Bioinformatic approach highlights cellular compositions in endometriosis organoids (EOs)

By obtaining published bulk RNA sequencing data of two menstrual fluid (MFO), two non- treated endometrial biopsies (EMO-N), and three hormone-treated endometrial biopsy-derived (EMO-H) organoids, we established a reference gene matrix from a published single-cell RNA sequencing (scRNAseq) collection of healthy eutopic endometrium, eutopic endometrium (EuE), ectopic peritoneal lesions (EcP) and the adjacent regions to these (EcPA), and ectopic ovarian lesions (EcO) for the deconvolution of major cell types in organoid models.

We performed scRNAseq analysis on 31 samples, identifying 58,072 evaluable single cells after filtering and doublet removal. Cells were classified into epithelial, endothelial, stromal, myeloid, and lymphocyte types using 325 marker genes. The major cell types varied by tissue condition: lymphocytes and stromal cells were most abundant in EuE, EcO, and healthy endometrium. The organoid models (MFOs) showed distinct cellular composition, resembling healthy endometrium more than EuE, with 53% lymphocytes, 24% stromal, 10% epithelial, 8% myeloid, and 5% endothelial cells. EMOs had 25% lymphocytes, 27% epithelial, 24% stromal, 16% endothelial, and 8% myeloid cells. Subclassifying EMOs into EMO-N (higher epithelial fraction, 30%) and EMO-H (25%) revealed differences in epithelial content, with EMO-Ns showing a greater epithelial proportion.

We suggest highlighting cellular heterogeneity within the EOs and revealing that the method of sample collection and patient-specific factors might affect the distribution of major cell types represented in EOs. Furthermore, the presence of non-epithelial cells in EOs should be considered when treatment tests are done using EOs. Key words Endometrial organoids, single-cell sequencing, cell heterogeneity