University of Leeds, Leeds, LS2 9JT, United Kingdom

Abstract Text:

Early pregnancy events are diverse amongst Eutherian mammals, with differences in key processes including implantation, placentation and morphology of embryo development. Communication between the endometrium and developing conceptus (embryo and extraembryonic membranes) during the peri-implantation period is critical in ensuring successful pregnancy, as most pregnancy loss occurs during this time. A key function of the endometrium is provision of secretions via uterine luminal fluid (ULF) from the glandular epithelium (GE) to support growth and development until placentation occurs. Studying this function has become easier in some species given the development of endometrial organoids. These 3D in vitro model organoid systems derived from endometrial glandular epithelial cells, can be cultured in vitro and respond to physiological cues. These organoids have been developed in multiple species including human, pig and mouse. Here, we aimed to culture porcine endometrial organoids to 1) investigate transcriptome changes over time, 2) determine the optimal stage to use this model to test function, and 3) examine how the transcriptional landscape differs to other mammals exhibiting different implantation strategies. Porcine glandular epithelial cells were isolated by dissection and enzymatic digestion from slaughterhouse reproductive tracts and cultured in Cultrex 2 extracellular matrix at 38.5^oC, 5% CO₂ (n=3). Organoids were imaged for morphology assessment, and RNA extracted and analysed by RNASeq (NovaSeqXPlus) at the following time points: Day 2, 5, 7, 9, passage 1, 2 and 3. Differential expression analysis by DeSeq2 negative binomial distribution model identified 2098 differentially expressed genes (padj < 0.05) at day 5 compared to day 2, 282 at day 7 vs 5, but only 1 significantly upregulated gene (TFF2) at day 9 vs 7. Following passage, 41 genes differed at passage 1 compared to day 9, 1892 at passage 2 vs passage 1 and 885 at passage 3 vs passage 2. Interestingly, no transcripts were differentially expressed across all time point comparisons i.e. these are temporal or passage-specific transcriptome changes. Genes upregulated in passage 2 compared to 1, and 3 compared to 2 are significantly overrepresented (FDR < 0.05) in processes regulating apoptosis, suggesting the organoids are declining. Gene expression was most stable (only 1 upregulated gene) between day 7 and day 9, which indicates that this may be the optimal period for treatment. Lists of genes with detected expression in all replicates at day 9 were compiled, and compared to those expressed (in all replicates) at day 9 in bovine endometrial organoids, human endometrial organoid control samples and mouse wild-type untreated endometrial organoid samples. Over 48% of genes were expressed in all 4 species, with 1496 unique to human, 571 to porcine, 555 to bovine and 3068 to murine. Four hundred genes were shared only between human and porcine organoids, 450 between porcine and bovine, 228 between bovine and murine, 326 between bovine and human, and 466 between human and murine. Downstream analysis was performed, with interesting highlights including overrepresentation of genes associated with control of stem cell and epithelial cell proliferation attributed to genes that are shared between cow, pig and human organoids; and hormone secretion overrepresented in those shared between cow and human. Overall, these data further characterise endometrial organoids as a tool for in vitro investigation of endometrial function and peri-implantation communication; and identify shared or exclusive gene regulation in different species models.